Below is an alphabetical listing, by presenter last name, of the abstracts presented at NEEPI2015. Thank you to all the outstanding presenters and their co-authors!  Slides, where permitted to be posted, are in separate category folders, according to the conference agenda.  The categories are: Infectious Disease, Chronic Disease, Occupational / Environmental Health, Cross-Cutting, Late Breakers and Rapid Fire. Here is the final conference agenda. (NEEPI2015 FINAL AGENDA.pdf)

Adjoian, Tamar (Adjoian_CH_1045.pdf)

Determining density of unhealthy outdoor advertising in NYC by neighborhood poverty level: Methods and approaches    Rachel Dannefer, MPH, MIA

Calpurnyia Roberts, PhD Shannon Farley, DrPH, MPH    Background: Advertising influences purchasing behavior, and the corporate practice of targeting advertising for unhealthy products to specific populations, including communities of color and children, is well-established. Considerable research has been done on outdoor advertising for alcohol and tobacco products. However, despite widespread acceptance in the clinical and public health community that many other products – including sugary drinks and unhealthy food – also pose risks to population health, little research has been done on outdoor advertising of these products in general and among vulnerable populations. This study is designed to estimate the density of outdoor, stationary advertising in New York City (NYC) overall and in low- and high-poverty neighborhoods for four target product categories: 1) beverages, including sugary drinks, low-calorie drinks, water/seltzer and other non-alcoholic beverages; 2) food products, including fresh produce, fast food, and sweets; 3) tobacco products (including electronic nicotine delivery systems such as e-cigarettes); and 4) alcohol.     Methods: Images of all street-level outdoor, stationary advertising will be captured for retail-dense street segments in a sample of low-, medium-, and high-poverty neighborhoods in NYC in summer of 2015.  NYC Department of City Planning 2014 Primary Land Use Tax lot Output (PLUTO) data was used to identify “retail-dense” street segments (block faces), defined as street segments where at least 50% of the doorway entrances on either side of the street are zoned as retail establishments.  To ensure that each street segment chosen had at least 2 doorways associated with retail lots, streets segments with fewer than four doorway entrances were excluded. Census tract data was used to determine poverty level for each street segment using the following three categories: low-poverty, where fewer than 10% of residents live below the federal poverty threshold; medium poverty, where 10 to <20% of residents live below the threshold; and high poverty, where 20% of residents or more live below the threshold. A spatially random sample of retail-dense street segments was selected, stratified by neighborhood-level poverty (low, medium, and high) and NYC borough. The target sample size was 1,050 street segments.  Images of street-level advertising will be recorded using cameras over a one month period.  Data coders will code ads for their content related to relevant products.  Non-alcoholic beverages, sugary drinks and their low-calorie counterparts will be identified as carbonated soft drinks, energy drinks, sports drinks, fruit drinks, iced tea, and other sugary drinks.  Water/seltzer, coffee drinks, and other beverages will also be coded. Food products will include coding for fast food, sweets, fresh fruits and vegetables, and other foods. Tobacco products will be identified as those which contain tobacco or are otherwise related to tobacco (rolling papers, pipes) and electronic nicotine delivery systems, such as e-cigarettes.  Alcoholic beverages, including beer, wine, hard liquor, and others, will be coded as “alcohol”.    Planned Analyses: Descriptive statistics (means, medians, and proportions) will be computed for each ad category per 500-foot street segment, for NYC overall, and by neighborhood poverty level, with separate analyses looking more closely at differences between low vs. high poverty neighborhoods. Bivariate analyses will be conducted to assess differences in the density of the target ad categories by neighborhood poverty.

 

 

Ackelsberg, Joel Rakeman, Jennifer (8_Ackelsberg_LB_945.pdf)

Three recent laboratory-associated Brucella incidents resulting in >70 exposures, New York City, 2015.  Since April 2015, three laboratory-confirmed cases of Brucella melitensis were reported to the NYC Department of Health and Mental Hygiene (DOHMH). All patients reported symptoms consistent with brucellosis, traveled to a country where the disease is endemic, and consumed unpasteurized milk or a milk product while there. Brucellosis was not suspected by clinicians, and the hospital laboratories were not notified to consider this diagnosis and to take appropriate precautions when handling clinical specimens and isolates. In all three laboratories, Brucella was not suspected, even though the organisms were slow-growing and with Gram stain features consistent with this organism. Microbiology work with the unknown B. melitensis isolates was performed on open benches, including procedures that could aerosolize brucellae. American Society for Microbiology and Association of Public Health Laboratories guidance for recognizing, ruling out and referring biological threat agents was not followed by the three laboratories, delaying diagnosis and, in one case, treatment of an at-risk patient with appropriate antibiotics. A total of 73 laboratory workers had either high- or low-risk exposures and will undergo serological, fever and symptom surveillance for 6 months, or until January 2016 for those most recently exposed. Approximately 80% of those with high-risk exposures started a regimen of post-exposure prophylaxis. To date, no laboratory workers have been diagnosed with brucellosis. DOHMH and the New York State Department of Health have recommended measures to clinicians and clinical laboratories to prevent additional laboratory-associated Brucella incidents from occurring.

 

 

Benowitz, Isaac

Surveillance and Monitoring for Ebola Virus Disease — New York City, 2014–2015        Joel Ackelsberg, MD  Sharon Balter, MD Jennifer Baumgartner, MSPH  Catherine Dentinger, FNP, MS Annie Fine, MD Scott Harper, MD  Keren Landman, MD Fabienne Laraque, MD Ellen Lee, MD Celia Quinn, MD Sally Slavinski, DVM Ann Winters, MD

Don Weiss, MD Jay K. Varma, MD Marcelle Layton, MD  Lucretia Jones, DrPH   Background: The 2014–2015 Ebola virus disease (Ebola) outbreak in West Africa has led to imported Ebola cases into the United States through travelers and health care workers who had recently been in Ebola-affected areas. In August 2014, the New York City Department of Health and Mental Hygiene (DOHMH) established enhanced passive surveillance to rapidly identify and isolate persons with Ebola presenting to health care settings in order to prevent transmission. In October 2014, DOHMH established phone-based active monitoring of persons arriving from Ebola-affected areas following the start of airport-based traveler screening by CDC. We describe our experience triaging calls regarding illness in persons returning from Ebola-affected areas.    Methods: Starting in August 2014, DOHMH defined persons under investigation (PUIs) as having temperature >101.5ºF and compatible symptoms such as severe headache, muscle pain, vomiting, diarrhea, abdominal pain, or unexplained bleeding within 21 days of being in an Ebola-affected area; later guidance lowered and then removed the temperature requirement. We performed outreach, asking clinicians to ascertain travel and exposure history and to isolate PUIs and report them immediately to DOHMH. We also recommended testing for other travel-related diseases. Ebola testing by polymerase chain reaction (PCR) was prioritized for persons with known exposures or with highly suggestive symptoms or laboratory results. In early August, one hospital was designated to manage potential or confirmed cases, and in October this was expanded to four hospitals. Some PUIs with mild symptoms were monitored at home. Persons under active monitoring were asked to record their body temperature twice daily for 21 days after leaving an Ebola-affected area. DOHMH called each person under active monitoring once daily to ask about fever, vomiting, diarrhea, or unexplained bleeding. Persons who reported symptoms or temperature >100.0ºF were evaluated by a medical epidemiologist who obtained additional symptom and exposure information and determined whether the person should be evaluated in a health care setting, isolated and monitored more closely at home, or returned to routine active monitoring.    Results: During August 2014–June 2015, we identified 54 PUIs through passive surveillance (50%) and active monitoring (50%). One (2%) had a known risk factor for Ebola exposure and tested positive for Ebola by PCR. All others reported no known risk factors for Ebola exposure: diagnoses included malaria (28%), renal failure (4%), diabetic ketoacidosis (2%), and stroke (2%). Others had fever alone (20%), GI symptoms (19%), respiratory symptoms (15%), or other symptoms (13%) that resolved without a diagnosis or medical care. Most (67%) sought medical care, but some (33%) were isolated at home until they had spontaneous resolution of symptoms.    Conclusions: We identified 54 PUIs, but only one had Ebola. Malaria was the most common diagnosis. Most illness resolved without any medical care and many persons were able to remain at home without medical evaluation. Providers should suspect other travel-related diseases in persons traveling from Ebola-affected areas and promptly perform appropriate diagnostic testing."

 

 

Bicking Kinsey, Cara   Paoline, Julie

Postpartum Invasive Group A Streptococcal Disease — Southeast Pennsylvania, 2015        Postpartum invasive group A streptococcal disease (iGAS) occurs among approximately 6/10,000 live births in the United States. Although rare, postpartum iGAS can have severe maternal sequelae, including wound infection and dehiscence, hysterectomy, sepsis, and death. Breastfeeding and maternal-infant bonding can be severely disrupted. Invasive GAS cases are reportable in Pennsylvania; however, postpartum iGAS cases are not consistently investigated.  On March 30, 2015, the Pennsylvania Department of Health was notified by the Montgomery County Health Department of an ongoing investigation of postpartum iGAS cases. Three patients had given birth in the same facility and developed iGAS in the previous four months. Routine surveillance of laboratory-confirmed iGAS identified postpartum cases in two additional nearby facilities during January 1, 2015–April 18, 2015. We completed investigations for all three facilities including case confirmation, case-finding and infection control review. We provided recommendations for infection control and enhanced surveillance. All healthcare workers (HCWs) with epidemiologic links to patients were screened for GAS colonization according to Centers for Disease Control and Prevention (CDC) guidance. One colonized HCW was identified and treated with antibiotics. Isolates from the HCW and four patients in the three facilities were emm typed by CDC. All five isolates differed by emm type. We identified no major lapses in infection control; however, we noted opportunities for improvement related to hand hygiene and instrument reprocessing. We found no laboratory or epidemiologic evidence that cases associated with these facilities were linked to each other. Because women might be colonized with GAS at the time of delivery and the birthing process offers a mode of invasive introduction, we cannot confirm that healthcare–associated transmission occurred.  Although these cases of postpartum iGAS were likely unrelated, vigilant surveillance is essential for healthcare–associated infection prevention. We recommend that infection control staff in acute care facilities maintain surveillance for postpartum iGAS. In addition to surveillance, obstetric facilities have a vital need to provide ongoing education opportunities for HCWs, including proper hand hygiene, education about postpartum iGAS, and best practices review to reduce the risk for disease transmission. Prompt investigation and response by public health might prevent additional cases of postpartum iGAS.

 

 

Bicking Kinsey, Cara

Pseudomonas aeruginosa in a Neonatal Intensive Care Unit — California, 2013–2014  Background: In 2013, a neonatal intensive care unit (NICU) experienced a Pseudomonas aeruginosa outbreak with 2 deaths. Infection control measures, water remediation, and intermittent use of point-of-use water filters followed. In September 2014, CDC was notified of additional cases. We investigated to identify risk factors and prevent further cases.  Methods: We defined a case as the first positive P. aeruginosa culture from an NICU patient during June 2013–September 2014. We reviewed medical records from the study period, and matched 1:1 by birth weight, NICU case-patients to control-patients. We performed a case-control analysis by using conditional logistic regression with pooled matched strata adjusted for age at culture. In September 2014, we observed infection control procedures, obtained environmental cultures, and typed isolates by using pulsed-field gel electrophoresis (PFGE).  Results: We identified 31 cases. Case-patients were more likely to have had a peripherally inserted central catheter (adjusted odds ratio [aOR] = 7.65; 95% confidence interval [CI] = 2.03–28.76), invasive ventilation (aOR = 8.26; 95% CI = 1.80–37.83), or exposure to water without point-of-use filters (aOR = 48.14; 95% CI = 9.49–∞) £7 days before culture. We observed deficiencies in hand hygiene. Of 45 environmental samples, 31 (69%) produced P. aeruginosa; 94% of these were water-related. Isolates from the 2 most recent case-patients were indistinguishable by PFGE from water-related samples obtained from the case-patients’ rooms.  Conclusions: This P. aeruginosa outbreak was attributed to contaminated water. Inadequate hand hygiene might have contributed to higher risk for transmission to neonates with invasive devices. Point-of-use filters might be effective in preventing P. aeruginosa cases. Further water-system remediation, continuous use of point-of-use filters, and optimal hand hygiene were recommended.

 

 

Bush, Kathleen

Impact of Heat Index on Emergency Department Admissions in the Northeast        Rebecca Lincoln  Dennis Holt  Julia Gold Melissa N. Eliot  Gregory A. Wellenius     Background: Heat-related morbidity and mortality is a growing public health concern. An aging population and increasing urbanization, combined with rising temperatures and more frequent heat waves, make the Northeast region particularly vulnerable. The primary objectives of this study are to: 1) evaluate the association between maximum daily heat index and daily emergency department (ED) visits in the Northeast (Maine-New Hampshire-Rhode Island); 2) determine whether the existing National Weather Service (NWS) excessive heat warning threshold is sufficiently protective to prevent excess morbidity in the Northeast; and 3) inform local public health action in response to extreme heat events.     Methods: We used distributed lag time-series models to evaluate the association between heat index and daily emergency department admissions in 7 cities in NH, 7 cities in ME, and in the state of RI between May and September from 2000 to 20010. Hourly weather station data from the NOAA National Climatic Data Center (NCDC) were used to calculate daily maximum heat index. For NH and ME, the geographic unit of analysis was defined as all towns in a 10-mile radius surrounding each weather station. All-cause daily ED visits were obtained from hospital discharge data in each state. We applied overdispersed Poisson constrained distributed lag models controlling for long-term time trends, day of week, and federal holidays to each study site. All models considered heat index over the previous 0-7 days and allowed for non-linear exposure-response functions. City-specific risk estimates were then pooled in a meta-analysis to provide a single regional estimate.     Results: This analysis focuses on the New England and covers 100% of Rhode Island, 60% of Maine, and 66% of New Hampshire, for a total of 74% of the population in the 3 study states, as well as 19% of New England. Maximum daily heat indices from May-September ranged from 38.8°F to 110.2°F, with a median of 74°F. We observed an increase in risk of all-cause ED visits at higher temperatures. In a pooled analysis of NH and RI, same-day risk of ED admissions on a 90°F day compared with an 80°F day increased by 1.0% (95% CI: 0.6%, 1.5%), and cumulative risk over a 1-week period increased by 3.3% (95% CI: 2.2%, 4.4%).     Conclusions: This project is a unique demonstration of a regional collaboration across multiple state health departments and universities. Our preliminary results suggest that there may be adverse health impacts associated with heat index before the current NWS excessive heat warning threshold. These locally-relevant results will be communicated to NWS partners in an attempt to revise the current threshold and will inform future local public health actions.

  

 

Cervantes, Kimberly

Maximizing provision of epidemiology services through assessing local health department capabilities and needs       Background: In January 2014, the New Jersey Department of Health (NJDOH) centralized epidemiology capacity by reducing 21 county-based epidemiologists to 11 epidemiologists housed at NJDOH within a newly-created Regional Epidemiology Program (REP). During the program’s first year, focus was placed on providing a consistent set of high-quality epidemiological response services to the 101 local health departments (LHDs). While based in Trenton, eight REP field epidemiologists were assigned to one of five regions in the state with some regions being assigned more than one epidemiologist. The regions used for assignment had been previously established for public health planning purposes with the number of epidemiologists assigned to a region being determined by the number of local health departments, communicable disease volume, population and the number of health care facilities. After the program’s first year, REP realized that the initial system used to assign epidemiology capacity may not have been optimal given the highly variable needs and capacities for communicable disease response at local health departments. In an effort to provide the best epidemiology capacity to all local health departments, an internal assessment of communicable disease response and epidemiology needs capacity was conducted by REP.      Methods: The Regional Epidemiologist assigned primary responsibility for each county prepared an assessment sheet for each LHD looking at communicable disease strengths and areas for improvement; REP successes and challenges faced when working with each LHD; special needs or requests for epidemiological assistance; and current field support strategies. Each LHD assessment was then discussed by the Epidemiologist and the REP management team, at which time a REP action plan was proposed. After the meeting, the REP Coordinator modified the LHD assessments based on the group discussion, drafted an action plan for each LHD, and also a County-based summary/plan of action. Issues that have global applicability across multiple health departments were also identified. Based on each LHD’s assessment and County summary, the REP Management team then assessed each County as high, medium, or low complexity/need. In addition to information provided by the LHD assessments, REP used the following additional criteria for determining complexity/need: number of LHDs; communicable disease volume / burden; special population(s) served; county or regional coordination mechanisms; LHD communicable disease staffing capabilities, responsiveness, and needs; onsite field presence needs; and political concerns.    Results: One-hundred one local health department assessments were completed and reviewed by the REP management team. Action plans and improvement strategies were defined. The assessment results were summarized at the County level, with additional overarching strategies to improve communicable disease capacity and response. Several global items were identified for NJDOH to address statewide. Out of the 21 counties, 9 were classified as “high effort,” in terms of REP time and assistance, 6 as “medium effort,” and 6 as “low effort.” Recognizing that counties can be assigned as “high effort” for different reasons, a workload analysis was conducted for each epidemiologist looking at communicable disease activity by county. Based on the county categorization and workload analysis, regional assignments were modified to maximize epidemiological response across the state.    Conclusion: Conducting an internal assessment of LHD communicable disease capacity and needs was an important tool that offered a qualitative and semi-quantitative methodology to identify areas of greatest need and to analyze whether the current program structure was best able to provide the level of epidemiological assistance required for each county. The assessment process combined practical experience gained from working with local health departments with a more global analysis of workload factors, looking at disease burden, political implications, and partnership/coordination opportunities. Based on the assessment process, the structure of epidemiology response services within REP was modified to better provide epidemiological services across the state.

 

Cervantes, Kimberly  

Forensic Epidemiology in Action – Plague /  A Joint Public Health / FBI Investigation      Background: Public health and law enforcement officials in NJ have been preparing for how to best respond to a bioterrorism threat since the 2001 terrorist attacks. While complementary in some areas, each agency’s goals and objectives are distinct, with different investigational procedures and methodologies as well as issues related to information sharing. Although public health and law enforcement agencies have collaborated on training courses and preparedness exercises, the first actual joint public health / law enforcement interview in NJ was conducted in September 2014 in response to a suspected case of plague.    Methods: The New Jersey Department of Health’s (NJDOH) Communicable Disease Service was notified by a clinical laboratory that they had a routine patient specimen that was consistent with Yersinia pestis. Following standard protocol, NJDOH performed additional testing at the Public Health Environmental and Agricultural Laboratory and worked with hospital infection prevention staff to collect initial clinical and epidemiological data. Confirmatory laboratory tests were positive for Y. pestis and when a natural exposure was not immediately identified, NJDOH notified law enforcement / FBI to discuss the potential threat. Both parties agreed to conduct a joint public health/law enforcement interview with the patient to identify the source of exposure and to determine if a criminal element was involved. : Public health and law enforcement partners clarified each agency’s information gathering objectives and roles and responsibilities prior to the interview.    Results An interview team was quickly assembled to interview the patient and family members. The team identified several challenges related to the patient interview. The patient had an atypical clinical presentation and as such no known incubation period. Establishing a rapport and environment of mutual trust was difficult with the FBI present. The patient didn’t speak English, which added a level of complexity with how the interview was conducted. No known natural or intentional exposure to Y. pestis was identified. Through continued investigation, it was revealed that specimen contamination at the clinical laboratory resulted in the false identification of Y. pestis and the subsequent public health response.    Conclusion: The investigation demonstrated that existing preparedness plans and protocols were effective for prompt communication and notification, and that having routine communicable disease capability and trained personnel from both public health and law enforcement was essential in responding to a suspect bioterrorism event.

 

Choden, Tsering   Hennessy, Robin (Choden_ID_345 B.pdf)

Evaluating the Classification of Congenital Syphilis Investigations in NYC, 2012          BACKGROUND: Congenital syphilis (CS) is a devastating yet preventable disease. To ensure that jurisdictions use a consistent approach to classifying and subsequently reporting cases of this nationally notifiable disease, the Centers for Disease Control and Prevention (CDC) developed a case definition that classifies infants as non-cases, laboratory-confirmed cases, syphilitic stillbirth or probable CS. Since the application of the CS case definition can be complex, in the late 1980s CDC also developed an algorithm to help public health workers with case determination (the ‘old algorithm’). However, use of the old algorithm resulted in case classifications that differed significantly from those arrived at using the case definition. Inconsistent methods of CS classification may have occurred across public health departments, potentially affecting national trends in CS data. In 2013, the CDC released a revised algorithm that is consistent with the case definition. New York City (NYC) uses the case definition to classify CS. However, during the time the old algorithm was available, it may have influenced case classification.   METHODS: We reviewed NYC surveillance data for CS investigations conducted among infants born between 1/1/2012 and 12/31/12. We abstracted maternal and infant data required for case classification including: prenatal care, testing and treatment, and clinical manifestations. We then evaluated each investigation using 1) the case definition, 2) the old algorithm, and 3) the revised algorithm. Using the case definition as the gold standard, we calculated the sensitivities and specificities of the old and revised algorithm, and the classification that NYC had assigned after investigation.  RESULTS: A total of 166 infants born in NYC in 2012 were investigated for congenital syphilis, there were no stillbirths or laboratory-confirmed cases. The case definition classified 8/166 (4.8%) of these as probable cases. Using the old algorithm, 69/166 were classified as probable cases (sensitivity, 87.5% (7/8), specificity, 60.7% (96/158)). Missing or unknown information was responsible for the classification of nearly all (93.5%, 58/62) of the investigations deemed probable cases by the old algorithm and non-cases by case definition. The revised algorithm had a sensitivity and specificity of 100% compared to the case definition. NYC’s case classification had a sensitivity of 50% (4/8) and a specificity of 100% (158/158) when compared with the case definition. One of the probable cases misclassified by NYC (as well as the by the old algorithm) had documentation of adequate maternal treatment with an appropriate serological response; however, the infant evaluation indicated abnormalities and should have been reported as a case. The three remaining probable cases that were missed by NYC differ slightly from each other but in all three cases the infants had elevated values for cerebrospinal fluid white blood cells or protein.  CONCLUSIONS: Applying the CS case definition accurately and consistently is challenging. In NYC, despite using the CS case definition (rather than the old algorithm) for classifying investigations, 50% of the probable cases in 2012 were erroneously classified as non-cases by investigation staff. It is critical to train staff to properly apply case definitions, monitor adherence and review protocols as needed. In an effort to address gaps and align CS prevention activities across the program, NYC recently convened a cross cutting workgroup to review policies, procedures, staff training and assessment process of CS investigations. Continued efforts to accurately classify and report cases is essential at the local level for programmatic and policy development. Other jurisdictions should assure that their program consistently uses the case definition when reporting congenital syphilis.

 

 

Choy, Catherine (FINALVer_Choy.pdf)

Adapting Evidence-based Hypertension Prevention and Healthy Promotion Strategies for Asian American Populations     Simona Kwon, DrPH, MPH

Nadia Islam, PhD  Shilpa Patel, MPH   Catlin Rideout, MPH  Jennifer Zanowiak, MA Stella Yi, PhD, MPH   Background: Asian Americans, the fastest growing U.S. racial/ethnic minority group, face higher risks of stroke and hypertension compared to whites; yet, they report low rates of physical activity and poor nutritional practices compared to other groups. Population- and evidence-based strategies (EBS) represent the gold standard for chronic disease health promotion in the overall population but may be a mismatch for Asian Americans, a cultural and linguistic minority population with characteristics that differ from the intended audience. For diverse populations, adaptation of EBSs is likely to produce more effective relevant programs. The goal of the Racial and Ethnic Approaches to Community Health for Asian Americans (REACH FAR) Project is to use a community-partnered approach to implement culturally adapted/tailored EBSs to address hypertension and improve healthy eating options for Asian Americans (i.e. Asian Indians, Bangladeshis, Filipino and Koreans) living in NYC/NJ.       Methods:  Principles of community-based participatory research and social marketing were used to adapt EBS hypertension prevention and policy, systems and environmental healthy eating strategies, including the Million Hearts campaign for the targeted Asian communities     Results: We will present the process, including the creation of a stakeholder workgroup to oversee the adaption process, the consensus decision-making, and the types and levels of adaptations (deep or surface level cultural adaptations) made to the EBSs.       Conclusions: EB strategies to improve policy, systems, and environmental outcomes are critical to mitigate AANHPI health disparities. Community engagement and social marketing strategies are needed to enhance message relevancy, and diffusion and sustainability of EBSs to reach Asian American or other underserved racial/ethnic audiences."

 

 

Clogher, Paula  Meek, James (Clogher_CC_1045.pdf)

Quantifying the Uptake of Wireless Phone Use in Connecticut, 2000-2014; Impact on Random-Digit-Dial Survey Methodology.       Background: Between 2006 and 2014 the National Center for Health Statistics (NCHS) estimate of American households with wireless only service increased from 12.8% to 44%. Until the 1990’s, with nearly 90% landline coverage in American households, random-digit-dial (RDD) phone sampling of households with landline phones was a cost effective and statistically sound method used by public health researchers to identify control subjects for case-control studies. With rapid uptake of wireless technology the validity of RDD landline sampling model has been questioned making it increasingly important to validate the national estimates of wireless phone use. In Connecticut, FoodNet (CT FN) conducts case-control studies to identify risk factors for foodborne disease based on statewide population-based surveillance for laboratory confirmed cases of foodborne pathogens. We describe the uptake of wireless phone use in the cohort of incident foodborne disease cases identified by CT FN from 2000-2014.  Methods: North American Local Exchange NPA NXX Database, NALENND™, a database that provides the line type (e.g., landline, wireless) of a phone numbers was purchased and merged with CT FN data using SAS. Time was aggregated to three five-year increments: 2000-2004, 2005-2009, 2010-2014. Proportions and rates were calculated to describe the distribution of landline and wireless phone use among CT FN cases by demographic characteristics and pathogen.  Results: From 2000-2014 wireless phone use increased 6-fold (3.5% to 26%) and landline use decreased 23.5%. Landline use decreased in all case age groups except for those >50. The highest rate of wireless use (30.5%) was in the 0-<1 age group and the greatest change in wireless use, an 18-fold increase, was in the 50-59 age group. The percent change in wireless line use was greatest among black (700%) and Asian (1000%) groups. The rate of change in wireless uptake varied widely by pathogen with a low of 22% for cases with Yersinia to a high 937% for cases with Shiga toxin E. coli.  Conclusion: Wireless phone use by CT FN cases increased in all age and racial-ethnic groups and grew most rapidly among non-white groups. Meanwhile, landline use is decreasing among children and young adults. High rates of wireless use and decreasing rates of landline use in pediatric cases most likely represents use patterns of the parent or guardian of the child. Use of RDD for control recruitment in CT may bias results due to underrepresentation of the young and minority groups with no landline. Additionally, the observations that the rate of wireless use and change over time varied dramatically by CT FN pathogen suggests that wireless use may be linked to behaviors or characteristics specific to exposure which may further contribute to confounding.

 

 

Cohen, Chari (Cohen_ID_145.pdf)

Chronic hepatitis B infection among high-risk Asian and Pacific Islander communities in Philadelphia     Chronic hepatitis B (CHB) is a serious liver infection caused by the hepatitis B virus (HBV). Two million people in the U.S. have CHB, of which 50% are Asians or Pacific Islanders (API). Up to 25% will die prematurely of cirrhosis or liver cancer if unmanaged. It is estimated that 75% of infected APIs in the U.S. remain unaware of their CHB, putting them at greater risk of morbidity and mortality. Due to the lack of systematic surveillance, and the lack of API inclusion in surveillance, CHB infection rates among APIs remain dramatically underestimated. This study sought to better understand the CHB prevalence, and risk factors associated with infection and protection, among high-risk API communities in Philadelphia. Community-based HBV screenings were conducted for 2,047 APIs from 2007-2014. An IRB-approved survey collected information on demographics, healthcare usage, and past/family history of CHB. Blood samples confirmed HBV infection and protection status. Data were analyzed using SAS 9.3. Descriptive statistics, frequency analysis, and logistic regression were used to describe the participant population. Of 2,047 people, 58% were female; 98% were foreign-born; 87% were limited-English proficient; 36% had health insurance; 12% knew of a family member having HBV; 19% reported being previously screened; 8% tested positive for infection; and 58% were protected. Controlling for all other variables (at p<0.5), infected individuals tended to be male, between the ages of 18‐29, disclosed that they had been previously tested for HBV, had a high school diploma or less, and had either a family member with HBV or HCC, or lived with an infected person; protected individuals in the sample tended to be between the ages of 18‐29 and born in Korea; susceptible individuals tended to be over age 55, with no family history of HBV, no previous HBV test, and born in a country with no (or a limited) national HBV vaccine program. Data collected at community events, in addition to routinely-collected surveillance data, can help understand local CHB prevalence and risk factors among APIs. These data can be used to develop targeted interventions to address the CHB needs of API communities in Philadelphia.

 

 

DiLonardo, Steve

Identification of areas with high concentrations of persons with poorly-controlled diabetes in NYC using spatial pattern analyses    Co-author 1: Qun Jiang, MA is the Director of A1C Registry Surveillance for the Division of Primary Care and Prevention at NYC DOHMH    Co-author 2: Winfred Y. Wu, MD is the Executive Director of the Development Team for the Division of Primary Care and Prevention at NYC DOHMH   The New York City Department of Health and Mental Hygiene (DOHMH) established the A1C Registry (Registry) in 2006 to enable DOHMH to respond to the diabetes epidemic.  The Registry contains A1C test results of New York City (NYC) residents reported from clinical laboratories.  As part of the existing Registry surveillance activities, we developed a spatial pattern analysis which was used to information a place-based initiative.       Using 2013 calendar year Registry data, we geocoded and plotted ~80,000 data points onto an ArcGIS map of New York City, with each point representing adults with diabetes in poor control (A1C greater than 9%). We utilized the ArcGIS point density analysis with the goal of identifying areas of high density. The point density analysis calculated the density of points (representing persons with diabetes in poor control) in a defined cell size and within a defined radius on a moving basis to create a density gradient map. The cell and radius size and number of density gradients were optimized for precision. High density areas, which represent a cluster of cells of the highest density gradients, guided the identity of neighborhoods with the highest occurrence of adults with diabetes in poor control.    For this analysis, the point density analysis indicated that a cell size of 100 square feet within a 0.1 mile radius neighborhood was optimal for defining a neighborhood with a density of adults with poorly controlled diabetes. In general, the highest density areas coincided with lower income neighborhoods and the relative percentage of adults with diabetes in poor control was roughly one order-of-magnitude greater than the percentage of area representing the highest density gradients. The largest areas of poorly controlled diabetes were principally situated in four neighborhoods – Central Brooklyn, East Harlem, Washington Heights, and the South Bronx.  Given the high concentration of adults with poorly controlled diabetes in East Harlem and the presence of a local health department office, a community health worker initiative to address this burden of disease was implemented within a year of completing the analysis.     Findings from the point density analysis are an important surveillance tool as they can inform public health decision making, particularly when situating place-based interventions intended to reach the greatest numbers of the at-risk population.

 

 

Dinitz-Sklar, Jill     Kratz, Natalie (DinitzSklar_ID_1045.pdf)

No Need to Reinvent the Wheel - Adapting Existing Systems for Active Monitoring During the Ebola Outbreak Response in New Jersey 2014-2015   Background: In response to the Ebola virus disease (EVD) outbreak in West Africa, Center for Disease Control and Prevention (CDC) recommended that states actively monitor all travelers arriving in the US from an EVD-affected area beginning October 27, 2014. Active monitoring requires conducting a risk assessment for each traveler upon arrival and daily contact with travelers who report temperatures and any EVD-like symptoms for 21 days following departure from the EVD-affected area. New Jersey (NJ) anticipated a large volume of travelers would need to be monitored due to a significant Liberian population, and NJ’s proximity to New York City.  The New Jersey Department of Health (NJDOH) utilized the existing Communicable Disease Reporting and Surveillance System (CDRSS) and Hippocrates, NJ’s emergency preparedness and response system, to capture the necessary data and communicate with local health departments (LHDs), which were responsible for actively monitoring travelers.  CDRSS is accessible from remote locations and allows for interoperability at both the state and local level.  Also, it allows for clear assignment of individuals to the LHD where the person either resides or is staying in NJ.  LHD users, as well as NJDOH, have real-time access to case reports in a secure environment that also allows for seamless communications.  Data are housed on a secure server that has the capability to generate reports for analytic purposes.  Methods: To monitor travelers, NJDOH enters report information received from the CDC Division of Global Migration and Quarantine into CDRSS, which automatically notifies LHDs of new reports. LHDs conduct risk assessments using an embedded Hippocrates survey, enter temperatures, and communicate relevant information to NJDOH using CDRSS during the 21 day monitoring period. When those under active monitoring relocate to a different jurisdiction within NJ, LHDs are able to transfer individuals using CDRSS.  LHD closes cases after the monitoring period ends or when the individual is transferred outside of NJ. NJDOH reviews closed reports for completeness. Using a numeric field unneeded for EVD monitoring purposes, NJDOH is able to enter the number of days each individual is monitored and appropriately reimburse LHDs for additional work. NJDOH generates reports for internal purposes and at CDC’s request using data stored in CDRSS.  Results:  LHDs and NJDOH staff were already familiar with the CDRSS system, which allowed for minimal training to be required during the implementation of active monitoring.  Both risk assessments and temperature logs are able to be housed within the CDRSS system using existing infrastructure.  Using the existing CDRSS and Hippocrates systems, NJDOH has been able to actively monitor over 1000 individuals as of July 2015, and securely manage, maintain, and analyze data for internal, state and national stakeholders.    Conclusion: CDRSS poses some challenges in that the system was not set up to create ad-hoc modules. To store data requested by stakeholders, but not available in CDRSS, such as days monitored, unused fields in reports were used by staff. Since it is a secure site and already in use by disease investigators both at LHDs and NJDOH, it was the most efficient system to use for the response, despite the restrictions on adding variables. The use of CDRSS to actively monitor travelers also assisted in facilitating dialogue to make changes to the existing CDRSS system; as there is a redesign planned, moving forward the need to respond to emergent issues will be incorporated into the new system.

 

 

Erdogdu, Pinar (Erdogdu_CC_1045.pdf)

Better, Stronger, Faster: Making the Case for Adding Data Fields to Syndromic Surveillance,  New Jersey, 2015     Authors:  Pinar Erdogdu, MPH1, Teresa Hamby, MSPH1, Stella Tsai, PhD, CIH1   1 NJ Department of Health, Communicable Disease Service  Introduction:  In New Jersey, real-time emergency department (ED) data are currently received from acute care and satellite EDs by Health Monitoring Systems Inc.’s (HMS) EpiCenter system, which collects, manages and analyzes ED registration data for syndromic surveillance, and provides alerts to state and local health departments for surveillance anomalies.   EpiCenter receives chief complaint data which are pre-diagnostic information from 78 of 80 acute care and satellite EDs.  The growing need for more specific information raises the possibility that other data elements from EDs such as triage notes can be of utility in detecting outbreaks e.g., Ebola without a significant delay.  This study evaluates the inclusion of a new data stream, triage notes, in EpiCenter to detect the recent increased usage of synthetic cannabinoids.   Methods:  In April 2015, the New Jersey Poison Information and Education System (NJPIES) reported an increase in number of calls to their center requesting consultation regarding synthetic cannabinoid reactions in ED patients.  NJPIES provided a list of call data for the New Jersey Department of Health (NJDOH) to evaluate the chief complaint included in EpiCenter.   NJDOH created a custom classification in EpiCenter to evaluate the increased synthetic cannabinoid-related ED visits using chief complaint data. DOH staff included the keywords of “black magic”, “black mamba”, “cloud 9”, “cloud 10”, “incense”, “k2”, “legal high”, “pot potpourri”, “spice”, “synthetic marijuana”, “voodoo doll”, “wicked x”, and “zombie” which were obtained from the New York City Department of Health and Mental Hygiene.  Staff also included additional keywords, “agitation”, “k-2”, ”moon rocks”, “seizure”, ”skunk”, and ”yucatan” to better characterize the related event.    For comparison, NJDOH performed a text search using the same keywords included in the custom classifier to evaluate the synthetic cannabinoid related ED visits within the triage notes fields from three emergency departments currently providing that information.            Results:  Between April 2015 - May 2015, using the key words, 18 (36%) out of 50 ED visits were identified via chief complaint data and 32 (64%) of the ED visits were not captured by the custom classification due to the non-specific keywords.  Among the identified ED visits, the most commonly used keywords were “seizure” and “marijuana”.  Of the 50 calls, 6 ED visits were admitted to hospitals that submitted triage notes data in EpiCenter.   Using the same keywords in triage notes query, 5 (83 %) out of 6 ED visits were identified.  The most commonly used keywords were “k2” and “marijuana”.  Conclusions:  Review of synthetic cannabinoid calls provided by NJPIES suggested that triage notes included more specific keywords and captured most of the calls.  The inclusion of triage notes has been initiated in NJ and this new data source will provide vital information to syndromic surveillance, which is expected to lead to an increase in early detection and response to health events like Ebola, Enterovirus D68, and, as outlined here, drug overdose surveillance.      Acknowledgements: Bruce E. Ruck, PharmD, RPh, Elizabeth Kostial"

 

 

Fiddner, Jennifer (Fiddner_ID_345 B.pdf)

Enhanced Lyme disease surveillance, Allegheny County, 2014   Background:   Classification of Lyme disease cases depends on clinical and laboratory criteria.  Between 2004-2008, the Allegheny County Health Department (ACHD) received <50 reports per year and was able to investigate all, confirming <30 cases per year.  In 2009, the number of positive laboratory tests reported began to increase and health department staff were unable to investigate cases due to time constraints, resulting in very few confirmed cases. In 2014, ACHD implemented enhanced Lyme disease surveillance with the assistance of an AmeriCorps member and student interns. The purpose of this initiative was to more accurately classify cases, better quantify the overall burden of Lyme disease within the county, and disseminate prevention recommendations to the public.   Methods:  All 2014 Lyme disease reports were investigated. If laboratory reports lacked physician contact information, staff made numerous phone calls to obtain valid fax numbers for providers. Providers were then faxed a one-page form and asked to provide demographic and clinical information. When providers could not be located after several attempts, patients were interviewed directly. The additional information collected was entered in the PA-NEDSS database. Paper reports submitted to ACHD by providers with clinical information were entered into the PA-NEDSS database.  Results:  Of the 1,455 cases reported in 2014, clinical information was obtained on 1,287. Of these, 822 (64%)  were classified as “confirmed” or “probable” based on Centers for Disease Control and Prevention 2011 case classification criteria. Males accounted for 57% of these cases. There was a bimodal distribution of cases by age group. The highest incidence occurred in children in the 5-9 year age group, followed by adults in the 55-59 year age group. The presence of erythema migrans (EM) was associated with 64% of cases.   Conclusions:  The proper classification of Lyme disease cases requires access to both pertinent demographic, laboratory and clinical data. Most cases are reported electronically by laboratories and do not include clinical information.  By devoting substantial staff and volunteer time, ACHD was able to confirm hundreds of cases, verifying the huge increase in Lyme disease in the county over the past 5 years and confirming a demographic profile similar to that in other endemic areas. Because of limited staff availability, ACHD plans to investigate a sample of Lyme disease reports to estimate the number of cases in future years, an established procedure followed in several other endemic areas.

 

 

Figgatt, Mary (Figgatt_LB.pdf)

Results from a Pilot Surveillance System of Harmful Algal Bloom-associated Illnesses in New York State, 2015    Lloyd Wilson, PhD, MA   David Dziewulski, PhD Eric Wiegert, MPH  James Hyde, MS  Introduction: Harmful algal blooms (HABs) in freshwater are defined as excessive growths of cyanobacteria and associated organisms with the ability to produce toxins that can be dangerous to humans or animals. These toxins include: microcystins, anatoxins, cylindrospermopsins, saxitoxins, and lipopolysaccharides. Exposure to HABs can occur through recreational water activities or drinking contaminated water, and human and animals illnesses associated with HABs have been documented to range from mild contact irritation to organ failure or death. While HABs and associated illnesses have been previously documented in New York State (NYS), the NYS Department of Health, Center for Environmental Health (CEH) designed a pilot surveillance system for HAB-associated illnesses during 2015 summer season to better address issues such under reporting and topic unfamiliarity within the public and healthcare providers.  Methods: The approach was to identify counties in which HABs were known to occur and also those with local health departments interested in participating in enhanced surveillance of illness potentially related to HABs. CEH administered a questionnaire to all Local Health Departments (LHDs) to identify those that would participate in the 2015 pilot surveillance system. The pilot surveillance system built on previously existing programmatic tools and activities such as: HAB notifications, beach closures, water advisories, and illness investigations. New HAB specific activities introduced included: (1) increased public outreach about the health concerns, (2) notification of illness directed at physicians, veterinarians and the general public, (3) surveillance of hospital data, and (4) increased partnership with other agencies and organizations.   Results: A total of 16 of the 62 counties in NYS a priori participated in the 2015 pilot surveillance program. In addition, other LHDs became involved in the pilot surveillance activities due to known HABs and/or suspected HAB-associated illness reports. Human and animal cases were reported throughout the summer season. Cases were identified through reports from other agencies and organizations, self-reporting to LHDs, hospital data, and reports from physicians and veterinarians. Multiple illness clusters were identified. Each report was evaluated based of the Centers for Disease Control and Prevention HAB-associated case classification. The epidemiologic and environmental aspects of all cases and clusters will be discussed during this presentation.   Discussion: HABs are an emerging public health concern that need to be better understood clinically and environmentally. This presentation will provide details about the methods, overall results, limitations, and strengths of the 2015 pilot HAB-associated illness surveillance system."

 

 

Fitzhenry, Robert  Weiss, Don (Fitzhenry1_ID_1045.pdf)

Shoe-Leather Epidemiology in the 21st Century     Co-authors: Sharon K. Greene, Robert W. Mathes, Robert Fitzhenry, Ramona Lall, Vasudha Reddy, HaeNa Waechter, Mike Antwi, Paula Del Rosso, Marie Dorsinville, Beth Nivin, Lucretia E. Jones, Kristen Forney, Eric R. Peterson, Hannah Mandel, Jennifer Baumgartner, Sharon Balter, Annie D. Fine, Ellen H. Lee and Marcelle C. Layton      The term “shoe-leather epidemiology” invokes images of public health field staff scuffing their soles in pursuit of clues in outbreak investigations and is perhaps homage to the groundbreaking investigations of John Snow during 19th century cholera outbreaks (1). For the 21st century, the image requires a makeover to reflect how advances in technology (e.g., electronic medical records, social media, electronic laboratory reporting, and routine geocoding) are improving the efficiency of field investigations and disease cluster detection (2).     Traditionally, medical chart review, phone calls to infection control practitioners or both have been needed to complete case investigations, sometimes involving repeated calls and several hours of commuting. Hospital personnel are under ever-increasing demands on their time, and delays in case investigations can be worrisome. The Bureau of Communicable Disease (BCD) of the New York City (NYC) Department of Health and Mental Hygiene uses Regional Health Information Organizations, saving staff time by remotely logging in to retrieve laboratory and other data from their desktop computers.    A prototypical use of social media involves identifying contacts to case-patients with reportable diseases, such as invasive meningococcal disease (IMD). BCD used social media to find contacts of a fatal IMD case in 2010 (3) and again during a prolonged outbreak of IMD among men who have sex with men, Facebook, Grindr, and Manhunt were used both to send messages to potential contacts and to post messages to individuals at risk (4). Cell phone logs have also been used to identify IMD contacts and uncover links among cases. Subscription databases, e.g., Lexis-Nexis Accurint and CLEAR, have proven useful for locating address and contact information for contacts of reportable disease cases and West African travelers from Ebola Virus Disease-affected countries under active monitoring. BCD additionally uses the business review website Yelp to scan restaurant reviews for foodborne illness key words as an adjunct to traditional notifications of restaurant-associated outbreaks (5).    BCD receives >6,000 laboratory or provider reports weekly. The advent of electronic data transmission has necessitated algorithms to rapidly process data and facilitate disease cluster recognition. To detect temporal and spatio-temporal clusters of reportable diseases, BCD uses a modification of the historical limits method (6) and the prospective space-time permutation scan statistic in SaTScan (7). Reportable disease cases are routinely geocoded, and the unique building identification number is used to alert investigators to the occurrence of certain diseases in particular buildings (e.g., influenza in long-term care facilities or Legionella in residential housing) (8). In addition, BCD employs emergency department visit chief complaint surveillance to scan for syndromes and key words of concern. Examples include: “exposure to meningitis” to facilitate specialized testing of suspected IMD cases; “tattoo” as part of an outbreak investigation (9); and “K2” to assist colleagues in monitoring adverse events from synthetic marijuana (10).      Traditional shoe-leather epidemiology and innovative disease investigation methods will always be needed. Epidemiologists now have a new array of tools to support creative public health investigations.     References  1. Koo et al. In Snow’s Footsteps: Commentary on Shoe-Leather and Applied Epidemiology. Am J Epi 2010;172:737-9.  2. Minniear et al. Using electronic surveys to gather information on physician practices during a response to a local epidemic—Rhode Island, 2011. Ann Epi 2013;23:521-23.  3. Gounder et al. Using the internet to trace contacts of a fatal meningococcal case-NYC, 2010. J Public Health Manag Pract 2012;18:379-81.  4. Kratz et al. Community-Based Outbreak of Neisseria meningitidis Serogroup C Infection in Men who Have Sex with Men, NYC, New York, USA, 2010-2013. EID 2015;21:1379-86.  5. Harrison et al. Using Online Reviews by Restaurant patrons to Identify Unreported Cases of Foodborne Illness—NYC, 2012-13. MMWR 2014;63:441-45.  6. Levin‐Rector et al. Refining historical limits method to improve disease cluster detection, NYC, New York, USA. EID 2014;21:265‐272.   7. Kulldorff et al. A space-time permutation scan statistic for disease outbreak detection. PLoS Medicine 2005;2:e59.  8. Levin‐Rector et al. Building‐level analyses to prospectively detect influenza outbreaks in adult long‐term care facilities—NYC, 2013‐2014. Am J Infect Control 2015; May 8.pii: S0196-6553 (15) 00225.  9. Kotzen et al. Using Syndromic Surveillance to Investigate Tattoo-Related Skin Infections in NYC. PLoS One 2015;doi:10.1371/journal.pone.0130468.  10. Nolan. Increase in Synthetic Cannabinoid-Related Emergency Department Visits Detected by Syndromic Surveillance—NYC, 2014. Presented at:Council of State and Territorial Epidemiologists Annual Conference. Boston, MA, June 15, 2015

 

Fitzhenry, Robert (Fitzhenry_ID_1045.pdf)

Recurrent clusters of legionellosis in a New York City hotel    Background: Hotel A, a 32-floor, 1,331-room facility located in midtown Manhattan in New York City (NYC), has been the source of recurrent cases of legionellosis, with previous clusters in 1997 and 2013. During the 2013 investigation, water sampling of the hotel’s plumbing system and cooling tower revealed Legionella pneumophila in water tanks, the cooling tower, and guest room water outlets. A copper-silver disinfection system was installed, and routine cleaning of the cooling tower was increased from annually to semiannually. During renovations, which were completed in 2013, many risers were replaced, and new water and ice dispensers were installed in guest areas. In June 2015, the NYC Department of Health and Mental Hygiene (DOHMH) was notified of a new cluster of legionellosis cases among foreign travelers who stayed at Hotel A. The ill travelers were initially reported to the Centers for Disease Control and Prevention (CDC) by the European Legionnaires’ Disease Surveillance Network (ELDSNET). ELDSNET issued an alert about the hotel and requested follow-up on the investigation.    Methods: On May 28th, DOHMH received its first report of a European traveler with legionellosis who had stayed at Hotel A during his incubation period. During June 20–July 7, eight additional legionellosis cases were reported among European travelers to NYC; six had stayed at Hotel A. DOHMH sent an Epi-X notification to all U.S. health departments calling for legionellosis cases with a history of visiting NYC from May 1 onward, and asked to be notified by hotel management if any employees developed symptoms of pneumonia. DOHMH reviewed travel questionnaires administered to cases by the European Centers for Disease Control (ECDC) and reviewed NYC case reports to determine if other NYC cases were linked to Hotel A.     On June 27, an environmental consultant employed by a European tour operator sampled the hotel’s hot and cold water systems, cooling tower, and guest rooms. At DOHMH’s recommendation, samples were split with a U.S.-based consultant for testing in a CDC ELITE-certified laboratory. The cooling tower was cleaned on June 28, and on June 30, DOHMH personnel resampled the hot and cold water systems, guest rooms, and ice and water dispensers. A total of 174 samples were taken (European consultant [25], U.S. consultant [125], DOHMH [24]). Samples collected by DOHMH were split for polymerase chain reaction (PCR) analysis at the New York State Department of Health ─ Wadsworth Center Laboratory (WC) and culture at the NYC Public Health Laboratory (PHL).    Results: The seven case-patients in 2015 who lodged at Hotel A ranged in age from 41 to 69 years; four were from England, and one each was from Germany, France, and the Netherlands. Check-in dates ranged from May 6–June 20, and symptom onset dates ranged from May 16–June 28. All were diagnosed by urine antigen testing. Five of seven case-patients completed interviews. Four had three other common local exposures in addition to the hotel, including a monument and a park, which have large decorative fountains.     The hotel has two hot water zones supplied by separate water heating systems and a cooling tower located at the level of the third floor and bordered by three exterior walls of the hotel. Potable water sampling on May 18, 2015 demonstrated <1 CFU/mL of Legionella species in all samples. The cooling tower had not been retested since September 2013. Cases reported staying in nine rooms in total (two changed rooms during their stay). Three case rooms faced the cooling tower and seven were in the block closest to the cooling tower.     On July 2, WC reported that 23 of 24 potable water samples were negative for L. pneumophila DNA by PCR, and one was inconclusive. The inconclusive sample, from a 27th floor water dispenser, was positive for L. pneumophila serogroup 1 by culture at the PHL. On July 7, DOHMH was notified that by the European consultant that eight cooling tower water samples were preliminarily positive for L. pneumophila by culture; multiple cooling tower samples and two guest room samples were confirmed by the U.S. consultant by culture as positive for L. pneumophila serogroups 1, 4, and 6, with concentrations ranging from 400 to >3,000 colony-forming units per milliliter (CFU/mL) in the cooling tower samples, and from 3 to 20 CFU/mL.  All other potable water system samples taken by both DOHMH and the European consultant were negative. Final reports are still pending.    Conclusion: Preliminary results suggest Hotel A’s cooling tower was the source of Legionella pneumonia in seven European travelers; it is unclear whether this was also the source of the hotel’s previous Legionella outbreaks. Our findings highlight the need to take into account a cooling tower’s position relative to case patient rooms when assessing its role as a Legionella source and to consider cooling tower disinfection independently of the disinfection of potable water systems.

 

 

Garnes, Preston (Garnes_CC_1045.pdf)

The Usefulness of Linking an Immunization Information System to a Disease Surveillance Database     The Usefulness of Linking an Immunization Information System to a Disease Surveillance Database  Preston Garnes, MPH, Robert Arciuolo, MPH, CPH, Alexandra Ternier, MPH, Vassiliki Papadouka, PhD, MPH, Jennifer Rosen, MD, Tamara Brantley, MPH, CPH, and Jane R. Zucker, MD, MSc.  Bureau of Immunization,   New York City Department of Health and Mental Hygiene,   Long Island City, N.Y.    Background: In October 2011, the New York City (NYC) Department of Health and Mental Hygiene (DOHMH) linked its immunization information system (IIS), the Citywide Immunization Registry (CIR), to its disease surveillance database, Maven. The CIR is a population-based database that collects immunizations administered to all children aged 0-18 years residing in NYC. Maven is a surveillance and case management system used by the Bureau of Immunization (BOI) staff for the investigation of vaccine-preventable diseases (VPDs). The integration of the two systems allows BOI surveillance staff to query the CIR and automatically upload relevant immunization information to Maven. The connection also allows staff to request the CIR be updated with additional immunization records that are located during the investigations.     Objective: To describe how the linkage between NYC’s CIR and Maven databases affected BOI’s VPD investigations during the first three years of implementation.    Methods: Immunization data were extracted from Maven for measles, mumps, rubella, pertussis, and Streptococcal pneumoniae case and contact investigations conducted between 2012 and 2014. The following components were evaluated: 1) the number of investigations for which an immunization record was found in the CIR through the Maven-CIR linkage; 2) the number of immunization doses prior to the investigation event date (i.e. case onset date or contact exposure date) that were uploaded to Maven from the CIR; 3) the number of investigations for which surveillance staff requested immunizations be added to the CIR; 4) self-reported surveillance staff practices before and after the Maven-CIR linkage; and 5) the number of immunization variables documented in the surveillance database before and after the linkage.       Results: A total of 12,705 records were investigated from 2012 to 2014: 3,654 were cases and 9,046 were contacts. Immunization records were found in the CIR for 5,166 investigations (40.7%) and 10,480 immunization doses administered prior to the event date were uploaded from the CIR to Maven. Data were stratified by disease and of the records found nearly half (47.8%) were associated with measles disease. During the three-year period, surveillance staff requested immunizations to be added to the CIR for a total of 632 investigations (12.2 %). Prior to the linkage, staff obtained immunization histories from providers and/or the patients directly, or had to look up each patient individually in the CIR, which can be time-consuming. Staff reported that the Maven-CIR linkage saved time during investigations by facilitating more rapid ascertainment of patients’ vaccination status. Staff also indicated that using the CIR has additional benefits to surveillance investigations beyond immunization dose information including access to information on parents, siblings, and immunizing provider of the cases or contacts. Lastly, more immunization variables can be documented in the surveillance database after the linkage as compared to before the linkage (i.e., type of vaccine, if vaccine is a valid dose, lot number, etc.). This has allowed for more detailed investigations and decreased follow-up activities.     Conclusion: The Maven-CIR linkage made VPD investigations more efficient. Since the linkage, immunization records are found and automatically uploaded for a large proportion of surveillance investigations greatly facilitating identification and follow up of exposed individuals. The linkage has also helped with the completeness of the CIR since surveillance staff frequently request for missing vaccinations to be added to the CIR.

 

 

German, Susan (German_EO_145.pdf)

Use of New Jersey Poison Information and Education System (NJPIES) during Hurricane Sandy   Pauline Thomas, MD  Amy Davidow, PhD  Steven Marcus, MD

Bruce Ruck, PharmD, RPh  Title:  Use of New Jersey Poison Information and Education System (NJPIES) during Hurricane Sandy    Introduction:  Hurricane Sandy, the most destructive storm of the 2012 Atlantic hurricane season, led to widespread evacuations, flooding, power outages, and gasoline shortages, as well as building, infrastructure, and environmental damage.  These events, in turn, resulted in exposures to potentially toxic substances and a need for information related to poisons/toxins, in the home and environment; unintentional misuse or therapeutic errors regarding medicine; and food storage and other issues related to lack of refrigeration.  NJPIES, which operates New Jersey’s only Poison Control Center, was able to stay open 24/7 before, during and after the storm, providing expert advice on a wide variety of topics.  This study characterized the use of NJPIES in the days immediately preceding, during, and after the storm to determine opportunities for targeted public health education and intervention.    Methods:  A retrospective review of electronic NJPIES case data from October 1, 2012 through December 31, 2012 was conducted.. Most Sandy-related cases had been coded as such by NJPIES staff either retroactively via case review including case narratives, or, once the code was established, in real-time.  Additional Sandy-related cases, identified by performing case narrative review, expanded the set of Sandy-related cases. Descriptive analyses were performed for case frequencies, exposure substances, type of information requested, caller site, exposure site, and gender.  The date of the first case designated as Sandy-related was October 26, 2012, when evacuation advisories were first announced.    Results:  Among a total case frequency of 10,806 from October 26, through December, 2012, 498 were deemed Sandy-related.  Sandy-related case volume peaked on November 1 (n=79).  During landfall, total case volume dipped slightly.  The five most frequent Sandy-related exposures or information requests were gasoline exposure (n=160), carbon monoxide (CO) exposure (n=100), poison information (n=51), food poisoning/spoilage information (n=48), and water contamination/information (n=16).  Gasoline and CO exposure cases were characterized as follows: 6.3% of gasoline exposure cases were initiated by a healthcare facility, whereas 64% of CO cases were initiated in this manner; males comprised 86.9% of gasoline cases vs. 38% of CO cases; and 85.6% of gasoline and 81% of CO exposures occurred at the patient's own residence.    Conclusions:  NJPIES provided essential guidance and information for callers throughout New Jersey before, during, and after the storm. This study identified the need for enhanced public health education and intervention particularly regarding gasoline siphoning as well as proper use of gasoline-powered generators and cleaning and cooking equipment in order to prevent carbon monoxide poisoning."

Gleeson, Jessie Impact of Drinking Water Source on Associations of Gastrointestinal Illness and Extreme Rainfall in New Jersey    Gastrointestinal illness (GI) has been associated with heavy rainfall. Storm events and periods of heavy rainfall and runoff can result in increased raw water turbidity and are positively associated with increased levels of bacteriological contaminants in raw water. Surface water supplies are open to the environment and runoff can directly influence the presence of contaminants. Associations of GI with weather, water quality parameters, and water source have been previously investigated. Specifically, increases in GI disease have been correlated to extreme rain events among persons receiving drinking water from a surface water source.      We estimated associations of extreme rainfall events and inpatient hospitalizations for GI in New Jersey using a case-crossover study design. All cases of GI hospitalization from 2009-2013 were extracted with a final sample size of 47,527. The day exactly a week before and a week after a case was admitted to the hospital, were assigned as control days. Each case-day and control-day was assigned an estimated daily precipitation value. The 90th percentile value of rainfall was categorized as extreme rainfall. Conditional logistic regression was used to create a ratio of case-days to control-days to estimate odds ratios of GI occurring on days with extreme rainfall. All analyses were stratified by warm versus cold season and controlled for temperature and humidity. Each case was geocoded to a water system layer for New Jersey and assigned into a ground water, surface water, or ‘other’ category. Estimates were stratified by drinking water source to determine how water source modifies effect estimates.     Initial findings indicate associations with extreme rainfall events and an increased risk of hospitalization for GI during the warm season. Stratification by water source, found study participants residing in a surface water source on the day of an extreme rainfall event during the warm season are at an increased risk of hospitalization for GI while those from a ground water or ‘other’ water source are not associated with an increased risk. More analyses are underway to assess lagged effects and stratification by age.     Water systems with surface water sources may play an important role in preventing GI hospitalizations during and immediately following extreme rainfall events.

 

Goun, Barbara  Opiekun, Richard  (Goun_EO_145 .pdf)

Background:  Environmental and Occupational Surveillance Activities at NJDOH: A New Use of Real Time Emergency Department Registration Data   Richard E. Opiekun, PhD; Marija Borjan, PhD; Jerald A. Fagliano, PhD; New Jersey Department of Health     Real-time emergency department (ED) data are currently received from 78 of 80 acute care and satellite EDs in New Jersey by Health Monitoring Systems Inc.’s (HMS) EpiCenter system.  EpiCenter collects, manages & analyzes ED registration information for syndromic surveillance and provides alerts to state and local Health Departments on an increasing number of health-related syndromes.  The New Jersey Department of Health (NJDOH) began using EpiCenter in 2005, and initially monitored the overall volume of ED registrations within a subset of NJ EDs.  In 2011, NJDOH was funded to expand participation statewide.  As of July 2015, EpiCenter receives data for greater than 97% of all NJ ED facilities, and is utilized by multiple NJDOH programs for both communicable and non-communicable syndromic surveillance.  Variables used for surveillance include: date and time of ED registration; facility name; patient’s age, sex, residential zip code; and chief complaint.  EpiCenter uses text within the chief complaint field to classify visits and NJDOH staff work with HMS to determine appropriate thresholds for generation of alerts and web-based data displays.   Methods:    In 2012, the Environmental and Occupational Health Surveillance Program began working with NJDOH Communicable Disease Service partners to conduct real time environmental and occupational health surveillance via EpiCenter for ED visits related to a variety of causes:  chemical exposures; carbon monoxide (CO) poisonings; extreme weather-related injuries; heat and cold-related illnesses; and occupational injuries, illnesses, and poisonings.  DOH staff receive automatic real-time electronic notifications whenever there is a threshold number of patient ED registrations related to selected environmental or occupation exposures.  Staff quickly review the cases using the secure EpiCenter website, and when appropriate contact the ED’s Nurse Manager to gather more information.  If needed, follow up activities may involve staff from local/county health agencies, hazmat teams, or fire departments.  The presentation will include screen shots of sample EpiCenter surveillance methodology, including: e-mail alerts, facility screens, and temporal and spatial analyses of recent elevated occurrences of diverse environmental and occupational illnesses detected via EpiCenter.    Results:     During the past year, EpiCenter tracking of CO poisonings and heat-related ED visits has generated three NJDOH press releases, numerous NJDOH tweets, and timely enhanced public health agency and media attention to these issues.  Surveillance of occupational events has detected diverse exposures and illnesses including: an exposure to toluene vapor at a nail polish manufacturing plant; six cases of pesticide exposure at a produce repacking facility; and carbon monoxide poisoning of two police officers due to exhaust problems in their vehicle.      Conclusions:    NJDOH has found EpiCenter to provide a useful real-time snap shot of environmental and occupational disease in a variety of situations and for diverse disease syndromes.   While EpiCenter ED data has been found to have low sensitivity and specificity when compared with complete billing records for ED visits for heat-related illnesses, EpiCenter provides a timely tripwire for environmental and occupational syndromic surveillance.

 

Hadler, James (Hadler.pdf)

Reduction in HPV-associated High Grade Cervical Lesion Incidence in Connecticut 2008-2014: Evidence for Herd Immunity  Reduction in HPV-associated High Grade Cervical Lesion Incidence in Connecticut 2008-2014: Evidence for Herd Immunity  J Hadler, P Julian, K Higgins, J Meek, L Sosa, L Niccolai  BACKGROUND:  With vaccine licensing in 2006, human papillomavirus (HPV) related cervical disease caused by HPV types 16 and 18 became vaccine preventable. In 2008, the CDC Emerging Infections Program established a surveillance system to monitor the population impact of HPV vaccine on the incidence of cervical intraepithelial neoplasia (CIN) 2 and higher (HPV IMPACT).  Five EIP sites were funded to conduct pathology laboratory based surveillance for CIN2, CIN3 and adenocarcinoma in situ (collectively, CIN2+). In Connecticut (CT), CIN2+ became a reportable condition beginning January 2008.  METHODS:   We examined changes in CIN2+ incidence in CT from 2008 (baseline) to 2014 by age group (21-24, 25-29, 30-34, 35-39), age groups for whom routine screening continues to be recommended, and census tract-level poverty. We compared observed changes in 21-24 year olds with those expected based on the percentage of CIN2+ that were vaccine types 16/18 (54.3%), the potential for cross-protection among non-vaccine types (9.6% of all CIN2+), and estimated vaccination rates (66%).   RESULTS:  In 2008, there were 2164 cases of CIN2+, incidence 516.3 per 100,000 21-39 year old women. The decrease began in 2010 and has dropped steadily the past 3 years. From 2008 to 2014, overall incidence decreased by 21.0% (p<0.01). By age group, women 21-24 years had a decrease from 877.6/100,000 population to 376.2, a 57.1% decrease (p=0.00004). The observed decrease in 21-24 year old women is higher than the maximum expected decrease (42.2%), relative decrease 1.36 (95%CI 1.20-1.61). To achieve the same decrease as was observed, at least 89% of 21-24 year olds would have had to be vaccinated.   CONCLUSIONS:  There has been a progressive decrease in CIN2+ diagnoses in CT, mostly since 2010. The decrease in the 21-24 yo subgroup is greater than would be expected based on assumptions including high vaccine effectiveness, high vaccination rates, and cross protection against non-vaccine types. Thus, it appears that population vaccination rates are now high enough to reduce the probability of HPV exposure in the unvaccinated, leading to herd immunity. Surveillance data in CT demonstrate a substantial population-level impact of HPV vaccine, particularly in women in their early 20s.

 

 

Hamilton, Lindsay  Semple, Shereen (Hamilton_ID_1045.pdf)

Lassa Fever in New Jersey, 2015   Background:  On May 25, 2015, the Centers for Disease Control and Prevention (CDC) confirmed the presence of Lassa virus in a New Jersey (NJ) resident with recent travel to Liberia. The individual had been under active monitoring (AM) for Ebola virus disease (EVD) but had not reported symptoms to the local health department (LHD); suspicion for Lassa fever (LF) was not reported until later in the illness course.   LF is an acute viral illness caused by Lassa virus, an arenavirus endemic to West Africa, where persons primarily become infected by ingesting or inhaling excreta from multimammate rats. Symptoms present one to 21 days following exposure, and 1% of cases are fatal. Secondary cases may result from exposure to infected body fluids.    Methods:  Contact tracing was performed among community/family members and healthcare workers who were assigned to high or low risk classifications based on responses to a standardized questionnaire assessing contact with infected body fluids and use of personal protective equipment.  Contacts were monitored for 21 days using the AM protocol implemented for EVD monitoring.     Results:  A total of 217 potentially exposed contacts were identified, of which 214 (98.6%) were interviewed. Of those interviewed, 181 (83.4%) reported possible exposures; of those exposed, 15 (8.3%) were classified as high risk and 166 (91.7%) as low risk.  Daily symptom monitoring of contacts concluded on June 18, the end of the 21-day incubation period. Hospital A monitored 81 (44.8%) low risk contacts. Hospital B monitored 65 (35.9%) contacts, including one laboratory worker classified as high risk. LHDs monitored 35 (19.3%) contacts from the community, including 13 family members and a private physician classified as high risk. No secondary cases were identified.    Conclusion:  Although secondary transmission of LF is less likely to occur in the U.S., health care providers need to remain vigilant about Lassa and other viral hemorrhagic fevers, to ensure timely identification, treatment and containment of disease.  Also, stakeholders should address stigma that may be associated with travelers from EVD-affected countries, to minimize barriers to reporting symptoms and risk factors. Lastly, use of current EVD monitoring systems and protocols was valuable for rapid implementation of large-scale contact tracing.  Kim Cervantes  Nicole Mazur   Diana Theriault  Julia Wells Christina Tan

 

Hamilton, Lindsay Hamilton_ID_345 A.pdf

Surveying for Superbugs: A Survey on Carbapenem-Resistant Enterobacteriaceae (CRE) Rates and Laboratory Testing in New Jersey, 2014     Background: Multi-drug resistant bacteria are a growing concern in the United States and New Jersey (NJ).  Patients infected with these bacteria face a higher mortality rate, and these infections are significantly more challenging to clinically manage. Carbapenem-resistant Enterobacteriaeae (CRE), in particular, is estimated to have a mortality rate of 48-70% nationally and is detected in around 5% of acute care hospital facilities according to the Centers for Disease Control and Prevention (CDC).  Currently, the incidence of CRE infection in the state of NJ is unknown. Determining the approximate rate of CRE could potentially lead to an appropriate state response regarding the control and prevention of resistant and multi-drug resistant bacteria.    Methods: In order to quantify the burden of and comprehend the testing practices for CRE in NJ, the New Jersey Department of Health (NJDOH) developed a survey in Hippocrates, a NJ-specific application aimed at capturing, managing, displaying, and disseminating health information.  The survey was sent electronically to all acute care hospital laboratories in NJ in February 2015 using a distribution list from the NJ Public Health and Environmental Laboratories.  All survey responses were exported and summarized in Microsoft Excel 2010 and SAS 9.3.    Results: Lab supervisors from 56 of 72 (78%) acute care hospitals responded to the survey.  The majority of respondents (73%) reported that they performed susceptibility testing in onsite microbiology labs.  The estimated average incidence rate of CRE positive specimens in NJ was 1.64%.  Susceptibility cutoffs used by hospital labs varied greatly, and a few hospitals included screening specimens in addition to clinical isolates.  The top three CRE positive bacteria were Klebsiella spp., Enterobacter spp., and Escheria coli. Hospital labs were equally split between using culture-based testing methods, such as Modified Hodge Test, and non-DNA assays, such as Vitek 2.  A vast majority of hospital labs (89%) had protocols in place for informing the clinical team of CRE positive results.    Discussion: This survey was the first study in NJ to assess the rate of CRE and testing practices in NJ.  The results, however, provide only a rough estimate due to the fact that not all hospitals were able to successfully report their rates of CRE positive specimens in this survey and susceptibility cutoffs were not always comparable between hospitals which may have led to misclassification.  Since there may also have been volunteer bias involved with the NJ CRE rate estimate, further studies should be completed to get a more comprehensive and accurate picture of CRE rates in NJ.  Hospitals may now be better prepared to utilize appropriate antibiotics with positive CRE infections and NJDOH is more informed of the species to consider for future CRE reporting considerations.  Hospitals can also compare where they stand in their testing practices and capabilities when the results of the survey are disseminated back to the NJ hospitals, this will allow for a “standard of care” to emerge that encourages appropriate testing and notification of positive results to clinical staff.  NJDOH will next send out a second survey to infection preventionist (IP) staff of all acute care hospitals in order to better understand infection control practices that are being utilized upon notification of positive results to gain a more complete understanding of the hospital care of CRE infections and prevention of healthcare associated infections.

 

Jones, Lucretia (Jones_ID_1045.pdf)

Locating hard to reach travelers for Ebola Virus Disease active monitoring in New York City   Lan Li, MPH Carolina Pichardo, BA Brian Toro, BA  Paul McNamee, MPH  Background: In October 2014, the Centers for Disease Control and Prevention (CDC) began the Check and Report Ebola (CARE) program to improve both the screening of individuals traveling from countries affected by the Ebola Virus Disease (EVD) outbreak in West Africa and their post-arrival active monitoring for 21 days. In November 2014, CDC expanded their program to CARE Plus (+), which included giving each arriving traveler a pre-paid cell phone with at least 21 days of voice and text service to improve the chances of health departments successfully contacting travelers daily for active monitoring.  Methods: On October 25, 2014, the New York City Department of Health and Mental Hygiene (NYC DOHMH) began receiving files from CDC with contact information (phone, email, and address) for travelers who had arrived at JFK Airport since October 11, 2014 and began active monitoring of these travelers for EVD. DOHMH created an active monitoring call center (AMCC) and reassigned employees from their routine responsibilities to call travelers and process incoming calls. Each day, AMCC staff called travelers, solicited two temperatures and screened for signs and symptoms of EVD, and recorded responses in a database. When a traveler was unable to be reached by AMCC for two consecutive days, the traveler was referred to the DOHMH Surveillance and Epidemiology Emergency Response Group’s Field Surveillance Unit (FSU).  Activated under DOHMH’s Incident Command System, FSU consists of approximately 200 DOHMH staff from six different bureaus. FSU staff are trained in patient/provider outreach, interviewing, and medical chart abstraction. Upon being assigned a traveler to contact, FSU staff initially attempted to contact the traveler or his or her listed emergency contact by phone. FSU staff used subscription-based people search databases to search for travelers or their emergency contacts. After calling all identified phone numbers and leaving messages for the traveler to call AMCC, FSU staff would send an email to the traveler, if an email address was available. If there was no response by phone or email after 2-3 hours, a team of two FSU staff would conduct a site visit to the traveler’s home, hotel, or local address. If the traveler was not found, a letter from DOHMH explaining the need to speak with the traveler was placed under the door or left at the hotel’s front desk. If there was no response from the traveler by the next day, the traveler’s information was referred to the NYC Police Department for additional follow-up. We used c2 tests to compare the proportions of FSU referrals before and after phones were added to CARE kits that (1) required additional contact information and (2) required a field visit.   Results: DOHMH conducted active monitoring for 2941 travelers arriving from October 11, 2014 to May 31, 2015. Of these, FSU received a total of 235 (8%) referrals for travelers the AMCC was unable to reach after two days. These referrals represented 165 unique travelers; during their 21-day monitoring period, 118 travelers were referred once, 31 referred twice, 11 referred three times, three referred four times, and two referred five times.  Of the 235 referrals, 80 (34%) were successfully contacted by FSU by phone within a day of referral. Sixty-seven (29%) required a field visit, of which 58 (87%) were to valid addresses resulting in FSU staff speaking with the traveler in person on 17 (29%) of these 58 visits. FSU spoke with the traveler’s household member, neighbor, building manager, or hotel clerk in the remaining 41 (71%) of the 58 visits and left the DOHMH letter with someone or placed it under the door. Of the 235 total referrals, 78 (33%) had other outcomes (e.g. left NYC or contact made after 24 hours) and for 10 (4%) FSU was unable to proceed with any outreach due to lack of valid contact information.  In the first four weeks of active monitoring before phones were given to travelers, FSU had to identify additional contact information for 41 of 136 (30%) of referrals; after phones were distributed, only 14 of 99 (14%) required identifying additional information (c2 P=0.004). Before phones were distributed, 49 of 136 (36%) FSU referrals required in a field visit; after phones were distributed, 18 of 99 (18%) referrals required in a field visit (c2 P=0.003).  Conclusions: Active monitoring of travelers who are at low, but not zero risk for EVD for 21 days is a labor-intensive activity for a health department. Distributing pre-paid cell phones facilitated efforts to contact travelers who were difficult to reach. However, valuable time and resources were still expended on sending field staff to locate and make contact with travelers who did not respond to contact attempts. Having a unit of staff trained and experienced in locating contact information, finding hard-to-reach people in the field, and interviewing reluctant clients was critical.

 

Jordan, Heather (Jordan_CH_1045.pdf)

Findings of the New Jersey Amyotrophic Lateral Sclerosis Surveillance project, 2009-2011   Jerald Fagliano, PhD  Lindsay Rechtman, MPH, MCHES  Daniel Lefkowitz, PhD  Kevin Henry, PhD  Richard Opiekun, PhD  Wendy Kaye, PhD   Background: Amyotrophic lateral sclerosis (ALS), or Lou Gehrig’s disease, is a rare, incurable, progressive neurological disease. Five to 10% of ALS cases are familial, and causes of remaining cases are not well understood. Research suggests a complex etiology with genetic and environmental risk factors. To monitor incidence and prevalence of ALS in the United States (US), the federal Agency for Toxic Substances and Disease Registry (ATSDR) maintains the Congressionally-mandated National ALS Registry.     Because ALS is a non-reportable disease in the US, the Registry identifies ALS cases using national administrative datasets (Medicare, Medicaid, Veterans Administration) and patient self-enrollment through a web portal. This non-traditional case ascertainment requires validation against conventional surveillance data collection techniques; therefore, ATSDR established surveillance projects in three states (Florida, New Jersey, and Texas) and eight metropolitan areas (Atlanta, Baltimore, Chicago, Detroit, Las Vegas, Los Angeles, Philadelphia, and San Francisco). This presentation describes the findings from the New Jersey project.    Methods: The New Jersey (NJ) Department of Health conducted a state-wide surveillance project to identify ALS cases. Neurologists in the region submitted case reports for patients who were residents of NJ, under their care between January 1, 2009 and December 31, 2011, and met the El Escorial criteria. Diagnosis was verified on a subset of reported cases through expert review of medical records. Mortality records were queried for the period 2009-2013 using the ICD-10 G12.2 code for motor neuron disease and key term literals. Query results were used for supplemental case identification and to determine survival of incident cases for the period 2009-2011. State-wide age-adjusted prevalence and incidence rates were calculated. Geographic clustering of ALS incidence was examined using a spatial scan statistic at the census tract level. The relationship between socioeconomic status (SES), measured as census tract median income, and ALS incidence was assessed using a multivariate Poisson regression model. Survival estimates by age, sex, race, and ethnicity were calculated using the Kaplan-Meier estimator for incident cases diagnosed between 2009 and 2011 and followed until death or December 31, 2013.    Results: Twenty-five percent (168/679) of neurologists diagnosed and/or cared for ALS patients and 90% (152/168) reported cases. A total of 965 case reports were collected and 21% (199) were cases reported more than one time resulting in 764 unique cases. Point prevalence was 4.4 cases per 100,000 persons as of December 31, 2011. Of 764 reported cases, 493 were incident for the period January 1, 2009 through December 31, 2011. The average annual age-adjusted ALS incidence rate was 1.7 cases per 100,000 person-years. ALS was more common in men than women, and incidence increased with age. ALS rates were higher in whites and non-Hispanics than other racial or ethnic groups. Age-adjusted ALS incidence rates increased by census tract average income. After adjusting for age, sex, and race, ALS incidence was significantly higher in the highest versus lowest income quartile. No statistically significant geographic clusters of ALS incidence were found. Median survival from diagnosis to death for incident cases was 21 months, and increased age was a predictor of shorter survival.     Conclusions: The project expands our understanding of ALS occurrence, and provides the first state-wide ALS incidence, prevalence, and survival rates by demographic groups. Age-adjusted point prevalence, annual average age-adjusted incidence rates, and survival are consistent with previously published literature. While no geographic clusters were identified, geographic variation of ALS incidence in NJ appears to be influenced by SES and race. More research is needed to explain how demographic and socioeconomic factors affect ALS incidence. Incidence and prevalence data were synthesized to construct an ALS Indicator, pilot-tested, and uploaded for display on the NJ Environmental Public Health Tracking (EPHT) health data portal. The project illustrates how Tracking partnerships can enhance chronic disease surveillance and help make chronic disease data available to the general public, health planners, and advocacy organizations.     Study Support: This study was supported by McKing Consulting Corporation through a contract funded by the Agency for Toxic Substances and Disease Registry (Contract #GS00F0042P).

 

Jordan, Heather   Marshall, Elizabeth (Jordan_EO_145.pdf)

Occupational Health Effects of Hurricane Sandy    Margaret Lumia, PhD  Shou-En Lou, PhD Marija Borjan, PhD Joel Swerdel, PhD  Daniel Lefkowitz, PhD  Pinar Erdogdu, MPH  Zhengyang Shi, MPH    Background: Work-related injuries and illnesses as a result of natural disasters are a serious public health concern. As a direct result of the effects of Sandy, seven work-related fatalities occurred in NJ, many of whom were working in a response capacity. Of the seven work-related fatalities identified as directly related to the storm, three were tree care workers.     This project aimed to 1) summarize work-related injuries and acute illnesses in NJ after Sandy through retrospective analyses of statewide data sources, identify gaps in existing data sources, and provide recommendations for strategies for future occupational health surveillance; and 2) to conduct focus groups among three first responder worker populations, Emergency Medical Services (EMS) responders, tree care company employers and employees, and 3) disaster volunteers, to understand, reduce, or eliminate adverse health impacts by providing recommendations for educational and outreach materials.    Methods: Emergency discharge (ED) and hospital discharge (HD) data among those 18-65 years old with a principal discharge diagnosis of unintentional injury, including death certificates, were obtained from the NJ Department of Health. Discharges were grouped by location of hospital by county and categorized as having high, medium, or low impact from Sandy based on data from FEMA and other sources. Discharges were considered work-related if the primary payer was Workers Compensation or if they included codes suggesting a work location. Poisson regression was utilized to compare the risk of work-related injury during different periods over the year following Sandy landfall (10/29/12-10/29/13) with corresponding periods from the three previous years (2009-2011). Subgroup analyses evaluated risks among those at generally highest risk (men in high impact counties), by diagnosis category (ICD-9), and by mechanism of injury (E-Code).    Nine focus groups (n=90), three per worker group, were conducted using structured focus group guides to ensure uniform coverage of topic areas. Participants were asked to describe similarities and differences between “regular” job/volunteer duties and compare them to Sandy-related job/volunteer duties, to describe exposures to contaminants, to describe the use of PPEs and other safety equipment, to characterize the physical and mental health outcomes among the target population, and to pilot test questions to be included in a longer electronic survey to be deployed at a later time.     Results: There were 8,511 work-related injuries in the first quarter post-Sandy in NJ, 8.8% of all injuries among those 18-65. Compared to previous years, the week immediately following Sandy (when much of the state had limited electrical power) showed a significant decline in total work-related injuries (RR=0.85 (95%CI: 0.78, 0.92)) and statewide there was no overall increase in any quarter post-Sandy. However, high impact counties showed an elevated risk of work-related injuries in the first 9 months post-Sandy, especially among men in the third quarter after Sandy (May-July) Quarter 3: RR=1.10 (1.01, 1.20 The greatest excesses occurred in the in falls; cut/pierce; struck by/against; and overexertion injuries. Black and Hispanic workers also showed an elevated rate of work-related injuries compared to white workers.    Preliminary focus group results suggest that although these worker groups need to be ready to “face the unexpected” in their day-to-day jobs, during the storm they faced additional hazards, such as contaminated floodwaters, downed power lines, storm damaged trees, and extra-long shifts. Respondents reported that training, communication, personal protection equipment, and team work were important factors in keeping workers safe during storm response and recovery.    Conclusions: The rate of work-related injuries in Sandy’s high impact area increased after the hurricane. Based on timing and type of injury, the greatest impact in work-related injury may be associated with rebuilding and recovery rather than initial response. sing workers compensation as payer will injuries will under-estimate work-related injuries after a natural disaster but did show some increases.  Information from focus groups confirmed the presence of hazardous conditions that might lead to those injuries and suggested prevention strategies.

 

Kaplan-Dobbs, Marissa

Public Health Detailing: Promoting Judicious Opioid Prescribing in Bronx County, New York  AUTHORS: Marissa Kaplan-Dobbs, Ellenie Tuazon, Denise Paone, Hillary Kunins, Jessica Kattan  BACKGROUND: Opioid analgesic (OA) overdose is a public health crisis in New York City (NYC). In 2013, Bronx County had the second highest rate of OA-involved overdose deaths among NYC’s five counties (3.4 deaths per 100,000 residents). In 2013, Bronx residents also had the second highest rates of OA prescriptions and of high-dose prescriptions filled (OA prescriptions with dosages >100 morphine milligram equivalents [MME]). In response, the NYC Department of Health and Mental Hygiene (DOHMH) conducted a targeted public health detailing campaign about judicious OA prescribing in the Bronx, modeled after a similar, successful campaign in Staten Island, the NYC county with the highest rate of OA-involved overdose deaths. To evaluate knowledge change, a brief survey was administered to detailed health care providers (HCPs).   METHODS: The Bronx detailing campaign was conducted during May–June, 2015 targeting Bronx HCPs including all physicians and physician assistants practicing primary care, infectious disease, surgery, and pain medicine. The campaign consisted of brief (~15 minute) visits with HCPs during which DOHMH representatives discussed DOHMH prescribing recommendations and provided supporting resources. The prescribing recommendations were: a 3-day supply of OAs is usually sufficient for acute pain; avoid prescribing OAs to patients taking benzodiazepines; and avoid prescribing high-dose OA prescriptions. Representatives attempted an initial and follow-up visit for each HCP, and administered a three-question survey about the recommendations at the visits. The proportion of answers consistent with recommendations was compared at baseline and follow-up.   RESULTS: A total of 972 HCPs had initial visits, of whom 814 (84%) had follow-up visits. At baseline, 50% (484/972) of HCPs responded that a 3-day supply is usually sufficient for acute pain, compared to 75% (611/814) on follow-up;  at baseline, 94% (910/972) indicated being somewhat or very concerned about prescribing OAs to someone already taking benzodiazepines, compared to 97% (789/814) on follow-up; at baseline, 9% (90/972) correctly determined DOHMH’s high dose threshold, compared to 62% (503/814) on follow-up.  CONCLUSIONS: The detailing campaign successfully increased HCP knowledge about OA prescribing recommendations. Further analysis will be performed to assess campaign’s impact on mortality and prescribing patterns. Public health detailing should be considered by other jurisdictions as part of a comprehensive strategy to promote judicious OA prescribing.

 

 

Kogut, Sarah (Kogut_LB.pdf)

New York State Counties with High Potentially-Avoidable Antibiotic Prescribing for Upper Respiratory Infections: Identification and Intervention   Introduction:  The Centers for Disease Control and Prevention (CDC) reports that each year in the United States at least 2 million people are infected with antibiotic-resistant bacteria leading to at least 23,000 deaths. Inappropriate use of antibiotics over the years has rendered some antibiotics ineffective against certain bacteria. The outpatient setting is a key focus for educational efforts on appropriate use of antibiotics. Studies have shown high rates of inappropriate antibiotic prescribing when patients are given a diagnosis of ‘cold’, ‘acute upper respiratory infection’ or ‘acute bronchitis’.    Evidence is limited on how to affect antibiotic prescribing in the outpatient setting. Some studies support healthcare provider (HCP) audit and feedback as an effective mechanism to elicit change in practice behaviors. This CDC funded project by the New York State Department of Health aims to affect outpatient antibiotic prescribing practices utilizing a form of state-level audit and feedback by providing HCPs with county level data on antibiotic prescribing practices for a specific clinical scenario, along with promoting CDC Get Smart: Know When Antibiotics Work educational materials.     Methods:  New York State Department of Health (NYSDOH) analyzed New York State Medicaid recipient administrative claims and prescription data to identify outpatient antibiotic prescribing trends from January 1, 2013 to December 31, 2013. Analysis included persons aged 3 months to 64 years of age who received a primary diagnosis of acute upper respiratory infection (URI) by ICD-9 code in an outpatient setting and subsequently filled a prescription for an antibiotic within four days. Initial analysis indicated higher rates of potentially avoidable prescriptions among adults than children. Indirect standardization was used to risk adjust based on age, principal diagnosis and visit type (emergency department (ED), institutional outpatient, or professional outpatient). County level rates were established based on the zip codes of provider practices. Those with the highest rates were deemed as having the highest potentially avoidable antibiotic prescribing, and providers practicing in these counties were targeted for intervention. A “Dear Provider” letter signed by the NYSDOH Commissioner of Health and a map identifying county-level rate categories were sent to primary care, family practice, internal medicine and emergency medicine HCPs who would be expected to see adult patients in outpatient settings. A second mailing was sent within a week and included Get Smart: Know When Antibiotics Work educational materials, including a viral prescription pad.     Results:  Among 150,379 eligible visits by NYS Medicaid enrollees aged 18 to 64 who were diagnosed with an acute URI in an outpatient setting, 25 to 64% of patients filled an antibiotic prescription within four days of their visit. Among the 62 counties in NYS, eleven counties were identified as having the highest rates (55 to 64%) of antibiotic prescriptions filled. The “Dear Provider” letter and accompanying map were sent to over 2,800 providers in these eleven counties. In addition to highlighting the issue of high rates of potentially avoidable antibiotic prescribing for acute URI, providers were enlisted to become provider “champions” or “standard-bearers” for antibiotic resistance education among their local peers.     Conclusions:  This intervention is in progress, however initial interest in the program has been high. Ongoing evaluation of the impact of this intervention will continue in the setting of additional collaborations and several local HCPs who have expressed interest in the program. Outreach to local health departments to notify them of the intervention has generated multiple opportunities for partnership. External partner opportunities have arisen with a private insurance company. These data will also serve as a baseline for comparison in the future as this public health intervention continues.    Authors:  Sarah Kogut, MPH, CIC  Mary Beth Wenger, ABJ  Emily Lutterloh, MD, MPH   Valerie Haley, PhD  Stephen Goins, MS  Tatiana Ledneva, MS  Mary Beth Conroy, MPH Elizabeth Dufort, MD, FAAP"

 

 

Kline, Kelly

Pregnancy as the main driver of sex disparities among adult influenza hospitalizations across 14 FluSurv-NET Sites, 2010-2012   Authors: Kelly Kline, MPH1, James L. Hadler, MD, MPH1, Linda Niccolai, PhD1, Kimberly Yousey-Hindes, MPH1, Pam Daily Kirley, MPH2, Lisa Miller, MD, MSPH3, Evan J. Anderson, MD4,5, Kemi Oni, MPH6, Maya L. Monroe, MPH7, Susan R. Bohm, MS8, Ruth Lynfield, MD9, Marisa Bargsten, MPH10, Shelley M. Zansky, PhD11, Krista Lung, BS12, Ann Thomas, MD, MPH13, Mary Lou Lindegren, MD, MPH14, Jessica Cohen, MPH15, 16    Author affiliations: 1Connecticut Emerging Infections Program, Yale School of Public Health, New Haven, CT, USA; 2California Emerging Infections Program, Oakland, CA, USA; 3Colorado Department of Public Health and Environment, Denver, CO, USA; 4Emory University School of Medicine, Atlanta, GA, USA, 5Atlanta Veterans Affairs Medical Center, Atlanta, GA, USA; 6Iowa Department of Public Health, Des Moines, IA, USA; 7Maryland Department of Health and Mental Hygiene, Baltimore, MD, USA; 8Michigan Department of Health and Human Services, Lansing, MI, USA; 9Minnesota Department of Health, St. Paul, MN, USA; 10New Mexico Department of Health, Santa Fe, NM, USA; 11New York State Department of Health, Albany, NY, USA; 12Ohio Department of Health, Columbus, OH, USA; 13Oregon Public Health Division, Portland, OR, USA; 14Vanderbilt University School of Medicine, Nashville, TN, USA,15Influenza Division, National Center for Immunization and Respiratory Diseases, CDC, Atlanta, GA, USA, 16Atlanta Research and Education Foundation, Atlanta, GA, USA    Background: Previous studies from the Emerging Infections Program found that adult females were more likely to have influenza-related hospitalizations than males. To identify groups of women at higher risk than men, we used data from 14 FluSurv-NET (FSN) sites that conduct active population-based surveillance for laboratory-confirmed influenza-related hospitalizations among residents of 76 counties.    Methods: We used 6,292 laboratory-confirmed adult (>18 years) cases collected by the 14 FSN sites during the 2010-11 and 2011-12 influenza seasons, 2010 US Census (denominators), and 2008-2012 American Community Survey (percentage in a census tract below the federal poverty level) to calculate overall age-adjusted and age group-specific female:male incidence rate ratios (IRR) by race/ethnicity and socioeconomic status (SES). The percentage of US women 18-49 years with live births in 2010 applied to site denominators was used to calculate the number of pregnant women.    Results: Overall, 55% of cases were female. Female IRRs were highest for 18-49 year-old women of low SES (IRR 1.50, 95% CI 1.30-1.74) and of Hispanic ethnicity (IRR 1.70, 95% CI 1.34-2.17). This difference disappeared after adjusting for pregnancy. Overall, 26% of 1,083 hospitalized 18-49 year olds were pregnant. Pregnant 18-49 year olds were more likely to have influenza-related hospitalizations than their non-pregnant counterparts (relative risk [RR] 5.23, 95% CI 4.57-5.99) but were no more likely to have been vaccinated (25.5% vs 27.8%). While the RR of influenza-related hospitalization of pregnant women was similar across the 14 FSN sites, the vaccination rates of pregnant cases varied widely (range: 9.1–57.1%, p<0.01).     Conclusions: Pregnant 18-49 year-old women were at much higher risk of influenza-related hospitalization than other 18-49 year olds, with a similar level of risk across population-based surveillance sites in 14 states. Vaccination rates of pregnant women were low but highly variable between sites. Improving influenza vaccination rates in pregnant women should be emphasized given the higher risk and substantial burden of disease in this population.

 

Kulkarni, Prathit

Evacuations as a Result of Superstorm Sandy — An Analysis of the 2014 New Jersey Behavioral Risk Factor Survey

Prathit A. Kulkarni, Hui Gu, Stella Tsai, Marian Passannante, Soyeon Kim, Pauline A. Thomas, Christina G. Tan, Amy Davidow

Background: Superstorm Sandy made landfall in New Jersey on October 29, 2012. It was the second-costliest hurricane in U.S. history. The purpose of this analysis was to estimate the extent of Superstorm Sandy-related evacuations in New Jersey and determine factors that were associated with the decision to evacuate.   Methods: We analyzed data from the 2014 New Jersey Behavioral Risk Factor Survey (NJBRFS). NJBRFS is part of the larger, nationwide Behavioral Risk Factor Surveillance System (BRFSS) coordinated by the Centers for Disease Control and Prevention. For the 2014 NJBRFS, questions specific to residents’ experience with Superstorm Sandy were added. Data were analyzed by using SAS® 9.3 (SAS Institute, Incorporated, Cary, North Carolina) survey procedures. Analysis of BRFSS data requires complex weighting methodology to account for its sampling design; weighting as recommended by BRFSS was used. We estimated the proportion of survey respondents who evacuated at any time because of Hurricane Sandy and the proportion of evacuees who evacuated before, during, or after Superstorm Sandy. We similarly evaluated the duration evacuees remained away from their home before returning. In univariate analysis, factors examined for association with the decision to evacuate were as follows: age, sex, race, education, marital status, having children, angina or coronary heart disease, history of myocardial infarction, history of stroke, asthma, depression, kidney disease, diabetes, presence of ≥1 of the previously listed chronic medical conditions, renting versus owning a home, length of time residing in the home, and living in a municipality that was more heavily affected by the storm. A classification system developed through previous work estimated the socioeconomic impact of Superstorm Sandy across different New Jersey municipalities by using multiple severity indicators, such as residential home damage and gasoline shortage. Univariate analysis was performed for all survey respondents and among those residing in municipalities that sustained the most socioeconomic impact from the storm. Multivariable regression analysis is in progress.  Results: A total of 13,025 respondents participated in the 2014 NJBRFS. An estimated 12.7% (95% confidence interval [CI] 11.8%–13.6%) of New Jersey residents evacuated their home because of Superstorm Sandy. Extrapolating to the estimated 2014 New Jersey adult population (6,926,000), the number of persons who evacuated was estimated to be 900,000. Among persons who evacuated, 28.5% (95% CI 25.1%–31.9%, estimated 261,000 persons) evacuated before the storm, 25.4% (95% CI 22.1%–28.7%, estimated 225,000 persons) evacuated during the storm, and 44.5% (95% CI 40.7%–48.3%, estimated 396,000 persons) evacuated after the storm. Persons who evacuated their home were away from their home for the following periods: 7.3% (95% CI 5.0%–9.6%) for <1 day; 62.7% (95% CI 59.0%–66.4%) from 1 day to 1 week; 19.4% (95% CI 16.5%–22.3%) from >1 week to <1 month; and 7.1% (95% CI 5.2%–9.0%) for >1 month. Factors significantly associated with the decision to evacuate among the entire population in univariate analysis were as follows: female sex (odds ratio [OR] 1.37; 95% CI 1.16–1.63), married versus unmarried persons (OR 0.79; 95% CI 0.67–0.93), renting compared with owning a home (OR 1.33; 95% CI 1.11–1.58), and living in a municipality more heavily affected by Superstorm Sandy (OR 1.82; 95% CI 1.53–2.16). Among persons living in municipalities more heavily affected by the storm, females (OR 1.49; 95% CI 1.16–1.92) were significantly associated with the decision to evacuate.

Conclusions: A substantial number of New Jersey residents evacuated their home as a result of Superstorm Sandy. Among evacuees, the largest group of persons evacuated after the storm had made landfall, and the majority remained away from their home from 1 day to 1 week before returning. Females were more likely to have evacuated their homes among all respondents and among those residing in municipalities that were most heavily affected by the storm. Conversely, persons who were unmarried or who rented their home were more likely to evacuate overall, but not among those living in the most heavily affected municipalities, perhaps indicating that the prospect of greater damage influenced their decision making. These results could potentially be used to direct future preparedness efforts for other natural disasters, possibly including educational campaigns and targeted messaging. Additional study is needed to more fully understand barriers to evacuation in preparation for future disasters.

  

 

Kulkarni, Prathit

Description of Persons Under Investigation for Ebola Virus Disease — New Jersey, August 2014–May 2015   Prathit A. Kulkarni, MD, Edward I. Lifshitz, MD, Barbara Carothers, LPN, Rosemary Kidder, RN, BSN, MPH, Shereen Semple, MPH, Christina G. Tan, MD, MPH   Background: Since the escalation of the outbreak of Ebola virus disease (Ebola) in August 2014, the New Jersey Department of Health (NJDOH) has provided guidance to clinicians on the management of persons under investigation (PUIs) for Ebola. On October 27, 2014, the Centers for Disease Control and Prevention instituted a policy of active monitoring (AM), which involves daily temperature and symptom reporting, for all persons returning to the U.S. from Liberia, Sierra Leone, and Guinea. The purpose of this analysis was to describe the clinical management and outcomes of PUIs in New Jersey during August 1, 2014–May 15, 2015 and to determine what proportion of those under AM became PUIs.

Methods: A PUI was defined as a person who had signs or symptoms consistent with Ebola, including temperature ≥100.4 or subjective fever, headache, fatigue, myalgias, abdominal pain, vomiting, diarrhea, or unexplained hemorrhage, and epidemiologic risk for Ebola, including travel to a country with widespread transmission, within 21 days before symptom onset. Clinical information was obtained from public health partners and treating physicians. Because of the known possibility of falsely negative results early in the disease, the decision to perform specific Ebola testing was based on the degree of exposure and clinical assessment. NJDOH also maintains a record of all persons who undergo AM in the state.

Results: During August 1, 2014–October 26, 2014, 12 PUIs underwent evaluation. After the initiation of AM, during October 27, 2014–May 15, 2015, 11 additional PUIs were evaluated, representing 1.3% of all 862 persons who underwent AM. The most common diagnosis was malaria, seen in six (26%) persons. Other common diagnoses included gastroenteritis and influenza. No diagnosis was made for eight (35%) PUIs, in whom Ebola was excluded based on rapid symptomatic improvement. Ebola testing was performed for one PUI with higher exposure risk, and this was negative.

Conclusions: Immediate Ebola testing was not needed for PUIs with low exposure risk. Identifying other diagnoses and carefully following the clinical course of PUIs were effective management strategies. Also, only a small percentage of those under AM became PUIs, indicating that most returning travelers remained healthy.

 

Kulkarni, Prathit

Methyl Bromide Release at a Condominium Resort — U.S. Virgin Islands, March 2015

Mary Anne Duncan, Michelle T. Watters, Leah T. Graziano, Elena Vaouli, Larry F. Cseh, John F. Risher, Maureen F. Orr, Tai C. Hunte-Ceasar, Esther M. Ellis           Background: On March 22, 2015, the Agency for Toxic Substances and Disease Registry (ATSDR) was notified by the U.S. Environmental Protection Agency (EPA) of a four-case cluster of suspected acute methyl bromide toxicity in a family vacationing at a condominium resort in the U.S. Virgin Islands (USVI). The family was transported to a hospital in USVI with progressive neurological symptoms that had developed during the preceding 24 hours, including weakness, myoclonus, fasciculations, altered sensorium, and word-finding difficulty. A preliminary investigation by the U.S. Virgin Islands Department of Health revealed that the pesticide methyl bromide, whose toxic effects matched the family’s symptoms, had been used two days before symptom onset to fumigate the building in which the family had been residing. The use of methyl bromide in homes and other residential settings is banned in the United States. An investigation conducted by the U.S. Virgin Islands Department of Planning and Natural Resources and EPA revealed that methyl bromide had also been used at the same condominium resort by the same pest control company during October 2014. An ATSDR Assessment of Chemical Exposures (ACE) team conducted an investigation to characterize the extent of exposures to methyl bromide at the resort, post-exposure adverse health effects, and the public health impact of inappropriate use of this chemical.   Methods: A contamination zone as a result of each fumigation was determined by examining the condominium resort’s physical layout. The resort comprised 11 separate residential buildings; therefore, no risk of methyl bromide affecting persons in a building other than those buildings that had been fumigated was present. On the basis of methyl bromide air concentration determined by initial EPA sampling and the expected decrease in air concentration of methyl bromide based on previous work, a 2-week timeframe after each fumigation date was used for the exposure period. Therefore, anyone who had resided in or had entered a fumigated building within two weeks of either fumigation date was considered to have been potentially exposed to methyl bromide. A standardized ACE health survey questionnaire was administered to potentially exposed persons to determine presence and severity of adverse health effects.

Results: In addition to the ill family, 37 persons who had potentially been exposed to methyl bromide at the condominium resort were identified; 20 (54%) persons were exposed from the October 2014 fumigation, 11 (30%) from the March 2015 fumigation, and one (3%) at both times. Exposure time was unknown for 5 (14%) persons. These 37 persons comprised 20 vacationers or visitors to the resort, eight resort staff, six emergency medical personnel, two police officers, and one pest control company applicator. Among these 37 persons, contact information was available for 20 persons. Sixteen of 20 (80%) persons were administered the standardized health survey either in person or by telephone. Six survey respondents reported post-exposure symptoms; the most common symptoms were headache, reported by 6/16 (38%) persons, and fatigue, which occurred in 4/16 (25%) respondents. Four of six exposed emergency medical personnel developed post-exposure symptoms; some emergency medical personnel noted that prompt notification about their exposure had not occurred.

Conclusions: Inappropriate use of methyl bromide in a residential setting resulted in certain unsuspecting persons being exposed to this toxic chemical. The most common post-exposure symptoms were headache and fatigue, and emergency medical personnel most commonly experienced adverse health effects from exposure. Pest control companies and others that use toxic compounds should be aware of and follow all regulations regarding restricted-use chemicals. In addition, consideration should be given to use of integrated pest management, which emphasizes an environmentally sensitive approach to pest control and minimizes use of nonspecific pesticides. Finally, a system for prompt notification of first responders who might have experienced a toxic exposure is recommended.

 

Landman, Keren (Landman_LB.pdf)

Fitzhenry, Robert   Outbreak of Community Acquired

Legionnaires¹ Disease in New York City, New York -- July-August 2015

Background: The New York City (NYC) Health Code requires providers and

laboratories to report positive laboratory tests for Legionella infection

to the NYC Department of Health and Mental Hygiene (DOHMH). DOHMH

investigates all urine antigen tests, sputum cultures, and reports of

four-fold or greater rise in acute and convalescent antibody titers.  To

detect potential concerning clusters of legionellosis, the DOHMH  conducts

an automated spatial-temporal analysis (SaTScan) to look for clustering by

neighborhood, and reviews building coordinates of cases to identify

clusters that occur in a high risk setting (e.g. adult care facility) or

in the same residential building.    Legionella infection rates are

historically higher in the Bronx than in the other four NYC counties.   On

July 17, 2015, SaTScan results alerted epidemiologists to an increase in

laboratory reports of infection with Legionella pneumophila serogroup

1(Lp1) in patients residing in seven zip codes of the South Bronx.

Methods: DOHMH initiated an investigation, prioritizing these reports for

rapid chart reviews to confirm case status and patient interviews to

identify common exposures.  Interviews of patients or their family members

did not identify a common exposure among most cases other than living or

spending time in the affected area of the South Bronx, and the geographic

distribution of cases was consistent with community-wide exposure to a

contaminated aerosol in the outdoor environment. Cooling towers in the

outbreak zone were identified, and water samples from cooling towers were

tested for the presence of Legionella by real-time PCR and culture.

Cooling towers were remediated immediately if water samples were positive

for Lp1 DNA by real-time PCR. A health alert was sent to NYC  health care

providers, encouraging them to test for Legionella infection in residents

of seven contiguous South Bronx zip codes (the ³outbreak zone²) presenting

with respiratory symptoms and to specifically obtain respiratory specimen

cultures for patients suspected of having Legionnaires¹ disease. Autopsy

samples from patients suspected to have died from Legionnaire¹s disease

were tested by both PCR and culture. Outreach teams visited

community-based organizations serving vulnerable populations in the

outbreak zone to alert them to the symptoms of Legionnaires¹ disease and

to encourage early evaluation of symptoms for individuals with risk

factors associated with Legionnaires¹ disease.

Results: As of August 21, 2015, 130 laboratory-confirmed cases of

Legionnaires¹ disease were reported among people with illness onset July

8-August 3 and a history of living in, working in, or visiting the

outbreak zone. The median age of case-patients was 55 years (range 29-90),

and 78 (60%) were male. Twenty-two (17%) case-patients were HIV-positive,

and 14 (11%) died.  Thirty-seven cooling towers in the affected area were

identified and tested, and 14 cooling towers positive for Lp1 DNA were

immediately remediated, including cooling tower A.  Legionella pneumophila

serogroup 1 (Lp1) isolates from 17 case-patients who live in the affected

area and 4 from case-patients who worked in or visited this area were

compared to environmental Lp1 isolates cultured from cooling towers in the

South Bronx using three different sub-typing methods: pulsed-field gel

electrophoresis (PFGE), sequence-based typing (SBT), and whole genome

sequencing (WGS). The Lp1 isolates from outbreak-associated cases were

indistinguishable from the Lp1 isolates cultured from the Opera House

Hotel cooling tower when they were compared using the three methods. In

contrast, Lp1 isolates cultured from all other cooling towers in the

affected area of the South Bronx did not match the case-patients¹

isolates.    On August 6, DOHMH ordered immediate inspection and

disinfection of all cooling towers in the city, and on August 13, the New

York City Council passed a bill mandating registration and annual testing,

disinfection, and maintenance of all cooling towers in NYC.  Conclusion:

We report the largest known outbreak of Legionnaires¹ disease in NYC, and

the largest known outbreak of community-acquired Legionnaires¹ disease in

the United States since the discovery of Legionella in 1977. All

outbreak-associated patient isolates of Lp1 were indistinguishable from

isolates obtained from cooling tower A, located in the center of the

outbreak zone. Screening of water sampled from cooling towers for the

presence of Lp1 DNA allowed public health resources to be directed on

those most likely to be involved in the outbreak, and led to the order to

remediate potentially contaminated cooling towers quickly while awaiting

final culture results. Although guidelines exist for the testing and

maintenance of cooling towers to mitigate Legionella risk, no large

municipalities or states in the United States mandated cooling tower

testing and maintenance by law prior to this outbreak.

 

Langenfeld, Maxine (Langenfeld_ID_145.pdf)

Determinants of seasonal flu vaccine hesitancy among university students in the US  Maxine Langenfeld MPH(c) (1); Elizabeth Marshall, PhD, MSPH (1); Sinae Kim, PhD (1) and Tefera Gezmu, PhD, MPH (2)  1. RBHS-School of Public Health; Rutgers, The State University of New Jersey;  2.Edward J. Bloustein School of Planning and Public Policy, Rutgers, The State University of New Jersey    Objective: Seasonal influenza vaccination has been shown to be effective among healthy young adults and provides protection to 70-90% of those who received it. In 2010 the Center for Disease Control’s Advisory Committee for Immunization Practices (CDC ACIP) expanded its recommendation for annual influenza vaccination, for the first time including all students regardless of residency status. Flu vaccination reduces healthcare provider visits, antibiotic use, the number of missed school/ workdays and poor school performance among college students. Despite overall increases in seasonal flu vaccine coverage over the past few years, research indicates that uptake among students is 30% at best. Such low uptake is unlikely due to lack of access to care, as full-time college students are typically required to carry health insurance in order to register for classes and most colleges offer free or reduced cost vaccination clinics through their health centers. Generally, vaccine hesitancy exists when acceptance of a vaccine is lower than expected given the availability of services; hesitancy is not limited to vaccine refusal, but includes delays in vaccination and low confidence in the safety or effectiveness of a vaccine. This study examined determinants of seasonal flu vaccine hesitancy among university students and assessed their overall attitudes towards the vaccine.      Methods: During the Spring 2015 semester we conducted a cross-sectional study of a convenience sample of students at a large public university using a computer-assisted self-administered questionnaire. Items for the questionnaire were developed based on the World Health Organization’s (WHO) model of vaccine hesitancy that includes measures of confidence, convenience and complacency. Vaccine uptake rates, related demographics and attitudes were summarized. Unadjusted and adjusted logistic regression modeling was used to explore demographic determinants of hesitancy. Associations between attitudes and vaccination status were tested using Chi-square tests.    Early Results: A total of 640 undergraduate (54%) and graduate students (45%) participated in the study. A majority of the respondents were female (67%), with an overall median age of 22 years. Half of the respondents reported receiving the flu vaccine for the 2014-2015 season. A majority (79%) of the respondents were vaccinated off-campus.  Adjusted logistic regression analysis indicated that health sciences students were twice as likely to have been vaccinated as students in other areas of study (aOR 2.04, 95% CI 1.3-3.1). Graduate students were 150% more likely to receive vaccination than undergraduate students (aOR 2.50, 95% CI 1.6-4.0).  Age, parental education and residency (on-campus vs. off-campus) were predictors of vaccine uptake only in the unadjusted models; gender, race and high-risk status did not show any association with vaccine uptake. Perceived risk and severity of the flu as well as perceived safety and effectiveness of vaccination were bivariately correlated with current vaccination status. Health sciences students held overall positive attitudes towards the vaccine more frequently than non-health sciences students.    Conclusion: Effective health promotion programs are needed to reduce flu vaccine hesitancy among college students. Health care workers and educators should engage students directly about the importance of the vaccine but also indirectly through family and friend networks to recommend the flu vaccine to university students. Future research exploring additional measures of hesitancy is needed to determine whether convenience is a significant predictor of vaccination status among university students."

Laramie, Angela

Troppy, T. Scott  Development of a standardized occupation reference code list for MAVEN: Collaboration between Infectious Disease and Occupational Health    T. Scott Troppy, MPH, PMP Susan Soliva, MPH Kathleen Fitzsimmons, MPHBackground:  The Massachusetts Virtual Epidemiologic Network (MAVEN) was deployed in 2006 by the Massachusetts Department of Public Health (MDPH) to serve as an integrated, Web-based disease infectious disease surveillance and case management system. It captures data used for case investigation and follow-up by MDPH epidemiologists and local boards of health (LBOH). The system captures information on approximately 90 reportable infectious diseases, and produces close to 150,000 disease reports annually.  Since its inception, MAVEN has included an occupation reference code list used by state epidemiologists and LBOH for demographic and risk purposes, but there was concern that the list was not comprehensive enough and did not correspond to standard coding systems, such as the Bureau of Labor Statistics Standard Occupational Classification (SOC).  Use of SOC would enable comparison of occupations within MAVEN and across data sets. The goal of this effort was to revise and expand the occupation pick list in MAVEN using occupation categories based on the SOC.  We aimed to keep the list to a length and format that would be useful and not overwhelming to those collecting the data.    Methods and Findings:  First, using data on workforce distribution from the Current Population Survey, we identified the most frequent occupations and industries among Massachusetts residents. We then used the existing MAVEN reference code list and cross-walked it to the SOC.  The final list contained more health care, food service and animal related occupations than the previous list in MAVEN and used occupations at different levels of the SOC coding structure since there were instances where there was not a one-to-one match and there were categories where more specificity would not serve a public health purpose (for example, administrative and office support).  The updated list was first integrated into MAVEN for the Foodborne Illness module and subsequently into the General Epidemiology, Tuberculosis, and Sexually Transmitted Disease modules in May 2015. Data using the first three months of the new occupation list will be analyzed in August to identify the most common occupations seen in a variety of diseases under surveillance. Distributions will be compared to previous years to identify any differences and evaluate the usefulness of the new occupation list.       Conclusion:  Using a collaborative approach, MDPH occupational health, MDPH Bureau of Infectious Disease staff successfully updated the occupation list in MAVEN and standardized it across diseases to allow for easier data entry and more accurate data collection.  Standardization will enable a better comparison of occupations within MAVEN and across MDPH data sets.  More complete and accurate occupation data will help to inform efforts to identify risk groups and prevent disease.

Lin-Mruk, PiI Debby (Lin-Mruk_EO_345.pdf)

Association of endometriosis and phthalate exposure: A Meta-analysis of observational studies   Qian Di, M.S   Rui Hu, M.S Yinyin Xu, M.S  Zhi Yu, M.S  Zhuoxian Zhao, M.S     The potential association between endometriosis and elevated phthalate metabolites level has been analyzed in several epidemiological studies. In order to establish whether phthalate exposure influences the risk of endometriosis, we summarize the evidence from published studies on this issue. We performed a meta-analysis of epidemiological studies published up to April 2014 using PubMed/MEDLINE and EMBASE. We computed summary odds ratio (OR) of endometriosis associated with different phthalate metabolite levels. Data were extracted from eligible studies and cross-checked by at least two reviewers, following the reporting format of Meta-analysis of Observational Studies in Epidemiology (MOOSE). We identified five eligible studies, including a total of 468 women with endometriosis and 1720 controls. The odds of endometriosis for the high urinary MEHP concentration group was 1.029 (0.562–1.877) folds higher than the low concentration group. The present meta-analysis provided no evidence for an association between urinary MEHP concentration and the risk of endometriosis.

 

Longeberger, Allison

Comparison of Electronic Laboratory Reporting and Laboratory-based Manual Reporting — Pennsylvania, 2007–2014 Kirsten Waller, MD, MPH  BACKGROUND: Standards-based electronic laboratory reporting (ELR) system has been in place since 2005 through Pennsylvania’s electronic disease surveillance system (PA-NEDSS); however, it was not available for most reportable conditions until 2007. Laboratories that do not participate in ELR must key-enter data manually into PA-NEDSS. Although laboratories undergo a rigorous ELR onboarding process, the quality of ELR data in Pennsylvania has not been systematically evaluated and routine feedback is not provided to laboratories. We compared the timeliness and completeness of ELR data with laboratory-based manual entry reports (non-ELR).  METHODS: We extracted ELR and non-ELR reports from 2007 through early December 2014 from PA-NEDSS relational databases. Lead, tuberculosis, sexually transmitted infection and human immunodeficiency virus reports were excluded. Odds ratios and confidence intervals were used to assess associations between missing data and report type (non-ELR versus ELR) by specimen source, specimen collected and received dates, test completed date, and patient address. Timeliness (number of days to report) was calculated by subtracting test result date from the date the report was received by PA-NEDSS.   RESULTS:  From January 1, 2007 through December 2, 2014, a total of 1,074,521 laboratory reports were received by PA-NEDSS; of these, 531824 (49.5%) were received via ELR. The proportion of reports received by ELR has increased over time from a low of 6.2% in 2007 to a high of 66.2% in 2014 with a simultaneous 400% increase in total laboratory reports received (45,527 to 227,798). In 2007, two laboratories used ELR for at least some conditions; in 2014 the number of laboratories participating in ELR had increased to 22. Non-ELR reports were more likely to have missing data for specimen source (16.33% versus 4.10%; OR: 4.57; 95% CI: 4.50–4.64), specimen collected date (7.43% versus 0.46%; OR 17.19; 95% CI: 16.50–17.91), laboratory received date (15.49% versus 0.15%; OR: 124.99; 95% CI: 116.47–134.14), and test completed date (12.37% versus 0.18%; OR: 80.41; 95% CI: 75.37–85.79); however, ELR reports were more likely to have missing address data (1.23% versus 4.83%; OR: 0.25; 95% CI: 0.24–0.25). ELR reports were more timely compared with non-ELR reports (median: 1 and 2 days respectively).  CONCLUSIONS:  Automation of reporting through ELR has resulted in more complete and timely disease reporting in Pennsylvania. However, despite rigorous ELR onboarding, missing data, particularly for patient address, remains an area of concern. Laboratories should receive standardized feedback to assure the quality of laboratory reporting data.

 

Martelon, MaryKate (Martelon_ID_345 B.pdf)

Automated Methods to Improve the Completeness of Key Data Elements in Infectious Disease Response   Susan Soliva, MPH Scott Troppy, MPH Gillian Haney, MPH   Background:  Certain information is vital to infectious disease follow-up including occupation for enteric diseases or blood transfusion history for babesiosis.  In January 2014, the Massachusetts Department of Public Health (MDPH) implemented new functionality in the state’s electronic surveillance and case management system, the Massachusetts Virtual Epidemiologic Network (MAVEN), to highlight the need to capture these critical data.  ISIS hypothesized that incorporation of this new functionality would prompt case investigators to answer these important questions.  Methods:  In January 2014, the MDPH, Bureau of Infectious Disease, Office of Integrated Surveillance and Informatics (ISIS) enabled a MAVEN function, called Concerns.  Concerns are color-coded text highlights which are specific by disease. Concerns appear in the main notification section of disease events and highlight questions that need to be completed during case follow-up.  For example, for enteric diseases, ISIS created the following concern: “Please note the case's occupation in the Demographic Question Package.” For babesiosis, ISIS created a concern: “Please note in the Risk/Exposure/Control & Prevention Question Package whether or not the case has received a blood transfusion, tissue products or organ transplant in the past year.”  The effect of concerns on completeness of these variables among confirmed cases in 2013 and 2014 was evaluated for statistical significance with Pearson’s chi-squared test for association with percent of completeness.  Cases where the official city was missing or listed as N/A, cases where the investigation was not completed due to lost to follow up, and where the investigation was deemed unnecessary by MDPH, were excluded from the analysis.   Results:  For salmonellosis in 2013, 48% (N=482) of 1013 confirmed cases were missing occupational information.  In 2014, 25% (N=278) of 1121 confirmed cases were missing occupational information.  This is a reduction of 47% of missing information (p<0.0001).  For babesiosis in 2013, 19% (N=71) of 366 confirmed cases were missing blood transfusion information.  In 2014, 10 % (N=45) of 467 confirmed cases were missing blood transfusion information (p<0.0001).  This is a reduction of 47% of missing information (p<0.0001).  Conclusions:  Incorporating concerns into MAVEN significantly reduced the percentage of missing information for vital questions in disease follow up.  Additional diseases should be considered for a similar approach to increasing completeness of important variables."

 

 

Maxted, Angie (Maxted_ID_1045.pdf)

Of Rats, Horses, and Zebras: An Old-World Hantavirus Investigation    Amy Burns, MS Pam Griffith, RN, BSN  Suzanne Osterhoudt, RN, BSN  Jennifer White, MPH   Background: Infection with New World hantaviruses, which can cause hantavirus pulmonary syndrome (HPS), is uncommon in the Northeast. Old World hantaviruses, such as Seoul virus, can cause hemorrhagic fever with renal syndrome (HFRS). Although Seoul virus has been detected among Norway rats in major U.S. cities, domestically acquired human infection is extremely rare. Objective: We investigated a report of Seoul hantavirus infection in rural New York.  Case Report and Results: In May 2015, a previously healthy woman aged 58 years developed self-reported fever to 103F, myalgia, cough with episodes of dyspnea on exertion, fatigue, right flank pain, and a multifocal, blanching, erythematous rash over much of her body. Each rash was approximately circular and up to several inches in diameter. Except for the rash and fatigue, the illness resolved rapidly and the patient did not require hospitalization. The patient was concerned about hantavirus infection; she reported a recent rat infestation of her basement home and that she had cleaned up ceiling insulation and a storage area that contained rat carcasses and excreta. Commercial hantavirus serology was positive for IgM antibodies. Complete blood count, blood chemistry panel, urinalysis, coagulation studies, and imaging studies of the chest and kidneys did not reveal any abnormalities consistent with HPS or HFRS. The rash was consistent with but not immediately recognized as multiple erythema migrans, thus doxycycline treatment was not immediately prescribed. Lyme disease was laboratory confirmed. Initial CDC testing detected the presence of IgM antibodies against Seoul hantavirus in an acute serum specimen, however, no IgM or IgG antibodies were detected in a convalescent specimen collected six weeks later. CDC ruled their own initial test result as well as the commercial test result false positives.  Conclusions: The patient’s significant rat exposure and self-concern for hantavirus likely influenced the healthcare provider’s decision to order hantavirus serology testing, despite the patient’s lack of typical HPS or HFRS clinical symptoms, signs, or bloodwork abnormalities. The falsely positive serologic test results likely led to heightened anxiety in the patient and additional, possibly unnecessary, laboratory and imaging studies. Despite Lyme disease being much more common than hantavirus infection in New York and the patient’s clinical presentation being consistent with Lyme disease, because of the rat exposure history the patient’s illness was not immediately recognized as Lyme disease and doxycycline treatment was delayed."

 

Mazur, Nicole (Mazur_RF.pdf)

Importance of Post-Exposure Prophylaxis Following Confirmed Case of Hepatitis A in a   Foodhandler  Nicole L. Mazur, MPH1, Rosemary Kidder, RN, BSN, MPH1, Lisa A. McHugh, MPH1  1Division of Epidemiology, Environmental and Occupational Health, New Jersey Department of Health;     Abstract  Introduction:.  On December 1, 2014, New Jersey Department of Health (NJDOH) was notified of a confirmed case of Hepatitis A virus (HAV). The individual was identified as a foodhandler, who had worked at a high volume restaurant during their infectious period, November 10th through December 1st. Given that HAV is transmitted through the fecal-oral route, there is great concern for potential secondary cases after contact with an ill foodhandler. Historically, HAV vaccine has been recommended as post-exposure prophylaxis (PEP) for those less than forty years of age, and HAV immunoglobulin (IG) has been recommended for those greater than forty years of age. Considering the potential large numbers of exposures a timely public health response was crucial.     Methods: The local health department (LHD) worked quickly to interview the foodhandler, trying to ascertain the dates they worked, and specific job duties. It was determined that the case had performed various foodhandling duties during his infectious period, including preparing ready to eat products, and was symptomatic during this time. Working with NJDOH, the LHD moved quickly to distribute public notification through an online bulletin, posted the day the case was identified and a subsequent press conference was held on December 2, 2014. They also began to coordinate vaccination clinics, and procuring HAV IG for those potentially exposed, in conjunction with surrounding jurisdictions. A LINCS message was sent, notifying all public health partners in NJ of the possible exposure, and an EpiX message was subsequently posted to notify those in other states. Three separate clinics offering vaccination were held. IG was recommended, but no large scale administration was performed.    Results: Pinpointing the number of potential exposures was unable to be determined, based upon the high volume of the restaurant, which did mostly cash transactions. With public notification, over 700 persons were vaccinated. Three additional cases of Hepatitis A were identified through surveillance, all of whom had eaten at the restaurant during the time the index case worked and was infectious. Subsequently it was discovered that two of the cases had been vaccinated after learning of their potential exposure, but were over the age of forty. The other case who was aware of the situation had chosen not to get PEP. On January 8th, a fourth case of confirmed Hepatitis A was identified, who had known of their exposure, but decided against PEP.    Discussion:   HAV infection in foodhandlers requires timely investigation and implementation of both immunoprophylaxis, and proper control measures. Assessing the foodhandler’s job responsibilities, calculating the period of infectivity and determining the time period in which the food handler worked are essential in providing public health recommendations. It is also crucial that those public health recommendations are consistent between both state and local health departments."

 

 

Mazur, Nicole

Lassoing Lassa: Contact Tracing in an Acute Care Facility Following a Case of Lassa Fever  Nicole L. Mazur, MPH1, Lindsay Hamilton, MPH1,2, Kim Cervantes, MPH1, Shereen Semple, MS1  1Division of Epidemiology, Environmental and Occupational Health, New Jersey Department of Health; 2CDC/CSTE Applied Epidemiology Fellowship, Atlanta, GA   Abstract  Introduction:.  On May 25, 2015, the Centers for Disease Control and Prevention (CDC) confirmed the presence of Lassa virus in a New Jersey (NJ) resident with recent travel to Liberia, arriving to the state on May 17, 2015. The individual had been under active monitoring (AM) for Ebola virus disease (EVD) but had not reported symptoms to the local health department (LHD); suspicion for Lassa fever (LF) was not recognized until later in the illness course. This NJ resident visited 2 acute care hospitals, Hospital A and Hospital B during the course of their illness, with two visits to Hospital A, and one visit to Hospital B. LF is an acute viral illness caused by Lassa virus, an arenavirus endemic to West Africa, where persons primarily become infected by ingesting or inhaling excreta from multimammate rats. Symptoms present one to 21 days following exposure, and 1% of cases are fatal. Secondary cases may result from exposure to infected body fluids.    Methods:   Contact tracing was performed by Hospital A to determine potentially exposed healthcare workers. These healthcare workers were then assigned by NJDOH to high or low risk classification categories based on responses to a standardized questionnaire assessing contact with infected body fluids and use of personal protective equipment.  Contacts were monitored for 21 days using the AM protocol implemented for EVD monitoring.      Results:   Hospital A identified a total of 95 potentially exposed contacts were identified, all of which were interviewed. Of those interviewed 81 reported possible exposures; all of which were classified as low risk. Hospital A used department supervisors to administer the risk assessment questionnaires, and to provide educational information to their employees. Working with their occupational health department, daily symptom and temperature monitoring was implemented. Employees were placed into three reporting groups randomly, each given a daily cutoff time for reporting, to ensure compliance. Hospital A’s administration had also made reporting a condition of employment.  Employees were instructed to make immediate contact with occupational health for any symptom onset rather than wait for their reporting time.  Hospital A’s administration would also send daily summary of monitoring and any symptoms of concern to NJDOH contact person. Daily symptom monitoring of contacts concluded on June 18, the end of the last contact’s 21-day incubation period.  No secondary cases were identified.      Conclusion:   Although secondary transmission of LF is less likely to occur in the U.S. than in endemic areas of West Africa, health care providers need to remain vigilant about Lassa and other viral hemorrhagic fevers to ensure timely identification, treatment and containment of disease. Use of current EVD monitoring protocols and establishment of NJDOH contact person was valuable for rapid implementation of large-scale contact tracing and effective communication and can be applied for various other infectious diseases."

 

Mehr, Jason (Mehr_ID_345 A.pdf)

Heart of the problem: Investigation of 3 bacterial endocarditis cases following oral surgery- New Jersey 2014       Background  Bacterial endocarditis is an infection of the heart valve or lining. Endocarditis typically occurs in individuals with underlying heart defects, artificial valves, or a history of endocarditis but in rare instances this infection can affect individuals without previous heart conditions. A single case of endocarditis is not reportable to the New Jersey Department of Health (NJDOH).  In October 2014, a healthcare provider reported that two patients were admitted to the same hospital with endocarditis, each following third molar extractions by the same oral surgeon. Blood cultures from both patients grew Enterococcus faecalis which literature suggests is the causative organism in only 5-10% of endocarditis cases.   Methods  In conjunction with the local health department (LHD), a public health investigation was initiated including patient interviews and medical records reviews.  The New Jersey Board of Dentistry (BOD) was notified of the investigation and informed NJDOH of a 2013 patient complaint of an endocarditis case following an extraction and bone graft performed in December 2012 by the same oral surgeon. This patient also had endocarditis caused by Enterococcus faecalis. Representatives from the New Jersey Division of Consumer Affairs Enforcement Bureau, LHD, and NJDOH performed an unannounced site visit to the surgeon’s office in November 2014.   Results  Patient interviews identified no other exposures or procedures. Breaches identified during the site visit include inadequate sterilization and disinfection practices, storage of sterile syringes and instruments outside their protective covers, handling of intravenous medications, routine use of single-dose medication vials for more than one patient, inadequate hand hygiene and glove use, and failure to maintain written infection prevention policies and procedures. NJDOH recommended the immediate disposal of all opened medication vials, sterile patient care items, and syringes and the reprocessing of all unwrapped surgical instruments. NJDOH recommended hiring an infection preventionist to ensure all policies and procedures are consistent with the Centers for Disease Control and Prevention’s guidelines and to review staff competencies.    Conclusion  Endocarditis is an uncommon but serious healthcare-associated infection. Since endocarditis is not reportable, this outbreak was brought to the attention of public health authorities by an astute clinician.  A close collaborative relationship with the BOD was essential in identifying an additional case and in conducting this investigation. Lastly, our investigation highlighted differences between best practice and common practice among oral surgeons and reinforced the need to optimize infection prevention practices in all outpatient facilities.

 

Melchreit Richard (Melchreit_ID_345 A.ppt.pdf)

Assessment of infection control and prevention capacity and infrastructure in acute care hospitals in Connecticut using National Healthcare Safety Network annual surveys    Since 2008, the 29 acute care hospitals in Connecticut have submitted annual surveys to the National Healthcare Safety Network (NHSN).  These data include information about the size and scope of facility infection prevention programs, laboratory testing capacity, and other data that are useful for assessing the infrastructure for HAI prevention in the state including the recent addition of data about facility antimicrobial stewardship programs.  Such data is now collected in all states and territories, as hospitals must use NHSN and fill out the annual survey to participate in the Centers for Medicare and Medicaid Services quality incentives programs.  NHSN’s annual surveys are an underutilized source for capacity assessment and program planning.  As assessment and mitigation of facility infection control program has increasingly been understood to be critical for both routine HAI prevention activities and for response to emerging infectious diseases (such as Ebola), robust and routine use of annual NHSN survey is important for state healthcare associated infections programs. The Connecticut HAI program has undertaken a systematic review of these data.  We will present a summary of our findings, and discuss how these data can be used for program planning and advocacy.

 

 

Mertz, Kristen

Protocol Development for Investigation of Single Cases of Legionnaires’ Disease, Allegheny County, PA  Introduction: Legionnaires’ disease is a waterborne respiratory disease that disproportionately affects immunocompromised individuals and the elderly.    Sharon Silvestri, RN   Lauren Torso, MPH   LuAnn Brink, PhD    The age-adjusted rate of Legionnaires’ disease in Allegheny County, PA is four times greater than the general U.S.  Due to this high incidence, follow-up on single cases of Legionnaires’ disease among non-community dwelling individuals in Allegheny County has historically been more aggressive than other parts of the U.S.  Our objective was to develop protocols for follow-up of single cases of Legionnaires’ disease to ensure consistency of approach and appropriate use of limited resources.      Methods: Protocols for environmental testing and health precautions were developed through consultation with multiple Allegheny County Health Department (ACHD) programs including the Infectious Disease Program, Public Health Laboratory, Public Drinking Water and the Bureau of Assessment, Statistics, and Epidemiology.  We conducted in-person meetings and conference calls to develop the protocols and revise them as needed.  Literature was reviewed and other public health departments were consulted to compare follow-up strategies.    Results: Protocols were developed for follow-up in the following settings: 1) hospitals 2) long-term care facilities (LTCF) 3) senior apartment buildings 4) all-ages apartment buildings 5) other community settings such as homes or workplaces.  Definite and possible transmission protocols were developed for each type of setting.  Definite transmission is defined as exposure to the facility for the patient’s entire 10-day incubation period.  Possible transmission is defined as exposure to the facility for 2 to 9 days of the incubation period.  ACHD resources were not allocated for definite or possible hospital transmission given greater infection prevention capacity at these facilities.  ACHD resources for environmental testing were deemed appropriate to offer for definite and possible nursing home transmission and definite senior apartment building transmission.    If health department resources were deemed inappropriate to offer, such as in the event of a single case of Legionnaires’ disease associated with an all ages apartment building, the protocol specified that a list of local laboratories available to conduct Legionella environmental testing should be provided to the facility.  If environmental tests yield positive results, the protocols offer steps for immediate remediation and continued testing with the goal of eliminating contamination in the building water system.  Restrictions on showering, drinking water and ice machine use were recommended until all environmental tests were negative.  Along with a protocol for each scenario, a packet of information was developed to be shared with the affected facility.  Packets include a letter describing recommendations for control, protocols for environmental testing and remediation, and a list of available laboratories for Legionella testing and companies available for remediation consultation.      Conclusion: The protocols have provided clearer direction for follow-up and have better defined how health department resources are utilized under each scenario.  Impact on prevention of future cases has yet to be determined.

 

Mertz, Kristen

Investigation of hepatitis C cases with limited resources in Allegheny County, PA   Jennifer Fiddner, MPH Vivian Liang, BA  Background:   In 2014, 2,364 persons with positive hepatitis C tests were reported to the Allegheny County Health Department (ACHD), an increase of 37% over the average number of reports received during the previous 10 years.  Of these cases, 1,149 (49%) were classified as “confirmed” based on positive RNA test results.   Given that most cases were reported from laboratories via the state’s electronic disease surveillance system (PA-NEDSS), no clinical or risk factor information was available, and ACHD lacked staff to investigate.  In 2015, AmeriCorps members assigned to ACHD are conducting targeted follow-up on reported cases.  The objective of this project is to correctly classify hepatitis C cases, identify possible sources of transmission and assess linkage to care.    Methods:  For persons with positive tests in 2015 who have not previously been reported, one-page forms are faxed to providers listed on electronic laboratory reports for purpose of collecting demographic, clinical, and risk factor information.  In addition, phone interviews on risk factors are conducted with persons <35 years of age with first-time positive test results under the assumption that this younger group is more likely to be recently infected.    Results:  In January 1 through July 17 of 2015, ACHD received positive hepatitis C test results on 1,789 individuals not previously confirmed as cases.  Of these, 465 (26%) are <35 years of age.  We have attempted to call 136 persons in this age group; of these, 57 (42%) had invalid phone numbers, 42 (31%) did not return calls, and 37 (27%) were successfully contacted.  Of those contacted, 25 (68%) reported past or current intravenous drug use, 14 (38%) reported >10 sex partners, and 13 (35%) reported a history of incarceration. Faxing investigation forms to providers has been delayed because of missing provider phone and fax numbers on electronic laboratory reports, requiring time-intensive follow-up by clerical staff.  Of 130 forms faxed to providers as of July 17, 25 were returned by July 24.  Faxing will continue throughout the year and results on case classification, risk factors, and treatment history will be compiled.    Conclusions:  Investigation of hepatitis C cases is resource intensive due to 1) large volume of reports, 2) transient nature of patient population, and 3) incomplete electronic lab reports making contacting providers difficult.   We plan to continue our efforts to investigate cases because of the serious nature of infections and the need to identify sources of ongoing transmission in the community.

 

Nelson, Deborah (Nelson_RF.pdf)

The Role of Childhood Violence Exposure, Self-esteem and Depressive Symptoms on Consistent Contraception Use among Young, Sexually Active Women    Huaqing Zhao, Phd  Rachel Corrado MS   Dimitrions M Mastrogiannnis, MD PhD   Stephen J Lepore, PhD   Objectives: This longitudinal assessment of young non-pregnant sexually active women examined the role of childhood violence exposures, perceptions of community violence, and levels of psychosocial constructs on the report of consistent, effective contraceptive use over a 9-month period.  Study Design: Eligible women were recruited from Philadelphia family planning clinics in 2013 and were followed for 9-months to longitudinally identify factors contributing to the type and consistency of contraception use. At baseline, a self-administered, computer-assisted interview was used to collect data on childhood physical and sexual violence, perceptions of community-level violence, current violence, substance use, and level of depressive symptoms, self-esteem, and sexual self-efficacy. Women were re-contacted 9-months later to assess their level of sexual activity, and the type/consistency of contraception use.  Results: Childhood sexual violence, low self-esteem, high depressive symptoms, and consistently high levels of depressive symptoms from baseline to follow-up were significantly related to reports of inconsistent/no contraception use over the follow-up period.  Given the strong correlation between self-esteem and depressive symptoms, a factor analysis was conducted. Exposure to childhood sexual violence and the factor incorporating high depressive symptoms and low self-esteem  were significantly and independently related to inconsistent/no contraception use over the 9 month follow-up period (aOR=1.91, 95% CI: 1.03-3.55 and aOR=1.27, 95% CI: 1.01-1.59 respectively).    Conclusions We found a strong relation between exposure to childhood sexual violence and inconsistent/no contraception use. Reporting high depressive symptoms at baseline, consistently high depressive symptoms over the follow-up period and low self-esteem levels were significantly, and independently, related to inconsistent/no contraception use among young, urban sexually active women.  These findings note the importance of trauma-informed approaches targeting increases in self-esteem and reductions in depressive symptoms may be most effective in improving consistent contraception use among young women.

 

Noonan-Toly, Candace (NoonanToly_ID_145.pdf)

A Method for Tracking Perinatal Hepatitis B in New York State      Objective:  Provide an easy way to track infants born to Hepatitis B surface antigen (HBsAg) positive women in New York State (NYS) excluding New York City (NYC).  Intro:  To prevent perinatal Hepatitis B virus transmission, the Advisory Committee on Immunization Practices (ACIP) recommends that infants born to HBsAg positive women receive post-exposure prophylaxis with hepatitis B vaccine and hepatitis B immune globulin (HBIG) within 12 hours of birth, and complete the 3-dose Hepatitis B series. To determine infant outcomes after post-exposure prophylaxis, ACIP recommends post-vaccination serologic testing (PVST) at age 9-18 months.   NYS excluding NYC, tracks on average, 300 infants of HBsAg positive mothers annually.  Our current Communicable Disease Electronic Surveillance System (CDESS) provides a robust infant tracking module for local health departments (LHDs) to enter and monitor vaccine information, add multiple infants per mother, and track patient movement and loss to follow-up.  The goal is to provide a tool for LHDs to analyze their infants’ data, by birth year cohort, with all of their current vaccination and serology information available in one record.  Methods:     The infant’s birth information, mother’s information and the provider information in CDESS is prepopulated from Newborn screening for each infant born to an HBsAg positive mother. LHDs enter the infant’s vaccination record, and the infant’s serology record and checks the tracking complete box along with the date completed and no more follow-up is necessary.    The CDESS data resides in Oracle tables. Using SAS V9.3, the complex relational data is smoothed to one record per infant.  This data is then converted to one comma separated values (CSV) file for each county and birth year cohort.   These CSV files are pushed to the Oracle database by using the Oracle SQL*Loader utility.  This loads the CSV files as a Character Large Object (CLOB) into an Oracle table in the CDESS database.    SAS runs automatically every night and sends the updated CSV files to CDESS.  The updated CSV files are immediately available for download by the LHDs. Using these files county users can filter their data by provider, dosage and age to be more proactive in protecting the health of babies in NYS.  Results:  The easy access of one file removes multiple steps in finding the infants’ vaccination and serology information.  Using these CSV files it is easily found that there are currently 259 babies being followed in the 2014 birth cohort.  Twenty babies of the original 285 birth cohort moved out of jurisdiction, while six cannot be located.  Of the 259, five did not receive HBIG at birth and four did not receive dose one at birth. By eight months of age 219 (85%) had received three Hepatitis B vaccinations.  Thirty-eight percent have already had their PVST.  Conclusions:  Removal of multiple steps in data retrieval saves time for the LHD.   The LHDs have an easy way to analyze their own data and be more proactive on their follow-up with physicians and families.  All infants within a jurisdiction can be monitored simultaneously providing easy decision making capabilities.  User friendly and easy ways to analyze data are the values for the users to collect good data.      Centers for Disease Control and Prevention. Postvaccination Serologic Testing Results for Infants Aged <=24 Months Exposed to Hepatitis B Virus at Birth – United States, 2008-2011. MMWR 2012 / Vol. 61 / No. 38: 768-71."

 

Okeke, Janice (Okeke_LB.pdf)

Effects of living in an underserved area on fruit and vegetable purchases among households participating in the New Jersey Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)  Introduction:  A growing body of research has highlighted an association between limited access to healthy food choices and increased risk of poor diets and obesity.  In response, public health strategies ranging from building or supporting healthy food retail outlets to improving transportation to grocery stores and farmers markets have been introduced as a way to improve access to healthy foods in areas often referred to as food deserts or underserved areas.  However, the actual effects of living in an underserved area on healthy food access are largely unknown.  Understanding and documenting these associations are essential to informing public health programs that strive to improve nutrition, particularly among underserved populations.  The purpose of this study was to assess the effects of living in an underserved area on fruit and vegetable purchases among low-income women and children participating in the New Jersey Special Supplemental Nutrition Program for Women, Infants, and Children (WIC).    Methods:  New Jersey WIC participants with continuous enrollment from June 2013 to May 2014 were eligible for the study.  Household addresses were geocoded to the census tract level and the Centers for Disease Control and Prevention (CDC) modified retail food environment index (mRFEI) was used to identify whether each household was located in an underserved area.  The mRFEI index measures the number of healthy and less healthy food retailers within census tracts across the United States.  Households in the study were divided into two groups:  those in areas that had at least one healthy food store (mRFEI≥1) and households located in underserved areas without any healthy food stores (mRFEI=0).  The New Jersey WIC program issues low-income women and children cash value vouchers (CVVs) every three months to purchase fruits and vegetables and these CVVs can be redeemed monthly at WIC-authorized vendors.  The proportion of money redeemed from CVVs for each WIC household was calculated by dividing the total CVV dollar amount redeemed by the total CVV dollar amount issued from June 2013 to May 2014.  Logistic regression was used to assess the association between living in an underserved area and complete CVV redemption, which was defined as a redemption proportion ≥70%.    Results:  Of the 30,078 WIC households in our study, 11.4% lived in underserved areas without a healthy food store.  The majority of households had a mother without a college degree (64.5%) and the majority of mothers were either non-Hispanic black (15.7%) or Hispanic (55.2%).  The odds of complete CVV redemption for fruits and vegetables was 31% lower for WIC households without a healthy food store in their census tract relative to WIC households in census tracts with at least one healthy food store after adjustment for mother’s race, mother’s education, mother’s age, and SNAP participation (OR=0.69, 95% CI:  0.62-0.76).    Conclusion:  The odds of complete CVV fruit and vegetable redemption was significantly lower for WIC households in underserved areas as measured by the mRFEI. Findings from this study suggest that public health efforts aimed at increasing access to healthy foods in these areas are warranted.  New Jersey’s WIC program has already begun to address this issue through policy change by mandating that all WIC-authorized vendors stock a minimum of two fruits and two vegetables in their stores by June 2014.  This includes WIC authorized small retail vendors (corner stores, bodegas) commonly located in areas without any healthy food stores.  Our study provides a framework for evaluating the impact of this policy change and other healthy corner store initiatives.

Paone, Denise  Tuazon, Ellenie  (Paone_CC_145.pdf)

New York City RxStat: A Public Health and Public Safety Partnership   The presenters will conduct a panel presentation on RxStat, an innovative New York City public health and public safety partnership: a public health approach to track drug use and associated health and safety consequences. Presenters will highlight the use of real-time data, provide examples of data driven initiatives, and the rapid deployment of public health and public safety resources to high need neighborhoods with rigorous follow-up to monitor impact. The panel will discuss how this model can be implemented by local jurisdictions to address opioid involved unintentional drug poisoning (overdose) deaths. Specific datasets to be discussed include drug poisoning (overdose) mortality data and prescription drug monitoring data to evaluate public health interventions, and syndromic surveillance data as an early warning system. Participants will learn how to use data to inform public health responses, including opioid prescribing guidelines, public health detailing, overdose prevention, and medically assisted treatment.  The proposed session will be interactive and provide ample opportunities for discussion.

 

 

Eleni Papadopoulos (Papadopoulos_EO_345.pdf)

(1), Reports of Adverse Health Effects Related to Synthetic Cannabinoid Use in New York State (April 1-June 30, 2015)         K. Cummings(2), J. Marraffa(3), K. Aldous(2), L. Li(2), N. Ahmad(2)    (1) University at Albany, School of Public Health, Albany, NY  (2) New York State Department of Health, Albany, NY  (3) Upstate New York Poison Control Center, Syracuse, NY     Reports of Adverse Health Effects Related to Synthetic Cannabinoid Use in New York State (April 1-June 30, 2015)     Background: Synthetic cannabinoids are laboratory-produced chemicals designed to act on cannabinoid receptors in the brain and marketed as 'legal' alternatives to illicit drugs. To evade detection in drug screens and avoid controlled substance laws, these compounds are chemically altered leading to various adverse health effects.  In 2011-2012, an outbreak of reported adverse health effects related to synthetic cannabinoid use prompted a regulatory response. During the 2011-2012 outbreak, compounds with the JWH (John W. Hoffman) and HU (Hebrew University) structural classes made up the majority of those identified. On April 16, 2015, the New York State Department of Health (NYSDOH) was notified by the Upstate Poison Control Center (PCC) of 15 reports of adverse health effects related to synthetic cannabinoids in County X. NYSDOH heightened surveillance efforts and issued health advisories and press releases, directing calls to the PCC. Through enhanced surveillance and case finding NYSDOH (1) assessed if the 2015 outbreak was an emerging public health threat and (2) compared the 2015 outbreak with the 2011-2012 outbreak.    Methods: We defined a case as a person reportedly exposed to synthetic cannabinoids in NYS during April 1 to June 30, 2015 or June 20, 2011 to July 29, 2012. We defined NYS as all counties, excluding those in NYC, and Westchester, Nassau, and Suffolk counties. For this analysis, we reviewed NYS data from the National Poison Data System (NPDS), a national database logging information reported by PCCs around the country. NPDS used a unified code to track all reports of synthetic marijuana, and collects information on demographics, exposures, and clinical effects. We defined the 2015 outbreak as April 1 to June 30, 2015, and the 2011-2012 outbreak as June 20, 2011 to July 29, 2012. Demographics were reported for each outbreak and chi square analyses were run between the outbreaks with SAS v. 9.4 to evaluate differences in hospital admissions, severity of effect, and common symptoms. We tested samples of synthetic cannabinoids obtained from 2015 cases using high resolution mass spectrometry.    Results: During the 2015 outbreak, we identified 335 cases. Of the 335 cases, there were 271 (80.9%) males with a median age of 27 years (range: 8 months-65 years). During the 2011-2012 outbreak, we identified 407 cases. Of the 407 cases, there were 312 (76.7%) males with a median age of 20 years (range: 1-68 years). During the 2015 outbreak, there was a 247% increase in average cases per week in comparison to the 2011-2012 outbreak. There was also an 89% increase in the proportion of cases admitted to a critical care unit during the 2015 outbreak when compared to the 2011-2012 outbreak (p=0.0009). The 2015 outbreak saw a 28% increase in the proportion of cases reporting a “major” or “moderate” effect compared to 2011-2012 outbreak (p=0.0122). During the 2015 outbreak, cases were significantly more likely to report experiencing agitation/irritation (p<0.0001), lethargy (p<0.0001), hypotension (p<0.0001), bradycardia (p<0.0001), coma (p=0.0005), confusion (p=0.0011), and respiratory distress (p=0.0065) when compared to cases during the 2011-2012 outbreak. Compounds identified from the 2015 outbreak samples included MAB-CHMINACA and XLR-11 as single or mixed products applied to plant material.     Conclusion: This investigation shows significant differences in clinical presentation following synthetic cannabinoid exposure, including significantly higher critical care admissions between 2015 cases compared to 2011-2012 cases. We identified chemical compounds used as synthetic cannabinoids in the 2015 outbreak that were not identified during the 2011-2012 outbreak. The significant difference in underlying chemical structure prompted the NYSDOH to propose regulation updates, adding new groups of chemical compounds based on those identified. Synthetic cannabinoids are a current public health threat in NYS and NYSDOH continues to monitor the outbreak.

 

 

Patel, Shilpa (Patel_CH_1045.pdf)

Understanding the Barriers and Facilitators to Colorectal Screening among South Asian Immigrants in NYC  Taher, MPH Yousra Yusuf, MPH Victoria Raveis, PhD  Simona Kwon, DrH Joseph Ravenell, MD  Nadia Islam, PhD  Background: South Asian immigrants, comprised of Asian Indian, Bangladeshi and Pakistani subgroups, remain poorly understood and are one of the most understudied racial/ethnic minority groups in the US. Aggregated data can mask further significant differences as Asian Indians, Bangladeshis, and Pakistanis have reported colorectal cancer (CRC) screening rates ranging from as low as 25% to 41%. The goal of this one-year study is to understand the barriers and facilitators of CRC screening among South Asian immigrants in NYC using a mixed methods, community-engaged approach.   Methods: Quantitative data from the NYC Community Health Survey, a large, city-wide cross-sectional dataset with detailed country-of-origin information from Asian American immigrants, were aggregated from 2009-2012 (n=1232) and descriptive and regression analyses were conducted to explore barriers and facilitators to colonoscopy screening. Qualitative data from in-depth interviews with NYC South Asian immigrants (n=42) collected information on knowledge, attitudes, and behaviors related to CRC screening, and provided contextual information on quantitative findings.   Results: Preliminary results suggest South Asian immigrants face a number of barriers to colonoscopy screening, including access to care factors such as health coverage type and having a regular PCP, sociodemographic factors such as years in country and language, and lifestyle factors. Qualitative results also suggest lack of knowledge is a significant barrier to screening. Both types of data reveal gender differences in colonoscopy screening behavior.   Conclusion: A confluence of poor access to healthcare, language barriers, and cultural and social beliefs play a role in CRC screening disparities among South Asian immigrants.

 

Pawlish, Karen (Pawlish_CH_1045.pdf)

Racial/ethnic Differences in Risk of Subsequent Invasive Breast Cancer Among Women Diagnosed with Invasive Breast Cancer and Ductal and Lobular Breast Carcinoma in situ in New Jersey, 1992-2012        Women diagnosed with breast cancer and breast carcinoma in situ are at increased risk for subsequent primary breast cancer, and this risk may vary by race and ethnicity. The risk of developing subsequent invasive breast cancer by race and ethnicity, age group, and histologic subtype was examined in a cohort of 136,671 New Jersey women diagnosed with invasive breast cancer and breast carcinoma in situ from 1992 to 2012, using data from the NJ State Cancer Registry. Standardized incidence ratios (SIR) for invasive breast cancer and 95% confidence intervals (CI) were calculated using the MP-SIR session of SEER*Stat. Compared to the NJ female population, risk of subsequent breast cancer was significantly elevated in the four racial/ethnic groups included in the analysis [whites: SIR=1.40, 95% CI 1.36-1.45; African Americans (AA): SIR=2.48, 95% CI 2.28-2.69; Asian/Pacific Islanders (API): SIR=2.26, 95% CI 1.85-2.73; Hispanics: SIR=2.10, 95% CI 1.86-2.37]. The risk for subsequent breast cancer was significantly elevated during the first 5 years, 5-10 years, and 10+ years after diagnosis of the index cancer. The risk for subsequent breast cancer was highest among the youngest women diagnosed with the initial cancer before age 40, in particular among younger AA and Hispanic women (SIR= 8.88, 95% CI=7.01-11.10; SIR=6.54, 95% CI=4.68-8.91, respectively). The risk for subsequent invasive breast cancer was significantly higher among women initially diagnosed with lobular carcinoma in situ (LCIS) (SIRs 2.96, 4.74, 4.48, 4.57 in whites, AA, API and Hispanics, respectively). Risk for contralateral breast cancer was higher than that for ipsilateral breast cancer. Our findings support the importance of continued surveillance of breast cancer patients, especially AA women, women diagnosed at younger ages, and LCIS patients. The risk of subsequent breast cancer continued to be elevated more than ten years after diagnosis.

 

 

Rabatsky-Her, Terry (Rabatsky-Ehr_LB.pdf)

The Sustainability of Traveler Monitoring for Emerging Infectious Diseases: Lessons Learned from Active Monitoring of Traveler’s from West Africa for Ebola Virus Disease in CT, October 2014–July 2015.  Title: The Sustainability of Traveler Monitoring for Emerging Infectious Diseases: Lessons Learned from Active Monitoring of Traveler’s from West Africa for Ebola Virus Disease in CT, October 2014–July 2015.    Authors: Terry Rabatsky-Ehr, Susan Petit, Jocelyn Mullins, and Mathew Cartter.    Author affiliations: Connecticut Department of Public Health, Epidemiology and Emerging Infections Program, Hartford, CT.    Background: Since October 2014, all travelers from countries experiencing ongoing active transmission of Ebola Virus Disease (EVD) have been routed to five U.S. international airports where they undergo health screening and risk assessment per current Centers for Disease Control and Prevention guidelines. The Connecticut (CT) Department of Public Health (DPH) is responsible for interviewing travelers and determining the level of monitoring and movement restrictions. Local health departments (LHD) are responsible for all follow-up of the travelers during the monitoring period. We conducted an evaluation after approximately 10 months of triaging and monitoring of travelers returning to CT from EVD affected countries to identify strengths and weaknesses of the system, find areas where work can be reduced, streamlined, or improved, and make recommendations for reducing staff burden and improving workflow.     Methods: The evaluation was conducted by interviewing participants in the triage and monitoring process and analyzing information collected in the traveler monitoring database. Areas of focus for the evaluation included timeliness, usefulness, completeness, acceptability, sensitivity, and specificity of the system for desired outcomes. Strengths and weaknesses of the system were determined.    Results: During October 16, 2014–July 28, 2015 152 travelers were triaged and monitored by the CT DPH. Of these, 139 (91%) came directly to CT after arrival in the US.  Of these, 129 (93%) were interviewed and the LHD notified within 1 day of arrival. Of the 13 travelers who came to CT from other states and were then monitored by CT DPH, all 13 were triaged and LHD notified within 1 day of arrival in CT. No risk assessment performed by CT DPH differed from that determined by CDC screening. No traveler was classified as lost to follow up during their monitoring period. Since monitoring began at least 3 travelers were evaluated at emergency departments (ED) without first contacting the LHD or DPH. As a result, the hospital ED was not prepared in advance to receive the travelers resulting in a delay in the diagnosis and unnecessary testing for Ebola virus. Strengths of the system were timeliness of triage and completeness of monitoring. Weaknesses included redundancy and complexity because triage, monitoring, and oversight is spread across staff and departments, and unequal burden on staff since travelers are disproportionally destined for a small number of LHDs.      Conclusions: The system is meeting goals of timeliness and completeness of triage and monitoring. However, to make this system acceptable and sustainable the redundancies should be reduced and the work burden be distributed more equally among staff. Given that CT DPH anticipates the addition of triage and monitoring of persons potentially exposed to MERS-CoV and highly pathogenic avian influenza virus, modifications will make the system more adaptable and sustainable should additional diseases be added.

 

 

Rahman, Mohammad

Title: In-utero arsenic exposure, Early marriage, and Preterm birth in Bangladesh      Background: In-utero arsenic exposure, early marriage, and low gestational weight gain (GWG) have been identified as risk factors for preterm birth. Whether the effect of arsenic and early marriage on preterm birth is direct or mediated through pathways related to GWG remains unknown.  Objective: To evaluate the relationship between in-utero arsenic exposure, early marriage, and preterm birth considering GWG as a mediator.   Methods: A prospective cohort of pregnant women was recruited in Bangladesh during 2008-2011.  Of 1,184 singleton live births, 1,181 subjects with complete exposure and outcome data were included in the analysis. Arsenic was measured in personal drinking water during the first trimester.  Age of marriage and other socio-demographic information was collected using structured questionnaire. GWG (lb/week) was measured during the second and third trimesters.   Mediation analysis was used to estimate the direct and indirect effects of arsenic and early marriage on preterm birth. GWG was considered a common mediating variable.     Results: The risk of preterm birth increased with increased natural log water arsenic (RR=1.10; 95% CI:1.02−1.16) after adjusting for confounders; most of which was the direct  effect of arsenic (RR=1.09; 95% CI:1.02−1.15), only 11.5% was mediated through decreasing GWG (RR=1.01; 95% CI:1.00−1.02).Women who married before age 18 had more than twice the risk of preterm birth (RR=2.08; 95% CI:1.10−2.81) compared to those who were married  ≥18 years. The direct and indirect effect risk ratios for early marriage on preterm birth were 2.05 (95% CI:1.10−2.75) and 1.02 (95% CI:1.00−1.05) respectively. Among women married before age 18, GWG significantly decreased the risk of preterm birth (RR=0.62; 95% CI:0.40−0.98); however, this effect was not observed among women married ≥18 years   Conclusion: In-utero arsenic exposure, early marriage and gestational weight gain were associated with preterm birth. However, the effect of arsenic and early marriage on preterm birth appears to be direct and partly mediated through GWG.

 

 

Rausch-Phung, Elizabeth (Rausch-phung_ID_145.pdf)

Implementation of Revised School Immunization Requirements and Impact on Immunization Coverage in New York State       In 2014, the New York State Department of Health (NYSDOH) established new immunization requirements for day care, nursery school, prekindergarten and school entry and attendance in order to better conform to Advisory Committee on Immunization Practices (ACIP) recommendations and schedules. The new requirements took effect on July 1, 2014. Principal among the new requirements were: 1) an increase in the required number of doses of diphtheria, tetanus, and pertussis-containing (DTaP) vaccines from a three doses to an age-appropriate number of doses as established by the ACIP childhood immunization schedule; 2) a multi-year rollout of an increase in the required number of doses of polio vaccine from a three doses to an age-appropriate number of doses as established by the ACIP childhood immunization schedule; 3) a multi-year rollout of an increase in the required number of doses of varicella vaccine from a single dose to two doses; 4) a requirement that all doses be valid according to the minimum intervals established by the ACIP; and 5) a requirement that students in process of completing the required immunization series complete them at the intervals established by the ACIP catch-up schedule. Preliminary kindergarten vaccination coverage demonstrates that NYS was able to achieve high kindergarten coverage levels with age-appropriate doses of DTaP vaccine (97.5%) and with two doses of varicella vaccine (96.4%) in the 2014-2015 school year, however additional regulatory changes were necessary in order to fully implement the new requirements. This presentation will discuss lessons learned from the implementation of the new NYS school immunization requirements and will present final NYS prekindergarten through grade twelve school immunization coverage and exemption data for the 2014-2015 school year.

 

Rhee, Jongeun

A Case-Control Study of Multiple Occupational Exposures and Lung Cancer Risk in the Northeastern USA Xiaomei Liao, PhD

David Christiani, MD, MPH   Objective: To investigate the relationships between occupational exposures to multiple substances and the risk of lung cancer in a Northeastern US population.   Methods: A hospital based case-control study was conducted from 1992 to 2012. The risk of lung cancer in relation to exposure to multiple occupational substances was examined in two different ways: total carcinogenic score based on the classifications by IARC and total number of occupational exposures (substances) reported. The association between occupational exposures and lung cancer was modeled, adjusting for smoking status and second hand smoking history.   Results: There was an increased risk of lung cancer in relation to total carcinogenic score (OR=1.17, 95% CI: 1.01 1.35, per unit). Overall, the exposure of an increasing number of suspected lung carcinogens was significantly associated with positive trend in risk (p-value= 0.05). Exposure to traffic exhaust was a significant risk factor of lung cancer (OR=1.89, 95% CI: 1.12 3.18).   Conclusions: There was an increased risk of lung cancer with increasing occupational exposure to total lung carcinogens as measured by a cumulative score, and with increasing number of occupational exposures to known or suspected carcinogens.    Key words: Occupational exposures, lung cancer, carcinogen score"

 

 

Robyn, Misha

DVM1,2,     Q Fever Outbreak among Travelers to Germany Linked to Live Cell Therapy — United States and Canada, 2014     Alexandra P. Newman, DVM2, Michael Amato, MPH3, Mary Walawander3, Cynthia Kothe4, James D. Nerone4, Cynthia Pomerantz4, Casey Barton Behravesh, DVM5, Holly M. Biggs, MD1,5, F. Scott Dahlgren5, Emily G. Pieracci, DVM1,5, Yvonne Whitfield, MPH6, Doug Sider, MD6, Omar Ozaldin, MSc7; Lisa Berger, MD7; Peter A. Buck, DVM8, Mark Downing, MD9,10, Debra Blog, MD2 (Author affiliations at end of text)    Background: During September–November 2014, the New York State Department of Health (NYSDOH) was notified of five residents who had tested seropositive for Coxiella burnetii, the bacterium that causes Q fever. All had symptoms compatible with Q fever and traveled to Germany during May 2014 for live cell therapy, an alternative medicine practice, unavailable in the United States, of injecting animal cells into humans. Testing was prompted after a Canadian resident with a similar history received a Q fever diagnosis; German authorities identified a Q fever-positive sheep flock as the cells source. We investigated to characterize the outbreak.  Methods: A case was defined as a person who received live cell therapy in Germany during May, experienced signs and symptoms compatible with Q fever, and a single IgG titer ≥1:128 to C. burnetii phase II antigen. Patients were interviewed and NYSDOH notified the Centers for Disease Control and Prevention. Clinical and exposure information was reported to Germany.  Results: Patients received sheep cell injections by the same physician. Median age was 61 years; three (60%) were female. Signs and symptoms began ~1–7 days after exposure and lasted ~10–90 days; 80% reported fever, sweats, and fatigue, and 60% reported headache, chills, and malaise. None were hospitalized. No additional cases were identified.   Conclusions: Epidemiologic evidence indicates live cell therapy was the likely source of this outbreak. This is the first Q fever outbreak associated with live cell therapy, although other serious adverse events have been reported. Clinicians should be aware of this practice and consider zoonotic disease potential among patients receiving live cell therapy.     1Epidemic Intelligence Service, CDC; 2New York State Department of Health; 3Erie County Department of Health; 4Ulster County Department of Health and Mental Health; 5Division of Vector-Borne Diseases, National Center for Emerging and Zoonotic Infectious Diseases, CDC; 6Public Health Ontario; 7Toronto Public Health; 8Centre for Food-borne, Environmental and Zoonotic Infectious Diseases, Public Health Agency of Canada; 9Saint Joseph’s Health Centre, Toronto, Ontario; 10Department of Medicine, University of Toronto

 

Roche, Lisa (Roche_CH_1045.pdf)

Disparities in female breast cancer stage at diagnosis - a spatial-temporal analysis   Xiaoling Niu, MS   Antoinette M. Stroup, PhD

Kevin A. Henry, PhD     Background:  Despite improvements in breast cancer stage at diagnosis, mortality and survival, disparities persist. Methods:  We used a space-time scan statistic in SaTScan to identify estimated geographic areas and time periods with significantly high proportions of female breast cancer diagnosed at the distant stage and in-situ separately. Results:  One estimated cluster (Cluster 1) of distant stage breast cancer was identified in northeastern New Jersey during 1997-2011 (RR=1.35, p<0.001). Two estimated clusters (Clusters 2 and 3) of in-situ breast cancer were found in northeastern New Jersey during 2004-2011 and in central New Jersey during 2006-2011 (RRs=1.35, 1.24 respectively, p<0.001). Cluster 1 contains relatively high percentages of minority and low socioeconomic status (SES) breast cancer cases and populations. Clusters 2 and 3 have relatively high percentages of non-minority and high SES breast cancer cases and populations.  Conclusion:  While there have been improvements, disparities continue. These study results can be used to target interventions."

Ross, Kathleen   Patel, Ami (Ross_ID_145.pdf)

Substance Abuse Behaviors among Newly Reported Hepatitis C Virus Positive Youth – January 2012 through June 2013 in Philadelphia    Kathleen Ross1, 2, Ami Patel PhD MPH2, 3  1 Department of Epidemiology and Biostatistics, School of Public Health, Drexel University, Philadelphia, PA, USA  2Acute Communicable Disease Program, Division of Disease Control, Philadelphia Department of Public Health, Philadelphia, PA, USA  3Centers for Disease Control and Prevention, Atlanta, GA, USA    ABSTRACT  Background – The Centers for Disease Control and Prevention has reported that in 2012, the incidence of HCV increased in every age group compared to earlier years with the largest increase in individuals <30 years of age. Injection drug use (IDU) has been identified as a primary risk factor for acquiring Hepatitis C Virus (HCV) infection among this age group. To better identify risk factors associated with HCV infection among youth, the Philadelphia Department of Public Health (PDPH) initiated an enhanced surveillance system. This study describes general epidemiology characteristics and substance abuse behaviors of HCV positive youth in Philadelphia.     Methods – Cases were defined as individuals enrolled into the enhanced surveillance system between the ages of 13-30 years of age with a positive HCV test result from January 1, 2012 to June 30, 2013, who were newly reported to PDPH. Demographic, clinical, and risk factor data were collected via a phone interview of the case. Cases were asked about substance abuse ever or within the six months prior to their positive test. Cases reporting a history of injection drugs were also asked additional questions related to introduction of injection drugs, type of drugs used, frequency of use, needle and syringe use, and who the case injects with. Utilizing SAS 9.3, descriptive analyses and logistic regression models were performed.    Results – Total, 269 cases were included in this study. Cases were predominately white (72%), non-Hispanic (80%), and evenly distributed among males (47%) and females (53%). Ninety-three percent of cases were ≥ 19 years of age. History of substance abuse was reported by 174 of 215 cases (81%) and of these individuals, 128 (74%) reported injection drug use. The mean initiation age of injection drug use is 21 years old. White cases were more likely (OR 8.12; CI, 3.99-16.45) than non-white cases to report a history of injection drug use. Non-Hispanic cases were more likely (OR 3.70; CI, 2.09-6.54) than Hispanic cases to report a history of injection drug use. Cases reporting a history of injection drugs were also asked questions related to their substance abuse behaviors. Eighty-three percent reported heroin as the first drug ever injected and 83% reported heroin as the drug currently used most often. Males were more likely (OR 2.14; CI, 1.01-4.55) than females to report being introduced to injection drugs by friends and females were more likely (OR 3.22; CI, 1.19-8.74) than males to report being introduced to injection drugs by a sex partner. Cases 19-25 years of age were more likely (OR 4.47; CI, 1.35-14.78) than cases 26-30 years of age to report injecting mostly with friends. Cases 26-30 years of age were more likely (OR 9.9; CI, 2.29-43.00) than cases 19-25 years of age to report injecting mostly with a sex partner. While only 57% of IDUs reported ever sharing a needle or syringe with another individual, 87% reported reusing their own needle or syringe and 87% reported sharing drug preparation equipment.     Conclusions – The results suggest that gender and age group are associated with substance abuse behaviors. There were significant differences observed between gender and age groups, in regard to who introduced the case to injection drugs and who the case injects with most frequently. The results also suggest a lack of injection safety knowledge. Recommendations based off the findings from this study call for greater harm-reduction efforts, specifically regarding injection safety. Additionally, there is a need to create social network level interventions instead of solely focusing on the individual.    KEYWORDS  Hepatitis C virus; substance abuse; injection drug use; youth

 

 

Saffa, Alhaji (Saffa_ID_1045.pdf)

Active Monitoring for Ebola Virus Disease — New York City, 2014–2015  Anna Tate, MPH  Ifeoma Ezeoke, MPHBackground   The ongoing Ebola virus disease (Ebola) outbreak in West Africa has primarily affected Guinea, Liberia, and Sierra Leone, resulting in >25,000 Ebola cases with >10,000 deaths. Imported Ebola cases have also occurred in the United States, including New York City (NYC). To facilitate early recognition of Ebola cases, the NYC Department of Health and Mental Hygiene (DOHMH) actively monitors persons who have risk factors for exposure to Ebola virus such as persons who have treated patients with Ebola or persons who have recently travelled from an Ebola-affected country. Active monitoring requires daily communication by public health personnel with persons who have potentially been exposed to an infectious agent to check whether an illness has occurred. Here we describe the DOHMH experience with active monitoring.      Methods  DOHMH receives notification on persons who require active monitoring for Ebola (clients) through communication with the Centers for Disease Control and Prevention or with NYC-based healthcare facilities or through other DOHMH investigations. Clients undergoing active monitoring for Ebola are asked to make contact via telephone with DOHMH once a day for 21 days after their last potential Ebola virus exposure (the incubation period of the Ebola virus.) If clients are not reached on their final day of their monitoring period, additional attempts are made to contact clients the following day; if these additional attempts are unsuccessful, clients are considered lost to follow-up. During the daily phone contact, clients report two temperature readings and whether they have experienced any vomiting, diarrhea, or unexplained bleeding or bruising. Additional monitoring measures (involving twice daily contact with clients, one of which entails direct visualization such as through video conferencing technology) are applied for clients who have engaged in activities that put them at higher risk for an Ebola virus exposure (such as providing direct care for patients in an Ebola-affected country.) Clients who experience any of the aforementioned symptoms or who report a fever (temperature ≥100.0°F) are referred to a medical epidemiologist for further evaluation over the phone. Clients are reminded each day to call DOHMH if they develop an illness or if they go to a hospital or emergency room. These same procedures are applied to persons with a potential exposure to Lassa virus.    Results  During February 1, 2015–July 20, 2015, DOHMH monitored 2161 clients, of which 2125 (98.3%) were considered to be at low (but not zero) risk for Ebola and 53.3% were male. During the peak of monitoring, >300 clients were monitored per day. The age distributions of clients were as follows: 3.3% were aged <5 years, 6.1% were aged 5–19 years, 86.7% were aged 20–64 years, and 3.9% were aged ≥65 years. The indication for active monitoring for all clients was because of recent travel to an Ebola-affected country: 1323 clients reported travel to Guinea, 566 to Liberia, and 439 to Sierra Leone (some clients traveled to >1 country; monitoring of persons who traveled to Liberia ceased on June 17.) During this period, four clients were lost to follow-up. Thirty clients (1.4%) reported fever, six reported diarrhea, four reported vomiting, and 0 reported unexplained bleeding or bruising, representing a total of 35 distinct clients. 1 presented to a healthcare facility without contacting DOHMH as she had been instructed to do. During May–June, 2015, one additional client who had not traveled to an Ebola-affected country underwent active monitoring because of a potential exposure to Lassa virus (but not Ebola virus.)    Discussion  The DOHMH active monitoring operation has monitored a large volume of clients. The vast majority of clients were deemed to be in the low (but not zero) risk category and remained under active monitoring for the recommended duration. The high proportion of ill clients who presented to a healthcare facility without also contacting DOHMH serves as a reminder for healthcare providers to take thorough travel histories when evaluating patients. Because active monitoring for Ebola will continue for the foreseeable future, and because active monitoring as described here could possibly be applied to infectious diseases other than Ebola and Lassa fever, additional data are needed on ways to promote compliance with active monitoring procedures, to understand the psychosocial impacts of active monitoring and to improve efficiency of active monitoring.

 

 

Sanderson, Jennifer

Increasing the Efficiency and Yield of a Tuberculosis Contact Investigation through Electronic Data Systems Matching – New York City, 2013   Lisa Trieu, MPH

Douglas Proops, MD MPH Shama Desai Ahuja, PhD MPH    Background: Transmission of tuberculosis (TB) in healthcare settings remains a risk to the health of patients and healthcare workers (HCWs). When a person with TB disease exposes others in a healthcare setting, infection preventionists at the facility are expected to lead a contact investigation (CI) to identify contacts. These investigations can require extensive resources and detailed record review; therefore, electronic health data can be a useful tool to improve the timeliness and accuracy of healthcare-associated CIs.       Methods: In 09/2013, the New York City Bureau of TB Control (BTBC) and Hospital A initiated a CI around a HCW with infectious TB who worked in a maternity ward. We employed two data systems in the response: hospital-based electronic health records (EHRs) to identify exposed patients, and an electronic immunization registry (EIR) to obtain contact information for exposed infants and their healthcare providers. During 10 months of follow-up, two matches were conducted with the EIR to capture changes in contact information. To assess the impact of EHRs we calculated the proportion of patients identified as exposed among those in the maternity ward, and to evaluate the EIR we quantified the amount of new information from this registry.           Results: 954 patients received care in the maternity ward during the HCW’s infectious period. Review of EHRs narrowed TB exposure to 145 mothers and 140 infants with ≥30 minutes of HCW interaction (30%). Over 99% of exposed infants were identified in the EIR data matches. The first EIR linkage yielded a new telephone and/or resident address for 49% of infants and a different provider for 52% of infants compared with data obtained from the EHR. Matching with the EIR a second time provided updated provider information for 30% of infants. During 10 months of follow-up, most mothers (59%) and infants (80%) had evaluation results reported to BTBC.                Conclusions: Electronic data matching improved the efficiency and yield of this CI. Hospital EHRs helped identify patients with direct TB exposure, enabling the concentration of resources on those at greatest risk for TB infection. Also, gathering accurate contact information from the EIR allowed us to successfully locate and notify patients. These findings underscore the importance of exploring how electronic health data can improve epidemiologic investigations."

 

Soliva, Susan, MPH Kathleen Fitzsimmons, MPH

Background:  The Massachusetts Virtual Epidemiologic Network (MAVEN) was deployed in 2006 by the Massachusetts Department of Public Health (MDPH) to serve as an integrated, Web-based disease infectious disease surveillance and case management system. It captures data used for case investigation and follow-up by MDPH epidemiologists and local boards of health (LBOH). The system captures information on approximately 90 reportable infectious diseases, and produces close to 150,000 disease reports annually.  Since its inception, MAVEN has included an occupation reference code list used by state epidemiologists and LBOH for demographic and risk purposes, but there was concern that the list was not comprehensive enough and did not correspond to standard coding systems, such as the Bureau of Labor Statistics Standard Occupational Classification (SOC).  Use of SOC would enable comparison of occupations within MAVEN and across data sets. The goal of this effort was to revise and expand the occupation pick list in MAVEN using occupation categories based on the SOC.  We aimed to keep the list to a length and format that would be useful and not overwhelming to those collecting the data.    Methods and Findings:  First, using data on workforce distribution from the Current Population Survey, we identified the most frequent occupations and industries among Massachusetts residents. We then used the existing MAVEN reference code list and cross-walked it to the SOC.  The final list contained more health care, food service and animal related occupations than the previous list in MAVEN and used occupations at different levels of the SOC coding structure since there were instances where there was not a one-to-one match and there were categories where more specificity would not serve a public health purpose (for example, administrative and office support).  The updated list was first integrated into MAVEN for the Foodborne Illness module and subsequently into the General Epidemiology, Tuberculosis, and Sexually Transmitted Disease modules in May 2015. Data using the first three months of the new occupation list will be analyzed in August to identify the most common occupations seen in a variety of diseases under surveillance. Distributions will be compared to previous years to identify any differences and evaluate the usefulness of the new occupation list.       Conclusion:  Using a collaborative approach, MDPH occupational health, MDPH Bureau of Infectious Disease staff successfully updated the occupation list in MAVEN and standardized it across diseases to allow for easier data entry and more accurate data collection.  Standardization will enable a better comparison of occupations within MAVEN and across MDPH data sets.  More complete and accurate occupation data will help to inform efforts to identify risk groups and prevent disease.

 

 

***** Soliva, Susan

Tackling Hepatitis C Surveillance Challenges Using MAVEN, an Electronic Surveillance and Case Management System    BACKGROUND:    The Massachusetts Department of Public Health (MDPH) receives approximately 8,000 newly reported cases of Hepatitis C (HCV) cases annually.  This represents a significant surveillance challenge in terms of collecting, triaging and investigating cases.   MDPH uses an electronic surveillance and case management system, the Massachusetts Virtual Epidemiologic Network (MAVEN).  MAVEN utilizes workflows to identify potential cases of disease (events) based on a determined set of criteria.  Once identified, specific actions will be taken, such as sending out a case report form (CRF), assigning the event to a local health department for investigation, or reviewing clinical and laboratory information for case classification. In 2015, the Division of Epidemiology and Immunization collaborated with the Office of Integrated Surveillance and Informatics Services to more accurately identify and investigate suspect acute HCV events through expanded use of MAVEN to electronically triage incoming HCV reports.    METHODS:  All case triage processes are managed through MAVEN workflows. MDPH mails a standard, one-page HCV CRF to the ordering provider for completion on all newly reported cases who are under 15 years of age or between the ages of 30-69 years.  This form collects basic clinical and risk history information.  Newly reported cases between 15 and 29 years or 70 years and older are pre-identified as suspect acute cases, and providers are mailed a two page Acute HCV CRF.  Reminders are sent if no response is received within a month.     The Acute HCV Identification workflow identifies suspect acute HCV cases less than 15 years of age or over 29 years of age who may require further investigation.  Individuals enter this workflow if the event date is within the past year and at least one of the following four criteria are met: evidence of seroconversion, jaundice or gastroenteritis or an elevated ALT value (>400).  Events will also enter this workflow if they are reported via an electronic medical record feed Massachusetts has implemented with select providers in the state to report only acute HCV cases.      Epidemiologists will review all events in this workflow to identify events that require further follow-up and assign to a local health department as needed.      The Acute HCV Pending Investigation workflow acts as an intermediate location for all events assigned to local health departments for further investigation.      The Acute HCV Final Review workflow captures acute HCV cases not yet assigned a surveillance case classification status (confirmed, suspect, revoked) for whom an investigation has been completed, either by a local health department assigned the case or a provider who completed and returned an Acute HCV Case Report Form.      Due to the potential risk of healthcare-associated transmission in the 70 years and older age group, if no CRF is received (no response within 60 days of the second mail out), events will enter the Suspect Acute Hepatitis C Events (70 years and older) with no Acute CRF Completed workflow to be reviewed by an epidemiologist and assigned out to a local health department for further follow-up if necessary.     RESULTS:  All of the workflows described were successfully implemented and have been operational since April 2015.  There are currently 57 events in the identification workflow, 59 pending acute hepatitis C investigations and 96 cases in the final review workflow to be assigned a status.      CONCLUSIONS:  Expanded use of workflows in MAVEN has allowed for more efficient identification and tracking of suspect acute HCV events.  The team plans to compare the overall number of acute HCV cases identified prior to and after the workflow revisions were implemented to see if more confirmed acute cases were identified.  The team is also planning to compare completeness of clinical and risk history data for acute cases identified before and after the workflow revisions were implemented.

 

******Soliva, Susan, MPH Kathleen Fitzsimmons, MPH

Background:  The Massachusetts Virtual Epidemiologic Network (MAVEN) was deployed in 2006 by the Massachusetts Department of Public Health (MDPH) to serve as an integrated, Web-based disease infectious disease surveillance and case management system. It captures data used for case investigation and follow-up by MDPH epidemiologists and local boards of health (LBOH). The system captures information on approximately 90 reportable infectious diseases, and produces close to 150,000 disease reports annually.  Since its inception, MAVEN has included an occupation reference code list used by state epidemiologists and LBOH for demographic and risk purposes, but there was concern that the list was not comprehensive enough and did not correspond to standard coding systems, such as the Bureau of Labor Statistics Standard Occupational Classification (SOC).  Use of SOC would enable comparison of occupations within MAVEN and across data sets. The goal of this effort was to revise and expand the occupation pick list in MAVEN using occupation categories based on the SOC.  We aimed to keep the list to a length and format that would be useful and not overwhelming to those collecting the data.    Methods and Findings:  First, using data on workforce distribution from the Current Population Survey, we identified the most frequent occupations and industries among Massachusetts residents. We then used the existing MAVEN reference code list and cross-walked it to the SOC.  The final list contained more health care, food service and animal related occupations than the previous list in MAVEN and used occupations at different levels of the SOC coding structure since there were instances where there was not a one-to-one match and there were categories where more specificity would not serve a public health purpose (for example, administrative and office support).  The updated list was first integrated into MAVEN for the Foodborne Illness module and subsequently into the General Epidemiology, Tuberculosis, and Sexually Transmitted Disease modules in May 2015. Data using the first three months of the new occupation list will be analyzed in August to identify the most common occupations seen in a variety of diseases under surveillance. Distributions will be compared to previous years to identify any differences and evaluate the usefulness of the new occupation list.       Conclusion:  Using a collaborative approach, MDPH occupational health, MDPH Bureau of Infectious Disease staff successfully updated the occupation list in MAVEN and standardized it across diseases to allow for easier data entry and more accurate data collection.  Standardization will enable a better comparison of occupations within MAVEN and across MDPH data sets.  More complete and accurate occupation data will help to inform efforts to identify risk groups and prevent disease.

 

Taggert, Elizabeth

VARICELLA-ZOSTER VIRUS AND STROKE IN AN IMMUNOCOMPROMISED PATIENT: A CASE REPORT    Background: Primary varicella-zoster virus (VZV) infection causes varicella while VZV reactivation causes herpes zoster later in life. Although these infections are typically self-limiting, serious complications can occur, including pneumonia, encephalitis, and vasculitis. VZV is a risk factor for vasculopathies including stroke.  Infectious etiology of vascular events should be considered in immunocompromised patients. We report a case of VZV vasculitis as the reason for a bilateral stroke in an immunocompromised individual that was identified through case-based public health surveillance data.      Case report:  A 63-year-old black, non-Hispanic female presented to a local Emergency Department (ED) complaining of a vesicular rash on her left flank for four days, pain for 1-2 weeks, and shortness of breath and cough for a few days. She was discharged with the diagnoses of herpes zoster and Chronic Obstructive Pulmonary Disease (COPD). No antiviral treatment was prescribed at this time. Six weeks later, the patient underwent a partial nephrectomy due to a recent diagnosis of renal cell carcinoma. The following week, the patient presented to another local ED complaining of left side weakness and difficulty speaking.  A computerized tomography (CT) scan and magnetic resonance imaging (MRI) of the head showed hypoattenuation involving the left basal ganglia and hypodensities bilaterally. She was admitted to the stroke unit, experienced seizures and was intubated. A second MRI showed worsening infarctions. Given the recent history of herpes zoster, a lumbar puncture was performed three days later to evaluate for possible infectious causes and intravenous acyclovir was empirically prescribed. Cerebrospinal fluid tested positive for VZV by Polymerase Chain Reaction and positive for VZV IgG antibody. Serum also tested positive for VZV IgG antibody (7.62; >1.10 is considered positive).  During her hospitalization, she was newly diagnosed with HIV, with a CD4 count of 133. With no improvement one week later, she was discharged to hospice. Three days later, the patient expired.      Discussion: This case highlights the potential for unusually severe complications of VZV infections particularly in immunocompromised individuals. Although rare, VZV may explain some vascular events. Unlike other viruses, VZV can replicate in cerebral vessels. VZV-related strokes have been previously documented in the literature.  In 2010, Ciccone et al. reported a similar case where a 5-year-old had a stroke 3 months after VZV reactivation.  In 2009, Kang et al. described the increased risk of stroke in the year following a herpes zoster infection in a population- based study in Taiwan. Vascular events may not be an immediate complication of VZV, but may trigger a slow progression of complications even after the rash resolves.  From 2006-2014 in Philadelphia, there were 2880 varicella and 5063 herpes zoster cases, of which, 58 (2%) varicella and 204 (4%) herpes zoster cases were hospitalized. The underlying medical conditions, immunocompromised status and reasons for hospitalization of these cases will be explored further. Continued research is needed to better understand the prevalence of these complications.

 

Talbot, Elizabeth    Tammer, Roza

I can see clearly now, the yeast is gone: Investigating a cluster of Candida interface infections and positive donor tissue cultures following corneal transplant surgery  Edmund Tsui, MD Michael E. Zegans, MD  Erin Fogel, MD  Katrina Hansen, MPH Elizabeth R. Daly, MPH James Noble, MD  Lynda Caine, MPH  Jessa Fogel  Shifts in procedures used to address corneal endothelial failure have contributed to greater usage of eye bank-prepared, pre-cut tissue in the US. Certain characteristics of a newer, more widely used procedure called Descemet’s Stripping Automated Endothelial Keratoplasty [DSAEK], including the usage of such tissue, may introduce increased risk for fungal infections in the recipient eye. The use of corneoscleral rim cultures prior to DSAEK has been adopted by some surgeons, but is not a standard practice. We describe two confirmed cases of Candida spp. infection and two presumptively treated patients following a cluster of seven positive rim cultures out of 99 pre-cut corneas used by a single surgeon over a three-year period. We discuss the public health investigation, current literature, and highlight important considerations regarding clinical and public health practice associated with a cluster of fungal eye infections and positive rim cultures. Patient microbiota, environmental or instrument contamination, transport media contamination, and tissue bank preparation and storage practices may be potential sources of Candida spp..  However, it is also possible that donor corneal tissue prepared specifically for DSAEK was contaminated, because there were no fungal infections associated with other corneal transplant procedures performed by this surgeon over the same time period, and because the positive corneoscleral rim cultures matched the organism found in two of the infections. We describe several interventions that may better describe the true rate, and reduce risk, of post-operative fungal eye infections, including routine donor rim cultures, supplementation of storage media with antifungal agents, careful monitoring of donor tissue warming times, and appropriate engagement with public health and Federal regulatory agencies leading to improved surveillance, investigation, and policy change."

 

Thomas, Pauline (Thomas_CH_1045.pdf)

Emergency department (ED) and hospitalization use for chronic obstructive pulmonary disease (COPD) or Asthma in older adults following Hurricane Sandy, New Jersey, 2012     Authors:  Mangala Rajan, Pauline Thomas, Amy Davidow, Christina Tan    Background and Purpose:   Individuals with chronic conditions requiring ongoing management with medications and electrical medical equipment are particularly vulnerable during environmental disruption from severe weather events, such as Hurricane Sandy, which made landfall in New Jersey (NJ) on October 29, 2012.  This study examines ED utilization and hospitalization for respiratory complications (asthma and COPD) among NJ adults older than 60 following Hurricane Sandy.  Methods:  NJ Hospital Discharge Data Collection System (NJDDCS) data on inpatient admissions and ED visits for adults older than 60 years were obtained for the years 2008-2012.  Data included patient demographics, admission/discharge dates, and ICD9 coded diagnoses.   Asthma- and COPD- related ED visits and hospitalizations were defined using the Preventable Quality Indicator #5 ICD9-based definition from the Agency of Healthcare Research and Quality.  ED visits and hospitalizations were examined separately. US Census 2010 data were used to generate rates. Level of impact was based on a composite score using duration of power outage, extent of residential and commercial municipal damage, and FEMA municipal assistance. Sandy Period was defined as October 28-December 28, 2012 (2 month period starting the day prior to storm landfall), and the pre-hurricane period was December 29, 2011–October 27, 2012.  To account for seasonality, Sandy Period ED and hospitalization rates were compared to pre-hurricane rates in 2012, and to the same two periods in 2008-2011 combined.  Municipalities where > 40% of households were poor or asset constrained were classified as having a low socioeconomic status (SES).  Associations between rates, years, and SES, by Sandy Period were tested using Poisson regression.  Results:  In 2012 there were 23,878 ED visits plus hospitalizations for asthma and COPD in adults over 60.  The yearly rate of ED visits for asthma and COPD was 44.1 per 100,000, significantly higher than 33.9 per 100,000 in 2008 to 2011 combined.  The hospitalization rate was actually slightly lower in 2012 at 80.1 vs. 83.6 per 100,000 for the other years. An increase occurred in both ED visits and hospitalizations during the Sandy Period. The pre-hurricane ED visit rate in 2012 was 42 vs. 57 per 100,000 during the Sandy Period.  For the prior 4 years, the rate of ED visits was 34 per 100,000 for both time periods (October 28-December 28, and December 29-October 27).  For hospitalizations, in 2012 the rate rose from 77 to 94 per 100,000.  In the prior years, for the two time periods, the rates were 84 and 81 per 100,000.   Comparing areas with high vs. low hurricane impact, rates in pre and post hurricane time periods were similar in 2012, with the post-hurricane increase present even in low impact areas.   In 2012, adults aged >75 had higher hospitalization than those aged 61-74 (115 vs. 62 per 100,000); however, ED rates were comparable (45 vs. 43 per 100,000).  Women had higher rates than men (hospitalization: 87 vs. 72 per 100,000, ED: 47 vs. 41). African Americans had higher rates than whites or Hispanics at 119 per 100,000 for hospitalization and 97 for ED visits.  Rates of ED visits and hospitalization for respiratory illness varied dramatically by SES.  In 2012, lower SES areas had a hospitalization rate of 99 per 100,000 and an ED visit rate of 66 per 100,000, compared with 64 and 30 in the higher SES areas.  After adjusting for SES, the Sandy Period increases were still significant in 2012.   Conclusions  Hurricane Sandy had a significant effect on management of respiratory health of older adults in NJ with increases in the rates of ED visits and hospitalizations not seen in earlier years. Older women in poorer and minority neighborhoods are especially vulnerable. Primary care providers and others caring for older adults with respiratory compromise should consider disaster response planning in their care maintenance visits.

 

 

Toprani, Amita   Hadler, James (Toprani_CC_145.pdf)

Trends in Mortality Disparities by Area-based Poverty in New York City, 1990–2010    Background  Area-based poverty has consistently been associated with higher mortality rates but the relationship between poverty and mortality can change over time. We examine the relationship between neighborhood poverty and mortality in New York City (NYC) from 1990 to 2010 to document trends in mortality disparities over time and determine the causes of death for which disparities are currently greatest.    Methods  We used NYC and New York state mortality data for years 1990, 2000 and 2010 to calculate all-cause and cause-specific age-adjusted death rates (AADR) by the proportion of persons in a census tract living below the federal poverty threshold, or census tract poverty (CTP) in each year. We calculated mortality disparities, measured as the difference in AADR between the lowest and highest CTP groups, within and between race/ethnicity, nativity and sex categories.    Results  We observed higher all-cause AADRs with higher CTP for each year for all race/ethnicities, both sexes and the US-born. Mortality disparities decreased progressively from 1990 to 2010 overall, for each race/ethnic group, and for most causes of death. The difference in AADR between the highest and lowest CTP groups decreased from 5.65 per 1,000 population in 1990 to 2.55 in 2010. The mortality disparity decreased most among non-Hispanic whites (from 8.48 in 1990 to 2.16 in 2010), followed by Hispanics (4.73 to 1.66), non-Hispanic blacks (6.91 to 4.23) and non-Hispanic Asians (from 3.02 in 2000 to 1.08 in 2010). Among the US born, the difference in AADR between the highest and lowest poverty groups decreased from 6.41 in 1990 to 2.93 in 2010. Between 1990 and 2010, the disparity between high and low poverty groups lessened for all top 10 causes of death except ‘diabetes mellitus’ and ‘essential hypertension and renal disease.’ The largest contributors to mortality disparities in 2010 were heart disease (0.49 deaths per 1000 population), HIV (0.22) and diabetes (0.18), together accounting for 35% of the total.     Conclusions  Progress was made in narrowing socioeconomic disparities in mortality from 1990 to 2010 but substantial disparities remain. Future efforts toward achieving health equity in mortality in NYC should focus on the areas contributing most to current disparities.

Tran, Olivia (Tran_CC_1045.pdf)

Matching Infectious Disease Surveillance Data to Identify Syndemics in New York City (2000–2013)    Jyotsna Ramachandran Li Chen, MPH  Jennifer Fuld, PhD    Background: HIV, tuberculosis (TB), sexually transmitted diseases (STDs), and viral hepatitis continue to burden the U.S. population. Co-infection with multiple pathogens can occur because of shared risk factors for exposure and because, in some cases, infection with one pathogen increases susceptibility to other infections. Some infections may also increase the risk of non-infectious co-morbidities and vice-versa, such as diabetes. Matching surveillance data across different disease registries provides an opportunity to understand syndemics in which ≥2 diseases interact synergistically to increase illness and death. In 2014, as part of the Centers for Disease Control and Prevention’s Program Collaboration and Service Integration initiative, the New York City (NYC) Department of Health and Mental Hygiene (DOHMH) conducted its second data match to improve identification of syndemics, promote collaboration across public health programs, and better target delivery of integrated services. We describe the matching methodology used most recently by DOHMH.  Methods: The 2014 match of surveillance data included HIV, TB, STDs, hepatitis B (HBV), and hepatitis C (HCV) reports during 2000–2013, vital statistics mortality data during 2000–2013 and hemoglobin A1C reports during 2006-2013. An additional 30 communicable diseases were also matched. Separate event- and person-level datasets from each disease registry were cleaned and de-duplicated before the match. A deterministic method was used to link records across multiple diseases registries. Matches were established using 14 keys comprised of combinations of first name, last name, date of birth, and social security number, with the higher numbered keys having more flexibility. Following the match, discordance in static demographic variables between disease datasets was addressed and geospatial data, based on address at report, were included to explore distribution of diseases by area and neighborhood-level poverty. A final parent table, including static demographic variables for all individuals who did and did not match, was created. Each record was de-identified and assigned a unique ID. Multiple events for acute diseases were included, and a relational database was created.  Results: The final dataset contained nearly 1.7 million individuals with a report of ≥1 condition, of which approximately 13% had a report of ≥2 conditions. Separation of event- and person-level data allowed us to conduct a more streamlined match while still maintaining data on multiple or repeated diagnoses. Due to the complex nature of the data and the potential recurrence of acute diseases, the implementation of a simplified parent table allowed for more efficient querying of data. In addition, standardization of static variables and implementation of a hierarchy to address discordance allowed for improved data integration and usability of matched data.  Discussion: During the 2014 match, we added more recent years of data, streamlined our matching methodology, implemented a parent table to improve database efficiency and included 30 additional communicable diseases. Identifying new and efficient ways to link surveillance data will allow for more timely identification of syndemics. The lessons learned and methods employed by NYC DOHMH may be useful to other jurisdictions interested in data matching.

 

 

Troppy, Scott (Troppy_ID_345 B.pdf)

Collaboration Creating and Utilizing a new Foodborne Illness Complaint Module    Gillian Haney, MPH  Susan Soliva, MPH  Johanna Vostok, MPH  Emily Harvey, MPH

   BACKGROUND:  In 2014, the Massachusetts Department of Public Health (MDPH) Bureau of Infectious Disease (BID) Office of Integrated Surveillance and Informatics (ISIS), in collaboration with the MDPH Bureau of Environmental Health Food Protection Program (FPP), created a new foodborne illness (FBI) complaint module in the Massachusetts Virtual Epidemiologic Network (MAVEN), the state’s integrated, web-based disease surveillance and case management system for infectious diseases. The goals of the new module were to integrate an existing, legacy Microsoft Access database used by FPP and BID into MAVEN to streamline the reporting of complaints, improve communication between state agencies, and promote collaboration with local public health partners. To achieve these goals, the module needed to serve three purposes: 1) act as a repository for consumer complaints regarding potential sources of foodborne illness; 2) link confirmed cases of disease to corresponding environmental inspections and outcomes, and; 3) house meeting minutes of the Working Group on Foodborne Illness Control (WGFIC), a collaboration of epidemiology, environmental, and laboratory staff working to investigate foodborne outbreaks.    .    METHODS: The FBI module consists of four question packages of administrative data, complainant information, food history, and WGFIC meeting minutes. The module allows tracking by facility and retail type for monthly and annual reports. Consumer complaints are entered by the FPP, BID, and local boards of health (LBOH).  Complaints are reviewed by FPP staff to determine which may require further investigation by the WGFIC.  Workflows were created to allow for efficient notification to FPP of new complaints and tracking of events being monitored by WGFIC.  Reports were created to quickly aggregate complaint data entered into the FBI module, as well as to generate printable WGFIC meeting minutes. An online training was created for BID epidemiologists, FPP inspectors, and LBOH followed by in-person trainings by ISIS.     RESULTS: In 2014, 38 multiple and 347 single foodborne complaints were entered into MAVEN by LBOH, BID, and FPP staff.    CONCLUSIONS: The FBI module was successfully implemented allowing for more efficient tracking of FBI complaints and linking laboratory-confirmed cases to corresponding complaints., Training for epidemiologists, laboratorians, inspectors, and LBOH partners increased knowledge and communication around tracking and managing foodborne complaints. The module allows an electronic linkage of confirmed foodborne illness cases from the MAVEN surveillance module to a FBI complaint."

 

 

Troppy, Scott (Troppy_ID_2_345 B.pdf)

Novel use of a surveillance system to enhance communication with general public and local boards of health    Gillian Haney, MPH  Molly Crocket, MPH

 MaryKate Martelon, MPH  Susan Soliva, MPH  BACKGROUND:   The Massachusetts Department of Public Health (MDPH), Bureau of Infectious Diseases (BID) utilized legacy and siloed MS Access database to manage calls from the general public, and contact information for staff responsible for case investigation and follow-up of infectious diseases in 351 cities/towns. In order to integrate these activities with infectious disease surveillance and follow-up activities, the BID Office of Integrated Surveillance and Informatics (ISIS) enhanced the capacity of our existing surveillance system, the Massachusetts Virtual Epidemiologic Network (MAVEN), in 2009, 2013 and 2014.     METHODS: In response to database failure due to the large volume of calls during the 2009 H1N1 pandemic, ISIS migrated existing variables and added new questions and workflows to MAVEN to manage all calls made to the main phone line at MDPH BID. Each call is now logged into MAVEN and given a unique event ID, and programmatic assignment to a workflow for General Epidemiology, Vaccine Preventable Disease Epidemiology or ISIS.  Once assigned to a workflow, appropriate staff respond to the call and document the outcome. A Communication (COM) event was created in 2013 to support dissemination of a surveillance Key Indicator Report for each city/town. In 2014, a set of questions were developed for COM events to track contact information for MDPH and LBOH staff, with allowance for local editing. This new question package allows MDPH staff to locate appropriate LBOH staff for disease follow-up and case investigation direction.     RESULTS:  In 2013, there were 6,569 and in 2014 there were 6,860+ “On-Call events”, questions from LBOH or from the general public and clinicians, created for the three programmatic areas. The epidemiologists responding to infectious disease events now use the LBOH question package for identifying public health personnel, contact information, office hours for LBOH and Visiting Nurses Association (VNA) coverage information. The web-based system can be accessed after business hours and on weekends by on-call epidemiologists for communicating with LBOH.    CONCLUSIONS:  Integrating On-Call and Communication capacity within MAVEN has allowed MDPH to retire aging databases and improve communication between MDPH and LBOH and with the general public. The LBOH now have the ability to update their contact and communication information directly and in real-time."

 

 

Tsai, Stella (Tsai_CC_1045.pdf)

Development and Application of Syndromic Surveillance for Severe Weather Events in New Jersey Following Hurricane Sandy   Teresa Hamby, MSPH

Alvin Chu, EdD   Hui Gu, MS Jessie Gleason, PhD Gabrielle Goodrow Jerald Fagliano, PhD  Hurricane ‘Superstorm’ Sandy struck New Jersey on October 29, 2012, causing harm to the health of New Jersey residents and billions of dollars of damage to businesses, transportation, and infrastructure. Monitoring health outcomes for increased illness and injury due to a severe weather event is important in measuring the severity of conditions and the efficacy of state response, as well as in emergency response preparations for future severe weather events. Following the experience with Hurricane Sandy and the foreseeable need to be prepared for future severe weather events, NJDOH initiated a project to develop a suite of more than twenty indicators in EpiCenter, an online system which collects emergency department chief complaint data in real-time, to perform syndromic surveillance of extreme weather–related conditions.    The development of each indicator followed a two-stage process for keyword refinement using diagnostic codes with chief complaint information. The statistical measures of sensitivity, specificity, and positive predictive value were computed for both the initial keyword list and the final keyword list.    Indicators for health concerns related to Sandy were evaluated against the same one-month, three-month, and one-year periods from the following year (October 2013-2014). Peaks are accurately captured for most conditions including three disrupted outpatient medical care outcomes (dialysis, medication refills, and oxygen needs), carbon monoxide poisoning, asthma, anxiety and adjustment disorders, and substance abuse (methadone/opiate/heroin related), in the month following Hurricane Sandy. For some conditions, longer surveillance and/or restriction to areas most impacted is necessary to see the impact of the storm.  Overall, this endeavor has provided NJDOH with a clearer picture of the effects of Hurricane Sandy and has yielded valuable information on how the state should prepare to monitor the effects of the next severe weather event."

 

 

Umukoro,  Peter (Umukoro_EO_345.pdf)

Long-term Metal PM2.5 Exposures Decrease Cardiac Acceleration and Deceleration Capacities in Welders    Tianteng, Fan, MD ScD Jinming, Zhang, MD MPH

Jennifer, Cavallari, ScD Shona, Fang, ScD   Chensheng, Lu, PhD   Xihong, Lin, PhD  Murray, Mittleman, MD DrPH  Background: Studies have documented associations between short-term particulate exposures (PM2.5) and heart rate variability (HRV), both in the general population and among occupational groups. However, there are limitations in using HRV as an outcome in these studies because of its drawback in failing to account for differences in heart rate. Acceleration Capacity (AC) and Deceleration Capacity (DC) which measure the variability during speeding up and slowing down of the heart rate respectively, address this limitation. The associations between AC and DC with long-term exposure to fine particulates especially in occupational group settings are unknown.  Objective: To investigate the associations between AC and DC with long-term occupational PM2.5 exposures.  Methods: A panel of fifty (50) male boilermaker welders, mean age 39 years, had their PM2.5 exposure measured during 3-6 typical welding work shifts within sampling periods between January 2010 and June 2012. Chronic exposure indices (CEI) of particulates were then calculated in retrospect over the work life of the study participants to date. Resting electrocardiograms (ECG) over ten minutes were also recorded using a Holter monitor and analyzed in the time domain to obtain the AC and DC. Linear regression was used to assess the association between CEI PM2.5 exposure and each of AC and DC, controlling for age, acute effects of welding exposure, and time of day when ECG reading was obtained.  Results: Mean (SD) CEI for PM2.5 exposure was 1.25 (1.77) mg/m3-years and ranged from 0.01 – 10.4 mg/m3-years. Negative crude exposure-response associations were found for both AC and DC with increased chronic PM2.5 exposure. On adjusting for age, a 1 mg/m3-year increase in CEI for PM2.5 was associated with a decrease of 1.46 (95% CI:  0.23, 2.69) msec resting AC, and a decrease of 0.99 (95% CI: -0.08, 2.06) msec resting DC.  Conclusion: Long-term metal particulate exposures decrease cardiac accelerations and decelerations."

 

Wang, Xiaohong (Michelle)

Increasing Adequate Treatment of Gonorrhea Patients in New York State  Candace Noonan-Toly, BA  Salvatore Currenti, MPH  Alison Muse, MPH  Hwa-Gan Chang, PHD   Objective  Delay the emergence of antibiotic resistant gonorrhea by increasing the percentage of patients in New York State (NYS), excluding New York City (NYC), receiving adequate gonorrhea treatment.   Introduction  In NYS (excluding NYC) greater than 13,000 cases of gonorrhea were reported in 2013 and 2014 via the Communicable Disease Electronic Surveillance System (CDESS).  From November 2013 through May 2014, only 61% of cases were adequately treated with a regimen recommended by the Centers for Disease Control and Prevention (CDC) STD Treatment Guidelines for gonorrhea1, and 29% were missing treatment information. From November 2014 through May 2015, Local Health Departments (LHDs) were offered a performance incentive to increase the percentage of adequately treated cases.  Monthly reports of gonorrhea cases and treatment data from CDESS were posted to help LHDs monitor case treatment status and data completeness.    Methods  For the performance incentive, adequate treatment was defined as those cases with confirmed gonococcal infection that were treated in accordance with a regimen recommended by the CDC STD Treatment Guidelines.  To assess LHD performance, a tool was developed in CDESS to allow LHDs to download cases including patient demographic information, correct and complete treatment type, dosage and dates of administration by using SAS v 9.3 to create datasets and exporting them into comma separated value (CSV) files with one record per patient. These CSV files were uploaded to the CDESS Oracle database using a character large object (CLOB).  LHDs used these files to identify cases with inadequate treatment and conduct patient and provider follow up to promote appropriate treatment.  In addition, LHDs acted upon incomplete records to improve surveillance documentation of gonorrhea treatment data.    Results  From November 2013 through May 2014, 61% of cases received adequate treatment and 29% had missing treatment information. From November 2014 through May 2015, 92% of cases received adequate treatment with only 2.6% missing data.  Reasons for inadequate treatment included treatment with only one of two recommended antibiotics, with an antibiotic not included in recommended regimens, or incorrect dosage. Twenty percent fewer patients were considered to be inadequately treated because they received only one treatment in 2014-2015 than in 2013-2014.     Conclusions  Based on surveillance data, the percentage of adequately treated gonorrhea patients increased by 31% during the performance incentive period. The completeness of surveillance data on gonorrhea treatment also improved by 26%.   Improvements in LHD performance led to improved patient outcomes.  Maintaining these successful outcomes beyond the incentive period will be critical to preventing antibiotic resistant gonorrhea.   References  1 Centers for Disease Control and Prevention. Update to CDC’s Sexually Transmitted Diseases Treatment Guidelines 2010: Oral Cephalosporins No Longer a Recommended Treatment for Gonococcal Infections. MMWR 2012 / Vol. 61 / No. 31: 590-94."

 

 

Wenger, Mary Beth

New York State Counties With High Potentially-Avoidable Antibiotic Prescribing for Upper Respiratory Infections: Identification and Intervention"  Introduction:  The Centers for Disease Control and Prevention (CDC) reports that each year in the United States at least 2 million people are infected with antibiotic-resistant bacteria leading to at least 23,000 deaths. Inappropriate use of antibiotics over the years has rendered some antibiotics ineffective against certain bacteria. The outpatient setting is a key focus for educational efforts on appropriate use of antibiotics. Studies have shown high rates of inappropriate antibiotic prescribing when patients are given a diagnosis of ‘cold’, ‘acute upper respiratory infection’ or ‘acute bronchitis’.    Evidence is limited on how to affect antibiotic prescribing in the outpatient setting. Some studies support healthcare provider (HCP) audit and feedback as an effective mechanism to elicit change in practice behaviors. This CDC funded project by the New York State Department of Health aims to affect outpatient antibiotic prescribing practices utilizing a form of state-level audit and feedback by providing HCPs with county level data on antibiotic prescribing practices for a specific clinical scenario, along with promoting CDC Get Smart: Know When Antibiotics Work educational materials.     Methods:  New York State Department of Health (NYSDOH) analyzed New York State Medicaid recipient administrative claims and prescription data to identify outpatient antibiotic prescribing trends from January 1, 2013 to December 31, 2013. Analysis included persons aged 3 months to 64 years of age who received a primary diagnosis of acute upper respiratory infection (URI) by ICD-9 code in an outpatient setting and subsequently filled a prescription for an antibiotic within four days. Initial analysis indicated higher rates of potentially avoidable prescriptions among adults than children. Indirect standardization was used to risk adjust based on age, principal diagnosis and visit type (emergency department (ED), institutional outpatient, or professional outpatient). County level rates were established based on the zip codes of provider practices. Those with the highest rates were deemed as having the highest potentially avoidable antibiotic prescribing, and providers practicing in these counties were targeted for intervention. A “Dear Provider” letter signed by the NYSDOH Commissioner of Health and a map identifying county-level rate categories were sent to primary care, family practice, internal medicine and emergency medicine HCPs who would be expected to see adult patients in outpatient settings. A second mailing was sent within a week and included Get Smart: Know When Antibiotics Work educational materials, including a viral prescription pad.     Results:  Among 150,379 eligible visits by NYS Medicaid enrollees aged 18 to 64 who were diagnosed with an acute URI in an outpatient setting, 25 to 64% of patients filled an antibiotic prescription within four days of their visit. Among the 62 counties in NYS, eleven counties were identified as having the highest rates (55 to 64%) of antibiotic prescriptions filled. The “Dear Provider” letter and accompanying map were sent to over 2,800 providers in these eleven counties. In addition to highlighting the issue of high rates of potentially avoidable antibiotic prescribing for acute URI, providers were enlisted to become provider “champions” or “standard-bearers” for antibiotic resistance education among their local peers.     Conclusions:  This intervention is in progress, however initial interest in the program has been high. Ongoing evaluation of the impact of this intervention will continue in the setting of additional collaborations and several local HCPs who have expressed interest in the program. Outreach to local health departments to notify them of the intervention has generated multiple opportunities for partnership. External partner opportunities have arisen with a private insurance company. These data will also serve as a baseline for comparison in the future as this public health intervention continues.    Authors:  Sarah Kogut, MPH, CIC  Mary Beth Wenger, ABJ

 

 

Wheeler-Martin, Katie (Wheeler-martin_LB.pdf)

Outbreak of C. perfringens at a summer camp in Dutchess County, NY - July 2015  Background:  A summer educational camp for high school students reported an outbreak of gastrointestinal illness among over 70 participants and staff on July 22, 2015.  The camp took place at a college in Dutchess County.  No other cases of illness were reported among students, staff, or campers at other summer programs taking place at the college at the same time.  Methods:  Dutchess County Department of Health (DCDOH) staff visited the college and conducted a preliminary interview with camp staff.  Given the short time frame of the outbreak, a point-source exposure was strongly suspected.  It was learned that a catered meal was shared on the evening of July 21st between the hours of 4:30-6:30 pm.  The catered meal was not consumed by any other programs at the college.  Remaining samples of the meal were prepared for lab testing, two stool specimens were collected from ill participants, and a restaurant inspection was conducted at the catering premises. DCDOH also administered a food history and illness questionnaire with camp participants.       Results:  121 camp students and staff completed the questionnaire.  A total of 72 respondents met a case definition of diarrhea with onset in the seven days prior to the survey; vomiting was also reported by a small number of cases (15%).  The earliest reported onset of diarrhea was around 7:00 pm on July 21st and the latest was 8:00 pm on July 23rd, with peak onset between 6:00 am and 12:00 noon on July 22nd (53%).  Among participants who consumed the catered dinner on July 21st, the attack rate was 67%.  The median incubation period was estimated to be 12 hours from the time of the suspect source meal.  Menu items consisted of chicken, beef, pork, oxtail, and fish, with sides of plain rice, rice with peas, red beans, and plantains.  Water and iced tea were also served.  Specific attack rates were high for nearly all items consumed from the dinner menu, with the highest individual attack rate ratios for roast pork (1.67, n=20 ill), fish (1.49, n=1 ill), red beans (1.37, n=14 ill), and plain rice (1.35, n=43 ill).  Combined attack rate ratios were 4.22 for any meat (n=71 ill) and 1.61 for any rice (n=69 ill), however very few individuals consumed meat without rice (n=7) or rice without meat (n=6).   Meals were served in individual containers with meat and rice touching, which were reportedly plated at 3:30 pm.  On 8/14/2015, DCDOH learned that the two stool specimens were positive for C. perfringens at a significant level; earlier test results were negative for norovirus.  Laboratory results on the food samples are pending.  The restaurant inspection identified several violations related to food storage and handling.  Conclusion:   The findings strongly suggest that the outbreak of diarrhea was due to exposure to C. perfringens from a catered meal served on July 21, 2015

Wright-Woolcock, Stacey  Adolphe, Camille Lee  (8_Wrightwoolcock_RF[1].pdf)

Ellen Partnering with Travel Agents to Develop Health Education Outreach among New York City Residents Visiting Friends and Relatives Abroad, 2013-2014    Stacey Wright-Woolcock1, Stefan Hagmann2, Stefanie Erskine3, Emily Jentes3, Edward T. Ryan4, Regina LaRocque4, Ana-Maria Emeh2, Kimberly Valcin1, Camille Adolphe1, Lucretia Jones1, Catherine Dentinger1, Ellen H. Lee1    1Bureau of Communicable Disease, New York City Department of Health and Mental Hygiene, Queens, New York  2Division of Pediatric Infectious Diseases, Bronx Lebanon Hospital Center, Bronx, New York; and Department of Pediatrics, Albert Einstein College of Medicine, Bronx, New York  3Division of Global Migration and Quarantine, Centers for Disease Control and Prevention, Atlanta, Georgia  4Travelers' Advice and Immunization Center, Massachusetts General Hospital and Harvard Medical School, Boston, Massachussetts    BACKGROUND: New York City (NYC) residents traveling overseas to visit friends and relatives (VFR) account for many cases of preventable diseases, including malaria and typhoid, that are diagnosed upon their return to NYC.  The NYC Department of Health and Mental Hygiene (DOHMH) Bureau of Communicable Disease and partners (Bronx-Lebanon Hospital [BXL], Massachusetts General Hospital, and Centers for Disease Control and Prevention) recruited NYC travel agencies serving VFRs for participation in developing an initiative aimed at improving travelers’ health.         METHODS: We identified 122 NYC travel agencies via internet search, contacted them by telephone, and made in-person visits to recruit participants. To elicit ideas for a collaboration promoting health among VFR travelers, we administered a questionnaire to travel agents expressing interest and invited them to a training/focus group, sponsored by DOHMH, regarding VFR traveler health.    RESULTS: We reached 71 travel agencies by telephone; 32 initially expressed interest. Of these, 15 travel agents scheduled in-person visits with DOHMH and BXL staff, and 12 completed questionnaires. Seven agents attended the training/focus group.  From the questionnaire, all 12 agents reported that their transactions for each VFR client involved at least two interactions, either by telephone or in-person visit.  All seven attending the training/focus group reported they would use an online application that provides customized travel health information and offer health information to clients during interactions, unless the information was perceived to potentially discourage travel. Agents suggested a successful initiative would include a media campaign, outreach to travel industry leaders and trade shows, and creating a trademark to recognize travel agencies participating in the initiative.      CONCLUSION: Travel agents working with VFR travelers described repeated contact with their VFR clients. Successful collaboration with travel agents on public health initiatives could entail periodic contact from the health department to provide updates on international public health news, a media campaign, and formal recognition, such as a certificate or plaque, for the efforts of the travel agents.