Polling for public health in New Jersey

Background: Surveillance, which underlies the Ten Essential Public Health Services, is the collection, analysis, and dissemination of health-related information for the purpose of prevention. Surveillance has evolved over a half-century from focusing on counting reports of diseases to a broader focus on hazards and effective interventions. To assess the feasibility of collecting state-based information on health to guide public health action, we explored affordable ways to collect periodic state-wide data.   At Rutgers University, the Department of Preventive Medicine and Community Health at New Jersey Medical School partnered with the Eagleton Center for Public Interest Polling to create the Rutgers-Eagleton Public Health Series (REPHS).

Method: The REPHS is a subset of questions in the Rutgers-Eagleton Poll, a periodic random-digit-dialed telephone survey of New Jersey residents.  Our first REPHS took place from July 28- August 5, 2014.  Health questions were added onto a regularly planned political poll. Respondents were asked 21 health-related questions on topics including distracted driving, driving while drowsy, knowledge and access of poison control, and the placement of home carbon monoxide monitors. The cost of asking these additional questions was $10,000.    

Results:  5487 respondents were reached by random-digit-dialing of whom 16% (N=871) agreed to participate. Preliminary analyses were completed within 5 days.  A series of press releases was then disseminated to local news outlets resulting in wide coverage of the information gathered.   Over 70 news articles were published citing the poll results through over 20 news outlets around the state.   The prevalence estimates that were collected were consistent with national surveys indicating a reliable method for the surveillance of health behaviors and knowledge. More detailed analyses of the data are ongoing and will result in public health policy recommendations and publications.

Discussion: Creative state-based partnerships can result in valuable and inexpensive surveillance data that can influence public health actions.  Limitations include the inability to provide precise estimates for sub-populations which may be at differential risk; however we could identify general trends.  Social desirability bias is another limit of telephone surveys which likely yielded conservative estimates of behavior.   The advantages of this approach include affordable cost, partnership with experts in state-based survey methods, quick dissemination of results, and the ability to re-survey with sufficient frequency to track trends.