NYC HANES 2013-2014: Accomplishments, Challenges, and Lessons Learned From Conducting a Population Health Surveillance Study

Background: The New York City Health and Nutrition Examination Survey (NYC HANES), conducted in 2004 and again in 2013-2014, was modeled after the National Health and Nutrition Examination Survey (NHANES). Both NYC HANES provide valuable citywide data on prevalence, awareness, treatment, and control of chronic conditions; environmental toxin exposure; and mental health screening. Blood, urine, and saliva specimens allow for objective measures of disease and risk factors, as well as for a repository for future research. We describe lessons learned from the second NYC HANES, which differed from the first by collecting data primarily in participants’ homes (versus in clinics), and by subcontracting with a vendor to collect data for most of the study (versus NYC Health Department staff in 2004).

Methods: NYC HANES was a population-based, cross-sectional survey of NYC adults, with a three-stage cluster sampling design, and a questionnaire, physical exam, and specimen collection. Interviews were conducted in English, Spanish, Russian, and Chinese, with telephone translation for other languages. Portable centrifuges and overnight shipping helped ensure timely processing of biological specimens. Participants were compensated $100, increased to $200 the last two weeks of the study. Outreach efforts included key messages stressing both personal and civic benefits, a website and video, a Facebook page, contacting building management, neighborhood blog postings and live presentations, and endorsements from official and community leaders.

Results: Final sample size was 1545 participants, with a response rate of 36%. A total of 1236 (82%) participants completed the study at home, and 1230 provided blood samples (80%). The increased incentive improved response: 78 interviews (5.1% of the total) were completed during the last 12 days of the study, compared with 121 (7.8%) the previous 30 days. Overall response rate was below that achieved in 2004 in spite of the modified study design and innovative outreach. Possible explanations include difficulty gaining access to locked buildings, lower acceptance of unsolicited home visits, greater reluctance toward surveys, remote supervision of data collection, and insufficient incentives.

Conclusions: Lessons from the second NYC HANES should be useful to others planning field surveys or biospecimen collection. While declines in response rates observed in NYC HANES mirror those of surveys nationwide, participants responded positively to the home interviews and specimen collection. To reach targeted sample sizes, though, researchers must be increasingly resourceful.