Effects of Efforts to Increase Response Rates on a Workplace Chronic Condition Screening Survey

• Published in: Medical care 2002-09, Vol.40 (9), p.752-760
• Main Authors: Wang, Philip S., Beck, Arne L., McKenas, David K., Meneades, Laurie M., Pronk, Nicolaas P., Saylor, John S., Simon, Gregory E., Walters, Ellen E., Kessler, Ronald C.
• Format: Article
• Language: English
• Subjects: Absenteeism; Chi-Square Distribution; Chronic Disease - epidemiology; Chronic diseases; Employment; Female; Financial incentives; Health Surveys; Humans; Interpersonal Relations; Job performance; Mailings; Male; Medical conditions; Patient care; Prevalence; Response rates; Risk Assessment; Sampling Studies; Telephone interviews; Telephones; United States - epidemiology
• ISSN: 0025-7079; 1537-1948
• DOI: 10.1097/00005650-200209000-00005

Objective. Expanded health risk appraisal (HRA) surveys can help employers target chronic conditions for outreach or disease management interventions by providing data on the prevalences of conditions and their effects on work performance. However, concerns exist about the accuracy of this data because most HRAs have low response rates. We evaluated these concerns by examining the prevalences and work impairments associated with chronic conditions across four HRA subsamples that differed in intensity of recruitment effort. Methods. Two thousand five hundred thirty-nine working people were invited to complete an expanded HRA survey that included questions about chronic conditions, work impairments, and demographics. Condition prevalences and associations between conditions and work impairment were compared across subsamples who responded after a single mailing, after two mailings, and in a telephone interview after the mailings either with or without a $20 incentive. Results. Consistent with previous research, response rates varied dramatically across the four subsamples (from 20.1% with only one mailing to 67.7% with telephone administration and a financial incentive). However, the estimated prevalences of chronic conditions, levels of work impairment, and effects of chronic conditions on work impairment did not differ with intensity of recruitment effort. Conclusions. Expanded HRAs can provide useful data on the prevalences and work impairments associated with chronic conditions even if response rates are low. Confirmation of these results is required, however, in new samples. Additional research is also needed on innovative and cost-effective strategies to improve HRA response rates.