Author/Authors :
Y.-W. Chiu، نويسنده , , LJ Su، نويسنده ,
Abstract :
Purpose
Differences between study participants and nonparticipants are a concern in epidemiologic studies. Study results may be biased if those participating are not representative of the reference population for whom the research in question is targeted. The purpose of this study is to compare demographic characteristics of individuals who choose or refuse to participate in the Louisiana Healthy Aging Study.
Methods
Louisiana Healthy Aging Study is a population-based study which examines the determinants of healthy aging in adulthood in eight Louisiana parishes. The Louisiana Voter Registration File (VRF) and Medicare Beneficiary Enrollment Data Files (CMS) were used to generate sampling populations of age 20 to 64 and age 65 and older, respectively. An introductory letter with a prepaid postcard was mailed to potential participants and a follow-up letter was sent after two weeks of lag time from initial attempt. Potential subjects who did not respond to the mailing were contacted by phone to check on their potential interest in study participation. Chi-square test and logistic regression were used to determine what factors were associated with those participating in the study.
Results
Overall, 1729 potential subjects have been identified. Participation rate declined with age (p < 0.001). 54.87 % of people age 20 to 64 years old participated in the study, compared with 41.83 % in age group 65 to 89 years old, and 27.76 % in age group 90 years and older. There is no significant difference in participation rates between female and male (40.84 % vs. 43.10 %) and there is also no difference between African Americans and Whites (38.96 % vs. 42.63 %).
Conclusion
Studies involving in-clinic examination may under-represent elderly, especially those with health and physical difficulties, even transportation is offered. However, gender and race do not seem to affect the decision to participate in research studies.