Abstract P268: Survival Advantage of Participating in Cohort Studies Varies Over Time

Abstract

Background: Volunteers in cohort studies are reported to have lower mortality (survival advantage) over non-participants. However, it is unclear if the early survival advantage in study participants compared to non-participants remains over time. Hypothesis: Cohort participation will be associated with survival advantage that diminishes over time. Methods: We had mortality follow-up among people who were covered by the Washington County private census (90,286, about 92% of all 94,881 Washington County residents). This census data was used to compare volunteers in 3 cohort studies (ARIC, CLUE1 & 2) to non-participants at baseline and with respect to subsequent mortality adjust for demographics. Individual death data between 1975 and 2015 was obtained from Maryland Vital Statistics and National Death Index and analyzed using Kaplan-Meier graph. Cox regression models were used to estimate the hazard ratios for mortality comparing participants to non-participants in 5-year time periods since enrollment in each cohort. Results: The Washington County Private Census included 2,933 participants and 18,196 non-participants in ARIC study (recruited 1986-1989; born 1922-1943); 19,912 participants and 52,973 non-participants in CLUE1 (recruited 1974, born in 1877-1963); and 14,408 participants and 44,356 non-participants in CLUE2 (recruited 1989; born in 1893-1961). Participants in each study were more likely to be female, older, married, more educated and never smokers relative to non-participants. Cohort participation was associated with lower mortality as compared to non-participation during the first five years in all studies; this survival advantage diminished over time and after very long follow-up (>20 years) participants actually had higher mortality than non-participants (Figure). Conclusion: In conclusion, the survival advantage associated with cohort participation was strongest in the first 5 years after study enrollment and diminished overtime and even reversed with very long follow-up.