Variation in mortality patterns among the general population, study participants, and different types of nonparticipants: evidence from 25 years of follow-up.

The general willingness to participate in health surveys is decreasing, resulting in increasingly selected study populations. We aimed to examine relative mortality rates by different categories of nonparticipation. We included 14,223 men and women aged 25-74 years who were sampled in the Swiss centers of the international Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) Study (1983-1992) and followed-up until 2008. Logistic regression was used to describe factors associated with nonparticipation. We compared persons who refused to participate (n = 2,911), persons who did not go to their examination (n = 470), and persons that could not be contacted for inclusion in the MONICA Study (n = 989) with participants (n = 9,853). We used Cox regression to determine mortality hazard ratios for the various categories of nonparticipants versus participants or versus the 1990 Census population average. Persons who refused were more likely to be older and female. Compared with participants, those who refused had a higher risk of death (for men, hazard ratio = 1.34, 95% confidence interval: 1.17, 1.54; for women, hazard ratio = 1.31, 95% confidence interval: 1.12, 1.52), whereas persons who did not go to their examination and those who could not be contacted did not differ. Compared with the general (census) population, participants had a lower risk of all-cause mortality and those who refused had a higher risk of all-cause mortality. These variations were mainly due to cardiovascular disease and not cancer. Variations in relative mortality risks between study participants and the general population could depend on the proportion of subjects who actively refused to participate in the study. Considering the increasing participation reluctance, our findings underline the importance of carefully assessing the reasons for nonparticipation.