Do high participation rates improve effects of population-based general health checks?

The aim is to investigate if the effect of a health check differs between areas with different participation rates. The Inter99 population-based randomized lifestyle intervention study covered 73 areas within the suburbs of Copenhagen, Denmark. Adults aged 30-60years were randomly drawn from a population and were randomized to intervention group (n=11,483) or control group (n=47,122). Participation rates in the health check varied considerably between areas (mean 52%; range 35-85%). In separate survival analyses, area participation rate was included both as a continuous exposure variable and as a categorical variable (tertiles; low: 35-49%, meddle: 50-54%, high: 55-84%). All persons in the intervention and control group were followed in registers for 10-year total mortality and combined events (ischemic heart disease, stroke, or both). In adjusted models (including sociodemographic variables, ethnicity, number of children and comorbidity), among men, there was no difference in risk of death between areas with varying participation rates. Surprisingly, among women living in high-participation areas a significantly higher risk of all-cause mortality (HR: 1.32 [1.03-1.69]) was found in the intervention group (ref=controls). For both men and women, in no areas there was any difference between intervention and control group in incident IHD/stroke. Higher participation rates in population based health checks is probably unlikely to improve the effects of these, and may in worst case be harmful in subgroups of the population. Further well-designed studies within non-participation research should have high priority and are required to establish link between health checks and risk of death in subgroups of the population.