The participant effect: mortality in a community-based study compared to vital statistics.

The 20-year mortality experience of the community-based Evans County Heart Study population is compared to local, regional and national vital statistics. Deficit mortality occurred in the study population at younger ages while at older ages mortality was similar to or greater than vital statistics. This was particularly true for white and nonwhite males, whose mortality patterns were statistically significantly different from Evans Co. vital statistics (P less than 0.005). Nonwhite/white mortality ratios in the study were close to those observed in local vital statistics, particularly for males. Sex mortality ratios in the study population were lower than in vital statistics due to a stronger participant effect (lower mortality) in males. Evans Co. was an area of particularly high mortality for whites in the period 1960-1980 compared to other parts of Georgia and the U.S. Results of this study are similar to other reports of participant effects in epidemiologic follow-up studies; implications for bias in estimates of population levels of disease and of disease/exposure relationships are discussed.