Related Experiment Videos
Non-participation and mortality in a prospective study of cardiovascular disease
M Walker1, A G Shaper, D G Cook
1Department of Clinical Epidemiology and General Practice, Royal Free Hospital School of Medicine, London.
Insights
Men not joining a cardiovascular disease study had higher early mortality, linked to social factors, not age. However, cardiovascular disease death rates were similar, suggesting non-participation bias is minimal for this outcome.
Area of Science:
- Epidemiology
- Public Health
- Cardiovascular Disease Research
Background:
- Prospective studies are crucial for understanding disease etiology.
- Participant selection can introduce bias, affecting generalizability.
- The British Regional Heart Study investigated cardiovascular disease in men.
Purpose of the Study:
- To assess the characteristics of men who did not participate in the British Regional Heart Study.
- To compare the mortality rates of participants and non-participants.
- To determine if non-participation bias affects cardiovascular disease outcome analysis.
Main Methods:
- Comparison of demographic and occupational data between participants and non-participants.
- Analysis of total mortality rates during the follow-up period.
- Stratified analysis of mortality by cause of death (neoplasms, cardiovascular disease, other causes).
Main Results:
- Non-participants were younger, more likely unmarried, and less skilled workers.
- Non-participants exhibited significantly higher total mortality in the initial follow-up years, which later normalized.
- Excess early mortality in non-participants was not attributable to age but linked to social characteristics and other causes of death.
- Cardiovascular disease mortality rates were similar between participants and non-participants.
Conclusions:
- Social characteristics of non-participants significantly influenced their overall mortality.
- The study's findings on cardiovascular disease mortality are unlikely to be biased by non-participation.
- Adjustments for social factors are necessary when interpreting overall mortality in this cohort.
Abstract:
Men who did not participate in a prospective study of cardiovascular disease (The British Regional Heart Study) were younger than the participants, more likely to be unmarried, and more likely to be less skilled workers. In the first three years of follow-up, their total mortality rate was significantly higher than that of the participants; thereafter it declined to levels not significantly different from those of the participants. This excess of early deaths could not be attributed to age. There was a small but non significant excess mortality in non-participants due to neoplasms and cardiovascular disease and a somewhat greater excess from all other causes combined. The social characteristics of the non-participant population appear to contribute to their significantly higher total mortality rate, and allowance needs to be made for this in interpreting the study findings. However the death rate from cardiovascular disease was similar in participants and non-participants, suggesting that any analysis related to this particular cause of death should not be biased by non-participation.