Representative pure risk estimation by using data from epidemiologic studies, surveys, and registries: estimating risks for minority subgroups

Abstract Representative risk estimation is fundamental to clinical decision-making. However, risks are often estimated from non-representative epidemiologic studies, which usually under-represent minorities. Model-based methods use population registries to improve external validity of risk estimation but assume hazard ratios are generalisable from samples to the target finite population. ‘Pseudoweighting’ methods improve representativeness of studies by using an external probability-based survey as the reference, but the resulting estimators can be biased due to propensity model misspecification and inefficient due to highly variable pseudoweights or small sample sizes of minorities in the cohort and/or survey. We propose a two-step pseudoweighting procedure that post-stratifies the event rates among age/race/sex strata in the pseudoweighted cohort to the population rates, to produce efficient and robust pure risk estimation (i.e. a cause-specific absolute risk in the absence of competing events). For developing an all-cause mortality risk model representative for the USA, our findings suggest that hazard ratios for minorities are not generalisable, and that surveys can have inadequate numbers of events for minorities. Post-stratification on event rates is crucial for obtaining reliable risk estimation for minority subgroups.