Skip to main content

Table 1 Simulation studies: prevalence estimate by four methods

From: Prevalence estimation by joint use of big data and health survey: a demonstration study using electronic health records in New York city

True Population PrevalencePrevalence Estimate (95% CI)
Prevalence (p1) based on outcome in health survey (Y1)Prevalence (p2) based on outcome in EHR (Y2)Health Survey (n1 = 500)Post-stratified EHRMosteller estimatorSubject-level imputation estimator
0.30.300.3000.2990.3000.303
0.30.310.3000.3090.3030.302
0.30.320.2990.3190.3050.302
0.30.330.2980.3290.3050.303
0.30.350.3000.3490.3080.304
  1. The size of health survey (n1) and the size of subjects linked between two sources (n12) are both 500