DETERMINING THE VALUE OF ADDITIONAL SURROGATE EXPOSURE DATA FOR IMPROVING THE ESTIMATE OF AN ODDS RATIO

We consider the design of both cohort and case-control studies in whic h an initial ('stage 1') sample of complete data on an error-free dise ase indicator (D), a correct ('gold-standard') dichotomous exposure me asurement (X) and an error-prone exposure measurement (Z) are availabl e, We calculate the amount of additional information on the odds ratio relating D to X that one can obtain from a second ('stage 2') sample of measurements only on D and Z, If one allows for differential measur ement error in Z, there is often little advantage in having more than four times as much data in stage 2 data as in stage 1. With the assump tion that a non-differential measurement error model is reasonable, la rger amounts of stage 2 data can be useful, Simulations indicate that stage 1 samples of modest size (50 cases in case-control studies and 5 0 failures in cohort studies) yield sufficiently reliable estimates of needed parameters to assist in determining an appropriate size for th e stage 2 sample, These ideas apply in settings either where the amoun t of stage 1 data is limited and fixed by external constraints or wher e one has gathered stage 1 data in advance to avoid collecting superfl uous stage 2 data.