Effect of non-response bias on estimates of the numbers of patients with intractable diseases based on nationwide epidemiological surveys of Japan

Objectives: Estimates of the numbers of patients with intractable diseases in Japan use data from nationwide epidemiological surveys under the assumption that the mean number of patients among hospitals that respond to the survey is equal to that among hospitals that do not respond ( "the assumption of no response bias" ). We examine the validity of this assumption. Methods: Data from nationwide epidemiological surveys of eosinophilic granulomatosis with polyangiitis (EGPA) and familial Mediterranean fever (FMF) were used. Information from initial questionnaires and from second surveys, of hospitals that did not respond to the initial surveys, were combined in a statistical model to estimate rates of response. Results: The proportion of hospitals that had patients with EGPA and FMF was higher among hospitals that responded to the initial survey than among those that responded to the second survey. The ratio of the response rate for hospitals without affected patients to that for those with affected patients was estimated to be 0.86 for EGPA and 0.90 for FMF. The ratio of the number of patients estimated under the assumption of no response bias to that based on the estimated ratio of response rate between in hospitals with and without affected patients was 1.11 for EGPA and 1.09 for FMF. Conclusions: The number of patients with EGPA and FMF estimated under the assumption of no response bias was only about 10% greater than the true number, suggesting that the assumption was almost completely valid.

Abstract licence: CC BY

7 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.