Predicting risk among non-respondents in prospective studies

Area of Science:

• Rehabilitation Medicine
• Epidemiology
• Health Services Research

Background:

• Follow-up studies are crucial for assessing outcomes in chronically disabled patients.
• Non-response bias is a significant concern that can impact the validity of study findings.
• Previous research suggests response rates above 80% may still harbor bias.

Purpose of the Study:

• To investigate potential non-response bias in a follow-up study of chronically disabled patients.
• To assess the impact of non-response on employment status estimates.
• To explore methods for obtaining reliable estimates despite non-response.

Main Methods:

• A follow-up study of 2,011 chronically disabled patients was conducted.
• Response rates at 6-month and 1-year follow-ups were 82.5% and 73.3%, respectively.
• Employment status of non-respondents was obtained from indirect sources; response delay was analyzed.

Main Results:

• Non-respondents had a lower employment rate than respondents.
• Non-response was associated with age, social class, previous employment, and disability type, but not the outcome.
• Late respondents' employment rates mirrored non-respondents', suggesting bias even with high initial response.

Conclusions:

• Risk estimates can be biased even with response rates over 80%.
• Risk factor prevalence among non-respondents does not reliably indicate the extent of bias.
• Extrapolating respondent data based on response delay provides reliable estimates.