Cut-point shift and index shift in self-reported health.

Area of Science:

• Health Economics
• Biostatistics
• Health Services Research

Background:

• Concerns exist regarding differential reporting in health surveys, potentially invalidating comparisons and health inequality measures.
• Reporting heterogeneity across population subgroups can compromise the accuracy of health assessments.

Purpose of the Study:

• To propose and illustrate a statistical test for differential reporting in ordered response models.
• To differentiate between cut-point shift and index shift in health reporting.
• To assess reporting heterogeneity in Canadian health data.

Main Methods:

• Development of a novel statistical test for differential reporting in ordered response models.
• Application of the test to Canadian National Population Health Survey data.
• Comparison of self-assessed health (five-point scale) with the McMaster Health Utility Index Mark 3 (HUI3).

Main Results:

• The proposed method successfully distinguished between cut-point shift and index shift.
• Clear evidence of index shifting and cut-point shifting was found for age and gender subgroups.
• No significant evidence of differential reporting was detected for income, education, or language subgroups.

Conclusions:

• Differential reporting, specifically index and cut-point shifting, significantly impacts health measures for certain demographic groups (age, gender).
• The developed statistical test is valuable for identifying and understanding reporting heterogeneity in health surveys.
• Findings highlight the need to account for reporting differences to ensure accurate health inequality assessments.