Impact of differential item functioning on age and gender differences in functional disability

Area of Science:

• Gerontology
• Sociology
• Health Services Research

Background:

• Estimates of functional disability may be biased by differential item functioning (DIF).
• DIF occurs when individuals from different groups have varying probabilities of responding to an item, given their disability level.
• This bias can affect the accuracy of group comparisons in disability severity.

Purpose of the Study:

• To investigate the impact of DIF on estimates of age and gender differences in functional disability severity.
• To determine the extent to which DIF influences perceived disability levels across different demographic groups.
• To assess the need for DIF adjustments in disability research.

Main Methods:

• Utilized data from the 1994/1995 National Health Interview Survey Disability Supplement.
• Analyzed 5,750 adult respondents with functional limitations in activities of daily living or instrumental activities of daily living.
• Employed multiple-indicator/multiple-cause models to estimate disability as a latent trait, examining age (18-39, 40-69, 70+) and gender groups.

Main Results:

• Nine survey items exhibited significant DIF by age or gender, particularly for "shopping" and "money management."
• Unadjusted analyses indicated lower disability in middle-aged adults compared to the elderly, and in women compared to men among non-elderly.
• After adjusting for DIF, these age and gender disparities in disability severity became non-significant.

Conclusions:

• Comparisons of functional disability across sociodemographic groups necessitate accounting for DIF.
• Future research should explore the underlying causes of DIF in disability measures.
• Developing alternative question wordings to minimize DIF is recommended for improved accuracy.