Trends and Socioeconomic Inequalities in PFQ-Defined Functional Limitation Among U.S. Adults: NHANES 2005-2018.

BACKGROUND

Difficulty with physical functioning is relevant to disability surveillance and rehabilitation planning, but national trends and socioeconomic inequalities in higher-burden PFQ-defined functional limitations remain incompletely characterised.

METHODS

We analysed seven 2-year cycles of NHANES (2005-2018) among adults aged ≥ 20 years. The primary outcome was PFQ-defined functional limitation using a prespecified operational threshold of ≥ 2 difficulties across 20 physical functioning items. Survey-weighted logistic regression estimated per-2-year change (odds ratios [ORs]) overall and within the poverty-income ratio (PIR) and education strata; Wald tests evaluated time × SES interactions. Because the outcome was common, log-link models additionally estimated prevalence ratios (PRs). Supplementary analyses examined alternative thresholds, multiple imputation for missing PIR, alternative handling of response 5 ('do not do this activity'), categorical survey cycle with adjusted prevalence estimates, and PFQ063 as supportive descriptive evidence among participants reporting limitation.

RESULTS

The weighted prevalence of PFQ-defined functional limitation increased from 22.1% in 2005-2006 to 30.1% in 2017-2018. Each 2-year cycle was associated with higher odds (OR 1.08, 95% CI 1.05-1.12) and higher prevalence (PR 1.05, 95% CI 1.03-1.08) of limitation. Socioeconomic gradients persisted throughout the study period. Evidence for differential trends was limited overall, with little heterogeneity on the OR scale, modest heterogeneity by PIR on the PR scale, and minimal heterogeneity by education. In multiple-imputation analyses, results were materially similar to the complete-case findings (OR per cycle 1.093 vs. 1.092; PR 1.058 vs. 1.059). Among participants reporting any limitation, musculoskeletal-related causes identified by PFQ063 were associated with a higher likelihood of meeting the ≥ 2-difficulty threshold (OR 2.64, 95% CI 2.35-2.96; PR 1.19, 95% CI 1.16-1.21), which should be interpreted as supportive descriptive evidence rather than validation.

CONCLUSIONS

PFQ-defined functional limitation increased modestly from 2005 to 2018 and remained socially patterned among U.S. adults. These findings suggest a rising population burden of more severe self-reported functional limitations, while the consistency of supplementary analyses supports the robustness of the overall interpretation.