Journal of the American Medical Directors Association · Oct 29, 2026

Care-Need Level Worsening After Hip Fracture Surgery in Kyoto, Japan: Population-Based Cohort Study

By Sasaki K, Takahashi Y, Nishikawa Y et al.

OBJECTIVES: To describe 3-year transitions of care-need level after hip fracture surgery and identify factors associated with subsequent nonfatal worsening among day-180 survivors. DESIGN: Retrospective cohort study using linked administrative medical and long-term care insurance claims. SETTING AND PARTICIPANTS: Adults aged ≥75 years in Kyoto City, Japan, who underwent hip fracture surgery between 2013 and 2018. The day-180 landmark cohort included 6556 patients who survived to day 180 and were not certified as care-need level 5 before surgery or at the landmark. METHODS: Follow-up started at day 180. The primary estimand was subsequent nonfatal care-need level worsening among patients alive and still at risk at the landmark; death after the landmark was treated as a competing event. We estimated cumulative incidence functions and Fine-Gray models with covariate-by-log-time interactions. Supplementary analyses examined a death-inclusive composite end point and alternative landmarks. RESULTS: The cumulative incidence of worsening was 10.6% at 12 months, 28.3% at 24 months, and 37.4% at 36 months after surgery. Longer initial claims-based rehabilitation continuity during the postoperative 6 months was associated with a lower incidence of subsequent nonfatal worsening (adjusted subdistribution hazard ratio per month, 0.961; 95% CI, 0.941-0.982). The supplementary composite end point analysis showed no clear association between rehabilitation continuity and worsening or death (hazard ratio, 0.988; 95% CI, 0.972-1.006). Associations of higher Hospital Frailty Risk Score and preoperative care-need levels with worsening varied over time. CONCLUSIONS AND IMPLICATIONS: Among day-180 survivors after hip fracture surgery, preoperative frailty and care-need levels showed time-varying associations with subsequent nonfatal worsening. Longer initial claims-based rehabilitation continuity was associated with lower incidence of nonfatal worsening but should be interpreted as a claims-based association, not evidence of a causal effect or mortality-inclusive benefit. These findings support ongoing risk assessment and periodic reassessment of care needs beyond the early postoperative period.

doi.org/10.1016/j.jamda.2026.106413PubMed 42956453Populations