# Exploring the association between Frailty Index and Knee osteoarthritis in middle-aged and older Chinese adults: A cross-sectional analysis of data from the China Health and Retirement Longitudinal Study

**Authors:** Na Zhu, Hongming Lin, Jia Yu, Rong Yang, Danhao Tao, Guoxin Huang, Da Qian, Yu Liu, Bingqian Chen

PMC · DOI: 10.1371/journal.pone.0343370 · PLOS One · 2026-03-27

## TL;DR

This study finds a strong link between frailty and knee osteoarthritis in older Chinese adults, suggesting targeted interventions may help.

## Contribution

The study reveals a nonlinear relationship between frailty and knee osteoarthritis using a large-scale Chinese dataset.

## Key findings

- Higher Frailty Index scores are strongly correlated with increased knee osteoarthritis prevalence.
- A significant nonlinear relationship exists between frailty and symptomatic knee osteoarthritis.
- Random Forest Machine model achieved high predictive accuracy (AUC 0.90).

## Abstract

To explore the relationship between Frailty Index (FI) and Knee osteoarthritis (KOA) in Chinese people aged 45 and above.

Data on symptomatic KOA, FI, and over ten covariates were from the China Health and Retirement Longitudinal Study (CHARLS). Logistic regression and restricted cubic splines (RCS) were used to analyze the correlation. Subgroup and interaction analyses were also conducted, and FI was divided into quartiles to assess result stability. Six machine learning constructs were used to build predictive models, which were then evaluated.

Among 29,105 participants, 5,012 had symptomatic KOA. Weighted multivariate logistic results showed a correlation between KOA and frailty (OR 3.63(3.48,3.78)), present across quartiles. Subgroup analyses revealed potential effects of various factors on their association, with significant interactions found for gender, education, place of residence, alcohol consumption, hypertension, and dyslipidemia. RCS results indicated a logarithmic non-linear relationship between symptomatic KOA and frailty. The area under the curve of the Random Forest Machine model is 0.90, and it performs best on the calibration curve and decision curve analysis.

This study showed a nonlinear relationship between FI and KOA in middle-aged and older Chinese adults, with higher FI scores linked to higher KOA prevalence and a significant inflection point, informing focused KOA interventions in aging populations.

## Linked entities

- **Diseases:** dyslipidemia (MONDO:0002525)

## Full-text entities

- **Genes:** INS (insulin) [NCBI Gene 3630] {aka IDDM, IDDM1, IDDM2, ILPR, IRDN, MODY10}, IL10 (interleukin 10) [NCBI Gene 3586] {aka CSIF, GVHDS, IL-10, IL10A, TGIF}, TNF (tumor necrosis factor) [NCBI Gene 7124] {aka DIF, IMD127, TNF-alpha, TNFA, TNFSF2, TNLG1F}, IL1A (interleukin 1 alpha) [NCBI Gene 3552] {aka IL-1 alpha, IL-1A, IL1, IL1-ALPHA, IL1F1}, CRP (C-reactive protein) [NCBI Gene 1401] {aka PTX1}, IL6 (interleukin 6) [NCBI Gene 3569] {aka BSF-2, BSF2, CDF, HGF, HSF, IFN-beta-2}
- **Diseases:** chronic pain (MESH:D059350), emphysema (MESH:D004646), underweight (MESH:D013851), Hypertension (MESH:D006973), Asthma (MESH:D001249), KOA (MESH:D020370), Chronic bronchitis (MESH:D029481), lipid abnormalities (MESH:D011017), Liver disease (MESH:D008107), osteoporosis (MESH:D010024), fractures (MESH:D050723), Synovial inflammation (MESH:D007249), Memory related disease (MESH:D008569), Estrogen deficiency (MESH:D056828), Dementia (MESH:D003704), hyperlipidemia (MESH:D006949), knee-joint degeneration (MESH:D000092443), loss of muscle mass (MESH:C536030), arthralgia (MESH:D018771), CHARLS (OMIM:603663), reduced muscle strength (MESH:D009135), cartilage damage (MESH:D002357), joint disease (MESH:D007592), brain atrophy (MESH:C566985), overweight (MESH:D050177), Chronic disease (MESH:D002908), wasting (MESH:D019282), knee pain (MESH:D046788), psychiatric problems (MESH:D001523), Gastric or duodenal ulcer (MESH:D013276), functional impairment (MESH:D003072), Parkinson's disease (MESH:D010300), arthritic condition (MESH:D015535), quadriceps muscle weakness (MESH:D018908), thrombocytosis (MESH:D013922), pain (MESH:D010146), degenerative disease (MESH:D019636), OA (MESH:D010003), knee joint injury (MESH:D007718), Chronic lung diseases (MESH:D029424), FI (MESH:D000073496), diabetes (MESH:D003920), leukocytosis (MESH:D007964), stroke (MESH:D020521), joint injuries (MESH:D000092464), dyslipidemia (MESH:D050171), obese (MESH:D009765), mobility restriction (MESH:D014086), Kidney disease (MESH:D007674), falls (MESH:C537863), diseases (MESH:D004194), sarcopenia (MESH:D055948)
- **Chemicals:** blood glucose (MESH:D001786), alcohol (MESH:D000438), lipid (MESH:D008055)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC13028503/full.md

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC13028503/full.md

## References

45 references — full list in the complete paper: https://tomesphere.com/paper/PMC13028503/full.md

---
Source: https://tomesphere.com/paper/PMC13028503