Diabetes mellitus and arthritis: is it a risk factor or comorbidity?: A systematic review and meta-analysis

BACKGROUND: Investigators have explored the association between diabetes mellitus and arthritis for a long time; however, there are uncertainties and inconsistencies among various studies. In this study, we tried to explore the relationship between diabetes mellitus and the overall risk of arthritis, as well as the potential modifiers for this relationship. METHODS: We conducted a comprehensive literature search through PubMed and identified 36 eligible studies. The overall analyses, subgroup analyses, as well as sensitivity analyses, were conducted to illustrate the association between diabetes mellitus and arthritis. Study quality was evaluated using the Newcastle-Ottawa Quality Assessment Scale. All statistical analyses were conducted using STATA SE version 13.0. RESULTS: In our study, 36 eligible studies were identified and involved in the meta-analysis. The overall association between diabetes mellitus and arthritis is 1.61 (95% confidence interval [CI]: 1.14–2.28, P = .007). The association exists only in nongouty arthritis, where we observed the estimated odds ratio (OR) 1.33 (95% CI: 1.05–1.67, P < .001). The opposite point estimates from different types of diabetes may indicate possible different associations for type I (OR: 0.98, 95% CI: 0.18–5.39, P = .985) or type II diabetes (OR: 1.28, 95% CI: 0.88–1.84, P = .194). CONCLUSION: Diabetes mellitus performs more likely as a comorbidity of arthritis rather than a risk factor; however, more studies will be helpful to increase the confidence of identifying the association between diabetes and arthritis.