Estimating the Future Risk of Osteoarthritis

Despite identifying 31 multivariable risk prediction models for calculating incidence of osteoarthritis (OA), researchers found the predictors lacking in applicability and usability in wider populations. The findings are published in Arthritis Care and Research.

“Our review identifies a general lack of inclusion of social stratifiers beyond age, sex, and occupation-associated risk,” the investigators mentioned. “Despite growing interest in multivariable prediction models for incident OA, focus remains predominantly on the knee, with reliance on data from a small pool of appropriate cohort data sets, and concerns over general population applicability.”

The team of 4 investigators searched PubMed, EMBASE, and Web of Science databases to include study populations from 15 unique data sources worldwide through December 2021.

The most common predictors of OA, regardless of the type, were age, body mass index, previous injuries, and occupational exposures. Focus remained predominantly on the knee (23), followed by hip (4), hand (3), and other OA (1).

However, “a reliance on a restricted number of cohort data sets, mainly from higher-income countries, and use of data sources that may be challenging to scale up in routine practice,” the authors stated. These key limitations challenge the applicability of existing prediction models in general populations.

Reference:
Appleyard T, Thomas MJ, Antcliff D and Peat G. Prediction models to estimate the future risk of osteoarthritis in the general population: A systematic review. Arthritis Care Res. 2023; 75: 1481-1493. DOI: https://doi.org/10.1002/acr.25035