Back

Urban-rural differences in physical therapy use among US adults with arthritis

McLaughlin, K. H.; Bove, A. M.; Minick, K. I.; Skolasky, R. L.

2025-12-17 rehabilitation medicine and physical therapy
10.64898/2025.12.16.25342395 medRxiv
Show abstract

PurposeThe purpose of this study was to analyze differences in physical therapy (PT) use among a nationally representative sample of adults with symptomatic arthritis according to the rurality of their residence. MethodsWe used data from the 2023 National Health Information Survey (adult sample) to identify individuals with symptomatic arthritis by using 2 survey items indicating whether a healthcare provider had diagnosed the participant with arthritis and whether the participant had experienced arthritis-related symptoms during the past 30 days. One survey item was used to identify whether the participant had participated in PT during the previous 12 months. Levels of rurality were defined, in ascending order, as "large central metropolitan," "large fringe metropolitan," "medium or small metropolitan," and "nonmetropolitan/rural," according to each participants county of residence. Univariate and multivariate statistics were used to determine the association of PT use with the level of rurality. National estimates were calculated using weighting variables. FindingsWe identified 5,749 adults (weighted = 40,358,683) meeting our definition of symptomatic arthritis. Compared to those living in large central metropolitan areas, participants living in medium or small metropolitan areas were 20% less likely to report PT use (weighted odds ratio: 0.80; 95% confidence interval: 0.66, 0.96) and those living in nonmetropolitan/rural areas were 30% less likely (weighted odds ratio: 0.70; 95% confidence interval: 0.56, 0.88). ConclusionsAmong adults with symptomatic arthritis, those living in more rural areas had lower odds of PT use than those living in less rural (more urban) areas.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.