Regional disparity in the use of a financial incentive for device-assisted rehabilitation: an ecological study in Kanagawa, Japan.
Daisuke Kubo, Takuya Hirose, Yushi Asaoka
PMID 42544284WHAT IT FOUND
In Kanagawa, Japan, billing for device-assisted rehab (FES, robotics) varies 67.7-fold between neighbouring areas.
No link was found to local stroke caseload or rehab capacity, pointing to the 2-month eligibility window as a likely structural barrier rather than a demand or supply problem.
Key findings
01Across nine secondary medical areas in Kanagawa, the standardized claim ratio for the device-assisted rehabilitation incentive ranged from 3.3 to 223.3; only Yokosuka–Miura (223.3) and Yokohama (136.3) met or exceeded the national average of 100.
02None of the exploratory correlations between incentive uptake and indicators of regional stroke demand or rehabilitation service capacity reached statistical significance.
03Excluding the highest-uptake area (Yokosuka–Miura) did not change the direction of any correlation but weakened their magnitudes; the authors note these analyses are exploratory given the small number of areas (n=9).
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Ecological design with only nine areas: no individual-level data, no causal inference, and high risk of ecological fallacy. Statistical power was very limited; the authors explicitly state the non-significant correlations may reflect a Type II error rather than a true absence of association. Claims data capture only publicly insured billing; out-of-pocket and automobile-insurance-funded sessions are invisible. The data cannot identify which specific devices were used or the frequency, intensity, or delivery details of the intervention. If a facility provided device-based rehab but did not submit the add-on claim, uptake is underestimated. Correlations were heavily sensitive to a single data point (Yokosuka–Miura), as shown by the sensitivity analysis.
Declared interests
The authors declare no conflict of interest and received no financial support for the research, authorship, or publication.
The easy way to misread this
Do not read the non-significant correlations as proof that stroke caseload or rehabilitation capacity do not influence uptake of FES and robotics. With only nine areas the study was underpowered to detect a moderate association, and ecological correlations cannot be interpreted as individual-level effects. The absence of a significant finding here is not evidence of absence.