More than sixty years after “I look up as I walk”, known worldwide as “Sukiyaki”, became Japan’s most famous song about walking, a new study finds that as walkers look up, their predicted healthcare costs come down. In addition, how steeply the costs descend changes dramatically with age. Physical activity has long been associated with lower healthcare costs, but most studies have relied on self-reported questionnaires and linear models. This has left the actual shape of the relationship, and how it evolves across the lifespan, largely hidden.
To uncover the connection between these variables, Associate Professor Haruka Kato at Osaka Metropolitan University’s Graduate School of Human Life and Ecology analyzed a linked dataset combining objective step counts from “A-Smile,” Osaka Prefecture’s official mobile health app, with administrative health insurance claims. The study covered 20,950 long-term app users aged 40 and over, which amounted to 338,706 person-months in total from 2020 to 2023. Generalized additive models, a statistical technique that estimates flexible curves rather than assuming straight lines, were used to map the data.
The results revealed a strikingly age-dependent landscape. Visualized as a three-dimensional surface of predicted costs across age and daily steps, the relationship was nearly flat for middle-aged adults aged 40–64. This group’s healthcare costs appeared to be driven by factors other than walking. However, among older adults 65 and over, a towering peak of predicted costs rose in the low-step region and fell away like a cliff as step counts increased. In other words, the gradient linking steps to healthcare costs steepens dramatically with age. Yet one feature remained constant across generations: predicted costs stopped declining and leveled off at around 11,000 steps a day, specifically 11,317 steps for middle-aged adults and 10,612 steps for older adults. The steepness of the slope changes with age, but the location of the plateau does not.
Our findings suggest that promoting walking among older adults could substantially contribute substantially to reducing healthcare expenditures because predicted costs are far more sensitive to daily steps in this group. Moreover, because the model estimates costs as a flexible function of age and daily steps, the model makes it possible to estimate the expected reduction in healthcare expenditures for any given increase in steps at any given age.”
Dr. Haruka Kato, Associate Professor, Osaka Metropolitan University’s Graduate School of Human Life and Ecology
To illustrate this, Dr. Kato has released an interactive version of the fitted model (https://harukakato-omu.github.io/steps-healthcare-cost-gam/), which displays the predicted expenditure and its confidence interval for any combination of age and daily steps within the study’s range.
The findings were published in SSM – Population Health.
Source:
Journal reference:
Kato, H. (2026). Non-linear association between mHealth-measured daily steps and subsequent healthcare expenditures among long-term app users in Osaka, Japan. SSM – Population Health. DOI: 10.1016/j.ssmph.2026.101958. https://www.sciencedirect.com/science/article/pii/S2352827326000601
Team Health Accessible
Health & Wellness Editorial Team
HealthAccessible editorial team delivers trusted, accessible, and evidence-based health information for everyone.