Urban-rural differences in the association between access to healthcare and health outcomes among older adults in China

Picture two older adults in China. Both are in their mid-seventies. Both live with some joint pain and early memory trouble. One lives in Shanghai and the other lives in a farming village in Sichuan province. On paper, both are covered by national health insurance — China hit 95 percent enrollment by 2015. But the village resident's insurance is the New Cooperative Medical Scheme, which carries higher out-of-pocket costs and lower benefits than the urban plans. The nearest clinic is far away, the family has moved to the city for work, and money is tight. When something goes wrong, care is delayed or skipped entirely. Zhang and colleagues set out to measure exactly what that difference costs — in disability, in cognitive decline, and in years of life. The study draws on four waves of the Chinese Longitudinal Healthy Longevity Survey, collected in 2005, 2008, 2011, and 2014. The sample included 26,604 adults aged 65 and older, covering nearly 90 percent of China's population across 23 provinces. Response rates across waves held near 98 percent, and the survey deliberately oversampled the oldest-old — people in their eighties, nineties, and beyond — to ensure robust data on aging and longevity. Together, those participants contributed over 48,000 person-wave observations. This is a big, carefully designed dataset, and it gives the findings unusual credibility. The central variable is straightforward: did you have adequate access to the healthcare you needed? Respondents answered yes or no. The gap that answer revealed was immediate. Among older adults in rural areas, nine point one percent reported inadequate access. Among urban older adults, the figure was five point four percent. That four-percentage-point difference sounds modest until you consider what drives it. About sixty percent of rural older adults who reported inadequate access cited financial constraint as the main barrier. Among urban older adults in the same situation, it was forty-five percent. Rural residents are disproportionately enrolled in the New Cooperative Medical Scheme, which Zhang and colleagues describe as less beneficial than urban plans — higher co-payments, lower coverage ceilings, and less financial protection overall. Add lower rural incomes, fewer pensions, and adult children who have migrated away, and the structural picture comes into focus. What the study then measured was where that structural gap shows up in the body. Zhang and colleagues tracked four outcomes: instrumental activities of daily living disability — things like shopping, cooking, walking a kilometer, or taking public transit; basic activities of daily living disability — bathing, dressing, eating, and continence; cognitive impairment, measured using a Chinese version of the Mini-Mental State Examination with a score below twenty-four indicating impairment; and all-cause mortality. For the disability and cognitive outcomes, they used logistic regression, building nested models that added sociodemographic factors, insurance enrollment, and health behaviors in sequence. For mortality, they used Cox proportional hazards models, which estimate how much faster one group reaches an endpoint — in this case, death — compared to another. Start with instrumental activities, the more complex daily tasks. Inadequate access to healthcare was associated with significantly higher odds of this kind of disability in both urban and rural older adults — but the rural effect was consistently larger. In urban areas, the fully adjusted odds ratio landed between one point fifty-eight and one point seventy-nine. In rural areas it was between one point ninety-five and two point thirty. Both are meaningful. But the rural number is telling you the odds of disability are roughly doubled compared to rural peers with adequate access. Now move to basic activities — bathing, dressing, the fundamentals of self-care. Here the urban association disappeared after adjustment. It was significant only in rural older adults, where the odds ratio ranged from one point eighty-nine in the most adjusted model up to three point zero five in the model controlling only for sociodemographics. A three-fold increase in the odds of needing help with basic daily tasks, tied to inadequate healthcare access — and the effect shows up only in rural areas. The cognitive impairment findings follow the same pattern, and they are the most striking. In rural older adults, inadequate access was associated with odds ratios between two point thirty-seven and three point nineteen across model specifications. In urban older adults, there was no significant association at all. Let that sit for a moment. The gap in healthcare access between a village clinic and an urban hospital is showing up not just in whether someone can dress themselves, but in whether their mind is intact. No similar link emerged for urban residents, even though inadequate access exists there too. Then there is mortality. The Cox models tell a consistent story. In rural older adults, inadequate access was associated with a twenty-nine percent higher mortality risk in the sociodemographic model, and that association persisted — twenty-four percent higher after adjusting for health behaviors, seventeen percent higher in the fully adjusted model including baseline health status. The rural mortality penalty remained statistically significant across every specification. In urban older adults, the initial hazard ratio was one point thirty-seven, but it fell to one point eighteen and lost statistical significance once health behaviors were accounted for. The pattern reinforces what the disability and cognition results already suggested: inadequate access carries a survival cost in rural areas that does not wash out with adjustment. Why is the rural effect so much larger? Zhang and colleagues are careful here — they cannot directly measure transportation barriers, distance, terrain, or objective facility quality, and they acknowledge that. But the mechanisms they point to are grounded in the data. Rural residents face higher financial barriers even with insurance. They rely more heavily on local village doctors and in-home care, and are less likely to use inpatient hospital services where better-equipped care lives. Urban older adults are more likely to reach hospitals and specialist care when something goes wrong. The paper also notes that adult children migrating to cities removes a practical layer of caregiving support for rural older adults — someone who might otherwise help arrange or accompany them to a clinic. The study also took care to address a methodological concern called endogeneity — the worry that the relationship runs backwards, that sicker people seek more care and therefore seem to have worse access. The authors measured access as a self-reported perception of adequacy rather than simply insurance enrollment, which better captures whether care was actually obtained. They adjusted for baseline health status, and their conclusions held. The policy argument flows directly from the findings. Near-universal insurance enrollment is not the same as adequate access to care. China achieved ninety-five percent coverage, but the data show that the nine point one percent of rural older adults who still report inadequate access face substantially worse outcomes across every domain measured — disability, cognition, and survival. Zhang and colleagues call for unifying China's fragmented urban and rural insurance programs, expanding New Cooperative Medical Scheme benefits to close the gap with urban plans, and improving rural healthcare infrastructure so that coverage actually translates into care people can reach and afford. The finding that matters most is not any single odds ratio. It is the consistency of the pattern. Across four health outcomes, across multiple model specifications, and in a sample of more than 26,000 people followed over nearly a decade, the rural penalty for inadequate healthcare access was larger, more robust, and more persistent than the urban one. As China's population ages — and it is aging fast — that pattern is not a statistical artifact. It is the consequence of a policy design that created two systems and called it universal coverage. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.