Digital Exclusion Linked to Disability and Lifespan in Multinational Study

The digital divide is no longer just a matter of inconvenience for older adults—it may be quietly subtracting years from their lives. A sweeping new multinational study published in SSM – Population Health has found that older people who lack access to the internet, a condition researchers call digital exclusion, face significantly higher risks of physical decline, impaired recovery from disability, and earlier death, ultimately losing several years of both total and healthy life expectancy. The research, led by Qian Liu and Fan Yang, analyzed longitudinal data from more than 124,000 adults aged 50 and older across 23 countries, drawing on five of the world’s most established aging cohorts: the China Health and Retirement Longitudinal Study (CHARLS), the Mexican Health and Aging Study (MHAS), the Survey of Health, Ageing and Retirement in Europe (SHARE), the English Longitudinal Study of Ageing (ELSA), and the Health and Retirement Study (HRS) in the United States.

The methodological strength of the study lies in its use of continuous-time multi-state Markov models, a statistical framework that treats functional health not as a fixed state but as a dynamic process. Rather than asking simply whether digitally excluded people are more likely to be disabled, the researchers modeled transitions between four distinct states: functional independence, mild functional limitations, severe functional limitations, and death. Functional capacity was measured using standardized assessments of six basic activities of daily living—such as dressing, bathing, and walking across a room—and six instrumental activities, including meal preparation, shopping, and managing medications. Mild limitation was defined as difficulty with at least one instrumental activity while remaining independent in basic ones, whereas severe limitation involved difficulty with any basic activity. Death was incorporated as an absorbing state, allowing the team to estimate how digital exclusion shaped not only the risk of entering disability but also the likelihood of recovering from it and the probability of dying at each level of function.

The results were striking in their consistency. Digital exclusion significantly increased the risk of progressing from functional independence to mild limitations in England (hazard ratio 1.47), China (2.56), the United States (1.98), and Europe (1.83). The transition from independence directly to severe limitations was also elevated across all settings, with hazard ratios ranging from 1.24 in Mexico to 1.81 in China. Among those already experiencing mild limitations, digital exclusion accelerated further deterioration in the United States and Europe. More troubling still, the study found that being offline impaired recovery: severely limited individuals who were digitally excluded were substantially less likely to regain functional independence in England (hazard ratio 0.74), the United States (0.55), and Europe (0.72), and digitally excluded people with severe limitations faced elevated mortality in England (1.59), the United States (1.36), and Europe (1.36).

The life expectancy calculations translate these transition dynamics into a metric with immediate public resonance. Using the fitted multi-state models, the researchers estimated total life expectancy and “marginal” life expectancy—the expected number of years spent in each functional state—at nine ages from 50 through 90. Digital exclusion was associated with a loss of functional life expectancy ranging from nearly four years at age 50 in Mexico to more than 14 years at age 50 in China, with statistically significant losses at every age examined. Total life expectancy losses followed a similar pattern, from roughly 2.8 years in Mexico to 8.15 years in the United States at age 50. In China and the United States, digitally excluded individuals also spent significantly more years living with mild and severe limitations, suggesting a dual penalty: shorter lives overall and a larger proportion of those lives constrained by disability.

Why the effects vary so dramatically between countries is one of the study’s most illuminating contributions. The authors propose a framework of “digital-healthcare coupling”: where medical resources, appointment registration, and care coordination have migrated overwhelmingly onto digital platforms, being offline compounds barriers to timely and appropriate care. China exemplifies this pattern. During the study period, tertiary hospitals—which concentrate the country’s best expertise and equipment—increasingly required electronic health cards linked to mobile payment systems for registration, consultation, payment, and prescriptions, effectively channeling digitally excluded patients toward resource-poor primary care facilities. Mexico sits at the opposite pole: walk-in pharmacy-adjacent clinics offering low-cost consultations without appointments, and traditional medicine practices used by as many as 12 to 60 percent of the population—particularly in rural areas—provided robust non-digital pathways to care, attenuating the marginal health cost of being offline.

The recovery findings follow a similar institutional logic. Impaired recovery from severe limitations appeared exclusively in high-income settings—England, the United States, and Europe—where health systems have increasingly embedded telerehabilitation and digital service delivery into rehabilitation care. In contrast, China’s “Internet Plus Healthcare” initiatives during the study years emphasized diagnosis and treatment through hospital-to-hospital networks rather than patient-facing rehabilitation, while both China and Mexico relied heavily on facility-based services and family caregiving rooted in traditions of filial piety and familismo. In such contexts, the incremental burden of digital exclusion on recovery appears weaker, underscoring that the health consequences of the digital divide are not universal but are shaped by how health care is organized and delivered.

The stratified analyses add further nuance. The harm of digital exclusion was greatest at younger ages: losing 14 functional life-years at age 50 versus just over 3 years at age 90 suggests that lacking digital access in midlife forecloses decades of accumulating “health capital”—better information, stronger social connection, and more effective use of health services. Counterintuitively, individuals without chronic diseases suffered larger life expectancy losses than those with chronic conditions, perhaps because healthy people depend on the internet for preventive resources such as lifestyle information, screening reminders, and monitoring tools, whereas chronically ill patients remain tethered to structured, often non-digital, clinical care routines. Sex differences also emerged: in China, men lost more total life expectancy while women accumulated more years with limitations, and in the United States women fared worse on both counts. Notably, urban–rural differences were minimal across all settings, suggesting that non-digital alternatives—or their absence—shape the digital divide’s health impact more than geography.

The mechanisms linking connectivity to physical function are plausible and multifaceted. Internet access facilitates health information seeking, telehealth consultations, real-time monitoring, and the purchasing of medications and health devices. It also sustains social participation: internet users engage in more social activities, experience less isolation, and adopt more beneficial health behaviors. Prior research cited in the study found that digital inclusion was associated with a 33 percent reduction in depression likelihood among retired adults and improved “intrinsic capacity” across cognitive, sensory, physical, and functional domains. Conversely, prior multinational work has linked digital exclusion to functional dependency, frailty, cognitive impairment, and depressive symptoms across dozens of countries. The new study extends this evidence base by showing that these associations translate into fundamentally different aging trajectories—differential risks of deterioration, recovery, and death—rather than static cross-sectional differences.

The authors are candid about the study’s limitations. Digital exclusion was operationalized differently across cohorts: most measured individual internet use, but the Mexican survey assessed only household internet access, and some cohorts applied frequency thresholds such as weekly use. The measure captures access—the foundational barrier—rather than the full multidimensional construct of digital literacy and usage competency. Functional limitations were self-reported difficulties rather than observed care dependency, and income data were unavailable or unreliable in some cohorts, particularly China, where over 40 percent item non-response and systematic misreporting complicated adjustment; the authors note this may have produced some upward bias in the Chinese estimates. Sensitivity analyses using simplified three-state models, ADL-only definitions, and additional covariates confirmed the robustness of the core findings.

The policy implications are difficult to ignore. As health systems worldwide accelerate digital transformation—from online appointment booking to app-based prescription services—the study warns that efficiency gains may come at the cost of equity, with the heaviest burden falling on those least equipped to bear it. The authors argue that digital inclusion should be treated as a modifiable determinant of healthy aging and integrated into national aging strategies, with interventions targeted particularly at midlife and “younger-old” populations, where the returns on connectivity are largest. Equally important, they caution, is the preservation of non-digital access channels—telephone appointments, walk-in clinics, face-to-face consultations—so that health care does not become contingent on a broadband connection. In an era when getting older increasingly means navigating a screen, the study suggests that the choice to digitize is also a choice about how many years, and what quality of years, societies are willing to let their most disconnected citizens lose.