The problem the researchers set out to tackle is twofold. First, health systems worldwide struggle to provide fair and equal access to primary care. In the United Kingdom, people living in areas of socioeconomic disadvantage consistently report lower satisfaction with the care they receive, and general practices in deprived areas have fewer doctors, less funding, and are more likely to be rated inadequate, all while serving patients with more complex, long-term health problems at younger ages. This is a textbook illustration of the “Inverse Care Law,” first articulated by Julian Tudor Hart in 1971, which holds that the availability of good medical care tends to vary inversely with the need for it in the population. Second, even where evidence exists, it is becoming nearly impossible to navigate. Primary care publications alone have risen by roughly 380 percent over the past two decades, and the average worldwide growth rate of academic output hovers around four percent per year. A full systematic review takes, on average, sixteen months from design to publication — by which point its findings may already be outdated.

Traditional systematic reviews, the gold standard for synthesising medical evidence, are labour-intensive and slow, and they rapidly fall behind the literature they are meant to summarise. Machine learning offers a way out. Prior work has identified dozens of tools that use machine learning techniques to assist with the systematic reviewing process, supporting everything from study selection to data extraction and gap identification. Yet relatively few studies have systematically combined these methods to support policymakers and practitioners working on health and care inequalities. Until now, no living evidence map existed describing how to address inequalities in and through primary care.

The research team built their Living Evidence Map using EPPI-Reviewer, systematic review management software developed by the EPPI Centre at University College London, together with its integrated suite of machine learning tools. Bibliographic records were drawn from OpenAlex, an open-access database containing more than 250 million scholarly works. At the heart of the workflow was a binary machine learning classifier — a model trained to classify each record as likely relevant or not relevant to the review question. The classifier was developed using 1,006 manually included title and abstract records and 22,426 excluded records, randomly assigned to training, calibration and evaluation sets with stratification by inclusion status. The model learned patterns in titles and abstracts associated with study relevance and assigned each incoming record a relevance score; records falling below a threshold were excluded from the screening pool entirely.

The team’s searches ran approximately monthly using two complementary approaches. Citation-based searches identified records linked to known relevant studies through citation relationships — papers that cited, were cited by, or were otherwise connected to included studies. Automated update searches used a model called ContReview, which combines information from citation links and article text to rank unscreened records by likely relevance. Human reviewers then screened articles in order of predicted relevance, with the screening pool continually re-ranked using an active machine learning approach, meaning the model improved as screening progressed. Screening continued until the rate of inclusion dropped, a standard stopping criterion in automated evidence synthesis.

The classifier’s performance was striking. On the evaluation set of 4,686 records, it achieved a recall of 0.965, meaning it correctly captured nearly 97 percent of relevant articles, while discarding 60.7 percent of records without any manual screening — a workload reduction that translates into months of saved reviewer time. Precision, at 0.105, was deliberately low: the model was tuned to prioritise catching everything relevant over keeping the screened pool small, a sensible trade-off when the cost of missing a key study outweighs the cost of screening a few extra irrelevant ones. Included articles were then manually coded for intervention type, disadvantaged population group, health or care outcome, and study design, with a ten percent sample audited by a second researcher to ensure accuracy.

The resulting map paints a vivid picture of where research attention has flowed — and where it has not. The team included 577 primary studies, 481 systematic reviews and six umbrella reviews, along with 154 minor contributions. Ethnic minority population groups emerged as by far the most frequently studied disadvantaged group, particularly in relation to education interventions, cultural tailoring, and chronic disease management. The single most heavily researched combination was education interventions for ethnic minorities, with 127 systematic reviews and 95 primary studies, followed closely by culturally competent care and advice and counselling interventions for the same groups. Latino and Hispanic populations were the most studied of all, followed by Black African and Caribbean and then Asian populations — a pattern the authors attribute to the predominance of studies originating in the United States.

In sharp contrast, gender and sexual minorities were the most underrepresented of all groups, with the fewest studies identified. The authors suggest this reflects the invisibility of these populations in research and a lack of routine data, since gender expression and sexual orientation are not systematically coded in health care practice, making it harder to target interventions. Notably absent from much of the map, too, were structural interventions — those addressing funding allocation, workforce distribution, and other upstream determinants of health. Such interventions were considerably less common than discrete, individual-level approaches such as education, counselling, and link workers. The researchers argue this is unsurprising but concerning: discrete interventions are easier to evaluate in conventional trial designs over short periods, whereas funding reforms and workforce policies are complex, slow-moving, and require long-term data. Funders, meanwhile, may prefer downstream interventions because they offer more direct, demonstrable benefits to individual patients.

Other patterns emerged in the conditions studied. Research on ethnic minority groups more frequently examined diabetes-related outcomes — with 87 systematic reviews and 94 primary studies on the topic — whereas studies of inclusion health groups, such as people experiencing homelessness or substance dependence, more commonly focused on cancer and substance misuse outcomes. Intriguingly, the team also found that the number of systematic reviews roughly matched the number of primary studies, a potentially unhealthy sign for the research ecosystem. For evidence synthesis to function well, there should always be far more primary research than reviews to draw upon. Recent analyses have found that the number of systematic reviews indexed in PubMed increased more than twenty-fold over two decades, reaching approximately eighty published per day by 2019.

The implications stretch well beyond primary care research. The Living Evidence Map, now publicly available through the Health Equity Evidence Centre, allows policymakers, commissioners and practitioners to explore the evidence interactively, spotting patterns and gaps in real time as new studies are added. The authors acknowledge limitations: the map does not yet capture intersectionality or multiple disadvantage, excludes grey literature and non-English studies, and is limited to high-income, UK-comparable contexts. Some relevant studies that do not mention specific disadvantaged groups in their titles and abstracts may also have been missed. Maintenance funding for living evidence resources remains an open question. Nevertheless, the study demonstrates that machine learning can transform evidence synthesis from a snapshot that ages quickly into a living, continuously updated resource — and it sends a clear message to research funders that the biggest gaps lie not in patient-level education programmes, but in the structural changes that could reshape who gets good care in the first place.

The researchers focused on Social and Digital Determinants of Health, or SD-DOH. Traditional social determinants include housing stability, food access, transportation, employment, education, neighborhood safety, and social connection. Digital determinants add another layer, including internet availability, device ownership, digital literacy, access to telehealth, the usability of online patient portals, and the ability to navigate increasingly technology-dependent health systems. These factors can determine whether a patient receives an appointment reminder, completes a telemedicine visit, finds trustworthy health information, or remains connected to clinicians after leaving a hospital. As healthcare rapidly adopts artificial intelligence, remote monitoring, online scheduling, and virtual care, unequal digital access can amplify existing racial, geographic, economic, and age-related disparities.

To investigate how organizations respond, the team conducted a PRISMA-ScR-guided scoping review and qualitative evidence synthesis. The researchers searched PubMed, Scopus, ProQuest, and CINAHL for U.S.-based peer-reviewed and gray literature published from January 2020 through June 2025. They identified 94 sources that described the perspectives of healthcare leaders or other organizational stakeholders on responsibility for addressing SD-DOH. Rather than treating responsibility as a single activity, the researchers analyzed the literature through two complementary frameworks: four strategic levers used by health systems and an expanded version of the National Academy of Medicine’s five-part social care framework.

The four strategic levers represent different levels of intervention. Enabling services are downstream activities delivered directly to patients, such as screening for social needs, providing transportation assistance, connecting people with food resources, or supporting digital access. Community partnerships operate at a midstream level, linking hospitals with community-based organizations, schools, local agencies, and public health departments. Policy advocacy is an upstream strategy that seeks to influence laws, regulations, reimbursement systems, and public investment. Direct community investment is another upstream approach, in which healthcare organizations commit financial resources, infrastructure, or institutional assets to address the conditions producing poor health. Together, these categories allowed the researchers to distinguish between helping individuals navigate existing barriers and attempting to reduce or remove the barriers themselves.

The second framework describes how organizations conceptualize their responsibilities. Awareness involves identifying patients’ social and digital needs, often through screening, electronic health record documentation, or community health assessments. Assistance means connecting individuals with services or resources. Adjustment refers to modifying clinical practices, communication, or care delivery to accommodate patients’ circumstances. Alignment involves coordinating healthcare organizations with community partners and broader systems. Advocacy describes efforts to influence public policy and structural conditions. The researchers added a sixth category—system governance and accountability—to capture leadership structures, performance measurement, institutional oversight, and mechanisms that make equity an organizational obligation rather than an optional project.

The results reveal a clear imbalance in the current landscape. Enabling services appeared in 86% of the initiatives identified in the literature, making them by far the most common form of equity-oriented action. Community partnerships followed at 70%. By comparison, policy advocacy appeared in 34% of initiatives, while direct community investment was present in only 9%. The pattern suggests that many health systems are willing to help patients manage immediate obstacles, but far fewer are consistently investing in or challenging the systems that create those obstacles. In practical terms, a hospital may help a patient arrange transportation or enroll in a food program, yet stop short of using its political influence or financial resources to address inadequate transit, food insecurity, broadband gaps, or unstable housing at their source.

The responsibility perspectives found in the literature were also multidimensional. Alignment was the most frequently observed perspective, appearing in 83% of the sources or initiatives analyzed. Assistance appeared in 68%, Adjustment in 65%, Awareness in 61%, and system governance and accountability in 60%. Advocacy was least common, appearing in 32%. This ordering suggests that healthcare leaders often see collaboration and coordination as central responsibilities, while direct political engagement remains more difficult to institutionalize. Alignment may include formal partnerships with community-based organizations, referral networks, shared care plans, joint health assessments, and coordinated responses to local needs. However, collaboration alone does not necessarily redistribute power or funding, and the review indicates that partnerships are most effective when supported by governance systems that clarify roles, measure outcomes, and sustain accountability.

The strongest connections between responsibility perspectives and strategic action appeared where organizations were closest to patient care. Assistance and Adjustment were each associated with 98% of enabling-service initiatives, showing that direct support and flexible care delivery tend to develop together. Alignment appeared in every community-partnership initiative and every direct-community-investment initiative, as well as in 83% of enabling-service initiatives. The relationship between Advocacy and upstream work was especially pronounced: Advocacy appeared in 88% of direct-community-investment initiatives, but in only 25% of enabling-service initiatives. System governance and accountability acted as a bridge across all four levers, appearing in 100% of direct-community-investment initiatives, 75% of policy-advocacy initiatives, 56% of community partnerships, and 54% of enabling-service initiatives.

The authors interpret these findings as evidence that U.S. health systems may be approaching an inflection point. Governance and accountability are already present in a substantial share of initiatives, meaning that organizations may possess some of the institutional tools needed to move beyond isolated programs. Those tools can include executive oversight, equity metrics, board-level reporting, community benefit requirements, dedicated budgets, workforce training, and integration of social and digital needs into electronic health records and quality-improvement systems. The challenge is converting recognition into durable organizational behavior. If equity goals are not linked to funding, leadership evaluation, clinical performance measures, and transparent reporting, they can remain vulnerable to changing priorities and short-term grants.

The review also suggests that policy advocacy may be a realistic next step for organizations that have already developed strong partnerships and internal alignment. Healthcare systems are among the largest employers and economic institutions in many communities, giving them potential influence over broadband expansion, housing policy, transportation, food access, Medicaid reimbursement, digital inclusion, and public health funding. Yet the researchers caution, implicitly through the evidence, that upstream responsibility should not replace direct services. Patients facing immediate needs still require practical assistance, while long-term improvements demand coordinated action across healthcare, government, business, education, and community organizations. The emerging model is therefore not a choice between treating individuals and changing systems, but a progression in which patient support, community alignment, governance, investment, and advocacy reinforce one another. As healthcare becomes more digital and health inequities remain deeply connected to social conditions, the study argues that responsibility for health can no longer be limited to what happens inside the examination room.

Subject of Research: Healthcare responsibility for addressing social and digital determinants of health in the United States

Article Title: Leader and organizational stakeholder perspectives on healthcare responsibility for addressing Social and Digital Determinants of Health (SD-DOH) in the United States: a scoping review and qualitative evidence synthesis

Article References: Rangachari, P., Thapa, A., Gari, S. N. B., et al. “Leader and organizational stakeholder perspectives on healthcare responsibility for addressing Social and Digital Determinants of Health (SD-DOH) in the United States: a scoping review and qualitative evidence synthesis.” BMC Public Health (2026).

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29142-x

Keywords: Social determinants of health; Social and digital determinants of health; Health system responsibility; Equity-oriented initiatives; Digital equity; Cross-sector collaboration; Shared accountability; Policy advocacy