Reducing disparities in health care requires coordinated action across multiple levels, from how hospitals collect patient data to how federal agencies set reporting standards. No single intervention closes the gap on its own. The most effective approaches combine better data infrastructure, community-based programs, policy incentives, and changes to how care is delivered day to day. Here’s what actually works, what’s still falling short, and where the biggest opportunities lie.
Better Data Collection as the Foundation
You can’t fix a disparity you can’t measure. That principle drives the most fundamental shift in health equity work: requiring hospitals and health systems to collect detailed demographic data from patients and then stratify their quality metrics by race, ethnicity, language, and social factors. The University of Chicago Medical Center, for example, built an equity lens into its institutional quality scorecard, allowing leaders to see exactly where outcomes diverge across patient groups rather than relying on system-wide averages that mask gaps.
At the federal level, the Office of Management and Budget updated its race and ethnicity data standards in 2024 for the first time since 1997. The revised framework now includes seven minimum categories, adding “Middle Eastern or North African” as a distinct group. These categories are understood as social and political constructs, not biological ones, and they exist to give federal agencies a common language so data can be compared across programs and institutions. Without standardized categories, disparities in one system can’t be meaningfully compared to another.
The Leapfrog Group, a hospital safety watchdog, now requires hospitals to report whether they collect patient self-reported demographic data, whether staff are trained in that collection, and whether quality and safety measures are stratified by demographics. This kind of external accountability pushes health systems beyond simply gathering data toward actually using it to find and fix inequities.
Screening for Social Needs in Clinical Settings
A growing number of hospitals and clinics now screen patients for social determinants of health: housing instability, food insecurity, transportation barriers, and similar factors that shape health outcomes far more than most clinical interventions. The challenge isn’t the screening itself. It’s what happens afterward.
A CDC-published study of emergency department patients in Utah found that among those who wanted referrals to social services after screening, only 49% were ultimately reached by a community referral line. That means roughly half the patients who identified a need and asked for help never connected with services. Screening without a reliable referral infrastructure can create frustration and erode trust, particularly in communities that already have reason to distrust the health care system. Effective programs invest as heavily in the follow-up process (dedicated staff, partnerships with community organizations, closed-loop referral tracking) as they do in the screening tool itself.
Community Health Workers Close the Gap
Community health workers are trusted members of the communities they serve, and they bridge the space between clinical care and daily life. They help patients navigate insurance, attend appointments, manage chronic conditions, and connect with food or housing resources. The evidence on their impact is strong, though the economics are nuanced.
CDC data shows the median cost of a community health worker intervention is about $329 per person per year. The median cost per quality-adjusted life year saved is $17,670, well below the $50,000 threshold that economists typically use to define a cost-effective intervention. One large study of an underserved urban population found a return on investment of 1.8 to 1 for the health plan. However, across multiple studies, direct health care cost savings didn’t always exceed the cost of the program itself. The value often shows up in harder-to-quantify outcomes: fewer missed appointments, better medication management, earlier detection of worsening conditions, and stronger patient engagement. For health systems willing to take a longer view, community health workers are one of the most evidence-backed tools for reaching populations that traditional care models miss.
Health Literacy and Plain-Language Communication
Complex medication instructions, dense discharge paperwork, and jargon-heavy provider conversations all widen disparities. Patients with lower health literacy are more likely to take medications incorrectly, miss follow-up care, and end up back in the hospital. Targeted literacy interventions can make a measurable difference.
In one study of older adults with coronary heart disease, a structured medication education program reduced the rate of dosing errors from 15.5% in the standard care group to 5.2% in the education group. Complications and hospital readmissions also dropped significantly. Another study found that adding simple pictograms to prescription labels significantly improved medication understanding among older adults compared to text-only labels. Techniques like “teach-back,” where patients explain their care instructions back to the provider in their own words, have also improved both comprehension and medication adherence among older populations.
These interventions don’t require expensive technology. They require intentional design: shorter sentences, visual aids, confirmation that patients actually understand what they’re being asked to do. Any health system serious about equity should treat plain-language communication as infrastructure, not a nice-to-have.
Closing the Digital Divide in Telehealth
Telehealth expanded rapidly during the pandemic, but it didn’t expand equally. In rural high-needs areas of the Southeast, only 43% of households subscribe to broadband, compared to 60% in urban high-needs areas. Device ownership follows the same pattern: while 88% of households across the Southeast own smartphones, that drops to 78% in rural high-needs communities. Only 56% of those households have a laptop, and just 44% have a tablet.
These numbers matter because telehealth is increasingly how patients access mental health services, chronic disease management, and specialist consultations. If your broadband is unreliable or your only internet device is a phone with a small screen, a video visit with a cardiologist isn’t a realistic option. Reducing this disparity means investing in broadband infrastructure, providing devices through health system or library lending programs, and ensuring that telehealth platforms work on low-bandwidth connections. It also means keeping in-person options available for patients who can’t or don’t want to use digital tools.
Rethinking Payment Models for Equity
Value-based payment models were designed to reward better outcomes rather than higher volume of services. In theory, this should benefit underserved populations by incentivizing preventive care and chronic disease management. In practice, the results are more complicated.
Provider organizations that serve the populations most likely to experience health inequities, including federally qualified health centers, Indian Health Care Providers, and rural health centers, are significantly less likely to participate in value-based payment models than other provider types. The upfront investment in data infrastructure, care coordination staff, and reporting systems can be prohibitive for smaller, safety-net providers. And when payment models adjust for cost of care without careful guardrails, they can actually reinforce inequitable spending patterns rather than correcting them.
Medicaid programs are experimenting with building explicit equity requirements into value-based contracts, such as requiring providers to stratify outcomes by race and ethnicity or tying bonuses to closing specific disparity gaps. The risk, as policy analysts have noted, is that without real accountability mechanisms, these requirements become a check-the-box exercise. The most promising designs pair financial incentives with technical assistance, helping under-resourced providers build the capacity to participate meaningfully.
What Implicit Bias Training Can and Can’t Do
Nearly every major health system now offers some form of implicit bias training for clinical staff. A systematic review of 77 such programs in U.S. health care institutions found that most are single sessions lasting about 5.5 hours on average. The evidence that these programs improve patient care is, so far, limited. Changes in bias scores on tests don’t reliably translate into changes in clinical behavior, and most programs haven’t been evaluated using patient outcomes like care satisfaction or treatment quality.
This doesn’t mean bias isn’t a problem. It means that a half-day workshop alone won’t solve it. More promising approaches embed equity into clinical workflows rather than relying on individual attitude change. For example, some hospitals now use a standardized health equity checklist during safety event reviews, prompting teams to examine whether social determinants, bias, or structural racism played a role in an adverse outcome. This shifts the focus from “train individuals to be less biased” to “build systems that catch and correct for bias.”
Hospital-Level Frameworks That Work
The most effective hospital strategies treat health equity not as a standalone initiative but as a dimension of quality and safety. A systems-based framework published in the Joint Commission Journal organizes institutional efforts into three maturity levels (fundamental, intermediate, advanced) across six domains: data collection and training, data validation, stratification and analysis, communicating findings, resolving equity gaps, and organizational culture. This gives hospitals a roadmap for progressing beyond basic data collection toward actually closing gaps.
CMS now requires hospitals to attest to a five-domain health equity structural measure. Hospitals must demonstrate that equity is part of their strategic plan, that they collect and analyze demographic data, that they participate in disparity-focused quality improvement, and that senior leadership, including the board of trustees, annually reviews equity performance indicators. These requirements create a minimum floor. The hospitals making the most progress go further: dedicating budget lines to equity goals, partnering with community organizations, and tying executive performance reviews to measurable progress on disparity reduction.
What distinguishes organizations that actually move the needle from those that stall at good intentions is sustained institutional commitment. Strategic plans that name priority populations, allocate specific resources, and set discrete action steps outperform vague pledges to “address disparities.” The work is ongoing, iterative, and uncomfortable. It requires looking at your own data honestly and acting on what it reveals.