SDoH data is information about the non-medical factors that shape a person’s health, things like housing stability, income, education, food access, and neighborhood safety. These social determinants of health are estimated to influence as much as 50 percent of health outcomes at the county level, while clinical care accounts for only about 20 percent. Collecting and using this data helps healthcare systems understand why patients get sick, not just what they’re sick with.
The Five Domains of SDoH Data
The federal framework from Healthy People 2030 organizes social determinants into five domains: Economic Stability, Education Access and Quality, Health Care Access and Quality, Neighborhood and Built Environment, and Social and Community Context. Each domain captures a different slice of a person’s life circumstances that can either protect or threaten their health.
In practice, the data points within these domains are concrete and specific. Economic Stability covers income, employment status, and whether someone lives below the federal poverty level. Education includes literacy and language barriers. Neighborhood and Built Environment tracks things like access to nutritious food, exposure to polluted air or water, and whether someone lives in a safe area. Social and Community Context captures discrimination, social isolation, and domestic violence. Together, these five domains paint a picture of daily life that clinical data alone misses entirely.
The relative weight of these factors is striking. One breakdown from the U.S. Department of Health and Human Services estimates that socioeconomic factors alone account for 47 percent of health outcomes, health behaviors for 34 percent, clinical care for 16 percent, and the physical environment for 3 percent. That ratio explains the growing urgency to capture SDoH data systematically rather than treating it as background noise.
How SDoH Data Gets Collected
Most SDoH data enters the healthcare system through screening tools administered during patient visits. The most widely adopted is the PRAPARE assessment (Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences), which contains 22 factors covering social background, housing security, material needs, employment, insurance status, and social isolation. A patient completing PRAPARE might be asked whether they’re worried about losing their housing, whether they need help with food or utilities, how often they see or talk to other people, and whether they feel safe where they live.
PRAPARE organizes its questions into clusters. A social background cluster covers language, ethnicity, education, and race. A social insecurities cluster addresses housing, material needs, transportation, access to healthcare, stress, domestic violence, and safety. Separate clusters cover insurance and employment, while standalone items capture income level relative to the federal poverty line, social isolation, and housing status. The tool is designed to align with standardized medical coding systems so that the answers can be stored and shared electronically.
Beyond formal screening tools, SDoH data also comes from community-level sources: Census data, public housing records, food desert maps, transportation network analyses, and environmental quality databases. Health systems increasingly layer these population-level datasets on top of individual screening results to get a fuller picture of a community’s needs.
How It’s Coded in Medical Records
When SDoH data is documented in a clinical setting, it uses a specific set of codes in the ICD-10-CM system, the same coding framework used for diagnoses and procedures. Categories Z55 through Z65 are reserved for social determinants, covering issues like housing instability, food insecurity, lack of transportation, unemployment, and problems related to education or literacy. These Z codes let a provider formally record that a patient faces a social barrier to health, making that information visible to anyone else involved in their care.
Standardization goes beyond coding. The Gravity Project, a national initiative working with HL7 International, has developed a comprehensive set of standardized SDoH data elements and value sets. These elements are built into the FHIR standard (the current framework for exchanging health data electronically), so that SDoH information recorded in one system can be read and understood by another. The federal government has reinforced this by including SDoH as a data class in the US Core Data for Interoperability (USCDI), which lists specific elements that electronic health records should be able to capture and exchange: food insecurity, housing instability, transportation insecurity, financial strain, social isolation, interpersonal violence, stress, inadequate housing, incarceration history, and more.
Why Standardization Matters
Without consistent standards, SDoH data collected at one clinic can’t be meaningfully compared or combined with data from another. A community health center might screen for food insecurity using one set of questions and categories while a hospital across town uses a completely different approach. Standardized coding and data exchange formats solve this by ensuring that “food insecurity” means the same thing in every system, enabling population-level analysis, coordinated referrals to social services, and tracking of whether interventions actually improve outcomes over time.
The USCDI SDoH data class now includes over 20 elements, ranging from housing instability and homelessness to childcare insecurity, clothing insecurity, refugee status, education level, and congregate living. This list continues to expand as the field matures, and it represents the minimum set of social factors that certified health IT systems are expected to handle.
Privacy Concerns With Social Data
SDoH data is more personal than a blood pressure reading. It can reveal that someone is experiencing domestic violence, is homeless, has been incarcerated, or is struggling to feed their family. Under HIPAA, this information can be used and shared for treatment purposes without requiring separate patient authorization. But disclosures to other authorities or for purposes unrelated to treatment or payment may require the patient’s explicit consent.
The technical infrastructure hasn’t fully caught up to these distinctions. Most electronic health record systems lack the ability to segment sensitive social data from the rest of a patient’s record, meaning they can’t easily restrict who sees what. This is a real problem: surveys have found that about one in eight patients avoids seeking care for sensitive conditions or withholds information from providers because of confidentiality concerns. If patients fear that disclosing housing instability or substance use will follow them in ways they can’t control, they may simply stop answering screening questions honestly.
Mental health records illustrate the tension. SDoH information collected by a therapist in psychotherapy notes receives stronger legal protection but becomes less available to other providers who might use it to coordinate care. The field is still working out how to balance comprehensive data sharing with the trust patients need to disclose sensitive information in the first place.
What Organizations Do With SDoH Data
Health systems use SDoH data at two levels. At the individual level, a provider who knows a patient lacks reliable transportation can connect them with ride services for appointments, or a care team that sees a food insecurity flag can refer the patient to a local food bank. These connections between clinical care and social services are increasingly formalized through closed-loop referral systems, where the health system can track whether the patient actually received help.
At the population level, aggregated SDoH data helps organizations identify patterns. A hospital might discover that readmission rates are highest among patients from a specific zip code with limited access to pharmacies, or that patients screening positive for social isolation have significantly worse outcomes for chronic conditions like diabetes and hypertension. These insights allow health systems, insurers, and public health agencies to direct resources where they’ll have the greatest impact, shifting from treating illness after it appears to addressing the conditions that cause it.