A Hospital Social Needs Index Would Help Hospitals Collaborate To Address Social Needs And Health Equity

In response to a recent Centers for Medicare and Medicaid Services (CMS) proposed rule for inpatient and long-term hospitals related to health equity, hospital stakeholders are ramping up efforts to build health equity into their core functions and considering what’s needed to do so operationally. The importance of addressing the social determinants of health (SDOH) to improve population health is front and center in these efforts.

In the fall of 2021, Cal Hospital Compare, a multistakeholder collaborative focused on providing consumers with hospital performance information in California, began to explore the use of social needs indices to inform its analysis of hospital clinical quality, patient experience, and patient safety. Using publicly available geographic social needs indices combined with hospital patient origin information, the methods team quantified the social needs of hospital patient populations. The approach provides a standardized method that can be used to compare and rank hospital-level social needs based on the patient population served. Importantly, the ranking can also be used to assess and address the impacts of social needs on specific hospital quality measures. The analysis showed that some hospital quality measures are substantially more correlated with social needs than others, and that hospitals within the same geographic area often serve similar high-risk patients, suggesting opportunities for collaboration.

Hospitals Address Health Equity

Hospitals are in a unique position to directly address challenges faced by their sociodemographically complex patients. Many hospitals, particularly those with advanced financial risk-sharing arrangements, are investing in efforts to address SDOH, although the level of investments and nature of interventions vary. Examples of population-level interventions include data analytic technologies to identify high-risk patient populations, screening programs connecting patients with community resources, and transition of care and community care management programs, including community economic development. Other direct patient interventions include identifying temporary housing for patients discharged from hospitals, connecting patients experiencing food insecurity to food banks, providing transportation for follow-up visits in collaboration with ride-hailing services, and connecting people experiencing interpersonal violence with community services.

There is concern that hospitals and health systems that are disproportionately serving patients with social disadvantage or higher-risk populations may fare poorly on quality rankings and receive financial penalties, and be less likely to receive financial rewards on value-based payment strategies, including incentive programs and risk-based models. Some suggest that the underlying differences in patient sociodemographic characteristics that lead to differences in health outcomes are out of the control of providers. Others are concerned that differences in outcomes between hospitals serving socially at-risk populations and those serving a more general population reflect disparities in the provision of health care. At a broader, policy level, recent Health Affairs Forefront articles have addressed the use of social risk to adjust Medicare payments to increase support for health care providers who serve high-needs patient populations and highlighted the need to pursue both individual and community approaches to addressing SDOH. Others have proposed combining individual-level data with geographic social needs indices to adjust payment to providers and recommended a comprehensive strategy for addressing social risk in Medicare’s value-based payment programs.

However, there is much work to be done to address SDOH more comprehensively. Challenges in addressing social needs include the lack of standardized, comprehensive, and shared patient-reported social data; difficulties in identifying actionable opportunities and community resources; and challenges in implementing financially sustainable targeted interventions for high-risk patients.

The Social Determinants Data Gap

Lessons learned and best practices are starting to emerge as hospitals, payers, employers, and other stakeholders try different approaches to invest in and address health disparities within their populations with social needs. A key challenge in efforts to address social needs is the collection, standardization, and sharing of social needs information. Research suggests providers and patients experience reluctance in gathering and sharing SDOH information, respectively. California hospitals have long collected race and ethnicity data and more recently social needs information directly from patients. This is the preferred approach.

Social needs information collected directly from individuals will likely improve and become standardized over time. The CMS proposed rule, released April 18, 2022, includes a Screening for Social Drivers of Health measure and Screen Positive Rate for Social Drivers of Health measure for the FY 2023 Hospital Inpatient Prospective Payment System, with mandatory reporting for calendar year 2024. This will establish ongoing infrastructure support for social needs information to be collected directly from individuals. However, in the short term, what is urgently needed is a refined set of initial priority social and demographic measures that is more granular, standardized, and comprehensive enough to facilitate tailored interventions and service delivery innovation.

Creating A Standardized Hospital Social Needs Score

Geographic measures of social need have been used for a variety of purposes, most recently by some states to address disparities in COVID-19 prevention, detection, treatment, and vaccination. Key national indices include the Social Vulnerability Index, Social Deprivation Index, and the Area Deprivation Index. California used its own Healthy Places Index (HPI), created by the Public Health Alliance of Southern California, as a basis for equitable COVID-19 re-opening criteria and vaccine distribution.

The California HPI combines 25 social needs metrics across eight domains into an overall score. The domains are: economic, education, health care, housing, neighborhood, clean environment, social, and transportation. The Public Health Alliance of Southern California recently updated and enhanced the data and methodology. We integrated census tract-level HPI scores with hospital-specific, ZIP code-level patient origin information from the California Department of Health Care Access and Information’s Open Data Portal. This approach has the advantage of assessing hospital social needs based on the geographic area in which patients reside rather than the geographic area in which the hospital resides.

The result is a single, unique hospital-level HPI score that supports a variety of quality performance analyses, social needs comparisons, and ways to develop interventions intended to improve health equity. We ranked the 312 hospitals included in Cal Hospital Compare reporting, the large majority of which are acute general hospitals, by their unique HPI score.

As part of our methodological development, we applied the same approach to the Social Vulnerability Index and the Area Deprivation Index. Hospital-level Social Vulnerability Index and HPI results were highly correlated. Correlations were substantially lower between the Area Deprivation Index and both the HPI and Social Vulnerability Index. We note that the approach is vulnerable to ecological bias since the social needs of the individual are proxied by the social needs of their ZIP code of residence.

Variation In Hospital Social Needs Within And Across Market Areas

Exhibit 1 shows results from 14 market areas in California that are sorted roughly from Northern California to Southern California from left to right. The figure presents the hospital-specific social needs index scores and the average scores based on the hospitals within the market area.

The Central Valley in California, which includes the southern San Joaquin Valley, an underresourced region, has the highest average hospital social need. This is a predominantly agricultural area with a high migrant farmworker population. In contrast, Bay Area hospitals have the lowest social need, consistent with the population of high-tech workers in that area.

Most importantly, within these broad market areas are local areas of patients with varying social needs as evidenced by the height of the vertical grey bars in each region. For example, the Los Angeles market area includes 75 hospitals, with varying proportions of high-risk, high-social-needs patient populations. From a policy viewpoint, the implication is that it is essential to consider the social needs of local populations at the most geographically granular level possible, using patient origin data.

Exhibit 1: Hospital-level social needs index across California market area, [2019]

Sources: Hospital patient origin data were sourced from the California Department of Health Care Access and Information. Healthy Places Index data were sourced from the Public Health Alliance of Southern California. Notes: The x denotes the average hospital-level social needs index (SNI) within the market area. The grey vertical bars represent the hospital-specific social needs index scores.

Opportunities For Geographic Area-Based Collaboration

An advantage of our approach is that the social needs of hospitals can be compared and ranked using a standardized methodology and data set. This can provide policy insights and help hospitals assess their social needs and develop collaborative interventions.

For example, the hospital with the highest social needs score in California is Martin Luther King Jr. Community Hospital, a critical safety-net provider of maternity, critical, and inpatient care, and specialty outpatient services located in South Central Los Angeles (LA)—market service area 11. The hospital serves some of the most vulnerable and marginalized patients, predominantly African American and Hispanic populations. The hospital’s impact in addressing social determinants of health in the surrounding communities in South LA has been recognized. In partnership with UCLA Health, they have implemented initiatives such as Recipes for Health to address food insecurity, Know Your Basics community outreach, prevention, and screening programs, as well as integrated behavioral health screenings during triage of emergency department visits regardless of presenting symptoms.

The analysis showed that multiple hospitals serve the same high-social-needs ZIP code in which Martin Luther King Jr. hospital resides, suggesting that there may be opportunities for cross-hospital collaboration.

As a counterpoint, the hospital that ranked fifth in the state is Adventist Health Clear Lake Hospital, part of a faith-based integrated health system in rural Northern California. The hospital serves predominantly White residents living in the surrounding rural communities. Both hospitals have very high social needs but serve very different populations.

Maximizing The Impact Of SDOH Investment On Quality

We examined the correlation between hospital-level social-need index scores and hospital performance scores across 88 hospital quality, safety, cancer volume, and patient experience measures. The exclusive breastfeeding measure had the highest correlation with the hospital HPI followed by patient experience and readmissions measures. In contrast, there was little correlation with health care-associated infection rates.

The stronger correlation with some measures raises the question: Can SDOH investment be targeted to address the structures and processes underlying the measures most impacted by social need? Are there best practices to be surfaced from the hospitals whose patients have high social needs that also exhibit high performance in specific measures?

In addition, through domain scores, the indices drill down valuably into the types of social needs facing the population. For example, transportation needs are greater for the rural hospital, whereas pollution is a greater problem for the urban hospital. The domain specificity produces analytics that are much more actionable both through collaboration and focus on specific quality measures.

Prioritizing measures with strong correlations as well as significant impact on outcomes and costs may optimize both SDOH financial investment and organizational attention.

Identification of high-performing hospitals that also have high social needs may also help identify best practices that other hospitals could learn from. For instance, our analysis identified that several hospitals that performed well on the breastfeeding measure were also among the hospitals whose patient populations had the highest social needs (see exhibit 2).

Exhibit 2: High breastfeeding rates and high social needs [2019]

Sources: Hospital patient origin data were sourced from the California Department of Health Care Access and Information. Healthy Places Index data were sourced from the Public Health Alliance of Southern California. Notes: The graphic shows hospitals grouped into quartiles along the horizontal axis; those whose populations have the highest social need on the left-most side. The vertical axis shows the range of hospital performance within each quartile. As indicated by the red circle, there are several hospitals with high performance in the quartile of hospitals whose patient populations have the highest social need.

Summary

We are at an early stage in understanding how to best measure and address social needs. There is much left to learn, and the field will, no doubt, rapidly evolve. However, the approach we suggest leverages data that are available now to take meaningful action.

Multisector collaboration with hospitals in high-social-needs areas to target specific quality measures through the development of focused interventions holds the promise of maximizing the impact of hospital SDOH investment and, ultimately, improving the health status of those with the highest need.

Authors’ Note

We wish to express our deep gratitude to the Public Health Alliance of Southern California for their technical support and stimulating discussions on potential use cases.