Abstract
Screening for social needs has gained traction as an approach to addressing social determinants of health, but it faces challenges regarding standardization, resource allocation, and follow-up care. The year-long study, conducted by the Association of American Medical Colleges, integrated data from conferences, surveys, and key informant interviews to examine the integration of social needs screening into health care services within Academic Health Systems (AHS). The authors’ analysis unveiled eight key themes, showcasing AHS’s active involvement in targeted social needs screening alongside persistent resource allocation obstacles. AHS are dedicated to efficiently identifying high-risk populations, fostering partnerships with community-based organizations, and embracing technology for closed-loop referrals. However, concerns endure about the utilization of reimbursement codes for social needs and regulatory compliance. AHS confront staffing issues, resource allocation intricacies, and the imperative for seamless integration across clinical and nonclinical departments. Notably, opportunities arise in standardized training, alignment of AHS priorities, exploration of social investment models, and engagement with state-level health information exchanges. Aligning clinical care, research pursuits, and community engagement endeavors holds promise for AHS in effectively addressing social needs.
Introduction
Social determinants of health (SDOH) play a significant role in health outcomes. People with higher levels of education, income, and access to resources generally enjoy better health outcomes, whereas those with lower levels of these resources tend to experience worse health outcomes. However, despite widespread recognition of the importance of addressing SDOH, there remains a lack of consensus on the role of health systems, much less the best approaches for doing so. 1 As offered in a landmark National Academies 2019 publication on SDOH, 2 an approach that has gained popularity is to screen individual patients for specific social needs that impact their health and well-being—such as for housing, food, and transportation 3 —and to connect those with needs to appropriate resources and support services. 4 It is important to note that while identifying social needs is a critical first step, screening alone is unlikely to move the needle on health inequities and outcomes. Current approaches to screening for patients’ social risk factors have raised concerns regarding the lack of standardization of processes and procedures, burden on patients and health care providers, and limited institutional resources. 5 –10 Furthermore, there are no uniform guidelines for addressing needs once identified. 11,12 Nevertheless, payors and accrediting bodies widely view screening for social needs as necessary to reduce health disparities and achieve health equity. For example, the U.S. Preventive Services Task Force has generally included assessments of social risks as part of other preventive health guidelines, and, as part of its Fiscal Year 2023 Hospital Inpatient Prospective Payment System, the Centers for Medicare and Medicaid Services (CMS) is now including the following two social risk screening quality measures: “Screening for Social Risk Factors,” which measures the proportion of patients who were screened for social risk factors at the time of hospital admission, and “Follow-Up Care for Social Risk Factors,” which measures the proportion of patients who were provided with follow-up care services related to their social risk factors. 13
In the face of this rapidly changing environment and as health systems take a more holistic approach to health care by addressing SDOH, 14,15 it is essential to continue improving processes and measures to ensure that individuals with social needs receive the support and care they require. Academic Health Systems (AHS) are uniquely positioned to contribute to the knowledge base regarding approaches for, and challenges related to, addressing social risk factors because of their integrated delivery models across the continuum of care. However, AHS are sometimes geographically and structurally heterogeneous organizations that face unique challenges as large educational and health delivery organizations with many different and diverse departments and mission areas. As the organization leading and serving the academic medicine community to improve the health of people everywhere, the Association of American Medical Colleges (AAMC) is interested in identifying key drivers and barriers to addressing SDOH and improvement opportunities that support the needs of AAMC’s member academic hospitals and health systems, their patients, and communities. The current study describes recent efforts to better understand the state of social needs screening in AAMC member institutions, amplify promising practices, and identify common barriers. The authors also discuss opportunities for the AAMC to support improvements that can be applied across AHS.
Materials and Methods
The AAMC held a virtual Integrating Quality Forum (IQF) on June 3, 2022, to bring together health care professionals from around the United States to collaborate on the status of social needs screening in AHS (eg, for patients’ more immediate food, transportation, housing needs) 3 and to explore how this information can be used to improve patient outcomes. A total of 103 attendees from 76 organizations representing AHS, medical specialty, and hospital associations participated in facilitated conversations with thought leaders, clinical administrators, population health personnel, and researchers. Topics included partnerships needed to act on social needs data, innovative approaches to social needs screening and services, the role of technology and health information exchanges, and the impact of social needs data on care delivery and patient outcomes. Themes were identified by AAMC staff and consultants that were then used to develop a post event survey to confirm and expand on the insights gained during the conference (see Supplementary Data S1).
In July-August 2022, surveys were distributed to AHS clinical leaders, including chief medical officers and faculty practice plan leaders, who helped identify individuals in social needs roles within AAMC member institutions. Respondents (n = 46) self-identified as Administrators/Directors (41%), Clinical Providers (24%), or Executives (24%) and represented public universities (28%), other health systems (28%, eg, FQHC), private universities (17%), and government (6%). Geographic distribution included urban (65%), rural (11%), and suburban (2%). Descriptive analyses of the survey data were conducted, and findings were presented during the AAMC’s in-person Learn, Serve, Lead annual meeting in November 2022 in Nashville, TN, where additional feedback was gathered (see Table 1 for participants).
Survey Respondent Organizations and Roles Within (N = 46)
Survey participants, national meeting session attendees, and other AAMC constituents (eg, Chief Medical Officers or their institutional designee) were given the opportunity to express their willingness to participate in individual key informant interviews about their specific experiences with social needs screening at AHS and what they viewed as critical issues to address. Interview participants included 6 social needs leaders representing different AHS areas, including pediatrics, adult medicine, academic public health, population health, and health IT/informatics from multiple (ie, Northwest, West, Midwest, Northeast, South) regions across the United States. Four participants were direct clinical care providers. Interviews were conducted using videoconferencing software, lasted approximately 1 hour, and were conducted by author A.W. in December 2022–January 2022.
The authors used rapid qualitative analysis 16,17 with purposeful data reduction activities. While interviews were recorded, they were not transcribed. Coinvestigators met and reviewed notes taken during interviews. Informed by a priori interest areas, investigators used thematic analysis across qualitative data sources that also included notes from the conference discussions. Themes from qualitative data sources were then combined with descriptive quantitative survey results. Finally, in an effort to further validate themes by member checking, 18 findings were summarized and refined with additional input from clinical leaders in the AAMC Integrating Quality Initiative Steering Committee in March 2023. Feedback from the subcommittee was incorporated into the final themes, discussion points, and implications presented below.
Results
AHS are engaged in targeted screening to identify patients with social needs
Most (89%, Table 2) survey respondents report that their institutions have implemented social needs screening; however, only 28% universally screen all patients. Most systems engaged in screening do so in primary/ambulatory care settings (74%), whereas 57% conduct screening in inpatient settings. Respondents specifically named a range of pediatric (28%) and adult (41%) specialty clinics as populations targeted for screening. The most common needs screened were for food (85%), transportation (83%), and housing (80%). Although all interview participants stated that screening questions are integrated in electronic health records (EHRs), surveys indicated that less than half (47%) use standard questions such as those offered through EHR vendors, with the balance indicating using other screening questions from existing (eg, PRAPARE) or internally developed instruments.
Social Needs Screening Characteristics
There is growing interest in using data to efficiently and effectively identify populations with significant social needs
Survey respondents noted that social needs screening efforts are largely driven out of a desire to address health or health care disparities (85%, Table 3). However, forum and interview participants emphasized the need to efficiently identify high-risk populations due to limited institutional resources. Most survey respondents (72%) reported inadequate workforce for screening and for responding to social needs (74%). Notably, in an open-ended manner, one-quarter (26%) of survey respondents mentioned alternative methods used by their health systems, such as leveraging public health data (eg, Social Vulnerability Indices) and data from vendors (eg, financial creditors’ data) to target resources. Steering committee feedback also highlighted the use of “metadata” on patient social needs and data collected by health care organizations for quality improvement, which aligned with the perspectives of interview participants who noted the growing importance of metadata in addressing new CMS quality measures within health systems.
Social Needs Screening Drivers and Barriers (N = 46, Multiple Responses Possible)
There are a large number of professional roles involved in social needs screening in AHS, but new roles can create isolation and accountability gaps
Survey results confirm that there is a range of interdisciplinary professional roles responsible for ensuring that social needs screening and outreach are conducted, with social workers being the most commonly named role responsible (72%). Other roles mentioned in surveys include case managers, community health workers, physicians, nurses, physician assistants, patient navigators, registration staff, and students (See Table 4).
Staff/Personnel Responsible for Ensuring Patients’ Social Needs are Assessed (N = 46, Multiple Responses Elicited)
Interviews revealed how logistical challenges arise when roles span multiple departments (eg, department of social work, department of internal medicine) and can make it difficult to determine and establish clear lines of accountability. In particular, interviews noted challenges establishing oversight, training, and connections with others in the role. As one participant noted, it has been challenging to create an “esprit de corps” to overcome the role isolation of those engaged in what is often emotionally challenging work with patients.
Open-ended survey responses reflected some confusion of where social needs are addressed, with 20% noting screening location (eg, clinic) or population, whereas others noted workforce at a more administrative level (eg, screening done by department of social work). Interviews added additional context to these survey responses as follows: participants noted that, while social needs screening programs and outreach efforts are often housed in nonclinical departments (eg, population health or quality), screening itself is generally conducted as part of clinic service delivery. One participant noted how these new roles created and managed by one department (eg, department of population health), but with daily work being conducted in another (eg, internal medicine clinics), can lead to confusion in accountability, oversight, training and, ultimately, job dissatisfaction.
AHS are seeking ways to make access to needed social resources more timely, effective, and patient-centered but the approaches are varied
It appears that AHS have largely adopted traditional referral pathways [eg, screening in EHR by clinicians with a referral to a social worker or community health worker who conducts outreach] for patients with identified social needs. Interviews expanded by noting that responses after patients screen positive for social needs are variable, ranging from referrals to in-clinic roles (psychology, social work) funded through reimbursement and/or grants to community health worker roles funded by specialized Medicaid funding, to case management funded by departments of nursing. While seemingly infrequent, key informants also shared examples of on-site social needs resources (eg, a food bank or medical–legal partnership located within a clinic). There was robust discussion and excitement among IQF participants regarding the need for AHS to bring community resources to clinical sites.
AHS rely on partnerships with community resource providers, but are seeking ways to track outcomes and understand value
The majority of survey participants (85%, Table 2) reported referring patients to community-based organizations (CBOs) for addressing social needs, but IQF attendees and key informants communicated skepticism about CBOs’ ability to meet these needs. In fact, survey and interview participants stressed the importance of “closed loop” referrals between AHS and CBOs to effectively understand the return on investment, with 70% of survey respondents citing quality performance and improved patient outcomes as key motivators for adopting social needs interventions.
To facilitate these closed-loop referrals, participants universally called for the development and adoption of technology. While CBO “aggregators” such as 211, Findhelp, and UniteUs were mentioned as options, there were mixed feelings about their value, mainly due to cost concerns. Questions were raised about how CBO aggregators affect funding for health systems and CBOs, as well as issues surrounding patient consent, data governance, and the capabilities of state-level health information exchanges to support secure data exchanges. Key informants noted awareness of CBO referral databases embedded within EHR systems, but mentioned that their organizations typically use in-house developed community referral tracking systems managed by outreach departments (eg, Social Work, Case Management).
Departments of population health are leading social care integration in AHS, creating opportunities to integrate clinical care, community outreach, and academic partnerships but challenges remain in clarifying and communicating the mission and role of such nonclinical departments and other champion groups within AHS
Key informant interviews revealed that, while addressing social needs is well within the purpose and scope of nonclinical departments such as population health, decisions regarding how referrals and outreach are conducted are often made by clinical departments or even individual clinics. This appears to stem from the belief that screening should be matched with available clinical resources.
While departments of population health are most commonly named across all participants as responsible for social screening efforts, key informants noted some variation in approach and results. It was noted that some departments are focused on outcomes data and efficiency rather than on building CBO partnerships and collaborations necessary to be responsive to the social needs of their patients and the community. In fact, despite 35% of survey responses indicating that organizational reputation is a significant driver for social needs programs in health systems, interview and forum participants reported difficulties identifying a person or role responsible for community engagement efforts in their own AHS.
Finally, in forum discussions, open-ended survey responses, and interviews, it was recognized that academic partners in public health, social work, and nursing are engaged in addressing social needs and that medical and other health professional students are involved in outreach efforts. However, key informants noted that gaps remain regarding understanding how to best integrate SDOH teaching and research efforts with efforts in clinical operations.
AHS are concerned about using coding systems to document social needs and the lack of funding for the resources needed to screen or meet patient needs
One third (39%) of survey respondents reported using Z-codes, which are a set of diagnosis codes used to identify Medicaid patients who have SDOH or complex medical or behavioral health conditions. It has been purported that use of Medicaid Z-codes can help Medicaid agencies and health care providers better understand and address SDOH. However, there appears to be variable interest in linking social needs screening to Z-codes without additional reimbursement for screening or outreach. In addition, key informants also expressed concerns about how codes may be used to justify service cuts versus expansion due to variability in how screenings are targeted and data analyzed. An example given was that, depending on the population and methods used to screen, data could reflect artificially low social needs. Of those who report using Z-codes, half (50%) involve care provider entry with the balance handled through coding/billing services. Importantly, almost one-third of respondents (36%) whose organizations screen for social needs did not know whether or how their organization used Z-codes (Table 5).
Use of ICD-10 Z-codes (N = 36)
As shown in Table 3, in multiple ways, respondents noted that a primary barrier to social needs screening and outreach was a lack of resources. Key informants noted some success in taking advantage of targeted Medicaid funding opportunities by directing efforts toward specific resources for certain high need populations, such as for housing for women in postpartum period and patients who have had transplants. However, all noted difficulty in finding resources meeting patients’ social needs on a larger scale. In addition, although no specific programs were offered, key informants noted a general awareness of potential financial mechanisms and strategies used to fund and support social programs and organizations.
AHS are preparing to respond to state and federal regulations regarding the need to document and respond to social needs
Survey respondents noted that payer contractual requirements (41%), quality ratings (20%), and accreditation requirements (13%) drove social needs screening efforts (Table 3). Survey responses highlighted how some payor contracts (41%) also mandated social needs screening. Key informants confirmed this, mentioning state-level Medicaid programs seeking detailed approaches from health systems for addressing social needs. However, there was variability between states, and the new CMS social risk screening quality measures are expected to amplify screening and impact the intervention landscape.
Key informants raised concerns about varying investments by health systems and suggested that regulatory oversight through Community Benefit Program regulations for nonprofit/tax-exempt AHS could ensure broader social needs addressing. Because community benefit regulations vary by state, and compliance methods differ, key informants proposed standardizing these regulations to prioritize underserved populations in tax-exempt health care organizations’ community benefit activities.
Discussion
This year-long effort examined the experiences of social needs screening in AHS, using data gathered from conferences, surveys, and interviews with key informants involved in implementing social needs screening as a part of health care delivery. The authors’ findings confirmed prevalent concerns that addressing patients’ social needs requires significant investments both within and outside health systems. Investments to meet the identified needs, not only to screen for them, are what will ultimately determine the impact on health outcomes. The data also uncovered new capacity building opportunities for AHS to consider, such as improved training, engagement, integration, coordination, communication, and aligning incentives for implementing social care programs. Although different health systems may have varying approaches to overcoming these challenges, these data highlight that there are many commonalities across AHS that may spur them to take action on social needs beyond screening, which is a means to more effectively connecting patients to needed resources and supports.
Through these data, it becomes clear that AHS face significant challenges in allocating resources for staffing social needs screenings and interventions. However, the data also reveal growing interest in population and public health, which presents opportunities for addressing these challenges. The use of open-source data, in combination with community assessments and partnerships with community organizations, can help health care organizations better understand the social needs of patient populations, leading to new investments and ideas for addressing patients’ social needs. 19 For instance, AHS could use these data to direct limited resources to fund outreach programs in high-need areas 20 –22 or provide support for essential needs such as housing or food. In addition, community-level data can be leveraged to align resource needs with regulatory requirements, like community benefit regulations, 23,24 which ensure that tax-exempt health care organizations fulfill their obligations to the community by providing benefits that address the community’s health needs and are transparent and accountable in their provision with the health needs of the community. 25
There is a nationwide effort to link social needs assessments to regulatory requirements, particularly as part of Medicaid programs. 26 However, the data also reveal widespread concerns about the challenge of obtaining reimbursement for social needs using supplemental ICD-10 “Z-codes.” Despite the potential for these codes to capture social and environmental factors affecting a patient’s health, such as education and literacy or lifestyle and personal health factors, the data show that there is a lack of enthusiasm for their usage. It is well known that medical diagnosis coding is affected by inaccuracies, inconsistencies, and institutional variability in resources to track trends in patient care and outcomes. As such, it will be important to ensure that Z-codes not reflect the capability of institutions to accurately integrate the codes rather than the true prevalence of population social needs.
A trend consistently noted throughout the data is how AHS are seeking ways to understand the benefits of addressing patients’ social needs as part of clinical care. Forum and survey participants indicated that their systems are looking to contracts with CBO aggregators or other organizations or groups that work to connect, coordinate, and collaborate with multiple CBOs. CBO aggregators provide a wide range of services but, for health systems, their importance appears to rest with the CBO aggregators’ ability to provide technical assistance to improve AHS’ capacity to bridge data exchange between the health system and CBOs and collecting and analyzing data on the effectiveness of AHS programs and services to inform health system program development, evaluation, and investment. However, the data revealed concerns about CBO aggregators as follows: open ended comments in surveys specifically questioned their value, and key informants expressed concerns about whether they divert needed resources. Furthermore, key informants noted a need for clear data governance and patient consent process efforts to make referrals “closed loop.” This aligns with nationwide efforts to ensure that data exchange is possible and standardized, while also adhering to privacy laws. 27
It is clear that AHS are dedicated to addressing social needs and reducing health disparities. They are investing heavily in talent, time, and technology to address patients’ social needs, but most efforts still focus on traditional funding models through assessments, referrals, documentation, and reimbursement. 14,15 However, the data emphasize the complexity of addressing social needs that may require a shift from traditional metrics, analytical approaches, and policy analyses toward models requiring facilitating engagement and tracking outcomes over time. 28 –31 Outcomes such as service utilization may not be sensitive enough or appropriate for many health disparity populations. Furthermore, while advocating for reimbursement is important, interviews revealed that more of these traditional efforts for gaining resources should be coupled with increased investment into social programs. Examples include impact investing, social bonds, pay-for-success models, community development financial institutions, social enterprises, and community shares. These programs aim to generate measurable social and environmental impact while also providing a financial return to investors 32,33 and, while there is some debate about their use, may be an opportunity for AHS to explore.
There are a number of champion groups leading efforts to address social needs in AHS and that this is appropriate given the expansive complex goal of improving the health of populations and reducing health disparities through the provision of social resources. 34 However, there appears to be wide variability in the specific roles of the clinical, nonclinical, and academic departments involved where some departments charged with social care integration may not have decision making authority in clinical operations where social needs screening and outreach most commonly takes place. Furthermore, it was noted that a strong focus on system-level program development and evaluation may sometimes be at odds with the realities of roles and funding enacted at the clinic level, something that has been noted as a challenge with the adoption of departments of population health by AHS in general. 35 Finally, participants noted a lack of clarity regarding where the responsibility for community engagement is within an AHS and seen as needing further delineation and development. 36 Effectively integrating social services and clinical care requires expertise and collaboration spanning clinical care coordination, health policy, population health management, advanced data analytics, and community engagement. Given the challenges of coordinating the diverse functions involved in social needs efforts across different settings within an AHS and into the community, it is not surprising that there are many opportunities to continue improving communication and coordination among all champion groups.
Conclusions
The current study examines social needs screening within AHS and highlights the multifaceted challenges and promising opportunities in addressing patients’ social needs. AHS are dedicated to this endeavor, but face hurdles related to resource allocation, reimbursement codes, 37 –39 data integration, and coordination among various departments. The integration of social care into clinical practice is a complex process that requires ongoing adaptation and collaboration. These findings underscore the need for standardized training, alignment of priorities, exploration of social investment models, and engagement with state-level health information exchanges. 8,40,1 In addition, efforts to bridge gaps between clinical, nonclinical, and academic departments within AHS and clarify community engagement responsibilities can enhance the effectiveness of social needs initiatives. 34 –36 While AHS are pioneers in health care, there is no one-size-fits-all approach, and these insights provide valuable guidance for AHS as they continue to refine their strategies for addressing SDOH.
Relevant Reporting Checklist
As per the EQUATOR Reporting Guideline Decision Tree, the authors used the COREQ checklist for reporting the Qualitative aspect of this study.
Footnotes
Data Availability
The deidentified data that support the findings of this study are available on request from the AAMC. The data are not publicly available due to privacy or ethical restrictions.
Ethics Approval
This study was approved by the University of Utah Institutional Review Board (IRB# 00154704).
Authors’ Contributions
J.B. Conceptualization, Methodology, Resources, Writing—Review & Editing, Supervision, and Project Administration. A.W. Conceptualization, Methodology, Software, Formal Analysis, Investigation, Resources, Data Curation, Writing—Original Draft, and Visualization. R.M.: Conceptualization, Writing—Review & Editing, Supervision, and Funding Acquisition.
Author Disclosure Statement
Dr. A.W. discloses: consulting fees paid by Association of American Medical Colleges, payment/honoraria paid by Northwell Health and Boston Children’s Hospital, and meeting/travel support for NIH and PCORI Study Sections. Other authors have no disclosures to report.
Funding Information
This study was supported by the Association of American Medical Colleges. The content is solely the responsibility of the authors and does not necessarily represent the official view of the AAMC.
Supplementary Material
Supplementary Data S1
Supplementary Data S2
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
