Abstract
Background:
Collecting patients’ social determinants of health data allows primary care providers to identify needed services that can address patients’ social needs and improve their overall health. Patients’ social needs may also be addressed through collaborations between legal and primary care teams, known as medical-legal partnerships (MLPs). We sought to understand patients’ experiences with social needs screening (SNS) and MLPs in primary care.
Methods:
In 2024, we conducted a web-based survey with a national sample of 502 U.S. adults who had visited a primary care provider in the prior year. The survey assessed patients’ experiences with SNS, preferences of SNS delivery and referrals, and perceptions of MLPs. Analyses used multivariable logistic regression models to assess correlates of willingness to use MLPs.
Results:
Thirty-eight percent of participants (n = 194) reported having completed a SNS. Of those who completed a SNS, 51% were offered a referral to address their social needs, including 19% who were referred to legal assistance. Approximately 53% of all participants said they were willing to use MLPs. Participants were more willing to use MLPs if they live in rural communities (Odds Ratio [OR] = 1.79, 95% Confidence Interval [CI]: 1.07-3.01) and were aware of SNS (OR = 1.47, 95% CI: 1.05-2.06) and receive public benefits (OR = 0.67, 95% CI: 0.48-0.93).
Conclusion:
MLPs are perceived favorably among patients. Establishing MLPs in rural areas may help address social determinants of health in these communities. Next steps in this research include understanding how to build trust in legal representatives in communities with negative experiences with the justice system.
Introduction
Social determinants of health (SDOH) account for at least 50% of a person’s health outcomes.1,2 Health care organizations, understanding its impact, began collecting patient information related to SDOH, otherwise known as unmet health-related social needs, in the form of social needs screening (SNS). While SNS standardize the collection of SDOH data with the promise to provide better care to patients, the uptake of SNS is low within these settings, despite Centers for Medicare and Medicaid Services (CMS) requirements.3-5 When social needs are captured using ICD-10 Z codes, the CMS data show a 1% utilization rate. 5 When other screening methods like the Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences (PRAPARE) tool is used, SNS utilization rates are, at most, 30%.3,4
Previous research has shown that patients are amenable to SNS, however, they express concerns related to data privacy and stigma. 6 The primary concern with data privacy is how this data will be stored and used.7,8 Specifically, patients want their SNS information to be secure to avoid other members of the community learning about their social needs or facing discrimination based on their SNS information. 7 The second concern, stigma, has shown to be a driving reason for patients neither reporting their social needs nor accepting a referral. 9 Currently, people of color and those experiencing poverty have a higher prevalence of social needs, and stigma can further prevent them from receiving the help they need.6,9
Patients’ perceptions are mixed on the appropriateness of SNS in the primary care setting. Most patients feel that the tool is appropriately placed in health care settings7,10 and are willing to speak with their providers about their social needs and accept assistance through patient navigation.11,12 On the other hand, some patients express that providers are not required to assist them with their social needs. 13
Following SNS, referrals connect patients to various resources that can mitigate the identified social need. One of the lesser-known referral resources is medical-legal partnerships (MLPs). MLPs assist patients by providing legal assistance and representation to address social needs. 14 When used, MLPs have shown improvements in patients’ health outcomes, including improvement in stress levels, reductions in hospital visits, and fewer missed medical appointments.15-18
One major limitation in the study of the use of SNS, including MLPs, is that they have only been evaluated in limited capacities by focusing on identified disparate communities or operating in individualized practice areas. Therefore, it is important to understand patients’ experiences and preferences related to SNS more broadly. Further, MLPs have been understudied despite their potential to address social needs comprehensively. Therefore, using a national sample of U.S. adult patients, the present study sought to (1) understand patient experiences with SNS, (2) assess their preferences related to SNS, and (3) evaluate whether MLPs are a viable resource from the patient perspective.
Methods
Participants and Procedures
Survey questions were developed based on literature, adapted questions from other surveys, and feedback from primary care providers (e.g., physicians, nurses, and social workers). Questions from the survey are available by contacting the corresponding author. Participants were members of an existing market research panel of U.S. adults maintained by Qualtrics, a commercial software and survey research company. The panel employs a range of recruitment methodologies to produce a sampling frame that is not overly reliant on any distinct demographic group but is designed to mirror the national representation percentages. 19 Invitations to our online survey were emailed to a random sample of panel members, and participants responded by visiting the survey and completing the eligibility screener. Eligible participants were adults between 18 and 64 years old, living in the United States, who have visited a primary care provider in the last year. We selected the upper age limit of 64 years old because the number of services that are available to the elder population of 65 years old and older increases due to their retirement age and eligibility for social security benefits. Primary care provider was defined as an outpatient practitioner, excluding long-term care and rehabilitative care providers. As this was an exploratory study, we did not conduct an a priori power calculation. Instead, our sample size of 502 was determined based on methodological and practical considerations. The first of these is that a sample of 500 or more participants allows for meaningful descriptive analysis of subgroups while maintaining adequate cell sizes for stable estimates. The second is that the sample size balanced the need for adequate data for a doctoral student-led project with available budget and timeline. Eligible adults provided implied consent and completed the survey in November 2024. The average completion time for the survey was 8 minutes. Prior to distributing to the Qualtrics panel, the survey was pilot tested among a small sample of individuals.
This study received Institutional Review Board approval on August 8, 2024.
Measures
Demographic Variables
Sociodemographic variables included age, ethnicity, race, and sex of participants. The survey also asked about the reason for the participant’s last primary care visit (i.e., well visit or annual physical, sick visit, other), their health insurance type, annual household income, whether they receive public benefits, and zip code (categorized by urbanity or rurality). Levels of urbanity are based on the USDA’s Rural Urban Commuting Area (RUCA) codes. Participants’ zip codes were used to link to RUCA codes, which were then classified as rural or urban. Urban was classified as RUCA codes 1-5 and rural was classified as RUCA codes 6-10.
Experience Using Social Needs Screening
The survey assessed SNS experience in 2 ways: (1) offered a SNS at their last primary care visit (yes/no) and (2) offered a SNS in the last year (yes/no). Based on whether participants were offered a SNS, they were asked whether they completed the SNS (yes/no). Those who had filled out a SNS were asked about the social needs included in the screening. Only participants who had not completed a SNS were asked whether they thought their primary care clinic should implement a SNS. Participants who agreed with implementing SNS into their primary care clinic were asked which social needs their provider should screen for from a list of 10 options: food insecurity, housing, health insurance coverage, childcare, legal counseling, immigration, transportation, utilities, domestic violence/safety in the home, and other.
Values Regarding Social Needs Screening
Participants who had filled out a SNS or believed that SNS should be implemented into their primary care clinic were asked whether they believe that SNS (1) strengthened the relationship with their provider, (2) gave the provider a better understanding of their health needs, (3) gave the provider a better understanding of their social needs, and (4) improved their health care experience. All items used a 5-point Likert scale, ranging from 1 = strongly agree to 5 = strongly disagree.
Social Needs Screening Data Collection
The survey assessed how participants filled out their SNS and their preferences regarding best practices for filling out a SNS. Participants who have filled out a SNS were asked how they have completed their SNS. Response options included (1) using the patient portal, (2) through an online form, (3) completing a paper form, (4) clinical staff asked the questions, and (5) other. Participants who responded that the “clinic staff asked the questions” were then asked to identify the clinic staff member who asked them the questions. Participants were asked to identify when they completed the SNS including (1) before the appointment, (2) in the waiting room, (3) in the exam room, and (4) after the visit. Using the same time frames, all participants, irrespective of their history in filling out a SNS were asked their preferred administration method and time frame for completing a SNS. Lastly, participants identified how often they preferred to complete a SNS including (1) at every visit, (2) 2 or 3 times a year, (3) once a year, and (4) only when they ask to do so.
Social Needs Screening Use During the Patient-Provider Interaction
The next set of items queried how patient’s SNS was used by their provider. Participants who had filled out a SNS were asked to identify which clinic staff member talked to them about their SNS responses. The clinic staff members included the physician, nurse, and social worker. Participants were also able to select no one. Those who selected clinic staff estimated how long the discussion of their social needs lasted. Following this response, participants were asked to say whether the amount of time discussing their social needs was enough. Finally, participants who had ever filled out a SNS were asked to rate their comfort in discussing their social needs with their provider, perception of their provider’s comfort discussing their social needs, and their comfort with the SNS information being included in their medical records.
Social Needs Screening Referrals
The survey asked participants who had filled out a SNS about their preferences related to referrals. Those who did use the referral were asked to select which resources they were referred to including food pantry or grocery store gift card, emergency shelter address, and gas card/ride share gift card. Response options were crafted based on their ability to address various social needs like food insecurity, housing, and transportation issues. Those who did not use the referral were asked to provide a reason for not using it. Response options included: (1) social need was not an emergency concern, (2) already receiving help for the social need, and (3) did not want help with a social need. Participants who were not offered a referral were asked to rate their willingness to receive a SNS referral. Next, participants were asked their preferred method to receive the referral. Using the same response options to participants who did receive resources, these participants identified the resources to which they would like to be referred. Finally, using a 5-point Likert scale, all participants rated their preferred referral delivery method, including (1) the resource being available in the primary care office, (2) the provider provides information about the resource to the patient, (3) the provider contacts the resource on the patient’s behalf, and (4) social workers or case managers connecting the patient to the resource.
Medical-Legal Partnerships
The final set of items asked participants about their opinions related to MLPs. After defining MLPs as “a resource where patients with social needs can receive free or low-cost legal services,” participants were asked to identify their comfort talking with someone from an MLP about their social needs, with responses ranging from 1 = very uncomfortable to 5 = very comfortable. Then participants were asked, in general, what types of social needs they think MLPs can assist with, using the same list of social needs asked in prior sections of the survey. Participants were able to select “no, I am not interested” as one of the above options. Penultimately, participants were asked about their willingness to seek an MLP to help with their social needs, with options ranging from 1 = very unwilling to 5 = very willing. Lastly participants were asked to identify whether they believed MLPs can assist with (1) addressing their social needs, (2) improving their overall health, (3) providing long-term solutions to their social needs, (4) improving their quality of life, and (5) finding a solution to their social need quickly.
Data Analysis
Sociodemographic variables were reported for the full sample using frequencies and means. Participants were dichotomized based on whether they completed a SNS or not. Differences between these 2 groups were assessed using the Fisher’s Exact test. All variables, irrespective of their statistical significance in the bivariate analysis, were included in the logistic regression models for adjustment (e.g., race/ethnicity, sex, annual income, locality). We conducted separate multivariable logistic regressions to assess the associations between participant sociodemographic factors and use of SNS and willingness to use MLPs. SNS receipt status was also included in the logistic regression model for MLP. We reported Odds Ratios (OR) and their 95% Confidence Intervals (CI). Analyses were conducted using SAS and all statistical tests were performed at the 0.05 significance level. Accounting for the skip patterns in the study survey, statistical models only included participants that were eligible to respond to the survey items assessed.
Results
Participant Characteristics
A total of 502 adults responded to the survey. The mean age of the participants was 45 years (SD: 13.4 years). There was a balanced distribution of sex with 51.3% of participants being female. Most of the participants were non-Hispanic White (68.2%), followed by non-Hispanic Black (13%) and Hispanic (12.9%). Most participants (61.2%) reported making less than $50 000 per year. Geographically, most of the participants live in urban areas (89.2%) based on the locality. Participants were split based on whether they have ever received a SNS (38.7%) or have never received a SNS (61.3%; Table 1). Statistically significant differences between the two groups were found when comparing race/ethnicity, income, health insurance status, and public benefits recipient status.
Participant Characteristics (n = 502).
Of those who had never been offered or filled out a SNS, 36.8% believe that SNS should be implemented in their primary care office. The most common social needs that participants felt should be screened for was consistent between the two groups with differing prioritization. Participants who have filled out a SNS report actually being screened for housing, food insecurity, and health insurance coverage most often (Table 2). Participants who had ever filled out a SNS or believed SNS should be implemented into primary care prefer being screened for housing, health insurance coverage, and food insecurity, respectively.
Social Needs Screening Categories by Actual, Preferred, and MLP Assistance Ability.
Participants can select more than 1 option.
Only participants who responded in the affirmative of filling out a SNS or believing SNS should be implemented were asked to answer this question.
Social Needs Screening Completion
SNS completion methods experienced by participants and preferred methods of completion varied (Table 3). Of the participants who were screened by clinic staff, 33% (5/15) of them were asked by nurses. Those who preferred clinical staff screen them for social needs most requested nurses (56%, 62/111), physicians (54%, 60/111), and physician assistants (42%, 47/111) as the individuals completing the screening.
Actual Experience and Preferences for Social Needs Screening Administration.
Only participants who responded in the affirmative for being asked to fill out a SNS in the last visit, the last year, or filled out the SNS.
Eligibility to answer these questions hinged on whether the respondents either reported completing a SNS through a member of the clinical staff asking the questions or respondents preference that a member of the clinical staff asks the questions.
Patient-Provider Interaction with Social Needs Screening
Physicians (38%, 75/194), physician assistants (30%, 59/194), and nurses (25%, 48/194) were the top 3 primary care providers who participants reported talking with them about their social needs. Forty-two percent (81/194) of participants report their primary care provider spending between 3-5 min discussing their social needs. One hundred thirty-seven participants (70.6%, 137/194) agree that the amount of time spent was enough. Forty-eight participants (24.6%, 48/194), irrespective of the amount of time spent discussing their social needs, wanted more time. Participants who had never filled out a SNS (n = 308) preferred to discuss their social needs with their physician (56%) as opposed to nurses (29%), physician assistants (28%), social workers/case managers (27%), and nurse practitioners (26%).
Referrals
Of those who have completed SNS, approximately half (51%, 99/194) of them were offered a referral. Among those participants who received a referral, 78 of them (79%) used the referral. The most highly indicated need was food pantry information or a grocery store gift card (48%, 48/99). Only 19% of participants (19/99) were referred to legal assistance. Among the entire sample of participants, 90.4% (454/502) said that they were willing to receive a referral. When asked how they would prefer to receive the referral, over half (52%, 236/454) of participants prefer a note in the patient portal connecting them to the resource. The most preferred referral resources were food pantry or grocery store gift card (60%, 273/454), gas card or ride share gift card (35%, 158/454), emergency shelter information (20%, 90/454), and legal aid (20%, 93/454).
Medical-Legal Partnerships
Most participants reported feeling comfortable talking with someone from an MLP about their social needs and 53.1% of participants would seek out MLP services to address their social needs. The top 3 social needs participants believed MLPs could help with were food insecurity (41.3%), housing (34.2%), and health insurance coverage (30.6%) (Table 2). In the multivariable model (Table 4), participants who live in rural areas have 79% higher odds (OR = 1.79, 95% CI: 1.07-3.01) of reporting willingness to use MLPs than those who do not live in rural areas. Similarly, participants who completed a SNS have 47% (OR = 1.47, 95% CI: 1.05-2.06), higher odds of reporting willingness compared to those who did not complete it. On the other hand, non-Hispanic Blacks and those who receive public benefits have 38% (OR = 0.62, 95% CI: 0.38-1.00) and 33% (OR = 0.67, 95% CI: 0.48-0.93) lower odds, respectively, of reporting willingness to use MLPs compared to their counterparts.
Correlates of Patients’ Comfort Using and Willingness to Use MLPs.
Discussion
This study aimed to understand patient experiences with and perceptions of SNS and willingness to use MLPs in a sample of U.S. adult patients. Results from the present study converge with current literature surrounding patient experiences with SNS in two ways. First, the sample of those who had ever received and filled out a SNS matched the prevalence outlined in the literature whereby between 20-30% of patients are screened for social needs.4,20 Solutions to address these low screening rates include ensuring privacy of the information, patient autonomy, and clear education for patients about the screening process. 21 More research should be done on the effectiveness of these proposed solutions through implementation science frameworks. Second, the most common social needs being screened for, food insecurity, health insurance, and housing, match the topics for which most patients prefer to be screened. Additionally, the most urgent social needs identified by health care facilities, match the same social needs identified in our study.5,22 This shows congruence between the preferences of the patients and the needs observed by primary care providers. Overall, patients in our study and literature perceive SNS positively when they have actually completed a screening. For other people who have never completed a SNS, they do not see the value in screening practices.
The major finding of the study included statistically significant information about participants from rural settings. Patients from rural communities tend to be more willing and more comfortable seeking legal representation to address their social needs. Despite this important finding, accessibility is a persistent issue that can limit rural patients obtaining this service. Research by the Legal Services Corporation (LSC) found that 94% of people seeking legal assistance on civil issues did not receive any or enough help due to their low availability. 23 These “legal deserts” make it difficult for rural communities to find legal representation for their social needs. 24 To mitigate the prevalence of legal deserts, courts, legal aid organizations, and law schools have begun offering remote services or mobile law offices. 25 The former approach, unfortunately, can prove moot because of internet connectivity issues. 25 This lack of accessibility directly connects to help as one of the most prevalent problem areas in civil legal problems in healthcare. 23 Battling these access issues requires an investment in internet connectivity for rural areas and leveraging partnerships between legal representatives and community-based organizations like integrating MLPs into local health care facilities.
MLPs exist in various settings like law schools, legal aid organizations, and primary care centers. According to the National Center for Medical-Legal Partnerships, there are at least 163 MLPs that operate in Health Resources and Services Administration (HRSA)-funded health centers. 26 More research needs to be done on how to successfully integrate MLPs into primary care settings. Integration into the patient workflow needs to be comprehensive, starting with screening practices for social needs to the referral protocol. Better integration could not only increase visibility of trust of legal representation but also increase access to legal representation to address social needs.
Strengths and Limitations
The strength of this study includes the national reach of the survey. Participants were from a national sample of U.S. adults who have received primary care recently. The survey respondents included members from multiple racial/ethnic subgroups across the US. In addition, this is among the first studies assessing willingness to use MLPs from a national sample of individuals engaged in primary care.
The study also has several limitations. One is that the survey did not collect qualitative data, limiting our ability to gather more context behind their responses. Second, the representative sample by race and ethnicity is not identical to the current national averages. While the non-Hispanic Black population is proportional to the current US population, the Hispanic sample is not. Third, the study questionnaire does not assess whether participants understand the definition of an MLP. This limits our understanding of their comprehension of legal representation for social needs. Fourth, the data analysis does not include interactions between different demographic variables. Finally, as this study relied on self-reported experiences of the participants, there could be issues related to recall bias. Additionally, relative to recall bias, people with some of the highest social needs may not be regular utilizers of primary care. Other biases exist such as sampling and interest biases. Sampling bias may exist where Qualtrics targets specific participants with survey invitations. Interest bias may have occurred and led to a skewing of the data based on participants’ approval or disapproval of the SNS concept.
Conclusion
Results from this study suggest that, upon exposure, patients are willing to participate in SNS as part of their primary care. Information from SNS can be used to connect patients to community resources and as is the resource of interest, legal services that patients believe can assist with a host of social needs like housing instability, food insecurity, and health insurance issues. These are the top 3 concerns of primary care patients in this study. More research needs to be done on ways to make SNS more accessible to patients by identifying best ways to integrate these tools into the patient workflow, specifically focusing on legal intervention as a referral to addressing the identified social needs.
Footnotes
Acknowledgements
The project described was partially supported by a Translational Science Fellowship award to Jamelia Graham for her doctoral work.
Ethical Considerations
The protocol for this study was approved by the Pennsylvania State University’s Institutional Review Board on August 8, 2024 (STUDY00025635).
Author Contributions
• Jamelia Graham – conceptualization, data collection, data analysis, drafting manuscript, and editing manuscript
• Jennifer L. Kraschnewski – drafting and editing manuscript
• Sophia Allen – drafting and editing manuscript
• Medha Makhlouf – drafting and editing manuscript
• Josheili Llavona-Ortiz – conceptualization, data analysis
• William Calo – data collection, data analysis, drafting manuscript, editing manuscript
Funding
The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: Jamelia Graham received a Translational Science Fellowship award supported by the National Center for Advancing Translational Sciences, Grant U54 TR002014-05A1. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data available on request from the authors.
