Uncategorized

A health-related social needs referral program for Medicaid beneficiaries treated in an emergency department

a b s t r a c t

Background: social determinants of health play an important role in health outcomes. This study sought to evaluate the effectiveness of a SDH screening and health-related social needs (HRSNs) referral program in an emergency department (ED) setting with adult Medicaid beneficiaries.

Methods: Between November 2016 and March 2017 we enrolled adult Medicaid patients in a prospective cohort study. Research assistants (RAs) completed an SDH screening survey with participants and asked them if they needed assistance with HRSNs related to medical, behavioral health, wellness, housing, food, legal and job train- ing issues. RAs referred participants to community-based organizations (CBO) for their top three HRSNs. Patients referred to at least one CBO were phoned a month later to determine whether their HRSN was addressed and CBOs also reported their assistance rates within four months of the ED visit.

Results: Of the 505 patients enrolled, 69% were female, 82% completed high school, and 57% reported working. Most participants (85%) requested assistance for at least one HRSN. Almost half (44%) receivED referrals to three different agencies. Help with housing (70%), medical issues (51%), and finding food (42%) were the most common. Among the 430 subjects referred to >=1 agency, 76% completed the follow-up interview. Few patients reported receiving help from the referral agencies (5% for a wellness program to 15% for medical services). Refer- ral agencies generally reported even lower assistance rates (0% for job training to 17% for medical services).

Conclusion: The majority of adult Medicaid patients treated in our ED wanted assistance with one or more HRSN. The passive referral system we implemented resulted in few patients receiving assistance from the referral agency, regardless of whether measured by self-report or by agency.

(C) 2021

  1. Introduction

The influence of the health care system on health is modest com- pared to the role of social determinants of health (SDH) (i.e. socio- economic status, living and working conditions, health behaviors, etc) [1,2]. Comparisons of social to medical care spending ratios demon- strate better health outcomes among countries and states that spend more on social compared to medical care services [3,4]. Recognizing that SDH must be addressed to improve health outcomes, an increasing number of health care organizations, health care systems and individual providers are assessing and attempting to improve their patients’ social and economic circumstances by addressing their health-related social

? Meetings: This abstract was presented at the 2018 Society for Academic Emergency Medicine Annual Meeting and the 2018 Academy Health Annual Research Meeting.

* Corresponding author at: Paige Kulie can be reached at 2120 L Street, NW, Suite 450, Washington, DC, 20037, United States of America.

E-mail addresses: [email protected], [email protected] (P. Kulie).

needs (HRSNs) [5-7]. HRSNs, such as housing instability, food insecurity, and lack of transportation, are caused by adverse SDH and may be alle- viated by referring patients to the appropriate social service agency

(s) in the community [8].

A number of Intervention studies have evaluated the effectiveness of addressing HRSNs in the healthcare setting. These studies have varied considerably in terms of study design, intervention intensity, patient population and outcomes measured. The most common study designs have been cohort [9-11], quasi-experimental [12] and randomized con- trolled trials [13-16]. The HRSN interventions have ranged in intensity from a one-time provision of written resource information [9] to patient navigation and/or social worker services on single [16] or multiple occa- sions [10,11,14,15]. Some studies have focused on a specific patient population such as high utilizers [12] or the publicly insured [13] whereas others have included all patients that seek treatment at a healthcare setting [9-11,14]. Most studies have measured patient- reported outcomes including resource assistance [9,10,16] or health [14], rarely have studies included objective measures of health [11] or

https://doi.org/10.1016/j.ajem.2021.03.069

0735-6757/(C) 2021

healthcare utilization [12,15]. Given the methodological heterogeneity of the studies conducted to date, it is not surprising that the evidence of the effectiveness of HRSN interventions has been mixed.

Relatively few HRSN intervention studies have been conducted in the ED setting [17,18], despite the high prevalence of HRSNs among pa- tients who seek care in the ED. [19-21] Wallace et al. conducted a feasi- bility study on a convenience sample of 210 ED patients to determine whether they could develop a screening and referral program [17]. While the results support program feasibility, they note that universal HRSN screening and referral in the ED setting would require staff train- ing, integration of screening results into clinical-decision making and technology that supports bidirectional communication between EDs and community-based organizations (CBOs). In another study aimed at addressing ED patients’ social needs, Losonczy et al. employed a more intensive intervention, similar to a Health Leads model [22]. Vol- unteer undergraduate students screened ED patients for HRSNs and on intervention days referred 154 participants to community resources or consulted with a social worker or lawyer if more intensive interven- tion was required. On non intervention days, 305 participants selected as controls received usual care. Among the 94 intervention and control participants successfully interviewed six months later (20%), there was no significant difference in the resolution of the primary HRSN by study group [18].

The purpose of this study was to determine whether we could suc- cessfully address a substantial proportion of adult ED patients’ HRSNs by offering them a referral to CBOs who provided the appropriate ser- vices. Because a higher proportion of patients of Lower socioeconomic status have greater HRSNs compared to patients with higher income, our study focused on Medicaid beneficiaries [9,10]. Based on previous studies that reported a third or more of patients receiving resource as- sistance [10,12,23], we hypothesized that our passive HRSN referral sys- tem would result in at least one-quarter of participants getting at least one HRSN fulfilled within a month of the index ED visit.

  1. Methods
    1. Study design and setting

We conducted a prospective cohort study to assess the effectiveness of a passive HRSN referral program in an ED setting for adults insured by the District of Columbia (DC) Medicaid program. We conducted our study at an urban, academic ED that has an annual volume of approxi- mately 71,000 visits and an admission rate of 18%. The ED is staffed by attending and resident physicians, mid-level providers, nurses and technicians. Approximately one-third of patients treated at this ED are insured by Medicaid. This study was approved by the Institutional Review Board.

    1. Selection of participants

To be eligible for the study, patients had to be 18 years or older, have a non- life-threatening Emergency Severity Index score, be insured solely by DC Medicaid and approved for ED discharge. Patients were ex- cluded if they were unable to understand consent, were non-English speaking, were also insured by Medicare, or did not have access to a phone.

Trained research assistants (RAs) screened and enrolled eligible pa- tients between 9 AM and 10 PM Monday-Friday and between 9 AM and 6 PM Saturday-Sunday from November 7, 2016 until March 31, 2017. Because multiple studies were being conducted at the ED simulta- neously, it was possible for a RA to miss an eligible patient while enroll- ing for another study so our study is a convenience sample.

Patients who enrolled in the study signed a written consent form and agreed to complete an SDH assessment survey with a RA. They also gave the study team permission to abstract relevant study informa- tion from the patient registration system (i.e. date of birth, sex, and

Medicaid beneficiary number), to obtain a copy of their Medicaid claims, and to share participant contact information with the referral CBOs. At the time of study enrollment, participants provided their email address, two telephone numbers and their residential address.

    1. Measurements

The data for this study come from patient interviews, the collaborat- ing CBOs and DC Medicaid claims. We obtained a copy of the DC Medic- aid claims for all participants for four months after the enrollment visit from the Department of Health Care Finance, the DC government agency responsible for the DC Medicaid program so that we could mea- sure healthcare utilization post enrollment.

Prior to ED discharge, a RA administered an SDH screening and HRSN questionnaire to each participant (see supplement for copy of question- naire). The purpose of the questionnaire was to screen subjects for SDH risk factors, to identify HRSNs and to document referrals made to com- munity agencies to address those needs. Our survey was guided by the World Health Organization (WHO) SDH Conceptual framework [24] which distinguishes between structural and intermediary SDH. Accord- ing to the WHO model, structural determinants of health such as race and gender, education and occupation determine one’s socioeconomic status . SES, in turn, influences intermediary SDH such as material circumstances, health behaviors, and psychosocial factors.

For structural determinants of health, our survey included questions about highest level of education achieved [25] and usual major activity. For intermediary SDH, we asked participants about their material cir- cumstances including questions about food insecurity [26], stable hous- ing [27], substandard housing conditions [28], material hardship [29], and neighborhood safety [28]. We also asked about health behaviors such as usual source of medical care [30], current alcohol use, drug use [31], and/or Tobacco use [32]. The questionnaire also included questions about psychosocial factors such as social support, marital status, whether living with children, history of spending time in jail or prison, and depression [33].

In addition to measuring SDH, we asked participants whether they needed assistance with any of the following health-related needs:

(1) medical (i.e. transportation to a medical appointment, finding a doctor, dentist or case manager, and assistance with managing their medications); (2) behavioral health (i.e. finding counseling and/or sub- stance abuse treatment services); (3) wellness (i.e. finding a wellness program); (4) food (i.e. finding a food pantry); (5) housing (i.e. paying utilities, paying rent/mortgage, resolving issues with a landlord, evic- tion, finding affordable housing and addressing substandard housing conditions); (6) legal issues; and (7) job training. After they told us which needs they wanted assistance with, we asked them to rank their top three HRSNs.

    1. Procedures

Based on the unmet needs identified, the RA then referred the partic- ipant to up to three different agencies based on what the top three unmet needs were and the services provided by the participating CBOs. The RAs recorded the CBO referrals in the study database for each participant. Prior to starting the study, we identified 10 community agencies that addressed different types of health-related social and medical needs who agreed to accept referrals from us and to confirm whether those we referred obtained services from the agency (see Fig. 1). For example, we partnered with Housing Counseling Services, an agency which helps low-income people with finding affordable housing, housing condition problems, eviction, etc.

The RAs made the referrals to each partner agency through Aunt Bertha, an online search engine that allows people to identify community-based agencies in their area that provide health-related so- cial services [34]. We used Aunt Bertha to send our referrals to our se- lected community organizations and to provide participants with

Table Church Food Pantry

Behavioral Health

Department of Behavioral Health

Unity Health (FQHC)

Referral Agencies

Legal

Neighborhood Legal Services Program

UDC Workforce Development

Allen Chapel Food Pantry

AmeriHealth (MCO)

Foggy Bottom Food Pantry

Housing Counseling Services

Our Door

Housing

Job Training

Food

Wellness

Medical

Fig. 1. Illustration of the 10 community-based organizations that we referred our participants to for their health-related social needs.

relevant agency information. RAs entered the participant’s email ad- dress into the Aunt Bertha website and selected the appropriate agency for the identified need. The participant then received an email with the contact information for the agency and the agency received an email with the referral which included the participant’s name and reason for referral. If a participant did not have an email address or did not wish to give his email address, a dummy email address was used so that the agency still received the referral information.

All referred participants also received a hard copy of the agency con- tact information which the RAs printed from the Aunt Bertha website and gave the participants prior to ED discharge. The Aunt Bertha print out contained the agency name, address, phone number, hours of operation, and a small map of the location of the agency. If a patient did not list any needs, a referral was not made, and the patient received no further information.

    1. Follow-Up

Participants referred to one or more agencies received a telephone call approximately one month after their index visit to determine if their HRSN(s) had been met. The RAs attempted to contact each re- ferred participant up to seven times before the participant was consid- ered lost to follow-up. For each type of need, the RA asked the participant if they had gotten that HRSN resolved. If the participant said yes, we asked the participant how it was resolved (i.e. with help from the referral agency, with help from family/friends, on their own, etc.). If the need was still unmet, we asked the participant whether they had contacted the referral agency to help them. If they said yes, we asked what happened (i.e. left message, scheduled an appointment, etc.) and if they said no, we asked why not (too busy, too difficult to get there, not feeling well, etc.). Participants who completed the telephone follow-up interview received a $5 gift card.

Approximately 4 months after the index ED visit, we sent a list of the referrals to each of our partner agencies and asked them to document whether the referred participants had received any assistance. We were able to obtain this information from all but one of the ten organi- zations. Unity Health, a large federally qualified health center (FQHC), that we referred our patients to when the participant stated they needed help finding a doctor did not provide the information to us. In- stead, we used the DC Medicaid claims to determine whether the re- ferred participants had visited a doctor at the FQHC. Among those we referred to the FQHC, if the participant had at least one claim from

Unity Health for an office visit to a physician after the index ED visit, we counted this as receiving assistance from the agency.

    1. Outcomes

Our primary outcome was determining the proportion of partici- pants with a HRSN who received assistance with their need from the referral agency. As a secondary outcome, we also measured why partic- ipants did or did not get their need met.

    1. Analysis

We conducted a descriptive analysis. First, we compared partici- pants’ demographics, structural and intermediary determinants of health by the number of referrals made to different agencies using a chi-square test of homogeneity. Differences were considered significant if the p-value associated with the test statistic was <= 0.05. Second, we re- port the assistance rates by agency and report mechanism (i.e. agency- reported vs self-reported) and their 95% confidence interval. Finally, we report the number and reasons needs were met or remained unmet by type of need. All analyses were completed with SAS 9.4 (Cary, NC).

  1. Results
    1. Characteristics of study subjects

Of the 1074 patients screened, 833 (78%) were eligible. Of 241 inel- igible patients, the most common reasons for ineligibility were the in- ability to understand consent (31%) or non-English speaking (24%). Among eligible patients, 505 enrolled (60%), 247 refused (30%), 72 felt too sick/uncomfortable (9%) and 9 (1%) were discharged before the RA could complete enrollment. The majority of enrolled participants were black (98%), female (69%), and had earned a high school diploma (82%). Almost half of participants (46%) were between the ages of 18-34 and 57% reported working full- or part-time.

Table 1 displays the demographic and SDH characteristics of our study sample overall and by the number of referrals each participant re- ceived. The study sample had a high prevalence of food insecurity (54%) and housing instability (39%). A relatively small percentage of patients described their neighborhood as safe (22%). Only 15% of the sample did not ask for assistance with at least one HRSN. The prevalence of many social risk factors, particularly the intermediary SDH, were

Table 1

Percent distribution of social determinants of health by number of health-related social need referrals

Characteristic Overall Number of referrals to different organizations

0

1

2

3

TOTAL

505

75

89

117

224

A. Demographics

Age?

18-34

46%

56%

47%

52%

39%

35-44

15%

16%

13%

15%

16%

44-54

21%

19%

13%

14%

29%

55 and over

18%

9%

26%

20%

16%

Sex

Female

69%

72%

75%

76%

61%

Male

31%

28%

25%

24%

39%

B. Structural determinants of health

Education?

<High school degree

18%

9%

17%

22%

18%

High school degree/GED

40%

43%

35%

32%

46%

Some college/vocational

31%

33%

36%

32%

29%

Bachelor’s degree or higher

Major Activity?

11%

15%

12%

15%

7%

Working full-time +- school

38%

51%

50%

39%

28%

Working part-time +- school

19%

17%

19%

15%

22%

Looking for work

17%

13%

10%

16%

21%

School

6%

7%

4%

7%

5%

Homemaker/retired

6%

4%

5%

6%

6%

Disabled

14%

8%

10%

17%

17%

C. Intermediary determinants of health

Type of Medicaid insurance?

Fee for service (Prime)

28%

17%

21%

30%

32%

Managed care

72%

83%

79%

70%

68%

Usual healthcare provider?

73%

79%

78%

74%

68%

Food insecure? Stable housing?

54%

25%

29%

59%

71%

Yes

61%

88%

75%

62%

45%

Unstable presently

20%

8%

12%

15%

30%

Worried about in the future

19%

4%

12%

22%

25%

Trouble Paying Bills in Past 12 Months? Neighborhood safety?

50%

16%

38%

55%

63%

Safe

22%

33%

24%

19%

19%

Somewhat safe

46%

51%

53%

46%

41%

Unsafe/very unsafe

32%

16%

24%

34%

38%

Drinks alcohol daily

7%

5%

3%

6%

9%

Uses drugs other than those prescribed for medical reasons?

Currently smokes?

15%

36%

8%

25%

9%

31%

15%

32%

20%

44%

Marital status

Married/living with partner

11%

13%

15%

9%

10%

follow-up interview stated they received assistance from the housing agency. According to the agency data, participants were most likely to receive assistance for medical-related needs (16.5%), behavioral health (4.6%) and food (4.7%).

Among the 326 participants who completed the follow-up interview, we made 728 referrals to address unmet HRSNs. The partic- ipants reported that 148 needs were met (20%) by the time of the 1-month follow-up interview, most commonly by the referral agency (56%) (see Table 2). Participants were most likely to report that the referral agency met their need if it was a medical (15%) or food (13%) need. Among the 580 referrals made where the needs remained unmet, only 12% of those referred attempted to contact the referral agency. Most often, the participant told us that they planned to con- tact the agency (47%) or that they were too busy or had forgotten (20%).

  1. Discussion

The adult Medicaid beneficiaries who participated in our SDH screening and HRSN referral program reported a high prevalence of so- cial adversities such as housing instability, food insecurity and trouble paying bills. The vast majority of participants requested assistance with at least one HRSN. The most common unmet needs were related to housing, food insecurity, medical and job training. Our HRSN referral program was not associated with a substantial proportion of Medicaid beneficiaries getting their HRSNs met.

It was not surprising to find a high prevalence of HRSNs among adult

Medicaid beneficiaries who sought treatment in the ED setting. How- ever, we had hoped to successfully address many of their needs by pro- viding them with information and a referral. Our passive intervention helped relatively few patients, regardless of whether assistance was re- ported by the agency (i.e. 1% to 17%) or patient-reported (i.e. 5% to 15%). The higher assistance rates reported by the patient as opposed to the agency is likely due to social desirability bias. This finding is worth not- ing since many studies have relied on patient-reported assistance mea- sures when evaluating the effectiveness of HRSN interventions.

There are many possible explanations for the low assistance rates from our HRSN referral program. First, studies that have reported higher success rates addressing HRSNs with adults (i.e. > 25%) have relied on a more active intervention (i.e. patient navigation) with multiple contacts with each patient to address their unmet needs [10-12,14,15]. Account- able Health Communities (AHCs), funded by the Centers for Medicare and Medicaid Innovations Demonstration Project, is presently evaluat-

* p <= 0.05.

a According to the 2-item Patient Health Questionnaire.

associated with an increasing number of referrals to community-based agencies.

3.2. Main results

Fig. 2 shows the percentage of participants who received help from a referral agency for one of their top three identified needs based on agency information and self-report. Of the 430 participants referred to at least one agency, we successfully completed a telephone follow-up interview with 326 (76%). The housing agency received the most refer- rals (N = 344). The housing agency reported assisting 1.5% of referred participants whereas 10.7% of those who completed the telephone

Even though they requested help, they may not have wanted or may not have had the time and energy needed to follow-up on the referral. Many of our participants reported that they had not contacted the refer- ral agency when we contacted them one month after the ED visit but stated that they planned to at a future time.

Third, our low assistance rates may also be related to the consider- able inconvenience and scarcity that people confront when trying to ob- tain community resources. While food insecurity may seem like one of the more easily addressed social needs, the location and limited hours of food pantries make it difficult for people to get there. A commentary by DeMarchis et al. reinforces the idea that patients may be less likely to follow-up when they perceive the resource to be unavailable, insuffi- cient or they had a bad prior experience with the agency [37]. Housing issues were common among our study sample but difficult to address. Our community referral agency is located in an area of the city that is not metro-accessible and is only open during regular business hours. Additionally, the limited funding they have to help people with their

Widowed/separated/divorced

18%

8%

18%

21%

21%

ing different intensities of referral interventions at 29 sites [35,36].

Never married

71%

79%

67%

71%

68%

However, the planned AHC Track 1 (Assistance) which was similar to

Living with children?

66%

44%

54%

46%

36%

our passive referral system, is not included in the evaluation because

Has someone to help in emotional crisis?

Has someone to Take Him/Her to Doctor? Has someone to Lend Him/Her $100?

82%

82%

60%

92%

96%

84%

88%

83%

67%

82%

85%

54%

72%

71%

46%

of a lack of qualified applications.

Second, the majority of the HRSNs that participants requested help

Screened positive for depression?a

53%

20%

34%

65%

65%

with are likely not new problems for them but rather chronic ones.

24 Agency

Participant

16.5

15

13.4

10.7

10.5

10.3

8.5

4.7

4.6

5

1.5

1.4

0

0

22

20

18

16

14

12

10

8

6

4

2

0

Food Pantry

Housing Program

Legal Program

Medical (MCO or

Behavioral Health

Job Training Program

Wellness Center

N=149 N=112

N=344 N=261

N=73 N=57

FQHC)

N=188 N=133

Agency

N=65 N=47

N=109 N=78

N=52 N=40

Fig. 2. Percentage Who Had Need Addressed As Reported By Agency And By Participant. Below each agency we report the number referred to the agency (denominator for agency %) and the number of people referred to the agency who completed the telephone follow-up interview (denominator for participant %).

rent and/or utilities runs out within the First two days of each month. If EDs implement a HRSN referral program, it is recommended that they build a strong alignment with CBOs and ensure that the organizations have the resources to address patients’ HRSNs [38].

Finally, it is also possible that we did not use a valid method of iden- tifying patients who will benefit from a HRSN referral [37]. Simply ask- ing patients whether they want assistance with a HRSN may not be enough. Although we asked people to prioritize their needs, we did not ask more detailed questions about the perceived urgency of these needs, what prior efforts they had made to address their needs, or other contextual factors (i.e. competing demands, cultural competency, attitudes towards assistance, etc) [39]. We do not have enough

experience in the healthcare setting to understand the extent to which lack of an uptake in HRSN assistance is because of a measurement problem or related more fundamentally to the accessibility and quality of the referred services [37].

This program could potentially improve HRSN uptake rates with a more active intervention that includes multiple contact attempts by ED staff until the unmet need is resolved or by co-locating community agency staff or benefits at the ED/hospital. Patients may have time dur- ing the ED visit or immediately afterwards to speak with community agency staff and receive needed services. In addition, the ideal ED refer- ral program would include an electronic health record system that allowed bi-directional information related to patients’ social needs to

Table 2

Reasons why subjects received or did not receive assistance among those who completed follow-up interview

Overall

Food

Housing

Legal services

Medical

Behavioral health

Job training

Wellness

Number of referralsa

728

112

261

57

133

47

78

40

Need Met:

148

34

46

11

35

7

12

3

By referral agency

83

15

28

6

20

4

8

2

By another agency

18

6

7

1

2

2

0

0

Figured it out on my own

33

11

7

3

9

1

1

1

Help from family/friends

6

2

1

1

2

0

0

0

No longer a problem

7

0

2

0

2

0

3

0

Missing

1

0

1

0

0

0

0

0

Need Not Met:

580

78

215

46

98

40

66

37

Contacted agency and appointment scheduled

29

1

14

0

8

4

1

1

Contacted agency, no answer/left message

39

0

22

3

5

2

2

5

Plan to contact agency

274

43

93

25

52

14

34

13

Too busy/forgot to contact agency

114

14

40

8

18

12

16

6

Lost agency contact information

22

1

10

0

5

3

2

1

Too difficult to get to agency

11

7

2

0

0

0

2

0

Too sick

19

3

5

2

3

2

1

3

Unable to help/doesn’t think agency can help

20

3

13

1

1

0

1

1

Thought agency would contact subject

11

0

5

1

2

0

2

1

Involved with another agency

6

0

3

0

2

0

0

1

Doesn’t want help for this

19

3

2

3

2

3

3

3

Missing

16

3

6

3

0

0

2

2

a This table is referral-based, not subject-based because the subject gave a reason for each referral and they could be different.

be shared (assuming patient consent) between healthcare providers and CBO staff.

The results of our study must be considered in the context of the fol- lowing limitations. First, there is no validated SDH or HRSN survey that is widely accepted. Many of the questions in our survey were from past validated surveys and scales. It is possible that some of the differences in prevalence of HRSNs between our study and others is due to variations in measurement. However, given that this is a new field, it is important to test different questions and methods and learn from each other about how best to measure SDH and HRSNs. Second, the generalizability of this study is limited because it was conducted in a single ED in an urban environment and it focused on English-speaking Medicaid bene- ficiaries. The results are likely to be meaningfully different in other set- tings and patient populations. In addition, not only was this a convenience sample but approximately one-third of the Medicaid pa- tients we approached, refused to participate. Third, our telephone follow-up period may have been too short to capture the assistance pro- vided by agencies, however the agency-reported data was not mea- sured until four months after the ED visit and resulted in lower rather than higher assistance rates than might have been expected with a lon- ger follow-up period. Finally, we used Medicaid claims as a proxy for agency-reported follow up to our federally qualified health center, Unity Health Care. This likely overestimated the percentage of patients who had a medical need meet because the reason for participants’ visit to the FQHC during the follow-up period may not have had anything to do with our referral.

In conclusion, this is the first ED-based HRSN referral program to be conducted with adult Medicaid beneficiaries. The SDH screening re- vealed that adult Medicaid beneficiaries who receive medical care in the ED face substantial social adversities and have a high prevalence of HRSNs. Although we partnered with CBOs to address patients’ needs, our passive intervention did not resolve the HRSNs of the majority of our Medicaid sample.

Author contributions

Conceptualization: PK and MLM; Formal analysis: PK, MLM and ES; Methodology: PK and MLM; Project administration: PK, MLM and SJ; Supervision: PK, MLM and SJ; Writing – original draft: PK and MLM; Writing – review and editing: PK, MLM, ES and SJ.

Declaration of Competing Interest

No commercial conflicts of interest.

References

  1. Frieden TR. A framework for public health action: the health impact pyramid. Am J Public Health. 2010;100(4):590-5.
  2. McGinnis JM, Williams-Russo P, Knickman JR. The case for more active policy atten- tion to health promotion. Health Aff (Millwood). 2002;21(2):78-93.
  3. Bradley EH, Elkins BR, Herrin J, Elbel B. Health and social services expenditures: as- sociations with health outcomes. BMJ Qual Saf. 2011;20(10):826-31.
  4. Bradley EH, Canavan M, Rogan E, et al. Variation in health outcomes: the role of spending on social services, public health, and health care, 2000-09. Health Aff (Millwood). 2016;35(5):760-8.
  5. Billioux A, Verlander K, Anthony S, Alley D. Standardized screening for health- related social needs in clinical settings: The accountable health communities screen- ing tool. Washington, DC: National Academy of Medicine; 2017.
  6. Artiga S, Hinton E. Beyond Health Care: The Role of Social Determinants in Promot- ing Health and Health Equity; 2018.
  7. National Academies of Sciences, Engineering, and Medicine. Integrating Social Care into the Delivery of Health Care: Moving Upstream to Improve the Nation’s Health. Washington, DC: The National Academies Press; 2019.
  8. National Academies of Sciences, Engineering, and Medicine. Investing in Interven- tions that Address Non-Medical Health-Related Social Needs: Proceedings of a Workshop. Washington, DC: National Academies Press; 2019.
  9. Buitron de la Vega P, Losi S, Sprague Martinez L, et al. Implementing an EHR-based Screening and Referral System to Address Social Determinants of Health in Primary Care. Med Care. 2019;57(Suppl 6 Suppl 2) S133-s139.
  10. Berkowitz SA, Hulberg AC, Hong C, et al. Addressing basic resource needs to improve primary care quality: a community collaboration programme. BMJ Qual Saf. 2016;25 (3):164-72.
  11. Berkowitz SA, Hulberg AC, Standish S, Reznor G, Atlas SJ. Addressing unmet basic re- source needs as part of chronic Cardiometabolic disease management. JAMA Intern Med. 2017;177(2):244-52.
  12. Schickedanz A, Sharp A, Hu YR, et al. Impact of social needs navigation on utilization among high utilizers in a large integrated health system: a quasi-experimental study. J Gen Intern Med. 2019;34(11):2382-9.
  13. Lindau ST, Makelarski JA, Abramsohn EM, et al. CommunityRx: a real-world con- trolled clinical trial of a scalable, low-intensity community resource referral inter- vention. Am J Public Health. 2019;109(4):600-6.
  14. Gottlieb LM, Hessler D, Long D, et al. Effects of social needs screening and in-person service navigation on child health: a randomized clinical trial. JAMA Pediatr. 2016; 170(11):e162521.
  15. Sege R, Preer G, Morton SJ, et al. Medical-legal strategies to improve infant health care: a randomized trial. Pediatrics. 2015;136(1):97-106.
  16. Garg A, Toy S, Tripodis Y, Silverstein M, Freeman E. Addressing social determinants of health at well child care visits: a cluster RCT. Pediatrics. 2015;135(2):e296-304.
  17. Wallace AS, Luther B, Guo JW, Wang CY, Sisler S, Wong B. Implementing a social de- terminants screening and referral infrastructure during routine emergency depart- ment visits, Utah, 2017-2018. Prev Chronic Dis. 2020;17:E45.
  18. Losonczy LI, Hsieh D, Wang M, et al. The Highland health advocates: a preliminary evaluation of a novel programme addressing the social needs of emergency depart- ment patients. Emerg Med J. 2017;34(9):599-605.
  19. Doran KM, Kunzler NM, Mijanovich T, Lang SW, Rubin A, Testa PA, et al. Homeless- ness and other social determinants of health among emergency department pa- tients. J Soc Distress Homeless. 2016;25(2):71-7.
  20. Malecha PW, Williams JH, Kunzler NM, Goldfrank LR, Alter HJ, Doran KM. Material needs of emergency department patients: a systematic review. Acad Emerg Med. 2018;25(3):330-59.
  21. Samuels-Kalow ME, Boggs KM, Cash RE, et al. Screening for health-related social needs of emergency department patients. Ann Emerg Med. 2021;77(1):62-8.
  22. Health Leads. https://healthleadsusa.org/. Accessed 12/20/2020, 2020.
  23. Garg A, Sarkar S, Marino M, Onie R, Solomon BS. Linking urban families to commu- nity resources in the context of pediatric primary care. Patient Educ Couns. 2010;79 (2):251-4.
  24. Solar OI. A conceptual framework of action on the social determinants of health. Ge- neva: World Health Organization; 2010.
  25. American Community Survey (ACS). http://www2.census.gov/programs-surveys/ acs/methodology/questionnaires/2016/quest16.pdf; 2016.
  26. Hager ER, Quigg AM, Black MM, et al. Development and validity of a 2-item screen to identify families at risk for food insecurity. Pediatrics. 2010;126(1):e26-32.
  27. Montgomery AE, Fargo JD, Byrne TH, Kane VR, Culhane DP. Universal screening for homelessness and risk for homelessness in the Veterans Health Administration. Am J Public Health. 2013;103(Suppl. 2):S210-1.
  28. Desmond M, Shollenberger T. Forced displacement from rental housing: prevalence and neighborhood consequences. Demography. 2015;52(5):1751-72.
  29. Ouellette T, Burstein N, Long D, Beecroft E. Measures of Material Hardship: Final Re- port. Washington, D.C; 2004.
  30. Statistics NCfH. National Health Interview Survey: NHIS Data, Questionnaires and Related Documentation. https://www.cdc.gov/nchs/nhis/data-questionnaires- documentation.htm; 2017. Accessed 8/8/2016, 2016.
  31. Skinner HA. The drug abuse screening test. Addict Behav. 1982;7(4):363-71.
  32. Group GATSC. Tobacco questions for surveys: A subset of questions from the global adult tobacco survey (GATS). Atlanta: Georgia; 2011.
  33. Kroenke K, Spitzer RL, Williams JB. The patient health Questionnaire-2: validity of a two-item depression screener. Med Care. 2003;41(11):1284-92.
  34. Find help. https://www.findhelp.org/?ref=ab_redirect. Accessed 8/1/2016, 2016.
  35. Alley DE, Asomugha CN, Conway PH, Sanghavi DM. Accountable health communities-addressing social needs through Medicare and Medicaid. N Engl J Med. 2016;374(1):8-11.
  36. Gottlieb L, Colvin JD, Fleegler E, Hessler D, Garg A, Adler N. Evaluating the account- able health communities demonstration project. J Gen Intern Med. 2017;32(3): 345-9.
  37. De Marchis EH, Alderwick H, Gottlieb LM. Do patients want Help addressing social risks? J Am Board Fam Med. 2020;33(2):170-5.
  38. Garg A, Boynton-Jarrett R, Dworkin PH. Avoiding the unintended consequences of screening for social determinants of health. Jama. 2016;316(8):813-4.
  39. Byhoff E, De Marchis EH, Hessler D, et al. Part II: a qualitative study of social risk screening acceptability in patients and caregivers. Am J Prev Med. 2019;57(6 Suppl 1):S38-46.