Transforming Nursing

© Meharry Medical College Journal of Health Care for the Poor and Underserved 31 (2020): 382–397.

Screening for Social Determinants of Health in Free and Charitable Clinics in North Carolina

Deepak Palakshappa, MD MSHP Mark Scheerer

Charles Th omas Alexander Semelka, MD Kristie L. Foley, PhD

Abstract: Free and charitable clinics care for patients at risk of unmet social needs, but limited research is available describing what these clinics do to address social determinants of health (SDH). We conducted a survey of free and charitable clinics in North Carolina to determine the proportion that screen for SDH. Clinics that were members of the North Carolina Association of Free and Charitable Clinics were eligible (n=67). Of the 55 clinics that completed the survey, 34 (61.8%) reported always screening for at least one SDH. Th e majority that were screening provided information about community resources. Twenty- seven (49.1%) reported that they followed-up with patients at the next visit to determine if they received the resources. Th e primary barrier to screening was a lack of personnel. Future research should focus on how to implement SDH interventions in clinics with limited resources serving patients likely to have a high need for social services.

Key words: Social determinants of health, free clinics, community health

The social determinants of health (SDH), or the areas in which people are born, grow, work, and age, have a profound impact on health.1– 3 Negative SDH include

such circumstances as unstable housing, limited access to healthy food, and lack of transportation. Numerous studies have found that negative SDH can worsen morbidity and mortality.4– 11 Historically, addressing unmet social needs have primarily been the focus of public health or policy communities. Increasingly though, national health care organizations and publications, such as the National Academy of Medicine, Healthy People 2020, and the American Academy of Pediatrics, have recommended that health systems address negative SDH as a routine part of clinical care to improve population health and reduce disparities.11– 17

With these national recommendations and the growing recognition of the negative eff ects of unmet social needs, several health care systems have begun to test strategies addressing negative SDH as part of health care delivery models.18– 25 For example, the

ORIGINAL PAPER

DEEPAK PALAKSHAPPA is an Assistant Professor at the Wake Forest School of Medicine. MARK SCHEERER is Deputy Director of the North Carolina Association of Free & Charitable Clinics. CHARLES SEMELKA is a physician at the Wake Forest School of Medicine. KRISTIE FOLEY is a Professor at the Wake Forest School of Medicine, Winston- Salem, NC. Please address all correspondence to: Deepak Palakshappa, Department of Internal Medicine, Wake Forest School of Medicine, Winston- Salem, NC 27157, Phone: 336-716-1795; Fax: 336-716-7359; E-mail: dpalaksh@wakehealth .edu.

383Palakshappa, Scheerer, Semelka, and Foley

Centers for Medicare and Medicaid (CMS) have recently started testing their Account- able Health Communities (AHC) Model. Th e AHC models are focused on improving patients’ health by addressing unmet health- related social needs through better con- nections between clinical care and community resources.13,26,27

Although there is a growing evidence base for strategies to address negative SDH in clinical settings, one important component of the U.S. health care safety- net that has received less attention are free and charitable clinics. Free and charitable clinics are volunteer- based 501(c) tax- exempt safety- net health care organizations that provide med- ical, dental, and pharmaceutical care for uninsured and underinsured individuals. Free and charitable clinics restrict eligibility to individuals who are uninsured, underinsured, or have limited access to primary care, specialty care, or prescription medications. Th e clinics receive little or no state or federal funding, and almost half of free and charitable clinics in the U.S. have operating budgets of less than $100,000 (consult https:// www .nafcclinics .org). Th ere are currently 1,400 free and charitable clinics in the U.S. that serve over two million individuals annually.28– 30 Despite the substantial progress made by means of the Aff ordable Care Act, over 20 million Americans remain uninsured, and free and charitable clinics play a critical role in caring for these individuals.30,31 Many of these individuals are at risk of having unmet health- related social needs.32 Yet, little is known about how free and charitable clinics identify and address the SDH. Further, an increasing number of health care institutions and clinics are implement- ing strategies to address patients’ unmet social needs, but there is a paucity of data on what these institutions and clinics are doing. As free and charitable clinics see patients at highest risk of having negative SDH, it is especially important to understand what these clinics are currently doing to screen and address patients’ unmet social needs and what barriers may exist to implementing systematic screening across the clinics.

To fi ll this gap in the literature, we worked with the North Carolina Association of Free and Charitable Clinics (NCAFCC, http:// ncafcc .org/ ). Th e NCAFCC is a non- profi t organization consisting of 67 free and charitable clinics located throughout North Carolina. It is one of the largest statewide associations of free and charitable clinics in the country, and, as of yet, North Carolina has not expanded Medicaid as part of the Aff ordable Care Act. Th is makes North Carolina an ideal setting to study SDH screening in free and charitable clinics. Th e primary objective of this study was to describe the current SDH screening and referral practices among free and charitable clinics in North Carolina. We specifi cally evaluated how many clinics were screening for SDH, which SDH clinics were screening for and how oft en, clinic staff ’s views of the importance of screening for and addressing patients’ unmet social needs, and any barriers to screening. A secondary objective was to evaluate if there were specifi c clinic characteristics associated with an increased likelihood of clinics screening for SDH.

Methods

Study design and population. We worked with the NCAFCC to conduct a cross- sectional study of all its members. Th e 67 free and charitable clinics that are members of NCAFCC provide care to over 80,000 uninsured and underinsured patients a year, and the clinics are located in urban and rural areas throughout the state. Th e clinics

384 Social determinants in free and charitable clinics

are members of the NCAFCC, but operate independently in deciding what services they provide.28 All free and charitable clinics that were members of the NCAFCC were eligible for this study. We sent a web- based survey to the directors of the 67 clinics between March– May 2018 to determine each clinic’s SDH screening practices. Th e clinics were sent up to fi ve email reminders during this period encouraging them to complete the survey, and all clinics that completed the survey received a $25 gift card. We combined the results of the SDH survey with data from the NCAFCC’s annual outcomes survey to further characterize clinics that were screening for SDH compared with those that were not.

Measures. Social determinants of health survey. Th rough a detailed review of lit- erature and in partnership with NCAFCC staff members, we developed a web- based survey to assess free and charitable clinics current SDH screening practices using the Research Electronic Data Capture (REDCap). Prior interventions addressing SDH in clinical care settings have primarily focused on screening patients for a particular SDH, referring patients with unmet social needs to resources in the community (either by providing information or directly connecting them to a community organization), and following-up with patients to see if they received the resources.18,22,23,25 Th e questions included in our survey were designed to assess if free and charitable clinics in North Carolina were following a similar care model. Our survey assessed what negative SDH clinics were screening for and how oft en, how clinics addressed patients’ unmet social needs, and if there was a process in place to follow-up with patients to see if they received the resources. We included the option in the survey for free text responses to off er clinics the opportunity to use diff erent methods to address the SDH domains included in our survey. We also assessed the clinics’ attitudes toward addressing SDH and what, if any, barriers may prevent the clinic from implementing systematic SDH screening. Th e SDH domains included in the survey related to the AHC Health- Related Social Needs (HRSN) Screening Tool. Th e AHC HRSN tool was developed by the Centers for Medicare and Medicaid Services (CMS) Center in conjunction with national experts. Th e tool is currently being tested as part of CMS’s AHC model.13,33 Th ere are several diff erent validated SDH screening tools available.18,21,34,35 We chose to base our SDH survey questions on the AHC tool because it included fi ve domains (housing instability, food insecurity, transportation, utilities, and interpersonal safety) that have been shown in prior research to be associated with worse health outcomes or increased health care utilization.13

Th e survey included nine questions about how oft en a staff member or volunteer asked about the screening questions included in the AHC HRSN tool (Table 1). Responses included always, sometimes, or never. Using branching logic, clinics that responded always or sometimes were asked, “What do you do when patients report these issues?” Responses included that the clinic either directly connected patients to resources in the community, provided information about community resources, or listed a free text response. We also included items in the survey to assess to what extent clinics felt they should address each of the fi ve SDH domains. Th is was assessed on a 5-point Likert scale from strongly agree to strongly disagree. We also asked clinics what the primary barriers were to addressing SDH in the clinic, and how the clinic follows up to see if patients who were referred to community resources received those resources.

385Palakshappa, Scheerer, Semelka, and Foley

Table 1. SOCIAL DETERMINANTS OF HEALTH SURVEY QUESTION ITEMS

Question Responses

1. How oft en does a staff member/volunteer as patients about their housing situation (for example if the patient has a place to live)?

Always, sometimes, never

2. How oft en does a staff member/volunteer as patients if they are having problems with their housing (for example issues with inadequate heat, smoke detectors not working, or mold in the house)?

Always, sometimes, never

3. How oft en does a staff member/volunteer ask patients they can aff ord to buy food or groceries?

Always, sometimes, never

4. How oft en does a staff member/volunteer ask patients if lack of transportation kept them from medical appointments, meetings, work, or from getting things needed for daily living?

Always, sometimes, never

5. How oft en does a staff member/volunteer ask patients if the electric, gas, oil, or water company threatened to shut off services in their home?

Always, sometimes, never

6. How oft en does a staff member/volunteer ask patients if anyone, including family, physically hurt them?

Always, sometimes, never

7. How oft en does a staff member/volunteer ask if anyone, including family, insult, or talk down to them?

Always, sometimes, never

8. How oft en does a staff member/volunteer ask patients in anyone, including family, threatened them with harm?

Always, sometimes, never

9. How oft en does a staff member/volunteer ask patients if anyone, including family, scream, or curse at them?

Always, sometimes, never

To what extent do you agree or disagree with the following statement: 1. It is essential that our clinic personnel directly connect patients with resources in the community for the following services: Housing, food, transportation, payment of utilities, interpersonal safety

5- point Likert scale (strongly agree– strongly disagree)

2. It is essential for our patients’ health that our clinic personnel provide information about community resources: Housing, food, transportation, payment of utilities, interpersonal safety

5- point Likert scale (strongly agree– strongly disagree)

There are many reasons why your clinic may not be able to directly connect your patients with various social services. Which of the following apply to your clinic: Which is the #1 reason why your clinic may not directly connect your patients with social services? If you do directly connect patients to community resources or provide information about community resources to patients, what do you do to follow up to see if patients accessed resources?

386 Social determinants in free and charitable clinics

Annual outcomes survey. We combined data from the SDH survey with the NCAFCC 2017 annual outcomes survey. Th e NCAFCC surveys all member clinics about the ser- vices they provided over the prior year. For the 2017 annual outcomes survey, clinics were surveyed in February 2018 to assess the clinic characteristics from 1/ 1/ 2017– 12/ 31/ 2017. Th e survey included questions such as the size of the clinic, the number and demographic characteristics of the patients served, and the types of services provided (medical, dental, pharmacy).36

For this study, we combined data from the annual outcomes to the SDH survey to determine if there were clinic characteristics (including characteristics of the clinics and the patients cared for at the clinics) associated with screening for SDH (Table 2). Th e characteristics included the total number of patients seen in the prior year, the race/ ethnicity of the patients seen (White, Black, Hispanic, other), the age of the patients, the proportion of patients seen who were female, and the proportion of patients seen who preferred a language other than English. We also included the type of services clinics provided.

Statistical analysis. We used descriptive statistics, such as means and frequencies, to characterize the study population and how the clinics currently screen and address the SDH domains. We performed bivariate analyses to test the association between SDH screening (if a clinic always screened for at least one SDH) and clinic characteristics using Fisher’s exact test for categorical variables and non- parametric Wilcoxon rank- sum test for continuous variables. We tested the association between SDH screening and total population, race, proportion of patients who were female, proportion who preferred a language other than English, and if the clinic provided enabling services (e.g., transportation, interpreter services) using logistic regression. We used a two- sided hypothesis test and considered an α < .05 signifi cant. All analyses were conducted using Stata 15.1 (StataCorp, College Station, TX). Th is study of the care provided at free and charitable clinics was not considered by the Wake Forest School of Medicine Institutional Review Board to constitute human subjects research.

Table 2. QUESTIONS FROM NORTH CAROLINA ASSOCIATION OF FREE AND CHARITABLE CLINICS ANNUAL OUTCOMES SURVEY INCLUDED IN STUDYA

1. Number of unduplicated patients served by your organization: 2. Number of patients within each race/ethnic group: 3. Number of female patients: 4. Number of patients within each age category (<18; 18–64; 65+) 5. Number of unduplicated patients best served in a language other than English: 6. Check all applicable service(s) that the clinic currently provides at the facility site (medical; dental; pharmaceutical; behavioral health; enabling services

Notes: aAll questions are specifi c to January 1, 2017- December 31, 2017.

387Palakshappa, Scheerer, Semelka, and Foley

Results

Clinic characteristics. Of the 67 clinics that were eligible, 55 (82%) completed the sur- vey. Th e 55 clinics provided care for 76,038 patients in 2017 (mean number of patients per clinic 1,435 patients; Table 3). Th e majority of patients seen were of White race, female, and between 18 and 65 years of age. Ninety percent of the clinics that responded off ered medical services. Clinics that responded to the SDH survey did not signifi cantly diff er from non respondents in the number or the demographic characteristics of the patients seen. Clinics that did respond were signifi cantly more likely to off er medical services than non- respondents (90.7% vs 40.0%, p=.02).

Screening for SDH. Of the 55 clinics, 34 clinics (61.8%) reported always screening for at least one of the SDH domains (housing, food insecurity, transportation, utilities, or interpersonal violence) assessed. One clinic reported always screening for all fi ve of the SDH assessed. Th e most common negative SDH clinics reported always screening for were transportation, if a patient had housing, and food insecurity (Table 4). Clin- ics were least likely to screen patients for problems with their housing, utilities, and

Table 3. CHARACTERISTICS OF FREE AND CHARITABLE CLINICS

Clinics (N=53)a

N (%)

Number of patients seen in prior year (mean; range)

76,038 (mean: 1,435; range: 144–18,298)

Race White 34,848 (45.9%) Black/African American 19,971 (26.3%) Hispanic 17,631 (23.2%) Other 3,827 (5.0%)

Number of female patients last year 45,341 (59.6%) Age

<18 years of age 959 (1.3%) ≥18 and <65 years of age 72,167 (94.5%) ≥65 years of age 3,210 (4.2%)

Language other than English 14,476 (19.0%) Type of services provided

Medical 48 (90.6%) Dental 17 (32.1) Pharmacy 44 (83.0%) Behavioral Health (mental health, substance use) 28 (52.8%) Enabling services (e.g. transportation, interpreter) 44 (83.0%)

Notes: a2 clinics who completed social determinants of health survey did not complete annual outcomes survey.

Ta bl

e 4.

N U

M BE

R O

F FR

EE A

N D

C H

A RI

TA BL

E C

LI N

IC S

SC RE

EN IN

G A

N D

A D

D RE

SS IN

G S

O C

IA L

D ET

ER M

IN A

N TS

O F

H EA

LT H

A

 F re

qu en

cy w

hi ch

it em

s a re

sc re

en ed

H ow

sc re

en in

g ite

m s a

re a

dd re

ss ed

D ir

ec tly

co

nn ec

te d

to re

so ur

ce s

N (%

)

Pr ov

id ed

w ith

in

fo rm

at io

n N

(% )

O

th er

N

(% )

A lw

ay s

N (%

)

So m

et im

es N

(% )

N

ev er

N (%

)

D oe s p

at ie nt

h av e ho

us in g

21 (3

8. 2)

28 (5

0. 9)

6 (1 0.

9) 13

(2 6. 5)

30 (6

1. 22 )

6 (1 2. 2)

Pr ob

lem s w

ith h ou

sin g (e .g . m

ol d in

th e ho

m e)

3 (5 .5 )

37 (6

7. 3)

15 (2

7. 3)

7 (1 7. 5)

25 (6

2. 5)

8 (2 0. 0)

Fo od

In se cu rit y

15 (2

7. 3)

32 (5

8. 2)

8 (1 4. 6)

17 (3

6. 2)

24 (5

1. 1)

6 (1 2. 8)

La ck

o f t ra ns po

rt at io n

23 (4

1. 8)

28 (5

0. 9)

4 (7 .3 )

12 (2

3. 5)

30 (5

8. 8)

9 (1 7. 7)

C on

ce rn s a

bo ut

u til iti es

(e .g . e le ct ric

, g as )

4 (7 .3 )

26 (4

7. 3)

25 (4

5. 5)

4 (1

3. 3)

21 (7

0. 0)

5 (1 6. 7)

H as

p at ie nt

b ee n ph

ys ic al ly

h ur t

13 (2

3. 6)

33 (6

0. 0)

9 (1 6. 4)

24 (5

2. 2)

15 (3

2. 6)

7 (1 5. 2)

In su lte

d 4 (7 .3 )

22 (4

0. 0)

29 (5

2. 7)

8 (3 0. 8)

14 (5

3. 9)

4 (1 5. 4)

Th re at en ed

w ith

h ar m

7 (1 2. 7)

35 (6

3. 6)

13 (2

3. 6)

20 (4

7. 6)

16 (3

8. 1)

6 (1 4.

3) Sc re am

ed o r c

ur se

d at

4 (7 .3 )

18 (3

2. 7)

33 (6

0. 0)

8 (3 6. 4)

10 (4

5. 5)

4 (1 8. 2)

No te

s: a T he

fi rs t p

ar t o f t he

ta bl e s ho

w s t he

fr eq ue nc y w

ith w hi ch

fr ee

an d ch ar ita bl e c

lin ic s s cr ee n fo r s oc ia l d et er m in an ts of

h ea lth

(h ou

sin g,

fo od

in se cu rit y, tra

ns po

rt at io n,

ut ili tie

s, an

d in te rp er so na l v io le nc e) . T

he se

co nd

p ar t o

f t he

ta bl e s

ho w s h

ow cl in ic s, w ho

sc re en

a lw ay s o

r s om

et im

es , a

dd re ss

e ac h so ci al

d et er m in an t o

f h ea lth

.

389Palakshappa, Scheerer, Semelka, and Foley

if patients had been insulted or screamed at by someone (outside the clinic visit). We did not fi nd signifi cant diff erences between clinics that screened for SDH (if a clinic always screened for at least one SDH) and any of the clinic characteristics tested in bivariate or multivariable analyses.

Addressing SDH. Over 90% of clinics either agreed or strongly agreed that the clinic should provide information about resources to patients who may have trouble with housing, food insecurity, transportation, utilities, or interpersonal violence. Over 80% either agreed or strongly agreed the clinic should directly connect or refer patients with these unmet social needs to resources in the community. Th e primary barriers clinics reported to addressing the SDH were a lack of specifi c personnel (40.7%), such as a social worker, or enough personnel (33.3%) to address patients’ needs.

Of the clinics who were screening, the majority addressed the patients’ unmet social needs by providing information about resources in the community (Table 2). Th ree of the clinics reported having a dedicated person at the clinic, such as a social worker or patient advocate, who was able to address patients’ unmet social needs in the clinic.

For all clinics that either directly referred or provided information about local resources, 27 (49.1%) reported that they followed up with patients at the next visit to determine if she or he received the resources. Ten (18.2%) reported that the clinic would call the patient, and six (10.9%) reported they directly contacted the resource organization to see if the patient received the resource. Eleven (20%) reported that they did not have any specifi c method to follow up with patients to determine if patients received the resources.

Discussion

Increasingly, national organizations are recommending that health care systems identify and address SDH as a routine part of clinical care.11– 16 Despite their importance to the U.S. health care safety- net, though, there are limited data evaluating SDH screening in free and charitable clinics. Th is is the fi rst study to evaluate SDH screening practices in free and charitable clinics, and we found that over 60% of free and charitable clinics in North Carolina reported systematically screening for at least one SDH. Th e majority of clinics agreed or strongly agreed that the clinic should address patients’ health- related social needs. A lack of personnel to assist with addressing patients’ unmet social needs, though, was the primary barrier reported by clinics to further integrating SDH screening.

Th e results of our study have important clinical and public health implications. First, there is a growing evidence base evaluating interventions designed to screen and address SDH in clinical settings,18,21,23 but there are limited data evaluating how many practices in the U.S. are currently screening or which SDH are being screened for by clinics. In this study, we found that over 60% of clinics reported always screening for at least one SDH, and the most common unmet social needs for which clinics reported screening were transportation, housing, and food insecurity. Similar to our results, one study found that 59% of health centers in Michigan were routinely screening for SDH. Th e primary SDH domains that health centers in Michigan screened for were demo- graphic characteristics, employment status, family and living arrangements, and mental health.37

390 Social determinants in free and charitable clinics

Research in this area is relatively new, but one reason for the lack of data about how many health systems have incorporated SDH screening into clinical care may be due to the lack of consensus on which screening tool clinics should use and what SDH clin- ics should identify.38 Several screening tools have recently been developed to identify patients’ unmet social needs in clinical settings.34 Th e Center for Medicare & Medicaid Services (CMS), in conjunction with national experts, developed the HRSN screener to evaluate SDH screening across all of the 31 organizations that are participating in the AHC model.13,33 Th is could be an important step in standardizing SDH screening across practices. Additionally, many state Medicaid programs are considering SDH screening tools to implement.24 Th e North Carolina Department of Health and Human Services has recently developed a SDH screening tool that includes similar domains as the AHC HRSN screener and will be incorporated in North Carolina’s Medicaid Managed Care program.39 An important step to standardizing screening across clinical care settings is identifying what health care systems and clinics are currently doing to screen and address patients’ unmet social needs. Our study provides new data on what free and charitable clinics in North Carolina are doing to screen and address SDH. We fi nd it encouraging that over 60% of clinics surveyed in this study reported screening for at least one SDH. Continued research to understand what and how clinicians and clinics are currently screening for SDH could help identify best practices and how to standard- ize SDH screening across health systems.37 Future research is also need to understand why a clinic may, or may not, screen for a particular unmet social need or if there are specifi c determinants that lead clinics to implement SDH screening.40

A second important fi nding from this study was that 90% of free and charitable clinics surveyed agreed or strongly agreed that the clinic should provide resources to address patients’ unmet social needs. For many individuals who lack health insurance or are underinsured, free and charitable clinics play an important role in providing medical, dental, and pharmacy care.30 Individuals lacking health insurance are also at increased risk of negative SDH, such as lack of housing and food insecurity, and the detrimental eff ects that these negative SDH can have on their health. If the U.S. health care system is going to play an increasing role in addressing negative SDH, free and charitable clinics will play a critical part in identifying and addressing the needs of patients at highest risk of having unmet health- related social needs. Although this study was limited to free and charitable clinics in North Carolina, we found that the majority of clinics recognized the eff ect of these social issues on their patients and were interested in addressing their patients’ needs as 90% believed the clinic should provide resources to patients to address SDH. As we did not specifi cally evaluate why a clinic may not be interested in addressing patients’ social needs, it is unclear why 10% of clinics did not either strongly agree or agree that the clinic should provide resources. In free text responses, several of these clinics reported limited community resources to address patients’ social needs or the clinic primarily seeing patients during off - hours, when many social service agencies may be closed. Future research is needed to determine if there are systemic community or clinic logistical barriers that aff ect clinics’ interest and ability to address SDH.

A third fi nding from our study was that the major barrier that prevented free and charitable clinics from systematically screening and addressing SDH was a lack of per-

391Palakshappa, Scheerer, Semelka, and Foley

sonnel. Our results are similar to those of prior studies that have found that clinicians recognize the impact that unmet social needs have on patients’ health, but clinicians feel they lack the time or knowledge to address SDH in clinical settings.25,41,42 Lack of personnel may be a particularly important barrier for free and charitable clinics as these clinics oft en rely heavily on volunteers to provide care. Th is could explain the variability in if and what SDH clinics were able to address. Th e one clinic that reported always screening for all of the SDH domains assessed reported that the clinic had a dedicated patient advocate who could assist patients’ with their unmet social needs. Free and charitable clinics, as well as other safety- net clinics, are likely to see patients at highest risk for unmet social needs, but they oft en have limited operating budgets and staff . Clinic and policy- level interventions will need to account for this lack of personnel. One potential solution to addressing the lack of personnel in clinics is health information technology. Health information technology tools have shown promise in identifying and addressing SDH in pediatric practices and could help in overcoming barriers to screening and addressing SDH in free and charitable clinics.19,25,43 Health information technology tools could also provide a means for clinics to follow up to assess if patients received resources, as 20% of clinics in this study reported that they did not have a method to follow up to see if patient received resources.44

Fourth, this study provides a baseline understanding of SDH screening practices in free and charitable clinics in a state that is poised to institute several initiatives to address SDH. Similar to its counterparts in other states, the North Carolina Depart- ment of Health and Human Services is planning on implementing a number of initia- tives to address SDH (https:// www .ncdhhs .gov/ about/ department-initiatives/ healthy - opportunities). Th ese initiatives include developing a standard set of SDH screen- ing questions, building a statewide coordinated care network to electronically refer patients to community resources, incorporating SDH strategies in Medicaid Managed Care plans, and developing a community health worker initiative. Even though we did not fi nd signifi cant diff erences in clinic characteristics and SDH screening, there was variability among the clinics in how they reported addressing patients’ unmet social needs. For example, 52% of clinics reported directly connecting patients who had been physically abused to community resources, while only 13% of clinics reported directly connecting patients with concerns about utilities (such as gas, water, and electricity) to resources. Th is could be due to perceived diff erences in the clinic’s role in address- ing the diff erent SDH (interpersonal violence compared with utilities). It could also be due to what resources in the community are available and the ability of clinics to refer patients directly to these resources. Although the results of our study may not be generalizable to other states, they provide important information on how clinics are addressing SDH prior to many of these statewide changes. As several states have plans or have recently begun to implement initiatives to address patients’ unmet social needs in health care settings, our results provide important information about the variability of SDH screening practices across free and charitable clinics and the barri- ers to implementing screening in these clinics. A future next step would be evaluating how these statewide initiatives aff ect how free and charitable clinics screen and ad- dress SDH.

Th ere are several limitations to this study that should be acknowledged. First, the

392 Social determinants in free and charitable clinics

survey was limited to free and charitable clinics that were members of the NCAFCC, which may limit generalizability. North Carolina is in the top fi ve of states with the largest number of free and charitable clinics and NCAFCC is one of the largest free and charitable clinic associations in the country. Additionally, the study provides an important starting point for understanding the variability in SDH screening and referral practices between free and charitable clinics in a large, diverse state that has relevance to other states with a large proportion of free and charitable clinics serving diverse low-income populations. Second, the results of our study were based on self- report. We did not directly assess how oft en clinics were screening for SDH or how clinics were addressing patients’ unmet social needs. It is possible the clinic directors interpreted the questions diff erently from how the researchers intended. Th ird, this study focused on the frequency with which clinics were screening for one of the SDH domains assessed, and not how clinics were screening for them. Further research would be needed to understand what tools or questions clinics are using to identify SDH and how (whether through use of in-person or written questionnaires, for example) clinics are screening. Fourth, our results are limited to the SDH that were assessed as part of the survey. We did not evaluate if clinics were screening for other SDH that have been shown to aff ect health outcomes, such as employment, education level, or social support. We focused on the fi ve domains we chose as they are included in CMS AHC HRSN tool and the North Carolina Department of Health and Human Services SDH screening tool, and these domains could become standard clinical practice in the state or nationally.

Conclusion. National organizations are increasingly recommending that clinicians and health care systems identify and address SDH in order to improve population health and reduce health disparities. Few data exist, though, describing what health care organizations are currently doing to assess patients’ unmet social needs. Th is is particularly true for free and charitable clinics that play a critical role in providing care and serving as a medical home for many uninsured patients. We found that over 60% of clinics were systematically screening for housing issues, food insecurity, utility needs, lack of transportation, or interpersonal violence. Although the majority of clinics felt the clinic should address patients’ health- related social needs, a lack of personnel was the primary barrier reported by clinics to further integrating SDH screening. Further research is needed to understand how to implement SDH programs in diverse settings, particularly at facilities with limited resources or personnel, and if addressing SDH in a clinical setting ultimately improves patient and population health.

Acknowledgments

Th is study was supported in part by Northwest Area Health Education Center of Wake Forest School of Medicine to help develop best practices to address social determi- nants of health. We would like to acknowledge the REDCap services of Wake Forest Clinical and Translational Science Institute (WF CTSI), which is supported by the National Center for Advancing Translational Sciences (NCATS), National Institutes of Health, through grant award number UL1TR001420. No funding organization or sponsor had a role in the study design; collection, analysis, and interpretation of the

393Palakshappa, Scheerer, Semelka, and Foley

data; writing the report; and the decision to submit the report for publication. Por- tions of this study were presented at the 2019 North Carolina American College of Physicians in Raleigh, NC.

References 1. Solar O, Irwin A. A conceptual framework for action on the social determinants of

health. Social determinants of health discussion paper 2 (policy and practice). Geneva, World Health Organization, 2010. Available at: http:// www .who.int/ sdhconference / resources/ ConceptualframeworkforactiononSDH_eng .pdf.

2. Havranek EP, Mujahid MS, Barr DA, et al. Social determinants of risk and outcomes for cardiovascular disease: a scientifi c statement from the American Heart Associa- tion. Circulation. 2015 Sep 1;132(9):873– 98. Epub 2015 Aug 3. https:// doi .org/ 10.1161/ CIR.0000000000000228 PMid:26240271

3. Cantor MN, Th orpe L. Integrating data on social determinants of health into electronic health records. Health Aff (Millwood). 2018 Apr;37(4):585– 90. https:// doi .org/ 10.1377/ hlthaff .2017.1252 PMid:29608369

4. Chetty R, Stepner M, Abraham S, et  al. Th e association between income and life expectancy in the United States, 2001– 2014. JAMA. 2016 Apr 26;315(16):1750– 66. https:// doi .org/ 10.1001/ jama.2016.4226 PMid:27063997 PMCid:PMC4866586

5. Mensah GA, Mokdad AH, Ford ES, et al. State of disparities in cardiovascular health in the United States. Circulation. 2005 Mar 15;111(10):1233– 41. https:// doi .org/ 10.1161/ 01.CIR.0000158136.76824.04 PMid:15769763

6. Schulz AJ, Kannan S, Dvonch JT, et al. Social and physical environments and disparities in risk for cardiovascular disease: the healthy environments partnership conceptual model. Environ Health Perspect. 2005 Dec;113(12):1817– 25. https:// doi .org/ 10.1289/ ehp.7913 PMid:16330371 PMCid:PMC1314928

7. Adler NE, Stewart J. Health disparities across the lifespan: meaning, methods, and mechanisms. Ann N Y Acad Sci. 2010 Feb;1186:5– 23. https:// doi .org/ 10.1111/ j.1749-6632.2009.05337.x PMid:20201865

8. Galea S, Tracy M, Hoggatt KJ, et al. Estimated deaths attributable to social factors in the United States. Am J Public Health. 2011 Aug;101(8):1456– 65. Epub 2011 Jun 16. https:// doi .org/ 10.2105/ AJPH.2010.300086 PMid:21680937 PMCid:PMC3134519

9. Jemal A, Th un MJ, Ward EE, et al. Mortality from leading causes by education and race in the United States, 2001. Am J Prev Med. 2008 Jan;34(1):1– 8. https:// doi .org/ 10.1016/ j.amepre.2007.09.017 PMid:18083444

10. Fitzpatrick T, Rosella LC, Calzavara A, et al. Looking beyond income and education: socioeconomic status gradients among future high- cost users of health care. Am J Prev Med. 2015 Apr;49(2):161– 71. Epub 2015 May 8. https:// doi .org/ 10.1016/ S0749-3797(15)00286-X

394 Social determinants in free and charitable clinics

11. Council On Community Pediatrics. Poverty and child health in the United States. Pediatrics. 2016 Apr;137(4). https:// doi .org/ 10.1542/ peds.2016-0339 PMid:26962238

12. Dzau VJ, McClellan MB, McGinnis JM, et al. Vital directions for health and health care: priorities from a national academy of medicine initiative. JAMA. 2017 Apr 11;317(14):1461– 70. https:// doi .org/ 10.1001/ jama.2017.1964 PMid:28324029

13. Billioux A, Verlander K, Anthony S, et al. Standardized screening for health- related social needs in clinical settings: the accountable health communities screening tool. National Washington, DC: Academy of Medicine, 2017. Available at: https:// nam .edu/ wp- content/ uploads/ 2017/ 05/ Standardized- Screening- for- Health- Related- Social - Needs-in-Clinical- Settings .pdf. https:// doi .org/ 10.31478/ 201705b

14. U. S. Department of Health and Huan Services. Healthy People 2020: an opportunity to address societal determinants of health in the United States. Secretary’s Advisory Committee on National Health Promotion and Disease Prevention Objectives for 2020. Washington, DC: U.S. Department of Health & Human Services, 2010. Avail- able at: https:// www .healthypeople .gov/ 2010/ hp2020/ advisory/ societaldeterminants health .htm.

15. Adler NE, Stead WW. Patients in context—EHR capture of social and behavioral determinants of health. N Engl J Med. 2015 Feb 19;372(8):698– 701. https:// doi .org/ 10.1056/ NEJMp1413945 PMid:25693009

16. Council on Community Pediatrics. Promoting food security for all children. Pediatrics. 2015 Nov;136(5):e1431– 8. https:// doi .org/ 10.1542/ peds.2015-3301 PMid:26498462

17. Daniel H, Bornstein SS, Kane GC. Addressing social determinants to improve patient care and promote health equity: an american college of physicians position paper. Ann Intern Med. 2018 Apr 17;168(8):577– 8. https:// doi .org/ 10.7326/ M17-2441 PMid:29677265

18. Gottlieb LM, Wing H, Adler NE. A systematic review of interventions on patients’ social and economic needs. Am J Prev Med. 2017 Nov;53(5):719– 29. Epub 2017 Jul 5. https:// doi .org/ 10.1016/ j.amepre.2017.05.011 PMid:28688725

19. Gottlieb L, Hessler D, Long D, et  al. A randomized trial on screening for social determinants of health: the iScreen study. Pediatrics. 2014 Dec;134(6):e1611– 1618. https:// doi .org/ 10.1542/ peds.2014-1439 PMid:25367545

20. Gottlieb LM, Hessler D, Long D, et al. Eff ects of social needs screening and in-person service navigation on child health: a randomized clinical trial. JAMA Pediatr. 2016 Sep 7;170(11):e162521. Epub 2016 Nov 7. https:// doi .org/ 10.1001/ jamapediatrics.2016.2521 PMid:27599265

21. Beck AF, Cohen AJ, Colvin JD, et  al. Perspectives from the society for pediatric

395Palakshappa, Scheerer, Semelka, and Foley

research: interventions targeting social needs in pediatric clinical care. Pediatr Res. 2018 Jul;84(1):10– 21. Epub 2018 May 23. https:// doi .org/ 10.1038/ s41390-018-0012-1 PMid:29795202

22. Garg A, Toy S, Tripodis Y, et al. Addressing social determinants of health at well child care visits: a cluster RCT. Pediatrics. Feb 2015;135(2):e296– 304. Epub 2015 Jan 5. https:// doi .org/ 10.1542/ peds.2014-2888 PMid:25560448 PMCid:PMC4306802

23. Alderwick HAJ, Gottlieb LM, Fichtenberg CM, et al. Social prescribing in the U.S. and England: emerging interventions to address patients’ social needs. Am J Prev Med. 2018 May;54(5):715– 8. Epub 2018 Mar 15. https:// doi .org/ 10.1016/ j.amepre.2018.01.039 PMid:29551326

24. Gottlieb LM, Garcia K, Wing H, et  al. Clinical interventions addressing non- medical health determinants in Medicaid managed care. Am J Manag Care. 2016 May;22(5):370– 6.

25. DeVoe JE, Bazemore AW, Cottrell EK, et al. Perspectives in primary care: a conceptual framework and path for integrating social determinants of health into primary care practice. Ann Fam Med. 2016 Mar;14(2):104– 8. https:// doi .org/ 10.1370/ afm.1903 PMid:26951584 PMCid:PMC4781512

26. Alley DE, Asomugha CN, Conway PH, et  al. Accountable health communities— addressing social needs through Medicare and Medicaid. N Engl J Med. 2016 Jan 7;374(1):8– 11. Epub 2016 Jan 5. https:// doi .org/ 10.1056/ NEJMp1512532 PMid:26731305

27. Gottlieb L, Colvin JD, Fleegler E, et al. Evaluating the Accountable Health Communi- ties Demonstration Project. J Gen Intern Med. 2017 Mar 2;32(3):345– 9. Epub 2016 Nov 14. https:// doi .org/ 10.1007/ s11606-016-3920-y PMid:27844261 PMCid:PMC5331008

28. Th e National Association of Free & Charitable Clinics (NAFC). NAFC home page [website]. Alexandria, VA: NAFC, 2016. Available at: http:// www .nafcclinics .org.

29. Rubin R. Half- century aft er “summer of love,” free clinics still play vital role. JAMA. 2017 Aug 15;318(7):598– 600. https:// doi .org/ 10.1001/ jama.2017.8631 PMid:28746712

30. arnell JS. Free clinics in the United States: a nationwide survey. Arch Intern Med. 2010 Jun 14;170(11):946– 53. https:// doi .org/ 10.1001/ archinternmed.2010.107 PMid:20548006

31. Swan GA, Foley KL. Th e perceived impact of the Patient Protection and Aff ordable Care Act on North Carolina’s free clinics. N C Med J. 2016 Jan– Feb;77(1):23– 9. https:// doi .org/ 10.18043/ ncm.77.1.23 PMid:26763240

32. Smith S, Malinak D, Chang J, et al. Implementation of a food insecurity screening and referral program in student- run free clinics in San Diego, California. Prev Med Rep. 2017 Mar;5:134– 9.

396 Social determinants in free and charitable clinics

https:// doi .org/ 10.1016/ j.pmedr.2016.12.007 PMid:27990340 PMCid:PMC5157787

33. Centers for Medicare & Medicaid Services. Accountable Health Communities Model [interactive map]. Baltimore, MD: U S Centers for Medicare and Medicaid Services, 2019. Available at: https:// innovation.cms .gov/ initiatives/ ahcm/.

34. Andermann A. Screening for social determinants of health in clinical care: moving from the margins to the mainstream. Public Health Rev. 2018 Jun 22;39:19. https:// doi .org/ 10.1186/ s40985-018-0094-7 PMid:29977645 PMCid:PMC6014006

35. Garg A, Butz AM, Dworkin PH, et al. Screening for basic social needs at a medical home for low-income children. Clin Pediatr (Phila). 2009 Jan;48(1):32– 6. https:// doi .org/ 10.1177/ 0009922808320602 PMid:18566347

36. Riley T, Baisden J. Th e Health Outcomes Initiative of the North Carolina Association of Free Clinics. N C Med J. 2014 May– Jun;75(3):218– 9. https:// doi .org/ 10.18043/ ncm.75.3.218 PMid:24830500

37. Byhoff E, Cohen AJ, Hamati MC, et al. Screening for social determinants of health in Michigan health centers. J Am Board Fam Med. 2017 Jul– Aug;30(4):418– 27. https:// doi .org/ 10.3122/ jabfm.2017.04.170079 PMid:28720624

38. Gottlieb L, Cottrell EK, Park B, et al. Advancing social prescribing with implementa- tion science. J Am Board Fam Med. 2018 May– Jun;31(3):315– 21. https:// doi .org/ 10.3122/ jabfm.2018.03.170249 PMid:29743213

39. North Carolina Department of Health and Human Services. Using standardized soclal determinants of health screening questions to identify and assist patients with unmet health- related resource needs in North Carolina. Raleigh, NC: North Carolina Department of Health and Human Services, 2018. Available at: https:// fi les.nc .gov / ncdhhs/ documents/ SDOH- Screening- Tool_Paper_FINAL_20180405 .pdf.

40. Byhoff E, Garg A, Pellicer M, et al. Provider and staff feedback on screening for social and behavioral determinants of health for pediatric patients. J AM Board Fam Med. 2019 May;32(3):297– 306. https:// doi .org/ 10.3122/ jabfm.2019.03.180276 PMid:31068394

41. Hoisington AT, Braverman MT, Hargunani DE, et al. Health care providers’ attention to food insecurity in households with children. Prev Med. 2012 Sep;55(3):219– 22. https:// doi .org/ 10.1016/ j.ypmed.2012.06.007 PMid:22710141

42. Byhoff E, Freund KM, Garg A. Accelerating the implementation of social determinants of health interventions in internal medicine. J Gen Intern Med. 2018 Feb;33(2):223– 5. https:// doi .org/ 10.1007/ s11606-017-4230-8 PMid:29188543 PMCid:PMC5789112

43. Hassan A, Scherer EA, Pikcilingis A, et al. Improving social determinants of health: eff ectiveness of a web- based intervention. Am J Prev Med. 2015 Dec;49(6):822– 31. Epub 2015 Jul 26. https:// doi .org/ 10.1016/ j.amepre.2015.04.023 PMid:26215831

397Palakshappa, Scheerer, Semelka, and Foley

44. Cottrell EK, Gold R, Likumahuwa S, et al. Using health information technology to bring social determinants of health into primary care: a conceptual framework to guide research. J Health Care Poor Underserved. 2018;29(3):949– 63. https:// doi .org/ 10.1353/ hpu.2018.0071 PMid:30122675 PMCid:PMC6779030

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.