Research ArticleOriginal Research

Adoption of Social Determinants of Health EHR Tools by Community Health Centers

Rachel Gold, Arwen Bunce, Stuart Cowburn, Katie Dambrun, Marla Dearing, Mary Middendorf, Ned Mossman, Celine Hollombe, Peter Mahr, Gerardo Melgar, James Davis, Laura Gottlieb and Erika Cottrell
The Annals of Family Medicine September 2018, 16 (5) 399-407; DOI: https://doi.org/10.1370/afm.2275

Article Figures & Data

Tables

  • Table 1

    SDH Data Tools and SDH Domains

    SDH Data ToolsDescription
    SDH Data Collection ToolsIncluded 14 SDH screening questions based on PRAPARE and National Academy of Medicine recommendations. Data collection modes included: data-entry flowsheets accessible by diverse clinic staff, a print version for patients to complete after which the data would be entered by CHC staff into a flowsheet, and an online portal form that patients could complete before the visit.
    SDH Summary ToolsPatient’s most recent SDH data displayed (as entered in flowsheets or elsewhere in the EHR), and past SDH-related referrals.
    SDH Data RostersAdded SDH-related data columns to the EHR’s panel management tools to identify patients who (1) had a pending visit (enabling e-mailing those with online portal accounts about completing SDH screening pre-appointment); (2) had a positive SDH screen and needed follow-up; or (3) were due for SDH screening.
    Problem ListCreated a new SDH class of problem list diagnoses, so that users could manually categorize SDH diagnoses in the problem list.
    SDH Referral ToolsBuilt as preference lists, to parallel the clinics’ processes for making clinical referrals. Worked with pilot clinics to create preference lists of local resources for addressing specific SDH needs. Used to add information about relevant resources to the patient’s after-visit summary and to identify resources that clinic staff could discuss with the patient.
    SDH Domainsa
    Alcohol usebEducationExposure to violence
    Race/ethnicitybFinancial resource strainPhysical inactivity
    Tobacco use and exposurebHousing insecuritySocial isolation
    DepressionbFood insecurityStress

    • CHC = community health centers; EHR = electronic health record; PRAPARE = Protocol for Responding to and Assessing Patient Assets, Risks, and Experiences; SDH = social determinants of health.

    • ↵a Wording defined in Supplemental Appendix 1, available at http://www.annfammed.org/content/16/5/399/suppl/DC1/.

    • ↵b Information on these domains is routinely documented elsewhere in the EHR so they were not included in the SDH data collection tool flowsheet. Responses, however, were pulled into the SDH summary tool.

  • Table 2

    Patient and Visit Characteristics of Patients Seen During the Study Period, and of Those Screened for SDH, by Clinic

    Patient CharacteristicsClinic AClinic BClinic C
    Total Patients No. (%)Screened Patients No. (%)Total Patients No. (%)Screened Patients No. (%)Total Patients No. (%)Screened Patients No. (%)
    Number of patients4,208 (100.0)602 (14.3)2,126 (100.0)379 (17.8)3,741 (100.0)149 (4.0)
    Race
     American Indian/AK Native122 (2.9)14 (2.3)56 (2.6)9 (2.4)39 (1.0)0 (0.0)
     Asian50 (1.2)5 (0.8)30 (1.4)4 (1.1)494 (13.2)9 (6.0)
     Black/African American62 (1.5)8 (1.3)31 (1.5)5 (1.3)303 (8.1)9 (6.0)
     Native Hawaiian/PI39 (0.9)6 (1.0)15 (0.7)5 (1.3)21 (0.6)2 (1.3)
     White3,798 (90.3)541 (89.9)1,726 (81.2)322 (85.0)2,569 (68.7)105 (70.5)
     Multiple races73 (1.7)12 (2.0)88 (4.1)13 (3.4)70 (1.9)4 (2.7)
     Unknown64 (1.5)16 (2.7)180 (8.5)21 (5.5)245 (6.5)20 (13.4)
    Hispanic
     Yes403 (9.6)40 (6.6)627 (29.5)75 (19.8)531 (14.2)22 (14.8)
     No3,710 (88.2)545 (90.5)1,457 (68.5)292 (77.0)3,110 (83.1)121 (81.2)
     Unknown95 (2.3)17 (2.8)42 (2.0)12 (3.2)100 (2.7)6 (4.0)
    Sex
     Female2,416 (57.4)315 (52.3)1,287 (60.5)200 (52.8)1,898 (50.7)42 (28.2)
     Male1,792 (42.6)287 (47.7)837 (39.4)178 (47.0)1,843 (49.3)107 (71.8)
     Unknown0 (0.0)0 (0.0)2 (0.1)1 (0.3)0 (0.0)0 (0.0)
    Age: 1st study period visit, y
     18–291,069 (25.4)169 (28.1)544 (25.6)42 (11.1)817 (21.8)48 (32.2)
     30–491,793 (42.6)241 (40.0)872 (41.0)142 (37.5)1,478 (39.5)69 (46.3)
     50–641,173 (27.9)170 (28.2)551 (25.9)140 (36.9)1,009 (27.0)27 (18.1)
     ≥65173 (4.1)22 (3.7)159 (7.5)55 (14.5)437 (11.7)5 (3.4)
    Homeless status
     Yes64 (1.5)7 (1.2)72 (3.4)13 (3.4)55 (1.5)1 (0.7)
     No1,858 (44.2)198 (32.9)713 (33.5)114 (30.1)1,299 (34.7)44 (29.5)
     Unknown2,286 (54.3)397 (65.9)1,341 (63.1)252 (66.5)2,387 (63.8)104 (69.8)
    Migrant/seasonal worker
     Yes13 (0.3)3 (0.5)47 (2.2)0 (0)7 (0.2)0 (0)
     No1,911 (45.4)200 (33.2)728 (34.2)131 (34.6)936 (25.0)24 (16.1)
     Unknown2,284 (54.3)399 (66.3)1,351 (63.5)248 (65.4)2,798 (74.8)125 (83.9)
    Primary payer
     Medicaid2,957 (70.3)416 (69.1)1,189 (55.9)193 (50.9)2,313 (61.8)84 (56.4)
     Medicare455 (10.8)52 (8.6)215 (10.1)78 (20.6)567 (15.2)13 (8.7)
     Other public11 (0.3)2 (0.3)9 (0.4)0 (0.0)5 (0.1)0 (0.0)
     Private264 (6.3)39 (6.5)299 (14.1)40 (10.6)94 (2.5)4 (2.7)
     Uninsured521 (12.4)93 (15.4)414 (19.5)68 (17.9)762 (20.4)48 (32.2)
    Primary language
     English3,915 (93.0)582 (96.7)1,703 (80.1)330 (87.1)2,761 (73.8)126 (84.6)
     Spanish189 (4.5)6 (1.0)418 (19.7)48 (12.7)336 (9.0)12 (8.1)
     Other56 (1.3)6 (1.0)4 (0.2)1 (0.3)639 (17.1)11 (7.4)
     Unknown48 (1.1)8 (1.3)1 (0.0)0 (0.0)5 (0.1)0 (0.0)
    Veteran status
     Yes118 (2.8)26 (4.3)87 (4.1)17 (4.5)78 (2.1)10 (6.7)
     No4,049 (96.2)566 (94.0)2,032 (95.6)360 (95.0)3,358 (89.8)112 (75.2)
     Unknown41 (1.0)10 (1.7)7 (0.3)2 (0.5)305 (8.2)27 (18.1)
    Diabetes status
     Yes557 (13.2)68 (11.3)279 (13.1)110 (29.0)531 (14.2)7 (4.7)
     No3,651 (86.8)532 (88.4)1,847 (86.9)269 (71.0)3,210 (85.8)142 (95.3)
     Unknown0 (0.0)2 (0.3)0 (0.0)0 (0.0)0 (0.0)0 (0.0)
    New/established patients
     New patients699 (16.6)311 (51.7)239 (11.2)57 (15.0)1,251 (33.4)142 (95.3)
     Established patients3,509 (83.4)291 (48.3)1,887 (88.8)322 (85.0)2,490 (66.6)7 (4.7)
    Number of visits13,990 (100.0)611 (4.4)8,162 (100.0)385 (4.7)16,281 (100.0)149 (0.9)
    Type of practioner
     MD, DO, Locum Tenens2,686 (19.2)7 (1.1)3,892 (47.7)209 (54.3)7,663 (47.1)149 (100.0)
     NP, PA8,827 (63.1)24 (3.9)2,577 (31.6)76 (19.7)4,399 (27.0)0 (0.0)
     RN, LPN, CHN1,359 (9.7)187 (30.6)1,427 (17.5)79 (20.5)2,428 (14.9)0 (0.0)
     MA1,049 (7.5)1 (0.2)59 (0.7)0 (0.0)109 (0.7)0 (0.0)
     BHS, LCSW…0 (0.0)181 (2.2)18 (4.7)1,550 (9.5)0 (0.0)
     Eligibility specialist…392 (64.2)…0 (0.0)…0 (0.0)
     Other69 (0.5)0 (0.0)26 (0.3)3 (0.8)132 (0.8)0 (0.0)
    Clincian status
     Primary care clinician7,119 (50.9)22 (3.6)4,439 (54.4)204 (53.0)9,151 (56.2)145 (97.3)
     Other6,871 (49.1)589 (96.4)3,723 (45.6)181 (47.0)7,130 (43.8)4 (2.7)

    • AK = Alaska; BHS = behavioral health specialist, CHN = community health nurse; DO = doctor of osteopathy; LCSW = licensed clinical social worker; LPN = licensed practical nurse; MA = medical assistant; MD = doctor of medicine; NP = nurse practitioner; PA = physician’s assistant; PI = Pacific Islander; RN = registered nurse.

  • Table 3

    Patients Screened for SDH Over Time, by Study Clinic (N = 1,130)

    MonthDistinct Patients Screened, No.
    Clinic A (n = 602)Clinic B (n = 379)Clinic C (n = 149)
    Jul 20161269
    Aug 2016321331
    Sep 201684819
    Oct 201678719
    Nov 201678427
    Dec 2016702212
    Jan 2017782319
    Feb 2017522113
    Mar 201733100
    Apr 201721510
    May 2017301010
    Jun 201728310
    Jul 201717620

    • SDH = social determinants of health.

    • Note: Clinic C data based on encounters with 1 provider. SDH screening stopped in Clinic C in February 2017 for reassessment of workflows and EHR access policies.

  • Table 4

    Screening Results and Referral Rates

    Study ClinicPatients Screened, No.Screened Patients With SDH DomainDomains for Patients With Positive Screen, No. (%)Patients With Positive Screen and Matching
    Positive SDH Screen, No. (%)SDH Referral, No. (%)SDH Referrala, No. (%)Problem List dx, No. (%)
    A602583 (96.8)141 (23.4)Financial resource strain426 (70.8)105 (24.6)22 (5.2)
    Housing insecurity206 (34.2)60 (29.1)19 (9.2)
    Food insecurity331 (55.0)91 (27.5)22 (6.6)
    Intimate partner violence175 (29.1)1 (0.6)0 (0.0)
    Inadequate physical activity311 (51.7)0 (0.0)0 (0.0)
    Social isolation433 (71.9)12 (2.8)4 (0.9)
    Stress436 (72.4)0 (0.0)2 (0.5)
    B379367 (96.8)26 (6.8)Financial resource strain277 (73.1)1 (0.4)0 (0.0)
    Housing insecurity103 (27.2)3 (2.9)0 (0.0)
    Food insecurity216 (57.0)0 (0.0)0 (0.0)
    Intimate partner violence94 (24.8)0 (0.0)0 (0.0)
    Inadequate physical activity167 (44.1)0 (0.0)0 (0.0)
    Social isolation235 (62.0)0 (0.0)0 (0.0)
    Stress253 (66.8)0 (0.0)2 (0.8)
    C149148 (99.3)44 (29.5)Financial resource strain107 (71.8)3 (2.8)1 (0.9)
    Housing insecurity56 (37.6)3 (5.4)1 (1.8)
    Food insecurity86 (57.7)2 (2.3)0 (0.0)
    Intimate partner violence36 (24.2)0 (0.0)0 (0.0)
    Inadequate physical activity63 (42.3)0 (0.0)0 (0.0)
    Social isolation111 (74.5)1 (0.9)0 (0.0)
    Stress107 (71.8)1 (0.9)8 (7.5)

    • dx = diagnosis; EHR = electronic health record; SDH = social determinants of health.

    • ↵a Referrals were matched to screening domains based on evaluation of EHR documentation associated with the referral order, including type and/or specialty of the referral provider and diagnoses associated with the referral.

Additional Files

  • Supplemental Appendixes 1-3

    Supplemental Appendixes 1-3

    Files in this Data Supplement:

  • The Article in Brief

    Adoption of Social Determinants of Health EHR Tools by Community Health Centers

    Rachel Gold , and colleagues

    Background A growing awareness that social factors--the conditions in which people live, work and play--influence health suggests that it is crucial to document such information in patients' electronic health records. This pilot study assesses the feasibility of implementing electronic health record tools for collecting, reviewing, and acting on patient-reported social determinants of health data in community health centers.

    What This Study Found The study found that adopting EHR tools to systematically document social determinants of health in primary care is feasible, but substantial barriers exist. Researchers implemented social determinants data tools in three Pacific Northwest community health centers. Among 1,130 patients for whom social determinants data were collected, 97 to 99 percent (n = 1,098) had one or more social need documented in the EHR, with 210 (19 percent) receiving an EHR-documented social determinants referral. Fifteen to 21 percent of patients with a documented social need wanted help from the clinic to address the need. Although the study identified many barriers to implementing and designing tools and workflows, participating community health centers successfully documented social determinants in the EHR and continued to do so post-study.

    Implications

    • The authors explain that, to meet the growing national emphasis on documentation of social determinants of health in EHRs, a wide range of factors and substantial gaps in knowledge must be addressed.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Get Permissions

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles

Subjects

Keywords