INTRODUCTION

Interest in integrating social and medical care in the US health care system is growing.1 This is especially true in Bureau of Primary Healthcare (BPHC)–funded health centers, which disproportionately serve patients whose health is affected by socioeconomic disadvantage. Little is known about the prevalence of social risks2 or social care interventions in BPHC settings. The current study examines the prevalence of social risk factors and social care among patients served in BPHC-funded health centers.

METHODS

We analyzed data from the 2014 Health Center Patient Survey (HCPS), a cross-sectional survey administered by the Health Resources and Services Administration (HRSA). The HCPS was conducted with a random convenience sample of patients from health centers receiving BPHC grants, including Community Health Centers (CHCs), Migrant Health Centers (MHCs), Health Care for the Homeless (HCH), and Public Housing Primary Care (PHPC) programs. Details of HCPS sampling scheme and administration are described elsewhere.3,4 Survey data were merged with annually reported 2014 Uniform Data System (UDS) data, which include health center–reported characteristics.5

HCPS asked multiple questions related to patients’ social risks, including employment status, education level, housing stability, and household income. Patients were also asked if they had ever received social care, including assistance from their health center to obtain free medications, access medical transportation, apply for government benefits, or access other basic needs resources (e.g., housing, employment, childcare, food, or clothing/shoes). We examined patient- and health center–level predictors of both social risk and social assistance. Predictor variables from HCPS included patient sociodemographics, health status, health care utilization, clinic urbanicity, and grantee type. Health center–level predictors from UDS included proportion of total staff FTE dedicated to enabling services (non-clinical services to improve health care access and outcomes),6 panel size, and regional location.

Weighted frequencies of patient demographics, health, and social risk factors, as well as health center–reported characteristics, were calculated in the full sample, then again in only those reporting receiving assistance, stratifying by type of assistance received. Chi-squared tests were used to compare differences between stratified samples. All data analyses were conducted using Stata/SE 15.0.

RESULTS

Of the 7002 survey respondents, 4225 (60.3%) adult patients from 167 health centers had recorded responses to the measures of social risk factors or key study demographic variables and were included in the final study sample. Respondent-reported demographics, health status and health care utilization, social risk factors and social care, and health center–level characteristics are shown in Table 1. Assistance with different types of social care varied by patient and health center characteristics (Table 2).

Table 1 Descriptors of Patient Respondents to 2014 Health Center Patient Survey (n = 4225 Patients) and Matched Bureau of Primary Health Care-Funded Health Center Characteristics (n = 167 Health Centers)
Full size table
Table 2 Percentage of Patients Who Reported Receiving Assistance from Their Bureau of Primary Health Care–Funded Health Center, by Patient Demographics, Self-Reported Health Status and Health Care Utilization, Practice-Level Characteristics, and Self-Reported Social Risk Factors
Full size table

DISCUSSION

This study is the first to evaluate the prevalence of patient-reported social risk factors and delivery of social care in a nationally representative sample of federally funded health centers. The population surveyed reported relatively high levels of social risks and low levels of assistance. Patients primarily reported receiving assistance with health care access and other medical care–related social risks, like transportation. Fewer patients reported assistance around basic material needs like food or housing. Receiving assistance varied by sociodemographic, health care utilization, and health center characteristics. In general, higher proportions of patients with poorer health status reported assistance accessing medical transportation and obtaining free medications. At the health center level, patients of clinics with more enabling services staffing reported receiving more assistance with basic material needs resources.

The HCPS survey design should influence the interpretation of these findings. The cross-sectional survey relied entirely on patient-reported health indicators. Questions about assistance were not temporally bounded (e.g., “Have you ever received assistance with...”), nor did they have a corresponding question about whether the patient perceived a need for assistance. Additionally, questions about applying for benefits were ambiguous, but likely mostly pertained to enrolling in health insurance. Despite these limitations, the survey is a useful indicator of the prevalence of social risk factors and delivery of social care in federally funded health centers. We found that delivery of social care is uneven across health center settings, driven both by patient- and health center–level variables. This may suggest that equally complex patients may receive different services in different settings. As interest grows around integrated social and medical care delivery, policy innovations can use data about the prevalence of social risk and availability of social care initiatives to improve comprehensive health programs.