Identifying Social Determinants of Health in an Inpatient Pediatric Setting

Background: Social determinants of health (SDOH)—defined by the Centers for Disease Control and Prevention as conditions in the places where people live, learn, work, and play—affect a wide range of health risks and outcomes. A recent U.S. Census Bureau community survey, showed our local poverty rate was 18.2% compared to the national average of 11.8%. Our institution serves a large portion of these vulnerable patients, the majority (70%) of whom are on Medicaid. Although there have been negative outcome associations with positive SDOH screening in the Emergency Department and outpatient settings, there has been limited study of SDOH screening in the Inpatient Pediatric setting. This study aims to test the feasibility of SDOH screening on an Inpatient Pediatric service, to determine SDOH needs within our patient population, and to analyze association of positive SDOH with hospitalization metrics.

Methods: A 14-question SDOH screening tool was performed on families by physicians or medical students as an addition to the social history documentation. Families who screened positive were given a resource list or referred for social work consultation.

Results: During this study, there were 1263 admissions to the Inpatient Pediatric service and 577 (40%) of these admissions were screened with a current 6-week moving average at 60%. A total of 211 admissions (36.6%) had positive SDOH screens with 537 needs identified. Frequency of needs were: 1-45%, 2-22%, 3-12%, >4-21%. Using student’s t-test, there was a significant difference in length of stay (LOS): 86.2 hours for positive screens vs. 62.6 hours for negative screens (p=0.01). Average patient complexity score [1 (Healthy) - 9 (Catastrophic)] was 4.4 vs 3.8 (p=0.03) using chi-squared statistic. There was no difference in 7- or 30-day readmission rates even after stratifying for number of needs. Patients most frequently screened positive for food insecurity (n=89), requiring help with assistance programs (n=73), lack of social support (n=62), and difficulty paying for medical bills (n=59).

Conclusion: To date, this is the largest study determining SDOH needs in an inpatient pediatric population. This is the first study to show significant longer average LOS by 23.6 hours for patients with positive SDOH screens. We found no difference in readmission. We plan to further compare LOS to diagnosis-related group (DRG) mean LOS and perform multivariable modeling with specific SDOH needs. Using our data, we have been able to justify implementing screening to all our inpatient hospital sites and establishing partnership for an inpatient emergency food pantry.