Predicting and Addressing Elevated Social Risk Among Adolescents in the Emergency Department

Background: Social determinants of health (SDH) affect overall health outcomes and health care usage in adolescents. The emergency department (ED) often serves as a safety net for vulnerable adolescents and may be a strategic venue in which to identify and address SDH. The objectives of this analysis were 1) to identify predictors of multiple unmet social needs among adolescent ED patients and 2) to test the effects of a pilot social navigation program on adolescent ED recidivism and community resource use.

Methods: We administered a tablet-based survey assessing unmet social needs to adolescent patients ages 13-21 presenting to a two-campus urban pediatric ED. Responses were used to tabulate a social risk score (SRS) and provide community resource referrals. Patients were randomly placed into an intervention group which received social navigation tailored to identified social risks, or an enhanced usual care group which received a preprinted packet of community resources. Each enrolled patient then received follow up calls at 3 and 12 months to assess resource use. We performed chart review at 12 months to calculate the number of repeat ED visits. Primary outcomes were the SRS and number of repeat ED visits at 12 months. We performed multivariable logistic regression to measure associations between a high SRS (≥3) and obesity, frequent non-urgent visits, food insecurity, or an at-risk chief complaint. We used the Wilcoxon rank sum test to compare the median number of repeat visits in the intervention and control groups. We studied community resource use after ED discharge and repeat visits for at-risk CC as secondary outcomes.

Results: 737 adolescents were approached for participation. 384 adolescents (68.2% non-Hispanic Black, 63.3% female gender identity) completed the survey. 95.3% reported at least one unmet social need, and 57.3% identified ≥3 social needs. Adolescents with an at-risk CC had twice the odds of a high SRS; adolescents reporting food insecurity had nearly 10 times the odds of a high SRS (Table 1). Obesity and frequent non-urgent visits were not predictive of a high SRS. There was no significant difference between the intervention and enhanced usual care groups’ rate of any return ED visit or return visits for an at-risk CC (Table 2). Among patients who completed at least one follow up phone call (54.4%), there was more community resource use reported in the intervention group vs. the control group (Table 2).

Conclusion: There is a high burden of unmet social needs in adolescents who visit the ED. Those who report food insecurity or present for an at-risk chief complaint are likely to have multiple unmet needs. Adolescents who receive social navigation report increased community resource use. Future studies should focus on partnerships with community organizations who can provide sustained follow-up after ED social navigation.