307 - The Impact of Demographic Factors, Physician Factors, and Social Determinants of Health on Referrals to a Pediatric Community Health Worker Program

Poster Number: 307
Publication Number: 307.1879

Background: The pediatric community health worker (CHW) program was established in 2016 to support children with special health care needs1. CHW models engage peers to improve health outcomes through care coordination support while reducing health disparities. However, few studies have evaluated program factors that may affect whether caregivers enroll.
1 Costich MA, Peretz PJ, Davis JA, Stockwell MS, Matiz LA. Impact of a Community Health Worker Program to Support Caregivers of Children With Special Health Care Needs and Address Social Determinants of Health. Clinical Pediatrics. 2019;58(11-12):1315-1320.

Objective: To determine the demographic, social, or physician referral factors that may impact enrollment in the CHW program.

Design/Methods: Data from January 2020 to November 2022 was gathered with randomized sampling of patients of four academic practices. Charts were manually reviewed to abstract demographic, social, and referral information in the medical record. Categories were created for demographics, social determinants of health (SDOH), medical complexity based on local methodology2, and reasons for referral. Statistical analysis was done in R.
2 Matiz LA, Robbins-Milne L, Rausch JA. EMR Adaptations to Support the Identification and Risk Stratification of Children with Special Health Care Needs in the Medical Home. Matern Child Health J. 2019 Jul;23(7):919-924.

Results: There were no significant differences in age of enrollees, though younger children were more often referred (Table 1). There was a statistically significant difference in enrollment between Hispanic and non-Hispanic patients and similarly, the most prevalent language of enrollees was Spanish and the most declined was English. No differences were found in interpreter need or race (Table 1). 56% of enrollees endorsed at least 1 of 3 insecurities assessed vs 44% of the unenrolled group (Table 2), with each insecurity being individually significant. The most common reason for referral in both groups was for navigation and there was no difference between the groups based on medical complexity (Table 3). Resident referrals did not differ significantly regarding enrollment from attendings.

Conclusion(s): Ultimately, there are few intrinsic program factors affecting enrollment. Significant differences in ethnicity and language without difference in interpreter use may suggest cultural differences at play. Differences in SDOH support the role of CHWs in screening. The lack of statistical difference in the physician factors for referral was reassuring suggesting similarities in program knowledge.