Higher rates of routine screening and referrals for social needs in pediatric primary care were not associated with an increase in patient enrollment in community-based resources, a clinical study published in Pediatrics has found.

“Our results carry important implications for the integration of social care into pediatric primary care and may temper expectations for the impact of such social care systems implemented in real-world practice conditions on connecting families to resources,” the study’s authors wrote.

Ever since 2016 when the American Academy of Pediatrics first recommended screening for social needs such as nutrition or housing assistance, doing so has been an integral feature of pediatric primary care, according to Arvin Garg, MD, MPH, and his co-authors.

Previously, Garg and co-authors studied the impact of social needs screening and referrals during well-child visits at urban community health centers. Yet, according to Garg and co-authors, few data have been collected on the impact of this kind of screening in busy, private primary care pediatric practices with racially, ethnically, and socioeconomically diverse patient panels.

Garg is a professor of pediatrics and the vice chair of health equity at the UMass Chan Medical School and UMass Memorial Children’s Medical Center, both in Worcester, Massachusetts.

The type 2 hybrid effectiveness-implementation stepped wedge cluster trial, WE CARE assessed the effect of the social needs intervention on diverse patient families seen in 18 pediatric practices across 14 states in the US. Participating sites excluded federally qualified health centers (FQHCs) and practices that already had routine screening protocols in place.

The parents (N = 1882) of children aged 2 months to 10 years were enrolled either during usual care or WE CARE phases at their well-child visit and were seen for follow-up care at 3 months. All sites were part of the American Academy of Pediatrics’ Pediatric Research in Office Settings and Academic Pediatric Association Continuity Research Network.

In all, there were eight urban, six suburban, and four rural practices. These were grouped into three clusters of six practices. All 18 practices participated in usual care, core training, pilot, and WE CARE phases.

For the intervention, parents were given a self-administered screening tool to help determine their social needs, while clinicians were given access to practice-generated family resources materials. The screening instrument, available in both English and Spanish, was written at a third-grade level and was designed to identify six social needs and determine whether parents wanted assistance with them. It was designed to be completed in less than 5 minutes.

Staff at the participating facilities had 3 months to create their respective community resource books, which included one-pagers with between two and four local resources for each social need, with all information in both English and Spanish.

In the usual care phase,which lasted up to 6 months before practices switched to the study intervention, a cohort of parents was enrolled to complete a baseline survey and a 3-month follow-up survey. During the core training phase, key staff were trained to administer the intervention screening tool and designed the resource materials. Any barriers to implementation were addressed during the pilot phase.

In the WE CARE phase, practices implemented the intervention as standard of care for well-care visits for patients aged 2 months to 10 years. Practices could opt to expand the protocol to other visit types (sick visits, etc.) and age ranges. During this phase, a second cohort of parents was enrolled to complete surveys similar to parents in the usual care phase.

Of the 1882 enrolled, 842 parents were randomly assigned to the WE CARE arm, and 1040 were assigned to the usual care arm. In all, 64% of both cohorts combined completed follow-up. Across the two groups, 60% of children were publicly insured.

A significantly higher proportion of WE CARE vs usual care parents reported discussing social needs with their child’s clinician (91% vs 79%; adjusted odds ratio [aOR], 3.6; 95% CI, 2.6-4.8). Individuals in the intervention group also received at least one referral (20% vs 12%; aOR, 1.7; 95% CI, 1.3-2.2). At the 3-month follow-up, there were no self-reported differences in enrollment in new community resources between WE CARE and usual care parents (23% vs 21%; P = .63).

“I would say that overall, as the study progressed over time, we were not too surprised with this result,” Garg told Medscape Medical News. Garg suggested some potential reasons for the study’s negative result.

“Many families who requested help on the WE CARE screener did not get a referral; compared to our previous trial, our follow-up time was shorter and the intensity of the intervention was less, namely, we did not have research staff members assist with completing referrals or paperwork associated with public benefits; and lastly, [there were] inherent challenges with accessing a limited and insufficient social safety net.”

In an accompanying editorial, experts from the Cincinnati Children’s Hospital and the University of Cincinnati College of Medicine wrote, “Limited effects of interventions like WE CARE on resource enrollment and limited measurement of the impact of such interventions on meaningful, measurable medical and social outcomes are the crux of current debates about the how, what, and when of health-related social needs screening and response.”

The editorial was co-written by Alexandria N. Card, MD, a clinical fellow in hospital medicine; Adrienne W. Henize, JD, an associate professor and child health equity specialist; and Andrew F. Beck, MD, MPH, a professor of pediatrics at the University of Cincinnati and an attending physician in the Divisions of General and Community Pediatrics and Hospital Medicine at Cincinnati Children’s Hospital in Cincinnati.

The editorial writers concluded, “Screening does not always identify needs, nor does it always prompt referral to or enrollment in potentially helpful resources.” Future research, they said, should push for an examination of each step in the process from screening to response from the varied perspectives of clinicians, patients, family members, and community partners.

Jay W. Lee, MD, MPH said, “As a family physician who works with socially vulnerable patients at a [FQHC] network, [I can say] nearly all of our patients are screened for and subsequently referred to meet health-related social needs. And similar to the study population, many of our patients are unable to complete referrals despite enhanced screening and relationships with community-based organizations,” he said.

“The reasons are multifold,” said Lee, who is the medical director of Integrated Health Partners of Southern California in Long Beach, California. First, he said, like FQHCs, many community-based clinics are resource limited and are unable to meet the high volume of referrals. The result is some patients are able to access services but not all due to constrained supply. Second, many patients who experience health-related social needs are already overwhelmed. Even with a referral in hand, they may not have the bandwidth to follow through due to competing needs. Last, data systems do not talk with each other well, so a referral from a medical system often does not cross platforms in a way that makes it easy for organizations to share information, he said.

None of the authors or editorial writers reported conflicts of interest. Lee did not report any conflicts of interest.

Whitney McKnight is a freelance medical and health policy writer based in Lexington, Kentucky.