Expanding Developmental and Behavioral Health Capacity in Pediatric Primary Care

Abstract

The Development and Behavior Access Clinic (DBAC) deploys a general pediatrician with brief/intensive training and proctoring by developmental-behavioral pediatricians (DBP) to provide developmental-behavioral (DB) care to children referred with mild/moderate complexity needs as determined by a team of clinical psychologists. This pilot study utilizes visit data, chart review, and surveys to assess wait times, need for subspecialty care, and referring clinician satisfaction. In its first 18 months, DBAC decreased the need for subspecialty DB care, providing initial services for 44% of patients referred for DB pediatric care from the study site; 89% did not require subsequent subspecialty evaluation. Among DBAC participants, average wait times for DB care decreased from a baseline of 218 to 41 calendar days. This pilot study provides a model for building DB clinical skills among interested general pediatricians, decreasing wait times, and building the capacity of primary care settings to address mild-to-moderate complexity DB concerns.

Introduction

One in 6 US children aged 3 to 17 years have received a diagnosis of a developmental disability, ¹ and more than 1 in 5 US children experience a mental health disorder with severe impairment by ages 13 to 18 years. ² Despite increasing educational and implementation efforts, the majority of pediatric primary care clinicians report insufficient training and time to manage many developmental-behavioral (DB) and mental health concerns.^3-5 Approximately 70% of children and adolescents who need mental health services receive no treatment at all, ⁵ and up to 55% of children with neurodevelopmental conditions encounter difficulties in using services. ⁶ A shortage of subspecialty-trained developmental-behavioral pediatricians (DBP) and child and adolescent psychiatrists is an identified barrier to accessing care.^7,8

To help address this crisis, the “Development and Behavior Access Clinic” (DBAC) was launched to expand the capacity of an academic pediatric primary care clinic to diagnose and manage common DB and mental health conditions on-site. DBAC’s approach of deploying a DB-focused general pediatrician has shown success in other settings ⁹ and may have greater potential for sustainability than co-location of subspecialists, due to its more limited expectation for scarce subspecialists’ time. ¹⁰ This article describes the results of a pilot study of this model, focusing on feasibility, acceptability to referring primary care clinicians, and impact on wait times for DB services.

Methods

Clinical Model

The DBAC offers appointments for DB concerns of mild-to-moderate complexity/acuity where the referral concern is possible attention-deficit/hyperactivity disorder (ADHD), depression, anxiety, developmental delay, learning disability, disruptive/oppositional behavior, and/or common behavioral concerns such as sleep or feeding challenges. DBAC was staffed by a general pediatrician, selected for the role due to strong interest and clinical experience in child behavior and development, who had undergone specialized training in DB pediatrics and psychopharmacology (described below). This pediatrician was proctored by DB with experience in co-located models of DB pediatrics in primary care settings. Through a previously piloted, concurrent departmental initiative, ¹¹ a team of neuropsychologists and a nurse reviewed all new referrals to DB pediatrics and assigned them to DBAC or to DBP or other subspecialty clinic based on the clinical description of case severity and complexity. While there was not a formal algorithm for referral management, referrals were managed based on clinical expertise and knowledge of community resources as follows. First, referrals that were not appropriate to the subspecialty’s skillset, such as children with severe mental health conditions characterized by suicidal or homicidal ideation, self- or other-injurious behavior, and/or active psychosis, were forwarded to child and adolescent psychiatry either internally or in the community depending on insurance and family residence. Children likely needing more than 2 psychotropic medications or comprehensive neurodevelopmental testing were assigned to see a DBP or neuropsychological subspecialist. Children needing diagnostic testing for autism spectrum disorder were assigned to a DBP, since the general pediatrician lacked the technical expertise to provide this kind of testing; however, children with an autism diagnosis were seen in DBAC for assistance with treatment planning and care coordination. This pilot study was operated from June 2019 through September 2020 at a primary care pediatric clinic in an urban academic medical center.

Prior to DBAC’s launch, the general pediatrician completed the REACH Institute’s ¹² 16.25-hour “mini-fellowship” in pediatric psychopharmacology and a 34.5-hour Developmental-Behavioral Pediatric Board Review Course (DB:PREP) ¹³ sponsored by the American Academy of Pediatrics. The DBAC clinician also received training in the following developmental testing tools: the Capute Scales for cognitive and language delay in children aged 0 to 36 months, ¹⁴ the Screening Tool for Autism in Toddlers & Young Children (STAT), ¹⁵ the Bracken School Readiness Assessment, 3rd Edition (BRSA-3) to assess school-readiness skills in children aged 3 to 6, ¹⁶ and the Wide Range Achievement Test, 5th Edition (WRAT-5) for youth in grades kindergarten through 12 to monitor reading, spelling, and math skills. ¹⁷ All of these assessments were designed to take less than 30 minutes of face-to-face time and were only available at an in-office visit. For the first 3 months of the pilot study, the DBAC clinician also consulted by telephone with her DBP proctor for each DBAC case. After this initial period, the frequency of consultation decreased substantially, but still occurred intermittently when the general pediatrician had questions regarding diagnosis or treatment planning for DBAC patients.

DBAC started with a half-day clinic session per week co-located in the primary care clinic; due to demand, this expanded to a full-day clinic per week after 6 months. The COVID-19 pandemic forced a temporary closure of in-person clinic sessions, as services rapidly converted to telehealth platforms. Currently, DBAC operates on-site at the primary care clinic for a half-day session per week and by off-site telehealth for an additional 1 to 2 half-day sessions every week. New patient visits are 60 minutes and follow-up visits are 30 minutes; time-based billing is used. The DBAC clinician also provides informal consultation to other general pediatric attendings and regular teaching for pediatric residents.

Measures

Feasibility and sample demographics

To assess program feasibility, this study measured the number of unique children receiving clinical services from DBAC, the percentage of DBAC visit slots that were utilized, and clinic no-show rates during the pilot period, utilizing visit data obtained from the electronic health record. This quality improvement dataset also included basic demographic information, including gender, language, age at the time of visit, and payer type to describe the study sample and ascertain the characteristics of the children using this service.

Acceptability

To measure DBAC’s acceptability to referring clinicians, anonymous satisfaction surveys were emailed in October 2020 to all 15 general pediatric faculties associated with the practice in which DBAC is located. Recruitment emails with anonymous survey links were sent only once, as there was no way to track duplicate responses. To measure their satisfaction, clinicians used a virtual visual analog scale to indicate a satisfaction score between 0 and 100, where 0 = not at all satisfied and 100 = maximum satisfaction.

Impact on care access

DBAC sought to decrease wait times for DBP care. To measure wait time, a listing of all DBAC visits was obtained from the electronic health record, including date of referral, date of first visit, payer type, gender, date of birth, primary language, and visit type. Wait time was defined as the number of calendar days between the date of referral and date of first visit.

The initiative also sought to decrease the need for direct care by subspecialists, aiming to have less than 25% of DBAC patients requiring follow-up evaluation by DBP specialists. To measure the need for subsequent subspecialty evaluation, all DBAC patient charts were reviewed at the end of the 18-month pilot study to assess whether they saw a DBP, psychiatric, or neuropsychological subspecialist between the time of DBAC evaluation and the time of the chart review.

Data Analysis

Statistical analyses were primarily descriptive in nature. Counts, proportions, wait times (number of business days), and sample means and 95% confidence intervals were calculated from visit data using Microsoft Excel for Office 365. ¹⁸ Survey responses were used to calculate net promoter scores (NPS), which are calculated by subtracting the proportion of detractors (score ≤60/100) from the proportion of promoters (score ≥80/100). ¹⁹ An NPS >0 indicates that more clients are satisfied than dis-satisfied; NPS scores greater than or equal to 50/100 generally indicate positive word-of-mouth referrals and increased likelihood of expansion or growth. ²⁰

Ethics

This study protocol was approved by the study site’s Institutional Review Board.

Results

Sample Demographics and Feasibility

Eighty-nine unique children received care in DBAC between June 1, 2019, and September 30, 2020. The majority (75%) of the patients were male, English speaking (99%), and privately insured (63%). In comparison, 51% of primary care patients are male, 96% are English speaking, and 66% are privately insured. Individuals insured through Medicaid comprised 37% of DBAC patients, compared with 31% of the patients seen for primary care in this pediatric clinic. Of the demographic differences between the DBAC and primary care groups, only gender and average age were significantly different (P < .05), with DBAC patients more likely to be male and slightly younger (age 6.3 years DBAC, age 7.2 years primary care). See Table 1 for a more detailed comparison.

Table 1.

Demographics of DBAC Participants and Primary Care.

	DBAC (n = 89)		Primary care (n = 13 186)
	Number (%)	Mean (95% CI)-years	Number (%)	Mean (95% CI)-years
Gender
Male	67 (75)*		6769 (51)*
Female	22 (25)*		6417 (49)*
Age		6.3 (5.7-7.0)**		7.2 (7.1-7.3)**
Language
English	88 (99)		12 634 (96)
Spanish	1 (1)		268 (2)
Insurance
Private	56 (63)		8702 (66)
Government sponsored	33 (37)		4088 (31)

Chi-square tests for language and insurance types non-significant (P > .05).

Abbreviations: DBAC, Development and Behavior Access Clinic; CI, confidence interval.

^*χ² (1, N = 13 275) = 20.295, P < .001.

^**P < .05.

Nearly half (45%) of all DBAC visits occurred by telehealth, and the great majority of telehealth visits took place in the last 6 months of the study. Of 451 available DBAC visit slots, 366 (81.2%) were filled. Many patients required medication management, which in turn necessitated frequent follow-up visits for monitoring, while others required follow up for complex care coordination requiring more time than available in standard general pediatric visits. This need for frequent follow-up may have limited the number of new patients who could be seen. No-show rates for DBAC (14% of scheduled appointments) were comparable with those for the general pediatric clinic in which it was embedded (11%) and of the subspecialty DBP clinics (12%, June 2020 through September 2020).

The advent of widespread telehealth adoption in the early months of the COVID-19 pandemic was associated with a rise in visit volume, especially for video visits. On average, there were 13.9 completed DBAC visits per month prior to the widespread adoption of telehealth in March 2020, and 20.4 completed DBAC visits per month afterward, a statistically significant rise in monthly visit volume (P < .05). Of note, this increase was not associated with any change in clinical time. Following March 15, 2020, when shelter-in-place (SIP) orders began in the local area, there was a striking rise in the proportion of video visits from 13% to 70% of all completed visits, a statistically significant difference (P < .001). No-show rates dropped from 18% pre-SIP to 12% post-SIP across all visit types (both telehealth and in-person), but this drop was not statistically significant. No-show rates for video visits were similar to those for in-person visits both before and after SIP. See Table 2 for a detailed comparison of visit type, volume, no-show rates, and wait times before and after COVID-19-related SIP orders began.

Table 2.

Visit Volume and Average Wait Times (Pre- and Post-COVID).

Time period	Type of visit		Completed visits per month (95% CI)	No-show rate (95% CI)			Average wait time for new patient appointments (95% CI)
	In-person	Telehealth		Telehealth	In-person	Total
Pre-COVID (1 June 2019 to 14 March 2020)	87.4%*	12.6%*	13.9** (11.8-15.9)	17.6% (0.0%-35.8%)	17.9% (11.5%-24.2%)	17.8% (11.9%-23.8%)	41 days (30.3-51.7)
Post-COVID (15 March 2020 to 30 September 2020)	30%*	70%*	20.4** (16.2-24.7)	12.4% (6.7%-18.1%)	10% (3.4%-16.6%)	11.5% (7.2%-15.8%)	37 days (29.7-44.3)

Abbreviation: CI, confidence interval.

^*Pre-post-COVID difference significant at P < .05 level: χ² (1, N = 233) = 76.56, P < .001.

^**Pre-Post-COVID difference significant at P < .05 level.

Impact on Wait Times for DB Care

The opening of DBAC greatly reduced average wait times for DB pediatric services for eligible patients. At baseline, patients who had been referred to DB pediatrics prior to DBAC’s opening and who were later seen in DBAC waited an average of 218 calendar days to be seen. Patients assigned to DBAC after its opening waited a mean of 41 calendar days (95% confidence interval, 35-48 days). This represents a difference of 177 calendar days or approximately 6 months.

This run chart (Figure 1) shows the improvement in wait times when DBAC first opened followed by a slower decline until COVID-related clinic cancellations, followed by another decrease in average wait time when telehealth became widely available after the start of SIP. Overall, however, average wait times did not change significantly after the switch to telehealth: an average wait time of 37 (±7.29) business days after SIP began in March 2020, compared with 41 business days (±10.7) before SIP and the switch to telehealth. Of note, outside factors seem unlikely to have affected wait times: DBAC’s opening date did not correspond to any additional DBP or psychology hires, and overall referral volume for DBP services increased during the COVID pandemic, which would have been expected to lead to longer waits.

Figure 1.

Average wait time for DBAC appointment (calendar days).

Abbreviation: DBAC, Development and Behavior Access Clinic.

Impact on Need for Subspecialty Services

Of 204 total DBP referrals made from the primary care study site between May 1, 2019, and September 30, 2020, just over half (n = 104 or 51%) were deemed eligible for DBAC, initially reducing by half the number of children referred from this primary care clinic awaiting evaluation by DBPs. Of eligible patients, 89 (86% of DBAC-eligible patients or 44% of all DBP referrals) had at least one completed DBAC appointment. All 89 DBAC patient charts were reviewed for disposition following initial DBAC assessment. In total, 10 out of 89 (11%) children initially seen in DBAC subsequently required subspecialist care during the 18-month pilot period, generally by a DBP, psychiatrist, or psychologist for neurodevelopmental testing (including autism diagnostic testing) and/or clinical management. The proportion of patients requiring subspecialty care was lower than the program target of 25%. Altogether, 39% of the patients initially referred for DB subspecialty care from the primary care study site had their DB care needs met by a specially trained general pediatrician alone.

Referring Clinician Satisfaction

A total of n = 7 out of 15 referring physicians completed satisfaction surveys online over the course of 7 days at the end of the study period (response rate 47%). The median satisfaction score was 100 out of a possible 100, and the mean score was 95.7, indicating very high clinician satisfaction with DBAC services. The NPS was 86.

Discussion

During this 18-month pilot study, a general pediatrician with brief, intensive developmental-behavioral and mental health training was able to provide DB pediatric care to children with mild-to-moderate complexity needs, with proctoring by developmental-behavioral pediatricians (DBP). This new general pediatrician-staffed program (the DBAC) decreased the need for subspecialty care, providing DB services for approximately 39% of all patients referred for DBP care from this primary care pediatric site without the need for subsequent subspecialist intervention. Wait times for DB care dropped significantly for DBAC patients, from a baseline of 218 calendar days to 41 calendar days (P < .05). The opening of telehealth visits maintained similar low wait times while seeing a significant increase in DBAC’s monthly completed visit volume, suggesting that telehealth services may increase patient access even further. Satisfaction among referring clinicians was very high: their NPS (86) was in a range generally considered to be “excellent” and substantially higher than a recently quoted health care industry average NPS of 38, ²⁰ boding well for program growth.

Given the ongoing dearth of DB and child and adolescent psychiatric subspecialists and high levels of unmet need,^2,21 general pediatricians are increasingly asked to provide much of the developmental and mental health care currently being referred to subspecialists. ²² While much research documents unmet needs, fewer studies have examined models designed to address related gaps in clinical skill and care access. This pilot study suggests that, with appropriate training, subspecialty support, and adequate visit time, general pediatricians can provide timely care for many children referred for subspecialty DB services in a model that significantly decreased wait times for DB services.

Our pilot study also suggests telehealth as a promising means of expanding DBP care access. During the switch from office visits to predominantly telehealth clinics due to COVID-19-related SIP restrictions, average visit volume rose from 13.9 to 20.4 completed DBAC visits per month without change in clinical full-time equivalents (FTE). While these data are preliminary, they suggest that the shift to telehealth enabled DBAC to continue to serve increasing numbers of children even in pandemic conditions and that telehealth may make DB services even more accessible than visits located at the primary care site. Further research should continue to explore both telehealth and embedded primary care visits as complementary means to expand access to DB care.

This pilot study has multiple limitations rendering our conclusions preliminary. While parents also received a satisfaction survey at the same time as clinicians, the low response rate precluded the ability to draw meaningful conclusions. Future research should evaluate acceptability of this model to both caregivers and patients. Available QI data were limited and mostly descriptive in nature, with wait times as the predominant measure of access and no data on clinical outcomes. Additional research is also needed to assess the clinical outcomes associated with this model as well as potential benefits of co-location. The limited dataset also provided few measures of equity in care access other than payer type and language and provided little insight as to why the DBAC referral pool skewed so heavily male. (DBAC’s demographic may mirror diagnostic trends within DB pediatrics as a whole, as males are more likely to receive a neurodevelopmental diagnosis.)^1,23 Nevertheless, this pilot study demonstrates the viability of training and deploying a general pediatrician to provide clinical care often reserved for DB subspecialists, a model that deserves further development and study to help more general pediatricians provide similar care in their primary care clinics, meeting a critical workforce need.

Conclusion

This pilot study provides a model for building DB clinical skills among interested general pediatricians, decreasing wait times for patients, and building the capacity of pediatric primary care settings to address mild-to-moderate complexity/acuity DB clinical concerns within the child’s medical home. More research is needed regarding the clinical impact and outcomes associated with this approach.

Author Contributions

JJ: Led conception and design of study; oversaw data acquisition, analysis, and interpretation; drafted the manuscript; critically revised the manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy.

JT: Contributed to data acquisition, analysis, and interpretation; critically revised the manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy.

AS: Contributed to data interpretation; critically revised the manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy.

SPM-H: Contributed to conception and design of study; contributed to data interpretation; critically revised the manuscript; gave final approval; agrees to be accountable for all aspects of work ensuring integrity and accuracy.