Implementation and Evaluation of a Wraparound Virtual Care Program for Children with Medical Complexity

Abstract

Objectives:

Children and adolescents with medical complexity benefit from care coordination and specialized pediatric care, but many access barriers exist. We implemented a virtual wraparound model to support patients with medical complexity and their families and used an economic framework to measure outcomes.

Methods:

Children with medical complexity were identified and enrolled in a virtual complex care program with a dedicated multidisciplinary team, which provided care coordination, education, parental support, acute care triage, and virtual visits. A retrospective pre- and postanalysis of data obtained from the Hospital Industry Data Institute (HIDI) database measured inpatient, outpatient, and emergency department (ED) utilization and charges before implementation and during the 2-year program.

Results:

Eighty (n = 80) children were included in the economic evaluation, and 75 had sufficient data for analysis. Compared to the 12 months before enrollment, patients had a 35.3% reduction in hospitalizations (p = 0.0268), a 43.9% reduction in emergency visits (p = 0.0005), and a 16.9% reduction in overall charges (p = 0.1449). Parents expressed a high degree of satisfaction, with a 70% response rate and 90% satisfaction rate.

Conclusions:

We implemented a virtual care model to provide in-home support and care coordination for medically complex children and adolescents and used an economic framework to assess changes in utilization and cost. The program had high engagement rates and parent satisfaction, and a pre/postanalysis demonstrated statistically significant reduction in hospitalizations and ED visits for this high-cost population. Further economic evaluation is needed to determine sustainability of this model in a value-based payment system.

Introduction

Children with medical complexity (CMC) are a growing population of children with expensive, complex, and chronic medical needs that require a multitude of health care services. About 0.5–1% of all children in the United States account for a third of pediatric spending, and the population is growing significantly, partially because life-saving advances in medical technology have enabled the survival of many childhood-onset diseases.^1–7 Because of their disabilities, these patients are concentrated in Medicaid insurance plans, and they need dramatically more health care services than their peers. Pediatric medical complexity is a primary determinant of health care inequity.⁸ Of the 33 million children in Medicaid, the 2 million with medical complexity accrue 10 times the costs per year compared to other children, representing 10% of admissions to children's hospitals, 25% of hospital patient days, and 40–50% of hospital charges.^1,9,10 Their inpatient care is the highest proportion of their costs, accounting for as much as 80% of their total health care costs.¹

In the traditional, fee-for-service health care system, incentives are not aligned to keep these children healthy and out of the hospital, and without a team-based approach, care can be episodic and fragmented.³ Parents and caregivers face challenges managing their appointments, equipment, medications, supplies, and home nursing. They often suffer from anxiety and overwhelm stemming from their child's illness, and more than half of these families experience financial hardship or have a family member stop working to care for their child.^3,11 Confusion about care plans, which are exacerbated by health literacy factors, can lead to complications requiring hospitalization, and gaps in care from home health nursing shortages are a key driver for hospital admissions and readmissions.^12–15

Children's hospitals have successfully addressed some of these factors by developing specialized complex care clinics that provide care coordination and treatment.^16–19 However, geographic and other barriers still exist, and the burden of transporting technology-dependent children can be a profound challenge for families while also exposing these fragile patients to communicable diseases.²⁰ The lack of reimbursement for the level of support services and care coordination needed leads to challenges in creating scalable sustainable solutions for these children.^21,22

In this article, we describe implementation of a wraparound model of care for vulnerable, high-cost pediatric patients using virtual care. We used an economic framework²³ to design a retrospective cohort study to measure the preintervention and postintervention utilization for the inpatient, outpatient, and emergency department (ED) setting, cost of care, and patient satisfaction for CMC.

Methods

PROGRAM DESCRIPTION

The “vKids” program was a fully virtual program that provided wraparound care to address medical and social needs of patients and families using a virtual team of pediatric physicians, pediatric nurse practitioners, pediatric nurses, and social workers. The program was implemented at Mercy Virtual in Chesterfield, MO and served patients across five midwestern states from June 1, 2018 to May 15, 2020. It was not a virtual-only model—it was designed to complement established in-person care resources at Mercy. Patients who received care were 0–19 years and had chronic conditions affecting at least three body systems, as well as at least three ED visits or two inpatient stays in the 12 months before enrollment, and had previously received care at Mercy Health system. Children with primary clinical issues of chronic pain or oncology were not included in the intervention program. Mercy patients who met enrollment criteria were identified using an Electronic Medical Record search or were referred to the program by their primary care pediatrician or specialist.

SOFTWARE

Patients were provided with access to an app-based portal (Life Sciences Technology). The virtual platform gave parents access to personalized educational resources and real-time communication through text, audio, or video during program hours. Parents received a daily notification from the app asking if the parent had any new concerns about their child.

HARDWARE

Families who did not have access to a smart device were given an iPad with the app installed, and those who were able to use their own device downloaded the app. All families received equipment to take vital signs, including a thermometer, a pulse oximeter, and a blood pressure cuff. No advanced peripheral devices (such as stethoscopes, otoscopes, or specialized examination cameras) were provided.

PROGRAM COMPONENTS AND ENCOUNTER TYPES

Enrollment visit (60-min appointment with pediatric nurse and physician)

Upon enrollment, a comprehensive evaluation of all medical and social needs was performed by a physician or provider, including specialists involved in their care, medication reconciliation, and contingency planning. The clinical team assessed their home supplies and helped the family compile a rescue kit with medications, supplies, and equipment needed for their action plans. A social assessment was completed, and a social worker was assigned to families who had needs identified. The comprehensive care plan was shared with the child's primary pediatrician and all specialists involved in the care.

Technology installation visit (30-min appointment with navigator or primary pediatric nurse)

The caregiver had a visit with a navigator for training about the technology and program resources. During this visit, they tested connectivity and ensured that they had access to their personalized app on all devices desired with notifications turned on.

Daily digital touchpoint (automated)

Parents received an app notification daily asking: “Do you have any new concerns or needs for your child?” Parents were made aware of this notification during the enrollment appointment, and these “nudges” reminded them to check in with the care team, prompting early conversation if there were concerns. A positive response to the digital touchpoint notified the primary nurse, who then followed up with the family and escalated to the appropriate team member using predetermined protocols.

Proactive preventative visits (15-min appointment with primary pediatric nurse)

Regularly scheduled preventative visits occurred through video, phone, and messaging with families. During onboarding, the visits were scheduled weekly to provide opportunity for relationship building, care plan assessment, answering parental questions, and assessing the child's baseline state of health. Once a family demonstrated consistent program compliance and no active issues, the patient would have biweekly or monthly preventative visits. When issues would arise, patients would temporarily be moved back to a weekly schedule.

Acute care visits (15 or 30-min appointment with pediatric nurse or pediatric nurse practitioner)

Families could contact the clinical team about acute issues through message or phone. Predefined protocols directed the clinical team to escalate the visit to the most appropriate care team member (nurse, nurse practitioner, or physician through phone or video) or to in-person resources (diagnostic testing, clinic office setting, ED, or direct admission to inpatient setting) as needed.

Ad hoc care coordination and social services

The team engaged with patients, families, and caregivers and supported the connection to services outside the health system such as social services, school systems, financial resources, legal resources, and other community-based services. Parents could also receive referrals to mental wellness resources for themselves or a chaplain for spiritual support.

DATA EXTRACTION AND ANALYSIS

This observational study utilized a pre-post design with clinical program data combined with health care billing records to assess the economic impact on health care utilization. The economic evaluation was limited to patients in one state. The economic evaluation included patients in Missouri between the ages of 2 and 19. We compared pre-enrollment (beginning in October 2017 with an average of 1.4 years of predata) to postenrollment (after enrollment until March 31, 2020, with a minimum of 3 months and up to 24 months) utilization data. These data were sourced from the Missouri ED and Hospital UB04 records through the Hospital Industry Data Institute (HIDI), which is the custodian of the Missouri ED and Hospital Discharge data sets, to ensure that all records for care provided from any Missouri provider were included.

Clinical reports were generated from the electronic medical record (Epic) and the third-party software (Life Sciences Technology). The HIDI data reports were used to extract subject hospitalization and ED visit information, cost information, and covariate information for the pre-enrollment and postenrollment time periods. A summary score of organ involvement was generated based on the codes available in the HIDI data. The Epic data were used to assess under-ascertainment bias in the HIDI data.

The Institutional Review Board at Mercy Research reviewed and approved the study protocol. A Waiver of Consent/Assent and Health Insurance Portability and Accountability Act Authorization were approved for this study.

Data for study participants were extracted from the HIDI data set for all inpatient and outpatient visits across all hospitals in the state between October 1, 2017, and March 31, 2020.

Admission dates were not included in the data and were calculated based on the provided discharge date and recorded length of stay (LOS) for the visit, defined as Discharge date − LOS +1. The “Pre” period was defined as the time before enrollment date, and the “Post” period defined as the time of the enrollment date or later. Given that the data were only provided through the end of March 2020, any patients with enrollment in the program after January 1, 2020, were excluded from the analysis as they did not provide sufficient Post data (n = 4 patients). One study participant with no data returned for the inpatient or outpatient data during the 2.5 years was excluded on the assumption that data linkage was not successful. Study participants with either an inpatient or outpatient visit record were assumed to be successfully linked; thus, the lack of other types of visits/charges was set at 0.

To determine the time in the pre- (or post-) period, the number of days from October 1, 2017, to enrollment date (or enrollment date to March 31, 2020) was divided by 365 days (assumed per year). To assess annualized number of visits and total charges, the sum of the number of visit (or total charges) in the pre (or post) period was divided by the time in years of the pre (or post) period. The annualized number of days in hospital was measured as the sum of all days in the hospital across all visits in the pre (or post) period, divided by the time in years in the pre (or post) period.

Due to positively skewed data, annualized days in hospital, number of inpatient visits, outpatient visits, and ED visit differences were assessed assuming a negative binomial distribution. Total annualized charges were also positively skewed and assessed using a Gamma distribution with a log link. All analyses were conducted using SAS 9.4 (Cary, NC).

Results

The study cohort included 80 patients aged 2–19 years in Missouri. Figure 1 describes the 75 patients who had sufficient data to be included in the final pre- and postanalysis. Table 1 describes the demographics of the population analyzed.

Fig. 1.

Patients analyzed in pre/postanalysis.

Table 1.

Demographics (N = 75)

PATIENT CHARACTERISTIC	N (%)
Age at enrollment, mean (SD)	6.6 (4.5)
Female	31 (41.3)
Race
American Indian/Alaska Native	1 (1.3)
Asian	1 (1.3)
Black or African American	7 (9.3)
Caucasian	60 (80.0)
Hispanic	3 (4.0)
Multiracial	3 (4.0)
Months enrolled, mean (SD)	15.5 (4.9)

SD, standard deviation.

Pre- and postanalysis of hospitalizations, ED visits, and cost of care are displayed in Table 2. Compared to the 12 months before enrollment, patients in the program had a 35.3% reduction in hospitalizations (p = 0.0268), a 43.9% reduction in emergency visits (p = 0.0005), and a 16.9% reduction in overall charges (p = 0.1449).

Table 2.

Annualized Resource Uses and Charges for Pre-Enrollment Compared to Postenrollment for Children (N = 75)

ANNUALIZED (PATIENT/YEAR)	PRE-	POST-	DIFFERENCE FROM PRE-ENROLLMENT	p-VALUE UNDER APPROPRIATE DISTRIBUTION
	MEAN (SD), N = 75	MEAN (SD), N = 75	Δ MEAN, %
Years	1.4 (0.4)	1.1 (0.4)
Encounters
Days in hospital	7.7 (10.2)	5.6 (12.5)	−27.3	NS
Hospital admissions	1.7 (1.8)	1.1 (1.6)	−35.3	0.0268
ED visits	3.2 (2.6)	1.8 (2.2)	−43.8	0.0005
Outpatient visits	8.9 (8.8)	8.7 (9.8)	−2.2	NS
Charges ($USD)
Hospital charges	$63,826 (93,587)	$45,954 (89,250)	−28	0.3606
ED charges	$9,208 (9,385)	$6,303 (8,392)	−31.5	0.2235
Outpatient charges	$25,604 (32,216)	$29,727 (44,561)	16.1	0.5166
Overall chargesa	$98,639 (105,250)	$81,985 (102,314)	−16.9	0.1449

^a Controlling for age.

ED, emergency department.

ENGAGEMENT WITH PROGRAM RESOURCES

Families in the intervention program had an average of 13.4 touchpoints with the virtual clinical team per month (10 texts, 1.4 phone, 2 video). Compliance rate with scheduled visits was 74%. Parents responded to the daily digital touchpoint 29% of the time. Patient satisfaction surveys were distributed at the 60-, 120-, and 180-day marks and had a 70% response rate. Results of completed surveys are illustrated in Figure 2. A majority (90%) of survey questions reported that they agreed or strongly agreed with the satisfaction measures. Every negative response was followed up with a call from the program manager, and every patient who had said they “strongly disagreed” reported that they had unintentionally selected the negative response and verbally reported that they strongly agreed.

Fig. 2.

Aggregate patient satisfaction results. Questions about the experience with vKids included: 1. The team listened to me when we talked about my child's health. 2. The team respected my ideas and concerns about my child's care. 3. The team explained things in a way that I could understand. 4. The team worked with each other and with my child's other providers to improve the care my child received. 5. I felt satisfied with the care my child received from the team.

Discussion

The vKids program implemented a virtual wraparound model of proactive, comprehensive, technology-enabled care for CMC. Parents expressed a high degree of satisfaction with digital and virtual access to a centralized team for care coordination, education, triage, and acute care video visits. An economic framework was utilized to measure pre- and postutilization and costs of care.

Parents' assessment of their child's state of health is a predictor of health decline and impending hospitalization.²⁴ Parents in this program frequently utilized texting to communicate with the clinical team, allowing the team additional insight to identify children and family's needs for education or interventions. Patients in the vKids cohort had an average of 160 touch points per patient per year, which allowed for more day-to-day interaction about patient needs. The digital touchpoints were meant to be behavioral “nudges” for parents to do a mental check-in with their child's health, and parents were more likely to respond to the alert if they had a concern and not respond if they had no concerns. The virtual program operated during weekday hours and had a traditional phone-only exchange for after-hour issues.

The Coronavirus disease 2019 (COVID-19) pandemic accelerated the adoption of virtual care across the medical community.²³ Virtual care is rapidly evolving and provides new tools, processes, and methods for connecting and engaging with patients in their own homes and communities.^25–27 Prepandemic, physicians and providers had concerns about lack of established quality metrics for evaluating patients virtually, along with liability and other concerns.^26,28 The American Academy of Pediatrics' policy statement on telehealth states that telehealth interventions should be held to the same quality standards required for in-person care, and this was the operating philosophy for this program.²⁹ If a virtual visit led the clinician to the conclusion that an additional in-person examination or diagnostic testing would be needed, patients were escalated to in-person care, preferably with one of their own outpatient providers unless an emergent need was identified. No safety events, near misses, or care delays were reported during the operations of the program.

In a fee-for-service system, financial sustainability is based upon volume of relative value-based units. Although care management and complex care codes exist, the lack of payment mechanism for comprehensive value-based care through a resource-intensive model has prevented development of this model of care in the past.²⁹ Contracting with payors for remote patient monitoring codes (which are now reimbursed by Centers for Medicare & Medicaid Services but are not widely available for use in pediatrics)²⁷ could be one option to support the virtual model discussed in this article. Alternatively, value-based models of wraparound in-home care for complex adult patients have demonstrated success in Medicare populations,³⁰ but not in pediatrics, although capitation has been suggested as a possibility for this population.¹ In a capitated model of care, the multistakeholder value of wraparound, digital first care may exceed operational costs at scale, which could allow for future opportunities to grow similar models in the future to focus on keeping complex children out of the hospital setting.

LIMITATIONS

Our pre-/postevaluation did not include a control group, and further analysis would be needed to assess for the possibility of confounding factors or regression to the mean,³¹ which has previously been reported to compound pre- and poststudies involving CMC. The study was not powered to demonstrate statistical significance in total cost of care charges, and a much larger sample size would have been needed to find a minimally important difference in cost of care.³² This study did not control for possible confounding based on selection bias of the referral-based population. We did not bill for the wraparound services during the pilot program, so analysis of the cost of the total resources required compared to the cost savings is required to fully assess the sustainability of such an intervention in a value-based payment model.

Conclusions

We implemented a program to provide comprehensive wraparound virtual care for CMC and used an economic framework to measure pre- and postfinancial outcomes. Parents and families expressed a high level of program satisfaction, and the pre/postanalysis showed a reduction in hospitalizations and ED visits. Additional control group analysis and robust economic evaluation of the intervention are needed to determine sustainability, but initial results demonstrate an opportunity for value-based care for children with high medical costs and needs.

Authors' Contributions

Dr. Curfman conceptualized and designed the study, drafted the initial article, critically reviewed the article, and approved the final article as submitted. M.H. conceptualized and designed the study and critically reviewed the article and approved the final article as submitted. Dr. McSwain drafted portions of the initial article, critically reviewed the article, and approved the final article as submitted. Dr. Dooley and Dr. Simpson designed the data collection instruments, performed data curation and analysis, critically reviewed the article, and approved the final article as submitted.

All authors approved the final article as submitted and agree to be accountable for all aspects of the work.

Disclosure Statement

Dr. Curfman and Meghan Haycraft have an equity stake in Imagine Pediatrics. Dr. McSwain, Dr. Simpson, and Dr. Dooley have no disclosures.

Funding Information

This publication was supported, in part, by NIH/NCATS SPROUT-CTSA Collaborative Telehealth Network grant number U01TR002626. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The NIH had no role in the design and conduct of the study.