Impact of a Complex Care Management Model on Cost and Utilization Among Adolescents and Young Adults with Special Care and Health Needs

Abstract

Adolescents and young adults with special care and health needs in the United States–many of whom have Medicaid coverage–at the transition phase between pediatric and adult care often experience critical care gaps. To address this challenge, a new model–referred to as Comprehensive Care Clinic (CCC)–has been developed and implemented by Geisinger Health System since 2012. CCC comprises a care team, consisting of a generalist physician, advanced practitioner, pharmacist, and a nurse case manager, that develops and closely follows a coordinated care plan. This study examines the CCC impact on total cost of care and utilization by analyzing Geisinger Health Plan claims data obtained from 83 Medicaid patients enrolled in CCC. A set of multivariate regression models with patient fixed effects was estimated to obtain adjusted differences in cost and acute care utilization between the months in which the patients were enrolled and the months not enrolled in CCC. The results indicate that CCC enrollment was associated with a 28% reduction in per-member-per-month total cost ($3931 observed vs. $5451 expected; P = 0.028), driven by reductions in hospitalization and emergency department visits. This finding suggests a clinical redesign focused on adolescent and young adults with complex care needs can potentially reduce total cost and acute care utilization among such patients.

Introduction

Adolescents and young adults as well as children who are medically fragile have intensive care needs that are not met by existing health care models.¹ With improvements in medical treatments, young adults with childhood-onset chronic conditions (eg, cystic fibrosis, spina bifida, congenital heart disease, cerebral palsy, sickle cell disease) have greatly increased life expectancy with the vast majority surviving into adulthood.² Although all youth must transition from pediatric to adult-centered medical care, this process is especially difficult for the 500,000 to 750,000 children with special health care needs in the United States that transition annually to adult care.^3
–5 Adolescents and young adults with special care and health needs (AYASCHN) who are at this transition phase between pediatric and adult care are at considerable risk during this vulnerable time,^6,7 and face costly and dangerous gaps in care⁸ that can lead to potentially serious consequences including worsening health outcomes.^9,10

Substandard care for the AYASCHN population is characterized by the following gaps: fragmented and uncoordinated care, planning, and decision making; inadequate communication and poor health information management across providers; crisis-driven management with a tendency toward overmedicalization; inadequate support of family caregivers,¹¹ patients' psychosocial needs,^12,13 and clinician education and training.^13
–15 Traditional primary care practices lack resources to support care management for relatively small numbers of patients with complex health care needs,¹⁶ creating an opportunity for new sustainable care models to meet the needs of transitioning AYASCHN based on a patient-centered medical home (PCMH) model.^17
–19

This study examines a real-world complex care management model that focuses on AYASCHN, referred to as Comprehensive Care Clinic (CCC) that was developed and implemented by Geisinger Health System (Geisinger), by assessing CCC's impact on cost and care utilization. This study is focused on medically complex AYASCHN covered by Medicaid, as they account for the majority of the AYASCHN patient population nationwide and also represent a more vulnerable subpopulation because of their socioeconomic conditions.

Background

Complex care management programs have similarities related to performing 4 essential activities^20,21: (1) identifying and engaging complex patients at high risk for poor outcomes and unnecessary utilization; (2) performing comprehensive health assessments to identify problems that, if addressed through effective intervention, will improve care and reduce the need for expensive services; (3) working closely with patients and caregivers as well as other health care and social service providers; and (4) rapidly and effectively responding to changes in patients' conditions to avoid use of unnecessary services. In addition to specially trained teams tailored to the target population, complex care management strategies^16,21
–23 may include the following features: care plans, dedicated provider/case manager, targeted patient-safety initiatives (eg, enhanced medication reconciliation), patient/family self-management education for specific procedures and complications, same-day emergent care access, and health information technology to facilitate care management, coordination and communication (eg, access to real-time data, care plan documentation, decision support for exacerbations).

Geisinger's CCC operationalizes these complex care management features via a dedicated multidisciplinary care team that includes an internal medicine/pediatrics physician, an advanced practitioner, a registered nurse case manager, and a pharmacist. This team develops care plans to coordinate interactions among the team members as well as with the patient and family caregiver and other health care providers located outside the clinic. This care plan is embedded within each patient's electronic health record and maintained by the team members with input from the patient, family, and all medical care providers. Conceptually, Geisinger's complex care management model augments its PCMH model by incorporating CCC, which has been established within an outpatient general internal medicine clinic in Geisinger's tertiary care hospital in central Pennsylvania. Geisinger is the largest integrated health care delivery system in central and northeastern Pennsylvania and has implemented CCC in collaboration with Geisinger Health Plan (GHP), a full-service regional plan covering approximately 500,000 lives. GHP has been a Pennsylvania Medicaid managed care plan since March 2013; in its third year of operation, GHP currently serves more than 177,000 Medicaid members.

To identify the AYASCHN patient population eligible for care in CCC, the following criteria are used: reside within Geisinger's service area to ensure reasonable access; be older than the age of 15; typically use 20 or more medications; rely heavily on specialty care (3+ specialties, 3+ times in the past 2 years; 2+ genetic medicine visits); be technology-dependent (eg, ventilator, gastronomy tube, wheelchair bound); have certain diagnoses (eg, cystic fibrosis, spina bifida, cerebral palsy, congenital heart defect, transplant, chromosomal or congenital abnormalities); have a history of frequent emergency department (ED) visits and inpatient admissions; or be referred to complex care management via their specialists, primary care physicians, or self-identified by the patient or caregiver. All referrals are reviewed by CCC staff to ensure appropriateness of the referral.

The CCC's comprehensive assessment strategy consists of an extended primary care visit with the physician (approximately 60 minutes) scheduled at regular intervals (every 3 to 6 months, depending on each patient's complexity and need). Medication reconciliation is performed by the team pharmacist as well as the nurse case manager to identify possible adverse effects, medication interactions, required laboratory and other screening and monitoring tests, and to assess the effectiveness of the current medication regimen. Comprehensive case management is provided by a nurse case manager who is embedded in the internal medicine clinic and sees the patient and family caregiver together with the physician to help develop the care plan. Training and education of the patient and family caregivers for self-management are conducted that also enable them to recognize signs and symptoms of acute exacerbations indicative of a worsening condition and to follow instructions on how to respond. In addition, the CCC team also dedicates substantial time outside of each visit to coordinate, plan, follow, and document all relevant patient care and social services. Same-day access to CCC for emergent care needs is also available. Table 1 lists and describes the key components of Geisinger's complex care management model as embodied by CCC.

Table 1.

Complex Care Management Model Components for Adolescents and Young Adults with Special Care and Health Needs

Components	Description
Multidisciplinary care team	Primary care physician (internal medicine/pediatrics), advanced practitioner, registered nurse case manager, pharmacist
Comprehensive assessment	Extended regular visits (∼1 hour) and post-visit care coordination, follow-up, and documentation by care team (∼3–4 hours)
Medication reconciliation	Review for adverse effects, interactions, required laboratory tests, and regimen effectiveness
Case management	Comprehensive complex condition management to proactively identify and coordinate medical and social services to optimize patient health and prevent complications and exacerbations
Care plan	Maintained by the care team with input from patient, family, and medical care providers and embedded in electronic health records. Sections include:
	1. Outline of baseline characteristics of problems
	2. System-based early identification of exacerbations and management plans
	3. Care coordination inclusive of other health care providers and social services
	4. Routine self-management goals that are developmentally appropriate
	5. Transition readiness assessment and plan
Self-management education/training	Establish a common understanding of condition-related needs, taking into consideration patient/caregiver health literacy and capabilities; collaboratively develop a care plan to interact with other health care providers and to address care needs and potential exacerbations at home
Emergent care	Rapid response to emergent needs including same-day access to the care team
Caregiver support	Actively support caregivers by establishing a common understanding of their role and capabilities, provide access to information and care team using electronic patient portal and clinic, activation of county support services/waiver programs, guardianship, and respite care.
Information technology/Electronic health record capabilities	Implement online health management tool (“MyGeisinger”) for patients and caregivers to facilitate provider messaging, appointment scheduling, and access to electronic health record information and the embedded care plan (described above) to enhance interaction with care team and other providers.

Data

This study was conducted as a part of Geisinger Health System's quality improvement initiative in partnership with GHP for case management and analytic support. To capture the cost and utilization of care by each patient, health plan claims data from GHP were obtained, covering a 19-month period from March 1, 2013, through September 31, 2014. Although CCC was first implemented in July 2012, GHP had not begun to enroll Medicaid members until March of 2013. Consequently, only claims data since March of 2013 were available for this study. At the time this study was conducted, GHP claims data from September of 2014 were the most recent available. Claims data, rather than electronic health records, were used for the purposes of this study because claims data, in general, represent a more comprehensive view of each patient's care utilization and costs incurred. For instance, electronic health records from a single health care provider would miss ED visits or hospitalization in other non-Geisinger facilities. One potential drawback of using claims data is that because of coverage limitations, the claims data represented only the services covered by GHP Medicaid.

The dependent variables were total cost of care and utilization of acute care, defined as inpatient (IP) admissions and ED visits. Cost of care was defined as “allowed” amounts (ie, health plan payments to providers plus patient out-of-pocket costs) on a per-member-per-month (PMPM) basis. The unit of observation in this analysis was therefore member-month rather than each member. Because Medicaid requires little to no patient out-of-pocket cost sharing, most of the allowed amounts in the data represented GHP payments to the provider. The PMPM total cost of care also was broken down into 4 main cost components: IP, outpatient, professional, and prescription drug costs.

Methods

A set of multivariate regression models was estimated to obtain the CCC enrollment impact on the dependent variables. For total cost of care and each of the 4 cost components, a generalized linear model with log link and gamma distribution was used. For ED visits and acute IP admissions, a Poisson model was used to account for the fact that these dependent variables were count data. The key explanatory variable was an indicator variable for whether the member was enrolled in CCC in a given month of the study period. As explained in the following section, there was variation in the CCC enrollment status across the 83 patients as well as over the study period. The study team exploited this variation to estimate the CCC enrollment effect on the dependent variables.

There are 2 main potential sources of confounding: First, there may have been a selection bias in which the patients who enrolled in CCC might have been systematically less or more severely ill than those not enrolled in CCC and therefore incurred less cost and care utilization. This potential bias is mitigated by inclusion of patient fixed effects in the regression model. The patient fixed effects (ie, inclusion of a binary indicator variable for each of the 83 patients in the sample as covariates in the regression model) control for any time-invariant patient characteristics (eg, sex, race, comorbidity) that might confound the CCC enrollment effect. Furthermore, the study team included MEDai (MEDai, Inc., Orlando, FL) risk scores²⁴–a propriety patient risk profiling tool developed by a third-party vendor and used by GHP for its population health management efforts–as a covariate to further capture differences in the patient risk profiles. Higher MEDai risk scores indicate patients with worse health conditions at each month of observation.

The second source of confounding is the provision in Pennsylvania's Medicaid in which medically complex pediatric patients transition out of GHP coverage and into the state's waiver program after the patients graduate from schools. Specifically, under the waiver program, home care services are covered and paid for by the state rather than by GHP (aka “shift care”). The waiver program thus impacts the total cost of care as captured by the GHP claims data for certain patients in the sample, but not the rates of ED or acute IP admission because GHP continues to cover and pay for other services not subject to the waiver program. As the waiver program eligibility is largely dependent on the patient's age, age categories (<18, 18–21, and >21) are created and included as covariates in the regression models.

Other covariates in the model included indicator variables for calendar month and year to adjust for seasonality and yearly secular trends, as well as indicator variables for whether the patient was enrolled in GHP's case management program (operated independently of CCC), and whether the patient had primary care providers (PCPs) who were employed by Geisinger Health System at each given month of the study period. Furthermore, interaction effects between the case management enrollment status and the Geisinger PCP indicator variables as well as between the year indicator variable and the MEDai risk score variables also were included in the regression models. The first interaction effect was intended to reflect the fact that the patients who are enrolled in the GHP case management program and receive care from Geisinger-employed PCPs may experience care patterns that are different from those who either are not enrolled in GHP case management or have non-Geisinger PCPs, independently of CCC. The second interaction effect was intended to reflect the possibility that MEDai scores may be calculated differently in each year of the study. To account for the repeated observations in the data (ie, a patient can be observed as many as 19 times in the data), clustered standard errors were obtained and used to determine the statistical significance of the estimates.

Results

A review of the list of patients enrolled in CCC as of April of 2015 yielded 83 unique patients who also were members of the GHP Medicaid plan. As shown in Table 2, the most common primary diagnoses among these patients were autism spectrum disorder and spina bifida, and 38.6% of them were technology-dependent. For the complete list of primary diagnosis for all 83 patients in the sample, refer to Supplementary Table S1 (Supplementary Data are available in the online issue at www.liebertpub.com/pop). As shown in Table 3, there was variation in the lengths of overall observation and CCC exposure across the 83 patients as observed from their claims data. The majority had been members of GHP Medicaid plan for the entire duration of the 19-month study period. An additional 18 patients had been GHP Medicaid members for 18 months. The average length of observation across the 83 patients was 17 months. The 47 patients who transitioned into CCC at some point during the study period (March 2013-September 2014) were, on average, enrolled in CCC for approximately 8 months out of the 19-month period (Table 3).

Table 2.

Sampled Comprehensive Care Clinic Patient Profile

Patient Characteristics	n (%)
Sex
Male	44 (53.0)
Female	39 (47.0)
Race/ethnicity
African American	7 (8.4)
White	72 (86.8)
Hispanic	4 (4.8)
Primary diagnosis
Autism spectrum	14 (16.9)
Spina bifida	9 (10.8)
Chromosome abnormality	5 (6)
Profound intellectual disabilities	5 (6)
Down syndrome	4 (4.8)
Congenital heart syndrome	2 (2.4)
Cystic fibrosis	2 (2.4)
Neurodegenerative disease	2 (2.4)
Sacral agenesis	2 (2.4)
Other	38 (45.8)
Technology dependent (n = 32^a; 38.6%)
Gastro-jejunal tube	21 (25.3)
Wheelchair	24 (28.9)
Oxygen	5 (6.0)

Some patients are dependent on multiple technologies.

Table 3.

Length of Observation and Comprehensive Care Clinic Enrollment Status During the Study Period

	Frequency	Percent
Length of observation (months)^a
2	3	3.61
3	1	1.2
6	1	1.2
9	1	1.2
11	1	1.2
12	1	1.2
13	1	1.2
14	3	3.61
15	1	1.2
17	2	2.41
18	18	21.69
19	50	60.24
Total	83	100
CCC status during the study period
Not yet in CCC	19	22.89
Already in CCC	17	20.48
Transitioned to CCC	47	56.63
Total	83	100

The study period was March 2013 through September 2014.

Mean length of observation was 17 months.

CCC, comprehensive care clinic.

Table 4 indicates that CCC enrollment (represented by 658 member-month observations) had coincided with member-month observations that were characterized by patients who were older, more likely to be enrolled in GHP's case management program, and had higher risk scores relative to the member-months when the patients had not yet enrolled in CCC (represented by 767 member-month observations). Consequently, CCC enrollment appears to be associated with higher cost and care utilization, although the differences are not statistically significant. Table 4 therefore suggests that inclusion of the covariates in the regression models is justified.

Table 4.

Unadjusted Bivariate Comparisons

	Not CCC enrolled (n = 658)	CCC enrolled (n = 767)	P^a
Cost of care ($ per member per month)
Total	3320	3752	0.588
Inpatient acute	1022	721	0.664
Outpatient	267	364	0.333
Professional	1537	1853	0.525
Prescription drug	477	795	0.130
Utilization (count per member per month)
Inpatient acute admission	0.037	0.044	0.738
Emergency department visit	0.064	0.066	0.926
Age category (%)
<18 years	29.5	8.3	0.002
18–21 years	47.0	50.2	0.709
>21 years	23.6	41.5	0.024
Female (%)	44.1	51.4	0.412
% Under case management	17.0	70.7	<0.001
% Geisinger primary care physician	62.0	78.1	0.047
Mean MEDai risk scores	2.57	3.68	0.059

n represents the number of member-month observations rather than unique patients.

Clustered standard errors were used to reflect repeated observations for each patient.

CCC, comprehensive care clinic.

Tables 5 and 6 summarize the CCC enrollment impact on utilization and cost, as obtained from the estimated regression models (for the full regression results, refer to Supplementary Table S2). Table 5 indicates that during the study period, CCC enrollment was associated with approximately a 78% reduction in acute hospital admissions (P = 0.053) and approximately 60.3% in ED visits (P = 0.017) per member per month. Table 6 suggests that CCC enrollment was associated with an approximately 28% reduction in PMPM total cost of care ($3931 observed vs. $5451 expected; P = 0.028). The most significant source of the total cost reduction appears to be inpatient cost (P = 0.028).

Table 5.

Regression-Adjusted Comprehensive Care Clinic Impact on Acute Care Utilization (per Member per Month)

	No. of acute admissions	No. of ED visit
Observed	0.043	0.066
Expected	0.192	0.168
Difference	−0.149	−0.101
% Difference	−77.6%	−60.3%
P	0.053	0.017

ED, emergency department.

Table 6.

Regression-Adjusted Comprehensive Care Clinic Impact on Cost of Care ($ Allowed per Member per Month)

	Total	IP acute	OP	Pro	Rx
Observed	3931	975	379	1938	836
Expected	5451	2612	333	2273	672
Difference	−1520	−1638	46	−335	164
% Difference	−27.9	−62.7	13.9	−14.7	24.3
P	0.028	0.028	0.626	0.455	0.206

IP, inpatient; OP, outpatient; Pro, professional; Rx, prescription drug.

Discussion

These findings address an important gap in the literature. To date, there is little evidence of the effectiveness and no studies assessing cost outcomes of primary care-integrated models caring for a medically complex transition-age population.⁹ Many of the previous intervention studies have focused either on a population with a specific disease (eg, diabetes, asthma, epilepsy, spina bifida, sickle cell anemia),^25,26 or children who are mostly not of transition age and overall have lesser functional limitations than a complex population. Hence, the inclusion of diverse diagnostic groups in this study that are often small individually and of greater overall severity for an AYASCHN population is informative.

As shown in Table 2, the AYASCHN patients included in this study are characterized by severe conditions that generally lead to worsening health outcomes over time. In fact, many of the eligible AYASCHN patients are transitioned to CCC at end-stage disease, including muscular dystrophies, cystic fibrosis, and congenital heart disease (see Supplementary Table S1 for the complete list of primary diagnoses), requiring prolonged hospitalizations and change in therapies including chronic ventilation or initiation of hospice services. Therefore, it is expected that a simple comparison over time between the periods before and after CCC enrollment would show a pattern suggesting that CCC enrollment seems to be associated with higher cost and utilization, as shown by Table 4. Yet, the reported results obtained after controlling for the confounders as shown in Tables 5 and 6 indicate the opposite, implying that CCC is likely to have alleviated the increasing disease burden over time. This also implies that the reported results may be subject to a downward bias that underestimated the true CCC impact.

Geisinger's CCC experience may serve as an initial example demonstrating that this type of program can have a measurable impact on a health care resource-intensive AYASHCN Medicaid population. Although the magnitude of the estimated impacts seems large, it may simply reflect the extent of the problems and challenges inherent in caring for this population. At the same time, these results also imply that the significant challenges of managing and coordinating care for a Medicaid AYASCHN population can effectively be translated into patient care improvements and medical care cost reductions that can be realized rather than just hypothesized.²⁷

As noted earlier, CCC is a real-world complex care management model offering AYASCHN intensive primary care through an augmented PCMH. Although based on a generic PCMH, CCC goes beyond it by incorporating the complex care management model (as described in Table 1), which is not part of a generic PCMH, and by further developing its components for the AYASCHN population and family caregivers. Indeed, PCMH is the foundational element in providing transition services for all youth.²⁸ However, AYASCHN require an expanded process to address more complex health information and other issues including insurance, guardianship, and decision-making support. The results of this study therefore suggest that there are complex care management model features and characteristics that are critical in redesigning PCMH to further meet the needs of an AYASCHN population

This study is subject to several limitations. First, this was a retrospective observational study based on administratively collected data that are specific to a single integrated health care system, focused on a relatively small cohort of patients. Particularly, because CCC was limited to a single location with one team of dedicated health care professionals, the generalizability of the findings is unclear. Nevertheless, the CCC principle of an internal medicine-pediatrics trained physician combined with a nurse case manager, an advanced practitioner, and a clinical pharmacist with extended visit times is designed to be generalizable and scalable. Second, this study did not estimate the costs of implementing the CCC intervention (ie, resource costs for staff time, coordination and management services, electronic health record functionality, additional coordination infrastructure versus usual care), so a return on investment assessment remains an area of future research. Third, clinical and other patient- and family-centered outcomes (eg, quality of life, self-management, satisfaction, successful transition) were not specifically measured and thus remain areas of future research. Lastly, the exact mechanism through which the CCC achieved its impact has not been examined. For instance, it is not clear from the available data whether the observed reductions in cost and care utilization under CCC were attributable to enhanced caregiver engagement, improved care coordination across providers, or better medication management via pharmacist services. Further studies are necessary to provide more insights on this issue.

Conclusion

This study suggests that a complex care management model embodied by a clinical redesign incorporating team-based care, longer and more routine visits, a coordinated care plan, and clinic time for emergent care access can potentially reduce total medical care costs and acute care utilization for medically complex adolescent and young adult patients transitioning to adult medicine.

Footnotes

Author Disclosure Statement

Drs. Maeng, Snyder, and Davis, and Ms. Tomcavage are current employees of Geisinger Health System. There are no other current or foreseeable future conflicts of interest. The authors received no financial support for the research, authorship, and/or publication of this article.

References

Cohen

, Kuo

, Agrawal

, et al. Children with medical complexity: an emerging population for clinical and research initiatives. Pediatrics, 2011; 127:529–538.

Perrin

, Bloom

, Gortmaker

. The increase of childhood chronic conditions in the United States. JAMA, 2007; 297:2755–2759.

Lotstein

, McPherson

, Strickland

, Newacheck

. Transition planning for youth with special health care needs: results from the National Survey of Children with Special Health Care Needs. Pediatrics, 2005; 115:1562–1568.

Scal

, Ireland

. Addressing transition to adult health care for adolescents with special health care needs. Pediatrics, 2005; 115:1607–1612.

Goodman

, Hall

, Levin

, et al. Adults with chronic health conditions originating in childhood: inpatient experience in children's hospitals. Pediatrics, 2011; 128:5–13.

American Academy of Pediatrics; American Academy of Family Physicians; American College of Physicians-American Society of Internal Medicine. A consensus statement on health care transitions for young adults with special health care needs. Pediatrics, 2002; 110(6 pt 2):1304–1306.

Sharma

, O'Hare

, Antonelli

, Sawicki

. Transition care: future directions in education, health policy, and outcomes research. Acad Pediatr, 2014; 14:120–127.

Vaks

, Bensen

, Steidtmann

, et al. Better health, less spending: redesigning the transition from pediatric to adult healthcare for youth with chronic illness. Healthc (Amst), 2016; 4:57–68.

Bhawra

, Toulany

, Cohen

, Moore Hepburn

, Guttmann

. Primary care interventions to improve transition of youth with chronic health conditions from paediatric to adult healthcare: a systematic review. BMJ Open, 2016; 6:e011871.

10.

Berens

, Peacock

. Implementation of an academic adult primary care clinic for adolescents and young adults with complex, chronic childhood conditions. J Pediatr Rehabil Med, 2015; 8:3–12.

11.

Berry

, Hall

, et al. Inpatient growth and resource use in 28 children's hospitals: a longitudinal, multi-institutional study. JAMA Pediatr, 2013; 167:170–177.

12.

Okumura

, Kerr

, Cabana

, Davis

, Demonner

, Heisler

. Physician views on barriers to primary care for young adults with childhood-onset chronic disease. Pediatrics, 2010; 125:e748–e754.

13.

Peter

, Forke

, Ginsburg

, Schwarz

. Transition from pediatric to adult care: internists' perspectives. Pediatrics, 2009; 123:417–423.

14.

Okumura

, Heisler

, Davis

, Cabana

, Demonner

, Kerr

. Comfort of general internists and general pediatricians in providing care for young adults with chronic illnesses of childhood. J Gen Intern Med, 2008; 23:1621–1627.

15.

Patel

, O'Hare

. Residency training in transition of youth with childhood-onset chronic disease. Pediatrics, 2010; 126 suppl 3:S190–S193.

16.

Bodenheimer

, Berry-Millett

. Care management of patients with complex healthcare needs. Princeton, NJ: Robert Wood Johnson Foundation, 2009.

17.

Homer

, Klatka

, Romm

, et al. A review of the evidence for the medical home for children with special health care needs. Pediatrics, 2008; 122:e922–e937.

18.

Long

, Cabral

, Garg

. Are components of the medical home differentially associated with child health care utilization, health, and health promoting behavior outcomes?. Clin Pediatr, 2013; 52:423–432.

19.

Romaire

, Bell

, Grossman

. Medical home access and health care use and expenditures among children with special health care needs. Arch Pediatr Adolesc Med, 2012; 166:323–330.

20.

Brown

, Peikes

, Peterson

, Schore

, Razafindrakoto

. Six features of Medicare coordinated care demonstration programs that cut hospital admissions of high-risk patients. Health Aff (Millwood), 2012; 31:1156–1166.

21.

Hong

, Siegel

, Ferris

. Caring for high-need, high-cost patients: what makes for a successful care management program?. New York: The Commonwealth Fund, 2014.

22.

Adams

, Cohen

, Mahant

, Friedman

, Macculloch

, Nicholas

. Exploring the usefulness of comprehensive care plans for children with medical complexity (CMC): a qualitative study. BMC Pediatr, 2013; 13:10.

23.

Berry

, Agrawal

, Cohen

, Kuo

. The landscape of medical care for children with medical complexity. Alexandria, VA: Children's Hospital Association, 2013.

24.

Hauck

, Adler

, Mulla

. Clinical pathway care improves outcomes among patients hospitalized for community-acquired pneumonia. Ann Epidemiol, 2004; 14:669–675.

25.

Crowley

, Wolfe

, Lock

, McKee

. Improving the transition between paediatric and adult healthcare: a systematic review. Arch Dis Child, 2011; 96:548–553.

26.

Sattoe

, Peeters

, Hilberink

, Ista

, van Staa

. Evaluating outpatient transition clinics: a mixed-methods study protocol. BMJ Open, 2016; 6:e011926.

27.

Berry

, Hall

, Neff

, et al. Children with medical complexity and Medicaid: spending and cost savings. Health Aff, 2014; 33:2199–2206.

28.

American Academy of Pediatrics; American Academy of Family Physicians; American College of Physicians; Transitions Clinical Report Authoring Group. Supporting the health care transition from adolescence to adulthood in the medical home. Pediatrics, 2011; 128:182–200.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB