Reported costs of children with medical complexity

Abstract

Examining reported costs for Children with Medical Complexity (CMCs) is essential because costing and resource utilization studies influence policy and operational decisions. Our objectives were to (1) examine how authors identified CMCs in administrative databases, (2) compare reported costs for the CMC population in different study settings, and (3) analyze author recommendations related to reported costs. We undertook a systematic search of the following databases: Medical Literature Analysis and Retrieval System Online, Excerpta Medica dataBase, Cumulative Index to Nursing and Allied Health Literature, and Cochrane Library with a focus on CMCs as a heterogeneous group. The most common method used n = 11 (41%) to identify the CMC population in administrative data was the Complex Chronic Conditions methodology. The majority of included studies reported on health care service costs n = 24 (89%). Only n = 3 (11%) of the studies included costs from the family perspective. Author recommendations included standardizing how costs are reported and including the family perspective when making care delivery or policy decisions. Health system administrators and policymakers must consider the limitations of reported costs when assessing local costing studies or comparing costs across jurisdictions.

Keywords

Children with medical complexity health care costs cost-benefit analysis disabled children medical assistance chronic disease

Introduction

Children with Medical Complexity (CMCs) are a heterogeneous pediatric population that span a wide range of clinical diagnoses such as Neuromuscular and Cardiovascular chronic conditions and Congenital/genetic defects (Cohen et al., 2011). CMCs require considerable healthcare resources, have poor health outcomes, and place a substantial financial and psychological burden on caregivers (Berry et al., 2015; Cohen et al., 2011). While CMCs make up roughly 1% of the pediatric population, they account for 30% of pediatric health care resource use (Berry et al., 2015; Cohen et al., 2012a). Understanding healthcare and family costs are essential for clinicians, researchers, and policymakers to improve healthcare delivery and support for CMCs and their families.

Children with Medical Complexity share four key characteristics; the presence of one or more complex chronic conditions, reliance on technology for survival (e.g., feeding tubes or tracheostomies), high healthcare utilization (e.g., requiring several specialist care visits and frequent hospitalizations), and significant caregiver needs (e.g., care coordination, financial and social needs) (Cohen et al., 2011; Dewan and Cohen 2013). While identifying CMCs at an individual level is accomplished through clinical assessment, identifying CMCs at a population level can be challenging (Berry et al., 2015). According to Berry et al. (2015), the three main challenges to identifying CMCs at a population level are (1) how medical complexity is defined, (2) discrepancies in healthcare data collection and reporting, and (3) CMCs, unlike adults with medical complexity, tend to have multiple chronic conditions without any being dominant or prevalent across the population. In addition to the heterogeneity of CMCs, these challenges make it difficult to report on healthcare costs for this population accurately.

This review examines the type and methods of reporting costs for this population. Such a review is essential as cost information can shape health care policy, care service delivery and research priorities. To our knowledge, this is the first review of this kind.

Aims

This systematic review has three aims: (1) to determine how CMCs are identified, (2) to analyze and compare reported costs for the CMC population, and (3) to examine author recommendations related to the costs reported in each included study.

Methods

A protocol was developed (not published), using the Cochrane Handbook for Systematic Reviews of Interventions and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) method for systematic reviews (Moher et al., 2009).

Search strategy

We developed the search strategy in consultation with a research librarian and peer-reviewed by a second librarian. We searched for peer-reviewed English language articles in the following databases: Medical Literature Analysis and Retrieval System Online (Medline) (1946–April 22, 2019), Excerpta Medica dataBase (EMBASE) (1974–April 22, 2019), Cumulative Index to Nursing and Allied Health Literature (CINAHL) (inception to April 22, 2019), and Cochrane Library (inception to Current). An online search of non-peer-reviewed articles was also conducted using Google search engine but did not yield additional studies.

Selection criteria

Studies were included if they reported costs for a pediatric population with Neurological Impairment, Complex Chronic Conditions (CCCs), use for Technology Assistance, high resource use, or high family care needs as described in the conceptual framework for CMCs reported by Cohen et al., (2011). Studies not examining CMCs as a heterogeneous group were excluded.

Study designs included were primary studies and literature reviews; letters, commentaries and opinion pieces were excluded. The outcomes of interest were to examine how CMCs were identified, compare what and how cost estimates were reported and examine author recommendations related to the measurement of CMC-related costs.

Study selection

The selection of studies was a two-step process. The first screening involved examining study titles and abstracts against pre-determined selection criteria using Microsoft Excel© 2016 (Microsoft Corporation, Redmond, WA, USA) by two independent reviewers (MSi, MR). The second screening stage involved examining the full text of studies identified in stage one and was completed by the same reviewers (MSi, MR). Any differences underwent discussion and were resolved by consensus. A third reviewer was available in case the two reviewers could not reconcile differences but was not needed.

Data collection and analysis

A piloted data collection tool Microsoft Excel© 2016 (Microsoft Corporation, Redmond, WA, USA), was used to extract the following data: study details (e.g., author, year, country, study design, and objective), population characteristics (e.g., inclusion and exclusion criteria, sample size, participant age), and study costs and related outcomes (e.g., the definition used to identify CMCs, method used for evaluation of an intervention, costs reported and related author recommendations). One reviewer (MSi) extracted data, and a second reviewer (MSe) verified a random selection of studies.

Data were then organized by type of study topic. For example, some studies used descriptive analysis of healthcare administrative data, whereas others conducted an economic evaluation of a specified intervention. The types of costs analyzed included large-scale macro-costing, local level micro-costing, and economic assessment studies reporting costs for the population. Studies were then analyzed for similarities and differences in (1) how the study defined the CMC population, (2) type of costs reported, and (3) cost-related author recommendations.

Quality appraisal of the included studies was conducted in duplicate by two reviewers (MSi and MSe), and conflict was resolved through consensus. The Critical Appraisal Skills Program (CASP) (Critical Appraisal Skills Programme, 2020) for cohort studies (CASP Cohort Study Checklist) was used to assess 26 (96%) of the studies, while the CASP Economic Evaluation Checklist was used for one (3%) study (Fields et al., 2018).

The three domains of quality assessment in the CASP tool are validity of results, study precision and applicability. Results are included in Table 1. Each included study was rated using the tool’s assessment questions as follows:

1. Are results valid? This domain was summarized as either high, medium, or low validity based on clear objectives, appropriate cohort recruitment and accounting for outcome bias and confounding measures.

2. Are results precise? This domain was summarized as either precise, somewhat precise and not precise/unsure. We determined precision based on sample size and spread of the confidence intervals or inter-quartile.

3. Are results applicable? This domain category was summarized as either high, medium or low based on the generalizability of study results, the applicability of results to local CMC populations, and how similar they were to other evidence in the literature.

Table 1.

Characteristics of included studies and quality appraisal results.

First Author (Year) / Country	Study design / Sample size	Study focus	Method used to identify CMCs	Reported costs and data sources	Main findings	CASP^a evaluation results
First Author (Year) / Country	Study design / Sample size	Study focus	Focus of this review			Valid	Precise	Applicable
Descriptive studies
Agrawal et.al., (2016), USA	Retrospective Cohort, sample size = 48743	Describes characteristics and trends of pediatric high resource users in Medicaid	Three groups; (1) children whose health system costs fell below the top 5% and did not return to the top 5%	Utilization rates reported—CMCs relative to other pediatric populations	One half of the top 5% of children resource utilizers in Medicaid will not continue to be in the top 5% while the other half will either stay in the top 5% or intermittently fall off and return to the top 5%	High	Precise	Medium
			(2) children whose spending fell below the top 5% but returned to the top 5%	Administrative and claims data from 10 state Medicaid programs	—
			(3) children who’s spending consistently remained in the top 5%	—	—
Ananth et.al., (2005), USA	Retrospective Cohort, sample size = 1252	Describes characteristics of children who died in 2012 and hospital use of type and number of life-threatening complex chronic conditions	Feudtner et. al., definition of complex chronic conditions (CCCs)	Costs were derived by converting charges to costs and adjusting for inflation	Hospital use was greatest among children with hematologic/immunologic conditions and children with > 3 life-threatening complex chronic conditions (LTC-CCC)	High	Precise	Medium
			—	Hospital costs for last year of life by type and number of Complex Chronic Conditions	Hospital use for children with LTC-CCCs in the last year of life varies significantly across the type and number of conditions
			—	National administrative data from 40 US Children’s Hospitals	—
Bekmezian et.al., (2008), USA	Retrospective Cohort, sample size = 816	Compares costs of patients with subspecialty needs in a staff-only general pediatric hospitalist group with subspecialty/faculty/house staff gastroenterology, hematology and oncology groups	3M’s Clinical Risk Group categories	Reported actual direct hospital costs	Patients in a staff-only hospitalist group experienced a 38% reduction in LOS and 29% reduction in direct costs compared to patients admitted to a faculty/house staff system	High	Precise	Medium
Bekmezian et.al., (2008), USA	Retrospective Cohort, sample size = 816		—	Local administrative data and direct costing supplied by the study hospital	—	High	Precise	Medium
Berry et.al., (2011), USA	Retrospective Cohort, sample size = 1083	Describes characteristics of hospitalizations for children with multiple, chronic conditions	(1) Feudtner et. al, Complex Chronic Condition, (2) Neurologic impairment and (3) Technology assistance	Costs reported as utilization rates	Better provider communication during transition from hospital to home, can prevent hospital readmissions	Medium	Precise	Medium
Berry et.al., (2011), USA	Retrospective Cohort, sample size = 1083		—	Administrative Data from 4 pediatric hospitals	High readmission rates in children with technology	Medium	Precise	Medium
Berry et.al., (2017), USA	Retrospective Cohort, 2.26 million hospital discharges	Identifies which specific chronic conditions of children with multiple chronic conditions are associated with high hospital resource use	Healthcare Research and Quality’s Chronic Condition Indicators	Hospital costs derived by converting charges to costs and adjusting for inflation	CMCC accounted for over one-fourth of acute-care hospitalizations and one-half of all hospital dollars for US pediatric care in 2012	Medium	Precise	High
Berry et.al., (2017), USA	Retrospective Cohort, 2.26 million hospital discharges		—	National administrative data from all USA acute care hospital discharges	—	Medium	Precise	High
Berry et.al., (2014), USA	Retrospective Cohort, sample size = 214765	Describes expenditures of CMCs and present a business case for reducing cost and reinvesting in outpatient and community care management	Healthcare Research and Quality’s Chronic Condition Indicators	Reported costs derived using standard costing and administrative data	Cost savings from a small reduction in hospital days could pay for more expensive care management for a small proportion (most acute) of CMC	Medium	No / unsure	Medium
			—	Multiple administrative databases including public payer direct billing data (Medicaid)	While, less intensive care management could be delivered to a larger portion of CMC
			—	—	Business case did not report how much families of CMC spend on care management activities in their home
Chan et.al., (2016), USA	Cross-sectional, sample size = 136133	Describes the use of critical care resources for children with differing medical complexity in pediatric ICUs in tertiary-care children’s hospital	Critical care defined as admission to a pediatric ICU or pediatric cardiac ICU	Hospital costs derived by converting charges to costs and adjusting for inflation	ICU resource use was greatest among patients with complex chronic disease compared with children with no complex chronic disease or no chronic disease	Medium	No / unsure	Medium
Chan et.al., (2016), USA	Cross-sectional, sample size = 136133		Medical complexity categories, that is, no chronic condition, noncomplex chronic conditions, complex chronic conditions	National administrative data from 40 pediatric hospitals	—	Medium	No / unsure	Medium
Chirico et.al., (2019), USA	Retrospective Cohort, sample size = 224	Describes characteristics of patients who died in a community-based palliative program and examine cost differences between location of death	All children who were enrolled in Compassion Net (community based palliative care program) between 2008 and 2015	Reported costs derived using standard costing and administrative data	Location of death, malignancy diagnosis and participation in Medicaid all were significantly related to total cost in final year of life	High	Precise	Low
Chirico et.al., (2019), USA	Retrospective Cohort, sample size = 224		—	Local and national administrative databases linked with insurance claims data to derive costs	—	High	Precise	Low
Cohen et.al., (2012a,b), Canada	Retrospective Cohort, sample size = 15771	Evaluates health care utilization and costs in a population-based sample of CMC in Ontario, Canada	Complex chronic conditions. All children aged newborn to 16 years who were hospitalized between April 1, 2005 and March 31, 2007. ICD-10 diagnostic codes within 3 clinical categories relevant to CMC: NI, complex chronic conditions (CCCs), and TA (Technology Assistance)	Reported costs derived using standard costing and administrative data	Hospitals wanting to discharge patients sooner without increasing risk of readmission should consider targeting care coordination for this high user group	High	Precise	High
Cohen et.al., (2012a,b), Canada	Retrospective Cohort, sample size = 15771		—	Provincial linked administrative data with public payer costs	—	High	Precise	High
Coller et.al., (2018), USA	Randomized Control trial, sample size = 147	Examines the effect of a caregiver coaching intervention, Plans for Action and Care Transitions (PACT), on hospital use among CMCs within a complex care medical home at an urban tertiary medical center	Complex Chronic conditions (CCCs) as defined by Feudtner et al. and included need for technology assistance	Costs reported as charges for hospital and other interventions	Those enrolled in PACT had fewer all-cause 30-day readmissions and were less likely to experience at least 1 readmission.	High	No / unsure	Low
Coller et.al., (2018), USA	Randomized Control trial, sample size = 147		—	Institutional administrative database	—	High	No / unsure	Low
Gold et.al., (2016), USA	Retrospective Cohort, sample size = 954,018CMC at 44 children’s hospitals	Assesses the impact of, risk factors for, and variation across children’s hospitals regarding long LOS (10Institutional administrative databasedays) hospitalizations in CMC	3M’s Clinical Risk Group categories	Hospital costs reported by converting charges to costs and adjusting for inflation	The characteristics most strongly associated with LOS 10 days were use of the ICU, respiratory complex chronic condition, and transfer from another medical facility at admission	High	Precise	Medium
Gold et.al., (2016), USA			—	National administrative data	Efforts to reduce hospital costs in CMC might benefit from a focus on prolonged LOS	High	Precise	Medium
Murtagh Kurowski et.al., (2014), USA	Cross-sectional, sample size = 0–17 years = 577507 patients, 18–24 years = 482785 patients	Generates nationwide estimates for Emergency Department (ED) use by children and young adults with complex chronic conditions. Evaluate if being of the age for transition to adult care significantly affects the site of care and likelihood of hospital admission	Patients with one or more complex chronic conditions, between the ages of 0 and 24 years, who visited an ED at least once during 2008	Hospital costs reported as charges	The majority of visits by pediatric- and transition aged patients with chronic complex conditions are into general EDs.	High	Precise	Medium
Murtagh Kurowski et.al., (2014), USA			—	National administrative database	—	High	Precise	Medium
Ralston et.al., (2015) USA	Retrospective cohort, sample size = 11804	Describes patterns of care for a population-based cohort of children with medical complexity in northern New England	Complex Chronic conditions (CCCs) as defined by Feudtner et al	Reported costs derived using standard costing and administrative data	Hospital-specific resource utilization in children with medical complexity exhibits a high degree of variation, particularly around high cost, discretionary services	High	Precise	Medium
Ralston et.al., (2015) USA	Retrospective cohort, sample size = 11804		—	National administrative data	Need for identifying best practices for this growing patient population	High	Precise	Medium
Srivastav et.al., (2016) Australia	Retrospective Cohort, sample size = 48758	Describes the hospital costs, hospital types and differences across states and territories for CMCs cared for in Australian public hospitals	Children with complex chronic conditions, neurological impairment, and technology assistance	Reported costs derived using standard costing and utilization metrics	CMCs accounted for 32.1% of pediatric healthcare costs for a total cost of $620,948,769	Medium	Precise	High
Srivastav et.al., (2016) Australia	Retrospective Cohort, sample size = 48758		—	National administrative databases	—	Medium	Precise	High
Evaluation
Avritscher et.al., (2018), USA	Pre-post and 2-phase stepped wedge design, sample size = 280	Assesses sustainability and scalability of medical home for high-risk children with medical complexity	>2 hospitalizations or >1 pediatric intensive care unit (PICU) admission in the year before enrollment in the medical home intervention	Costs reported by converting charges to costs and adjusting for inflation	Findings support scalability of medical home type model of pediatric complex care	High	Precise	Medium
			—	Local hospital administrative data	Such models are not adequately reimbursed
			—	—	New alternative reimbursement mechanisms are needed
Casey et.al., (2011), USA	Pre-post without control, sample size = 225	Evaluates the effect of care coordination multidisciplinary clinic on Medicaid costs for CMC	Children with at least 2 chronic medical conditions and who were followed by at least 2 pediatric subspecialists	Local clinic costs and Medicaid data	The program showed a decrease in total Medicaid costs for CMC	High	No / unsure	Medium
Casey et.al., (2011), USA	Pre-post without control, sample size = 225		—	Local and state (Medicaid) databases	—	High	No / unsure	Medium
Cohen et.al., (2012), Canada	Pre-post without control, sample size = 81	Compares, pre- and post-implementation of integrated complex care coordination for CMC	(a) Children (<16 years) with a known and/or suspected diagnosis of a complex chronic condition that is associated with medical fragility; (b) technology assistance and, (c) involvement of multiple specialists	Reported costs derived using standard costing and administrative data	Formal partnerships between children’s hospitals, community hospitals, families and primary care providers, in care coordination is a promising model for complex care delivery	Medium	No / unsure	Medium
			—	Family costs derived through administered surveys	Improvements in health care utilization and family-centeredness of care are possible with little to no changes in parental perceptions of the child’s health
			—	Provincial administrative databases	—
Gay et.al., (2016), USA	Retrospective cohort, sample size = 2783 hospitalized children receiving post discharge HH services across 19 states and 7361 matched controls not discharged to HH services	Assesses the relationship between home health nursing care (HH) services and hospital use in Children	3 or more Complex Chronic Conditions Chronic condition defined as conditions expected to last >12 months and involve either several different organ systems or 1 organ system severely enough to require specialty pediatric care. Medical technology to maintain a child’s health status (e.g., gastrostomy, tracheostomy)	Costs estimated using charges to cost ratios adjusted for inflation	Children discharged to HH care experienced less hospital use than children with similar characteristics who did not use HH care.	Medium	No / unsure	Medium
Gay et.al., (2016), USA			—	Local Case Mix database and National administrative data	—	Medium	No / unsure	Medium
Goldhagen et.al., (2016), USA	Multi-method with primary data collection and retrospective cohort, sample size = 48 Children 53 Caregivers	Evaluates the results from a 2007 study to demonstrate the impact of a community PedsCare program (CBPPC) on Health-Related Quality of Life (HRQoL) and hospital utilization and costs	Children with chronic life-limiting conditions including children already enrolled in hospice	Costs reported using local institutional costs	Parents-caregivers reported overall positive perceptions with impaired emotional health, decision-making, social support, interaction and communication, child health and self-efficacy in caring for their children, with higher HRQoL scores associated with longer periods of enrollment	Low	No / unsure	Low
Goldhagen et.al., (2016), USA			—	Local institutional database	The utilization analysis showed reductions in hospital services during the pre-enrollment quarters compared to the quarters following their enrollment	Low	No / unsure	Low
Gordon et.al., (2007), USA	Pre-post without control, sample size = 227	Evaluates the impact of a tertiary care center special needs program (SNP)	Children with complex chronic conditions that involve several organ systems and require multiple specialists, technological supports, and community services	Hospital costs reported as charges	An investment of $400,000 per year was associated with more than 50% decrease in hospital days and $10.7M decrease in tertiary care center payments.	Low	No / unsure	Low
Gordon et.al., (2007), USA	Pre-post without control, sample size = 227		—	Local institutional costs	—	Low	No / unsure	Low
Ronis et.al., (2019), USA	Time study, sample size = 208	Quantifies the time required to perform non-reimbursed care coordination activities by a multidisciplinary care coordination program for CMC and to estimate the direct salary costs of that time	3 or more affected body systems Dependence on technology and/or reliance on caregivers for instrumental activities of daily living	Costs were reported by multiplying non-reimbursed worked hours by staff salary costs	In the context of a children’s hospital that is generally well staffed to care for these children, care coordination activities required conservative estimate of $145 to $210 in unreimbursed salary per child per month, not including fringe benefits, management or overhead costs	High	No / unsure	Medium
			Demographics	Staff hourly costs were based on national employment statistics	—
			Family characteristics	Hours of work recorded by participants	—
			Activities performed by staff that are not reimbursed and time spent on those activities	—	—
Shumskiy et.al., (2018), USA	Retrospective Cohort, sample size = 93121	Assesses correlations between years of Well Child Visits (WCV) with hospitalizations for CMC	Feudtner pediatric chronic complex conditions; Agency for Healthcare research and quality’s chronic condition indicators (CCI)	Health service utilization rates	Most Medicaid-insured CMC do not receive annual WCVs consistently over time	High	Precise	Medium
Shumskiy et.al., (2018), USA	Retrospective Cohort, sample size = 93121		—	State Medicaid database	Children with fewer annual WCVs have a higher likelihood of hospitalization	High	Precise	Medium
Simon et.al., (2017), USA	Cluster randomized, sample size = Control group n = 150. Intervention group n = 181. Primary Care Physicians = 200	Assesses whether children with medical complexity (CMC) exposed to a hospital-based comprehensive case management service (CCMS) experience improved health care quality, improved functional status, reduced hospital-based utilization, and/or reduced overall health care costs	3M Clinical Risk Group categories; Identification by participating primary care physicians	Reported costs derived using standard costing and administrative data	Exposure to a hospital-based CCMS focused on care coordination for CMC generally improved quality of care, did not change functional status or hospital-based utilization, and increased overall health care costs	Medium	Precise	Medium
			—	Program implementation costs; net cost after accounting for clinic revenues	Tiered care providing care coordination services commensurate with disease severity may be a more cost-effective approach for CMC
			—	Local institutional database and third-party payers	—
Van Orne et.al., (2018), USA	Quasi-experimental, sample size = 38 patients, 34 primary caregivers plus a similar size control group	Determines the effectiveness of a structured boot camp–style pre-discharge training program for caregivers of pediatric patients with complex medical conditions	Patients with complex medical conditions who are dependent on technology and have newly placed tracheostomies	Savings reported are derived from local administrative data	Statistically significant decreases were found in	High	Precise	Low
				All costs were not reported; however, cost savings were reported and calculated based on median Length of Stay for intervention group versus control group	Mean unit length of stay (92 vs 60 days), mean discharge training days (60 vs 16 days; and median total parental stress score
			—	Local institutional costs	More than 90% of caregiver participants were very satisfied with the program
			—	—	Cost savings were estimated at between $53 300 and $69 900 per patient
Walter et.al., (2018), USA	Retrospective Cohort, sample size = 7286570 -Nonchronic (76.8%)—Noncomplex chronic (16.6%)—Complex chronic (6.6%)	Assesses health care utilization and costs for children enrolled in private insurance plans in the United States	The Pediatric Medical Complexity Algorithm, a modification of the Chronic Disability Payment System, was used to identify and stratify children aged 0–17 years by levels of medical complexity	Reported costs derived using standard costing and administrative data	Families of children with medical complexity are 1.9–3.8 times as likely as those without chronically ill children to spend over US$1000 out of pocket per year on children’s medical services	High	Precise	Low
Walter et.al., (2018), USA			—	Commercial claims data (employer sponsored health insurance)	Medical spending on children is not the only source of financial burdens for these families	High	Precise	Low
Other
Thomson et.al. (2016), USA	Survey design, sample size = 167 CMCs (111 responded to survey)	Describes hardships experienced by families of children with medical complexity (CMC) and compare them with those experienced by families of children with asthma	Severe, chronic disease, and receive care from more than or equal to 3 clinical subspecialists	Costing using local administrative costing	Families of CMC may face greater challenges than previously recognized	Medium	Precise	Low
Thomson et.al. (2016), USA			—	Survey data	Families of CMC frequently report financial and social hardships, often at rates higher than families with asthma who were generally of lower socioeconomic status	Medium	Precise	Low
Fields et al. (2018), USA	Pre-Post without control. sample size = NA	Uses Lean methodologies to standardize processes, and eliminate waste in the provision of medical equipment	All children needing to access medical equipment supplies	Cost assumptions were based on median local staff salaries	The median lead time for processing DME/supply requests decreased from 50 days (range 42–80 days) to 3 days (range 1–17 days; p < .0001)	A different toll, the CASP economic Review, was used to evaluate this study
Fields et al. (2018), USA	Pre-Post without control. sample size = NA		—	Data collected at implementation of intervention	The annual cost of completing DME/supply requests decreased from an estimated $29,000 to $18,000	Although the intervention showed a positive economic benefit the results are not applicable outside of the study area

^aCritical Appraisal Skills Program (CASP) tool for cohort studies was used for every study except for Fields et al. (2018) where the CASP tool for Economic Evaluation was used.

We summarized the CASP Cohort evaluation tool findings as high, medium, or low based on clear objectives, appropriately recruited cohort, and adequately accounting for outcome bias and confounding measures. Precision was summarized as precise, somewhat precise and not precise/unsure based on sample size and spread of confidence intervals or inter-quartile range reported. Finally, applicability of results was summarized as high, medium or low based on study design and how similar the results were to other evidence.

Cost Methods and data sources are further summarized in Table 2—Cost Variable Summary.

Studies were grouped under Descriptive Studies category if they identified CMCs costs and characteristics but were not an evaluation of an intervention.

Studies were grouped under Evaluative Studies category if they included an evaluation of an intervention.

Studies were grouped under Other category if they did not fit with Descriptive or Evaluative Studies.

The CASP Economic Review Checklist was used to examine one process improvement study as it was not a cohort study.

Results

The initial search identified 4395 studies in total, and 3205 studies remained after removing duplicates, as illustrated in Figure 1. Following initial title and abstract screening, 160 titles remained for full-text screening. The full-text review identified 33 papers for data extraction; however, six papers were later excluded because they were conference presentations or full text was not available. A systematic review was excluded to prevent bias as it included similar studies to those in this review.

Figure 1.

Study Selection Flow Diagram

Of the n = 27 included studies, the majority n = 24 (89%) were from the United States; two were from Canada and one from Australia (Table 1). The earliest study was from 2007 (Gordon et al., 2007), and the most recent was published in 2019 (Chirico et al., 2019; Ronis et al., 2019). The ages of participants in the studies varied from birth to 18 years, while three studies included older participants up to 24 years old (Chirico et al., 2019; Goldhagen et al., 2016; Murtagh Kurowski et al., 2014). These studies were included in this review despite having participants older than 18 years as they reported on CMC’s transition from pediatric to adult care, an important consideration for this population. Table 1 outlines the included studies, summarizing how CMCs were identified, reported costs and data sources, author recommendations related to costs and qualitative appraisal results.

Identifying the children with medical complexity population

The majority n = 18 (67%) of studies included in this review used one of four CMC identification methods (Table 1) as described by Berry et al. (2015). The most common method used n = 11 (41%) was Feudtner et al. (2000) Complex Chronic Conditions. 3M Clinical Risk Groups 3M Science (2020), were used by n = 3 (11%), Chronic Condition Indicators Agency for Healthcare Research and Quality (2020), were used by n = 2 (7%), and Patient Medical Complexity Mangione-Smith (2014), was used by n = 1 (3%) of the studies. Technology assistance was identified in seven (25%) studies as part of their CMC definition.

Costing methods

Type of reported health care costs varied and included standard costing methodology, such as case costing or standard hospital costs n = 8 (30%), local direct care costs n = 7 (26%), costs converted from hospital charges n = 6 (22%), hospital charges not converted to costs n = 3 (15%), and utilization rates including relative utilization rates n = 3 (15%), as outlined in Table 1. Costs to CMC families was reported in n = 3 (11%) of studies. This information was captured in family surveys. Data sources varied and included national administrative databases, claims data, provincial or state-wide databases, private payer data, and local administrative and costing data.

A majority of studies n = 24 (89%) included hospital-related costs while only n = 3 (11%) included both health system and family costs, the most under-reported cost variable shown in this review. Ten (37%) studies reported on outpatient costs, the second most utilized cost variable. Cost variables are summarized in Table 2.

Table 2.

Cost variable summary.

Cost variable considered	# Of studies included cost variable
Hospital costs including admissions, hospital days, hospitalization rates, inpatient costs, length of stay, ICU days, post ICU days, total hospital costs, readmissions, and physician compensation. (Agrawal et al., 2016; Ananth et al., 2005; Avritscher et al., 2019; Bekmezian et al., 2008; Berry et al., 2011, 2014, 2017; Casey et al., 2011; Chan et al., 2016; Chirico et al., 2019; Cohen et al., 2012a, 2012b; Coller et al., 2018; Gay et al., 2016; Gold et al., 2016; Goldhagen et al., 2016; Gordon et al., 2007; Murtagh Kurowski et al., 2014; Ralston et al., 2015; Shumskiy et al., 2018; Simon et al., 2017; Srivastava et al., 2016; Thomson et al., 2016; Van Orne et al., 2018; Walter et al., 2018)	n = 24 (89%)
Outpatient costs including specialist visits, day surgery visits, clinic costs. (Agrawal et al., 2016; Avritscher et al., 2019; Berry et al., 2014; Casey et al., 2011; Cohen et al., 2012a, 2012b; Ralston et al., 2015; Shumskiy et al., 2018; Simon et al., 2017; Thomson et al., 2016; Walter et al., 2018)	n = 11 (40%)
Primary care (Agrawal et al., 2016; Avritscher et al., 2019; Berry et al., 2014; Casey et al., 2011; Cohen et al., 2012a, 2012b; Gordon et al., 2007; Shumskiy et al., 2018; Simon et al., 2017; Van Orne et al., 2018)	n = 10 (37%)
Emergency department (ED) costs including ED visits, type of ED (pediatric vs adult), and admission or transfers from ED. (Berry et al., 2014; Casey et al., 2011; Chan et al., 2016; Cohen et al., 2012a; Murtagh Kurowski et al., 2014; Ralston et al., 2015; Shumskiy et al., 2018; Srivastava et al., 2016; Walter et al., 2018)	n = 10 (37%)
Pharmaceutical (inpatient or outpatient) excluding out of pocket family costs. (Agrawal et al., 2016; Bekmezian et al., 2008; Berry et al., 2014; Casey et al., 2011; Chan et al., 2016; Cohen et al., 2012a; Goldhagen et al., 2016; Simon et al., 2017; Srivastava et al., 2016; Walter et al., 2018)	n = 10 (37%)
Diagnostic testing including medical imaging and/or laboratory (Avritscher et al., 2019; Bekmezian et al., 2008; Berry et al., 2014; Cohen et al., 2012a; Goldhagen et al., 2016; Ralston et al., 2015; Simon et al., 2017; Srivastava et al., 2016)	n = 9 (33%)
Home health care (does not include family or other in-kind home care contributions) (Agrawal et al., 2016; Berry et al., 2014; Chirico et al., 2019; Cohen et al., 2012a, 2012b; Gay et al., 2016; Gold et al., 2016; Simon et al., 2017)	n = 8 (29%)
Therapeutic care including physical therapy, occupational therapy and/or speech therapy. (Agrawal et al., 2016; Berry et al., 2014; Casey et al., 2011; Cohen et al., 2012a; Goldhagen et al., 2016; Simon et al., 2017; Van Orne et al., 2018; Walter et al., 2018)	n = 8 (29%)
Mental health services (inpatient and outpatient) (Agrawal et al., 2016; Berry et al., 2017; Chan et al., 2016; Cohen et al., 2012a; Srivastava et al., 2016; Walter et al., 2018)	n = 6 (22%)
Care coordination activities including family training programs (excluding home care). (Cohen et al., 2012b; Simon et al., 2017; Thomson et al., 2016; Van Orne et al., 2018)	n = 5 (18%)
Family costs including any out-of-pocket expenses and opportunity costs. (Cohen et al., 2012b; Thomson et al., 2016; Walter et al., 2018)	n = 3 (11%)

Note: Cost variables were reported differently throughout the included studies. For example, some hospital costs were reported as charges, converted to costs or reported as hospital days.

Children with medical complexity costs at a hospital, geographic or national level accounted for n = 14 (52%) descriptive studies. In evaluative studies, n = 11 (40%) included costing of a given intervention. Types of interventions included care coordination, medical home-like support models and CMC family/caregiver educational programs. The remaining two studies, n = 2 (7%), did not fit into the descriptive or evaluative category; one compared CMC family experience and financial hardship with families of asthmatic children, and the other reported a process improvement intervention to improve access to technology for CMCs (Table 1).

A review by Gold et al. (2016) found the prevalence of lengthy hospitalizations for CMC varied significantly by type of chronic condition and that no dominant chronic condition was evident. This lack of a dominant chronic condition presents a challenge in effectively measuring the burden of the disease as each specific ailment has different acuities and, therefore, different financial and social burdens on patients, families and the health system (Berry et al., 2011; Cohen et al., 2012a; Kuo and Houtrow, 2016).

All three (11%) studies that included families' costs identified a significant financial impact to CMC families/caregivers. After examining 7.2 million children’s records, including CMCs enrolled in private insurance plans in the United States, Walter et al. (2018) found that families of CMCs are almost four times more likely to spend over $1000 out of pocket per year than non-CMCs. The authors also noted that this did not include any lost income due to reduced working hours to care for a CMC.

Only two studies (7%) in this review included a health state assessment such as cost-utility analysis or Patient Reported Outcome measures for families/caregivers. Families and caregivers of CMCs are more likely to score their health lower than parents of healthy children or children with single chronic diseases (Dewan and Cohen, 2013).

Cohen et al. (2012b) and Goldhagen et al. (2016) included Health State Quality of Life measurement as part of the evaluation of care coordination. Both studies suggest improving care delivery models such as partnerships between children’s hospitals and care coordination programs can positively affect family caregiver quality of life, and have a positive impact on healthcare costs for the child and the caregiver.

Author recommendations

Table 3 (included as supplemental materials) summarizes the author recommendations and outlines four high-level themes; health spending and resource utilization, care coordination, service delivery and family perspective. Care coordination and a medical home-like model for CMCs post-acute care discharge was the focus of eight (30%) (Avritscher et al., 2019; Casey et al., 2011; Cohen et al., 2012b; Gay et al., 2016; Goldhagen et al., 2016; Gordon et al., 2007; Simon et al., 2017; Van Orne et al., 2018) included studies. All but one of the care coordination evaluation studies reported cost savings to the health system through reduced length of stay, hospital admissions, and readmissions. Unlike the rest of the intervention studies, Simon et al. (2017) found that while hospital-based comprehensive case management service improved quality of care for the CMC population, it did not change hospital-based utilization and increased overall health care costs.

Table 3.

Summary of author recommendations.

Topic theme	Summary of author recommendations
Health spending and resource utilization patterns for CMCs	•Efforts to reduce hospital costs for CMC may benefit from focusing on long LOS. (Bekmezian et al., 2008; Gold et al., 2016)
	•Measuring variation in costs of medical and surgical pediatric conditions across hospital types might provide more context for cost variation. (Srivastava et al., 2016)
	•A publicly available database of healthcare costs and utilization to assess longitudinal trends in spending and health outcomes for CMC is needed. (Berry et al., 2014)
	•New alternative reimbursement methods for programs to support high-risk medically complex children are needed. (Avritscher et al., 2019; Gordon et al., 2007; Ronis et al., 2019)
Care coordination	•Consider policies or programs that enhance and promote consistent utilization of primary care services for CMCs (Agrawal et al., 2016; Shumskiy et al., 2018)
	•Identifying CMC populations that are more likely to benefit from care coordination programs is beneficial for health system costs and families. It is also important to assess the effectiveness of different approaches to care. (Berry et al., 2014; Casey et al., 2011; Cohen et al., 2012b; Gay et al., 2016; Simon et al., 2017; Van Orne et al., 2018)
	•Further studies, examining care coordination type programs, should include more rigorous designs such as RCTs and studies that examine the population for a longer time frame.(Gay et al., 2016)
	•Families, health systems and payers may all benefit from community-based pediatric palliative care programs for CMCs. (Chirico et al., 2019; Goldhagen et al., 2016)
Service delivery	•Hospitals might benefit from exploring the efficiency and effectiveness of inpatient care delivery for CMCs with a chronic mental health condition. (Berry et al., 2017)
	•Consider recruiting clinicians and staff with the clinical competency to care for CMCs with the most common underlying conditions for example, neurological impairment. (Berry et al., 2011)
	•Standardized care protocols for clinical practice including for common admission diagnoses to the ICU may be useful for CMCs. (Chan et al., 2016; Ralston et al., 2015)
	•Using lean methods can be beneficial to improving equipment supply processing time and is beneficial to patients and staff. (Fields et al., 2018)
	•Future research for CMC population should include young adults who are cared for outside of pediatric-specific care center. (Murtagh Kurowski et al., 2014)
Family perspective	•Need to identify key outcomes most important to patients, families, health care providers and payers. (Chirico et al., 2019; Goldhagen et al., 2016; Thomson et al., 2016; Walter et al., 2018)
Family perspective	•Need to better understand the relationship between family-experienced hardships and resource utilization of CMC, as well as to identify characteristics of CMC (e.g., CCC category, type of technology dependence, receipt of home healthcare) associated with the most hardships. (Cohen et al., 2012b; Thomson et al., 2016)

Themes from the included studies are summarized.

Further research on health economic assessment recommendations, especially those considering the family perspective are needed.

Another set of recommendations addressed clinical service delivery for CMCs. A key finding was the need for standardized protocols for clinical practice for CMCs as a means of enhancing care for CMCs and reducing cost variation (Chan et al., 2016; Ralston et al., 2015; Srivastava et al., 2016). Srivastava et al. (2016) suggested measuring variation in medical and surgical pediatric conditions costs might provide more insight into cost variation. Similarly, Berry et al. (2017) found better provider communication during transition from hospital to home could prevent hospital readmissions. Cohen et al. (2012b) reported formal partnerships between children’s hospitals, community hospitals, and primary care are beneficial in reducing CMC care costs. Alternative reimbursement methods for programs supporting CMCs were also recommended (Avritscher et al., 2019; Gordon et al., 2007; Ronis et al., 2019)

Quality of included studies

We assessed the quality of included studies using the CASP tool to examine strength of study designs and applicability of author recommendations. Table 1 includes a summary of the quality appraisal results.

Using the Cohort CASP tool, 15 (56%) of the studies rated as high validity, 9 (33%) medium validity, and two (7%) were low validity. For precision, 17 (63%) were rated as precise, and 9 (33%) were rated as not precise or unsure. In applicability, seven (26%) studies were rated as highly applicable, 16 (59%) were medium applicability, and seven (26%) were rated as low applicability. Finally, using the economic evaluation CASP tool, one (3%) study showed favorable results in reduced operating costs; the savings are specific to the local operational context and may not be applicable in other areas due to different operational structures and processes.

Discussion

This review aimed to examine reported costs for the CMC population, including methods used to identify CMCs and related recommendations. Results indicate that most reported costs are from the healthcare system’s perspective, with significant variation in types and methods used to represent these costs for CMCs. Cost-related author recommendations covered healthcare spending, resource utilization, care coordination, service delivery and the need to include family perspective in reporting costs for this population.

Identifying children with medical complexity

Identifying CMCs in administrative data can be challenging and may affect reported costs for the population (Berry et al., 2015). CMCs are identified in health administrative data using International Classification of Diseases (ICD) ICD-9 and 10 diagnostic codes using various methods.

Our review included studies using all four CMC identification methods based on Berry et al. (2015). While all these methods are capable of identifying CMCs, they may yield different cohorts. For example, Feudtner et al. (2000) would not identify noncomplex Chronic Conditions such as behavioral/mental health conditions, whereas the other three methods would. While Complex Chronic Conditions are clearly defined in Feudtner et al. (2000) and Patient Medical Complexity Algorithm methods, there is no clear definition for Complex Chronic Conditions in the Clinical Risk Groups and Chronic Condition Indicator methods (Berry et al., 2015).

Another limitation of all four methods is the dependence on administrative data, as they do not typically include patient and family-related costs or outcomes. Berry et al. (2015) describe strengths and limitations of each method in detail. While it may be possible to compensate for limitations of these methods through study design, it makes it more challenging to compare costs and outcomes between studies. Future studies need to carefully consider challenges and limitations of using administrative data to identify CMCs and choose the most appropriate option for their study.

Reported costs—health system perspective

Variation in health resource utilization and other types of reported costs in the included studies is partially caused by differences in how data is collected and stored in administrative databases. While administrative databases are a reliable and accessible source of information, they also have well-documented limitations. Riley (2009) described several limitations of claims databases, such as the potential for claims data to contain biases as providers are inherently incentivized to maximize payments, limiting generalizability to a larger population. Limitations with claims data were especially applicable to for-profit health systems, such as the case with many of the studies included in this review.

Healthcare costs were reported as local costs converted by hospital-specific ratios ((AHRQ), 2020) or reported as hospital charges, making it difficult to compare costs across studies. This variation, along with variation in type of health care service cost included in costing data, provides an incomplete picture of CMC resource use. For example, only six (22%) of included studies reported on costs or utilization of mental health resources, even though according to Berry et al. (2017), mental health conditions were the most prevalent chronic conditions in acute-care hospitalizations and are associated with a significant number of pediatric hospital days.

Costs were also reported as a component of evaluating a given intervention, namely care coordination or patient/family education type programs. Consistent with a systematic review by Cohen et al. (2011), we found that most evaluations in the included studies are quasi-experimental and did not include control groups. Nonetheless, intervention studies indicate that examining characteristics of the local CMC population is recommended as some CMCs may benefit from more intensive and often costlier care coordination services than other CMCs. This differentiation in CMC cohort needs is an essential finding for jurisdictions looking to design or improve their CMC care coordination programs.

Standardizing clinical care where appropriate was also identified as an important quality improvement activity that can reduce health system costs and improve care for CMCs and their families (Chan et al., 2016; Ralston et al., 2015). Chan et al. (2016) reported that CMCs make up the largest pediatric Intensive Care Unit (ICU) population, and that standardized care protocols for common admissions, and standard use of palliative care services can improve care for this population. A systematic review by Coller et al. (2014) found that discharge points, transfer from another medical facility, home care, technology supports, and post-discharge follow-up were important in reducing preventable hospitalizations for CMCs (Coller et al., 2014). Our study shows that many of these service categories were not consistently reported in included studies.

Healthcare activities not reimbursed by funding agencies were another area that was not widely considered. Imperfect measurement of reimbursed care activities is a potential barrier to effective and sustainable care coordination efforts (Avritscher et al., 2019; Gordon et al., 2007; Ronis et al., 2019). Ronis et al. (2019) estimated that healthcare staff provided approximately $2400 worth of unreimbursed care coordination time per CMC per year in undocumented activities such as review of medical records, paperwork, and multidisciplinary team meetings. Considering these costs in the design of CMC care programs ensures adequate funding and long-term sustainability.

The context of how healthcare is funded is also an important consideration. For example in the United States, while the Medicare program provides public health insurance for some, the private patient funded third-party insurance is the predominant model. Canada and Australia have more publicly funded healthcare services than the U.S. These important differences should also be considered when using healthcare costing data for policy decisions.

Family/caregiver perspective

Evidence from this review suggests a substantial gap in reporting of costs incurred by families and caregivers of CMCs. Included studies that reported CMC family costs showed that families of CMCs endure a substantial financial burden. For example, Walter et al. (2018), after examining 7.2 million children’s records including CMCs that were enrolled in private insurance plans in the United States, found that families of CMCs are almost four times more likely to spend over $1000 out of pocket per year than non-CMCs. The authors also noted that this did not include any lost income due to reduction in working hours to care for a child with medical complexity. Another study by Thomson et al. (2016) showed that families of CMCs experienced more financial and social hardships.

Measuring and reporting family costs can be more arduous as it often involves survey design and implementation. Excluding family costs from program evaluation not only provides incomplete information but risks transferring the cost burden from the health system to individual CMC families, causing harm to patients and families. Prudent policy makers and funders would carefully examine the implications of funding decisions given the complexity of accurately measuring the patient and family burden for this population. It is critical to consider funding decisions from a larger system perspective that include social and healthcare system supports for this population.

Limitations

This study has several limitations that must be considered. We examined CMCs as a heterogeneous cohort, so costing evaluations of individual complex chronic diseases were not included. As a result, disease-specific costing assessments that were excluded may provide further insight into the financial burden of CMCs. A lack of European studies, with most of included papers from the United States n=24 (89%) is another limitation. This result may be due to the use of the term Children with Medical Complexity, the definition used for CMCs in this search, or because non-English papers were excluded. More research is needed to ascertain differences between North American and other countries' definitions of CMCs. Only peer-reviewed studies were included in this review, and unpublished studies may have different results or provide more insights into the gaps identified.

Implications for practice

While CMCs make up only a small fraction of all children (usually less than 1%), they are consistently identified as the highest utilizers of health care resources (Agrawal et al., 2016; Berry et al., 2017; Chan et al., 2016; Cohen et al., 2012a; Srivastava et al., 2016). Calculating family/caregiver costs alongside health system costs may provide more insight on policy or operational gaps (Berry et al., 2014).

Healthcare administrators, policymakers, and researchers can benefit from the findings of this review by considering limitations of costing data when making policy or operational decisions. The type of data used for cost analysis, including claims data, charge data, or publicly funded system data, impacts how costing information is used and compared across jurisdictions. Variables such as unfunded care coordination time and examining trends over time would also help provide a clearer picture of costs for the local CMC population and inform local care policy or delivery decisions. (Agrawal et al., 2016; Berry et al., 2014; Shumskiy et al., 2018; Srivastava et al., 2016).

Conclusion

Caring for CMCs requires a large, disproportionate amount of healthcare resources and exerts financial and psychosocial strain on families. This systematic review has identified essential issues, including inconsistent reporting of healthcare costs and a need to consider the family perspective. Finally, identifying standardized methodologies for reporting costs for this population will improve the comparability and applicability of results across different jurisdictions.

Supplemental Material

Supplemental Material—Reported costs of children with medical complexity—A systematic review

Supplemental Material for Reported costs of children with medical complexity—A systematic review by Michael Sidra, Meghan Sebastianski, Arto and Sholeh Rahman in Journal of Child Health Care

Footnotes

Acknowledgments

The authors would like to thank Ms Linda G. Slatter for her support in development and execution of the search strategy and Dr David Wyatt Johnson for his review and recommendations for the manuscript. Dr David Wyatt Johnson, MD Senior Medical Director Maternal, Newborn, Child and Youth Strategic Clinical Network at Alberta Health Services. Linda G. Slatter BA, BEd, MLIS, John W. Scott Health Sciences Library (retired).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Michael Sidra

Arto Ohinmaa

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

3M Science. Applied to Life (2020) 3m™ clinical risk groups (crgs). Available at: https://www.3m.com/3M/en_US/health-information-systems-us/providers/grouping-and-classification/crgs/

Ananth

Melvin

Feudtner

, et al. (2005) Hospital Use in the Last Year of Life for Children With Life-Threatening Complex Chronic Conditions. Pediatrics 136: 938–46.

Agency for Healthcare Research and Quality (2020, May) Chronic condition indicator (cci) for icd-9-cm. Available at: https://hcup-us.ahrq.gov/toolssoftware/chronic/chronic.jsp

Agrawal

Hall

Cohen

, et al. (2016) Trends in health care spending for children in medicaid with high resource use. Pediatrics 138(4): e20160682.

Avritscher

EBC

Mosquera

Tyson

, et al. (2019) Post-Trial Sustainability and Scalability of the Benefits of a Medical Home for High-Risk Children with Medical Complexity. The Journal of Pediatrics 206: 232–239. e233.

Bekmezian

Chung

Yazdani

. (2008) Staff-only pediatric hospitalist care of patients with medically complex subspecialty conditions in a major teaching hospital. Archives of pediatrics & adolescent medicine 162: 975–80.

Berry

Agrawal

Kuo

, et al. (2011) Characteristics of hospitalizations for patients who use a structured clinical care program for children with medical complexity. The Journal of Pediatrics 159(2): 284–290.

Berry

Ash

Cohen

, et al. (2017) Contributions of children with multiple chronic conditions to pediatric hospitalizations in the United States: a retrospective cohort analysis. Hospital Pediatrics 7(7): 365–372.

Berry

Hall

Cohen

, et al. (2015) Ways to identify children with medical complexity and the importance of why. The Journal of Pediatrics 167(2): 229–237.

10.

Berry

Hall

Neff

, et al. (2014) Children with medical complexity and Medicaid: spending and cost savings. Health Affairs (Project Hope) 33(12): 2199–2206.

11.

Casey

Lyle

Bird

, et al. (2011) Effect of hospital-based comprehensive care clinic on health costs for Medicaid-insured medically complex children. Archives of Pediatrics & Adolescent Medicine 165(5): 392–398.

12.

Chan

Rodean

Richardson

, et al. (2016) Pediatric critical care resource use by children with medical complexity. The Journal of Pediatrics 177: 197–203. e191.

13.

Chirico

Donnelly

Gupton

, et al. (2019) Costs of care and location of death in community-based pediatric palliative care. Journal of Palliative Medicine 22: 517–521. DOI: 10.1089/jpm.2018.0276

14.

Cohen

Berry

Camacho

, et al. (2012a) Patterns and costs of health care use of children with medical complexity. Pediatrics 130(6): e1463–e1470.

15.

Cohen

Kuo

Agrawal

, et al. (2011) Children with medical complexity: an emerging population for clinical and research initiatives. Pediatrics 127(3): 529–538.

16.

Cohen

Lacombe-Duncan

Spalding

, et al. (2012b) Integrated complex care coordination for children with medical complexity: a mixed-methods evaluation of tertiary care-community collaboration. BMC Health Services Research 12: 366.

17.

Coller

Klitzner

Lerner

, et al. (2018) Complex Care Hospital Use and Postdischarge Coaching: A Randomized Controlled Trial. Pediatrics 142: 1–9.

18.

Coller

Nelson

Sklansky

, et al. (2014) Preventing hospitalizations in children with medical complexity: a systematic review. Pediatrics 134(6): e1628–1647.

19.

Critical Appraisal Skills Programme (2020) Casp checklists.

20.

Dewan

TCE

Cohen

(2013) Children with medical complexity in Canada. Paediatric Child Health 18(10): 518–522.

21.

Feudtner

Christakis

Connell

(2000) Pediatric deaths attributable to complex chronic conditions: a population-based study of Washington State, 1980-1997. Pediatrics 106: 205–9.

22.

Fields

Neogi

Schoettker

, et al. (2018). Using Lean methodologies to streamline processing of requests for durable medical equipment and supplies for children with complex conditions. Healthcare (Amsterdam, Netherlands) 6: 245–52.

23.

Gay

Thurm

Hall

, et al. (2016) Home health nursing care and hospital use for medically complex children. Pediatrics 138(5): e20160530. (Coller et al., 2014).

24.

Gold

Hall

Shah

, et al. (2016) Long length of hospital stay in children with medical complexity. Journal of Hospital Medicine 11(11): 750–756.

25.

Goldhagen

Fafard

Komatz

, et al. (2016) Community-based pediatric palliative care for health-related quality of life, hospital utilization and costs lessons learned from a pilot study. BMC Palliative Care 15: 73.

26.

Gordon

Colby

Bartelt

, et al. (2007) A tertiary care-primary care partnership model for medically complex and fragile children and youth with special health care needs. Archives of Pediatrics & Adolescent Medicine 161(10): 937–944.

27.

Kuo

Houtrow

Norwood

, et al. (2016) Recognition and management of medical complexity. Pediatrics 138(6): e20163021.

28.

Mangione-

Smith

(2014). Pediatric Medical Complexity Algorithm: A New Method to Stratify Children by Medical Complexity. Pediatrics 133.

29.

Moher

Liberati

Tetzlaff

, et al. (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6: e1000097.

30.

Murtagh Kurowski

Byczkowski

Grupp-Phelan

(2014) Comparison of emergency care delivered to children and young adults with complex chronic conditions between pediatric and general emergency departments. Academic Emergency Medicine: Official Journal of the Society for Academic Emergency Medicine 21(7): 778–784.

31.

Ralston

Harrison

Wasserman

, et al. (2015) Hospital variation in health care utilization by children with medical complexity. Pediatrics 136(5): 860–867.

32.

Riley

(2009) Administrative and claims records as sources of health care cost data. Medical Care 47(7): S51–S55.

33.

Ronis

Grossberg

Allen

, et al. (2019) Estimated nonreimbursed costs for care coordination for children with medical complexity. Pediatrics 143(1): e20173562.

34.

Shumskiy

Richardson

Brar

, et al. (2018) Well-child visits of medicaid-insured children with medical complexity. The Journal of Pediatrics 199: 223–230. e222.

35.

Simon

Whitlock

Haaland

, et al. (2017) Effectiveness of a Comprehensive Case Management Service for Children With Medical Complexity. Pediatrics 140(6): e20171641.

36.

Srivastava

Downie

Hall

, et al. (2016) Costs of children with medical complexity in Australian public hospitals. Journal of Paediatrics and Child Health 52(5): 566–571.

37.

Thomson

Shah

Simmons

, et al. (2016) Financial and Social Hardships in Families of Children with Medical Complexity. The Journal of Pediatrics 172: 187–93. e1.

38.

Van Orne

Branson

Cazzell

(2018) Boot camp for caregivers of children with medically complex conditions. AACN Advanced Critical Care 29(4): 382–392.

39.

Walter

Ellis

Yuan

(2018) Health care utilization and spending among privately insured children with medical complexity. Journal of Child Health Care: For Professionals Working with Children in the Hospital and Community 23(2): 213–231. DOI: 10.1177/1367493518785778

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

0.67 MB