Health Care Models for Persons with Multiple Chronic Conditions from Populations that Experience Health Disparities: A Scoping Review

Abstract

Almost half of U.S. adults have multiple chronic conditions (MCC), and the prevalence of MCC has significantly increased for racial and/or ethnic minority groups, especially among those aged 45–64 years. Yet, little is known about evidence-based health care models for managing MCC in these populations. The overall objective of this scoping review was to identify the breadth of literature testing health care delivery models or components of models to improve the management of MCC for populations that experience health disparities. The databases of CINAHL Plus, Embase, PubMed, and Scopus were searched for original articles from 2016 to 2023. Included studies had to assess a health care delivery model, intervention, approach, or strategy for improving the management of two or more chronic conditions among U.S. adults. Using Covidence, each record was independently assessed by two reviewers and relevant data about the study, health care model, population studied, and outcomes were extracted. Out of 9583 initially screened records, 17 met the inclusion criteria, of which 5 (29%) were randomized controlled trials. Most (82%) studies focused on the management of psychiatric and physical chronic conditions. The most cited care model was the Patient-Centered Medical Home (41%). Most studies (82%) were conducted within clinical settings: primary care (n = 9), specialty care (n = 4), and behavioral health (n = 2). All studies documented positive improvements in patient outcomes, including fourteen (82%) studies that measured outcomes related to service utilization and eleven (65%) studies that measured clinical outcomes. Four studies (24%) measured cost-related outcomes. While the Chronic Care Model was developed almost 30 years ago, the applicable evidence for MCC is sparse for populations experiencing health disparities. There is an opportunity for research to develop, adapt, integrate, and implement evidence-based health care models for MCC to improve clinically significant health outcomes that align with the patient goal needs.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11606-025-09491-w.

BACKGROUND

Almost half (42%) of adults in the United States (U.S.) have multiple chronic conditions (MCC),¹ defined as having two or more chronic conditions such as diabetes mellitus, cardiovascular disease (CVD), chronic obstructive pulmonary disease (COPD), depression, or anxiety. The prevalence of MCC has significantly increased for racial and/or ethnic minorities, especially among those aged 45–64 years.^2,3 The burden of MCC is twice as high among adults living in poverty,⁴ and there is a significant concentration of MCC in the U.S. continental southeast region.⁵ Despite the accumulation of evidence supporting effective treatments and the development of care guidelines for some of the most common chronic conditions, these advances have not been equitably implemented.^6–9 The proportion of people achieving treatment and control goals has consistently been lower among populations experiencing health disparities.^10–14 This often results in high cost of medical services, including emergency department visits and preventable hospitalizations, related to specific chronic conditions^15–27 and MCC^28–32 among the uninsured or underinsured³³ and those in rural communities.^33,34

The management of MCC presents unique challenges beyond a single chronic condition, such as treatment conflicts, complex care coordination, care prioritization, and self-management burden.³⁵ Evidence-based health care models implemented at the health system or clinic level could be leveraged to address these challenges that unduly impact populations and communities. Health care models refer to the systematic organization and delivery of high-quality patient-centered care.^36,37 For example, the Chronic Care Model emphasizes the assurance of productive interactions between the patient and care team and proposes the integration of six elements: health care organization, community resources, patient self-management support, delivery system design, health care provider decision support, and clinical information system.³⁸ Other health care models in which components of the Chronic Care Model have been expanded or enhanced include the Patient-Centered Medical Home,^39,40 eHealth Enhanced Chronic Health Care Model,⁴¹ Community-Based Transition Model,⁴² Model for Developing Complex Interventions in Nursing,⁴³ Home-Based Model,⁴⁴ Integrated Delivery Systems Model,⁴⁵ Team-Based Care,⁴⁶ Family Management Framework,⁴⁷ and others.^48–50 Recently, the Value-Based Care Model has gained increasing interest.^51–53 Furthermore, evidence-based multiple disease-specific guidelines that account for the concurrent treatment of MCC are needed as they presently do not exist.⁵⁴ While some health care models have demonstrated improvements in health outcomes in a single condition, we are unaware if these models have been shown to improve optimal management and control of two or more coexisting chronic conditions specifically for populations that experience health disparities.

Therefore, we performed a scoping review of the literature over an eight-year period to identify studies of health care models for persons with MCC from populations that experience health disparities and to determine which health care models or components of these models reduced health disparities. We were particularly interested in studies that tested and described health care models that successfully implemented evidence-based care or practice guidelines for at least two coexisting chronic conditions, especially within the context of low-resource settings or settings serving a high need population.

METHODS

We chose the scoping review approach to comprehensively map the current literature and identify research gaps in this field.⁵⁵ We used the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist to report this review (Appendix 1).⁵⁶

Eligibility Criteria

Eligible studies had to be conducted in the U.S. with adults aged 18 years and older. At least one of the conditions in the study had to be a physical health condition and both chronic conditions had to be independent and not a consequence of each other (e.g., diabetes mellitus and diabetic neuropathy). Analytical studies, with or without a comparison group, were included if they assessed any health care model, intervention, approach, or strategy for improving the management or coordination of comprehensive health care and outcomes of at least two coexisting chronic conditions. Outcomes had to be related to the MCC, including any process of care (e.g., hemoglobin A1 C (A1 C) recommendation or Patient Health Questionnaire- 9 (PHQ- 9) testing frequency) or health outcome measured at the patient level (e.g., attaining A1 C recommended level or PHQ- 9 score). Examples of eligible chronic conditions, along with the detailed screening criteria, are stated in Appendix 2.

During the full-text review screening, studies were included if at least half of the study population was from a population experiencing health disparities or if the study was conducted in a health care setting that served such a population (e.g., rural community health center or Medicaid serving clinic). We used the National Institutes of Health (NIH) designated populations experiencing health disparities as of 2023,⁵⁷ which includes racial and/or ethnic minority groups (i.e., American Indian or Alaska Native, Asian, Black or African American, Latino or Hispanic, Native Hawaiian and Pacific Islander), socioeconomically disadvantaged populations (e.g., Medicaid beneficiaries), sexual and gender minority groups, rural underserved communities, and persons with disabilities.

Information Sources and Search

A biomedical librarian (AAL) conducted literature searches using four electronic databases: CINAHL Plus (Ebscohost), Embase (Elsevier), PubMed (US National Library of Medicine), and Scopus (Elsevier). A combination of keywords and controlled vocabulary (e.g., CINAHL Subject Headings, EMTREE, and MeSH) for each concept (e.g., chronic condition, health care model, United States) was created by the biomedical librarian with input from the review team members. An initial search was conducted in March 2022 with an update done in January 2024, and a revised search with additional search terms completed in February 2024. The original search was planned as a six-year review to capture the most current literature but was extended two additional years to add search terms to further explore health care models (e.g., integrated care, patient group care, mobile clinics) identified from the screening of the first set of results; the other parts of the search strategy for the other concepts remained the same. The final search strategies for the original and revised searches are in Appendix 3. The searches were limited to those published from January 2016 through December 2023 in English and excluded publication types specified in the exclusion criteria. All database results were exported to EndNote 20 (Clarivate Analytics), duplicates identified, and unique records exported to Covidence (Veritas Health Innovations) for screening.

Selection of Sources of Evidence

Two-levels of screening were completed in Covidence. First the titles and abstracts were screened independently by two reviewers using the eligibility criteria. Second, records that proceeded to full-text review were screened independently by two reviewers. Discrepancies between reviewers were arbitrated by a different third reviewer for both steps of screening. The title and abstract screening procedures were tested on a random set of records to ensure a shared mental model and interpretation of eligibility and exclusion criteria among reviewers.

Data Collection and Data Items

Covidence was used for data collection. Two reviewers independently collected the data from each included article. A third reviewer made the final decision on any data discrepancies identified using Covidence’s consensus feature. Before commencing data collection, the data collection form, data items, and process was tested by the reviewers and then further refined.

The following data items were collected: year published, title, study aims, study design, population description, inclusion and exclusion criteria, total number of participants, populations of interest using NIH definitions, domain of MCC (i.e., psychiatric and physical conditions or only physical conditions), list of chronic diseases studied, name of health care model, type of model (i.e., chronic care model, patient-centered medical home, collaborative care, integrated care, or other), elements of health care model or component(s) being addressed,⁵⁸ length of follow-up, description of health care setting, comparator group, selection and description of outcomes measured (i.e., clinical, service utilization, function status, quality of life, feasibility, sustainability, other), level of outcomes measured, article explicitly state clinically significant improvement, clinical threshold, and report proportion of population that achieved optimal clinical outcomes, and summary of key findings.

Statistical Analysis

The data collected in Covidence was downloaded and imported into SAS, version 9.4 (SAS Institute, Cary, NC) to calculate and report descriptive statistics of the included studies. A summary table of the included studies listed by study design (i.e., randomized controlled trials (RCTs) versus non-randomized) was created and frequencies and percentages were calculated for the key variables extracted.

RESULTS

Study Selection

The searches identified 17,440 records, which are outlined in the PRISMA flow diagram in Fig. 1. After deduplication, the titles and abstracts of 9583 records were screened. Of these, 9408 were excluded and 175 records screened at the full-text level. Most exclusions were due to not being focused on the management of two or more coexisting chronic conditions (n = 84) followed by not being about a health care model (n = 21). After full-text review, 17 records were included in the final analytic data set (Tables 1 and 2).

Figure 1.

PRISMA flow diagram of included and excluded studies.

Table 1.

Included Studies by Study Design (n = 17)

First author last name, year published	Health care model	Model components and health care setting	Chronic conditions studied	Number enrolled	Length of follow-up	Patient outcomes
Randomized controlled trial (n = 5, 29%)
Chan, 202359	Streamlined Unified Meaningfully Managed Interdisciplinary Team (SUMMIT)	Co-located medical and behavioral multidisciplinary primary care team at a FQHC	Two or more chronic medical conditions (e.g., diabetes, CHF, or liver disease) or chronic condition and SUD or mental illness	159	6 months	= Hospitalization  = ED use  + Primary care visits  + Behavioral health visits  + Feasibility  = Patient-reported outcomes (activation experience, HRQOL)  + Self-rated health
Shah, 202360	Diabetes Research, Education, and Action for Minorities (DREAM) Atlanta intervention	Telehealth intervention with health education sessions and action plan development with community health workers	Diabetes and hypertension	183	6 months	+ Hypertension control  = HbA1c level  + Weight loss  + Feasibility (retention rate)
Felker, 202261	Technologies to improve drug Adherence and Reinforce Guideline-based Exercise Targets in patients with Heart Failure and Diabetes Mellitus (TARGET-HF-DM) trial	Telehealth intervention with pillbox and text messages for medication management teaching at 6 clinical sites	Heart failure and diabetes	187	6 months	+ Physical activity  + Quality of life*  = Medication adherence  = Heart failure clinic status  = Glycemic control (HbA1c)
Junkins, 202162	Telemedicine-administered, culturally adapted cognitive behavioral therapy for depression and antiretroviral therapy adherence (CBT-AD) approach	Telemedicine-delivered cognitive behavioral therapy at 4 HIV care outpatient clinics	HIV and depression	22	6 months	= Feasibility/acceptability  = Depression symptoms  = Viral load suppression  = Treatment adherence
Errichetti, 202063	Reverse Co-located Integrated Care	Multidisciplinary care team for medical care at a behavioral health clinic	Serious mental illness (bipolar, schizophrenia, depression, or a combination of these diagnoses) and comorbid chronic disease (diabetes)	416	12 months	+ Hypertension control  + Glycemic control (HbA1c)  + Body mass index  = Total cholesterol  = Depressive symptoms (PHQ- 9 score)
Non-randomized clinical trial (n = 12; 71%)
Steinman, 202364	Program to Encourage Active, Rewarding Lives (PEARLS)	Home-based care program developed with social service organization for self-management, psychoeducation, support, coordination, and mental health specialty care	Depression and another comorbidity: CVD, diabetes, gastrointestinal, nervous system, pulmonary, renal, SUD	164	24 months	+ Inpatient hospitalizations  = ED visits  + Nursing home days  + 12-month all-cause mortality  = Depression score  = Self-rated health
Germack, 202265	Patient-Aligned Care Team (PACT) initiative	Team-based care implementation within the Veterans Health Administration (831 primary care outpatient clinics)	Mental health condition (MDD, psychosis/schizophrenia, bipolar disorder, anxiety, PTSD, SUD) and a comorbid physical health condition (chronic pain or arthritis, hypertension, COPD, diabetes, thyroid disorders, CAD, cardiac arrhythmias)	1,444,942	12 months	+ Hospitalization rates
Grove, 202066	Community Care of North Carolina (CCNC) program	Medical home enrollment within primary care practices to improve primary care access	Schizophrenia and/or depression, and diagnosis of hypertension, hyperlipidemia, seizure disorder, COPD, diabetes, or asthma	83,819	36 months	+ ED use  + Psychiatric facility inpatient stays  + Outpatient visits  + Medical expenditures
Schuttner, 202067	Patient-Aligned Care Team (PACT) initiative	Team-based care, care management, and expanded access and care within the Veterans Health Administration 900 + primary care clinics	≥ 3 chronic diseases in ≥ 3 body systems (hypertension, diabetes, depression, ischemic heart disease, alcohol use disorder, CHF)	318,764	12 months	+ Quality metric for those with 3–4 chronic conditions  + Glycemic control  + Lipid control  = Depression screening
Irwin, 201968	Bridge Intervention: Person-Centered Collaborative Care for Serious Mental Illness and Cancer	Team care with psychiatry and cancer care at a cancer center	SMI (schizophrenia, schizoaffective disorder, bipolar disorder, or MDD with prior psychiatric hospitalization) and cancer (thoracic, gastrointestinal, breast, or head and neck)	25	3 months	+ Psychiatric illness severity (BPRS)  = Depression symptoms (PHQ- 9)  = Quality of life  + Feasibility/acceptability  + Implementation
Crits-Christoph, 201869	Pennsylvania Chronic Care Initiative (CCI)	Practice transformation with coordination processes for behavioral and medical care at 137 primary care practices	HIV and one of 4 comorbid chronic medical conditions (diabetes, COPD, asthma, and CHF) and comorbid behavioral health conditions psychiatric (MDD, schizophrenia/ schizoaffective disorder, bipolar disorder, PTSD, and anxiety disorders) and/or SUD	2879	36 months	= ED use  + All inpatient services  = All outpatient services  + Total inpatient  + Total cost savings
Gilmer, 201870	Behavioral Health Integration and Complex Care Initiative (BHICCI)	Practice transformation including collaborative and complex care management within health systems and organizations including FQHC (primary care), multispecialty clinics, and behavioral health clinics	Severe mental illness – schizophrenia, bipolar disorder, and severe major depression; diabetes, obesity, hypertension	6699	12 months	+ Systolic blood pressure  + HbA1c level  + Depression (PHQ- 9)  + Body mass index  + Cost savings (inpatient only)
Matzke, 201871	Pharmacist–physician collaborative care model	Team-based pharmacist– physician collaborative care approach implemented at 6 hospitals and 22 PCMH practices within a single health system	CHF, hypertension, hyperlipidemia, diabetes mellitus, asthma, COPD, and depression	4960	12 months	+ HbA1c  + Blood pressure  = LDL  = Total Cholesterol  = ED use  + Hospitalizations  + Cost savings
Swietek, 201872	Community Care of North Carolina (CCNC) Patient-Centered Medical Home Program	PCMH program that provides primary care, specialty care coordination, and care management within 14 not-for-profit care networks with 1600 primary care practices	Asthma, COPD, diabetes, hypertension, hyperlipidemia, seizure disorder, MDD, and schizophrenia	131,036	12 months	+ Diabetes metrics: HbA1c testing, attention for nephropathy, liver function, eye exam, lipid panel, medication  + Lipid testing  + Psychotherapy – short-acting b-agonists for asthma
Taber, 201873	Pharmacist-led, Technology-Aided, Education Intervention	Pharmacist-led encounters at clinic with telehealth monitoring of glucose and blood pressure home devices	Kidney transplant recipients with diabetes and hypertension	60	6 months	= Blood pressure control*  = HgA1c < 7%*  = LDL, triglycerides, or HDL  + Medication errors reduction  = Medication adherence
Rhodes, 201674	Pennsylvania Chronic Care Initiative (CCI)	PCMH enrollment at 96 primary care practices	Chronic medical condition—diabetes, COPD, asthma, heart failure; Comorbid behavioral health conditions—psychiatric (MDD, schizophrenia/schizoaffective disorder, bipolar disorder, PTSD anxiety disorders) and SUD (opioid, cocaine, alcohol)	22,210	12 months	+ ED visits  + Inpatient psychiatric utilization  + Cost savings
Rossom, 201775	Care of Mental, Physical and Substance use Syndromes (COMPASS) initiative	Intensive care management using treat-to-target guidelines of care at 172 primary care clinics within 18 health systems (integrated health systems, FQHCs, multisite physician practices, and individual practice associations)	Depression and diabetes or CVD	3363	11 months	+ Depression severity (PHQ9 score < 5)  + HbA1c goal < 8.0%  + Feasibility (depression care satisfaction)  = Hypertension control (< 140/< 90 mmHg)  = Feasibility (satisfaction)

Abbreviations: BPRS, Brief Psychiatric Rating Scale; CAD, coronary artery disease, CHF, chronic/congestive heart failure; COPD, chronic obstructive pulmonary disorder; CVD, cardiovascular disease; ED, emergency department; FQHC, Federally qualified health center; HIV, human immunodeficiency virus; HDL, high-density lipoprotein; HRQOL, health-related quality of life; LDL, low-density-lipoprotein; PCMH, patient-centered medical home; PTSD, posttraumatic stress disorder; MDD, major depressive disorder; SUD, substance use disorder

+ improvements in outcomes, = no changes in outcomes, – worse outcomes, * study assessed clinical significance

Table 2.

Characteristics of Included Studies (n = 17)

	n	%
Types of health care model*
Patient-centered medical home	7	41.18
Collaborative care model	5	29.41
Chronic care model	3	17.65
Integrated care model	2	11.76
Other	5	29.41
Elements of the health care model*
Health system	10	58.82
Community resources	3	17.65
Delivery system design	14	82.35
Self-management support	7	41.18
Decision support	4	23.53
Clinical information systems	5	29.41
Study populations*
Lower socioeconomic status	13	76.47
Racial and ethnic minority groups	9	52.94
Rural underserved or rural clinic	6	35.29
People with disabilities	2	11.76
Types of comorbid conditions studied*
Diabetes	15	88.24
Cardiovascular disease	8	47.06
Hypertension	8	47.06
Serious mental illness	8	47.06
Depression	7	41.18
Heart failure	7	41.18
Chronic obstructive pulmonary disease	6	35.29
Asthma	5	29.41
Substance use disorder	5	29.41
Hyperlipidemia	3	17.65
Liver disease	2	11.76
Human immunodeficiency virus	2	11.76
Renal disease	2	11.76
Cancer	1	5.88
Outcomes measured*
Service utilization	14	82.35
Clinical outcomes	11	64.71
Quality of life	4	23.53
Feasibility	5	29.41
Cost	4	23.53
Sustainability	1	5.88
Level outcomes measured*
Patient	17	100.00
Health system(s)/clinic(s)	6	35.29
Care team	2	11.76
Societal/policy	1	5.88
Clinically significant improvements stated
Yes	3	17.65
No	14	82.35

*More than one option could be selected. Therefore, percent column will not sum to 100%

Characteristics of Included Studies and Multiple Chronic Conditions

Among the 17 included studies, 71% (n = 12) were observational study designs (e.g., non-randomized quasi-experimental or cohort studies) and 29% (n = 5) were RCTs (Table 1). Most (82%) studies focused on the management of one psychiatric and one physical chronic condition while three studies focused exclusively on two physical chronic conditions, specifically diabetes mellitus and CVD. The most common psychiatric condition of study was serious mental illness (n = 8), including major depressive disorder, schizophrenia, bipolar disorder, and posttraumatic stress disorder (Table 2). Five studies also included substance use disorder. Diabetes mellitus (n = 15), CVD (n = 8), hypertension (n = 8), and COPD (n = 6) were the most common physical chronic conditions studied.

Characteristics of the Health Care Model and Health Care Setting

The Patient-Centered Medical Home was the most cited health care model (n = 7) followed by the Collaborative Care Model (n = 5) and the Chronic Care Model (n = 3) (Tables 1 and 2). Four other health care models used telehealth (e.g., mHealth, technology). Most studies (82%) focused on the delivery system design through team-based care or case management followed by addressing the health system (59%) by co-locating services such as primary care and behavioral health. The integration of community resources (18%) was the element least addressed. Health care settings were diverse and included hospitals, primary care networks, federally qualified health centers, multispecialty clinics, and behavioral health clinics as well as telehealth. Three studies were conducted within health systems and 14 studies were conducted within the clinical setting, including primary care (n = 9), specialty care (n = 4), and behavioral health (n = 2). Two studies occurred within the Veterans Administration health care system.

Characteristics of the Study Population

Participants enrolled in these studies ranged from 22 to 1,444,942 persons with a mean follow-up of 12 months. Thirteen studies (76%) included populations of lower socioeconomic status, including Medicaid beneficiaries (n = 8) and people experiencing homelessness (n = 2). Nine studies (53%) included or focused on racial and/or ethnic minority groups. Six studies included rural populations or clinics within rural communities. Two studies (12%) included people with disabilities.

Study Outcomes

All studies demonstrated improvements in outcomes associated with the health care model assessed (Table 1). Fourteen (82%) studies measured outcomes related to service utilization, such as rates of hospitalizations, emergency room visits, primary care visits, as well as receipt of testing for specific conditions. Clinical outcomes (65%) were also commonly measured, including blood pressure level, A1 C level, depressive symptoms, and lipid testing. Of note, most studies on diabetes mellitus focused on glycemic control but not on other clinical goals and recommended guidelines of care. Four studies (24%) measured outcomes related to cost savings and cost expenditures. Five studies assessed feasibility and only one assessed sustainability. All studies measured patient-level outcomes, and six studies measured outcomes at the clinic or health system level.

DISCUSSION

The aim of this scoping review was to identify the breadth of literature testing health care models or components of models to improve the management of MCC for populations that experience health disparities. While there have been other published reviews on MCC,^76–80 to our knowledge this is the first review to evaluate health care models specifically for the management of two or more coexisting chronic conditions for these populations. We identified 17 studies over an eight-year period from the literature. Research on the management of one physical and one psychiatric health conditions was more common than for two physical health conditions. The Patient-Centered Medical Home was the most cited health care model, with most studies focused on the delivery of team-based care or case management followed by addressing the organization of care delivered at the health system level. Most interventions occurred within a clinic setting, with a particular focus in primary care. Populations of focus included persons from lower socioeconomic status and racial and/or ethnic minority groups. While health care models for individual chronic diseases were developed almost 30 years ago,⁸¹ the applicable evidence for MCC remains sparse.

This was not a systematic review, and therefore the effectiveness of the models and study quality were not assessed. From the extensive literature search, we identified five unique health care models tested for persons with MCC from populations that experience health disparities. The integrated care and telehealth models were the only models tested in RCTs. Although positive patient outcomes were reported among the 17 studies, the study characteristics and outcomes measured across studies varied. For example, only three studies assessed clinical significance of changes in patient-level outcomes,^61,70,73 of which two studies compared outcomes by race and ethnicity.^73,75 Specifically, Taber et al. telehealth intervention reported an 18% increase of the study population reaching a blood pressure target < 140/90 mmHg, which was more significant among African American participants than in non-African American participants.⁷³ On the other hand, there were no significant changes in the percent attaining blood pressure < 130/80 mmHg. However, African American participants did experience statistically significant reductions in both systolic and diastolic blood pressure compared to non-African American participants. They also reported a 14% increase among those reaching A1 C < 7%, which was significant among non-African American participants only. Both African American and non-African American participants experienced reductions in A1 C similar in magnitude. Gilmer et al. collaborative care study reported statistically significant improvements in systolic blood pressure, A1 C, body mass index, and the PHQ- 9 score at follow-up among those who exceeded the initial screening threshold.⁷⁰ Those improvements did not reach clinical significance among any of the racial and/or ethnic groups represented in the study. Using the Kansas City Cardiomyopathy Questionnaire to assess quality of life, Felker et al. telehealth RCT demonstrated clinically significant improvement in 51% of patients receiving a mobile health intervention designed to increase guideline-based goals, which was 12% higher compared to the control group.⁶¹

Additionally, we would expect health care models for MCC to be tailored for specific populations given the undue burden of social determinants of health. Three studies centered on specific racial and/or ethnic minority groups, including South Asian immigrants,⁶⁰ African American women,⁶² and Hispanic population.⁶³ Two additional studies addressed social needs along with the management of MCC. Steinman et al. was the only home-based collaborative care model developed with trusted social service organizations to address non-clinical care issues such as housing and food insecurity.⁶⁴ Chan et al. specifically addressed the needs of individuals with MCC experiencing homelessness by implementing a co-located multidisciplinary team that included care management for social needs.⁵⁹ Strengthening the social and clinical care integration for patients with MCC who are also experiencing transportation, financial hardship, inadequate housing, food insecurity, or cultural/language barriers is a critical area for future research.⁸²

The limited or inconsistent efficacy or effectiveness of different health care models on the improvement of clinical outcomes in persons with MCC has been documented in other reviews and in the global literature. For instance, Barajas-Nava et al. systematic review of 25 RCTs in health care models for adults aged 60 + with MCC reported that most of these interventions were focused on education, and mostly in the domain of patient-clinician communication, and that 57% were ineffective at providing any benefits.⁷⁹ Eriksen et al. assessed the effectiveness of 20 RCTs on health-related quality of life, mental health, and mortality.⁸⁰ The barriers to fully implementing or integrating various health care models globally have also been documented. Longhini et al. highlighted the need to assess caregivers-related outcomes and clearer description of interventions in organizational models of pediatric primary care worldwide.⁸³ Ludwick et al. underlined the readiness required, especially financial and resources, which is key for a successful implementation of community health worker-led integrated care in Ethiopia.⁸⁴ Pinter et al. described financial barriers, shortage of professionals, and lack of training as the main barriers to the implementation of integrated models of care in Asia, whereas financial incentives seemed to increase their success.⁸⁵ The authors also observed that performance assessments were rarely embedded in the models. Wilson et al. highlighted the very limited studies of health care models for persons with MCC in Canada, and the call for patient-centered guidelines of care, transdisciplinary care teams, and research.⁸⁶ Rohwer et al. pointed out the worldwide heterogeneity of integrated models, with inconsistent integration and measured outcomes.⁸⁷ These examples denote universal challenges revolving around the increasing prevalence and awareness of MCC, and invoke comprehensive rethinking of approaches to prevent, diagnose, and effectively treat MCC that both consider the cost-effective or cost-saving approaches for long-term sustainability.

Considering the complexities of health care delivery and population health, we highlight important future directions. Research is needed to better understand how to organize and deliver care for MCC to attain optimal health outcomes based on evidence-based care guidelines and patient-centered goals. However, the exclusion of persons with MCC from clinical trials precludes the generation of necessary evidence that would inform the optimal integration of new treatments into real-life clinical settings.^37,88,89 Health care models should be adapted to patients’ health care needs, personal goals, resources, and limitations. Examples of patient-centered goals or personal health outcomes may include what the patients hope to achieve, affordability and out-of-pocket expenses, lost wages, time burden for health care, physical and mental function and independence, well-being, life expectancy, social and occupational engagement, burden of multiple diagnostic and surveillance tests, procedures, or treatment, and the complexity of self-management tasks.⁹⁰ The lack of focus on reducing health disparities highlights an opportunity area for future development and tailoring of health care models. Examples include the creation of high-functioning interprofessional care teams that can effectively coordinate care within and between care settings and during care transitions,⁹¹ the integration of supportive personnel in a comprehensive and holistic care plan,⁹² and the development of algorithms, decision support tools, and telemedicine technology for use by clinicians, patients, and caregivers to engage in shared decision-making and communication.⁹³ Furthermore, understanding how best to implement these models across complex and diverse health care settings, including assessing its feasibility, affordability, and sustainability, would be critical for their long-term success and to have implications for policy and planning.⁹⁴ As noted, only three studies reported on clinical significance, and those studies did not achieve optimal outcomes for all patients in their study population. Methods and measures of determining delivery of optimal health care and achieving health outcomes for all are also needed when health care models are tested while also balancing the individual care needs for different patients.^95,96

Several limitations should be noted when interpreting our findings. First, despite a robust search strategy of peer-reviewed published articles that was inclusive of many chronic conditions, it is possible that some relevant articles were not identified or missed. While many articles were screened in this review, many failed to manage two or more chronic conditions, often focused on managing one condition among populations with multiple conditions. This strict inclusion criteria limited the scope of our findings, but also highlights the dearth of research. Additionally, ten studies included many chronic conditions in their inclusion criteria rather than focusing on the management of two or three conditions, which made the interpretation of findings for specific co-occurring conditions difficult. Lastly, at least half of the study population had to be from a NIH-designated population that experiences health disparities or be in a health care setting that served such population. However, not all articles clearly described their study population and as such, we only included articles if the population sociodemographic characteristics explicitly aligned with our search criteria. Despite these limitations, the literature on health care models for MCC is emerging,⁷⁶ and our scoping review is one of the first to highlight the opportunities to advance the science and can be used to inform future reviews with a formal quality evaluation.

While there has been a call for a fundamental change in the care for persons with MCC over the last several decades, health care delivery research is nascent for the management of multiple, coexisting chronic conditions, especially among populations that experience health disparities. Our ability to advance health care delivery research and population health is constrained when research studies do not reflect the real-world experiences of populations that experience health disparities and MCC. There is an opportunity for multilevel, multicomponent, innovative research to develop adapt, integrate, and implement evidence-based health care models or strategies for MCC to improve clinically significant health outcomes that align with patient goal needs.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

Open access funding provided by the National Institutes of Health. This article was prepared as part of the authors’ official duties as employees of the US Federal Government. The statements in this report are those of the authors and do not necessarily represent the official position of the National Institutes of Health, National Institute on Minority Health and Health Disparities, National Cancer Institute, or other federal agencies.

Declarations

Conflict of Interest

The authors declare that they do not have a conflict of interest.