The Chronic Care Model and Diabetes Management in US Primary Care Settings: A Systematic Review

1Diabetes is a major cause of heart disease and stroke among adults in the United States and is the leading cause of nontraumatic lower-extremity amputations, new cases of blindness, and kidney failure. In 2010, the Centers for Disease Control and Prevention reported that 25.6 million, or 11.3%, of US adults aged 20 or older had diagnosed or undiagnosed diabetes. Comprehensive models of care, such as the original Chronic Care Model (CCM), advocate for evidence-based health care system changes that meet the needs of growing numbers of people who have chronic disease. CCM was developed to provide patients with self-management skills and tracking systems. The model represents a well-rounded approach to restructuring medical care through partnerships between health systems and communities.

2CCM comprises 6 components that are hypothesized to affect functional and clinical outcomes associated with disease management. The 6 components are 1) health system — organization of health care (ie, providing leadership for securing resources and removing barriers to care), 2) self-management support (ie, facilitating skills-based learning and patient empowerment), 3) decision support (ie, providing guidance for implementing evidence-based care), 4) delivery system design (ie, coordinating care processes), 5) clinical information systems (ie, tracking progress through reporting outcomes to patients and providers), and 6) community resources and policies (ie, sustaining care by using community-based resources and public health policy).

3The sum of these CCM component parts are purported to create more effective health care delivery systems that institute mechanisms for decision support, link health care systems to community resources and policies, deliver comprehensive self-management support services for patients, and operate and manage patient-centered clinical information systems. Despite evidence indicating widespread application of CCM to multiple illnesses, such as diabetes, congestive heart failure, and asthma, no summative reviews have investigated how CCM has been applied in diabetes care. The objective of this study was to determine how CCM has been applied in US primary care settings to provide care for people who have diabetes and also to describe outcomes of CCM implementation.

4This study identified English-language peer-reviewed research articles describing CCM-based interventions for managing type 1 and type 2 diabetes in US primary care settings (ie, hospital-network outpatient clinics, private practices, and community health centers). We collected articles from the Cochrane database of systematic reviews by using 2 distinct searches for “chronic care model” and “diabet*,” which were combined by using the word “and.” We also collected articles via EBSCOhost from the CINAHL database and the Health Source: Nursing/Academic Edition database by using the Boolean phrase search function for “chronic care model” (and) “diabet*.” These databases are all repositories for original health science research studies. Each database was separately searched. We conducted our analysis in October 2011.

5Inclusion criteria specified that studies 1) be published after the formal inception of the original CCM (1999); 2) use the original CCM instead of the expanded CCM; and 3) describe CCM-based interventions to manage and treat diabetes in US primary care settings. We searched for articles published between January 1999 and October 2011. We excluded studies that took place outside of the United States, reported secondary data, or represented an editorial, commentary, or a literature review. We identified 155 studies (Figure) and reviewed them in 3 steps. First we reviewed the abstracts; 76 manuscripts met inclusion criteria, and 79 were excluded. We then reviewed the full articles; 43 articles were retained, and 33 were excluded. After additional review, we excluded 27 articles and retained 16 for data extraction.

6As recommended by the Centre for Reviews and Dissemination systematic review guidelines, we created the following categories to systematically assess the 16 studies and gain an understanding of the methods used and the outcomes associated with CCM application: study design, sample size, setting, participant demographics, primary and secondary outcomes measured, data collection instruments used, statistical tests used, and major findings. We determined which of the 6 CCM components had been applied to each intervention and how the component(s) had been applied. We then qualitatively assessed the outcomes of each component that was applied in each study. The study selection process was conducted by 1 author (K.D.) and then repeated by the same author to ensure accurate selection; any discrepancies were analyzed and resolved on the basis of the inclusion and exclusion criteria. The same author (K.D.) extracted the data. Another author (M.S.) critiqued the data to identify any inconsistencies between data presented in the studies and the data extracted for the review, posed questions for further clarification on all extracted content, and then reviewed and synthesized the extracted data for accurate presentation within the context of the CCM.

7The 16 studies (Table 1) included 9 randomized controlled trials, 2 prospective cohort studies, 3 natural experiments, 1 qualitative study, and 1 cross-sectional study. Study settings included academic-affiliated primary care practices, private practices, community health centers, safety net clinics, and a hospital. Only 6 of 16 studies implemented all 6 CCM components (Table 2). The studies focused primarily on people aged 50 to 70 years.

8Support from health care leaders stimulated organizational changes. Engaging the governing boards of health care systems resulted in support for institutionalizing the CCM approach, which was associated with HbA1c reductions of at least 1% during 12 months and improved foot care. Two studies revised the health care system to redefine health care team roles (eg, nurses, instead of PCPs, became responsible for conducting foot examinations). These changes improved the quality of diabetes care and rates of eye examinations, and were associated with improved HbA1c levels, blood pressure, cholesterol, and weight. Health system reorganization also helped to establish diabetes self-management training programs that identified and intervened with patients at risk for developing complications and improved clinical and behavioral outcomes.

9We found that diabetes self-management education (DSME) generally improved psychosocial and clinical outcomes in patients with diabetes. Twelve of 16 studies administered individual DSME sessions, and 9 studies administered group sessions using both group- and individual-level approaches. Facilitators, such as Certified Diabetes Educators (CDEs) or nurses, provided instruction on various topics, such as medication compliance, goal setting, foot care, and interpretation of laboratory results. Follow-up telephone calls allowed clinicians to monitor patient progress toward meeting diabetes-management goals that were set during individual office visits. For example, Schillinger et al found that weekly automated (prerecorded) tailored telephone calls from nurses were associated with improvements in interpersonal processes of care, physical activity and function, and slightly better metabolic outcomes (eg, HbA1c, blood pressure, cholesterol). Lyles et al found that the use of a secure e-mail connection and a smartphone to upload glucose readings via a wireless Bluetooth device allowed some participants to feel better connected with their nurse case manager. However, some participants found this communication system to be unstructured and preferred regular interaction (eg, face-to-face) with their nurse case manager; some participants found the smartphones to be frustrating because of technical difficulties associated with these unfamiliar technologies. Other studies reported that computer-based interactive diabetes self-management training modules and toolkits were supplemented by a “diabetes passport” or “diabetes care record” that listed goals, action plans, and laboratory results so that patients and providers could monitor performance and progress in diabetes care.

10Specialized decision support services for diabetes care were provided to PCPs (eg, endocrinologists) and nurse practitioners via telephone and e-mail, problem-based learning meetings, and telemedicine technology. Individual patient reports were also provided to health care teams for reviewing clinical trends (eg, HbA1c, blood pressure, lipids) and initiating clinical responses to laboratory results (eg, medication adjustments). Training PCPs on evidence-based guidelines and methods for implementing CCM resulted in improved PCP adherence to clinical guidelines, including the American Diabetes Association (ADA) Standards of Care and Institute for Clinical Systems Improvement (ICSI) Clinical Guidelines for Hypertension, Diabetes, and Hyperlipidemia. In several studies, this training was associated with improved diabetes knowledge among patients and improved levels of HbA1c and high-density lipoprotein (HDL) cholesterol.

11Implementation of ADA standards of care and ICSI clinical guidelines resulted in innovative diabetes care delivery in PCP offices. For example, ADA standards require that people with diabetes receive DSME to “optimize metabolic control, prevent and manage complications, and maximize quality of life in a cost-effective manner”. To address barriers to care, such as poor diabetes knowledge, low awareness of educational service accessibility, and lack of psychosocial support, PCPs streamlined DSME services by offering “diabetes days” and planned visits exclusively for people with diabetes. Instituting these programs in PCP offices allowed for better communication between CDEs, PCPs, and patients, which contributed to lower HbA1c levels; better adherence to medication and adjustment processes; and stronger support networks located in more personalized settings. One study even noted that providing DSME programs in PCP offices instead of hospital settings resulted in a 2- to 3-fold increase in the number of patients reached with diabetes education.

12Collaborative clinical information systems using disease registries and electronic medical records enabled multiple health care providers (eg, PCPs, nurse practitioners, nurses, CDEs, physician assistants, medical assistants) to review detailed reports on laboratory and examination results and identify lapses in diabetes care (eg, missed visits, laboratory appointments, and examinations). These systems helped patients and providers set self-management goals and review progress reports to determine whether patients met their predetermined goals. Improved tracking (ie, using electronic patient registries or electronic medical records) of individual health outcomes (eg, HbA1c trends) provided an expedient way to manage patient information and also improved provider responses (eg, medication adjustment) to clinical data. For example, the Medical Archival Retrieval System (MARS) stored data and generated robust reports for providers on laboratory results, visits, medications, health insurance, comorbid conditions, medical procedures, and billing charges. MARS also served as a tool for administrators to gauge fiscal outcomes associated with placing CDEs in primary care sites to deliver DSME.

13Seven studies specified strategies for using community resources and forming public policy. Collaborations between community leaders and physicians and between pharmaceutical companies and health plans led to support for PCP training sessions on how to use CCM for diabetes management. Hospital and PCP collaborations within the community, such as partnerships between the University of Pittsburgh Medical Center and western Pennsylvania community hospitals and PCP offices, provided greater access to funding, information systems, and administrative support for CCM implementation.

14The findings of these studies contribute to a qualitative understanding of the relationship between the application of CCM components and diabetes outcomes in US primary care settings. Although the original CCM has been critiqued for not adequately meeting the needs of diverse patient populations with diabetes, our systematic review supports the idea that CCM-based interventions are generally effective for managing diabetes in US primary care settings.

15One meta-analysis determined that no single component of the CCM was imperative for improved outcomes. However, it is important to determine the combination of components that will likely produce optimal patient and provider outcomes. Our review suggested that incorporating multiple components together in the same intervention can help facilitate better CCM implementation (eg, using the decision-support component to train providers on guidelines such as the ADA Standards of Care and using the delivery system design component to remodel the care delivery process to provide self-management support through DSME in PCP offices).

16In several studies, organizational leaders in health care systems initiated system-level reorganizations that facilitated more comprehensive and coordinated diabetes care. Changing staff roles and responsibilities to more efficiently treat diabetes was 1 strategy that produced clinical benefits. Reorganized care can also support better training programs for patients to help them self-manage diabetes. Future system-level CCM reorganizations should create clear access points for providers to intervene with patients who are at risk for diabetes complications. Some organizations have already begun to do so. For example, the Rockwood Clinic Foundation revised its mission statement to include fundraising for research and development in new methods of chronic care delivery, which has resulted in increased funding for training materials, glucometers, blood pressure monitors, and laboratories.

17In several studies, providing administrative support to train PCPs in implementing evidence-based care was associated with improved patient engagement that led to positive health outcomes. Future studies should examine the effects of continuing education for ADA Standards of Care and ICSI clinical guidelines on CCM decision support among providers. It is important to determine whether provider training delivered through telecommunication and distance learning technologies can provide ample decision-support training to PCPs. Another area worth investigating is whether the longitudinal use of decision support in different primary care practice settings (eg, private practices, community health centers, hospitals) improves patient outcomes.

18Delivery system design was identified as an important strategy for integrating DSME into primary care settings through addressing patient barriers to care such as accessibility to DSME and availability of staff to assist with diabetes care. Our review supports the idea that DSME improves psychosocial and clinical outcomes. DSME fostered learning about proactive diabetes self-care practices and self-management skills. When ADA-accredited DSME occurs in primary care settings, PCPs are able to provide patients with personalized access to CDEs, who are likely funded through third-party health insurers. Offering DSME in primary care settings, rather than solely hospital settings, enhances the reach of such programs in a more intimate, socially supportive venue. Future DSME for primary care patients should continue to cover the ADA content areas for diabetes self-management, and strategies for delivering DSME should be evaluated by assessing the comparative effectiveness of group- and individual-level DSME approaches.

19Only 1 study in our review conducted weekly, skill-based learning sessions for racial/ethnic minority groups on healthful cooking modifications for traditional foods and snacks. This type of culturally appropriate self-management support was associated with a greater number of participants who had an HbA1c measurement of less than 7% and a fewer number of participants who had an HbA1c measurement of greater than 10%. Other culturally tailored non-CCM interventions have demonstrated larger absolute reductions in HbA1c than nontailored interventions. Given the large number of racial/ethnic minority populations in the United States who are at high risk for type 2 diabetes (eg, African Americans, Hispanics, American Indians, Asian Americans, Pacific Islanders), future research should focus on culturally tailored DSME in primary care settings. Cultural factors (eg, food preparation, views of illness) should be considered when designing, implementing, and evaluating DSME for these underserved groups. It is also noteworthy that none of the reviewed studies addressed the needs of pediatric patients diagnosed with either type 1 or type 2 diabetes. Diabetes is becoming more common in children and adolescents; Rapley and Davidson have advocated for the adoption of CCM programs aimed at adolescent patients with diabetes to help bridge the gap between pediatric and adult care.

20More personalized, patient-centered interactions (eg. individual office visits) help patients and providers set behavioral and clinical goals that can be monitored through clinical information systems. Many studies used disease registries and electronic medical records to establish patient goals, monitor patient progress, and determine lapses in patient care. Assimilating clinical information systems into user-friendly, portable digital technologies (ie, smartphones, iPads) may enable patients and providers to view and respond to laboratory results more regularly. For older populations of chronic disease patients (the age group sampled in most of the reviewed studies), training programs on the use of digital technologies for diabetes self-management may reduce the anxiety and barriers to access that may currently exist. Involving patients in exploratory focus groups to inform the development of assistive technologies can customize educational technology and address usability concerns among unique patient populations. Future studies on diabetes self-management support within the broader CCM framework should attempt to refine the use of information and communications technologies to empower, engage, and educate patients.

21Finally, community-level partnerships pooled human and fiscal resources to provide diabetes management services. However, strategies for using community resources and developing policies were described in only 7 studies. A meta-analysis also found that few studies addressed the community resources and policies component of CCM. More public-private partnerships need to be developed between providers and community organizations to address barriers to care and explore culturally appropriate community-based services (eg, cooking classes, exercise programs, nutrition counseling, self-monitoring assistance) for underserved populations and neighborhoods. Other models have sought to improve the community resources and policies component of the CCM. The Innovative Care for Chronic Conditions (ICCC) model espoused by the World Health Organization is comparable to the Expanded Chronic Care Model proposed by Barr and colleagues; it introduces prevention efforts, social determinants of health, and enhanced community participation as core components of chronic disease care. The ICCC has a larger focus on supporting “positive policy environments” (ie, partnerships, legislative frameworks, human resource allocation, leadership, and financing) in community and health care organizations. Future studies should investigate how different derivations of CCM components contribute to changes in diabetes care within primary care settings.

22This study had several limitations. We used only a few search terms, so all relevant studies may not have been identified. Only 1 person selected the studies for inclusion in our review. Future studies should use the multiple-rater approach for study selection and data extraction as outlined by the Centre for Reviews and Dissemination systematic review guidelines. We did not conduct a meta-analysis because we did not have access to primary data, and the variability in study design did not allow us to pool data. Future research could include a meta-analysis of data from randomized controlled trials to evaluate the methodological quality of quantitative studies that have tested the effectiveness of CCM for managing diabetes.

23In conclusion, our study provides evidence that CCM is effective in improving the health of people who have diabetes and receive care in primary care settings. The model accounts for health services at various levels in the diabetes care process. Positive clinical outcomes have been cited as indicators of CCM’s success in diabetes management. Far less emphasis has been placed on measuring the process outcomes of CCM that help lead to functional and clinical improvements. Process outcomes (eg, self-efficacy for disease management and clinical decision making, perceived social support, knowledge of diabetes self-care practices) are all indicators that need to be assessed. These assessments could enable health care administrators and professionals to determine how CCM could become further integrated into primary health care initiatives in diabetes.