Integrated care (IC) models have continued to rise in popularity and utilization over the past several years. Specifically, there is an increased interest in using IC models to address common chronic diseases like diabetes and depression. There is, as a result, more Randomized Controlled Trials that have examined the impact of different types of IC models to address diabetes and depression. However, there are no meta-analytic reviews that summarize the existing research, synthesize the effects of IC models for depression and diabetes, and identify the next steps for research and practice. I recently sought to address these questions in a recent meta-analysis. Within this blog, I will review some of the highlights of the review. I will also identify some of the gaps regarding how the RCTs reported on the IC models and behavioral interventions used in their respective studies. I will then review a fidelity paper from a recent RCT that seeks to address the common gaps in reporting within integrated care RCTs.
Integrated Care Models Improve Outcomes for Co-Occurring Diabetes and Depressive Symptoms
People with diabetes are about twice as likely to develop depressive symptoms. Further, people with diabetes report that their diabetes care teams rarely (less than 50% of the time) assess their mental health generally or their depressive symptoms more specifically. When people present with diabetes and depressive symptoms and do not have these health concerns addressed simultaneously, their diabetes and depressive outcomes suffer. IC models provide a potential solution, but there is limited understanding of their comparative effectiveness. Our meta-analysis addresses this knowledge gap. We synthesized findings from 31 RCTs involving 8,843 participants to evaluate the impact of IC on glycemic control (HbA1c) and depressive symptoms.
Our results showed people have better outcomes for their diabetes and depression when receiving IC compared with usual medical treatment. Specifically, we found a moderate difference between groups where those receiving IC had greater reductions in HbA1c compared to those receiving usual care (d = −0.36, 95% CI −0.52 to −0.21). In addition, we found a large effect for depressive symptoms where those receiving integrated care had better outcomes (mean reduction of 0.72 points, 95% CI −1.15 to −0.28).
We were then interested in comparing whether there were differences in outcomes explained by the type of IC model, behavioral intervention, and treatment provider. We found that studies that used a combination of behavioral interventions—such as behavioral activation and cognitive behavioral therapy—demonstrated even greater improvements in glycemic control. However, there were no significant differences regarding the type of IC model used, behavioral intervention provided, or which type of professional delivered the care. This is encouraging as it promotes flexibility when selecting what kind of integrated care model one selects. In summary, integrated care works and we may consider shifting the conversation from “What model works best” to “Let’s make sure that we increase the implementation of IC!”
The Fidelity Challenge: Strengthening Intervention Reporting and Implementation
When performing the meta-analysis, I noticed that many studies had unclear reporting about what kind of IC model they were using, and how they were structuring their behavioral health visits. For example, several studies neglected to identify the type of IC model they were utilizing and how they were structuring their visits to align with the identified model. To mitigate this, we used resources to create checklists to assess what type of model was used in the RCT based on their description of the behavioral intervention, where it was performed, and the degree to which the interventionist collaborated with the medical team. To address this gap, we recently wrote a fidelity paper for our RCT that tested solution-focused brief interventions within a primary care behavioral health (PCBH) model.
Our study focused on providing a comprehensive strategy to monitor the fidelity of the solution-focused brief intervention while ensuring that the intervention adhered to the visit structure consistent with a PCBH model. We summarize the practical and feasible strategies that we used to include a solution-focused self-report screener for the behavioral interventionist to fill-out after visits, a survey that assesses growth in solution-focused core constructs, charting templates that include prompts consistent with the delivery of a solution-focused intervention within a PCBH context, and more. We hope that this fidelity paper can provide a template for researchers, clinicians, and administrators to enhance the reporting of IC models, and the behavioral interventions delivered within them.
Implications for Clinical Innovation and Research
Together, these studies point to several key takeaways for advancing integrated care:
1. Integrated Care Works and Should Be Scaled Up. Our meta-analysis demonstrated that IC models improve both glycemic control and depressive symptoms. Given that no single model, intervention type, or provider stood out as superior, the focus should shift from identifying the “best” IC model to increasing the implementation and accessibility of IC more broadly.
2. Flexibility in Implementation is Key. Since different IC models, behavioral interventions, and provider types yielded similar improvements, organizations have the flexibility to tailor IC approaches to their specific resources and populations. This adaptability supports wider dissemination of IC without the need for rigid adherence to a single model.
3. Clearer Reporting of IC Models is Needed. Our meta-analysis revealed substantial inconsistencies in how studies described their IC models and behavioral health interventions. Standardized reporting guidelines are necessary to enhance clarity in research and practice. Future RCTs should explicitly define their IC model, visit structure, and intervention components to facilitate replication and comparison across studies.
4. Fidelity Monitoring Must Become Routine. Our fidelity study highlights the need for structured intervention tracking to ensure that behavioral interventions are delivered as intended within IC models. We provide a practical framework—including interventionist self-report screeners, structured charting templates, and outcome surveys—that can be adopted in research and clinical practice. Embedding fidelity monitoring into routine workflows will enhance intervention consistency, support training efforts, and improve patient outcomes.
5. Bridging Research and Practice Requires Standardized Tools. Researchers and clinicians must collaborate to develop standardized measures for intervention reporting, fidelity tracking, and outcome assessment in IC. By creating practical tools that can be easily integrated into clinical workflows, we can ensure that research findings translate into real-world impact.
As integrated care continues to expand, these insights can guide the field toward more effective, scalable, and sustainable models. The next step is not just studying IC—but ensuring it is implemented well and equitably across healthcare settings.
Links to Fidelity Paper and Meta-Analysis
Here is the link to the fidelity paper if you are interested in reading the full text: https://link.springer.com/article/10.1007/s10880-025-10063-7
Below is a link to the full-text version of the meta-analysis, “Effects of Integrated Care Approaches to Address Co-occurring Depression and Diabetes: A Systematic Review and Meta-analysis.” If you are unable to access the full text, please email Zach Cooper at zach.cooper@uga.edu for assistance.
Share a Thought, Leave a Comment!
Leave a Reply