Collaborative care is a complex intervention for treating people with chronic disease, developed by the originators of the Chronic Care Model (Wagner et al., 1996). Collaborative care has been shown to be significantly more effective for treating depression than usual care (Archer et al., 2012) and has been previously discussed by the Elves (for example see Jennifer’s 2014 blog and Ben’s 2016 blog).
Collaborative care involves a case manager who works with the GP and links up the patient with relevant health professionals and services while providing ongoing one-to-one support and therapy.
Collaborative care for depression in primary care has four major components (Gunn et al., 2006):
- A multiprofessional approach to patient care, with a GP and at least one other health professional (e.g., nurse, psychologist, psychiatrist)
- A structured management plan, including both therapy and medication
- Scheduled patient follow-ups
- Enhanced interprofessional communication, with regular discussion especially between mental health and primary care staff.
However, despite being widely recommended in national guidelines, implementation of collaborative care in routine practice is still rare because of limited resources, a lack of understanding of what it actually involves, and uncertainty regarding the key components.
This study by Schillok and colleagues (2025) builds on previous reviews of collaborative care by employing an Individual Participant Data (IPD) Meta-Analysis to address the last of these issues.
Collaborative care is a complex intervention for chronic disease that has been shown to be significantly more effective for depression than usual care. Yet, implementation in routine practice is rare.
Methods
To assess which components of collaborative care are the biggest drivers of its effectiveness in reducing symptoms of depression in primary care, the authors searched five databases alongside reference lists of systematic reviews to identify relevant studies. Studies were eligible for inclusion if they were randomised controlled trials (RCTs) or cluster RCTs that compared the effect of collaborative care with usual care in adults with depression in a primary care setting. Depression severity had to be measured on a continuous scale via self-report, which the authors standardised using the instrument’s cut-off value for mild depression.
Individual participant data were collected for demographic characteristics and depression outcomes (baseline and follow-up) from trial authors. Data was extracted using Gunn et al.’s (2006) framework, with a fifth component added (increased patient and family activation) which represents recent emphasis in primary care studies on patient and carer involvement. This model was included in the first analysis as the independent variable, but collinearity between two of the components (with a Pearson Correlation Coefficient of 0.61) caused suppression effects. Suppressors are variables that when added to a regression model change the original relationship between a predictor and the outcome – and can make it stronger, weaker or no longer significant. A Principle Components Analysis (PCA) was carried out to ensure highly correlated items were grouped together and the resulting new four component model (see below) was used in the final analysis, which employed linear mixed models with random nested effects.
Results
Out of 74 eligible studies, data were successfully collected from 35 RCTs which included 20,046 participants (57.3% of all eligible) and 38 comparisons. The majority of participants were female (n = 13,709) with a mean age 50.8 years (SD = 16.5, range = 18-95 years).
Course of depression over time
The treatment group receiving collaborative care showed a reduction in depression severity compared with usual care at each timepoint, which persisted for the first 24 months. A corresponding forest plot for the overall effect at 4-6 months revealed a small significant effect over usual care (standardised mean difference [SMD] = -0.20, 95% CI [-0.26 to -0.15], I2 = 58.4%).
Association between depression outcome and intervention components
The components of collaborative care after the PCA were:
- Patient-centred care (e.g., consideration of patient preference, goal setting, involvement of community or cultural background)
- Therapeutic treatment strategy (e.g., manual-based therapy, involvement of family/friends, counselling)
- Measurement-based care (e.g., ad-hoc emergency communication, shared medical record)
- Integrated mental health care (e.g., case manager with mental health background, relapse prevention plan).
A significant interaction effect with the largest effect size was found between depression outcomes and the collaborative care component ‘therapeutic treatment strategy’ (effect size = -0.07, p <.001 this="" was="" the="" most="" effective="" component="" of="" collaborative="" care="" interventions="" however="">significant interactions were found for all other components, with smaller effect sizes (patient centred care: -0.04, p = .03; measurement-based care: -0.04, p = .02; integrated mental health care: -0.04, p = .01).
Across 35 randomised controlled trials with 20,000+ participants, patients receiving collaborative care consistently showed greater reductions in depression severity compared to care at usual over 24 months.
Conclusions
This IPD meta-analysis found that:
The derived collaborative care component labeled therapeutic treatment strategy, including its main items manual based-therapy and involvement of family or friends, was the most effective for reducing depression severity in primary care.
The other collaborative care components contributed to this reduction, too, and may potentially interact with each other. As such, the authors concluded that all components should be implemented to some extent when using collaborative care in this population, but that the elements of ‘therapeutic treatment strategy’ warrant the prime focus, as they have the greatest potential for impacting success.
‘Therapeutic treatment strategy’, including manual based-therapy and involvement of family or friends, was the most effective component of collaborative care for reducing depression severity in primary care.
Strengths and limitations
According to the authors, this is:
the most rigorous methodological examination of the most effective collaborative care components to date and the first attempt to synthesise these data also incorporating IPD.
I agree with the first part of that statement but note that some of the authors have previously used IPD to examine the relationship between chronic physical conditions and the effectiveness of collaborative care for depression (Panagioti et al., 2016).
Initially they used a modified version of the widely accepted model of collaborative care, but followed this with an empirical model based on actual data and derived from PCA (Principle Components Analysis). They checked for publication bias and used comprehensive literature searches with strict inclusion criteria to mitigate study selection bias.
However:
- Only 57% of individual participant data was available across the RCTs eligible for inclusion, which is below the recommended 80% target.
- Inconsistent reporting of what was included in ‘collaborative care’ led to problems in recording details of the studies and coding the presence/absence of components of collaborative care interventions.
- The framework used to characterise the collaborative care model doesn’t account for variables such as the doctor-patient relationship and relationships between collaborating clinicians, which could influence treatment outcomes (Ciechanowskiet al., 2006; Hammarberg et al., 2019).
- The data also doesn’t allow for fully disentangling the relationship between the specific components of the intervention and the intensity with which these were applied. For example, how frequent was supervision with the mental health specialist, if provided, and did this have an impact on outcomes?
- More discussion and rationale for inclusion of the ‘increased patient and family activation’ component within collaborative care itself would also have been helpful, as this is a change to the widely accepted model used in most collaborative care research, even though its inclusion has been recommended (Menear et al., 2020). When interpreting the results of a review we need to understand exactly how and why such decisions were made.
This is the first attempt to identify the most effective components of collaborative care for depression using Individual Participant Data, providing guidance for study design and implementation.
Implications for practice
The primary finding that ‘therapeutic treatment strategy’ is the most effective component of collaborative care aligns with a previous meta-analysis that identified psychological interventions as the significant indicator of improved outcomes in collaborative care (Coventry et al., 2014). Based on this evidence, it seems that if you want to implement collaborative care effectively within services, it is essential to ensure this component is as streamlined as possible, both in its design and execution. That is not to say that the other components of collaborative care should be disregarded, as they are also important; however, their individual effects are smaller, meaning that these components may have less of an impact on depression outcomes.
While this study represents an important addition to the literature, it isn’t able to answer questions about the impact of therapeutic and interprofessional relationships in collaborative care nor how the different components interact. Personally, as a mental health professional who has implemented collaborative care in routine practice in the NHS (as a local modification of Improving Access to Psychological Therapies [IAPT]), I’d say that both are likely to matter. If collaborative care has the best evidence base for improving outcomes for depression in primary care (Archer et al., 2012), its implementation has been patchy because it requires not only considerable cultural change but also major organisational, professional and community buy-in. That’s why developing those patient/service user and professional relationships are so important for making it work.
Despite the evidence for and widespread recommendation of collaborative care, NHS England has continued to pursue the IAPT (NHS Talking Therapies) approach, which focuses mainly on psychological therapy, and draws its evidence base from studies of CBT which were not (unlike those for collaborative care) carried out in primary care (Carey et al., 2025). Given the importance of the psychological therapy component in collaborative care, some would argue that implementing that alone, as in IAPT, is enough, but that would be to ignore the influence of the other components of the model. What this paper tells us is that it’s essential to get the balance right. However, it doesn’t conclude that this alone is sufficient to achieve the best outcomes. We still need to know more about the ways in which elements of the model interact and how to best implement it in practice, which requires not only quantitative but qualitative methods.
While this meta-analysis and the developed model are helpful, we still need to know more about how the different components of collaborative care interact with each other for its successful implementation in primary care.
Statement of interests
I have been involved in implementing collaborative care both in routine NHS care and also in research trials. I am a co-author on 3 of the trial papers included in the review and the two previous systematic reviews cited, and I was a Harkness Fellow with Ed Wagner.
Links
Primary paper
Schillok, H., Gensichen, J., Panagioti, M., Gunn, J., Junker, L., Lukaschek, K., … & POKAL Group. (2025). Effective components of collaborative care for depression in primary care: an individual participant data meta-analysis. JAMA Psychiatry.
Other references
Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, D., Gask, L., … & Coventry, P. (2012). Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews, (10).
Carey, M., Kerr-Gaffney, J., Strawbridge, R., Hieronymus, F., McCutcheon, R. A., Young, A. H., & Jauhar, S. (2025). Are cognitive behavioural therapy, cognitive therapy, and behavioural activation for depression effective in primary care? A systematic review and meta-analysis. Journal of Affective Disorders.
Ciechanowski, P. S., Russo, J. E., Katon, W. J., Von Korff, M., Simon, G. E., Lin, E. H., … & Young, B. A. (2006). The association of patient relationship style and outcomes in collaborative care treatment for depression in patients with diabetes. Medical Care, 44(3), 283-291.
Coventry, P. A., Hudson, J. L., Kontopantelis, E., Archer, J., Richards, D. A., Gilbody, S., … & Bower, P. (2014). Characteristics of effective collaborative care for treatment of depression: a systematic review and meta-regression of 74 randomised controlled trials. PloS One, 9(9), e108114.
Gunn, J., Diggens, J., Hegarty, K., & Blashki, G. (2006). A systematic review of complex system interventions designed to increase recovery from depression in primary care. BMC Health Services Research, 6(1), 88.
Hammarberg, S. A. W., Hange, D., André, M., Udo, C., Svenningsson, I., Björkelund, C., … & Westman, J. (2019). Care managers can be useful for patients with depression but their role must be clear: a qualitative study of GPs’ experiences. Scandinavian Journal of Primary Health Care, 37(3), 273-282.
Hannigan, B. (2018). Collaborative care for depression: acceptable, effective and affordable. The Mental Elf.
Laidlaw, J. (2014). Collaborative care for adolescent depression: new RCT shows promise. The Mental Elf.
Menear, M., Dugas, M., Careau, E., Chouinard, M. C., Dogba, M. J., Gagnon, M. P., … & Légaré, F. (2020). Strategies for engaging patients and families in collaborative care programs for depression and anxiety disorders: A systematic review. Journal of Affective Disorders, 263, 528-539.
Panagioti, M., Bower, P., Kontopantelis, E., Lovell, K., Gilbody, S., Waheed, W., … & Coventry, P. A. (2016). Association between chronic physical conditions and the effectiveness of collaborative care for depression: an individual participant data meta-analysis. JAMA Psychiatry, 73(9), 978-989.
Wagner, E. H., Austin, B. T., & Von Korff, M. (1996). Organizing care for patients with chronic illness. The Milbank Quarterly, 511-544.
