Collaborative care for depression

Depression and chronic physical illness are both serious health problems, but it is their combination that is associated with an even worse health status.

For the management of chronic physical illness, recent advances are based on the Chronic Care Model, which promotes a more proactive, planned and population-based approach to disease management. Out of chronic care models, “Collaborative Care” (CC) is one of the most promising approaches and it typically includes a multi-professional approach, structured management, planned patient follow-ups and augmented inter-professional communication.

However, CC is a complex intervention and there is significant variation in its exact implementation across trials, as well as among the patient populations and contexts in which it was implemented. Just last week there was a Mental Elf blog by Jennifer Laidlaw discussing a promising trial of CC for adolescent depression published in JAMA.

In a recent meta-analysis in PLOS One, Coventry and collaborators looked at an updated data set of randomised controlled trials (RCT) of collaborative care with the goal of identifying factors associated with improvement in patient outcomes (this time focusing on adults) and in the process of care (Coventry et al, 2014).

Collaborative care often involves a medical doctor, a case manager (with training in depression), and a mental health specialist such as a psychiatrist. (The cheesy smiles and folded arms are very much optional).

Methods

The authors included studies that were RCTs or clustered RCTs of collaborative care delivered in primary care settings or community settings. Studies had to be conducted on adults with a primary diagnosis of depression or mixed anxiety and depressive disorder, according to an observer interview or validated self-report measures, and no restrictions were made about the severity or chronicity of depression. Finally, studies had to compare the effectiveness of collaborative care with standard or enhanced usual care.

The authors looked at two types of outcomes:

Change in self-reported or observer-rated depression scores, AND/OR
Change in the use of antidepressant medication, based on self-report or administrative records data

Ten theoretically plausible covariates that could influence the effectiveness of collaborative care on both depressive symptoms and medication use were defined a priori:

Country (US or non-US)
Recruitment (systematic identification or referral by clinicians)
Patient sample (antidepressant medication part of inclusion criteria or not)
Chronic physical health condition (present or absent)
Case manager professional background (mental health professional or not)
Intervention content (medication management versus psychological or both)
Number of sessions
Supervision frequency (ad hoc, scheduled or not applicable)
Enhanced usual care
Allocation concealment (low risk of bias or high risk of bias)

Random effects meta-regression was used to estimate regression coefficients with 95% confidence intervals (CI) for the relationship between study level covariates and depressive symptoms, and antidepressant use.

This review investigated changes in self-reported or observer-rated depression scores and antidepressant use (measured by self-report and administrative records data).

Results

74 trials met the inclusion criteria and were included in the meta-regression analysis. Out of these, 84 comparisons had data on depressive symptoms, and 59 comparisons had data on antidepressant use.

Only 29% of the comparisons included in the meta-regression were conducted outside the United States and only 4% were conducted in low to middle income countries.
Compared with usual care, collaborative care was associated with:
- improvements in depressive symptoms (standardized mean difference/SMD= -0.28, 95% CI -0.33 to -0.23)
- increased antidepressant use (relative risk/RR= 1.53, 95% 1.40 to 1.68)
For depressive symptoms, only psychological interventions remained a significant prediction: studies including psychological interventions (alone or combined with medication) reported greater improvements in depressive outcomes than studies that included only medication management.
For antidepressant use, systematic recruitment of participants (RR= 1.43, 95% CI 1.12 to 1.81) and the inclusion of participants with a chronic physical condition (RR= 1.32, 95% CI 1.05 to 1.65) were significant predictors: participants who were identified systematically and those with a chronic physical conditions adhered more to their antidepressant medications.
However, increased antidepressant use was not associated with improvement in depressive symptoms

Psychological interventions (alone or combined with medication) resulted in greater improvements in depressive outcomes

Conclusion

The authors concluded that:

Overall, collaborative care successfully improves both patient outcomes and the process of care for depression.

More specifically, they also emphasize that:

Our findings show that structured management plans that included psychological interventions either as a standalone therapy or in combination with antidepressant medication predicted reductions in depressive symptoms more so than collaborative care that only offered patients antidepressant medication.

Importantly, individual patient meta-analysis (i.e., using individual rather than study-level data) would be necessary to increase the possibility of detecting differential treatment effects across participants in trials and to allow for more complex modeling for the associations between treatment effects and characteristics of the patients, study design or trial setting.

Antidepressant use was not associated with improvement in symptoms.

Limitations

The authors were not able to include demographic variables in the regression models due to a lack of variability across studies.
Also, they were not able to model dose response relationships between treatment effects and case management sessions, as most trials did not report data about the frequency, intensity or duration of psychological treatments.
Depression severity was inconsistently reported in the included trials, which prevented the authors from conducting sensitivity analysis on severely depressed patients, which are known to be more responsive to psychological and pharmacological interventions.
Most included trials only reported short-term follow-up and this limited the possibility of examining moderators for the long term effectiveness of collaborative care.