Evaluating the Benefits and Affordability of a Program to Improve the Care of Common Mental Disorders in Primary Care

The goal of the research is to evaluate the clinical and cost-effectiveness of a Collaborative Stepped Care intervention for the treatment of Common Mental Disorders (CMD) in primary care in Goa, India.

The Collaborative Stepped Care intervention will be compared with Enhanced Usual care in a cluster randomized controlled trial in 24 primary care facilities; the first phase of the trial will be located in 12 government run Primary Health Centres, while the second phase will be located in 12 private family physician/general practice facilities. Approximately 2400 participants who screen positive on a screening questionnaire for CMD will be recruited from the 24 facilities (a harmonic mean of 100 per cluster per arm) of whom 2/3 will be ICD10 cases confirmed by a diagnostic interview (the Revised Clinical Interview Schedule). Follow-up interviews at home will be conducted by an independent team at 2, 6 and 12 months post recruitment. Allowing for 25% attrition during follow-up gives 650 patients in each arm with a confirmed CMD, or a total of 960 in each arm, contributing to analyses of the primary outcomes (see below).

The Collaborative Stepped Care intervention is a complex intervention comprising the following components: screening for detection of cases; psycho-education; antidepressants; inter-personal therapy; yoga; specialist supervision \& adherence management. The method of delivery will be based on the stepped-care model which provides treatments according to the patients' needs and incorporates strategies to overcome challenges to the integration of mental health in primary care in developing countries. This will be compared with Enhanced Usual Care which will comprise screening for detection of cases, provision of the diagnosis to the doctor, and making antidepressants available in the facility-all these components will be available in the Collaborative Stepped care facilities as well. The intervention will be delivered by two new, full-time, primary care human resources-the Health Assistant (in both arms) who will screen adult attenders; and in the Collaborative Stepped Care arms, the Health Counselor-who has been trained over 3 months, and a part-time Psychiatrist supporting the primary care team.

The primary analyses will be intention-to-treat, regardless of adherence to the intervention, and will be based on outcomes measured at 6 months. All analyses will take into account any within-facility clustering resulting from the cluster randomized design, and will include adjustment for any of the a-priori defined confounding factors for which randomization did not achieve balance between the two arms at baseline.

The primary research question being answered and the analysis outputs are:

• Does the Collaborative Stepped Care Intervention have an effect on the outcome of Definite Cases? Prevalence of CMD in the two arms amongst participants who had an International Classification of Diseases, tenth edition (ICD10) CMD at baseline
• Does the Collaborative Stepped Care Intervention have an effect on the burden of CMD amongst patients who are Possible Cases? Prevalence of CMD in the two arms in all screen-positive participants
• Does the Collaborative Stepped Care Intervention have an effect on the outcome of Depression Cases? Prevalence of CMD in the two arms amongst participants who had an ICD10 Depression at baseline
• Does the Collaborative Stepped Care Intervention have an effect on the incidence of ICD10 CMD amongst Sub-threshold Cases? Prevalence of CMD in the two arms in screen-positive participants who did not have an ICD10 CMD diagnosis at baseline.

Secondary analyses will include:

• Comparison of prevalences listed above at 2 months (as a measure of speed of recovery) and at 12 months (sustainability of effectiveness).
• Repeated measures analysis of CMD prevalence at 2, 6 and 12 months, to assess effect of the intervention over the whole 12 month follow-up period.
• Comparisons of Revised Clinical Interview Schedule (CISR) and General Health Questionnaire (GHQ) scores, and World Health Organisation Disability Assessment Schedule (WHO-DAS) scores as continuous outcomes (i) at each endpoint; and (ii) using repeated measures analyses combining information from the three endpoints.
• Interaction of gender, education and co-morbidity with intervention effects
• Interaction of facility type (private/public) with intervention effects.
• Interaction of baseline illness attribution (somatic/psychosocial) with intervention effects.
• Interaction of baseline severity of CISR scores with intervention effects
• Comparison of adverse events between the two arms at 2, 6 and 12 months.
• Per-protocol analysis:
• Adherence with minimum intervention (at least one meeting with the Health Counselor in the Collaborative Stepped Care arm)
• Adherence with optimal intervention-completion of intervention-i.e. discharge by the Health Counselor; or minimum therapeutic dose of either antidepressant (90% adherence over 90 days) or Interpersonal Psychotherapy (IPT) (6 sessions over 4 months).
• Dose response relationship with levels of adherence

For the economic analysis, health care costs and other patient- or family-borne costs will be computed and compared at 2, 6 and 12 months, and subsequently related to changes in health outcome: both the primary outcome measure of depression status and also Quality Adjusted Life Years [QALYs]. In the event that dominance is not shown, i.e. the intervention is more effective but the costs are also more than the usual care group, incremental cost-effectiveness ratios will be computed, together with their confidence intervals (using bootstrapping techniques to overcome expected skewness of cost data). Cost-effectiveness acceptability curves will also be derived in order to show the probability of any cost-effective advantages for the component interventions at a range of 'willingness to pay' threshold levels.

We will also carry out a nested qualitative evaluation with the objective is to assess the participants' perceptions of the quality of the care they received, their satisfaction with this care, the reasons why they did/did not take full advantage of the intervention offered, and the impact of their health problems on their daily lives, economic productivity and lives of family members. We will also explore their experience of being a trial participant (in particular, the experience of the informed consent procedure; reasons for participation; their understanding of randomization). Two in-depth interviews will be conducted with at least 96 participants, the first about 1 month after enrolment and the second interview about 6 months later. A second group of in-depth interviews will be carried out with members of the intervention team about the process of the intervention, its successes and limitations (on quality of care, on the PHC's overall function) and how future programs may address these. We aim to interview at least one doctor from each facility and one Health Counselor from each PHC (i.e. total of 24). Two in-depth interviews will be conducted with each staff member, the first about 2 months after the initiation of the program. The second interview will be conducted between 6 and 12 month after the initiation of the trial.