The aim of the project is to explore if a complementary addition of a care manager for depression treatment in the primary care setting has positive effects concerning reduction of the patients' depressive symptoms as well as improvement of daily function and quality of life in a short- and long-term perspective compared with the traditional Swedish primary care organization. The aim is also to evaluate the care manager function in a health economic perspective.

The specific aims are to develop and evaluate an evidence-based care manager program for patients with mild to moderate depression and to compare the evidence-based care manager program to treatment as usual (TAU) in terms of their short- and long-term effects on symptom remission, treatment adherence, care satisfaction, and self-efficacy as well as to perform a cost - effectiveness analysis of the care manager program, i.e. to evaluate the effects of the program on patients' health in relation to the running costs of the program.

Methods The research questions will be tested in a randomised controlled trial with follow ups 3 and 6 months in the primary care setting in cooperation between universities of Gothenburg, Karolinska Institute and primary care centres (PCC) of regions/county councils of Västra Götaland, and Dalarna.

Design A cluster randomized controlled trial of two groups (intervention and control). Randomization will be at the level of the primary care centers (PCCs).

Population of study Patients aged 18 and up that visit one of 20 different urban and rural PCCs.

Randomization All PCCs in the participating regions without employed onsite mental health professionals are invited to participate in the study. Around twenty practices will be matched into 10 blocks by depression treatment practices, thereafter 1 practice from each block will be randomly assigned to receive the care manager intervention program.

Intervention At the intervention practices, a district nurse will apply around 15-25% of working time as care coordinator for management of care for the patients with depression. Before patient recruitment begins, physicians and the nurse/district nurse (care manager) will participate in sessions for training in providing clinical services (patient information and education on patient's preferred treatment, assessment of patient's depressive symptoms and regular telephone contacts, follow up visits etc).

Recruitment of patients and Data collection Patients aged 18 and up that visit primary health care centres and are diagnosed by the General Practitioner (GP) with light/medium depression disorder according to PRIME-MD (4). The patients will be monitored up to 6 months after inclusion in the study by an individual interview (first visit, 3 and 6 months).

All parameters will be assessed at baseline and 3 and 6 months; prescription of antidepressants, number of days of labour lost, socio-demographic and economic variables, alcohol consumption, physical activity, and ethnicity. BDI-II, EuroQoL-5D, Work Ability Instrument (WAI) Karasek Job strain Model , medication, other treatment (psychological, counselling, or other).

For somatic health reasons, also blood pressure and p-glucose will be monitored, and blood samples for bio-bank storage.

Outcome criteria Change in: depression symptoms (BDI-II), work ability index (WAI), Job strain (Karasek), EQ-5D, QALYs from baseline to 3 and 6 months. Amount of sick listing (days, %), days of labour lost.

Statistical analysis Analysis of covariance, controlling for baseline value, will be used to estimate the overall treatment effectiveness (difference in score) at final follow-up. Subgroup analysis by gender as well as age adjustments will also be performed.

Power calculation The primary variable is the level of depression (as measured by the BDI-II) and analyzed in an ANCOVA model with repeated measures. In order to detect an effect of 3 units in the difference between the two groups, with a power of 80% and a significance level of 10% (two-sided) about 200 patients is needed in each group (9). The underlying assumption is a standard deviation in the group of 10 units, a within-subject correlation of 0.4, and a within-cluster-correlation of 0.1, i.e. a design effect of 1.9 to correct for having a cluster analysis.

Cost-effectiveness analysis A cost- effectiveness analysis (CEA) with quality-adjusted life years (QALY) as measure of effect will be performed.

The region Västra Götaland has proclaimed interest in the study, and financially supports the study.

Inclusion Criteria:

• diagnosed mild-moderate depression,
• 18 - 80 years,
• attending primary care center

Exclusion Criteria:

• schizophrenia,
• abuse/addiction,
• psychosis,
• bipolar syndrome,
• suicidal ideation or earlier suicide attempt,
• ongoing deep depression,
• generalized anxiety syndrome,
• not communicable in Swedish