Background: The comorbidity of pain and depression is associated with high disease burden for patients in terms of disability, wellbeing, and use of medical care. Patients with major and minor depression often present themselves with pain to a general practitioner and recognition of depression in such cases is low, but evolving. Also, physical symptoms, including pain, in major depressive disorder, predict a poorer response to treatment. A multi-faceted, patient-tailored treatment programme, like collaborative care, is promising. However, treatment of chronic pain conditions in depressive patients has, so far, received limited attention in research. Cost effectiveness of an integrated approach of pain in depressed patients has not been studied.This article describes the aims and design of a study to evaluate effects and costs of collaborative care with the antidepressant duloxetine for patients with pain symptoms and a depressive disorder, compared to collaborative care with placebo and compared to duloxetine alone.Methods/Design: This study is a placebo controlled double blind, three armed randomized multi centre trial. Patients with (sub)chronic pain and a depressive disorder are randomized to either a) collaborative care with duloxetine, b) collaborative care with placebo or c) duloxetine alone. 189 completers are needed to attain sufficient power to show a clinically significant effect of 0.6 SD on the primary outcome measures (PHQ-9 score). Data on depression, anxiety, mental and physical health, medication adherence, medication tolerability, quality of life, patient-doctor relationship, coping, health resource use and productivity will be collected at baseline and after three, six, nine and twelve months.In the collaborative care conditions a) and b), a care-manager provides Problem Solving Treatment and integrated symptom management guidance with a self-help manual, monitors depressive and pain symptoms, and refers patients to a physiotherapist for treatment according to a 'Graded Activity' protocol. A psychiatrist provides duloxetine or placebo and pain medication according to algorithms, and also monitors pain and depressive symptoms. In condition c), the psychiatrist prescribes duloxetine without collaborative care. After 12 weeks, the patient is referred back to the general practitioner with a consultation letter, with information for further treatment of the patient.Discussion: This study enables us to show the value of a closely monitored integrated treatment model above usual pharmacological treatment. Furthermore, a comparison with a placebo arm enables us to evaluate effectiveness of duloxetine in this population in a real life setting. Also, this study will provide evidence-based treatments and tools for their implementation in practice. This will facilitate generalization and implementation of results of this study. Moreover, patients included in this study are screened for pain symptoms, differentiating between nociceptive and neuropathic pain. Therefore, pain relief can be thoroughly evaluated. Trial registration: NTR1089. © 2013 de Heer et al.; licensee BioMed Central Ltd.