Duloxetine plus collaborative care treatment versus Duloxetine alone in the treatment of concomitant pain and depressive disorder. A randomized clinical trial in primary care.

Patients with depression often present with pain. The positive predictive value
for depressive disorder is 43 % for joints ache and 39% for backache (Gerber et
al., 1992). Pain is one of the associated symptoms of depressive disorder
(American Psychiatric Association, 1980). The percentage of chronic pain rises
from a non-depressed general population sample from minor depressed patients to
up to 70% in MDD patients (Ohayon & Schatzberg, 2003). Up to 85% of patients
with pain symptoms experience a depression (Bair, Robinson, Katon, & Kroenke,
2003;Kroenke, Spitzer, & Williams, 2001). An increased risk for depression in
patients with pain symptoms is found in groups without and with a somatic
disorder. Furthermore, the presence of a major depressive episode results in a
2-fold increase in painful physical symptoms, both in medically unexplained
and in medically explained conditions (Demyttenaere et al., 2006). Subjects
with pain symptoms have at least a 2-fold increased risk for depression
compared to subjects without pain (Bair et al., 2003) and the prevalence of
depression increases when the number of pain symptoms increases (Gureje et al.,
2007). The burden of pain and depression is high for patients (and doctors) in
terms of disability, wellbeing and use of medical care (Bair et al., 2003).

The primary objective of the study is to establish the effects of an integrated
intervention for depression and pain in primary care.

A two-armed cluster randomized trial with randomization between PCPs.
Randomization across PCPs will be used because the PCPs have a potential
influence on the outcome of the intervention. If subjects are randomly
assigned, there would be a possibility that transfer effects would influence
the outcome of the intervention (Van der Feltz-Cornelis & Ader, 2000).
Practices will be matched according to their location and size. It is an open
label study, single blinded that is the outcome assessment will be performed by
a blinded research assistant and the participants will not be aware if they are
in the experimental or control condition. Differences in the outcome measures
between the two conditions will be analyzed with Multi Level Analysis (MLA) to
correct for the clustered design. To control for possible skewness in the
randomized groups as far as distribution of confounders is concerned,
propensity scores will be calculated. Propensity scores are used to correct for
a possible randomization bias, they balance the observed covariates (Joffe &
Rosenbaum, 1999;Van der Feltz-Cornelis et al., 2006a). Possible confounders
such as age, gender, immigrant status, level of education, history of treatment
and life events will be taken as variables in the analysis. To assess the cost
effectiveness, we will apply a cost-utility analysis (CUA). The results will be
expressed as cost per Quality Adjusted Life Year (QALY). The economic
evaluation will be undertaken from a societal perspective. Hence, all relevant
effects and costs due to resource utilization within the healthcare (direct
medical costs) and costs due to production losses (productivity costs) will be
included. The study will take four years. Planned start of the study:
01.11.2007.

In the experimental arm, treatment with the antidepressant duloxetine will be
combined with psycho education in the form of a DVD provided by the PCP, a self
help manual containing a short (approximately 6-12 weeks) behavioral
intervention with focus on both pain symptoms and depression and including
relaxation techniques and exercise techniques, handed out by the care manager
(nurse practitioner) to the patient, and PST provided by the care manager. The
participating PCP and CM apply this integrated treatment of an antidepressant,
psycho-education, a self help manual and PST in a collaborative care model. The
patients progress will be monitored during the intervention by use of a numeric
rating scale for pain (NRS) and the PHQ9.
In the control condition, patients will only receive duloxetine.

The patients have to fill in a questionnaire at 4 different times. The patients
in both conditions will receive duloxetine daily. Furthermore, patients in the
intervention condition will receive a no-risc integrated treatment for pain and
depression.