Background: There is sufficient research evidence in favour of cognitive therapy in western world. However, only limited research has been carried out on its effectiveness in other countries. It is suggested that adaptations in content, format and delivery are needed before CBT can be employed in non-western cultures. We describe a preliminary evaluation of culturally adapted CBT for depression in Pakistan. Aims: We aimed to evaluate the efficacy of this culturally adapted CBT using a therapist manual. Method: In a randomized controlled trial we compared combination of CBT and antidepressants with antidepressants alone (treatment as usual) in primary care. Referred patients with ICD-10 diagnosis of depression were invited to participate and randomized to the intervention and control groups. Hospital Anxiety and Depression Scale (HADS) and Bradford Somatic Inventory (BSI) were used to measure changes in depression, anxiety and somatic symptoms. Results: Seventeen patients each were randomized to each arms of the trial. Except for financial status there were no differences between the two groups on various demographic variables. Patients receiving CBT showed statistically significant improvement on measures of depression (p < .001), anxiety (p < .001) and somatic symptoms (p < .000) as compared to antidepressant alone group. 82% patients attended six or more sessions of therapy. Conclusions: A culturally sensitive manualized CBT was effective in reducing symptoms of depression and anxiety in Pakistan.