Addition of CBT to antidepressant drug therapy more than trebles the odds of response for those with treatment-resistant depression, a primary care study has found.¹

Overall 469 patients aged 18-75 were recruited from 73 general practices across the UK. Participants met the ICD-10 criteria for depression and had a Beck depression inventory (BDI) score ≥ 14 despite treatment with antidepressants for a minimum of six weeks. Just under half the participants had a chronic physical health problem.

Two hundred and thirty five patients were randomised to continue usual care and 234 patients were randomised to receive CBT in addition to usual care. The individual CBT intervention consisted of at least 12 sessions, each lasting 50-60 minutes.

Response to treatment was defined as a reduction in depressive symptoms of at least 50% compared with baseline. At six months follow-up, 95 patients (46%) from the intervention group had responded to treatment compared with 46 (22%) from the usual care group, odds ratio 3.26 (95% CI: 2.10-5.06; P < 0.001), NNT = 4.

NICE recommends that patients without a chronic condition who have moderate to severe depression should be given combined treatment with antidepressants and individual CBT. Moderately depressed patients with a chronic condition should be offered a high-intensity psychological intervention, preferably group CBT, and severely depressed patients should be offered both individual CBT and an antidepressant.^2,3

However, patients may be reluctant to commit to CBT, high-intensity psychotherapy resources are limited, and the additional benefit from combined treatment over antidepressants alone is modest (NNT = 8) and may result from greater adherence to treatment rather than a genuine therapeutic effect.⁴

Therefore, the authors of the present study suggest an alternative treatment strategy in which antidepressants are used as first-line treatment and CBT is reserved as an add-on therapy for those who have not responded to adequate treatment after 6-8 weeks.

Severity rating scores are only an approximate guide to depression severity. In my opinion, the boundary between mild and moderate depression roughly corresponds to a PHQ-9 score ≥ 15; the corresponding HADS and BDI scores are ≥ 11 and ≥ 20 respectively.⁵

A BDI score ≥ 14 roughly corresponds to the boundary of mild depression. However, average baseline BDI score was 31.8, which suggests that most participants had moderate to severe depression (BDI 20-28 = moderate, BDI 29-63 = severe).⁶

Unfortunately, this was a heterogeneous group of patients, most of whom had chronic depression of at least two years duration. Therefore, the study is not a reliable guide to the management of patients who have not responded to an initial 6-8 week course of antidepressant treatment.

Nor do we know whether the apparent benefit was due to the specific intervention or a non-specific response to therapy. The comparison was with usual care rather than optimal antidepressant therapy: ideally the control group should have been managed according to a protocol specifying when and how antidepressant treatment should be adjusted in non-responders.

Patients with severe depression should be offered combination treatment. For patients with moderate depression, there is no good evidence to favour one treatment strategy over another and I suggest we offer a choice of:

• antidepressants alone

• antidepressants with CBT augmentation for non-responders at 6-8 weeks

• CBT alone

• antidepressants plus CBT.