Current issues of ACP Journal Club are published in Annals of Internal Medicine


Psychological therapy reduced depression earlier (4 months) but at 1 year was not better than usual general-practitioner care


ACP J Club. 2001 Jul-Aug;135:28. doi:10.7326/ACPJC-2001-135-1-028

Source Citation

Ward E, King M, Lloyd M, et al. Randomised controlled trial of non-directive counselling, cognitive-behaviour therapy, and usual general practitioner care for patients with depression. I: Clinical effectiveness. BMJ. 2000 Dec 2;321:1383-8. [PubMed ID: 11099284]



In patients with depression, is psychological therapy (nondirective counseling or cognitive behavioral therapy [CBT]) more effective than usual general-practitioner (GP) care?


Randomized (allocation concealed*), unblinded,* controlled trial with 1-year follow-up. This abstract reports the results of the randomized 3-way comparison only (197 of 464 patients).


13 general practices in northern London and 11 practices in greater Manchester, England.


197 patients who were ≥ 18 years of age (mean age 37 y, 77% women) and were depressed or depressed and anxious (score ≥ 14 on Beck Depression Inventory [BDI]). Exclusion criteria were serious suicidal intent, psychological therapy in previous 6 months, use of antidepressants, restricted mobility, organic brain syndromes, or inability to complete questionnaires. Follow-up was 91% at 4 months and 84% at 1 year.


Patients were allocated to nondirective counseling (n = 67), CBT (n = 63), or usual GP care (n = 67). Psychological therapy was given in the GP's office according to a manual developed for each group. Non-directive counseling was based on the work of Rogers†, and CBT involved problem formulation and staged intervention. Patients were initially offered 6 sessions (maximum of 12). Patients in the CBT group had a mean of 5.0 (SD 3.5) sessions, and 9 patients (14%) did not attend any sessions; in the nondirective-counseling group, patients had a mean of 6.4 (SD 4.2) sessions, and 7 patients (11%) did not attend any sessions.

Main outcome measure

Depression (score on the BDI).

Main results

Analysis was by intention to treat. At 4 months, patients in the psychological therapy groups had greater reductions in BDI scores than did those in the usual-GP-care group {mean score decreases 4.5, 95% CI 0.7 to 8.3 for CBT; and 5.7, CI 2.1 to 9.3 for nondirective counseling}‡; at 12 months, groups no longer differed {mean score differences -0.9, CI -4.2 to 2.2 for CBT vs usual-GP care; and 0.9, CI -2.4 to 4.2 for nondirective counseling vs usual-GP care}‡ (Table).


In patients with depression, psychological therapy was better than usual general practitioner care for reducing depression at 4 months, but the difference no longer existed at 1 year.

*See Glossary.

Rogers CR. On Becoming a Person: A Therapist's View of Psychotherapy. Boston: Houghton Mifflin; 1961.

‡Mean differences and CIs calculated from data in article.

Source of funding: NHS Executive, Health Technology Assessment Programme.

For correspondence: Professor M.B. King, Department of Psychiatry and Behavioural Sciences, Royal Free and University College Medical School, University College London, London NW3 2PF, England, UK. FAX 44-20-7830-2808.

Table. Nondirective counseling or cognitive behavioral therapy (CBT) vs usual general-practitioner (GP) care for depression§

Follow-up Mean BDI score (SD)
CBT Counseling GP care
Baseline 27.6 (8.4) 25.4 (8.6) 26.5 (8.9)
4 months 12.7 (9.5) 11.5 (7.7) 17.2 (11.9)
12 months 9.3 (8.8) 11.1 (9.3) 10.2 (8.5)

§BDI = Beck Depression Inventory.


The study by Ward and colleagues shows more clearly than ever that patients with depression want “talking treatments,” or psychotherapies. The study used a patient-preference design, which meant eligible participants had a choice of being randomly allocated to 1 of the 3 treatment conditions or, if they had a strong preference, to be given one of these without being randomly allocated but still providing data. Almost no one opted for usual care, and although participants did not show a great preference for either CBT or counseling, they used the preference group to avoid receiving usual care.

The popularity of psychological approaches is reflected by the proliferation of nondirective counseling in U.K. primary care. So far this growth has taken place in a vacuum of evidence for its effectiveness, indicating that in a patient-centered health service, lack of evidence is no obstacle to provision (1). This important and well-conducted trial shows that for patients with depression—most of whom had moderately severe symptoms (the mean BDI score was approximately 25 to 28)—nondirective counseling is as effective as the more established treatment of CBT and both are more effective, at least in the short term, than usual care. The investigators speculate on why they found an effect for counseling when another study (2) had shown no benefits. They suggest that it was their inclusion criteria, which indicated that the patient had to have substantial symptoms of depression, whereas other trials had used such open inclusion criteria that many participants scored too low on measures of depression to show any benefit.

Matthew Hotopf, MBBS, PhD
Guy's, King's and St. Thomas' School of Medicine
London, England, UK


1. Wessely S. The rise of counselling and the return of alienism. BMJ. 1996;313:158-60.

2. Friedli K, King MB, Lloyd M, Horder J. Randomised controlled assessment of non-directive psychotherapy versus routine general-practitioner care. Lancet. 1997;350:1662-5.