International Journal of Neuropsychopharmacology (2007), 10, 131–136. Copyright f 2006 CINP doi:10.1017/S1461145706006912

Cognitive therapy in relapse prevention in depression

S P EC I A L S EC T I O N

CINP

Eugene S. Paykel Department of Psychiatry, University of Cambridge, Cambridge, UK

Abstract

Received 14 October 2005; Reviewed 5 November 2005; Revised 9 November 2005; Accepted 13 November 2005; First published online 20 June 2006 Key words : Cognitive therapy, depression, relapse.

The therapeutic problem It has been recognized for some time that in spite of the efficacy of antidepressants in acute treatment, the longer term outcome of depression is problematic. Major follow-up studies published in the 1980s (Keller et al., 1984; Kiloh et al., 1988; Lee and Murray, 1988) showed high rates of relapse and recurrence. Initially it was possible to view these findings as reflecting a failure to apply routine continuation of antidepressant for 6 months to a year after acute treatment, and long-term maintenance of medication in some patients. However, it has become apparent more recently that the problem still remains despite full recognition of the need for longer term medication, and its reasonably adequate delivery. A series of studies from Cambridge illustrates the problem. In 1990 we began a prospective follow-up of depressed patients receiving psychiatric treatment,

Address for correspondence : Professor E. S. Paykel, Department of Psychiatry, University of Cambridge, Douglas House, 18E Trumpington Road, Cambridge CB2 2AH, UK. Tel.: 44-1223-741930 Fax : 44-1223-741929 E-mail : [email protected] Presented in the session on Psychological Treatments Combined with Pharmacotherapy in the CINP Presidential Symposium at the Regional Meeting of the CINP, Brisbane, Australia, December 2005.

mainly as hospital in-patients. One of the main aims was to determine whether longer outcome was still poor with modern treatment. Short-term outcome was good. The large majority of subjects (87 %), had achieved full remission by 15 months, and 94 % had achieved at least partial remission (Ramana et al., 1995). However, in the next 15 months, 40 % of those who had achieved at least partial remission, relapsed to major depression. These findings were very similar to those of the NIMH collaborative study (Keller et al., 1984), more than a decade earlier. There was one strong prognostic feature, presence of residual symptoms at remission. Among subjects scoring o8 on the Hamilton Depression Scale (HAMD) at remission, 76 % relapsed in the following 15 months, while among those with lower scores, only 25 % did so (Paykel et al., 1995). The high risk of relapse when remission was partial rather than complete had been recognized earlier, but was particularly striking in this study, and was important as pointing to a target patient group for relapse reduction. There was also a hint that failure to receive adequate medication was not the prime cause either of residual symptoms or relapse. Although by the time of relapse some patients were off medication, antidepressant treatment doses at the time of remission in subjects

Downloaded from http://ijnp.oxfordjournals.org/ by guest on January 15, 2017

This paper reviews recent advances in application of cognitive therapy (CBT) to a therapeutic problem in depression. Modern follow-up studies indicate that, in spite of the efficacy of pharmacotherapy, relapse and recurrence rates in some depressed patients remain high. This does not appear mainly due to failure to receive medication, but to reflect intractability of the disorder. In acute treatment, psychological treatments, although beneficial, are less cost-effective than antidepressants, due to high costs of therapists. Benefit which lasts longer, particularly if combined with medication, may therefore be particularly valuable. There have now been seven randomized controlled trials of cognitive therapy designed specifically to test relapse and recurrence prevention. All have shown significant benefit, which lasts beyond the cessation of therapy. The effect appears to be more on preventing symptom return than on lessening current symptoms, to summate well with continuation and maintenance antidepressant, and not to be due simply to enhanced medication adherence. Incorporation into routine clinical practice is now appropriate and recommendations are proposed.

132

E. S. Paykel

Cognitive therapy in relapse prevention There exists, therefore, a situation in which some patients with depression are highly prone to relapse and recurrence, in spite of good treatment with medication. This raises the obvious question as to whether a psychological treatment could be of help. In the last

10 yr, cognitive therapy has emerged as a strong candidate in this situation. Cognitive therapy as a treatment approach is now about 30 yr old (Beck, 2005). There is good evidence that it is effective in milder depression, and of comparable efficacy to antidepressants (Churchill et al., 2001), but less evidence that it is effective in severe depression. It is costly, due particularly to the cost of therapist time, for multiple sessions (Scott et al., 2003). It also requires a patient time commitment. Moreover in most countries there is a shortage of trained therapists. Although there is promising evidence for brief approaches and computerized adaptations (Proudfoot et al., 2003) it is still generally a second choice in acute treatment of adult depressives. If, however, benefit of preventing episodes extends over a longer time-frame after receipt of treatment and it can add to medication or be effective in those subjects who fail to benefit from medication, it clearly would have a useful therapeutic place. Evidence for benefit started to emerge in the 1980s. Follow-up studies of acute treatment trials comparing CBT with antidepressants showed significantly lower relapse rates in the groups which had received CBT (Blackburn et al., 1986; Evans et al., 1992; Simons et al., 1986) with non-significant results in three other studies (Kovacs et al., 1981; Miller et al., 1989; Shea et al., 1992). Such follow-up studies of acute trials cannot be conclusive (Paykel et al., 1999). First, a ‘differential sieve ’ may have occurred during acute treatment, whereby patients less likely to relapse responded to CBT. Second, antidepressant continuation was not always undertaken systematically or well controlled in these studies, and in one study where it was, differences in relapse rates between antidepressant and CBT groups were much reduced (Evans et al., 1992), as in a recent similar study in severe depression (Hollon et al., 2005). It really requires studies designed specifically for relapse reduction to test this effect. There have now been seven such trials, which are summarized in Table 1. The first studies were two by Fava and colleagues, one in residual depression (Fava et al., 1994, 1996, 1998a) and one in recurrent depressives (Fava et al., 1998b, 2004). Both showed substantial difference in relapse rates at the time-points of maximum differences, shown in the table, and these were reasonably sustained after more lengthy followup to 6 yr. These studies were small, with only 20 patients in each treatment group. In both also, drug treatment was withdrawn at the time of commencement of CBT, a procedure likely to inflate relapse rates in the control groups. In ordinary treatment practice,

Downloaded from http://ijnp.oxfordjournals.org/ by guest on January 15, 2017

with and without residual symptoms, and in those relapsing and not relapsing, did not differ or show suggestive trends (Paykel et al., 1995; Ramana et al., 1995). The Cambridge findings have more recently been extended to 10 yr (Kennedy et al., 2003). Taking a criterion for full recovery, 92 % of patients recovered, the great majority within 2 yr. However, over the 10 yr, 67 % experienced a recurrence of major depression, half of these by 24 months after recovery. Moreover, a comparison with published follow-up studies for depressed patient cohorts from mid- to late-1970s (Mueller et al., 1999; Surtees and Barkley, 1994) and 1960s (Kiloh et al., 1988; Lee and Murray, 1988) suggested that there had been little change in recurrence rates (Kennedy et al., 2003). In addition there were high levels of inter-episode subthreshold symptoms (Kennedy et al., 2004). In order to ascertain whether the problem really was failure to deliver adequate medication, a separate further study was carried out (Ramana et al., 1999, 2003) focusing on aftercare received in the 18 months following discharge from hospital, by 104 subjects with depression. Detailed medication histories were obtained from subjects and from case records, and were compared with published guideline standards for continuation and maintenance. In these subjects, with severe recurrent disorder, both prescribed levels and medication adherence were surprisingly good. Moreover, the majority of failures to receive adequate medication were due to patient refusal or non-adherence rather than prescriber failure. It is well known that people are distrustful of antidepressants, as shown in general population attitude studies (Paykel et al., 1998). Overall, there were some deficiencies in receipt of medication, but they were not sufficient to explain high relapse rates, nor were they evident particularly in relapsing patients. The preponderance of evidence suggests that, while use of antidepressants and mood stabilizers could be improved, the principal reason for high relapse rates is the intractability of the disorder to currently available medications, in some patients. Antidepressants appear to be more effective in inducing remission and recovery than in preventing further episodes.

CBT in relapse prevention Table 1. Controlled trials of cognitive therapy (CBT) in prevention of relapse and recurrence Relapse rates*

Study

Control ( %)

6 yr

35

70

3.5 yr

29

47

40

6 yr

25

80

145 156 76 187

16 mo. 8 mo. 15 mo. 2 yr

40 10 39 49

66a 31 62a,b 54b,c

40 158

Followup

* For studies with further follow-up, relapse rates are shown at time of maximum differences. a Significant only where >2 previous episodes; b significant interaction with n of episodes; c significant only where >4 previous episodes.

antidepressants should be continued for 9–12 months after acute treatment. The Cambridge–Newcastle study (Paykel et al., 1999) is one of the largest to date, including 158 patients, all with current residual symptoms rated o8 on the 17-item HAMD and o9 on the Beck Depression Inventory, following DSM-III-R major depression in the previous 18 months. All were receiving antidepressants in moderately high doses (mean daily dose equivalent to 33 mg fluoxetine for those on SSRIs, or 186 mg/d amitriptyline for those on TCAs) and were required to remain on these doses throughout the trial treatment phase of 5 months and followup phase of a further 12 months. Subjects were randomized to receive clinical management, or clinical management plus 16 sessions of CBT over 20 months from trained cognitive therapists, with two subsequent booster sessions. The trial was thus targeted at subjects known first to be only partially responsive to antidepressant ; second, to be relapse prone by virtue of their residual symptoms ; third, already receiving moderately high antidepressant doses without adequate responses, so that CBT would be indicated if effective, and fourth, with CBT delivered after acute treatment, rather than given in acute treatment to all depressives, which would be impractical. The study showed moderate-sized treatment effects, with relapse

reduced from 47 % in the control group to 29 % in the CBT group (Paykel et al., 1999). Benefit was not due to any effects on drug adherence, which was good in both groups. Effects on achievement of full remission and on symptom ratings were small, suggesting a specific relapse effect. There were some effects on social adjustment (Paykel et al., 1999; Scott et al., 2000). A cost analysis showed higher costs in the CBT group amounting to about £4500 (UK pounds) per relapse avoided (Scott et al., 2003), due largely to cost of therapists, not a negligible expenditure but generally a worthwhile one. Follow-up has recently been extended to 6 yr after randomization (Paykel et al., 2005). Effects on relapse gradually diminished over time, but lasted until 312 yr after the end of the CBT. There was also some benefit on residual symptom levels in this analysis. These findings confirm an advantage of CBT over medication : effects last for some time after cessation of the treatment. This would also improve the cost-benefit balance, since costs during treatment can be held against benefit over a substantially longer period. The study by Jarrett et al. (2001) used a different design. All subjects received acute-phase CBT and unmedicated responders were randomized either to 8 months (10 sessions) continuation cognitive therapy or a control group. Continuation cognitive therapy significantly reduced the relapse rate. In post-hoc analyses the effect was confined to depressions with first onset prior to age 18 yr. Teasdale et al. (2000) used an innovative cognitive approach, ‘mindfulness-based ’ cognitive therapy. The studies reviewed above all commenced CBT while depressive symptoms were present since cognitive therapists need some negative cognitions to work with and these may be hard for the patient to recapture, after the depression has gone. Teasdale et al. (2000) employed a technique using minor dysphoric feelings in well subjects to re-experience the feelings of depression and the negative cognitions, and to teach the patient to become more aware of these and accept them as mental events rather than as aspects of the self or necessarily true reflections of reality. In a trial of this variant of CBT delivered to recovered depressives in groups, the rate of recurrences was reduced compared with the control group, but only in subjects with o3 previous episodes. Ma and Teasdale (2004) undertook a smaller replication study with similar results. This time there was a significant interaction with number of episodes. In the smaller portion of the sample with only two episodes, CBT was a little worse than treatment as usual, but in those with o3 episodes (who also tended to be earlier

Downloaded from http://ijnp.oxfordjournals.org/ by guest on January 15, 2017

Residual depression Fava et al. (1994, 1996, 1998a) Paykel et al. (1999, 2005) Recurrent depression Fava et al. (1998b, 2004) Teasdale et al. (2000) Jarrett et al. (2001) Ma and Teasdale (2004) Bockting et al. (2005)

CBT ( %)

n

133

134

E. S. Paykel

in onset and to have episodes unrelated to life events), it significantly reduced the recurrence rate. The most recent study is that by Bockting et al. (2005). CBT was delivered in groups to 187 recovered depressives, but only 165 were available for analysis. Again there was a significant interaction with number of episodes, with a slightly higher relapse rate in CBT than in treatment as usual for those with few episodes, but a strong beneficial effect (46 % vs. 72 %) for those with o5 previous episodes. Remaining questions

Acknowledgements I am grateful to the many colleagues with whom I have undertaken studies, as indicated in the cited references. These studies were carried out with the support of the UK Medical Research Council. The views expressed are the author’s own. Statement of Interest The author has received honoraria or travel support from a number of companies which are involved with antidepressants. This has not influenced the present paper. References Beck AT (2005). The current state of cognitive therapy: a 40-year retrospective. Archives of General Psychiatry 62, 953–959. Bockting CL, Schene AH, Spinhoven P, Koeter MW, Wouters LF, Huyser J, Kamphuis JH (2005). Preventing relapse/recurrence in recurrent depression with cognitive therapy: a randomized controlled trial. Journal of Consulting and Clinical Psychology 7, 647–657. Blackburn IM, Eunson KM, Bishop S (1986). A two year naturalistic follow up of depressed patients treated with cognitive therapy, pharmacotherapy and a combination of both. Journal of Affective Disorders 10, 67–75. Churchill R, Hunot V, Corney R, Knapp M, McGuire H, Tylee A, Wessely S (2001). A systematic review of controlled trials of the effectiveness and cost-effectiveness of brief psychological treatments for depression. Health Technology Assessment 5, no. 35. Evans MD, Hollon SD, De Rubeis RJ, Piasecki JM, Grove WM, Garvey MJ, Tuason VB (1992). Differential relapse following cognitive therapy and pharmacotherapy

Downloaded from http://ijnp.oxfordjournals.org/ by guest on January 15, 2017

The efficacy of CBT in reducing relapse and recurrence rates is now clear. Some further questions remain to be answered. One major issue is the mechanism of action. It is not via enhanced medication compliance. In the Fava studies medication was withdrawn, and in the Jarrett study patients were not on medication. In the Cambridge–Newcastle study medication compliance was good, and did not differ between groups. Also in the Cambridge–Newcastle study, a special complex analysis was undertaken to examine possible specific cognitive mechanisms, looking for what changes during CBT predicted lessening of later relapse in the CBT compared with the control group (Teasdale et al., 2001). The key change appeared to be a movement from extreme cognitions either in a negative or a positive direction, towards the middle. Replication in other samples is still required. One less specific possibility is that CBT provides a coping framework for the patient whose symptoms are starting to return or worsen. A patient whose treatment has focused around medication and its benefits, and who is on high doses, can do little if symptoms are returning in spite of the medication, and can readily become demoralized, and lack a means of selfhelp. CBT can provide such a coping framework. The existing controlled trials do not fully rule out one further possibility : that some of the benefit is merely due to an increased amount of non-specific therapeutic contact. It is difficult to find a control group modality that fully controls for this, is plausible and motivating for patients and sustainable over a substantial number of sessions for therapists. It would seem unlikely that such effects of non-specific contact would produce differences in major relapse of the magnitude which have been found, in trials with assessors masked to treatment modality, but a portion of the effect could be attributable to this. It is also not clear how the effects of CBT in this context compare with other psychological therapies for depression, and particularly for interpersonal

psychotherapy (IPT). A controlled trial comparing CBT and IPT in relapse prevention is now needed. It would need to be in a large sample, since some benefit might be expected for each, and preferably should also include a lower contact control group. Lastly, the indications for CBT in relapse prevention require formulation and some consensus. The following recommendations are proposed : (1) CBT is indicated where there is prior evidence of vulnerability to relapse and recurrence, for instance by residual symptoms not responding fully to antidepressant, previous history of relapse or recurrences ; (2) it should be used in these circumstances as an adjunct to continuation or maintenance medication ; (3) preferably it should be started during improvement, while there are still some symptoms and negative cognitions are still accessible.

CBT in relapse prevention

differential relapse prevention effects. Journal of Consulting and Clinical Psychology 72, 31–40. Miller IW, Norman WG, Keitner GI (1989). Cognitivebehavioral treatment of depresssed inpatients: six- and twelve-month follow-up. American Journal of Psychiatry 146, 1274–1279. Mueller TI, Leon AC, Keller MB, Solomon DA, Endicott J, Coryell W, Warshaw M, Maser JD (1999). Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. American Journal of Psychiatry 156, 1000–1006. Paykel ES, Hart D, Priest RG (1998). Changes in public attitudes to depression during the Defeat Depression Campaign. British Journal of Psychiatry 173, 519–522. Paykel ES, Ramana R, Cooper Z, Hayhurst, H, Kerr J, Barocka A (1995). Residual symptoms after partial remission: an important outcome in depression. Psychological Medicine 25, 1171–1180. Paykel ES, Scott J, Cornwall PL, Abbott R, Crane C, Pope M, Johnson AL (2005). Duration of relapse prevention after cognitive therapy in residual depression: a follow-up of controlled trial. Psychological Medicine 35, 59–68. Paykel ES, Scott J, Teasdale JD, Johnson AL, Garland A, Moore R, Jenaway A, Cornwall PL, Hayhurst H, Abbott R, Pope M (1999). Prevention of relapse in residual depression by cognitive therapy. A controlled trial. Archives of General Psychiatry 56, 829–835. Proudfoot J, Goldberg D, Mann A, Everitt B, Marks I, Gray JA (2003). Computerized, interactive, multimedia cognitive-behavioural program for anxiety and depression in general practice. Psychological Medicine 33, 217–227. Ramana R, Paykel ES, Cooper Z, Hayhurst, H, Saxty, M, Surtees PG (1995). Remission and relapse in major depression: a two-year prospective follow-up study. Psychological Medicine 25, 1161–1170. Ramana R, Paykel ES, Melzer D, Mehta MA, Surtees PG (2003). Aftercare of depressed inpatients: service delivery and unmet needs. Social Psychiatry and Psychiatric Epidemiology 38, 109–115. Ramana R, Paykel ES, Surtees PG, Melzer D, Mehta M (1999). Medication received by patients with depression following the acute episode: adequacy and relation to outcome. British Journal of Psychiatry 174, 128–134. Scott J, Palmer S, Paykel E, Teasdale J, Hayhurst H (2003). Use of cognitive therapy for relapse prevention in chronic depression: cost effectiveness study. British Journal of Psychiatry 182, 221–227. Scott J, Teasdale JD, Paykel ES, Johnson AL, Abbott R, Hayhurst H, Moore R, Garland A (2000). Effects of cognitive therapy on psychological symptoms and social functioning in residual depression. British Journal of Psychiatry 177, 440–446. Shea MT, Elkin I, Linber S, Sotsky S, Watkins J, Collins J, Pilkonis P, Beckham E, Glass D, Dolan R,

Downloaded from http://ijnp.oxfordjournals.org/ by guest on January 15, 2017

for depression. Archives of General Psychiatry 49, 802–808. Fava GA, Grandi S, Zielezny M, Canestrari R, Morphy MA (1994). Cognitive behavioural treatment of residual symptoms in primary major depressive disorder. American Journal of Psychiatry 151, 1295–1299. Fava GA, Grandi S, Zielezny M, Rafanelli C, Canestrari R (1996). Four-year outcome for cognitive behavioural treatment of residual symptoms in major depression. American Journal of Psychiatry 153, 945–947. Fava GA, Rafanelli C, Grandi S, Canestrari R, Morphy MA (1998a). Six year outcome for cognitive behavioural treatment of residual symptoms in major depression. American Journal of Psychiatry 155, 1443–1445. Fava GA, Rafanelli C, Grandi S, Conti S, Belluardo P (1998b). Prevention of recurrent depression with cognitive behavioural therapy: preliminary findings. Archives of General Psychiatry 55, 816–820. Fava G, Ruini C, Finos L, Conti S, Grandi S (2004). Six-year outcome of cognitive behaviour therapy for prevention of recurrent depression. American Journal of Psychiatry 161, 1872–1876. Hollon SD, DeRubeis RJ, Shelton RC, Amsterdam JD, Salomon RM, O’Reardon JP, Lovett ML, Young PR, Haman KL, Freeman BB, Gallop R (2005). Prevention of relapse following cognitive therapy vs medications in moderate to severe depression. Archives of General Psychiatry 62, 417–422. Jarrett RB, Kraft D, Doyle J, Foster BM, Eaves CG, Silver PC (2001). Preventing recurrent depression using cognitive therapy with and without a continuation phase: a randomised clinical trial. Archives of General Psychiatry 58, 381–388. Keller MB, Klerman GL, Lavori PW, Coryell W, Endicott J, Taylor J (1984). Long-term outcome of episodes of major depression: clinical and public health significance. Journal of the American Medical Association 252, 788–792. Kennedy N, Abbott R, Paykel ES (2003). Remission and recurrence of depression in the maintenance era: long-term outcome in a Cambridge cohort. Psychological Medicine 33, 827–838. Kennedy N, Abbott R, Paykel ES (2004). Longitudinal syndromal and sub-syndromal symptoms after severe depression: 10-year follow-up study. British Journal of Psychiatry 184, 330–336. Kiloh LG, Andrews GA, Neilson M (1988). The long-term outcome of depressive illness. British Journal of Psychiatry 153, 752–757. Kovacs M, Rush AJ, Beck AT, Hollon SD (1981). Depressed outpatients treated with cognitive therapy or pharmacotherapy. Archives of General Psychiatry 38, 33–41. Lee AS, Murray RM (1988). The long-term outcome of Maudsley depressives. British Journal of Psychiatry 153, 741–751. Ma SH, Teasdale JD (2004). Mindfulness-based cognitive therapy for depression: replication and exploration of

135

136

E. S. Paykel

Parioff M (1992). Course of depressive symptoms over follow-up. Archives of General Psychiatry 49, 782–787. Simons AD, Murphy GE, Levine JL, Wetzel RD (1986). Cognitive therapy and pharmacotherapy of depression: sustained improvement over one year. Archives of General Psychiatry 43, 43–50. Surtees PG, Barkley C (1994). Future imperfect: the long-term outcome of depression. British Journal of Psychiatry 164, 327–341.

Teasdale JD, Scott J, Moore RG, Hayhurst H, Pope M, Paykel ES (2001). How does cognitive therapy prevent relapse in residual depression? Evidence from a controlled trial. Journal of Consulting and Clinical Psychology 69, 347–357. Teasdale JD, Segal ZV, Williams JM, Ridgeway VA, Soulsby JM, Lau MA (2000). Prevention of relapse/ recurrence in major depression by mindfulness-based cognitive therapy. Journal of Consulting and Clinical Psychology 68, 615–623.

Downloaded from http://ijnp.oxfordjournals.org/ by guest on January 15, 2017