Holmeslab.commons.yale.edu

Cogn Ther ResDOI 10.1007/s10608-007-9166-6 The Role of Cognitive-Behavioral Therapy andFluoxetine in Prevention of Recurrence of MajorDepressive Disorder Timothy J. Petersen Æ Joel A. Pava Æ Jacqueline Buchin ÆJohn D. Matthews Æ George I. Papakostas Æ Andrew A. Nierenberg ÆAvram J. Holmes Æ Ryan Bogdan Æ Lesley M. Graves ÆRebecca M. Harley Æ Maurizio Fava Ó Springer Science+Business Media, LLC 2007 This study evaluated the role of cognitive-behavioral therapy (CBT) and fluoxetine in preventing recurrence of a depressive episode during maintenance phase treatment forpatients with remitted major depressive disorder (MDD). Patients (n = 52) completed openacute fluoxetine treatment and sustained remission during a 28-week randomized continuationtreatment (CBT + fluoxetine vs. fluoxetine only). They were assigned to one of four main-tenance treatments: CBT + fluoxetine, CBT + placebo, fluoxetine only, and placebo only.
There were no statistically significant differences in MDD recurrence between maintenancetreatments, but continued antidepressant treatment (with or without CBT) provided an 18–21%lower MDD recurrence rate than placebo. These findings are consistent with those of recentantidepressant studies of chronic and recurrent MDD populations. Although sample sizes weresmall, CBT did not significantly lower rates of MDD recurrence.
CBT Á Fluoxetine Á Recurrence Á Relapse Á Major Depressive Disorder Á Major depressive disorder (MDD) is a highly recurrent, often chronic illness (Fava and KajiEven following successful acute phase treatment, patients remain at risk for relapse ofthe index depressive episode or the emergence of a new episode (recurrence). Patients not T. J. Petersen (&)Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital,15 Parkman Street, WAC 812, Boston, MA 02114, USAe-mail: tpetersen@partners.org J. A. Pava Á J. Buchin Á J. D. Matthews Á A. A. Nierenberg Á A. J. Holmes Á R. Bogdan ÁL. M. Graves Á R. M. Harley Á M. FavaDepartment of Psychology, Harvard University, Cambridge, MA 02138, USA G. I. PapakostasDepartment of Psychiatry, Massachusetts General Hospital, 50 Staniford Street, Suite 401, Boston,MA 02114, USA achieving full remission following the acute treatment phase (i.e., partial response withremaining residual symptoms) are at greater risk for both relapse and recurrence (Judd et al.
As a result, emphasis has been placed on evaluating treatments according to their abilityto produce acute phase remission and prevent relapse and recurrence.
Evidence indicates that continuation and maintenance of antidepressant treatment lowers rates of relapse and recurrence relative to placebo (Fava and Kaji ). In addition, bothinterpersonal psychotherapy and variants of cognitive-behavioral therapy (CBT) have shownefficacy in preventing relapse and recurrence relative to control treatments (Fava et al. 1998;Frank et al. Reynolds et al. ; Teasdale et al. ). Specifically, variants of CBTmodified to specifically address residual symptoms in patients who have recovered fromdepression appear to be most efficacious in preventing relapse and recurrence (Segal et al In spite of recent clinical advances, the precise role of antidepressants and CBT in pre- venting depressive recurrence during the maintenance phase of treatment remains unclear. Pastresearch suggests that medication maintenance prevents recurrence (Fava and Kaji ;Keller et al. ), and the addition of CBT methods to medication maintenance confers along-term treatment advantage (Paykel et al. ; Teasdale et al. This study sought toinvestigate the efficacy of different forms of maintenance treatment. To accomplish this goal,the present study compared rates of MDD recurrence and symptom change during mainte-nance phase treatment for remitted MDD patients who were randomized to one of fourtreatment groups: (a) antidepressant (fluoxetine) treatment only, (b) CBT plus antidepressanttreatment, (c) CBT plus placebo, and (d) placebo only. Patients had previously achievedsustained remission through open acute and randomized continuation treatment, the results ofwhich are described elsewhere (Perlis et al. ). Based upon previous research suggestingbeneficial effects of maintenance treatment, we hypothesized that patients receiving activetreatment (antidepressant and/or CBT) would show lower rates of MDD recurrence comparedto those receiving placebo. Furthermore, in light of research suggesting superiority of com-bined maintenance treatments, we hypothesized that those receiving a combination treatment(CBT + medication) would have a superior outcome relative to those receiving monotherapy(CBT or medication alone).
Participants were recruited through the Depression Clinical and Research Program (DCRP) ofthe Massachusetts General Hospital between 1992 and 1998. All participants signed writtenconsent approved by the Massachusetts General Hospital’s Institutional Review Board.
Patients were randomized to one of four 80-week maintenance treatment groups:CBT + placebo, CBT + medication (40 mg fluoxetine), medication (40 mg fluoxetine) only,and placebo only. Participants entered the maintenance phase of treatment after completing theacute and continuation phases of treatment (Table ).
The acute phase was an open, eight-week, fixed dose of fluoxetine at 20 mg/day (n = 391 patients; 55% female; mean age 39.8 ± 10.6; Fava et al. ). Sixty patients dropped outduring the acute phase of treatment (391 – 60 = 331). Of the remaining 331 patients, 101 (52men and 49 women; mean age, 41.6 ± 10.6 years) met criteria for partial response (n = 49) ornonresponse (n = 52). These patients were randomized, as part of a separate study, to fourweeks of double-blind treatment with high-dose fluoxetine (40–60 mg/day), fluoxetine pluslithium (300–600 mg/day), or fluoxetine plus desipramine (25–50 mg/day). Eighty-two Table 1 Demographic and clinical features by maintenance treatment group Note. HAMD-17 = 17-item Hamilton Rating Scale for Depression patients (33 men and 49 women; mean age 40.4 ± 9.1 years) met criteria for response andwere offered open-label treatment within the DCRP program. Of the 148 patients who metcriteria for full remission, 16 refused randomization to the continuation study described in thisreport. The remaining remitters (n = 132; 54.5% female; mean age 39.9 ± 10.3) to this acutephase treatment [remission = 17-item Hamilton Rating Scale for Depression (HAMD-17;Hamilton £7 for three consecutive weeks) were randomized to two, 28-week contin-uation phase treatment groups: (1) increase in fluoxetine to fixed dose 40 mg/day (n = 66) or,(2) the same increased fixed dose plus 12 weekly sessions and then seven biweekly CBTsessions (n = 66) (Perlis et al. ).
There were no significant differences between the two continuation treatment groups in rates of relapse, discontinuation, change in well-being, or residual symptoms and symptomquestionnaire scores. Rates of relapse were 6% (4/66) for the CBT + medication group and 8%(5/66) for the medication only group (Perlis et al. ). Of the 132 patients randomized to thecontinuation study groups, 47 discontinued the study (9 due to relapse), 30 exited the study toreceive open-label treatment of worsened depressive symptoms, and 55 met criteria for andwere randomized to maintenance phase treatments.
Participants were drug-free outpatients who met criteria for MDD, as diagnosed with theStructured Clinical Interview for DSM-III-R—Patient Edition (Spitzer et al. had aninitial 17-item Hamilton Rating Scale for Depression (HAM-D-17; Hamilton ) score [16,and were 18 to 65 years of age. Patients were also required to meet at least one of thefollowing criteria: history of three or more major depressive episodes, with the prior episodeno more than 2.5 years before the onset of the current episode; diagnosis of current episode aschronic (onset of continuous depressive symptoms ‡36 months prior to study); history of poorinterepisode recovery; or both MDD and dysthymia.
Exclusion criteria included pregnant women and women of child-bearing potential who were not using a medically accepted means of contraception, women of child-bearing potential taking a birth control pill, or women who were currently lactating. Also excluded were patientswith a serious risk of suicide, seizure disorder history, major unstable medical illness, historyof multiple adverse drug reactions or allergy to the study drugs, antisocial personality disorder,or a DSM-III-R comorbid diagnosis of axis I pathology other than anxiety disorders. Fur-thermore, patients currently using nonstudy related psychotropic drugs or exhibiting evidenceof hypothyroidism were excluded. Patients were excluded if their depression failed to respondin the past to a trial of (1) a higher dose of fluoxetine (60–80 mg/day), (2) the combination offluoxetine and desipramine, or (3) the combination of fluoxetine and lithium. Finally, patientswere excluded if they failed to respond during the course of their current major depressiveepisode to at least one adequate antidepressant trial, defined as six weeks or more of treatmentwith either [150 mg of imipramine (or its tricyclic equivalent) or [60 mg of phenelzine (orits monoamine oxidase inhibitor equivalent).
Fifty-five patients (54% women; mean age: 43.6 years) met criteria for remission at the end ofthe continuation treatment phase and were randomized to one of four maintenance treatments:CBT plus placebo (n = 12), CBT plus medication (n = 11), medication only (n = 15), andplacebo only (n = 17). Three patients (one in the CBT + placebo group, one in the medicationonly group, and one in the placebo group) discontinued the study before beginning mainte-nance phase treatment, resulting in the following group sizes for analyses: CBT plus placebo(n = 11), CBT plus medication (n = 11), medication only (n = 14), and placebo only (n = 16).
All three patients who discontinued were lost to follow up. With such low frequency, nodifferences in discontinuation were detected across the four groups. Patients were randomizedto one of the four maintenance treatment groups on the basis of their continuation treatmentassignment. Patients who received CBT during continuation treatment were eligible for ran-domization to either the CBT plus placebo or CBT plus medication maintenance arms. Patientswho did not receive CBT during continuation treatment were eligible for randomization toeither the medication only or placebo only maintenance arms.
Patients were seen for 20 monthly study visits by psychiatrists or psychologists. During each visit, patients were administered the 31-item version of the HAM-D (allowing the scoringof the HAMD-17; Hamilton ), and the Clinical Global Impressions—Severity (CGI-S)and Improvement (CGI-I) Scales (Guy ). To assess symptom change and therapeuticimprovement, patients completed the 92-item Symptom Questionnaire (SQ; Kellner theBeck Hopelessness Scale (BHS; Beck et al. the Beck Depression Inventory (BDI; Becket al. ), the Patient Global Impression of Improvement (PGI-I; Guy ), and the BeckAnxiety Inventory (BAI; Beck et al. at the conclusion of each study visit.
Cognitive therapy was conducted by highly trained doctoral-level psychologists according to a treatment manual adapted from Beck et al. and Mercier and Leahy ). CBTwas modified to address residual symptoms specifically and to enhance patient coping skills.
The therapy used for this study was designed specifically to target symptoms and issuescommon to remitted depressed patients, who are at high risk for relapse and recurrence. Threecontent domains are emphasized when working with remitted depressed patients. The first isrecovery, which involves working to resolve any residual symptoms that are present afterclinical remission. Such residual symptoms are common and include: irritability, neurovege-tative disturbances, and hopelessness. The second content area is re-entry, which entailsworking to increase a patient’s functioning in key life roles such as student, family member,spouse, and employee. An acute depressive episode typically results in lowered levels offunctioning in one or more of these areas, thus the gap between current and optimal psychosocial functioning may be significant. One common target of this content area isavoidant behavior, which is often activated by a patient to maintain tentative short-termstability but in turn prevents return to premorbid levels of functioning. The third content area,risk, involves focusing on maladaptive cognitive and behavioral patterns that contribute toheightened relapse rates. Such patterns include lack of assertiveness and self-care, as well asperfectionism and unrealistic self-expectations. The structure of therapy for the maintenancephase of this protocol (an 80-week period) consisted of seven biweekly, 50-min sessionsfollowed by 16 monthly, 50-min sessions. Each session followed a conventional cognitivetherapy format, which includes: symptom check, agenda setting, homework review, cognitiveand behavioral exercises for specific problem areas, and assignment of new homework. Fordetails on what specific techniques are matched to certain problem areas, a full treatmentmanual is available upon request (Pava et al. ). Psychopharmacologists followed astandard protocol for medication management visits (Fawcett et al. and were instructednot to engage in cognitive or behavioral interventions (Pava et al. ).
The primary study endpoint was depressive recurrence, defined as either meeting criteria for anew episode of MDD at any maintenance visit or scoring 15 or higher on the HAM-D-17 overtwo consecutive visits. Recurrence was confirmed by a follow-up visit one week later withanother clinician blind to treatment status. The chi-square method was used to compare MDDrecurrence rates across the four treatment groups. The Kaplan–Meier survival analysis wasalso utilized for time-to-recurrence or study discontinuation, with observations censored after80 weeks, following completion of this phase of the study. The Mantel–Cox (log-rank) testwas used to compare survival curves between study conditions. Paired t-tests were used tocompare within-group changes on secondary measures (from baseline to week 80 or studydiscontinuation), and factorial ANOVAs were used to compare groups. Effect sizes of within-group changes were calculated using the method suggested by Cohen FactorialANOVAs were used to compare mean outcome measure values across treatment groups for theentire maintenance phase. All values for a given outcome measure for each patient weretotaled and divided by the number of observed data points for that patient. All analyses wereconducted on the intent-to-treat sample, with last observation carried forward. All tests weretwo-tailed, with the threshold for statistical significance set at p \ 0.05.
The primary outcome of this study was rate of recurrence, which was not significantly differentacross the four treatment groups: CBT + placebo = 5/11 (45%); CBT + medication = 4/11(36%); medication only = 4/14 (29%); placebo = 8/16 (50%; X2 = 1.91, p = 0.19). Theoverall MDD recurrence rate for medication (with or without CBT) was not significantly lower(8/25, 32%) than the recurrence rate for placebo (50%). Kaplan–Meier survival curves wereconstructed with results censored at end of maintenance phase or at study discontinuation forreason other than recurrence, and there were no significant differences between groups (log-rank X2 = 2.37, df = 3, p [ 0.40). We also examined rates of recurrence between the twopatient groups defined as continuation treatment assignments (medication only andCBT + medication). Rates of recurrence, using this methodology, were: medication only =9/22 (41%) and CBT + medication = 12/30 (40%). Kaplan–Meier survival curves were Table 2 Mean change and effect size (Cohen’s d) of change during maintenance treatment CBT + placebo (n = 11) CBT + medication (n = 11) Medication (n = 14) Placebo (n = 16) Note. BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; HAMD-17 = 17-item Hamilton RatingScale for Depression; CGI-I = Clinical Global Impressions, Improvement; PGI-I = Patient Global Impressions,Improvement; CGI-S = Clinical Global Impressions, Severity; SQ = Symptom Questionnaire; BHS = BeckHopelessness Scale constructed for these results censored at end of maintenance phase or at study discontinuationfor reason other than recurrence, and there were no significant differences between groups(log-rank v2 = 0.14, df = 1, p = 0.71). For those who received CBT (with or without medi-cation), the mean number of psychotherapy sessions completed was 21.1 (SD = 2.5) of 23 forthe CBT + medication group and 21.6 (SD = 1.9) for the CBT + placebo group.
As seen in Table there were no significant differences in change scores between groups on the BAI, BDI, HAMD-17, CGI-I, PGI-I, CGI-S, and SQ (depression, anxiety, hostility,somatic), and BHS. The 52 patients randomized to maintenance treatment did not differsignificantly from the remainder of the sample (n = 339; all comparisons p [ 0.05) on any ofthe demographic characteristics.
As with a number of studies on the long-term effects of therapies for depression (e.g., Franket al. ; Paykel et al. the sample size in this investigation of maintenance treatmentsis small. Conclusions about the lack of statistically significant differences between groups therefore must be tentative. Nonetheless, a discussion of the findings can stimulate futureresearch on prevention efforts in the treatment of depression, which is increasingly beingcharacterized as a recurrent and chronic disorder (McCullough et al. An unexpectedfinding was that CBT did not provide further protection from recurrence when added topharmacotherapy. This finding is inconsistent with some recent studies in which CBT con-ferred an advantage in long-term treatment outcome when compared with antidepressanttreatments alone or with no treatment control conditions (Jarrett et al. Teasdale et al.
For example, in a recent study by Teasdale and colleagues (Teasdale et al. ), 145patients who had remitted or recovered by the end of acute phase pharmacologic treatmentwere randomized to receive medication management only or medication management plusMindfulness-Based Cognitive Therapy (MBCT). MBCT is a group intervention designed totrain recovered recurrently depressed patients to disengage from depressogenic thinking thatmay mediate relapse/recurrence. Relapse/recurrence to major depression was assessed over a60-week study period. Patients treated with MBCT experienced 30% fewer relapses/recur-rences compared to the usual care group. The Teasdale et al. () study included a largersample than the present study, a group rather than individual psychotherapy, and a differentcognitively-based therapy that included meditation. However, there was no placebo controlgroup, and medication was provided naturalistically, with no dosing restrictions or limitationson type of antidepressant used.
Jarrett and colleagues (studied 84 treatment responsive patients with recurrent MDD who were randomized to either eight months (10 sessions) of cognitive (CT) or a controlcondition (evaluation without CT). Over an eight-month period, only 10% of patients receivingCT relapsed, whereas 31% of those in the control condition relapsed. This study differs fromours in that patients did not receive antidepressant treatment during any phase of treatment. Inaddition, the psychotherapy utilized was CT, not CBT, and involved far fewer sessions.
However, the Jarrett et al. () study does suggest a prophylactic effect of CT whencompared with no treatment.
Two other recently published studies provide important additional data regarding the long- term impact of psychotherapeutic treatments for depression. Gelenberg et al. () examinedrecurrence rates in two maintenance treatment arms—nefazodone treatment (n = 76) andplacebo only (n = 74). All patients randomized to these two arms had achieved and sustainedclinical response during the acute and continuation treatment phases with either nefazodonealone or in combination with psychotherapy specifically designed for chronically depressedpatients (CBASP). Findings of this maintenance study, while supporting the importance oflong-term antidepressant maintenance treatment to prevent recurrence, did not reveal anyadded protective effect of psychotherapy. In a companion study, some patients from the samecohort who received maintenance psychotherapy sessions (up to 13 monthly sessions) hadlower recurrence rates (10.7%; n = 42) than patients receiving clinical management (CM) only(32.0%; n = 40) (Klein et al. This latter investigation is most similar to ours in thatpsychotherapy was provided throughout all phases of treatment.
Recent studies have also evaluated the ability of CT or CBT to resolve residual depressive symptoms and lower rates of relapse and recurrence. For example, Paykel et al. ran-domized 158 patients who responded to acute phase pharmacotherapy but still reportedresidual symptoms to either 16 CBT sessions plus medication or to usual CM (medicationonly). After this continuation treatment, patients were followed for one year. The cumulativerelapse rate was 47% for the CM group and 29% for the CBT group. Patients were allowed tohave up to a 30% increase in antidepressant dosing during continuation treatment. Our studydiffers in that dosing of fluoxetine was doubled after acute phase treatment, and only those who remitted in our continuation and maintenance study arms were included, thus excludingpatients with significant residual symptoms.
G.A. Fava and colleagues have also conducted a series of investigations (1998b,c) to evaluate CBT in resolving residual symptoms and reducing relapse/recurrence rates. In brief,43 patients were followed for a six-year period following acute phase antidepressant responseand randomization to either CBT or usual CM. CBT consisted of 10 biweekly sessions and wasmodified to specifically address residual symptoms and relapse prevention. All patients in theCM group had antidepressants tapered and ultimately discontinued during continuationtreatment. Relapse/recurrence rates for the CM and CBT groups were 35% vs. 15% at twoyear, 70% vs. 35% at four year, and 75% vs. 50% at six-year follow-up points.
G.A. Fava and colleagues then devised a modified form of cognitive therapy known as Well Being Therapy (WBT; Fava et al. ). The rationale for WBT is that continuation psy-chotherapy should not only target depressive cognitions and behaviors, but also ways toenhance well-being and modify problematic lifestyles. In a study that evaluated WBT vs. CMduring the continuation phase of treatment (Fava et al. ), patients’ antidepressantmedications were tapered after about three months into the continuation phase. Results indi-cate that over a period of two years, 80% of patients in the CM group experienced a depressiverelapse, as compared with 25% in the WBT group. Although it is unclear how WBT mighthave its prophylactic effects, this finding has led to increased interest in the use of thisspecialized form of psychotherapy to prevent depressive relapse and recurrences. The majordifference between Fava et al.’s work and ours is that patients in their protocols tapered andultimately discontinued antidepressant medications, whereas patients in our protocol remainedon an increased dosage throughout the continuation phase of treatment. In addition, the initialfocus of our post-acute treatment was on resolving residual depressive symptoms.
Other studies are consistent with our findings of no significant prophylactic effect of CT (Blackburn and Moore ; Frank et al. Shea et al. In analyses of follow-updata from the NIMH collaborative study, Shea and colleagues found that neither IPTnor CT provided additional benefit in terms of prevention of relapse and recurrence, whencompared with medication only and placebo conditions. Blackburn and Moore ran-domized 75 outpatients with recurrent major depression to three groups: 16 weeks of acutetreatment and two years’ maintenance treatment with antidepressants and maintenance anti-depressants; cognitive therapy and maintenance cognitive therapy; and antidepressants andmaintenance cognitive therapy. In the acute phase of treatment, all patients improved signif-icantly, and there were no significant differences among treatments or in the pattern ofimprovement over time. In the maintenance stage, CT therapy had a similar prophylactic effectto maintenance medication, and there were no differences in symptom improvement orrelapse/recurrence between the three treatment groups. Some patients in this protocol receivedacute phase CT, whereas patients in our protocol received only medication during acute phasetreatment.
Although the findings were not statistically significant, on some secondary measures we did see a trend toward separation of groups. For instance, the CBT + placebo and CBT + medi-cation groups demonstrated lower mean scores on the BAI, BDI, and SQ anxiety subscalewhen compared with the medication only group. With a larger sample, it is possible that suchdifferences would have reached statistical significance. We speculate that these results mayrepresent a tendency for CBT to provide a greater prophylactic effect than medications alone,as previous research mentioned above suggests.
As suggested by Segal et al. ), sequenced or crossover treatment in which the delivery of antidepressant medication is followed by structured psychotherapy appears to providesuperior prophylaxis than either treatment alone or concurrently combined. However, studies to date are few and generally include small sample sizes. Much work remains to furtherelucidate optimal forms of psychotherapy and how to integrate such psychosocial treatmentswith antidepressant medications.
Aside from insufficient power, another limitation of the study is that we did not include independent ratings of treatment quality and fidelity. As a result, it is possible that the qualityof CBT and/or adherence to the treatment manual differed between therapists. However, thetherapists who provided treatment in this study are formally trained and certified to provideCBT treatment for depression, which makes it more likely that psychotherapy was uniform. Asecond limitation is that patients randomized to maintenance treatment entered via two dif-ferent continuation phase pathways (CBT and medication or medication only) and thus mayrepresent groups with different degrees of relapse risk before receiving maintenance treatment.
In addition, given that patients were experiencing minimal depressive symptoms upon entry tothe maintenance phase (HAMD-17 mean score range = 2.8–5.5 across treatment groups), it isconceivable that there were limited targets for CBT interventions (e.g., residual depressivesymptoms). In this way, it is possible that there was less room for CBT to provide a pro-phylactic effect. Modified forms of CBT, such as Well Being Therapy (Fava et al. ) orMBCT (Teasdale et al. ) may be a better match for remitted patients. Future investiga-tions may benefit from a more streamlined design, where patients who respond or remit toacute phase treatment are randomized to maintenance treatment arms immediately, rather thanadding an intervening continuation phase. In addition, keeping continuation and maintenancephase antidepressant doses at the same level as in the acute phase would help to clarify thespecific relapse prevention effects of CBT. Such a design would also mitigate antidepressantside effects that can contribute to dropout or be confused for re-emergence of symptoms.
Even with these limitations, data from this study provide an important contribution to a relatively understudied area in the long-term treatment of depression and confirm theimportance of the design and execution of larger size, well-controlled, randomized trials toinvestigate the role of CBT-based psychotherapies and antidepressant medication in the pre-vention of depressive relapse and recurrence.
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893–897.
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571.
Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of pessimism: The hopelessness scale. Journal of Consulting and Clinical Psychology, 42, 861–865.
Blackburn, I. M., Eunson, K. M., & Bishop, S. (1986). A two-year naturalistic follow-up of depressed patients treated with cognitive therapy, pharmacotherapy, and a combination of both. Journal of AffectiveDisorders, 10, 67–75.
Blackburn, I. M., & Moore, R. G. (1997). Controlled acute and follow-up trial of cognitive therapy and pharmacotherapy in out-patients with recurrent depression. British Journal of Psychiatry, 171, 328–334.
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159.
Fava, G. A. (1999). Subclinical symptoms in mood disorders: Pathophysiological and therapeutic implications.
Psychological Medicine, 29, 47–61.
Fava, M., Alpert, J., Nierenberg, A., Lagomasino, I., Sonawalla, S., Tedlow, J., et al. (2002). Double-blind study of high-dose fluoxetine versus lithium or desipramine augmentation of fluoxetine in partial responders andnonresponders to fluoxetine. Journal of Clinical Psychopharmacology, 22, 379–387.
Fava, M., & Kaji, J. (1994). Continuation and maintenance treatments of major depressive disorder. Psychiatric Fava, G. A., Rafanelli, C., Cazzaro, M., Conti, S., & Grandi, S. (1998a). Well-being therapy: A novel psychotherapeutic approach for residual symptoms of affective disorders. Psychological Medicine, 28,475–480.
Fava, G. A., Rafanelli, C., Grandi, S., Canestrari, R., & Morphy, M. A. (1998b). Six-year outcome for cognitive behavioral treatment of residual symptoms in major depression. American Journal of Psychiatry, 155,1443–1445.
Fava, G. A., Rafanelli, C., Grandi, S., Conti, S., & Belluardo, P. (1998c). Prevention of recurrent depression with cognitive behavioral therapy: Preliminary findings. Archives of General Psychiatry, 55, 816–820.
Fawcett, J., Epstein, P., Fiester, S. J., Elkin, I., & Autry, J. H. (1987). Clinical management: Imipramine/placebo administration manual. Psychopharmacology Bulletin, 23, 309–323.
Frank, E., Kupfer, D. J., Perel, J. M., Cornes, C., Jarrett, D. B., Mallinger, A. G., et al. (1990). Three-year outcomes for maintenance therapies in recurrent depression. Archives of General Psychiatry, 47, 1093–1099.
Gelenberg, A. J., Trivedi, M. H., Rush, A. J., Thase, M. E., Howland, R., Klein, D. N., et al. (2003). Randomized, placebo-controlled trial of nefazodone maintenance treatment in preventing recurrence in chronicdepression. Biological Psychiatry, 54, 806–817.
Guy, W. (Ed.) (1976). ECDEU assessment manual for psychopharmacology, revised [DHEW Pub. No. (ADM) 76-338]. Rockville, MD: National Institute of Mental Health.
Hamilton, M. (1960). A rating scale for depression. Journal of Neurology and Neurosurgery Psychiatry, 23, 56– Jarrett, R. B., Kraft, D., Doyle, J., Foster, B. M., Eaves, G. G., & Silver, P. C. (2001). Preventing recurrent depression using cognitive therapy with and without a continuation phase. Archives of General Psychiatry,58, 381–388.
Judd, L. L., Akiskal, H. S., & Paulus, M. P. (1997). The role and clinical significance of subsyndromal depressive symptoms (SSD) in unipolar major depressive disorder. Journal of Affective Disorders, 45(1–2), 5–18.
Keller, M. B., Kocsis, J. H., Thase, M. E., Gelenberg, A. J., Rush, A. J., Koran, L., et al. (1998). Maintenance phase efficacy of sertraline for chronic depression: A randomized controlled trial. Journal of the AmericanMedical Association, 280, 1665–1672.
Kellner, R. (1987). A symptom questionnaire. Journal of Clinical Psychiatry, 48, 268–274.
Klein, D. N., Santiago, N. J., Vivian, D., Blalock, J. A., Kocsis, J. H., Markowitz, J. C., et al. (2004). Cognitive- behavioral analysis system of psychotherapy as a maintenance treatment for chronic depression. Journal ofConsulting and Clinical Psychology, 72, 681–688.
McCullough, J. P. Jr., Klein, D. N., Borian, F. E., et al. (2003). Group comparisons of DSM-IV subtypes of chronic depression: Validity of the distinctions, part 2. Journal of Abnormal Psychology, 112, 614–622.
Mercier, M. A., & Leahy, R. L. (1992). Cognitive therapy of dysthymia: A treatment manual. Unpublished manuscript, Columbia-Presbyterian Hospital, New York.
Pava, J. A., Fava, M., & Levenson, J. A. (1994). Integrating cognitive therapy and pharmacotherapy in the treatment and prophylaxis of depression: A novel approach. Psychotherapy and Psychosomatics, 61, 211–219.
Pava, J. A., McDermott, S. P., & Fava, M. (1996). Cognitive therapy of remitted depressives at high risk for relapse/recurrence: A treatment manual. Unpublished manuscript.
Paykel, E. S., Scott, J., Teasdale, J. D., Johnson, A. L., Garlan, A., Moore, R., et al. (1999). Prevention of relapse in residual depression by cognitive therapy. Archives of General Psychiatry, 56, 829–835.
Perlis, R. H., Nierenberg, A. A., Alpert, J. E., Pava, J., Matthews, J. D., Buchin, J., et al. (2002). Effects of adding cognitive therapy to fluoxetine dose increase on risk of relapse and residual depressive symptoms incontinuation treatment of major depressive disorder. Journal of Clinical Psychopharmacology, 22, 474–480.
Reimherr, F. W., Amsterdam, J. D., Quitkin, F. M., Rosenbaum, J. F., Fava, M., Zajecka, J., et al. (1998).
Optimal length of continuation therapy in depression: A prospective assessment during long-termfluoxetine treatment. American Journal of Psychiatry, 155, 1247–1253.
Reynolds, C. F. III, Frank, E., Perel, J. M., Imber, S. D., Cornes, C., Miller, M. D., et al. (1999) Nortriptyline and interpersonal psychotherapy as maintenance therapies for recurrent major depression: A randomizedcontrolled trial in patients older than 59 years. Journal of the American Medical Association, 281, 39–45.
Segal, Z., Vincent, P., & Levitt, A. (2002). Efficacy of combined, sequential and crossover psychotherapy and pharmacotherapy in improving outcomes in depression. Journal of Psychiatry and Neuroscience, 27, 281–290.
Shea, M. T., Elkin, I., Imber, S. D., Sotsky, S. M., Watkins, J. T., Collins, J. F., et al. (1992). Course of depressive symptoms over follow-up. Findings from the National Institute of Mental Health Treatment ofDepression Collaborative Research Program. Archives of General Psychiatry, 49, 782–787.
Spitzer, R. L., Williams, J. B. W., Gibbon, M., & First, M. B. (1989). Structured clinical interview for DSM-III- R—Patient Edition (SCID-P, 9/1/89 Version). New York: Biometrics Research.
Teasdale, J. D., Segal, Z. V., Williams, J. M., Ridgeway, V. A., Soulsby, J. M., & Lau, M. A. (2000). Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. Journal of Consultingand Clinical Psychology, 68, 615–623.

Source: http://holmeslab.commons.yale.edu/files/2013/11/Petersen-Cogn-Ther-Res2010.pdf