Maintenance treatment for obsessivecompulsive disorder: Findings from a naturalistic setting

Transcription

1 ANNALS OF CLINICAL PSYCHIATRY ANNALS OF CLINICAL PSYCHIATRY 2015;27(1):25-32 RESEARCH ARTICLE Maintenance treatment for obsessivecompulsive disorder: Findings from a naturalistic setting Eric D. Peselow, MD New York Medical College Richmond University Medical Center and Freedom From Fear Staten Island, New York, USA Demetria R. Pizano, MA Pepperdine University Malibu, California, USA Department of Psychiatry and Behavioral Neurosciences Cedars-Sinai Medical Center Los Angeles, California, USA Waguih William IsHak, MD, FAPA Department of Psychiatry and Behavioral Neurosciences Cedars-Sinai Medical Center Los Angeles, California, USA David Geffen School of Medicine University of California, Los Angeles Los Angeles, California, USA Dr. Peselow died March 22, 2014 CORRESPONDENCE Waguih William IsHak, MD, FAPA Cedars-Sinai Medical Center 8730 Alden Drive Thalians W-157 Los Angeles, CA USA BACKGROUND: Although the utility of medication in the acute treatment of adult obsessive-compulsive disorder (OCD) is well-established, the role of maintenance therapy is not as well-studied. This study examines the efficacy of long-term treatment for, and predictors of, stability in medicated patients with adult OCD. METHODS: Using the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), we retrospectively evaluated 84 OCD patients who responded to a 10- to 12-week, open-label, acute treatment in a naturalistic clinic setting. Patients were followed based on their medication response for 1 to 92 months (mean 34.3), or until they terminated therapy. We evaluated Y-BOCS scores every 6 months or sooner, if clinically indicated. RESULTS: Of the 84 patients, 39 (46.4%) responded, having relapsed within a 5-year period. Predictors of longer duration of stability were adjunctive cognitive-behavioral therapy (CBT), lack of comorbid disorders, lower Y-BOCS score after treatment, and larger decrease in Y-BOCS score during treatment phase. CONCLUSIONS: Our results show the importance of maintenance treatment of OCD, noting the benefits of long-term response to adjunctive CBT and of achieving maximal acute response. It is becoming crucial to develop larger maintenance studies with more uniform design to better assess the natural course of treated OCD and improve treatment strategies. waguih.ishak@cshs.org AACP.com Annals of Clinical Psychiatry Vol. 27 No. 1 February

2 MAINTENANCE TREATMENT FOR OCD INTRODUCTION Treatment of obsessive-compulsive disorder (OCD), with its lifetime prevalence reported at 2% to 3% and studies that have demonstrated its chronicity, 1,2 is a significant psychiatric challenge. Many controlled studies have shown the efficacy of acute treatment with medication, primarily selective serotonin reuptake inhibitors (SSRIs) and clomipramine. The Clomipramine Collaborative Study Group 3 found that patients had a mean decrease of 40% in their Y-BOCS score in the acute phase of treatment, when compared with >10% decrease in those taking placebo, leading to the approval of clomipramine as the first FDA-approved medication for OCD. 3 Significant decreases in Y-BOCS scores (20% to 25%) also were seen in similar placebo-controlled studies with fluoxetine, sertraline, fluvoxamine, and paroxetine. 4-7 Cognitivebehavioral therapy (CBT) techniques such as exposure with response prevention, with and without adjunctive medication, have been used and studied in the acute and subacute treatment phase. 8,9 However, few studies chronicle how patients respond to treatment over longer periods of time. Several double-blind, placebo-controlled studies have followed OCD patients for 1 to 2 years. These studies, including some with open-label continuation phases, generally show a significant continued response to medication vs placebo. Studies that evaluated the discontinuation of medication also have added information to our knowledge base, showing a significant advantage for, or at least a trend favoring, continuation of medication Few naturalistic studies of treated adult OCD have been undertaken. Naturalistic studies, that is, investigations of data culled from uncontrolled, non-randomized settings, most closely approximate the settings in which most psychiatrists treat patients. The results from these kinds of studies more likely reflect the patient characteristics and modes of treatment of clinically based psychiatrists. Our evaluation hopes to quantify treatment response and predictors of that response in a naturalistic setting for the vast majority of practitioners who treat OCD in this type of environment. This study retrospectively examines 5-year longterm follow-up (range 1 month to 7.7 years, mean 34.3 months) of medicated adult OCD patients. We used not only the degree of response but also the duration of response as an endpoint. We were curious about the likelihood of continued response to the patients acutely effective medication regimens. Differential response to SSRI vs clomipramine was examined. In addition, we questioned whether there are significant predictors of the duration and degree of response and wondered if the degree of initial response itself might predict a longer response. METHODS Subjects and study design We conducted a retrospective review of data gathered on 84 patients with OCD treated in a naturalistic setting at Freedom from Fear, a community outpatient clinic in the New York metropolitan area, from June 1993 to October Patients were followed for 1 to 92 months (mean 34.3 months) after acute treatment and response. All of these patients gave informed consent for treatment and collection of data on initiation of treatment. Measures at baseline All patients met DSM-IV criteria for OCD at baseline. Each patient was rated for OCD symptom severity at baseline with the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS). To meet criteria for the study, patients must have scored at least 20 on Y-BOCS. They were clinically evaluated for any psychiatric disorder during their evaluations, and 52 (61.9%) of the patients had a DSM-IV diagnosed comorbid depressive or anxiety disorder, while 32 (38.1%) did not have psychiatric comorbidities. We excluded patients with diagnosed bipolar disorder or schizophrenia from this study, as well as those with active substance abuse or dependence. Comorbid personality disorders were evaluated, but this information is not included in this review. Treatment and measures at follow-up All patients studied were treated with medication chosen on clinical grounds, such as consideration of a patient s past history of response to a particular medication or his/her sensitivity to side effects. Various types of agents were used, but all patients received either an SSRI and/or clomipramine. A 10- to 12-week acute medication management plan was used on all patients. At the end of this period, patients were rated once again with the Y-BOCS and had achieved a 35% decrease in their score, with a total score of 15. Forty-two (50.0%) patients received monotherapy 26 February 2015 Vol. 27 No. 1 Annals of Clinical Psychiatry

3 ANNALS OF CLINICAL PSYCHIATRY (25 received SSRIs alone, 17 clomipramine alone), while the other 42 (50.0%) received multiple medications in addition to an SSRI and/or clomipramine such as a standing benzodiazepine. Thirty patients (35.7%) had engaged in at least 12 consecutive sessions of CBT within the first 6 months of treatment, while 54 (64.3%) were not in CBT. The maintenance phase of treatment was considered to have begun after the acute treatment phase. Patients were maintained on the medication to which they responded in the acute phase. A Y-BOCS was administered to each patient every 6 months. Patients whose symptoms became clinically apparent before the 6-month interval were assessed with a Y-BOCS at the time an increase in symptoms was noted. All patients were classified at study endpoint as remaining stable, terminating well, or failed. Patients continued in the study and were considered stable as long as their Y-BOCS score remained >10% lower than baseline and 15. Patients who terminated well had maintained criteria for stability until they either dropped out of the study or discontinued or had a lowered dosage with the concurrence of their psychiatrist. Patients failed when their Y-BOCS score had reached <10% decrease from baseline and required a change in medication. TABLE 1 60-month survival analysis Months stable N = Patients starting interval Terminated well Statistical analysis To assess the percentage of patients remaining stable during maintenance treatment, we used Kaplan-Meier survival analysis. The analysis was continued until the 60-month (5-year) follow-up interval. For those who terminated well, the last observation continued forward (LOCF) was used in this analysis. Survival analysis through the 36-month (3-year) interval also was used in obtaining the percentage of patients remaining stable on SSRIs alone and those on clomipramine alone. A comparison of the SSRI survival analysis and clomipramine survival analysis were assessed for differences. T test analysis was used in the comparison of months stable for those that (a) received multiple medications vs monotherapy, (b) received adjunctive CBT vs those who did not, and (c) had comorbid depression/anxiety vs those without comorbidity. A Pearson correlation was employed to assess the probability that initial score, post acute phase score, and average improvement in score on Y-BOCS predicted how long patients remained stable. On retrospective analysis, and in order to compare our study to past studies that had subdivided maintenance patients, we divided our patients into groups by post-acute treatment response (complete response = Y-BOCS 0 to 7; mild OCD symptoms = Y-BOCS 8 to 11; moderate OCD symptoms = Y-BOCS 12 to 15). Analysis of variance/least significant difference planned comparison was used to assess the number of months patients were stable according to the degree of acute response. For all comparisons, P <.05 was considered statistically significant. Analyses were performed using SAS software, version 9.2 (SAS Institute Inc., Cary, NC). RESULTS Failed Probability (%) of remaining stable % % % % % % % % % % The probabilities of a patient remaining stable at 6-month intervals were obtained through the 60-month interval (TABLE 1). As expected, the probability of remaining stable decreased over time, with about 86% probability at 1 year, 74% at 2 years, 59% at 3 years, and 48% at 4 years. At the 60-month (5-year) mark after their acute response to treatment for OCD, patients had an approximately 41% probability of remaining stable (FIGURE 1). To evaluate whether SSRIs or clomipramine were more effective in treating OCD in the maintenance phase, we compared the probabilities of remaining stable in those receiving SSRIs alone and clomipramine alone (FIGURE 2). For the 25 patients who were maintained on SSRIs alone, the probability of remaining stable was 59.37%. While the probability of stability for the AACP.com Annals of Clinical Psychiatry Vol. 27 No. 1 February

4 MAINTENANCE TREATMENT FOR OCD FIGURE 1 60-month survival analysis % Remaining stable FIGURE 2 SSRI alone vs CMI alone % Remaining stable Months Months SSRI n = 25 CMI n = 17 Z = 0.87 P > CMI: clomipramine; SSRI: selective serotonin reuptake inhibitors; Z: z-test value. 17 patients on clomipramine alone was slightly higher (66.51%); the difference was not significant (Z = 0.87, P >.38). Other predictors of long-term response also were investigated. Initial Y-BOCS scores were not found to be good predictors of long-term response (r = ). However, those patients with lower post-acute Y-BOCS score (r = , P <.008) and, to a lesser extent, those with a greater degree of improvement from initial score (r = 0.219, P <.046) had a significantly increased probability of a longer period of stability (TABLE 2). To further delineate the degree of remission from OCD symptoms, we divided patients into discreet groups based on their Y-BOCS score after acute treatment (TABLE 2). Thirteen patients (15.5%) attained a Y-BOCS <8 and were considered to have achieved full remission. They remained stable for a mean of about 52 months (SD = 23.1). The 22 patients (26.2%) who reached a Y-BOCS of 8 to 11 were considered to have partial remission with mild symptoms and remained stable for a mean 35 months (SD = 23.4), while the 49 (58.3%) with Y-BOCS of 12 to 15 were labeled as having moderate symptoms and remained stable for a mean 30 months (SD = 22.3). These divisions were found to significantly differ in terms of probability of remaining stable (F = 4.87, P <.015). Full remission was found to predict a longer period of stability (vs mild symptoms, P <.035; vs moderate symptoms, P <.003) than partial remission. Mild symptoms were not significantly more predictive of a longer response to treatment than moderate symptoms (P >.39). Other variables also were analyzed for their predictive value (TABLE 3). Those patients with a comorbid depressive and/or anxiety disorder were significantly less likely to maintain stability (29 months stable, SD = 21.8) compared to those without comorbid illness (43 months stable, SD = 24.6) (t = 2.67, P <.01). The number of medications that a given patient needed in order to respond acutely did not predict length of response. Patients who were maintained on a single medication remained stable for approximately 35 months (SD = 24.0), while those receiving combination therapy were stable to a mean of 33 months (SD = 23.8) (t = 0.43, P >.66). Finally, we found that adjunctive CBT was a significant aid in keeping patients from experiencing a return of symptoms on their initial pharmacologic treatment. Those patients who participated in CBT sessions maintained stability for a mean of 44 months (SD = 23.4), while those who did not failed at a mean of 29 months (SD = 22.3). DISCUSSION Our evaluation of data collected in a naturalistic setting revealed some interesting findings. We found that patients have an approximately 40% likelihood of responding for 5 years to the medication on which they initially were stabilized; we also considered what patient and treatment factors would predict a continued period without significant worsening of symptoms. In summary, we found the following to increase significantly the period of stability: (a) the lower the Y-BOCS score and the greater the percent decrease in score following acute treatment, 28 February 2015 Vol. 27 No. 1 Annals of Clinical Psychiatry

5 ANNALS OF CLINICAL PSYCHIATRY (b) meeting criteria for full remission following acute treatment, (c) lack of comorbid depression and anxiety, and (d) adjunctive CBT. The type or number of medications required to achieve initial response did not have a significant impact on how long patients likely were to remain stable. The fact that clomipramine alone was not found to be superior to SSRIs alone in maintaining stability in OCD patients agrees with the majority of head-to-head trials of these medications. 14 Two meta-analyses have shown clomipramine to have superior efficacy to SSRIs, but given the differences in study design between studies, which we have addressed in comparing our own work with others in this area, caution must be exerted before drawing conclusions from these reports. 7,15 Few other studies have described OCD maintenance treatment in a naturalistic setting. Based on our review of the literature, we followed patients longer than any other contemporary naturalistic study of medicated OCD patients (mean 2.9 years, range 1 to 92 months). Six other naturalistic studies of the treated course of long-term OCD were found. Before comparing our study with others, we must emphasize that each study has differences in the way data was obtained and in the variables measured, thus severely limiting the conclusions drawn from comparisons with these studies. For example, Skoog and Skoog 16 presented a collection of data begun 40 years ago to investigate OCD in a naturalistic setting. This Swedish study prospectively examined a cohort of what were considered at that time to be OCD patients in the mid-1950s. Of those initially evaluated, 144 were found for a one-time follow-up 40 years later. This study found 40% of patients significantly improved and 20% completely recovered. Given that few of these patients were treated with biological modalities, and that symptoms were assessed using a different set of criteria and interview, any comparison between our study and this older study would be difficult. Rufer et al 17 conducted a single follow-up with participants between 6 to 8 years (mean 7.2 years) after initial end of trial. With an 81% response rate, 30 of the 37 participated in the follow-up evaluation, all of which were exposed to CBT, 19 patients (63%) received SSRI, while 11 were placed on placebo. An overall mean was found indicating an improvement of a score reduction to 16.4 (45% reduction) in comparison to the average score at pre-test. There were no significant changes between post-treatment and follow-up as well as no significant difference in the level of severity from post-treatment and follow-up. Unlike participants in our study, Rufer et TABLE 2 Acute response as predictor of long-term response Variable n (%) Initial Y-BOCS Y-BOCS score after acute treatment Average Y-BOCS improvement Full remission (Y-BOCS < 8) Partial remission/ mild symptoms (Y-BOCS 8-11) Partial remission/ moderate symptoms (Y-BOCS 12-15) Months stable (SD) 13 (15.5%) (23.1) Probability r = (NS) r = (P <.008) 22 (26.2%) (23.4) F = 4.87 P < (58.3%) (22.3) Full remission vs Mild symptoms P <.035. Full remission vs Moderate symptoms P <.003. Mild symptoms vs Moderate symptoms P >.39 (NS). Y-BOCS: Yale-Brown Obsessive-Compulsive Scale. al 17 reported all but 1 patient received additional forms of treatment, whether medication, CBT, or alternate forms of therapy after the end of the trial. Five other naturalistic studies of OCD have been published within the last 10 years that have much shorter follow-up but that use similar criteria and measure many of the same variables. Orloff et al 18 reported a retrospective review, similar to our review, of 85 patients treated with SSRIs, with 29% participating in adjunctive CBT. They, like Skoog and Skoog 16 looked at the data just once, at 1 to 3.5 years (mean 2.1 years) after initial OCD treatment. They found that 87% of all patients had improved >25% on Y-BOCS score at follow-up. Our study looked at the probability of maintaining a >10% improvement in Y-BOCS after an initial 35% improvement, so a comparison of degree of improvement is not possible. The authors were not able to correlate any variables, including lack of psychiatric comorbidity and CBT, with improvement at follow-up. Our study, in contrast, found that both lack of comorbidity and adjunctive CBT did indeed predict greater probability of remaining stable. Eisen et al 19 published prospective data of a 2-year follow-up of 66 patients pharmacologically treated. The AACP.com Annals of Clinical Psychiatry Vol. 27 No. 1 February

6 MAINTENANCE TREATMENT FOR OCD TABLE 3 Other predictive variables Variable n (%) Months stable (SD) Comorbidity present Comorbidity absent Single drug Combination therapy Underwent CBT No CBT CBT: cognitive-behavioral therapy. 52 (61.9%) 32 (38.1%) 42 (50%) 42 (50%) 30 (35.7%) 54 (64.3%) (21.8) (24.6) (24.0) (23.8) (23.4) 28.7 (22.3) Probability t = 2.67 P <.01 t = 0.43 P >.66 (NS) t = 3.03 P <.004 authors divided patients into those with full remission (Y-BOCS score 0 to 7), partial remission/mild (Y-BOCS 8-11), and partial remission/moderate (Y-BOCS 12 to 15). They found a 47% probability of a partial remission over 2 years (with 48% of those patients relapsing after achieving remission), and a 12% chance of full remission over the same time period. We looked at the same parameters for remission at the post-acute response evaluation. These subcategories of remission predicted duration of response in our study, with 15.5% who had reached full remission after acute treatment remaining stable for a mean months. This study evaluated other configurations of Y-BOCS improvements but not predictors of long-term response. Hantouche et al 20 prospectively followed 130 patients for 1 year. Different scales were used to measure response, including the Montgomery-Åsberg Depression Rating Scale and the National Institute of Mental Health Obsessive-Compulsive scale. This group investigated those that maintained acute response (43% to 64%) at both 6 months and 1 year, as measured by several scales, and those who relapsed or only responded after many months. Predictors of response such as better insight into illness were noted; our study did not examine their set of variables. Zitterl et al 21 reported a 1.5-year prospective follow-up of 70 patients who all received medications and adjunctive CBT as treatment for OCD. At follow-up, 43% of patients had responded, as defined by a Y-BOCS score decrease >25%, and 31.4% who also scored <16 on Y-BOCS after acute treatment. Zitterl et al 21 also noted that persistent depressive symptomatology correlated with worse prognosis, as also is suggested by our data. The authors then postulated that in their patients, depression appeared to be secondary to OCD and that treatment of OCD symptoms should be the focus of treatment in those patients with comorbidities. Alonso et al 22 examined 60 OCD patients for up to 5 years (mean 2.5 years) who were treated with medication, many of whom also received CBT. Like our study, this group reviewed data retrospectively and found that 63% of patients responded with 35% decrease in Y-BOCS score or a final score of 16. Although Alonso et al 22 assessed the number of patients who responded and our study presented the probability of remaining stable, we both have used the same criteria for response/maintaining stability. This group s response rate was higher by >20% compared with our assessment of the likelihood of stability. Also in contrast to our report, no predictive variables were found, including comorbid depression and adjunctive CBT. The comparisons between our study and other naturalistic studies appear in TABLE One can see the significant variance between our study and other studies in duration of mean follow-up (1 to 2.9 years), modes of treatment (for example, CBT), and response at follow-up (40% to 87%). The significant predictors of response also differed between studies, with some showing no predictors, and others showing the predictive value of factors such as insight into illness, depressive or anxiety comorbidity, adjunctive CBT, and a fuller decrease in OCD symptoms in the acute treatment phase. Unfortunately, comparisons between studies must be interpreted with caution because different parameters were used to measure response (for example, 25% vs 35% decrease in Y-BOCS score), different predictive factors were evaluated in different studies (for example, some studies did not examine the effect of comorbidity on response), and different analyses were performed to assess response (for example, % responders vs probability of remaining stable). In addition to the limitations in comparing our evaluation of data with other naturalistic studies, our study must be understood with its own set of limitations. As for all naturalistic studies, the treatment offered for an individual patient is based on the clinical judgment of the treating psychiatrist. The treatment choice could have been biased based on knowledge of 30 February 2015 Vol. 27 No. 1 Annals of Clinical Psychiatry

7 ANNALS OF CLINICAL PSYCHIATRY TABLE 4 Comparison with other recent adult naturalistic studies Study n Duration of follow-up (mean) Adjunctive CBT Response Orloff et al (1994) year (2.1) 29% 87% None Significant predictors of response Eisen et al (1999) year 18% 47% None (Greater acute decrease in Y-BOCS [trend]) Hantouche et al year 19% 43% to 64% More insight (2000) 20 Zitterl et al (2000) year 100% 43% Remission of any depressive symptomatology Alonso et al 60 5 year (2.5) 62% 63% CBT in those with mostly compulsions (2001) 22 CBT: cognitive-behavioral therapy; Y-BOCS: Yale-Brown Obsessive-Compulsive Scale. prior treatment or effect of that treatment on an individual. This bias, if any, is inherent in clinical practice. Once medications were chosen, our expectation was that patients would remain on the medication to which they initially responded. Also, our data was obtained retrospectively. Because the patients were not followed prospectively to a specific endpoint, there is significant variability in how long patients were seen in the clinic. The commonly used technique of LOCF was employed to evaluate those patients who left the study without relapse of OCD symptoms. We do not know how our results would be altered if these patients had remained as participants in our study. In addition, we did not consider patients to have failed if they had a worsening of symptoms (Y-BOCS score between 10% and 35% decrease from baseline) as long as they did not require a change in their medication regimen. Given our use of LOCF and our possible inclusion of those with some worsening of symptoms in the stable patient group, our results for the probability of remaining stable may be too optimistic. An analysis of LOCF as having failed and a further investigation of any patients who had a substantial worsening of symptoms but still met our criteria for stability could be done in the future to present a more potentially lower response probability. Finally, we do not have data available on the exact duration patients who were in CBT remained in therapy; we also do not know the exact CBT techniques used with each patient. We believe that the data in this study is of practical use to front-line clinicians who do not see patients under controlled conditions. While this study and the other naturalistic studies mentioned here are not double-blind, randomized, or placebo-controlled, they still provide quantifiable results from real-world clinical environments. The naturalistic data presented here, in addition to earlier studies, point to the need to develop more uniform criteria so that studies can be more easily compared. DISCLOSURES: The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. REFERENCES 1. Karno M, Golding JM, Sorenson SB, et al. The epidemiology of obsessive-compulsive disorder in five US communities. Arch Gen Psychiatry. 1988;45: Rasmussen SA, Eisen JL. The epidemiology and clinical features of obsessive compulsive disorder. Psychiatr Clin N Am. 1992;15: Clomipramine in the treatment of patients with obsessive-compulsive disorder. The Clomipramine Collaborative Study Group. Arch Gen Psychiatry. 1991; 48: Tollefson GD, Rampey AH Jr, Potvin JH, et al. A multicenter investigation of fixed-dose fluoxetine in the treatment of obsessive-compulsive disorder. Arch Gen Psychiatry. 1994;51: Greist J, Chouinard G, DuBoff E, et al. Double-blind comparison of three doses of sertraline and placebo in outpatients with obsessive-compulsive disorder. Arch Gen Psychiatry. 1995;52: Jenike MA, Hyman S, Baer L, et al. A controlled trial of fluvoxamine in obsessive-compulsive disorder: implications for a serotonergic theory. Am J Psychiatry. 1990;147: Greist JH, Jefferson JW, Kobak KA, et al. Efficacy and tolerability of serotonin transport inhibitors in obsessive-compulsive disorder: a meta-analysis. Arch Gen Psychiatry. 1995;52: Greist JH. Behavior therapy for obsessive-compulsive disorder. J Clin Psychiatry. 1994;55(suppl 10): Baer L. Getting control. Boston, MA: Little Brown & Co; Rasmussen SA, Eisen JL. Treatment strategies for chronic and refractory obsessive-compulsive disorder. J Clin Psychiatry. 1997;58(suppl 13): Ravizza L, Barzega G, Bellino S, et al. Drug treatment of obsessive-compulsive disorder (OCD): long-term trial with clomipramine and selective serotonin reuptake AACP.com Annals of Clinical Psychiatry Vol. 27 No. 1 February

8 MAINTENANCE TREATMENT FOR OCD inhibitors (SSRIs). Psychopharmacol Bull. 1996;32: Romano S, Goodman W, Tamura R, et al. Long-term treatment of obsessive-compulsive disorder after an acute response: a comparison of fluoxetine vs placebo. J Clin Psychopharmacol. 2001;21: Koran LM, Hackett E, Rubin A, et al. Efficacy of sertraline in the long-term treatment of obsessivecompulsive disorder. Am J Psychiatry. 2002;159: Pigott TA, Seay SM. A review of the efficacy of selective serotonin reuptake inhibitors in obsessivecompulsive disorder. J Clin Psychiatry. 1999;60: Piccinelli M, Pini S, Bellantuono C, et al. Efficacy of drug treatment in obsessive-compulsive disorder: a meta-analytic review. Br J Psychiatry. 1995;166: Skoog G, Skoog I. A 40-year follow-up of patients with obsessive-compulsive disorder. Arch Gen Psychiatry. 1999;56: Rufer M, Hand I, Alsleben H, et al. Long-term course and outcome of obsessive-compulsive patients after cognitive-behavioral therapy in combination with either fluvoxamine or placebo: a 7-year follow-up of a randomized double-blind trail. Eur Arch Psychiatry Clin Neurosci. 2004;255: Orloff LM, Battle MA, Baer L, et al. Long-term follow-up of 85 patients with obsessive-compulsive disorder. Am J Psychiatry. 1994;151: Eisen JL, Goodman WK, Keller MB, et al. Patterns of remission and relapse in obsessive-compulsive disorder: a 2-year prospective study. J Clin Psychiatry. 1999;60: ; quiz Hantouche EG, Bouhassira M, Lancrenon S. Prospective follow-up over a 12 month period of a cohort of 155 patients with obsessive-compulsive disorder: phase III National DRT-TOC Study [article in French]. Encephale. 2000;26: Zitterl W, Demal U, Aigner M, et al. Naturalistic course of obsessive compulsive disorder and comorbid depression. Longitudinal results of a prospective follow-up study of 74 actively treated patients. Psychopathology. 2000;33: Alonso P, Menchon JM, Pifarre J, et al. Long-term follow-up and predictors of clinical outcome in obsessivecompulsive patients treated with serotonin reuptake inhibitors and behavioral therapy. J Clin Psychiatry. 2001;62: February 2015 Vol. 27 No. 1 Annals of Clinical Psychiatry