Background : Treatment of depression with a single pharmaceutical agent often does not work, and several agents may be tried or combined to increase efficacy. Augmentation involves the addition of one or more medications to an existing antidepressant monotherapy to enhance mood and overall antidepressant response. Approximately 22% of individuals with unipolar depression are prescribed augmentation strategies. This study examined the effectiveness of augmentation strategies. Methods : A Medline search of studies published before January 1, 2007 was conducted to assess the extent of published data on the most frequently prescribed augmentation strategies. Studies with completed original data, sufficient efficacy data, and participants diagnosed with unipolar depression were included. Letters to the editor, preliminary data, data only presented at conferences, and small uncontrolled case reports were excluded. Results: 13 studies contained sufficient data to calculate an effect size. Mean estimated effect size of all 13 studies calculated with random effects was 0.1782 with a 95% confidence interval of -0.2513-0.6076. Conclusions : There are minimal published data examining antidepressant augmentation, and augmentation is a minimally effective treatment option.
Keywords: meta-analysis; depression; antidepressant; augmentation
Major depressive disorder affects approximately 14.8 million adults in the United States (Kessler, Chiu, Demler, & Walters, 2005) and is the leading cause of disability in the United States for persons between 15 and 44 years of age (World Health Organization [WHO], 2004). The volume of antidepressants sold, and presumed to be consumed, per day in the major developed world is reflective of the prevalence of depression. A comparison of retail and hospital sales of antidepressants in eight major developed countries in 1998 revealed an average of 28 defined daily doses (DDDs) per 1,000 population per day (McManus et al., 2000). The volume of antidepressants sold meets the recommended daily dosage for approximately 3% of the developed world's population.
Antidepressant therapy can be administered as a monotherapy, a series of monotherapies, or as augmentation. Monotherapy is defined as a patient taking only one prescribed antidepressant. A series of monotherapies occurs when administration of one antidepressant is stopped and a new antidepressant is administered. This pattern may repeat itself several times. Augmentation involves the addition of one or more medications to an existing antidepressant monotherapy to enhance mood and overall antidepressant response.
Treatment-resistant depression is commonly defined as an inadequate clinical response, including nonresponse, following at least two trials of properly prescribed antidepressant therapy among patients suffering from depression (Fava & Rush, 2006). It is estimated that at least 50% of individuals who begin treatment with antidepressant monotherapy do not respond, and as many as 30% of individuals treated for major depressive disorder do not benefit from a series of monotherapy trials (Thase, 2004). In addition, full and persistent remissions are uncommon in acute depression trials (Thase, 2004). For example, selective serotonin reuptake inhibitors and serotonin/norepinephrine reuptake inhibitors tend to show remission rates of 25% to 45% in acute trials (Thase, Entsuahm, & Rudolph, 2001). Only 20% to 30% of patients experience a full remission of their depressive symptoms during the first antidepressant trial (Fava & Davidson, 1996; Fava et al., 2003).
Prescribers may pursue a variety of strategies to increase the probability of full remission with antidepressant monotherapy (Fava & Rush, 2006). These options include (a) psychoeducation, (b) enhancing treatment adherence, (c) ensuring adequacy of antidepressant dose, (d) ensuring adequacy of antidepressant treatment duration, (e) prescribing antidepressant medications with relatively greater efficacy in specific subtypes or populations, and (f) psychotherapy (Fava & Rush, 2006). Practice guidelines concerning depression recommend the implementation of additional interventions when patients fail to respond to two or more trials of antidepressant monotherapy. Many prescribers recommend increased dosages when patients fail to achieve full remission (Fava et al., 2003; American Psychiatric Association, 2000; Trivedi, Shon, Crismon, & Key, 2000).
The rationale for augmentation strategies to enhance retention or to increase remission rates is supported by findings from empirical research. First, most patients with unipolar depression do not remit with initial antidepressant monotherapy; second, no monotherapy medication is robustly different from others in achieving remission; third, the lack of response with antidepressant monotherapy leads to high dropout rates among depressed patients; and fourth, the emergence of adverse side effects (e.g., agitation, insomnia) or persistence of some initial baseline symptoms (e.g., anxiety, insomnia) may lead to premature discontinuation from monotherapy (Fava & Rush, 2006; Thase, 2004). Thus, augmentation has the potential to be a viable clinical treatment option.
American Psychiatric Association treatment guidelines describe lithium, thyroid supplements, second antidepressants, second-generation antipsychotic medications, anticonvulsant medications, stimulants, and buspirone as potentially helpful augmenting agents (American Psychiatric Association, 2000; Dording, 2000; Shelton, 2003). Lithium is generally acknowledged to have the strongest empirical support; there are several randomized, controlled trials indicating that lithium augmentation reduces symptoms among patients with unipolar depression (Bauer et al., 2003; Bauer & Dopfmer, 1999; Shelton, 2003). In addition, lithium has been reported to have a specific antisuicide effect for patients with mood disorders (American Psychiatric Association, 2003; Baldessarini, Tondo, & Hennen, 2003). Fava and Rush (2006) state that lithium augmentation is the most wellsupported augmenting agent. These authors note that lithium was more commonly used in the 1980s.
There are noteworthy limitations to the augmentation strategies recommended by the American Psychiatric Association. First, researchers have noted a lack of data to inform the sequence in which augmentation strategies should be implemented or for identifying the types of patients for whom specific strategies might be most helpful (Fava et al., 2003; Stimpson, Agrawal, & Lewis, 2002). Further, it appears that the majority of medications suggested as augmenting agents have no studies, or few examining them as augmenting agents (Thase, Entsuahm, & Rudolph, 2001). Thus, it is an irony of current psychiatric practice that the most common augmentation strategies for the treatment of depression may be those with the least evidence of efficacy (Thase, 2004).
The STAR*D study was a large trial that included some augmentations (Rush et al., 2006). The STAR*D study was a practical effectiveness trial that had no placebo comparisons. In the trial, patients who had failed to respond to citalopram and agreed to be randomized to augmentation strategies received either 12 weeks of bupropion or buspirone added to their ongoing citalopram in phase two of the trial. Such augmentation resulted in up to a 39% remission rate in this trial, with a subsequent relapse rate of up to 67% (Rush et al., 2006). Subsequent augmentations had even lower remission and higher relapse rates. Without a placebo comparison, the interpretation of these results (i.e., whether they are positive or negative) is difficult. It could be argued that any further response was positive among those patients failing prior trials. However, one might also consider the results to be disappointing and reflect an unfavorable risk-benefit ratio given that about 4% of patients experienced a serious adverse event from the first augmentation and that about 13% had to discontinue the study because of intolerance of side effects.
In an evaluation of current augmentation prescribing patterns within Veteran Administration mental health settings, Valenstein et al. (2006) examined the frequency of antidepressant augmentation strategies prescribed to 244,859 patients with a diagnosis of unipolar depression and an antidepressant prescription during the fiscal year 2002. The vast majority (90%) of the patients in the sample were male, and 22% were over 65 years of age. Approximately 6% of the depressed patients had had a psychiatric hospitalization during the prior year. Comorbid anxiety and substance abuse disorders were common, with 28% having a diagnosis of posttraumatic stress disorder, 24% having a co-occurring diagnosis of anxiety disorder, and 21% having a co-occurring diagnosis of substance abuse during the study year.
Results of the Valenstein et al. (2006) analysis show that approximately 22% of the depressed VA patients treated in mental health settings received antidepressant augmentation, with approximately 4% receiving more than one augmenting agent during the year. The most popular augmenting agents were a second antidepressant received by 11% of patients, and a second-generation antipsychotic received by 7% of the patients. Approximately 4% of patients received augmentation with anticonvulsants, and 5% received other augmenting agents. Only 0.5% patients received augmentation with lithium.
Based on the results of Valenstein et al. (2006), Antonuccio, Yury, Valenstein, and Matuszak (2008) conducted a literature review of the most frequently prescribed augmentation strategies. This review indicated that 22 studies have been published that examine augmentation strategies for unipolar depression and that only six of these studies used placebo controls. Further, the review indicated that more than 40% (3 out of 7) of the most popular classes of augmentation strategies, and more than 55% (11 out of 19) of the most frequently prescribed combinations of drugs used for augmentation for unipolar depression have no published scientific support. Another main conclusion of this review was that safety was rarely the focus, with only eight studies summarizing safety results in a dedicated safety section.
Based on the Valenstein et al. (2006) sample it appears that antidepressant augmentation is a common practice. Contrastingly, the Antonuccio et al. (2008) review indicates that data addressing the potential risks and benefits of augmentation are limited. Therefore, the aim of the present study was to examine the efficacy of augmentation strategies through meta-analytical analyses of published articles. A meta-analysis was conducted from the data obtained from the Antonuccio et al. (2008) literature review. Cohen's d was calculated from depression inventories used in the studies and was combined to create an average effect size for augmentation strategies.
Studies in the current analysis were identified in a review conducted by Antonuccio et al. (2008). In that review, a Medline search of studies published before January 1, 2007 was conducted to assess the extent of published data on the most frequently prescribed classes of drug augmentation strategies identified by Valenstein et al. (2006). For each augmentation strategy the limits of Medline were set to search for published articles containing a combination of either the generic drug name (listed in Table 1), class of drug (listed in Table 1), or the term antidepressant in either the title, abstract, or keywords.
Antonuccio et al. (2008) restricted the review to studies that met each of the following criteria:
1. Published articles.
2. Completed original data articles. Letters to the editor, preliminary data, small uncontrolled case reports, and data only presented at conferences were excluded.
3. Studies with sufficient data to calculate an effect size.
4. Participants were diagnosed with unipolar depression.
To assess the efficacy of augmentation strategies, a meta-analysis was conducted with the published articles having sufficient obtainable data to calculate effect sizes. Metaanalysis is a statistical analytic method that combines data from several studies, thereby increasing the sample size and overcoming the power limitations of undersized studies (Thalheimer & Cook, 2002). Meta-analyses utilize effect sizes, or the differences between group means, as their main statistic. The effect size statistic is also an indication of the amount of variability in the dependent variable that can be accounted for by the treatment (Jacobson & Traux, 1991). Utilizing effect sizes inform us about the relative magnitude of the treatment and allows for a comparison of the magnitude of treatments from one experiment to another (Thalheimer & Cook, 2002).
For treatment efficacy, effect sizes were calculated as a comparison of the mean completed score of the treatment group and the mean baseline score of the treatment group. For studies listing only one primary source of measurement, this measure was used for the effect size. In studies reporting multiple measures of efficacy, the most commonly used measures among the identified studies were selected.
Calculation of Effect Size
Effect sizes were computed as Cohen's d, where a positive effect size represents improvement and a negative effect size represents a worsening of symptoms. Cohen (1988) defined effect sizes as "small, d = .2," "medium, d = .5," and "large, d = .8". In most cases, effect sizes were calculated for efficacy as the mean posttreatment score minus the mean pretreatment score, divided by the pooled standard deviation (Thalheimer & Cook, 2002). For studies that do not list standard deviations, the standard deviations were calculated as standard error times the square root of the number of participants.
Computation of the meta-analysis was done with the MIX (Meta-analysis with Interactive eXplanations) program (Bax, Yu, Ikeda, Tsurata, & Moons, 2006). Comparative data were loaded by means of association measures with their standard error. The random effects (RE) model was used for this analysis because RE models allow for the possibility that the population parameter values vary from study to study (Becker, 1996; Hedges, 1992). Further, fixed effects models typically manifest substantial type I bias and yield confidence intervals for mean effect sizes that are narrower than the nominal width, thereby overstating the degree of precision (Hunter & Schmidt, 2000). RE models do not share these biases (Hunter & Schmidt, 2000).
Using the search strategy described above, 22 studies of the most common augmentation strategies were identified. Unfortunately many studies chose to use the less preferred method of presenting percentage of patients who responded to medication and did not present data that would allow for a statistical evaluation of effectiveness. Thirteen studies contained enough data to calculate an effect size, had baseline and end-point data, and were not part of a series of augmentations. The number of participants in the analysis was 1,617, with the smallest study enrolling 7 participants and the largest 552. The duration that participants received monotherapy antidepressant ranged from 3 to 6 weeks, and the length of time participants were administered augmenting agents ranged primarily from 3 to 12 weeks, with one study lasting 76 weeks. The STAR*D (Rush et al., 2006) study was excluded from the sample because it is a series of trials designed to assess clinical practice as opposed to efficacy of augmentation. Data were collected based on treatment steps, and each step included different treatment options. Thus, data do not indicate the effectiveness of specific augmentation strategies.
The Hamilton Rating Scale for Depression (Hamilton, 1960) was the most widely used assessment instrument, and was a primary assessment instrument in 11 studies. The Hamilton Rating Scale for Depression is a 17-21-item observer-rated scale to assess presence and severity of depressive symptoms. It is one of the widely used instruments for clinical assessment of depressive symptoms. It is established as a highly reliable and valid assessment tool (Bech, 1990). The Montgomery-Asberg Depression Rating Scale ( Montgomery & Asberg, 1979) was used as the primary assessment instrument in two studies. The Montgomery- Asberg Depression Rating Scale was developed to be a simpler rating scale than the Hamilton Depression Scale while remaining responsive to change in the patient's state. Thus, it can be used to monitor a patient's depressive symptoms over time ( Montgomery & Asberg, 1979). For studies that used several instruments, the Hamilton Rating Scale for Depression data was used in calculations.
Mean estimated effect size of all 13 studies calculated with random effects was 0.1782 with a 95% confidence interval of -0.2513 to 0.6076. Table 2 displays the individual effect sizes for each study. There were insufficient data to create a placebo control mean estimate effect size to compare the treatment mean estimate effect size.
Eight studies used an antipsychotic augmenting agent; the mean estimated effect size for this analysis was 0.1561 with a 95% confidence interval of -0.2928 to 0.6052. Table 3 displays the individual effect sizes for each study. Five studies used an antidepressant augmenting agent. For this analysis the mean estimated effect size was 0.4046 with a 95% confidence interval of -1.0151 to 1.8243. Table 4 displays the individual effect sizes for each study.
Results of the present analysis indicate that effect sizes for antidepressant augmentation for unipolar depression are small. The overall mean estimated effect size of 0.1782 is small, as is the 0.1563 mean estimated effect size of the antipsychotic augmentation and the 0.4159 mean estimated effect size of the antidepressant augmentation. The confidence intervals for all of these calculations include 0 and are therefore not statistically signifi- cant. Unfortunately, there were insufficient data to calculate placebo effect sizes, thus estimating the proportion of drug response duplicated by placebo is not possible. Kirsch and Saperstein (1998) examined effect sizes of wait-list and no-treatment patients with a primary diagnosis of depression, as part of a larger meta-analysis examining antidepressants and psychotherapy. These authors reported a 0.37 pre-post effect size for wait-list and no-treatment control groups. Although Kirsch and Saperstein's analysis examined monotherapy antidepressant treatment and psychotherapy, the wait-list and no-treatment comparison does indicate the amount of change that is expected by extraneous variables or measurement error.
No underdosing or overdosing of medications was described in any of the studies involved in the analysis. Reported dosage levels are all within acceptable dosage levels. Dose-response levels were not calculated because included studies either maintained dosages in the acceptable ranges or only reported average dosages. Thus, dosage levels for all studies approximated the mean level of acceptable dosage for each drug.
Some limitations to our meta-analysis restrict the interpretation of the results. Fava (2006) points out that utilizing only published data can be a serious risk in that financial conflicts of interest may contribute to a publication bias. Including only published data may exclude an unknown number of unpublished studies. The extreme view of the socalled file drawer problem is that journals publish only the 5% of the studies that show type I errors, while the file drawers are filled with the 95% of the studies that show nonsignificant results (Antonuccio, Burns, & Danton, 2002). Thus, it is probable that effect sizes may be smaller than found in this analysis (Turner, Matthews, Linardatos, Tell, & Rosental, 2008).
All studies included in this analysis concluded that their specific data reflected a successful treatment. Results of the meta-analysis contradict this conclusion in that the overall effect size is small and nonsignificant. Effect size is an indication of the amount of variability in the dependent variable that can be accounted for by the treatment, whereas clinical significance refers to a treatment's ability to meet standards of efficacy set by consumers, clinicians, and researchers (Jacobson & Traux, 1991). A lack of placebo comparisons in the current analysis, and past research indicating that wait-list control groups produce similar effect sizes, cast further doubt on whether reductions in scores are clinically meaningful. Studies in the analysis do show a reduction in depression assessment scores, but no data were taken reflecting normative levels of functioning or an elimination of the presenting problems to aid in the determination of clinical significance.
Many of the medications examined in this analysis have been available for two decades, yet data on the most common augmentation strategies continue to be extremely limited. The recent publication of a major augmentation study for bipolar patients offers some hope that the science of drug augmentation is beginning to evolve (Sachs et al., 2007). In the mean time, it may be prudent for clinicians and patients to consider the sparse scientific data supporting most augmentation strategies when deciding whether or not to augment antidepressants. It is necessary to have well-controlled studies, that demonstrate superior efficacy when compared to placebo, to determine whether augmentation is an effective treatment option. Currently, the evidence available for antidepressant augmentation is weak and suggests that other scientifically supported treatment options, including psychological interventions, be considered for individuals who do not respond to monotherapy.
American Psychiatric Association. (2000). Practice guideline for the treatment of patients with major depressive disorder (revision). Washington, DC: Author.
American Psychiatric Association. (2003). Practice guideline for the assessment and treatment of patients with suicidal behaviors. American Journal of Psychiatry, 160 (Suppl.), 1-60; erratum, 161, 776.
Antonuccio, D. O., Burns, D. D., & Danton, W. G. (2002). Antidepressants: A triumph of marketing over science? Prevention and Treatment, 5, article 25. Retrieved September 1, from http:// www.antidepressantsfacts.com/2002-07-15-Antonuccio-therapy-vs-med.htm
Antonuccio, D., Yury, C. A., Valenstein, M., & Matuszak, J. (2008). Common augmentation strategies for depression: Findings show lack of evidence. Psychiatric Times, 25 (3), 21-22.
Baldessarini, R. J., Tondo, L., & Hennen, J. (2003). Lithium treatment and suicide risk in major affective disorders: Update and new findings. Journal of Clinical Psychiatry, 64, 44-52.
Bauer, M., Adli, M., Baethge, C., Berghofer, A., Sasse, J., Heinz, A., et al. (2003). Lithium augmentation therapy in refractory depression: Clinical evidence and neurobiological mechanisms. Canadian Journal of Psychiatry, 48, 440-448.
Bauer, M., & Dopfmer, S. (1999). Lithium augmentation in treatment-resistant depression: Metaanalysis of placebo-controlled studies. Journal of Clinical Psychopharmacology, 19, 427-434.
Bax, L., Yu, L. M., Ikeda, N., Tsuruta, N., & Moons, K. G. (2006). Development and validation of MIX: Comprehensive free software for meta-analysis of causal research data. BMC Medical Research Methodology, 6, 50.
Bech, P. (1990). Psychometric development of the Hamilton scales: The spectrum of depression, dysthymia and anxiety. In P. Bech & A. Coppen (Eds.), The Hamilton scales (pp. 72-79). Berlin: Springer-Verlag.
Becker, B. J. (1996). The generalizability of empirical research results. In C. P. Benbow & D. Lubinski (Eds.), Intellectual talent: Psychological and social issues (pp. 363-383). Baltimore: Johns Hopkins University Press.
Bondolfi, G., Chautems, C., Rochat, B., Bertschy, G., & Baumann, P. (1996). Non-response to citalopram in depressive patients: Pharmacokinetic and clinical consequences of a fluvoxamine augmentation. Psychopharmacology, 128 (4), 421-425.
Carpenter, L. L., Yasmin, S., & Price, L. H. (2002). A double-blind, placebo-controlled study of antidepressant augmentation with mirtazapine. Biological Psychiatry, 51, 183-181.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Corya, S. A., Andersen, S. W., Detke, H. C., Kelly, L. S., Van Campen, L. E., Sanger, T. M., et al. (2003). Long-term antidepressant efficacy and safety of olanzapine/fluoxetine combination: A 76-week open-label study. Journal of Clinical Psychiatry, 64, 1349-1356.
Corya, S. A., Williamson, D., Sanger, T. M., Briggs, S. D., Case, M., & Tollefson, G. (2006). A randomized, double-blind comparison of olanzapine/fluoxetine combination, olanzapine, fluoxetine, and venlafaxine in treatment-resistant depression. Depression and Anxiety, 23, 364-372.
Dording, C. M. (2000). Antidepressant augmentation and combinations. Psychiatric Clinics of North America, 23, 743-755.
Fava, G. A. (2006). The intellectual crisis of psychiatric research. Psychotherapy and Psychosomatics , 75, 202-208.
Fava, M., & Davidson, K. G. (1996). Definition and epidemiology of treatment-resistant depression. Psychiatric Clinics of North America, 19, 179-200.
Fava, M., & Rush, A. J. (2006). Current status of augmentation and combination treatments for major depressive disorder: A literature review and a proposal for a novel approach to improve practice. Psychotherapy and Psychosomatics, 75, 139-153.
Fava, M., Rush, A. J., Trivedi, M. H., Nierenberg, A. A., Thase, M. E., Sackeim, H. A., et al. (2003). Background and rationale for the sequenced treatment alternatives to relieve depression (STAR*D) study. Psychiatric Clinics of North America, 26, 457-494.
Ferreri, M., Lavergne, F., Berlin, I., Payan, C., & Puech, A. J. (2001). Benefits from mianserin augmentation of fluoxetine in patients with major depression non-responders to fluoxetine alone. Acta Psychiatrica Scandinavica, 103, 66-72.
Hamilton, M. A. (1960). Rating scale for depression. Journal of Neurology, 23, 56-62.
Hedges, L. V. (1992). Meta-analysis. Journal of Educational Statistics, 17, 279-296.
Hunter, J. E., & Schmidt, F. L. (2000). Fixed effects vs. random effects meta-analysis models: Implications for cumulative research knowledge. International Journal of Selection and Assessment, 8 (4), 275-292.
Jacobson, N. S., & Traux, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12-19.
Kennedy, S. H., McCann, S. M., Masellis, M., McIntyre, R. S., Raskin, J., McKay, G., et al. (2002). Combining bupropion sr with venlafaxine, paroxetine, or fluoxetine: A preliminary report on pharmacokinetic, therapeutic, and sexual dysfunction effects. Journal of Clinical Psychiatry, 63, 181-186.
Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (2005). Prevalence, severity, and comorbidity of twelve-month DSM-IV disorders in the National Comorbidity Survey Replication (NCS-R). Archives of General Psychiatry, 62 (6), 617-627.
Kirsch, I., & Saperstein, G. (1998). Listening to Prozac but hearing placebo: A meta-analysis of antidepressant medication. Prevention and Treatment, 1 (1), 2a.
Konig, F., Kippel, C., Petersdorff, T., Neuhoffer-Weiss, M., Wolfersdorf, M., & Kaschka, W. P. (2001). First experiences in combination therapy using olanzapine with SSRIs (citalopram, paroxetine) in delusional depression. Neuropsychobiology, 43, 170-174.
Lam, R. W., Hossie, H., Solomons, K., & Yatham, L. N. (2004). Citalopram and bupropion-sr: Combining versus switching in patients with treatment-resistant depression. Journal of Clinical Psychiatry, 65, 337-340.
McManus, P., Mant, A., Mitchell, P. B., Montgomery, W. S., Marley, J., & Auland, M. E. (2000). Recent trends in the use of antidepressant drugs in Australia, 1990-1998. The Medical Journal of Australia, 173, 458-461.
Montgomery, S. A., & Asberg, M. (1979). A new depression scale designed to be sensitive to change. British Journal of Psychiatry, 134, 382-389.
Papakostas, G. I., Petersen, T. J., Kinrys, G., Burns, A. M., Worthington III, J. J., Alpert, J. E., et al. (2005). Aripiprazole augmentation of selective serotonin reuptake inhibitors for treatmentresistant major depressive disorder. Journal of Clinical Psychiatry, 66, 1326-1330.
Papakostas, G. I., Petersen, T. J., Nierenberg, A. A., Murakami, J. L., Alpert, J. E., Rosenbaum, J. F., et al. (2004). Ziprasidone augmentation of selective serotonin reuptake inhibitors (SSRIs) for SSRI-resistant major depressive disorder. Journal of Clinical Psychiatry, 65, 217-221.
Rapaport, M. H., Gharabawi, G. M., Canusco, C. M., Mahmoud, R. A., Keller, M. B., Bossie, C. A., et al. (2006). Effects of risperidone augmentation in patients with treatment-resistant depression: Results of open-label treatment followed by double-blind continuation. Neuropsychopharmacology, 31, 2505-2513.
Rothschild, A. J., Williamson, D. J., Tohen, M. F., Schatzberg, A., Andersen, S. W., Van Campen, L. E., et al. (2004). A double-blind, randomized study of olanzapine and olanzapine/fluoxetine combination for major depression with psychotic features. Journal of Clinical Psychopharmacology, 24 (4), 365-373.
Rush, A. J., Trivedi, M. H., Wisniewski, S. R., Nierenberg, A. A., Stewart, J. W., Warden, D., et al. (2006). Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: A STAR*D report. American Journal of Psychiatry, 163, 1905-1917.
Sachs, G. S., Nierenberg, A. A., Calabrese, J. R., Marangell, L. B., Wisniewski, S. R., Gyulai, L., et al. (2007). Effectiveness of adjunctive antidepressant treatment for bipolar depression. New England Journal of Medicine, 356 (17), 1711-1722.
Shelton, R. C. (2003). The use of antidepressants in novel combination therapies. Journal of Clinical Psychiatry, 64 (Suppl. 2), 14-18.
Simon, J. S., & Nemeroff, C. B. (2005). Aripiprazole augmentation of antidepressants for the treatment of partially responding and nonresponding patients with major depressive disorder. Journal of Clinical Psychiatry, 66, 1216-1220.
Stimpson, N., Agrawal, N., & Lewis, G. (2002). Randomised controlled trials investigating pharmacological and psychological interventions for treatment-refractory depression: Systematic review. British Journal of Psychiatry, 181, 284-294.
Thalheimer, W., & Cook, S. (2002). How to calculate effect sizes from published research articles: A simplified methodology . Retrieved January 20, 2005, from http://worklearning.com/effect_ sizes.htm
Thase, M. E. (2004). Therapeutic alternatives for difficult-to-treat depression: A narrative review of the state of the evidence. CNS Spectrums, 9, 808-821.
Thase, M. E, Entsuahm, A. R., & Rudolph, R. L. (2001). Remission rates during treatment with venlafaxine or selective serotonin reuptake inhibitors. British Journal of Psychiatry, 178, 234-241.
Trivedi, M. H., Shon, S., Crismon, M. L., & Key, T. (2000). Texas implementation of medication algorithms (TIMA): Guidelines for treating major depressive disorder: TIMA physician procedure manual . Dallas, TX: University of Texas Southwestern Medical School at Dallas.
Turner, E. H., Matthews, A. M., Linardatos, E., Tell, R. A., & Rosental, R. (2008). Selective publication of antidepressant trials and its influence on apparent efficacy. New England Journal of Medicine, 358, 252-260.
Valenstein, M., McCarthy, J. F., Austin, K. L., Greden, J. F., Young, E. A. & Blow, F. C. (2006). What happened to lithium? Antidepressant augmentation in clinical settings. American Journal of Psychiatry, 163 (7), 1219-1225.
World Health Organization. (2004). The World Health Report 2004: Changing history, annex table 3: Burden of disease in DALYs by cause, sex, and mortality stratum in WHO regions, estimates for 2002. Geneva: Author.
Acknowledgments. Dr. Antonuccio reports the following competing interests: author of smoking cessation book Butt Out ; past recipient of NIDA and NCI funding for smoking cessation research; past recipient (1995) of funding from Marion Merrill Dow for research on the nicotine patch; workshops on the treatment of depression; expert witness on depression, nicotine dependence, or PTSD; private practice in clinical psychology. Drs Fisher, Valenstein, Matuszak, and Yury report no conflicts of interest concerning the subject matter of this article.
Craig A. Yury, PhD
University of Manitoba
Jane E. Fisher, PhD
David O. Antonuccio, PhD
University of Nevada, Reno
Marcia Valenstein, MD
Ann Arbor VA
Jeremy Matuszak, MD
University of Nevada, Reno
Correspondence regarding this article should be directed to Craig Yury, PhD, Department of Clinical Health Psychology, A4 Brandon Regional Health Center, 150 McTavish Ave E., Brandon, MB R7A 2B3. E-mail: firstname.lastname@example.org…