ABSTRACT FROM: Barnes TR, Leeson VC, Paton C, et al. Antidepressant Controlled Trial For Negative Symptoms In Schizophrenia (ACTIONS): a double-blind, placebo-controlled, randomised clinical trial. Health Technol Assess 2016;20:1–46.
What is already known on this topic?
Persistent negative symptoms which do not respond to antipsychotic regimens are experienced by 15–20% of patients with schizophrenia.1 Antidepressants have been the most studied augmentation strategy in these patients;2 however, previous findings are inconsistent in terms of their potential benefits.
Methods of the study
In this double-blind, placebo-controlled, parallel-group, 15-centre pragmatic trial, 62 patients with schizophrenia with persistent primary negative symptoms were randomised in a 1:1 ratio to receive either citalopram (20 mg/day) or placebo as an adjunctive therapy to the baseline ongoing stable dose of second generation antipsychotic treatment for 48 weeks. Patients were excluded in case of receiving any antidepressants or any drug that risked interaction with citalopram at baseline. Between-group differences in Heinrich's Quality of Life Scale (QLS) score at weeks 12 and 48, and Positive and Negative Syndrome Scale (PANSS) negative subscale score at week 12 were measures of primary outcomes. Calgary Depression Rating Scale for Schizophrenia (CDSS), PANSS total and subscales and a combination of scales for side effects were used to measure secondary outcomes at weeks 12 and 48. Finally, for purpose of health economics analysis, the European Quality of Life-5 Dimensions, 3 Levels (EQ-5D-3L) was used as primary measure alongside ICEpop CAPability measure for Adults (ICECAP-A), and numerous variables to account for all health and social services use related to the disease as well as the intervention and its potential side effects at weeks 12, 36, and 48. All outcomes were adjusted for baseline differences.
What does this paper add?
This pragmatic trial addresses an important challenge in treatment of patients with schizophrenia, especially those with persistent negative symptoms, from both clinical efficacy and cost-effectiveness aspects and therefore provides a relatively new perspective to this issue.
With the longest duration of follow-up among other trials of its kind, this study indicates that the long-term effects of citalopram are not significantly different from placebo on quality of life (end point difference in QLS score −0.14, 95% CI −10 to 10.1) or negative symptoms (PANSS negative subscale score −1.4, −5.2 to 2.3), as well as PANSS total score (−5.1, −13.7 to 3.4) or depressive symptoms (CDSS score −0.8, −3.1 to 1.4) in patients with schizophrenia.
Interestingly, no significant differences were reported between citalopram and placebo groups for long-term health economics in terms of service-use, side effects and quality-of-life/well-being (p values of 0.25 and 0.248 for EQ-5D-3L and ICECAP-A, respectively).
Limitations
In addition to the small baseline sample size (citalopram, n=30 and placebo, n=32), high rates of loss to follow-up (especially in the citalopram group) substantially reduced the power of this study and potentially introduced bias to the results (week-12 completers: citalopram, n=22 and placebo, n=25; week-48 completers: citalopram, n=19 and placebo, n=23).
Only mentioning that ‘all analyses were conducted under the intention-to-treat principle’ to assess primary and some secondary outcomes (ie, QLS and PANSS scores) is not a clear definition of intention-to-treat (ITT) analysis. In fact, it seems that per-protocol analysis (considering completers only) is applied instead. It is noteworthy that such an analysis may significantly distort the results of the statistical analysis and proper ITT analysis is particularly important in pragmatic trials.3
The sample included both inpatients and outpatients from 15 centres, which implies the involvement and training of many raters. However, no information is provided on the number of raters and inter-rater reliability in this study. The inter-rater reliability is very important in such a study considering the number of centres and also considering that patients had variety of diagnoses including schizophrenia, schizoaffective disorder, schizophreniform disorder and other psychotic disorders.
What next in research?
Future trials should recruit larger samples (the target recruitment for this study was 358 participants) and reduce the incidence of early dropouts from the study. This will require involvement of several centres and longer recruitment periods, which will turn up in higher costs.
Less heterogeneity in the sample and narrow range of antipsychotic regimens will provide valuable information regarding whether certain subgroups of patients with schizophrenia can benefit from a certain combination of antidepressants and antipsychotics.
Do these results change your practices and why?
No. Heterogeneity and inconsistency of the few published trials make it hard to draw a conclusion on the effects of adjunctive citalopram or other antidepressants in patients with schizophrenia, particularly in terms of negative symptoms. Meanwhile, avoiding unnecessary polypharmacy will be the choice until further comprehensive investigations provide us with consistent findings.
Footnotes
Competing interests: None declared.
Provenance and peer review: Commissioned; internally peer reviewed.
References
- 1.Buchanan RW. Persistent negative symptoms in schizophrenia: an overview. Schizophr Bull 2007;33:1013–22. 10.1093/schbul/sbl057 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barnes TR, Schizophrenia Consensus Group of British Association for Psychopharmacology. Evidence-based guidelines for the pharmacological treatment of schizophrenia: recommendations from the British Association for Psychopharmacology. J Psychopharmacol 2011;25:567–620. 10.1177/0269881110391123 [DOI] [PubMed] [Google Scholar]
- 3.Hernán MA, Hernández-Díaz S. Beyond the intention-to-treat in comparative effectiveness research. Clin Trials 2012;9:48–55. 10.1177/1740774511420743 [DOI] [PMC free article] [PubMed] [Google Scholar]