|
A clinical and economic evaluation of selective serotonin reuptake inhibitors in major depression |
Ausejo M, Glennie J L |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Selective serotonin reuptake inhibitors (SSRIs) for first line treatment of depression.
Type of intervention Treatment and rehabilitation.
Economic study type Cost-effectiveness analysis and cost-utility analysis.
Study population Male and female patients, over 18 years of age, and suffering from depression.
Setting Hospital and community. The economic study was carried out in Canada.
Dates to which data relate Effectiveness evidence was derived from studies published between 1980 and mid 1996. Estimates of resource use were determined using experts' opinion or authors' assumptions. Prices refer either to 1994 or 1996.
Source of effectiveness data The evidence for the final outcomes was derived from a synthesis of previously completed studies.
Modelling Decision analysis techniques were used to model the events occurring as patients follow one of three treatment paths:
(1) SSRI followed by a TCA in the case of drop-out;
(2) a TCA followed by an SSRI in the event of a drop-out; or
(3) TCA only.
Outcomes assessed in the review The outcomes assessed in the review were clinical efficacy, rates of improvement (completion rates and drop-out rates, as proxies for the acceptability of the medication) and frequency of adverse events after the antidepressant treatment. The clinical efficacy of the anti-depressants was measured by the effect size differences as reflected by the HRSD (Hamilton Rate Scale for Depression) score. The rates of improvement were measured by the rates of improvement either in the HRSD or CGI (Clinical Global Impression) scores or both. The number of adverse events determined the frequency of adverse events associated with the corresponding antidepressant.
Study designs and other criteria for inclusion in the review The clinical evidence was derived from the literature involving randomized controlled trials, clinical trials and reviews from the period 1980 - mid-1996. Then the source studies were limited to those involving randomized trials of one SSRI versus another SSRI, another antidepressant or placebo. These trials were carried out in a single center or simultaneously in several centres, in an inpatient or outpatient setting, and involved adult and elderly patients.
Sources searched to identify primary studies Not specified. A systematic search of the published, unpublished and computerised sources was carried out.
Criteria used to ensure the validity of primary studies The studies included were double blind, randomized controlled trials of antidepressant therapy with a duration between 4 and 12 weeks for major depression (DSMR-III-R or DSM-IV degree). The results of the trials had to be presented in one or more of the following forms: pre- and post-treatment HRSD scores; change in HRSD score following treatment; number of patients experiencing 50% or more improvement in HRSD score following treatment; number of patients whose Clinical Global Impression (CGI) scores improved by a lot or markedly following treatment; number of drop-outs; number of adverse events. When the adverse event was reported in less than 6 studies, or was reported in a trial involving less than 20 patients, or was very rare (although sometimes serious), the adverse event was excluded from the analysis.
Methods used to judge relevance and validity, and for extracting data Number of primary studies included 162 papers were included in the review.
Methods of combining primary studies Investigation of differences between primary studies The authors did not show how they investigated the differences between the primary studies with respect to study designs, outcomes measured, differences in participants etc. The authors only mentioned that, along with the total analysis, they carried out a subgroup analysis. The subgroups were formed according to the patient's age, dosing level, and treatment setting. The authors did not carry out statistical tests of homogeneity of the results obtained from the individual studies. The authors reported how differences in the primary studies affected the final estimates.
Results of the review The mean difference in effect size between using SSRIs and placebo was 0.55 (95% CI: 0.400.70). This difference was evaluated on the basis of the information extracted from 48 studies and was statistically significant, showing that the SSRIs have a moderate average antidepressant effect. The efficacy (change in HSRD scores) of one type of SSRI was compared to another type of SSRI by measuring the mean differences of their effect size. The mean difference varied between 0.047 and 0.113 depending on the SSRIs being compared and their 95% confidence intervals all crossed zero showing no significant difference between the efficacy of different SSRIs. The latter result was based on the information extracted from 10 studies. The mean difference in effect size of using SSRIs versus classical TCAs (imipramine, amitriptyline, or clomipramine), based on46 primary studies was -0.05 (95% CI: -0.180.09), showing a slight trend to superior results with TCAs. However, the result was not statistically significant. The mean difference in effect size of using SSRIs versus the other TCAs and antidepressants considered in the study, based on 117 primary studies, was -0.01 (95% CI: -0.08 -0.06), showing no statistically significant difference in the use of these two types of antidepressants. The results reported in the study were calculated using the HSRD score. The same conclusions, although the numerical results were not explicitly reported in the study, were reached using the CGI scores. The subgroup analysis also led to the same conclusions although the numerical results were also not reported in this study. Completion rates were evaluated for SSRIs versus placebo, SSRIs amongst themselves, and SSRIs versus other TCAs and other antidepressants, but showed no statistically significant differences between the different treatments. Drop-out rates (due to adverse events, lack of effect or worsening of symptoms) were evaluated for SSRIs versus TCAs but also showed no statistically significant differences between the different treatments. The subgroup analysis of the drop-out rates showed a statistically significant decrease in drop-outs of adults (versus elderly) and outpatients versus inpatients among the patients who were treated with SSRIs. Rates of frequency of adverse events were evaluated for SSRIs versusTCAs. Those adverse events which occurred significantly more frequently with SSRIs were: nausea, anorexia, diarrhea, anxiety, agitation, insomnia, and nervousness. Those adverse events which occurred significantly more frequently with TCAs were: dry mouth, constipation, blurred vision, and dizziness.
Methods used to derive estimates of effectiveness Estimates of effectiveness were also based on authors' assumptions.
Estimates of effectiveness and key assumptions The utility values were measured in terms of 8 alternative degrees of health states as in Revicki (1995). The main assumptions in the study were that drop-out rates did not depend on previous therapy, that during the first three months, no switch of medication occurred and that the TCA replace rate was the same as that for SSRIs.
Measure of benefits used in the economic analysis The authors assessed the clinical outcomes by a dichotomous measure: success (patients who did not drop out by the end of the time horizon) and failure (patients who dropped out). The cost-utility outcomes were measured in terms of quality-adjusted life years (QALYs). The authors used the literature to assess the health states.
Direct costs Costs were not discounted as the time horizon was less than 12 months. Costs included in the study were: costs of daily doses and increased daily doses of SSRI and TCA; daily cost of hospital stay; cost of a GP visit; cost of a psychiatrist consultation; and cost of laboratory tests. Only costs to the health care system were considered. The estimation of quantities was an estimation based on published information in other studies. The estimation of costs and quantities was based on a guess using actual data about the prices of these medications and rates of GP and psychiatric visits in Quebec. As stated by the authors, the perspective adopted was that of the Ministry of Health. Some of the costs were measured in 1996 Canadian dollars, and others in 1994 Canadian dollars. In the analysis, no cost was attributed to the management of adverse events, except within the sensitivity analysis.
Statistical analysis of costs Sensitivity analysis One-way sensitivity analysis was carried out, in which the values of the variable estimates for drop-out rates, replace rates, physician and psychiatrist costs, drug costs, hospitalisation costs and utility values were varied with respect to the base line case. Also, Newfoundland psychiatrist fees were used in the evaluation instead of the Quebec fees. The uncertainty investigated in the sensitivity analysis concerns the validity of data and analytical methods (model assumptions).
Estimated benefits used in the economic analysis In the first three months (short term) horizon, the base case results of the cost-effectiveness analysis showed that the expected probability of success (incremental success rate) was 5.6% greater than in the SSRI first then TCA strategy as compared to TCA-only and TCA first then SSRI strategies. In the nine months (medium term) horizon the SSRI first then TCA treatment strategy had almost identical effects to that of TCA first then SSRI. Both of these strategies had an incremental success rate of 0.3% and 0.039 incremental QALYs. In the case of TCA-only treatment the incremental success rate was 9%, and the incremental QALYs were 0.042. The side effects of treatment were considered.
Cost results Total intervention costs were not reported. In the first three months horizon, the base case analysis showed that the SSRI first then TCA treatment strategy had an incremental cost of Can$44 compared to the other two treatment strategies. In the first nine months horizon the incremental costs of the SSRI/TCA versus TCA/SSRI treatment strategy was Can$110 and the incremental cost of SSRI-first then TCA versus TCA-only treatment strategy was Can$ -406. The costs estimates took into account direct costs of side effects.
Synthesis of costs and benefits Benefits and costs were combined by calculating incremental cost/QALY gained. In the nine month horizon, the SSRI/TCA treatment strategy had a cost of Can$2,818 per QALY gained compared to the TCA/SSRI; the SSRI-first then TCA treatment strategy is dominant compared to TCA-only.
Authors' conclusions The overall analysis showed that there are no statistically significant differences between the efficacy (as measured in HRSD or CGI scores), completion rates and drop-out rates of the different antidepressants considered (individual SSRIs compared with each other and with TCAs or other antidepressants). These results held regardless of patient age, trial setting or dosing level. Adverse events such as nausea, anorexia, diarrhea, anxiety, agitation, insomnia, and nervousness are statistically more frequent with SSRIs and adverse events such as dry mouth, constipation, blurred vision, and dizziness are statistically more frequent with TCAs. The treatment strategy of SSRI-first then TCA yields the same success as that of TCA-first then SSRI, but is slightly more costly. However, when the SSRI-first then TCA treatment strategy is compared with the TCA-only, the first is both more successful and less costly, thus making it the dominant strategy in this case.
CRD COMMENTARY - Selection of comparators The reason for the choice of comparators is clear. Validity of estimate of measure of effectiveness The clinical evidence was derived by the authors using a systematic search of the literature. However, as acknowledged by the authors, the studies in the literature do not provide a long term analysis of the antidepressants' treatment effects (only between 4 and 12 weeks). Validity of estimate of costs Insufficient details were provided regarding the estimation of costs and quantities. The results may not be generalisable to other settings or countries. Other issues The uncertainties of the model were tested using sensitivity analyses. Implications of the study Further research is needed to investigate the long term effectiveness and cost-effectiveness of treatments for major depression.
Bibliographic details Ausejo M, Glennie J L. A clinical and economic evaluation of selective serotonin reuptake inhibitors in major depression. Ottawa, ON, Canada: Canadian Coordinating Office for Health Technology Assessment. Technology Overview: Pharmaceuticals, 11. 1997 Other publications of related interest This publication is based primarily on two technical reports:
Trindade E & Menon D. Selective Serotonin Reuptake Inhibitors (SSRIs) for major depression. Part I. Evaluation of the clinical literature. Ottawa, Canada: Canadian Coordinating Office for Health Technology Assessment, 1997.
Baladi J F. Selective Serotonin Reuptake Inhibitors (SSRIs) for major depression. Part II. The cost-effectiveness of SSRIs in treatment of depression. Ottawa, Canada: Canadian Coordinating Office for Health Technology Assessment, 1997.
Indexing Status Subject indexing assigned by CRD MeSH Canada; Clinical Trials as Topic; Cost-Benefit Analysis; Depression /drug therapy; Drug Costs; Government Agencies; Humans; Serotonin Uptake Inhibitors /economics /therapeutic use /adverse effects; Technology Assessment, Biomedical AccessionNumber 21998008039 Date bibliographic record published 30/11/1999 Date abstract record published 30/11/1999 |
|
|
|