|
Pharmacoeconomic analysis of venlafaxine in the treatment of major depressive disorder |
Einarson T R, Addis A, Iskedjian M |
|
|
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 Venlafaxine in the treatment of acute depressive disorders.
Economic study type Cost-effectiveness analysis.
Study population Patient characteristics are published elsewhere, with a meta-analysis of effectiveness data.
Setting Inpatient and outpatient settings were included. The economic analysis was carried out in Ontario, Canada.
Dates to which data relate Effectiveness data were collected over the period 1984-1996.
Source of effectiveness data Effectiveness data were based on a review and meta-analysis of the literature (unpublished). Correspondence from the author indicates that additional data were obtained from an expert panel and that the meta-analysis has since been published (see related publications).
Link between effectiveness and cost data Costing was undertaken retrospectively.
Modelling The pharmacoeconomic model used was a predictive model (i.e. based on decision analysis) with a 6-month time horizon.
Outcomes assessed in the review The outcomes assessed in the review were clinical rates of success (a 50% reduction in the depression score on either scale in patients who completed therapy), and the number of dropouts due to adverse drug reactions and due to efficacy failures.
Study designs and other criteria for inclusion in the review Only double-blind randomised controlled trials between any two of the drugs of interest were included in the meta-analysis. Patients were adult (aged over 18 years) inpatients or outpatients (but data were separated by patient type), who were either antidepressant naive or had undergone a 7 to 14 day washout period. The acceptable diagnosis for this analysis was major depressive disorder. Patients must have scored more than 15 on the Hamilton Rating Scale for Depression or more than 18 on the Montgomery and Asberg Depression Rating Scale. The drugs included oral venlafaxine, amitriptyline, imipramine, desipramine, nortriptyline, fluoxetine, sertraline, paroxetine, or fluvoxamine given in therapeutic doses for at least 4 weeks. Although patients could not be taking other antidepressants, lithium, or thyroid hormone, other drugs such as tranquillisers and hypnotics were acceptable. Patients could have no concomitant diseases which would interfere with diagnosis or treatment (e.g. metabolic, endocrine, psychiatric).
Sources searched to identify primary studies Medline, Embase, and International Pharmaceutical Abstracts were searched, as were references from retrieved articles and reviews.
Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Although not explicitly stated in the original paper, correspondence from the author indicates that data were extracted by two reviewers and adjudicated by a third reviewer.
Number of primary studies included 37 randomised controlled trials involving 2,953 patients (2,380 outpatients and 573 inpatients) were included in the review.
Methods of combining primary studies Meta-analysis. The methods used are not described in this report.
Investigation of differences between primary studies Results of the review The results of the meta-analysis of studies using various drug classes to treat acute major depressive disorder show that the range of overall success was 0.580 to 0.702 (including all drug categories and both outpatient and inpatient settings). The success rates for TCA's (inpatient/outpatient was 0.580/0.582, for SSRI's 0.586/0.616, and for SNRI's 0.623/0.702. The 95% confidence interval for each of these success rates was significant for a given drug/setting category. Statistical testing between drug categories was not presented. Correspondence from the author states that the results were not statistically different between classes of drugs due to their large standard errors.
The dropout rates (inpatient/outpatient combined) due to adverse effects were: SNRIs 0.159, SSRIs 0.179, TCAs 0.266. The dropout rates due to lack of efficacy were: SNRIs 0.056, SSRIs 0.094, and 0.086 for TCAs. The 95% confidence interval for each of these rates was significant for a given drug category. Statistical testing between drug categories was not presented.
Measure of benefits used in the economic analysis The measures of benefit were clinical rates of success (a 50% reduction in depression score on either scale in patients who completed therapy), and number of dropouts due to adverse drug reactions and due to efficacy failures. A decision tree was used to estimate the cost per success, and the cost per symptom-free day.
Direct costs Costs were not discounted, due to the short time-horizon (6 months). The cost data were presented separately, but not all quantities were presented (i.e. number of visits required). The costs included in the model were physician (GP) visit, physician (GP) follow-up, psychiatrist visit, psychiatrist follow-up, hospital stay (per day), physician service, electroconvulsive treatment (per session), cardiologist consultant, laboratory (initial regimen), and drug costs (1 month) for SNRIs, SSRIs, and TCAs. The cost perspective was that of the health service. The estimation of quantities was derived from an expert panel and review of the literature (dates not specified). Drug costs were determined from the Ontario Drug Benefit Formulary (1994), the PPS (July 1996), or the manufacturer's price list (date not stated). Costs for physician and laboratory services were taken from Ontario's Schedule of Benefits (1992), while hospital charges were taken from the Ontario hospital statistics (1991/1992). For venlafaxine, the authors used the manufacturer's price and for the other drug classes, they used the average cost for all drugs, based on published costs. The lowest cost of available products (i.e. the generic or lowest cost branded drug) was used and equal weights, rather than market share, were assigned to all drugs within each class. Incremental costs were reported.
Statistical analysis of costs The authors treated the cost data in the sensitivity analysis in a stochastic way. They assumed a standard deviation of 15%, producing a pseudo-confidence interval of +/- 30%. The authors then calculated a Z score to statistically test the difference between expected costs, using a pooled standard deviation weighted by the number of studies in the meta-analysis.
Sensitivity analysis One-way simple and probabilistic (Monte-Carlo simulation) sensitivity analyses were performed on success rates, drop out rates, the assumptions made by the expert panel, sample size of the primary studies and costs. The sensitivity analyses were carried out to test the variability in the data.
Estimated benefits used in the economic analysis As a result of the model used, the expected success rate for SNRIs was (outpatient/inpatient) 0.7891/0/7319, for SSRIs was0.7647/0.6649, and for TCAs was 0.6841/0.6608. The expected symptom-free days (in 6 months, outpatient/inpatient) for SNRIs was 103.86/99.08, for SSRIs was 98.22/88.38 and for TCAs was 87.42/88.69.
Cost results The medical costs, in Canadian dollars, for the pharmacoeconomic model were as follows:
physician (GP) visit (Can$51.40),
physician (GP) follow-up (Can$38.80),
psychiatrist visit (Can$114.55),
psychiatrist follow-up (Can$69.30),
hospital stay (per day) (Can$375.00),
physician service (Can$51.40),
electroconvulsive treatment (per session) (Can$29.10),
cardiologist consultant (Can$105.40),
and laboratory (initial regimen) (Can$127.29),
and one month drug costs were Can$46.80, Can$38.06 and Can$12.74 for serotonin-norepinephrine reuptake inhibitors, selective serotonin reuptake inhibitors, and tricyclic antidepressants, respectively.
Synthesis of costs and benefits The expected costs per success were Can$6,044 and Can$17,234 for venlafaxine, Can$6,634 and Can$20,874 for SSRIs, and Can$9,035 and Can$20,459 for TCAs in outpatients and inpatients, respectively. The respective expected costs per symptom-free day were Can$45.92 and Can$127.31 for venlafaxine, Can$51.64 and Can$157.04 for SSRIs, and Can$70.71 and Can$152.43 for TCAs. In the incremental cost per success and cost per symptom-free day, venlafaxine was dominant over the others.
Authors' conclusions Analysis concluded that venlafaxine was a cost-effective drug for the treatment of acute major depressive disorder (MDD) in adult outpatients as well as inpatients. It confirmed previous results that venlafaxine was cost effective in treating MDD when compared with SSRIs or TCAs. The dominance of venlafaxine in all of the pharmacoeconomic analyses provides evidence that its use is reasonable and that its adoption is warranted.
CRD Commentary The meta-analysis used to derive the effectiveness data was not presented in the original paper which limits the reproducibility of these results. The results of the panel of clinicians was also not presented. These assumptions were used in he modelling of resource use and this lack of reporting limits the generalisability of the study
Implications of the study More research is required to establish the dominance of venlafaxine over other antidepressant drug classes (SSRIs and TCAs).
Bibliographic details Einarson T R, Addis A, Iskedjian M. Pharmacoeconomic analysis of venlafaxine in the treatment of major depressive disorder. PharmacoEconomics 1997; 12(2): 286-296 Other publications of related interest Einarson T R, Addis A, Mittman N, et al. Meta-analysis of venlafaxine, SSRI's and TCA's in the treatment of major disorder. Can J Clin Pharmacol 1998;5(4):205-216.
Indexing Status Subject indexing assigned by NLM MeSH Antidepressive Agents, Second-Generation /economics /therapeutic use; Cyclohexanols /economics /therapeutic use; Depressive Disorder /drug therapy /economics; Humans; Models, Economic; Serotonin Uptake Inhibitors /economics /therapeutic use; Venlafaxine Hydrochloride AccessionNumber 21997008290 Date bibliographic record published 31/07/1999 Date abstract record published 31/07/1999 |
|
|
|