|Assessment of the cost effectiveness of travoprost versus latanoprost as single agents for treatment of glaucoma in France
|Payet S, Denis P, Berdeaux G, Launois R
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.
The aim was to assess the costs and effects of travoprost and latanoprost as single treatments for glaucoma in France. The authors concluded that travoprost was very cost-effective, producing fewer treatment changes, at a small increase in cost, which was acceptable to payers, over five years. In summary, with a few exceptions, the methods were transparent and clearly reported. The scope of the analysis and the available evidence tend to support the authors' conclusions.
Type of economic evaluation
The aim was to assess the costs and effects of travoprost and latanoprost as single treatments for glaucoma in France.
A course of travoprost was compared with latanoprost given once daily to ambulatory patients with primary open-angle glaucoma or ocular hypertension. Both these prostaglandin analogues were designed to lower the intra-ocular pressure (IOP).
A Markov model was used to synthesise the data from two observational studies (Denis, et al. 2006, 2004a, and 2004b see ‘Other Publications of Related Interest’ below for bibliographic details). These studies were selected to represent the actual clinical practice, to compare with the evidence on the efficacy of the two treatments from head-to-head randomised trials, in France. This analysis covered a five-year period, and the authors stated that the perspective was that of society.
The evidence for the clinical estimates was based on a cross-sectional, retrospective, observational study (Denis, et al. 2006). No major baseline differences were found between the two groups, and adequate controls were made for selection bias and potential confounding factors. The main clinical estimate was the "mean time to treatment change" while secondary endpoints were the "proportion of patients undergoing laser therapy or surgery", and the "proportion of patients still taking prostaglandin analogue monotherapy" after five years.
Monetary benefit and utility valuations:
Measure of benefit:
The measures of benefit used were "years without treatment change" and "avoided treatment changes".
The direct costs were those of the drugs, laser treatments and surgical operations since diagnosis, hospitalisations for glaucoma, and health professional visits. The resources were valued from national sources (the Vidal Compendium for drug costs, diagnosis related groups for hospitalisation, etc). The unit costs were used for out-patient visits and procedures. The costs for failed laser or surgical treatments were included. The monthly costs were estimated for each model health state. Discounting was applied at an annual rate of 3.5% and all prices were in year 2005 Euros (EUR). National inflationary adjustments to prices were made.
Analysis of uncertainty:
The parameter uncertainty was investigated using probabilistic sensitivity analysis. Full details of the distributions assigned and the methods used were reported and cost-effectiveness acceptability curves were plotted. Analyses were also performed on the three ranges of IOPs at the start of treatment, which were 20mmHg or less, 21 to 23mmHg, and 24mmHg or more.
The mean time to treatment change was 44.3 months for travoprost and 37.8 for latanoprost. The discounted travoprost costs were EUR 2,411 compared with EUR 2,360 for latanoprost. The incremental cost-effectiveness ratio (ICER), for travoprost over latanoprost, was EUR 95 per year without treatment change.
The results differed for the three ranges of baseline IOPs. For travoprost the percentage of patients who continued to receive treatment at five years was 55.6% for 20mmHg or less, 53.9% for 21 to 23mmHg, and 50.4% for 24mmHg or more, whereas for latanoprost it was 32.3% for 20mmHg or less, 26.1% for 21 to 23mmHg, and 26.1% for 24mmHg or more. The ICERs for travoprost over latanoprost were EUR 140 for 20mmHg or less, EUR 45, for 21 to 23mmHg, and EUR 123 for 24mmHg or more, per year without treatment change.
The probabilistic sensitivity analyses showed travoprost to be more cost-effective 95% of the time, if payers were willing to pay EUR 50 more for travoprost than for latanoprost to avoid treatment change.
The authors concluded that travoprost was cost-effective for treatment for glaucoma in comparison with latanoprost, especially for patients with IOP values from 21 to 23mmHg, as there was only a small additional cost for avoiding treatment change.
The two treatments were clearly described including their physiological mechanism, historical development for efficacy, and cost-effectiveness. The reader should decide whether these treatments are relevant in their own setting.
The effectiveness data were derived from a retrospective observational study, and their selection and the rationale for this were justified by the authors. In order to assess the validity of these estimates, it would be necessary to consult the original study by Denis, et al. 2006, although, the efficacy for travoprost in France was established. The efficacy results were clearly reported.
The direct medical costs appeared to reflect a health system perspective rather than the societal perspective stated. The sources of the resource use data were clearly stated, but the monthly health state costs and unit costs were not presented separately from their model formulas. Despite a small cost difference between the two groups, which may not have been statistically significant, the sensitivity analyses indicated that the cost-effectiveness ratios remained stable. The authors briefly discussed the differences in resource use in the UK setting (which was used for the treatment change data) as being a potential limitation, but stated that these were still "real-world" estimates and likely to reflect the French patterns of use.
Analysis and results:
The cost and effect analyses were appropriate and comprehensive, and were designed to reflect actual clinical practice. The base-case and sensitivity analysis results were fully reported and illustrated. The impact of structural uncertainty on the base-case findings was discussed with regard to altering the baseline IOP values. The generalisability of the results and the limitations of the economic model were identified and discussed, including the use of non-randomised data and limiting the analysis to five years, which was constrained by the data available.
Despite minor limitations with data transparency, the methods appear to have been appropriate and comprehensive. The conclusions reached by the authors reflect the scope of the analysis.
Supported by a grant from Alcon France SA.
Payet S, Denis P, Berdeaux G, Launois R. Assessment of the cost effectiveness of travoprost versus latanoprost as single agents for treatment of glaucoma in France. Clinical Drug Investigation 2008; 28(3): 183-198
Other publications of related interest
Denis PH, Launois R, Devaux M, et al. Comparison of diurnal IOP control by latanoprost versus travoprost: results of an observational survey. Clin Drug Invest 2006;26:703-14.
Denis P, Lafuma A, Berdeaux G. Medical predictive factors of glaucoma treatment costs. J Glaucoma 2004;13:283-90.
Denis P, Lafuma A, Berdeaux G. Medical outcome of glaucoma therapy from a nationwide representative survey. Clin Drug Investig 2004;24:343-52.
Subject indexing assigned by NLM
Age Factors; Aged; Antihypertensive Agents /administration & Cloprostenol /administration & Cost-Benefit Analysis /methods /statistics & Databases, Factual /statistics & Drug Costs /statistics & Feasibility Studies; Female; France; Glaucoma /drug therapy; Humans; Intraocular Pressure /drug effects; Male; Markov Chains; Proportional Hazards Models; Prostaglandins F, Synthetic /administration & Sex Factors; Survival Analysis; Time Factors; Travoprost; Treatment Outcome; derivatives /economics /therapeutic use; dosage /analogs & dosage /economics /therapeutic use; dosage /economics /therapeutic use; numerical data; numerical data; numerical data
Date bibliographic record published
Date abstract record published