|
In addition, we included in the model that a certain percent of all patients would suffer legal blindness despite therapy. The specific rates of legal blindness were taken from the literature for Denmark and the UK and general rates were used from the United States and Europe for Norway and Sweden. We assumed that the rates were associated for timolol maleate since the literature was published before the release of the prostaglandin analogs[14,15]. We also assumed that all legal blindness occurs only in patients in the unstable branches of the Markov model. Adjustments to the blindness rates were made for each country based on the percent of unstable patients in each prostaglandin analog group compared to timolol maleate.Table 1Unit costs of procedures by countries（略）
Markov Model economic Aspects The Markov model was completed with economic information from each country (Table 1). In addition, we also performed personal interviews with a practicing ophthalmologist within each country to obtain unit cost information for visits, procedures and therapy. Only direct medical costs were considered (i.e., cost of visits, procedures and therapy). For timolol maleate the cost of the generic products was used. Cost was measured in local monetary units and by converting currency values from Scandinavia to the United States dollar at the exchange rate for December 31, 2005 at Oanda.com (1 Swedish Krona=0.12565 US Dollar; 1 Norwegian Kroner =0.14750 US Dollar; 1 Danish Krone = 0.15846 US Dollar; Oanda.com). All analyses were from a payer perspective and cost results were discounted at 3% per annum.
Markov Model Sensitivity Analysis For the sensitivity testing the following analyses were performed using a oneway analyses for the following costs: a single unit cost of latanoprost, the cost of a routine visit the cost of blindness. In each case the parameter was changed by ±10% and 15%. The analyses were performed separately for each country. An adjustment of up to 10% was used an estimate that might be practically seen clinically because there are few guidelines in the literature regarding how much real variation for these cost exist in Scandinavia and efficacy among the three prostaglandins. Fifteen percent then was used to further test the robustness of the model since little change was observed at 10% levels. A noncost sensitivity analysis was chosen also using ±5% and 10% absolute change in the control rate of latanoprost.
RESULTS
Key assumptions of the study and how the Markov model was populated based on the previous literature. Latanoprost dominated (i.e.,was less expensive and more effective than) bimatoprost and travoprost in both Norway and Sweden.In contrast, in Denmark bimatoprost dominated travoprost and was slightly less expensive than latanoprost (25495 versus 25097 DKK).
However, in Denmark latanoprost was more effective than bimatoprost and travoprost. The incremental costeffectiveness ratio (ICER) for latanoprost to gain 1 QALY was 63011 Danish Krone ($9985) which was under the National Institute for Clinical Excellence (NICE) classification of 50000(US dollars) for a cost effective treatment(Table 2)[16]. Overall, initiating latanoprost treatment was approximately 4% less expensive in Sweden and Norway and the costs of all three medicines were within 1.5% of each other in Denmark. In all three countries the average cost over the life of the model was for 4565 for latanoprost, 4660 for travoprost and 4620 for bimatoprost (US dollars).
Effectiveness (years till progression) was within a narrow range (4.01554.0703) for all products in each country. Latanoprost effectiveness was greater than the other prostaglandins in all three countries by a range of 0.00380.0214 QALY. With sensitivity testing at up to ±15% for the cost of latanoprost, an uncomplicated visit or blindness, latanoprost still dominated travoprost and bimatoprost in Sweden and Norway. In contrast, in Denmark bimatoprost still dominated travoprost, was less effective than latanoprost, and remained less expensive than latanoprost (ICER,63010 DKK) for blindness as well as visit costs (range: 5877567246 DKK). With a more expensive latanoprost bottle cost the ICER ranged for ±10%, 80855 DKK, and ±15% 144150 DKK. However, when latanoprost unit cost was reduced by 10%15%, it dominated the other two products.
When the absolute control rate (24%30%) was adjusted for latanoprost by 5% and 10% (maximum relative increase or decrease in control up to 42%) latanoprost still dominated both prostaglandins when adjusted for greater control. However, bimatoprost dominated both prostaglandins with either level of less control for latanoprost.Table 2Cost effectiveness results from the Markov model (略)
DISCUSSION
Several previous studies have evaluated the Markov model with a time horizon of five years for glaucoma medications[4]. Le Pen and associates evaluated travoprost versus latanoprost and timolol maleate in Austria, France, Germany, the Netherlands and the United Kingdom. The health states were stable and progressed glaucoma. They assumed a greater efficacy with travoprost over latanoprost and timolol maleate. They concluded that travoprost economically dominated the other medicines in several countries.
Nordmann and coworkers evaluated first to fourth line treatment in glaucoma with a variety of medicines[4]. They concluded that a more powerful first or second line treatment would contribute to preservation of longterm vision. Christensen and associates evaluated bimatoprost as an alternative to filtration surgery for glaucoma patients on maximum tolerable medical therapy in Italy over four years. They concluded that bimatoprost was less expensive than filtration surgery and had important implications for waiting list planning.
The purpose of this study was to evaluate the costeffectiveness of latanoprost, travoprost and bimatoprost monotherapy in open angle glaucoma in Scandinavia using a Markov model. This study showed that latanoprost dominated both travoprost and bimatoprost in Norway and Sweden for effectiveness (years to glaucomatous progression) and cost and was more effective in Denmark. In contrast, in Denmark travoprost was dominated by bimatoprost which was slightly less expensive than latanoprost.
This cost advantage for latanoprost in Norway and Sweden, and its close proximity to the other prostaglandins in Denmark, existed despite a slightly less expensive per unit cost of travoprost (4%10%) and bimatoprost (0%10%). This indicated that while travoprost and bimatoprost may be less expensive shorttem, other factors associated with their use might increase costs and cause them to become more expensive than latanoprost longterm. The reason for the greater cost of travoprost and bimatoprost in our model most likely resulted from the greater persistency rates of latanoprost, derived from the literature, which allowed better maintenance of the original prescribed medication[911]. The lower persistency with travoprost and bimatoprost might have increased costs in several ways: first, the additional visits and procedures resulting from the need to change medicines due to more frequent side effects (i.e., conjunctival hyperemia); second, the need for an additional medicine to control intraocular pressure because timolol maleate (the medicine to which a nonpersistent patient was changed) less frequently controls intraocular pressure than a prostaglandin; and last, because of a slightly greater incidence of blindness associated again with the lower efficacy of timolol maleate[911].
上一页 [1] [2] [3] 下一页 |
