|
Estimating the cost effectiveness of atovaquone versus intravenous pentamidine in the treatment of mild-to-moderate Pneumocystis carinii pneumonia |
Zarkin G A, Bala M V, Wood L L, Bennett C L, Simpson K, Dohn M N |
|
|
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 Atovaquone, administered on an outpatient basis for treating mild to moderate Pneumocystis carinii pneumonia (PCP).
Economic study type Cost-effectiveness analysis.
Study population Adult patients with AIDS and histologically confirmed PCP who were intolerant of cotrimoxazole.
Setting Hospital and primary care. The economic study was carried out in North Carolina and Ohio, USA.
Dates to which data relate The data for the effectiveness analysis was obtained from a previous study published in 1994. The resource use and costs were estimated, in part, from data for Maryland from 1992, and a cost-charge ratio from a study published in 1992. The cost data were adjusted to 1995 prices.
Source of effectiveness data Effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was carried out retrospectively and not on the same patient sample as that used in the effectiveness analysis.
Study sample No power calculations were reported. 56 patients were included in the intervention group and 53 in the comparator group.
Study design The study was a randomized controlled trial. The duration of follow up was 8 weeks after the termination of initial acute therapy. The loss to follow up was not clearly reported (at most 11% in the intervention group and 8% in the comparator group).
Analysis of effectiveness The analysis was based on intention to treat. The primary health outcomes used were success rate, toxicity-related failure, lack of response and the rate of unevaluable patients. Mortality rates were reported (at 4 and 8 weeks after initial therapy). 'Success' was defined as sustained clinical improvement 4 weeks after termination of initial therapy without additional PCP treatment. 'Toxicity-related failure' was defined as "discontinuation of treatment as the result of the patient experiencing an adverse event and switching to alternative PCP medication for treatment". Lack of response (LOR) was defined in terms of clinical indicators with a consequent switch to the other treatment arm to that to which the patient was initially allocated, and completion of the alternative treatment (21 days of new, alternative, therapy after 10 days from the initiation of the study medication). The 'unevaluable patient' definition was used both for patients lost in follow up and non-compliant cases. There was no report of analysis of differences in demographic variables or prognostic features between groups.
Effectiveness results The success rate for the intervention group was 57%, whereas for the comparator it was 40%. The p value was 0.085. The failure rate was broken down into LOR rate, treatment-limiting toxicity (adverse effects), and unevaluable patients. The LOR rate for the intervention was 29% and for the comparator 17% (p = 0.18). The treatment-limiting toxicity rate was 4% for the intervention and 36% for the comparator (P<0.001). The unevaluable patient rates were 11% and 8%, respectively (p = 0.74). The mortality rates (at 4 weeks of follow up) were 13% and 8%, for intervention and comparator, respectively (p=0.53). The mortality rates at the end of follow up were 16% and 17%, respectively (p=1.00).
Clinical conclusions Atovaquone and pentamidine had similar rates of successful treatment and similar mortality rates. Atovaquone showed significantly fewer "treatment-limiting adverse effects".
Modelling A decision tree was used to estimate costs and benefits. A Monte-Carlo simulation was used to analyse the uncertainty behind the cost estimates.
Measure of benefits used in the economic analysis The benefit measure wassuccess rate, with success being defined as sustained clinical improvement 4 weeks after termination of initial therapy without additional PCP treatment.
Direct costs Costs were not required to be discounted due to the short follow-up period of the study. Quantities were analysed separately and cost items were reported separately. The costs measured were operating costs and costs of complications (monitoring for adverse effects). The cost boundary adopted was the hospital. The estimation of quantities was based on unpublished data and on experts' opinion. The (unit) cost data were based on actual data; the estimate of the per-day hospitalisation charge was based on "all hospital discharges for patients with PCP in Maryland, US, in 1992", in conjunction with a cost charge ratio reported for AIDS patients in a separate study published in 1992; the estimates of average charge for outpatient clinic visits and outpatient IV infusion were based on data from two "large US AIDS clinics situated in Washington, DC, and North Carolina" (cost-based estimates). The costs were reflated to 1995 prices.
Statistical analysis of costs The Monte Carlo method was used to estimate 95% confidence intervals (CI) for the difference in estimated costs between treatment strategies.
Sensitivity analysis The number of days of stay in hospital and the number of days of PCP therapy (in both cases for both therapies) were the parameters used in the sensitivity analysis. A one-way simple sensitivity analysis was performed. A Monte Carlo simulation was performed to treat all the decision model parameters as random variables.
Estimated benefits used in the economic analysis The success rate for the intervention group was 57%, whereas for the comparator it was 40% (p= 0.085).
Cost results The total expected cost for the intervention over the 21-day treatment period was $3,990 per patient. The corresponding figure for the comparator was $6,545. The difference in costs (intervention-comparator) was reported as having a 95% CI of -$5,632 to -$155.
Synthesis of costs and benefits Not applicable since the intervention was shown to be the dominant strategy. The sensitivity analysis did not alter the results with the exception of the assumed duration of hospitalisation in the intervention and comparator strategies, which varied from 0 to 3 days for the intervention and from 3 to 0 days for the comparator. In neither scenario was the difference in costs statistically significant at the 5% level (calculated CI using Monte Carlo technique).
Authors' conclusions The authors concluded that atovaquone was less expensive than pentamidine over a range of assumptions about days of hospitalisation and treatment duration. Also, "when (the authors) treated all the decision model parameters as random variables in the Monte Carlo simulation, atovaquone dominated pentamidine in terms of cost approximately 98% of the time. The clinical results of Dohn et al., in conjunction with the results of (this) decision tree model, suggest that the success rate of atovaquone is similar to that of pentamidine, and it has a lower expected cost."
CRD COMMENTARY - Selection of comparators The reason for the choice of comparator (pentamidine) was not clear. You, as a database user, should consider whether this is a widely used health technology in your own setting. Validity of estimate of measure of benefit The estimate of the measure of benefit used in the economic analysis is likely to be internally valid. Validity of estimate of costs The resource quantities were reported separately from the prices and adequate details of quantity/cost estimation were given. The authors themselves pointed out a weakness in the study, namely that the resource use data came from "clinical practice algorithms that were validated by discharge data, published reports and discussion with clinicians" rather than from the clinical trial itself. The only important costs not included were those associated with the 'unevaluable patient' subgroup. Other issues The authors' conclusions were justified, given the uncertainties in the data. The generalisability to other settings was addressed and appropriate comparisons were made with other studies. The results were not presented selectively. Source of funding Supported by the Burroughs Wellcome Co.
Bibliographic details Zarkin G A, Bala M V, Wood L L, Bennett C L, Simpson K, Dohn M N. Estimating the cost effectiveness of atovaquone versus intravenous pentamidine in the treatment of mild-to-moderate Pneumocystis carinii pneumonia. PharmacoEconomics 1996; 9(6): 525-534 Indexing Status Subject indexing assigned by NLM MeSH Antifungal Agents /economics /therapeutic use; Atovaquone; Costs and Cost Analysis; Decision Trees; Humans; Injections, Intravenous; Monte Carlo Method; Naphthoquinones /economics /therapeutic use; Pentamidine /economics /therapeutic use; Pneumonia, Pneumocystis /drug therapy /economics AccessionNumber 21996008261 Date bibliographic record published 31/07/1999 Date abstract record published 31/07/1999 |
|
|
|