|
Analysis of life-long strategies to prevent Pneumocystis carinii pneumonia in patients with variable HIV progression rates |
Wynia M K, Ioannidis J P, Lau J |
|
|
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 Four different strategies for the prevention of Pneumocystis carinii pneumonia (PCP) in AIDS patients were examined. Each strategy comprises at least five different drug regimens, to reflect the fact that patients often switch regimens due to failure or toxicity. The four strategies are referred to by their starting regimens as follows:
high-dose (also referred to as standard dose) trimethoprim-sulfamethoxazole (TS),
low-dose TS,
high-dose aerosolised pentamidine (AP),
monthly AP with a switch to low-dose TS at a specified time.
Dapsone and other alternative drugs were also part of these strategies.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised AIDS patients with no history of PCP, whose disease was progressing at a range of rates.
Setting The setting was secondary care. The economic cost study was carried out in Boston, USA.
Dates to which data relate The effectiveness analysis used studies published between 1989 and 1997. The resource use and cost data relate to 1995 and 1996.
Source of effectiveness data The effectiveness data were derived using a meta-analysis, a review of completed studies, and some assumptions.
Modelling A set of Markov decision models were used to estimate the lifetime efficacy, toxicity and costs associated with the four strategies at three different rates of disease progression (baseline, rapid and slow). The model terminated when 99% of the cohort of patients died. Twenty-nine health states were included in the model. The authors state that conducting a clinical trial on this issue is difficult because patients often switch medications and change doses.
Outcomes assessed in the review The outcomes examined included the probabilities of PCP, death, toxicity, successful desensitisation and dose reduction after a toxicity event, and the prophylactic efficacy of the various medications.
Study designs and other criteria for inclusion in the review The efficacy and toxicity data were taken from one meta-analysis (see Other Publications of Related Interest). No inclusion or exclusion criteria were stated for the other studies.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included Eighteen studies were included in the review.
Methods of combining primary studies Investigation of differences between primary studies Results of the review The results of the review include the following baseline parameter estimates:
a 2-year risk of PCP without prophylaxis of 40%;
an overall median survival of 3 years from the initiation of prophylaxis;
a 2-month probability of toxicity of 16% on the first cycle of standard-dose TS and 1.7% on subsequent cycles;
a 2-month probability of toxicity of 10% on the first cycle of low-dose TS and 1% on subsequent cycles;
a 2-month probability of toxicity of 0.5% on AP;
a probability of successful TS desensitisation of 50%;
a probability of successful dapsone dose reduction of 50%;
prophylactic efficacies of 97% for standard-dose TS, 95% for low-dose TS, 92% for standard-dose dapsone, 82% for low-dose dapsone, 75% for high-dose AP, 64% for standard-dose AP, and 64% for "alternative therapies".
Methods used to derive estimates of effectiveness The authors made some assumptions in order to construct their model.
Estimates of effectiveness and key assumptions The key assumption made by the authors was that TS was always used before dapsone, which was used before AP. TS toxicity was treated by desensitisation, while dapsone toxicity was treated by dose reduction. The authors provided details of their other assumptions in the article.
Measure of benefits used in the economic analysis No summary benefit measure was used. Therefore, this was a cost-consequences analysis.
Direct costs This analysis included the costs to the hospital, which were based on those incurred by the New England Medical Centre (Boston) for the treatment of AIDS patients. The inpatient care costs were derived from a database of patients admitted with PCP from October 1995 to June 1996 (n=59). These costs included the variable costs only; overheads were excluded. The toxicity costs were estimated to include two outpatient visits and the laboratory costs. The medication costs were derived from the pharmacy of the aforementioned hospital. There was no report of any discounting even though it is relevant. The quantities and the costs were not reported separately.
Indirect Costs The indirect costs were not included in this analysis.
Sensitivity analysis Sensitivity analyses were carried out on all the effectiveness and cost parameters of the model, in order to investigate variability in the data. The ranges were obtained from the literature. Some of the model assumptions were also varied. The article reported a limited number of results from the sensitivity analyses.
Estimated benefits used in the economic analysis For patients with typical and rapid disease progression, the standard- and low-dose TS strategies resulted in less than half the number of PCP cases per 100 patients, compared with the high AP and AP switch strategies. With slow disease progression, low-dose TS became the most effective strategy.
Side effects were also considered, specifically drug toxicity. The AP strategies resulted in fewer toxicity episodes than the low-dose TS strategy. The standard-TS strategy led to the most toxicity episodes.
The incremental differences in life expectancy between the strategies were small.
Cost results In the slow disease progression case, the average lifetime costs were estimated to be $4,323 for standard TS, $2,942 for low TS, $20,197 for high AP, and $5,041 for AP switch. Thus, the low-TS strategy was the least expensive strategy. The difference between it and the next cheapest strategy (standard TS) was greatest with slow-progressing disease ($1,381 lifetime average cost difference).
Synthesis of costs and benefits Authors' conclusions Trimethoprim-sulfamethoxazole (TS)-based strategies are superior to aerosolised pentamidine (AP)-based strategies in terms of their effectiveness and cost, though they are more toxic. For patients with slower disease progression (an expected survival of more than 3.8 years), a low-dose TS strategy is dominant.
CRD COMMENTARY - Selection of comparators The authors state that they chose the four strategies based on their clinical relevance given current information. You will have to compare these with current practice in your own settings.
Validity of estimate of measure of effectiveness The authors did not state that a systematic review was performed to derive the model parameters. They did, however, subject the parameter estimates to a sensitivity analysis using ranges identified in the literature. The validity of the results may be limited because the authors only compared four treatment regimens. In addition, the analysis may not be generalisable to the wider population because some patients may experience adverse reactions to TS.
Validity of estimate of measure of benefit No summary benefit measure was used as the authors suggest there is variability in the patients' preferences for avoiding PCP, compared with avoiding toxicity episodes. This study was, therefore, a cost-consequences analysis.
Validity of estimate of costs All the relevant, acute-care cost elements were included. The analysis does not appear to have included the community care costs. The unit costs were derived from an American hospital and were not adjusted for inflation. The future costs should have been discounted to reflect the fact that up-front costs (for example, the drugs) are worth more than future costs (for example, future episodes of PCP). A statistical analysis of the costs was not performed.
Other issues The authors compared both their model design and results with those of other studies. They pointed out that the published models have not accounted for the switching of regimens, and the need for treatment following toxicity episodes that occur in these patients. Despite using a more complicated model, the authors found their mortality and cost outcomes to be qualitatively similar to those in other studies.
The authors addressed the issue of the generalisability of trial efficacy findings to real-life use where non-compliance may be a problem. They adjusted some of their model inputs to account for non-compliance.
Overall, the model was well described and should be reproducible in other settings. The authors present clear recommendations that seem justified given the scope of their analysis. <IMPLICATIONS OF THE STUDY>> The authors suggest that, as patients with AIDS live longer with the help of highly active antiretroviral therapy, a low-dose TS strategy will be the preferred method of preventing PCP. More research is required on the issue of discontinuing PCP prophylaxis in some patients.
Source of funding Supported in part by the Agency for Health Care Policy and Research, grant numbers T32 HS00060 and RO1 HS07782; and the National Institutes of Health, grant number T32 A107389.
Bibliographic details Wynia M K, Ioannidis J P, Lau J. Analysis of life-long strategies to prevent Pneumocystis carinii pneumonia in patients with variable HIV progression rates. AIDS 1998; 12(11): 1317-1325 Other publications of related interest Ionnidis JP, Cappelleri JC, Skolnik PR, Lau J, Sacks HS. A meta-analysis of the relative efficacy and toxicity of Pneumocystis carinii prophylactic regimens. Archives of Internal Med 1996;156:177-88.
Indexing Status Subject indexing assigned by NLM MeSH AIDS-Related Opportunistic Infections /economics /physiopathology /prevention & Antifungal Agents /adverse effects /economics /therapeutic use; Disease Progression; Drug Costs; Health Care Costs; Humans; Life Expectancy; Markov Chains; Pneumonia, Pneumocystis /economics /physiopathology /prevention & Time Factors; Trimethoprim, Sulfamethoxazole Drug Combination /adverse effects /economics /therapeutic use; control; control AccessionNumber 21999006194 Date bibliographic record published 31/12/2002 Date abstract record published 31/12/2002 |
|
|
|