|
Cost-effectiveness analysis of inhaled zanamivir in the treatment of influenza A and B in high-risk patients |
Griffin A D, Perry A S, Fleming D 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 The health technology was the use of zanamivir 10mg twice daily for 5 days in the treatment of influenza in high-risk patients.
Economic study type Cost-effectiveness analysis; cost-utility analysis.
Study population The study population was a cohort of high-risk patients drawn from a series of zanamivir clinical trials conducted between 1995 and 1999. High-risk patients were defined as those who had chronic respiratory disease (including asthma and chronic obstructive pulmonary disease (COPD)); significant cardiovascular disease (excluding patients with hypertension as the sole diagnosis); were immuno-compromised; or were aged 65 years or older with or without underlying medical conditions. Additionally, in two studies patients with diabetes mellitus were also included. This definition was reported to be similar to that used to identify high-risk individuals appropriate for vaccination in the UK.
Setting The study setting was primary care. The economic analysis was carried out in the UK.
Dates to which data relate Effectiveness and resource use data were extraced from patient record forms held in the zanamivir clinical trial database for trials conducted between 1995 and 1999. Utility values were gathered during the 1999/2000 season. The price year was 1999.
Source of effectiveness data The evidence for the effectiveness outcomes was based on an integrated analysis of the zanamivir clinical trial database.
Modelling Bootstrapping (using a combination of Excel and SPSS) was used to generate confidence intervals, as well as cost-effectiveness acceptability curves.
Outcomes assessed in the review The outcome measures were time to return to normal activities as defined by the patient; time to alleviation of clinically significant symptoms; incidence of complications requiring use of antibacterials, and adverse events. Alleviation of symptoms was considered to have occurred when patients reported absence of fever (temperature less than 37.80 degree Centigrade and no feverishness), and recorded a symptom score of none or mild (or absent/minimal in two studies) for headache, cough, myalgia and sore throat. These conditions had to be maintained for a further 24 hours.
Study designs and other criteria for inclusion in the review The studies included in the review were those incorporating high risk patients, completed before or during the 1998-1999 winter season, which were performed to investigate the efficacy and safety of inhaled zanamivir (10 mg twice daily 5 days) for the treatment of confirmed influenza. All studies were randomised, double-blind, and placebo-controlled with 21- to 28-day follow-up. In total, 2,751 patients were recruited and, of these, 321 (12%) were considered high risk and 154 were randomised to zanamivir.
Sources searched to identify primary studies It was reported that the studies included in the review were from the main Glaxo Wellcome development programme completed before or during the 1998-1999 winter season; no more information was provided.
Criteria used to ensure the validity of primary studies All the studies included were double-blind. The analysis was based on intention to treat.
Methods used to judge relevance and validity, and for extracting data Number of primary studies included A total of 6 large multicentre, randomised, placebo-controlled clinical studies were included in the review.
Methods of combining primary studies The method used for the combination of the primary data was retrospective pooled analysis of data from high-risk patients. Median times to alleviation were calculated from Kaplan-Meier estimates of success rates; interquartile ranges were calculated in the same way. Confidence intervals (CIs) were calculated using the percentile bootstrap method.
Investigation of differences between primary studies The studies were different in terms of length of follow-up and sizes of the study groups. No homogeneity test was reported.
Results of the review The results of the review were as follows:
Zanamivir-treated high-risk patients had a treatment benefit of 2.5 days, (95% CI: 0.68 - 4.27) of normal activities gained compared with those given placebo, (p=0.015).
Patients treated with zanamivir returned to normal activities 3.0 days earlier (p=0.022) and had an 11% reduction, (p=0.039) in the median total symptom score over 1 to 5 days relative to those taking placebo.
Incidence of complications requiring antibiotic use was reduced by 43%, (95% CI: 1% - 67%; p=0.045) in the zanamivir treatment group compared with the placebo group for patients with confirmed influenza. A similar reduction of 37%, (95% CI: 2% - 59%; p=0.042) was observed for the intent to treat population.
Adverse events were reported to be of a similar nature and frequency between the two groups.
Measure of benefits used in the economic analysis The benefit measures were the number of days of normal activities, symptom-free days, complication averted, and quality-adjusted life-years (QALYs). Utility values were not gathered directly during the clinical trial programme. Utility estimates were collected from two sources:
firstly, typical UK general population patients (aged 18 years or over; n=21) who were asked to retrospectively complete one EuroQol instrument (EQ-5D) health questionnaire describing the entire period of their influenza illness and one EQ-5D for the most severe day within that period. The value describing the entire influenza period being used in the analysis; and
secondly, a UK General Practitioner panel (n=8) who used the EQ-5D questionnaire to rate the health-state for hypothetical influenza high-risk patients (these values were not used in the analysis).
Direct costs Costs were not discounted due to the short time frame (one-year) of the cost analysis. Resource use profile per patient was reported separately from the costs. Costs were disaggregated. The cost analysis covered the costs of healthcare contacts, hospitalisations, antibacterial use and over-the-counter (OTC) medications. The perspective adopted in the cost analysis appears to have been that of the health care system. Resource use data (unscheduled healthcare contacts) were collected as part of the case record form (CRF); based on the intention to treat principle (for both censored and uncensored patients). Data on hospitalisation were not collected routinely as part of the CRF; they were therefore gathered from the serious adverse event (SAE) forms reported by investigators. The standard UK unit costs were applied to the unscheduled healthcare contacts and antibacterial use in the trials. The price year was 1999.
Statistical analysis of costs 95% confidence intervals for incremental costs were generated through a bootstrap method.
Indirect Costs Indirect costs were not considered.
Sensitivity analysis Two cases of one-way sensitivity analyses were performed by excluding one hospitalised patient (one outlier) and by excluding SAE hospitalisation. Cost-effectiveness acceptability curves were generated to allow decision-makers to see the probability of zanamivir being cost-effective in relation to varying amounts they might be prepared to pay to secure the benefits. The curves were constructed by varying the maximum cost-effectiveness threshold and calculating the proportion of bootstrap resamples that would fall under each threshold. The cost-effectiveness acceptability curve was also generated through the parametric net benefit approach in order to corroborate the non-parametric bootstrap method.
Estimated benefits used in the economic analysis The mean benefit due to the use of the study drug was estimated to be: 2.5 days, (95% CI: 0.68 - 4.27) of normal activities gained; 2.0, (95% CI: 0.56 - 3.51) symptom-free days; and a 9% reduction in complications, (95% CI: 0 - 18%). Influenza was estimated to reduce utility by 0.883, (95% CI: 0.697 - 1.069) per day, demonstrating the debilitating effect of the disease. QALYs were reproted in the synthesis of costs and benefits, as shown below.
Cost results The incremental cost per patient treated with zanamivir was 44.94, (95% CI: -8 - 98) (total cost per patient was 27.21 for placebo and 72.15 for zanamivir) including all hospitalisations.
Excluding the hospitalisation costs for the one patient withdrawn from the study reduced the incremental cost to 21.67.
Excluding all inpatient hospitalisation costs, treatment with zanamivir increased the cost of treatment per patient by 23.52.
Synthesis of costs and benefits The incremental cost-effectiveness and cost-utility measures were cost per day of normal activities; cost per symptom free day; cost per complication averted; and cost per quality-adjusted life-year (QALY).
The study used a bootstrapping technique to generate 95% confidence intervals (CIs) around the resulting cost-effectiveness point estimates.
Excluding the effect of rare hospitalisation costs, the cost of gaining a day of normal activities was 9.50, (95% CI: 5 - 39); cost per symptom-free day was 11.56, (95% CI: 6 - 43); and cost per complication averted was 262, (95% CI: 90 - 1574).
Extrapolating a day of normal activities to a standard utility measure resulted in a cost per QALY of 3,900 excluding inpatient costs (7,490 including inpatient costs).
Cost-effectiveness acceptability curves demonstrated 90% certainty that zanamivir would be cost-effective at 8,000 per QALY.
Authors' conclusions Significant health benefits can be obtained with zanamivir treatment in high-risk patients. The cost per QALY for zanamivir in these patients compares well with that of other commonly used pharmacological interventions.
CRD COMMENTARY - Selection of comparators The strategy of using placebo was regarded as the comparator. This allowed the active value of the intervention to be evaluated. It was reported that, although amantadine is registered for prophylaxis and treatment of influenza A, it offers neither the scope for treatment (being only effective against influenza A) nor sufficiently widespread acceptance for it to be considered the standard care.
Validity of estimate of measure of benefit The internal validity of the effectiveness results is likely to be high due to the performance of a pooled analysis of randomised, double-blind trials. However, reporting the search strategies adopted for the review; and systematically investigating the differences between studies could have further strengthened the internal validity of the effectiveness analysis. Furthermore, it was noted that since there were more placebo patients than zanamivir patients whose symptoms were not alleviated at the end of studies, the extrapolation of censored data undervalues the effect of zanamivir.
Validity of estimate of measure of benefit:
Estimation of benefits was directly based on the effectiveness analysis. The utility values were not adopted from the same samples used in the effectiveness analysis. The use of a sample of the general population to elicit health utility changes is likely to over estimate QALY scores for the intervention as high risk patients for influenza (used in the analysis) will have lower "normal" utilities. The effect on cost-utility ratios of using the target population could be to raise the ratio. The wide range of benefit measures utilised appears to be very informative.
Validity of estimate of costs The positive aspects of the cost analysis which enhanced its validity were: adequate details of methods of cost estimation were given and resource use/cost profiles were reported; it appears that all important direct cost elements were included in the direct cost analysis; the price year and the perspective adopted in the cost analysis were specified; sensitivity analysis was performed on some parameters affecting costs. Although hospitalisations were not collected prospectively in the trials they were included as a sensitivity analysis based on SAE forms. The main weakness of the cost analysis was the fact that the effects of alternative procedures on indirect costs were not addressed.
Other issues The authors' conclusions appear to be justified given the sensitivity analysis performed and cost-effectiveness acceptability curves generated. Some comparisons were made with other studies. One of the limitations of this study was the fact that the impact on GPs of patients with influenza presenting outside 48 hours and patients with other respiratory infections was not evaluated.
Implications of the study The results suggest that health benefits can be gained from the use of zanamivir within acceptable economic thresholds, but it was suggested that uncertainties around the diagnostic rate and consultation patterns can only be reliably estimated once zanamivir is in routine use.
Bibliographic details Griffin A D, Perry A S, Fleming D M. Cost-effectiveness analysis of inhaled zanamivir in the treatment of influenza A and B in high-risk patients. PharmacoEconomics 2001; 19(3): 293-301 Other publications of related interest Lalezari J, Campion K, Keene O, Silagy C. Zanamivir for the treatment of influenza A and B infection in high-risk patients: a pooled analysis of randomized controlled trials. Archives of Internal Medicine 2001;161:212-7.
Indexing Status Subject indexing assigned by NLM MeSH Administration, Inhalation; Adult; Aged; Antiviral Agents /therapeutic use; Cost-Benefit Analysis; Female; Guanidines; Humans; Influenza A virus; Influenza B virus; Influenza, Human /drug therapy /economics; Male; Middle Aged; Pyrans; Sialic Acids /administration & Zanamivir; dosage /therapeutic use AccessionNumber 22001008109 Date bibliographic record published 30/09/2001 Date abstract record published 30/09/2001 |
|
|
|