|Economic impact of using an immunostimulating agent to prevent severe acute exacerbations in patients with chronic obstructive pulmonary disease
|Collet J P, Ducruet T, Haider S, Shapiro S, Robinson A, Renzi P M, Contandriopoulos A P, Ernst P
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 study examined the use of an immunostimulating agent (OM-85 BV) to prevent severe acute exacerbations (SAE) in patients with chronic obstructive pulmonary disease (COPD).
Type of intervention
Secondary prevention and treatment.
Economic study type
The study population comprised patients with a history of heavy smoking (20 pack-years or more) and a forced expiratory volume in 1 second (FEV1) between 20% and 70% of the predicted value, which was not reversible with salbutamol.
The setting was hospital. The economic study was carried out in Montreal, Quebec, Canada.
Dates to which data relate
The effectiveness and cost data were collected between September 1994 and June 1995. The price year reported was 1995.
Source of effectiveness data
Effectiveness data were derived from a single study. The effectiveness results from this single study are published by the authors elsewhere (see 'Other Publications of Related Interest' below).
Link between effectiveness and cost data
The costing was undertaken prospectively on the same patient population as that used in the effectiveness analysis.
The authors did not report that the sample size was calculated in the planning phase of the study to assure a certain power. Patients with moderate to severe COPD were recruited from 12 different institutions. These patients had a history of heavy smoking (20 pack-years or more) and a forced expiratory volume in 1 second (FEV1) between 20% and 70% of the predicted value, not reversible with salbutamol. The authors did not report whether the study sample was representative of the study population. The study included 381 patients: 191 in the OM-85 BV group, and 190 in the placebo group.
This was a double-blind, randomised controlled trial. The study was multi-centred (patients from 12 institutions were included in the effectiveness analysis). Randomisation was centralised and stratified by institution and degree of ventilatory impairment. The duration of follow-up was 6 months. There were 8 patients lost to follow-up: one left Canada (it was not reported to which group she belonged) and the other 8 died (2 in the OM-85 BV group, and 6 in the placebo group).
Analysis of effectiveness
The basis for the analysis of effectiveness seems to have been intention to treat. The primary health outcomes used in the analysis were: the risk of being hospitalised for a respiratory cause, the reduction in the risk for OM-85 BV in comparison with placebo, the length of hospital stay for a respiratory problem, the total number of days spent in hospital for a respiratory reason, and the average number of hospital days per patient. The authors also reported the number of hospitalisations for a non-respiratory cause and the total and mean length of stay for both the OM-85 BV and the placebo group. The groups were shown to be comparable at analysis in terms of baseline characteristics such as sex, age, education, dyspnea, FEV1, use of antibiotics, etc.
The effectiveness results were as follows:
The risk of being hospitalised for a respiratory cause was 31/191 (16.2%) for OM-85 BV and 44/190 (23.2%) for placebo, (p=0.089), and the reduction in the risk for OM-85 BV in comparison with placebo was 30%.
The mean length of hospital stay for a respiratory problem was 6.5 days (standard deviation (SD) = 8.3; median = 4) for OM-85 BV, and 11.3 days (SD = 16.1; median = 6) for placebo, (p=0.058).
The total number of hospital days for a respiratory reason among those patients in the OM-85 BV group was 287, and 642 for patients in the placebo group.
The average number of hospital days per patient in the OM-85 BV group was 1.5 days (SD = 5.0) and in the placebo group 3.4 (SD = 11.3), (p=0.037).
In the OM-85 BV group, the number of hospitalisations for a non-respiratory cause was 31, versus 29 for those patients in the placebo group.
The total length of stay due to non-respiratory causes was 189 days for the OM-85 BV group (mean = 6.5 days; SD = 7.3; median = 4) and 255 days for the placebo group (mean = 8.2 days; SD = 10.8; median = 6).
Patients treated with OM-85 BV had a lower risk of being hospitalised for respiratory problems (although the difference was not statistically significant), and the length of hospital stay due to respiratory problems was also lower than those in the placebo group. On the contrary, there were no differences with regard to risk of hospitalisation due to non-respiratory causes.
Measure of benefits used in the economic analysis
The summary measures of effectiveness used in the study were the number of hospitalisations for a respiratory cause, and the average number of hospitalisation days per patient, which were considered by the authors as a marker of disease severity.
Only some resource quantities were reported independently from the costs. The direct costs reported were those of the health service, and included: the costs for prescribing the immunostimulating agent (including the cost of the drug and the cost of the physician visit); the costs for ambulatory treatment of acute exacerbations (including medical visits to GPs or specialists, prescription and non-prescription medications, non-pharmaceutical therapy, and all tests and diagnostic procedures); the hospital-related costs; and the costs associated with the help needed by the patients in relation to acute exacerbations (personal care, housekeeping, meals, transportation and other activities). The cost of the drug in Europe was used for the estimation of the costs. The direct costs for ambulatory treatment of acute exacerbations were estimated according to the Regie d'Assurance Maladie du Quebec (RAMQ) documents. The hospital-related costs were modelled, and the algorithm included primary and secondary diagnoses, information about disease severity and comorbidity, allowance for prolonged duration of stay and allowance for death in hospital. The costs associated with the help needed by the patients were valued according to the average wage in Quebec (1995). Discounting was not carried out but, as the authors stated, it was not necessary due to the short duration of the trial (less than 1 year). The authors reported average costs. The price year was 1995.
Statistical analysis of costs
The means and standard deviations of all the direct costs included in the analysis were reported. T-tests were carried out to compare the mean costs per event in each treatment arm.
The authors estimated the indirect costs associated with the productivity losses produced by the effects of acute exacerbations on ability to work. These costs were estimated according to the mean hourly wage in Quebec (1995), based on actual data. Quantities and costs were analysed separately. Discounting was not performed because the period of study was shorter than 2 years. The price year was 1995.
The authors reported that both cost of treatment and length of hospitalisation may have been affected by the presence of outliers. In order better to reflect the patient distribution and variability, they computed the 95% confidence intervals (CIs) of costs and hospitalisation lengths, and the differences between OM-85 BV and placebo, by means of bootstrapping the study population 1000 times. They used different cost and effectiveness values within the limit of the variability observed in the study. Therefore, the areas of uncertainty investigated were variability in data.
Estimated benefits used in the economic analysis
See the 'Effectiveness Results' section.
The cost results were as follows:
The average cost per patient for hospitalisation related to a respiratory cause was Can$853 (SD = 2,459) for the OM-85 BV group versus Can$1,521 (SD = 4,122) for the placebo group, and the difference in cost per patient between OM-85 BV and placebo was -668 (95% CI: -1,350 - 14).
The average cost per patient for hospitalisations related to both respiratory and non-respiratory disease was Can$1,328 (SD = 3,172) for OM-85 BV) versus Can$2,279 (SD = 5,125), and the difference in the cost per patient between the groups was -951 (95% CI: -1,807 to -95).
The cost to prevent hospitalisations was higher in the OM-85 BV group (mean = Can$165; SD = Can$147) than in the placebo group (mean = Can$79; SD = Can$144).
The distribution of the difference between the groups, obtained by bootstrapping, was centred on Can$85 (95% CI: 58 - 115).
Synthesis of costs and benefits
The estimated benefits and costs were combined in a cost-effectiveness ratio (CER). The estimated extra cost for preventing one day of hospitalisation for a respiratory cause by using OM-85 BV was Can$45 (95% CI: 18 - 210), (while the average cost for one day of hospitalisation was Can$382). The estimated extra cost per patient to prevent hospitalisation was Can$85, (while the difference in hospitalisation-related cost per patient between OM-85 BV and placebo was Can$668 for respiratory causes, and Can$951 for all-cause hospitalisations).
Bootstrapping and sensitivity analysis showed that the results were robust and were not likely to be due to random variations. The difference between the groups regarding the mean days of hospitalisation for a respiratory reason had only a 2% chance of being observed by chance. When the cost of the drug was considered as being Can$200 (instead of Can$50, which was the value considered at analysis) the resultant CER showed that the extra cost of preventing one day of hospitalisation was Can$119 (95% CI: 49 - 195).
The use of an immunostimulating agent may decrease the financial burden of acute exacerbations in patients with COPD on both the health system and the patients, by reducing the severity of the exacerbations and therefore, the number of hospitalisations and the length of hospital stay per patient. The authors concluded that the use of immunostimulants may become a key element in the control of COPD, given the high prevalence of this disease worldwide and the high costs of acute exacerbations.
CRD COMMENTARY - Selection of comparators
The authors gave no justification for the choice of placebo as a comparator. You, as a user of this database, should assess whether placebo represents a valid comparator in your own setting.
Validity of estimate of measure of effectiveness
The analysis was based on a randomised controlled trial that seemed to be appropriate for the study question. The blinding method for assessment of the outcomes was not reported. Although, the study sample was not shown to be representative of the study population, the fact that the patients were recruited from 12 different institutions increased the likelihood that the study sample was representative. The patient groups were shown to be comparable at analysis with regard to some baseline characteristics such as sex, age, education, dyspnea, FEV1 and use of antibiotics. Many of the missing details may have been reported within the efficacy paper. (See 'Other publications of related interest'.)
Validity of estimate of measure of benefit
The estimation of benefits was obtained directly from the effectiveness analysis. The authors justified this choice as they considered the measures of health benefit included to be a marker of disease severity.
Validity of estimate of costs
It appears that all the costs relevant to the perspective adopted were included in the analysis. Costs and resources were not reported separately, and, although the price year was given, the fact that not all the resource quantities were given independently from the costs could make the generalisation of results to other settings difficult. Statistical and sensitivity analyses of costs and quantities were conducted. It was not clearly stated whether the indirect costs associated with lost productivity were included in the analysis. As the authors adopted a health service perspective, these costs did not need to be included. If included, the study may have overestimated the costs of both groups considered at analysis, favouring the OM-85 BV group in comparison with the placebo group because the number of days off work was higher among the patients in the placebo group. However, if a societal perspective were to have been adopted, these indirect costs would have to be included.
The authors made comparisons of their findings with those from other studies. The issue of generalisability of results to other settings was not addressed. The authors do not appear to have presented their results selectively. The authors' conclusions reflected the scope of the analysis, and the use of bootstrapping allowed the authors to show that the results obtained regarding effectiveness and costs were not likely to have been due to chance.
Implications of the study
The results of the study show that OM-85 BV is likely to reduce severity, hospitalisations, length of hospital stay and help needed by patients with COPD, and therefore, their associated costs. The authors state that an accurate identification of those patients most likely to be hospitalised (as the target population to be treated) is important in order to increase the efficiency of the drug and to optimise the benefits.
Source of funding
Funded by Jouveinal Inc. Dr J-P Collet was supported by a personal research grant from the Fond de la Recherche en Sante due Quebec.
Collet J P, Ducruet T, Haider S, Shapiro S, Robinson A, Renzi P M, Contandriopoulos A P, Ernst P. Economic impact of using an immunostimulating agent to prevent severe acute exacerbations in patients with chronic obstructive pulmonary disease. Canadian Respiratory Journal 2001; 8(1): 27-33
Other publications of related interest
Anthonisen N R. OM-85 BV for COPD. American Journal of Respiratory and Critical Care Medicine 1997;156: 1713-4.
Collet J P, Shapiro S, Ernst P, Renzi T, Ducruet T, Robinson A, for the PARI-IS Study Steering Committee and Research Group. Effects of an immunostimulating agent on acute exacerbations and hospitalisations in patients with chronic obstructive pulmonary disease. American Journal of Respiratory and Critical Care Medicine 1997;156:1719-24.
Efron B, Tibshirani J R. An Introduction to the Bootstrap (Monographs on Statistics and Applied Probabilities, No 57). London: Chapman & Hall/CRC Press, 1994.
Grove A K, Bergemann R, Kelle R. Preventive treatment of chronic bronchitis: a cost-effectiveness analysis for an immunoactive bacterial extract in Switzerland. British Journal of Medical Economics 1996;10:1-14.
Orcel B, Delclaux B, Baud M, Derenne J P. Oral immunization with bacterial extracts for protection against acute bronchitis. European Respiratory Journal 1994;7:446-52.
Subject indexing assigned by NLM
Adjuvants, Immunologic /administration & Bacteria; Cell Extracts; Cost-Benefit Analysis; Female; Health Care Costs; Humans; Length of Stay /economics; Lung Diseases, Obstructive /drug therapy /economics; Male; Prevalence; Severity of Illness Index; Treatment Outcome; dosage /economics /therapeutic use
Date bibliographic record published
Date abstract record published