|
Twelve months of isoniazid compared with four months of isoniazid and rifampin for persons with radiographic evidence of previous tuberculosis: an outcome and cost-effectiveness analysis |
Jasmer R M, Snyder D C, Chin D P, Hopewell P C, Cuthbert S S, Paz E A, Daley C L |
|
|
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 Isoniazid (ISO) was compared with isoniazid and rifampin (ISO-RIF) for the treatment of latent tuberculosis (TB) infection in individuals with radiographic evidence of prior TB and a positive tuberculin skin test, who had not received adequate treatment. In the ISO alone regimen, ISO was taken at a dose of 300 mg/day for 12 months. In the combined regimen, ISO 300 mg and RIF 600 mg were taken daily for 4 months. The two regimens were self-administered.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised a hypothetical cohort of individuals with radiographic evidence of prior TB, which had not been adequately treated, and a positive tuberculin skin test.
Setting The setting was likely to have been primary care. The economic study was conducted in the USA.
Dates to which data relate The effectiveness data were gathered from 1993 to 1996. Other effectiveness evidence and resource use information was derived from studies published between 1982 and 1997. The price year was 1999.
Source of effectiveness data The effectiveness evidence was derived from a single study, a review of completed studies, and some authors' assumptions.
Link between effectiveness and cost data The costing was conducted on a sample of patients different from that used in the effectiveness study.
Study sample The use of power calculations was not reported. All eligible patients who presented at the clinic considered in the study from 1993 to 1994 received ISO alone (n=545). All those who presented from 1995 to 1996 (after the ATS/CDC guidelines were published) received ISO-RIF (n=477). However, the authors stated that 4% of those patients enrolled in 1995 to 1996 started treatment on ISO alone. The mean age of the whole cohort of 1,022 patients was 52.3 (+/- 16.3) years, 57% of the sample were male and 93% were born outside the USA.
Study design This was a non-randomised comparative study that was conducted in a single centre, the San Francisco Tuberculosis Clinic in California (USA). It was unclear whether the study was conducted prospectively or retrospectively. Most of the patients were recruited in two different time frames. The actual length of follow-up was 5.5 years in the ISO group and 3.5 years in the ISO-RIF group. The patients were evaluated monthly to assess the adherence to therapy and potential side effects. The loss to follow-up was not reported.
Analysis of effectiveness It appears that all the patients included in the initial study sample were considered in the analysis of effectiveness. The outcome measures used were completion rates, the occurrence of adverse events and the development of active TB. The two groups were comparable at baseline in terms of their demographic and disease characteristics.
Effectiveness results In the ISO alone group, the completion rate was 79.8% for 12 months' treatment, while 3.4% self-stopped between 6 and 11 months' treatment, and 16.7% self-stopped after less than 6 months' treatment. In the ISO-RIF group, the completion rate was 83.6% for 4 months' treatment while 7.1% self-stopped between 2 and 3 months' treatment, and 9.2% self-stopped after less than 2 months' treatment.
There was no statistically significant difference in the percentage of completers for ISO alone (12 months) and ISO-RIF (4 months).
The rate of adverse events was 4.9% in the ISO alone group and 6.1% in the ISO-RIF group. The difference was not statistically significant.
There were 6 cases of active TB in the ISO alone group versus 3 in the ISO-RIF group, (p>0.05).
Clinical conclusions The effectiveness results, which were used in the decision model, showed that the two treatments were similar in terms of completion rates and the development of active TB.
Modelling A Markov model was used to estimate the costs and benefits of the three alternative strategies in a hypothetical cohort of 1,000 persons with radiographic evidence of prior TB, which had not been adequately treated, and a positive tuberculin skin test. The model considered patients aged from 52 to 99 years. Other details of the model, which was likely to have been deterministic, were not reported. The model was populated with data derived from three sources (see Source of Effectiveness Data).
Outcomes assessed in the review The outcomes assessed from the literature were:
the annual risk of developing active TB among patients with radiographic evidence of prior TB;
the death rate for patients who developed TB between the age of 52 and 64 years, and at 65 years or older;
the all-cause risk of death; and
the rate of reduction in the annual risk of TB.
Study designs and other criteria for inclusion in the review The authors stated that life expectancy was derived from US life tables, while the efficacy of the treatment was derived from a study of 8,428 patients. However, the design of the primary studies was not reported.
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 Four primary studies were included in the review.
Methods of combining primary studies Investigation of differences between primary studies Results of the review The annual risk of developing active TB among patients with radiographic evidence of prior TB was 1%.
The death rate was 7.4% for patients who developed TB between the age of 52 and 64 years, and 16.8% for those who developed TB when aged 65 years or older.
The all-cause risk of death was 3.61% in the age class 52 - 59 years, 4.72% in the age class 60 - 69 years, 7.09% in the age class 70 - 79 years, and 11.9% in the age class 80 years and older.
The rate of reduction in the annual risk of TB with ISO alone was 89% after 12 months, 67% for 6 to 11 months' treatment, and 0% for less than 6 months' treatment.
The rate of reduction in the annual risk of TB with ISO-RIF was 89% after 4 months, 67% for 2 to less than 4 months' treatment, and 0% for less than 2 months' treatment.
Methods used to derive estimates of effectiveness The authors made some assumptions that were used in the decision model.
Estimates of effectiveness and key assumptions It was assumed that no patients were infected with Mycobacterium tuberculosis strains that were resistant to ISO. It was also assumed that TB-related deaths occurred in the same year in which the active disease developed.
Measure of benefits used in the economic analysis The summary benefit measures were the expected numbers of TB cases, TB-related deaths and life expectancy. These were derived from the decision model. The number of TB cases was discounted at an annual rate of 3%.
Direct costs Discounting was relevant since lifetime costs were estimated. An annual rate of 3% was applied. The unit costs were not reported separately from the quantities of resources used. The health services in the economic evaluation were initial screening and treatment, including medications, physician visits, follow-up visits with nurses, chest radiograph, diagnostic tests, vitamin B6 and hospitalisations. The cost/resource boundary of the health care system was used. Resource use was estimated from a published study and authors' assumptions. The costs were derived from the San Francisco Department of Public Health and published estimates. All the costs were inflated to 1999 values using the medical care component of the US Consumer Price Index.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not considered.
Sensitivity analysis Threshold analyses, to identify the critical values at which the ISO alone strategy became more cost-effective than ISO-RIF, were conducted. One-way sensitivity analyses were also performed to assess the robustness of the estimated cost-savings. The rate of ISO resistance was varied and the duration of ISO alone treatment was shortened (from 12 to 9 months).
Estimated benefits used in the economic analysis The expected number of TB cases was 157.5 (116.9 discounted) with no treatment, 43.7 (32.1 discounted) with ISO alone and 34.7 (25.5 discounted) with ISO-RIF.
The expected number of TB-related deaths was 17.7 with no treatment, 5.0 with ISO alone and 4.0 with ISO-RIF.
The expected life expectancy would increase by 1.42 years with ISO alone and by 1.53 years with ISO-RIF.
Cost results The initial costs in the cohort of 1,000 patients were $593,757 (approximately $594 per patient) with ISO alone and $592,107 (approximately $592 per patient) with ISO-RIF.
The expected discounted costs of treating active disease and contract tracing and treatment (life-time analysis) were $1,861 per patient with no treatment, $539 with ISO alone and $405 with ISO-RIF.
The cost-savings per patient relative to no treatment were $720 with ISO alone and $864 with ISO-RIF. ISO-RIF led to cost-savings of $135 per patient in comparison with ISO alone.
Synthesis of costs and benefits An incremental cost-effectiveness ratio, to combine the costs and benefits, was calculated. However, both treatment strategies dominated the no treatment option, being both more effective and less expensive. ISO-RIF dominated ISO alone.
The threshold analysis showed that ISO alone led to cost-savings in comparison with ISO-RIF, but only when the cost of ISO-RIF doubled or the efficacy of ISO-RIF decreased by 8.6% to 81%.
The sensitivity analysis showed that even a small increase in ISO resistance (to 10%) would lead to cost-savings in the ISO-RIF group.
Reductions in the duration of ISO alone treatment (9 months) did not change the conclusions of the analysis. However, they did reduce the gap between the cost-savings observed with ISO alone and ISO-RIF.
Authors' conclusions Both isoniazid (ISO) alone and isoniazid with rifampin (ISO-RIF) led to cost-savings and gains in survival when compared with no treatment for patients with radiographic evidence of prior tuberculosis (TB), which was not adequately treated, and a positive tuberculin skin test. ISO-RIF was associated with a financial advantage over ISO alone, even when the latter was administered for 9 months.
CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparator was clear. The authors stated that ISO alone and ISO-RIF were treatments recommended by the ATS/CDC for the patients considered in the study. The strategy of no treatment was selected so as to assess the active value of the two treatments under evaluation. You should decide whether they represent valid comparators in your own setting.
Validity of estimate of measure of effectiveness The effectiveness evidence came from multiple sources. Some data were derived from a single observation study, which was not randomised. The baseline characteristics of the two groups were not reported, although the authors stated that the two groups were comparable. It was unclear whether the study was prospective or retrospective. The method of selecting the sample was reported, but it was not stated whether some patients were excluded from the initial study sample for any reason. Similarly, the loss to follow-up was not reported. The main limitations to the validity of the analysis were the lack of power calculations, the non-randomised design, and the fact that outcomes were gathered in two different timeframes. The latter issue is of particular importance since factors other than the study interventions could have affected the results of the analysis, although the authors stressed that the same protocol was applied during both periods. However, due to the design of the study, both selection and assessment bias might have occurred. The study sample comprised consecutive patients which were representative of the study population.
In terms of the evidence derived from the literature, a formal review of the literature was not conducted. In addition, there was little information on the types of studies used in the analysis. Issues such as the validity of the study and the methods used to combine primary estimates were not addressed. Finally, the authors did not justify their choice of assumptions and only one assumed estimate was tested in the sensitivity analysis. These issues tend to limit the internal validity of the analysis.
Validity of estimate of measure of benefit A group of benefit measures was used in the analysis. All of them favoured the ISO-RIF group and were not combined with the costs. The benefit measures were derived from the decision model, which was not described in detail. Only the number of TB cases was discounted. Life expectancy could be comparable with the benefits of other health care interventions. The impact of the interventions on quality of life was not assessed.
Validity of estimate of costs The authors explicitly reported the perspective adopted in the study. It appears that all the relevant costs have been included in the analysis. A breakdown of the categories of costs was provided. Since the indirect costs might have been relevant, it would have been interesting had a wider perspective been adopted. Information on the unit costs and the quantities of resources used was not provided satisfactorily, which reduces the possibility of replicating the study. The source of the cost data was reported. Resource use was not derived from the same sample of patients as that used in the effectiveness study. The price year was reported, which aids reflation exercises in other settings. The cost estimates were specific to the study setting. The sensitivity analyses investigated variations in some items.
Other issues The authors made few comparisons of their findings with those from other studies. They also did not address the issue of the generalisability of the study results to other settings. Some one-way sensitivity analyses were conducted but the overall external validity of the analysis appears to have been low. The study referred to a specific patient population and this was reflected in the conclusions of the analysis. The authors noted that conservative assumptions were made in the analysis.
Implications of the study The study results suggested that, regardless of the therapy used, patients at high-risk for ISO-resistant TB infections should be treated. ISO-RIF should be considered the treatment of choice due to the cost-savings and life expectancy gained in comparison with ISO alone.
Source of funding Supported by grants from the National Institutes of Health (A1 01549 and A1 34238).
Bibliographic details Jasmer R M, Snyder D C, Chin D P, Hopewell P C, Cuthbert S S, Paz E A, Daley C L. Twelve months of isoniazid compared with four months of isoniazid and rifampin for persons with radiographic evidence of previous tuberculosis: an outcome and cost-effectiveness analysis. American Journal of Respiratory and Critical Care Medicine 2000; 162(5): 1648-1652 Other publications of related interest Dutt AK, Moers D, Stead WW. Smear-and culture-negative pulmonary tuberculosis:four-month short-course chemotherapy. American Review of Respiratory Disease 1989;139:867-70.
Goldberg SV, Duchin JS, Shields T, Nolan CM. Four-month, four-drug preventive therapy for inactive pulmonary tuberculosis. American Journal of Respiratory and Critical Care Medicine 1999;160:508-12.
Indexing Status Subject indexing assigned by NLM MeSH Antitubercular Agents /administration & Cost-Benefit Analysis; Drug Administration Schedule; Drug Costs; Drug Therapy, Combination; Female; Health Care Costs; Humans; Isoniazid /administration & Life Expectancy; Male; Middle Aged; Outcome Assessment (Health Care); Patient Compliance; Recurrence; Rifampin /administration & San Francisco; Tuberculosis, Pulmonary /drug therapy /economics /radiography; dosage /adverse effects /economics; dosage /adverse effects /economics; dosage /economics AccessionNumber 22000001840 Date bibliographic record published 30/09/2004 Date abstract record published 30/09/2004 |
|
|
|