|
Cost effectiveness of ACE inhibitor treatment for patients with Type 1 diabetes mellitus |
Dong F B, Sorensen S W, Manninen D L, Thompson T J, Narayan V, Orians C E, Gregg E W, Eastman R C, Dasbach E J, Herman W H, Newman J M, Narva A S, Ballard D J, Engelgau M 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 initiation of treatment with an angiotensin-converting enzyme (ACE) inhibitor (captopril) was assessed in adult patients with Type I diabetes. Treatment immediately after diagnosis was compared with treatment after the onset of microalbuminuria.
Study population The study population comprised a hypothetical cohort of 10,000 adults newly diagnosed with Type I diabetes. Distribution of gender and race ethnicity within the cohort was representative of the general US population, that is, 50% female and 82.9% White, 11.0% Black, 5.8% Hispanic, 0.1% American Indian and 0.28% Asian/Other. The cohort was assumed to have an age at diagnosis of between 20 and 30 years, 25 years on average. The level of glycosylated haemoglobin (HbA1c) in this patient population was assumed to have a normal distribution, with a mean of 8.0% and standard deviation of 1.2%.
Setting The setting appears to have been secondary care (outpatient). The economic study was conducted in the USA.
Dates to which data relate The effectiveness evidence was derived from literature published from 1987 to 2000. The costs were derived from sources published from 1994 to 2000. The price year was 1999.
Source of effectiveness data The effectiveness data were derived from a review or synthesis of completed studies.
Modelling A semi-Markov Monte Carlo simulation model was built to assess the lifetime costs and benefits of early (1 year after the diagnosis of diabetes) versus late (only after the diagnosis of microalbuminuria) treatment with an ACE inhibitor in a hypothetical cohort of persons with newly diagnosed Type I diabetes. The model structure incorporated three microvascular complications (nephropathy, retinopathy and neuropathy). ACE inhibitor treatment was expected to directly affect only nephropathy, but indirectly it would affect all microvascular complications since life expectancy might be affected. Nephropathy was modelled as four sequential disease states, namely, normoalbuminuria, microalbuminuria, macroalbuminuria and end-stage renal disease (ESRD). Death was the final state of the model, and diabetes-related mortality was caused by either lower-extremity amputation (LEA), ESRD, or cardiovascular disease (CVD). The study cohort was followed from the diagnosis of diabetes until death or age 95 years, up to a maximum of 70 years.
Outcomes assessed in the review The outcomes assessed included:
the transition probabilities of nephropathy- and retinopathy-related health states;
the probabilities for neuropathy and LEA;
the relative risk reduction of the transition probabilities of nephropathy-related health states following ACE inhibitor treatment;
the mortality rates resulting from LEA, ESRD and CVD;
the risk of CVD in the presence of nephropathy;
the characteristics of the screening test for microalbuminuria; and
the hypothetical study population characteristics, including age, gender, race or ethnicity, and HbA1c values at diagnosis.
Study designs and other criteria for inclusion in the review Inclusion criteria for a review of any of the parameters were not explicitly stated. However, the majority of the effectiveness data were taken from randomised clinical trials. In addition, it was reported that major population surveys, epidemiological studies, clinical trials and other reported studies of persons with diabetes were used to determine the characteristics of the hypothetical study population.
Sources searched to identify primary studies Not stated. It was reported only that the studies used to determine the study population characteristics were identified from literature searches of MEDLINE (completed by an expert panel consisting of the paper's co-authors).
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 Three primary studies were included in the review. The results of one of the primary studies included were derived from three separate publications.
Methods of combining primary studies Some of the effectiveness parameters derived from primary studies were combined using a narrative method. These parameters were further examined in a sensitivity analysis.
Investigation of differences between primary studies Potential differences between the primary studies were not discussed.
Results of the review The ranges of annual transition probabilities of nephropathy-related health states without treatment with ACE inhibitor, depending on the duration of diabetes, were:
normo- to micro-albuminuria, from 1.42% (at 3 years) up to 4.38% (>/= 11 years);
micro- to macro-albuminuria, from 2.35% (at 3 years) falling to 0.82% (>/= 24 years);
macroalbuminuria to ESRD, 10% annually (at >/= 3 years).
The above transition probabilities were 0 for up to 2 years of diabetes.
The relative risk reductions for the transition probabilities due to ACE inhibitor treatment (captopril) were:
24% between normo- and micro-albuminuria;
65% between micro- and macro-albuminuria; and
50% between macroalbuminuria and ESRD.
The annual transition probabilities for retinopathy in patients with diabetes depended on the time since diagnosis and the different stages of retinopathy. Most retinopathy states developed only after 9 years of diabetes; for at least 9 years of disease the probabilities were:
0.1511 for background diabetic retinopathy;
0.0217 for macular oedema; and
between 0.0034 and 0.0248 (>/= 17 years) for proliferative diabetic retinopathy.
Additional probabilities were calculated for blindness resulting from either macular oedema or proliferative diabetic retinopathy if they remained untreated, or if they were treated.
The annual probability for peripheral neuropathy was 0.011249.
The annual probability for LEA was 0.0079 for patients with a disease history of at least 10 years, and 0 for patients with shorter disease history.
The mortality rates resulting from LEA, ESRD and CVD, as well as the risk of CVD in the presence of nephropathy, were not reported.
The characteristics of the study population were described in the 'Study Population' section.
Methods used to derive estimates of effectiveness Some of the effectiveness evidence was augmented by an expert panel, which consisted of the paper's co-authors.
Estimates of effectiveness and key assumptions For the base-case analysis, it was assumed that the screening test for microalbuminuria had 100% sensitivity and specificity.
Measure of benefits used in the economic analysis The primary outcome measures used were the cumulative lifetime incidence of ESRD, the average age at death, and the quality-adjusted life years (QALYs) as measured from the onset of diabetes. Utility values were assigned to year of life lived with major complications (blindness 0.69, ESRD 0.61, LEA 0.80) and were based on published literature. In order to ensure consistency of this model with similar published models, the utility values were selected from the literature with the criterion of having been used in other cost-effectiveness studies for diabetes.
Direct costs The direct costs included medical costs only. The costs were for drugs (ACE inhibitor), screening for microalbuminuria, renal, neurological and eye examinations, major diabetic complications (LEA, blindness, photocoagulation treatment, ESRD and CVD) and other routine treatment involving intensive glucose control, routine diabetes care and non-diabetes-related medical care. Resource use was reported separately for routine and intensive diabetes care, drug therapy and screening for microalbuminuria. The cost data were derived from literature published from 1994 to 2000. The drug costs were based on the Average Wholesale Price (AWP). The total lifetime costs were derived using modelling. The costs were adjusted to 1999 prices according to the US Bureau of Labor Statistics Medical Care Index and discounted at an annual rate of 3%. This rate corresponded roughly with US Treasury long-term inflation indexed bonds and had been used as the base-case rate in other cost-effectiveness analyses.
Statistical analysis of costs The costs were treated deterministically. No statistical analysis of the costs was undertaken.
Indirect Costs The indirect costs were not included in the analysis.
Sensitivity analysis A sensitivity analysis was carried out to test the robustness of the results under different assumptions and values of key parameters used in the base-case analysis. One-way sensitivity analyses were performed by varying the relative risk reduction on the transition probability between normo- and micro-albuminuria when ACE inhibitors were used, the drug costs, the characteristics (sensitivity and specificity) and the cost of the screening test for microalbuminuria, the age at diagnosis and the discount rate. The ranges used were based on authors' assumptions. In terms of drug costs, Health Care Finance Administration (HCFA) and market utilisation prices for generic captopril were tested.
Estimated benefits used in the economic analysis The benefits resulting from early treatment of patients with newly diagnosed Type I diabetes with ACE inhibitor were as follows:
cumulative lifetime incidence of ESRD, 4.51%;
average age at death, 63.4 years; and
total number of QALYs per patient, 20.456 over lifetime.
The benefits associated with treatment of patients with Type I diabetes with ACE inhibitor after the onset of microalbuminuria were as follows:
cumulative lifetime incidence of ESRD, 4.98%;
average age at death, 63.2 years; and
total number of QALYs per patient, 20.357 over lifetime.
Early treatment resulted in a 0.47% reduction in cumulative lifetime incidence of ESRD, an increase in average age of death equal to 0.3 years, and 0.099 additional QALYs per patient. The QALYs were discounted at an annual rate of 3%.
Cost results The total costs associated with early treatment of patients with Type I diabetes with ACE inhibitor were $130,460 per patient.
The total costs incurred by treating these patients with ACE inhibitor only after the diagnosis of microalbuminuria were $127,768 per patient.
Early treatment resulted in an incremental cost of $2,692 per patient.
The costs were estimated over lifetime and were discounted at an annual rate of 3%.
Synthesis of costs and benefits The costs and benefits were combined in the form of incremental cost-effectiveness ratios (ICERs), by dividing the incremental changes in costs by the incremental changes in QALYs for the two strategies.
The ICER of early ACE inhibitor treatment (captopril) compared with delaying treatment until the diagnosis of microalbuminuria was $27,143/QALY gained.
Separate sub-analyses were undertaken for different ages of the study population newly diagnosed with diabetes, and for different levels of HbA1c.
For an HbA1c level of 7%, the ICER ranged from $32,972/QALY (age at diagnosis 20 years) to $39,912/QALY (age at diagnosis 30 years).
For an HbA1c level of 8%, the ICER ranged from $28,783/QALY (age at diagnosis 20 years) to $30,674/QALY (age at diagnosis 30 years).
For an HbA1c level of 9%, the ICER ranged from $13,814/QALY (age at diagnosis 20 years) to $17,778/QALY (age at diagnosis 30 years).
The level of glycaemic control was a more important driver of cost-effectiveness than age.
The sensitivity analysis showed that the results were particularly sensitive to the relative risk reduction in the transition probability between normo- and micro-albuminuria caused by ACE inhibitors. At a 50% relative risk reduction the ICER became $8,814/QALY, while at a 10% relative risk reduction the ICER rose to $75,276/QALY.
The results were also sensitive to drug prices. At the HCFA reimbursement price for generic captopril, the ICER became $2,338/QALY.
When 30% sensitivity and 90% specificity of the screening test for microalbuminuria were assumed, the ICER fell to $3,674/QALY.
A 5% discount rate produced an ICER of approximately $41,000/QALY.
Cost-effectiveness was not significantly affected by variability in the screening test costs and by extending the lower age at diagnosis of diabetes to 10 years (while keeping initiation of ACE inhibitor treatment at 20 years).
Authors' conclusions Early treatment with an angiotensin-converting enzyme (ACE) inhibitor such as captopril provided modest benefit in the prevention of end-stage renal disease (ESRD), at reasonable cost-effectiveness from the US single-payer perspective, compared with delaying treatment until the diagnosis of microalbuminuria. This conclusion was sensitive to the extent that ACE inhibitors delayed the onset of microalbuminuria. Other factors, such as the patient's age and their glycaemic level, should be considered when deciding to initiate early treatment.
CRD COMMENTARY - Selection of comparators Although no explicit justification was given for the comparator used, the initiation of treatment with ACE inhibitors for patients with Type I diabetes mellitus only after the diagnosis of microalbuminuria represented routine practice in the authors' setting. You should consider whether the comparator reflects routine practice in your own setting.
Validity of estimate of measure of effectiveness It was not stated whether a systematic review of the literature had been undertaken. The authors used data from the available studies selectively. Although this is a common practice with models, it does not always ensure that the best data available are used in the model. One cannot be sure that all relevant literature was identified, although the estimates of effectiveness appear to have been derived credibly from the studies identified. The effectiveness estimates were combined using narrative methods. Potential differences between the primary studies were not discussed. Effectiveness parameters that were characterised by uncertainty were treated in sensitivity analysis.
Validity of estimate of measure of benefit The estimation of benefits was modelled. The semi-Markov Monte Carlo simulation model used for this purpose was appropriate, as it incorporated all health states of Type I diabetes that were likely to be affected by treatment. In addition, it allowed the estimation of lifetime benefits.
Validity of estimate of costs It was stated that the study adopted a single-party payer perspective. All costs relevant to this perspective were included in the analysis. In general, the costs and the quantities were not reported separately and this may hinder the generalisability of the results. However, resource use was estimated for cost elements involving drug therapy, screening for microalbuminuria, routine and intensive diabetes care. A sensitivity analysis of drug and screening costs was undertaken. The ranges of values used appear to have been appropriate. It was reported, as a potential limitation of the analysis, that although the perspective of a single payer was adopted, such a payer did not generally exist in the US (the closest example for such a payer was Medicare). Discounting was undertaken, which was appropriate since the costs were incurred during the lifetime of the patients. The date to which the prices referred was recorded, and this increases the reproducibility of the results.
Other issues The authors compared their findings with those of other studies and gave explanations for differences in the results. The issue of generalisability to other settings (US payers) was addressed. The results of the study were reported in full and the authors' conclusions reflected the scope of the analysis. The authors did not report any further limitations of their study.
Implications of the study The authors suggested that the initiation of early ACE inhibitor treatment for patients with Type I diabetes should be tailored to each individual, taking both age and the degree of glycaemic control into consideration. They felt that a recommendation of early treatment with ACE inhibitors for all patients with Type I diabetes was not justified. They pointed out the need for observational studies, or clinical trials, to determine the impact of early use of ACE inhibitors on transition probabilities from normoalbuminuria to microalbuminuria. In addition, they considered blood pressure, cholesterol levels and smoking status as factors that should be examined in the future, as they might lead to higher risks for both cardiovascular disease and neuropathy. Finally, they recommended that research be undertaken to develop effective strategies for reducing kidney disease among patients with Type I diabetes.
Source of funding Funded by the Centers for Disease Control and Prevention.
Bibliographic details Dong F B, Sorensen S W, Manninen D L, Thompson T J, Narayan V, Orians C E, Gregg E W, Eastman R C, Dasbach E J, Herman W H, Newman J M, Narva A S, Ballard D J, Engelgau M M. Cost effectiveness of ACE inhibitor treatment for patients with Type 1 diabetes mellitus. Pharamcoeconomics 2004; 22(15): 1015-1027 Indexing Status Subject indexing assigned by NLM MeSH Adult; Age Factors; Angiotensin-Converting Enzyme Inhibitors /economics /therapeutic use; Captopril /economics /therapeutic use; Cohort Studies; Cost-Benefit Analysis; Diabetes Complications /economics /prevention & Diabetes Mellitus, Type 1 /economics /complications /drug therapy; Humans; Markov Chains; Models, Economic; Monte Carlo Method; Quality of Life; Treatment Outcome; United States; control AccessionNumber 22004008405 Date bibliographic record published 31/10/2005 Date abstract record published 31/10/2005 |
|
|
|