|
Health and economic impact of combining metformin with nateglinide to achieve glycemic control: comparison of the lifetime costs of complications in the UK |
Ward A J, Salas M, Caro J J, Owens D |
|
|
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 This article assessed the impact of adding nateglinide (360 mg/day) to metformin (1,500 mg/day) in patients with glycaemic problems.
Economic study type Cost-effectiveness analysis and cost-utility analysis.
Study population The study was conducted with a hypothetical cohort of patients with Type 2 diabetes. Patient characteristics were assigned using distributions.
Setting The setting was primary care. The economic study was carried out in the UK.
Dates to which data relate The effectiveness data were obtained from studies published between 1984 and 2003. All complication costs were expressed in 1999 UK pounds sterling. The authors added that the prices of pharmaceutical products were from 2002, and that the costs of end-stage renal disease was estimated using data from 1996 and that this cost was not inflated.
Source of effectiveness data The effectiveness evidence was derived from a review and synthesis of published studies, some of them clinical trials, but no details were provided. Assumptions were made for some inputs, while risk equations were used for others.
Modelling A Monte Carlo simulation technique was used to simulate the risk of developing diabetes-related micro and macrovascular complication rates. The model had 30 annual cycles. Simulated patients were assigned clinical characteristics which were updated after each cycle. The risk of complications in each cycle was determined by functions. Patients entered the model and had some risk of dying and of developing micro and macrovascular complications. Micro complications were progressive but macro complications were treated as one-off events. There was a pre-model period of 7 years in which patients could accumulate complications, but the costs of treatment were not considered.
Outcomes assessed in the review The risks of death and of complications were assessed. Complications included nephropathy, retinopathy, neuropathy, end-stage renal disease, blindness, amputations, hypoglycaemia, foot ulcer, stroke and myocardial infarction.
Study designs and other criteria for inclusion in the review The authors gave no indication that a systematic review of the literature had been undertaken. The main epidemiological studies used were cited, as were the types of equations used to extrapolate and obtain data.
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 At least 12 studies were used to inform the model.
Methods of combining primary studies Investigation of differences between primary studies Results of the review The risks of death and complications were determined by statistical models and the distributions of clinical characteristics of the patients.
Methods used to derive estimates of effectiveness In some instances, the authors made assumptions about long-term predictions for effectiveness.
Estimates of effectiveness and key assumptions It was assumed that "after the initial improvement in glycemic control the HbA1c would begin to drift upward as it did with metformin and other hypoglycemic agents employed in the UK Prospective Diabetes Study".
Measure of benefits used in the economic analysis The measures of health benefits used were years of mean survival, life-years gained (LYG) and quality-adjusted life-years (QALYs). The method used to value health was not reported. The only information provided was that the values were obtained from the published literature and that point values were used for each of the main health states. The health benefits were discounted at a rate of 1.5%.
Direct costs The direct medical costs of complications were evaluated. The cost estimates were derived from regression equations taken from a publication in which resource use data were collected from the UK Prospective Diabetes Study (Clarke et al. 2003, see 'Other Publications of Related Interest' below for bibliographic details). The direct pharmaceutical costs were obtained from the Monthly Index of Medical Specialties. All complication costs were expressed in 1999 year terms, but it should be noted that the costs of end-stage renal disease were estimated based on 1996 data and were not inflated to 1999. In the case of complications, the inpatient costs of less severe complications were not taken into consideration. In the case of macrovascular costs, the treatment of milder conditions was also not included. The costs were discounted at a rate of 6%.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not reported.
Currency UK pounds sterling (). Conversion rates to US dollars ($) and Euros (EUR) were also reported. 1 = $1.7 = EUR 0.4.
Sensitivity analysis Several one-way sensitivity analyses were performed. The main parameters considered were:
the discount rate for health benefits and costs (no discount, 3% rate and 6% rate);
age (46.5 and 82.5);
the costs of complications (+/- 20% from the baseline results);
utilities (+/- 20% from the baseline results);
the upward drift of HbA1c (1.5% and 0% for metformin); and
the initial HbA1c before prescription (9.4% and 7.9%).
There was no information on the rationale used to determine the ranges used in the sensitivity analysis.
Estimated benefits used in the economic analysis Mean survival (discounted) was 13.5 (11.7) life-years with metformin and 13.9 (12.1) life-years with combination therapy. The incremental LYG (discounted) with combination therapy over metformin was 0.39 (0.32).
Mean survival (discounted) was 12.2 (10.7) QALYs with metformin and 12.6 (11.0) QALYs with combination therapy. The incremental quality-adjusted survival (discounted) was 0.46 (0.37).
Cost results In the metformin group, the mean cost of complications was 3,548 per patient and the total cost per patient was 5,093.
In the combination group, the mean cost of complications was 3,084 per patient and the total cost per patient was 7,159.
Hence, the difference between both treatments was -464 for complications and 2,066 for mean total costs.
Synthesis of costs and benefits The costs and benefits were synthesised by calculating the incremental cost per LYG and the incremental cost per QALY, comparing combination treatment versus metformin alone.
The incremental cost-effectiveness ratio was 5,403 for LYG and 4,500 for QALYs. The results for discounted LYG and QALYs were 6,772 and 5,609, respectively.
The model input that had the greatest impact on the estimated costs was the impact of the combination of nateglinide and metformin on HbA1c. On the other hand, the efficacy of the combination treatment on postprandial glucose had only a minor effect on costs. The sensitivity analysis showed that the results were consistent over broad ranges.
The population characteristics of the cohort have important implications for the results. The sensitivity analysis showed that combination treatment achieves the greatest savings when the patients are young, with longer duration of disease and poorer glycaemic control initially.
Authors' conclusions For patients who are not maintaining good glycaemic control, the combination of metformin and nateglinide is cost-effective and reduces the rate of diabetes-related complications. The increase in the cost of drug treatment is partially offset by the reduction in the cost of treating long-term diabetes complications.
CRD COMMENTARY - Selection of comparators The authors explicitly justified the selection of the treatments compared, mainly the importance of improving glycaemic control in diabetic patients. Nateglinide is one of the new agents developed for use with metformin for this purpose.
Validity of estimate of measure of effectiveness The estimates of the risk of complications were derived using statistical models from published studies, applied to patient data. The authors gave no indication that a systematic review of the literature was performed. It is therefore hard to tell whether the most appropriate data were used in the model. However, the validity and the predictive validity of the model were assessed in order to assure the implications of the study and the validity of the findings.
Validity of estimate of measure of benefit The authors reported LYG and QALYs as summary measures of health benefits. These are appropriate measures for detecting the impact of the interventions on quality and length of life, and will facilitate comparisons with the results of studies examining other interventions. However, there was no information on the method used to obtain the QALYs used. This hinders comparability since the method used to obtain QALYs influences the results obtained. An appropriate sensitivity analysis was performed to evaluate a range of utility values.
Validity of estimate of costs For the perspective adopted, all the relevant categories of cost were included in the analysis. The authors excluded costs of minor complications. It would have been useful had this decision been tested explicitly in a sensitivity analysis, although a range of costs were tested in the sensitivity analysis. The costs were not reported separately from the resources.
Other issues The authors stated that the results were consistent with the evidence found in other studies, although they recognised that some of the assumptions made about long-term effectiveness require more research to confirm the effectiveness values. Despite the check on the validity of the equations used to obtain data for the model, the fact that assumptions were used remains a limitation of the study. However, the sensitivity analysis demonstrated the robustness of the model, and how the combination of nateglinide and metformin can be a cost-effective therapy for patients with glycaemic control problems.
Implications of the study Glycaemic control is an important strategy in the treatment of a sub-group of diabetic patients.
Source of funding Funded in part by Novartis Pharma AG, UK.
Bibliographic details Ward A J, Salas M, Caro J J, Owens D. Health and economic impact of combining metformin with nateglinide to achieve glycemic control: comparison of the lifetime costs of complications in the UK. Cost Effectiveness and Resource Allocation 2004; 2(2) Other publications of related interest Glick H. The potential for CVD prevention by reducing postprandial hyperglycaemia. In: Proceedings of the IDEG. Acapulco: IDEG; 2000.
Cowie CC, Harris MI. Physical and metabolic characteristics of persons with diabetes. In: National Diabetes Data Group. Diabetes in America. 2nd ed. Bethesda (MD): National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 1995. p.179-220. NIH publication no.: 95-1468.
DECODE Study Group. Glucose tolerance and cardiovascular mortality: a comparison of fasting and 2-hour diagnostic criteria. Arch Intern Med 2001;161:397-404.
Clarke P, Gray A, Legood R, et al. The impact of diabetes-related complications on the health care costs: results from the United Kingdom Prospective Diabetes Study (UKPD Study no. 65). Diabetic Medicine 2003;20:442-50.
Indexing Status Subject indexing assigned by CRD MeSH Cost-Benefit Analysis; Diabetes Mellitus, Type 2; Great Britain; Humans; Hypoglycemic Agents; Metformin; Quality-Adjusted Life Years AccessionNumber 22005000825 Date bibliographic record published 31/05/2006 Date abstract record published 31/05/2006 |
|
|
|