|
Effect of varying threshold and selective versus universal strategies on the cost in gestational diabetes mellitus |
Larijani B, Hossein-Nezhad A, Vassigh A R |
|
|
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 authors assessed the screening method best-suited for gestational diabetes mellitus (GDM) in Iran. The screening and diagnostic strategies compared were selective versus universal strategies with thresholds of 130 and 140 mg/dL.
A two-step approach for screening and diagnosis was used. The 50-g glucose challenge test (GCT) was used as a screening test, while the 100-g oral glucose tolerance test (OGTT) was used as the diagnostic test. Two or more OGTT readings were considered diagnostic of GDM.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised pregnant women. No specific inclusion or exclusion criteria were reported. The selective strategies targeted women with at least one risk factor. The risk factors examined were known diabetes in first-degree relatives, history of poor obstetric outcome, polyhydramnios, history of having a macrosomic child, glycosuria, maternal age 35 or over, and obesity. Poor obstetric outcomes included spontaneous abortion, neonatal death, intrauterine foetal death, anomaly and preterm labour.
Setting The setting is likely to have been community care. The economic study was carried out in Tehran, Iran.
Dates to which data relate The dates to which the effectiveness evidence and resource use data related were not reported. The price year was 2002.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The same patients provided both the effectiveness and cost data. There was insufficient evidence to determine whether or not the costing had been carried out prospectively.
Study sample The authors did not report that power calculations were used to estimate the potential impact of chance on the results. They did not report whether any patients refused to participate or who were excluded from the initial sample. The authors appear to have included all pregnant women who entered the study setting during the dates of the study. A total of 2,146 pregnant women were enrolled in the study.
Study design The analysis was based on a diagnostic yield study. The study was carried out at four university teaching hospitals. High-risk patients were screened during their first antenatal visit. If they were found not to have GDM at an initial screening, they were re-tested between the 24th and 28th weeks of their gestation. The remaining women (without risk factors) were also screened for GDM between the 24th and 28th weeks of their pregnancy. All women with impaired glucose tolerance and/or symptoms suggestive of hyperglycaemia were followed up and re-tested between the 32nd and 36th weeks of pregnancy. After universal testing was completed, samples were re-tested as if selective or risk factor-based screening had been used.
Analysis of effectiveness The primary health outcomes were the prevalence of GDM and the sensitivities of the screening strategies. The reference strategy was universal screening at a threshold of 130 mg/dL. The authors reported summary statistics for the study participants.
Effectiveness results The prevalence of GDM was 4.7%.
The sensitivity was 100% for the "universal 130" strategy (the reference strategy), 88% for the "universal 140" strategy, 86% for the "selective 130" strategy and 77% for the "selective 140" strategy.
Clinical conclusions The authors did not report a clinical conclusion. The findings indicated that universal screening at a threshold of 130 mg/dL was the more sensitive strategy. Switching from the 130 to the 140 mg/dL threshold decreased case-detection sensitivity by 12%. The selective screening strategies were the least sensitive strategies.
Measure of benefits used in the economic analysis The measure of benefits used was the number of GDM cases detected. This was derived from the effectiveness analysis.
Direct costs The costs measured were those of the hospital. The costs were broken down into unit costs and quantities. The costs measured were for serum glucose determination, the test solution for the GCT, the test solution for the OGTT and phlebotomy. The authors reported that the costs were calculated based on the mean value of the public and private sector tariffs, assuming standard material and services. Discounting was not reported, but it would not have been relevant if the authors were interested only in the immediate screening and diagnostic costs. The price year was 2002.
Statistical analysis of costs The authors did not report any statistical analyses of the costs.
Indirect Costs The indirect costs were not included in the analysis.
Currency Iranian rials (IRR). The conversion rate to US dollars ($) was $1 = IRR 8,000.
Sensitivity analysis No sensitivity analysis was reported.
Estimated benefits used in the economic analysis The number of GDM cases detected was:
114 when using the "universal 130" strategy,
100 when using the "universal 140" strategy,
98 when using the "selective 130" strategy, and
88 when using the "selective 140" strategy.
Switching from the 130 to the 140 mg/dL threshold would result in 12% of GDM cases being missed when using the universal strategy. When switching from the "universal 130" strategy to the "selective 130" strategy, 14% of GDM cases would be missed. When switching from the "universal 130" strategy to the "selective 140" strategy, 23% of GDM cases would be missed.
Cost results The total costs were not reported.
The cost per pregnant woman was:
IRR 30,140 ($3.80) when using the "universal 130" strategy,
IRR 25,641 ($3.20) when using the "universal 140" strategy,
IRR 21,703 ($2.71) when using the "selective 130" strategy, and
IRR 19,124 ($2.39) when using the "selective 140" strategy.
Synthesis of costs and benefits The average costs per GDM case detected were reported.
The cost per GDM case detected was:
IRR 644,488 ($80.56) when using the "universal 130" strategy,
IRR 619,500 ($77.43) when using the "universal 140" strategy,
IRR 535,052 ($66.88) when using the "selective 130" strategy, and
IRR 525,044 ($65.63) when using the "selective 140" strategy.
Authors' conclusions The authors concluded "in the Tehran population and any other populations with similar demographic and socioeconomic conditions, where the baseline prevalence of GDM (gestational diabetes mellitus) is moderate to high and a significant number of cases are missed with selective screening, universal screening is the best option".
CRD COMMENTARY - Selection of comparators The choice of the reference threshold of 130 mg/dL was based on the Carpenter and Coustan criteria. However, the authors did not justify the alternative threshold (140 mg/dL). The choice of the risk factors examined for the selective screening strategies appears to have been based on the literature.
Validity of estimate of measure of effectiveness The analysis was based on a diagnostic yield study, which was appropriate for addressing the study question pertaining to the sensitivity of the screening strategies. Details of the study sample, such as demographics, were given to set the context for the reader and to enable an assessment of generalisability to other settings. The authors reported appropriate outcomes for a diagnostic yield study. However, the study design did not allow an assessment of the number of false positives and true negatives obtained with each screening strategy. It therefore provides a limited measure of the healthy benefits.
Validity of estimate of measure of benefit The measure of benefits was the number of GDM cases detected, which was derived from the effectiveness analysis. The choice of this measure limits the comparison of the results with those from other health interventions.
Validity of estimate of costs The costing analysis was carried out from the hospital perspective. The analysis appears to have focused on the immediate direct costs of screening and diagnosing patients. The cost-savings due to false positives and false negatives avoided were not included in the analysis. The unit costs and the quantities were reported separately, which will help the generalisability of the results to other settings. It was unclear whether the costs of overheads were incorporated in the unit cost estimates. Public and private sector tariffs were used. No statistical or sensitivity analyses of the costs were carried out, and this limits the interpretation of the results. Discounting was not relevant and, appropriately, was not carried out.
Other issues The authors drew comparisons with studies that had produced different results in terms of the cost per GDM detected. However, they noted that direct comparisons were difficult, owing to the variety of selection protocols and the wide range of GDM prevalence in the populations studied. Although the authors reported that cost-effectiveness or cost-benefit analyses had been published on the same topic, they did not compare their results with these studies. The issue of generalisability to other populations was addressed in the 'Discussion' section. The conclusions reported were an accurate reflection of the scope of the analysis and the results presented. The authors reported no limitations to their study.
Finally, the authors did not calculate the incremental cost-effectiveness ratios (ICERs). They reported the average cost per GDM detected, but did not compare the incremental cost per additional GDM detected. The correct approach would be to compare each strategy with the next most effective strategy, and then to identify that the "universal 140" strategy was dominated by the "universal 130" strategy only on the basis of the ICER (extended dominance). The decision could not be based on average costs. Care should be taken when using the results of this type of limited analysis.
Implications of the study The authors recommended a universal screening for populations in whom there is a significant prevalence of GDM.
Bibliographic details Larijani B, Hossein-Nezhad A, Vassigh A R. Effect of varying threshold and selective versus universal strategies on the cost in gestational diabetes mellitus. Archives of Iranian Medicine 2004; 7(4): 267-271 Other publications of related interest Everett WD. Screening for gestational diabetes: an analysis of health benefits and costs. American Journal of Preventive Medicine 1989;5:38-43.
Kitzmiller JL, Elixhauser A, Carr S, et al. Assessment of costs and benefits of management of gestational diabetes mellitus. Diabetes Care 1998;21 Suppl 2:B123-37.
Langer O, Rodriguez DA, Xenakis EM, McFarland MB, Berkus MD, Arredondo F. Intensified versus conventional management of gestational diabetes. American Journal of Obstetrics and Gynecology 1994;170:1036-46.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Blood Glucose /analysis; Costs and Cost Analysis; Diabetes, Gestational /diagnosis /epidemiology; Female; Gestational Age; Glucose Tolerance Test; Humans; Iran /epidemiology; Mass Screening /economics /methods; Pregnancy; Pregnancy, High-Risk; Risk Factors; Sensitivity and Specificity AccessionNumber 22005000464 Date bibliographic record published 31/12/2005 Date abstract record published 31/12/2005 |
|
|
|