|
A comparison of screening strategies for elevated blood lead levels |
Campbell J R, Paris M, Schaffer S J |
|
|
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 Screening strategies for the detection of elevated blood lead levels.
Economic study type Cost-effectiveness analysis.
Study population A hypothetical cohort of asymptomatic children aged 6 or younger.
Setting Hospital. The economic study was carried out in New York, USA.
Dates to which data relate The main effectiveness data were taken from published data of previously completed studies, and unpublished data dated 1992-96. The price date used to report the results was not stated.
Source of effectiveness data Effectiveness data were derived from reviews of previously completed studies and opinion.
Modelling A decision analysis model in SMLTREE was used to combine estimates of the costs per child screened for 3 universal and 2 selective screening strategies. The model included the cost of prompting a return visit and for confirmatory venipuncture lead assay and phlebotomy for children with a fingerstick blood lead level of >/= 0.72 micromol/L (15 microg/dL).
Outcomes assessed in the review The main outcomes were the prevalence ofblood lead levels, and sensitivity and specificity of diagnostic tests.
Study designs and other criteria for inclusion in the review The review included previously completed published and unpublished studies. The inclusion/exclusion criteria were not stated.
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 Overall, 19 studies were included, of which 10 were used to derive estimates of prevalence of blood lead levels. Five studies were used for the sensitivity of fingerstick and four for the specificity of fingerstick. 5 studies were used for the specificity of the risk assessment at blood lead levels <0.48 micromol/L (<10 microg/dL). Unpublished data were used for the sensitivity of the risk assessment at blood lead levels 0.48-0.68 micromol/L (10-14 microg/dL), and 3 studies were used for the risk assessment's sensitivity at levels greater than or equal to 0.72 micromol/L (15 microg/dL).
Methods of combining primary studies Investigation of differences between primary studies Not stated, even though it would have been applicable.
Results of the review The estimated prevalence of blood lead levelswas 1.5% for 0.48-0.68 micromol/L and 0.5% for >0.72 micromol/L. The sensitivity and specificity of fingerstick and the specificity of risk assessment at blood lead levels <0.48 micromol/L were 90%, 90% and 50%, respectively. The estimated sensitivity of risk assessmentat blood lead levels 0.48-0.60 was 65% and >0.72 micromol/Lwas 85%.
Methods used to derive estimates of effectiveness The authors assumed the diagnostic value of venipuncture tests, and the proportions of falsely elevated fingerstick screening tests (as compared to the gold standard venipuncture blood lead levels) falling into the categories of blood lead levels of 0.48-0.68 and >/= 0.72 micromol/L.
Estimates of effectiveness and key assumptions The authors assumed that phlebotomists were "entirely successful" at obtaining venipuncture specimens, without any contamination occurring at phlebotomy or during the assay, so that lead assays of venipuncture specimens reflect true blood lead levels (gold standard or reference test). It was further assumed by the authors that the 90% and 10% of children without lead toxicity who had a falsely elevated fingerstick screening test had fingerstick blood lead levels of 0.48-0.68 and >/= 0.72 micromol/L, respectively.
Measure of benefits used in the economic analysis Since the model assumed that all children who required a confirmatory venipuncture would actually obtain one, the strategies were assumed to yield similar benefits and, therefore, the economic study was based on the differences in costs only.
Direct costs The main quantities of resource use (number of individual tests) were implicitly stated. The costs associated with the screening strategies (including the cost of prompting a return visit for children with a fingerstick blood lead level >/= 0.72 micromol/L and confirmatory venipuncture) were includedAlthough no cost discounting was documented it seems that the time period covered by the analysis was far shorter than 1 year. The cost boundary adopted was that of the clinic. The date of the price data was not stated. The cost associated with the risk assessment (performed by the parent) was not included as it was considered to require a short period of time to complete, whilst the cost of medical visits was also excluded given local practice relating to well-child care visits and in view of the fact that a confirmatory venipuncture did not require a medical visit.
Sensitivity analysis A one-way sensitivity analysis was carried out on prevalence of elevated blood lead levels, risk assessment sensitivity and specificity, fingerstick blood lead assay sensitivity and specificity, blood leads assay cost and cost of follow-up for a child who has a fingerstick blood lead level of 0.72 micromol/L or higher.
Estimated benefits used in the economic analysis Cost results The cost per child screened (assuming a prevalence rate of disease of 2%) of strategies 1 through 5 was respectively, $9.07, $8.16, $10, $4.13 and $5.04. Strategy 2 had the lowest cost among the universal strategies (1-3) and strategy 4 had the lowest cost among the selective strategies (4-5).
Synthesis of costs and benefits Synthesis was not applicable. The sensitivity analysis showed that among the universal strategies, screening children by fingerstick had the lowest cost at a prevalence of less than 38% and a fingerstick blood lead assay specificity of greater than 62%. Among the selective strategies, screening high-risk children by fingerstick had the lowest cost at a prevalence of less than 38% and a fingerstick blood lead assay specificity of greater than 63%.
Authors' conclusions At a readily attainable specificity of the fingerstick blood lead assay, practices serving a patient population with a prevalence of elevated blood lead levels of less than 38% will have the lowest cost when a fingerstick screening strategy is used.
CRD COMMENTARY - Selection of comparators The reason for the choice of comparators is clear.
Validity of estimate of measure of benefit The estimate of measure of benefit used in the economic analysis may be questionable due to the uncertain scientific basis for the assumption of the diagnostic value of lead assays of venipuncture specimens.
Validity of estimate of costs Adequate details of methods of quantity/cost estimation were given. Although no important cost items appear to have been not omitted, the validity of the estimate of costs rests on the validity of the underlying effectiveness study results and the degree of scientific rigour of the literature search and review which served as the basis for the decision model used in the economic study. The latter cannot be assessed given the limited information provided by the authors.
Other issues The authors' conclusions were justified in terms of the sensitivity analysis, given the uncertainties in the data. The issue of generalisability to other settings was not addressed. Appropriate comparisons were made with other studies.
Implications of the study Further analysis of the data and studies may be necessary to validate the results of the study reviewed above.
Bibliographic details Campbell J R, Paris M, Schaffer S J. A comparison of screening strategies for elevated blood lead levels. Archives of Pediatrics and Adolescent Medicine 1996; 150(11): 1205-1208 Other publications of related interest Comment in: Archives ofPediatric and Adolescent Medicine 1997;151(6):636.
Indexing Status Subject indexing assigned by NLM MeSH Costs and Cost Analysis; Decision Support Techniques; Humans; Lead /blood; Phlebotomy /economics; Risk Assessment; Sensitivity and Specificity AccessionNumber 21996001026 Date bibliographic record published 30/06/1999 Date abstract record published 30/06/1999 |
|
|
|