|
Utilization of common inflammatory markers in new, symptomatic, primary care outpatients based on their cost-effectiveness |
Takemura Y, Ishida H, Inoue Y |
|
|
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 study investigated the relative effectiveness of five common inflammatory markers applied to newly symptomatic patients in a primary care setting. These markers were C-reactive protein (CRP), white blood cell count (WBC), erythrocyte sedimentation rate (ESR), sialic acid and protein fractionation.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised 540 newly symptomatic outpatients (irrespective of specific symptoms or disease classification). There were 250 males and 290 females, with an age range of 9 to 83 years. Patients referred with existing test results or tentative clinical diagnoses were excluded.
Setting The setting appears to have been secondary care, although the authors reported it as primary care (Comprehensive Medicine Clinics, National Defense Medical College Hospital in Tokorozawa and Nishi Ohmiya Hospital in Saitama-City, Japan). The economic study was carried out in Japan.
Dates to which data relate The dates when the data were collected for the costs and effectiveness analysis were not reported.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was undertaken on the same patient sample as that used in the effectiveness study, although it was unclear whether this was done prospectively.
Study sample Sample size calculations were not reported. The sample was selected from newly symptomatic outpatients (regardless of specific symptoms or disease category) seen by three specified physicians at the study setting. Of the 540 patients who entered the study, 177 were allocated a tentative initial diagnosis (TID) relating to an infection or inflammatory-related condition and were subsequently included in the study.
Study design This was a multi-centre, diagnostic cohort study. All of the patients received the tests, which were administered as a package. This package comprised 30 common diagnostic items, of which 5 were related to inflammatory markers. It was unclear whether the order in which the tests were performed would create bias. Physicians were not aware of the test results at the time they made a TID. The length of follow-up was not stated although, owing to the nature of the condition, substantial follow up was unlikely to be relevant.
Analysis of effectiveness All patients received all tests, and a complete set of results was available for each. The primary health outcome in this study was the impact of diagnostic testing on, or its contribution to, changes in the physician's diagnosis or decision-making following the TID. The test was labelled "effective" where these impacts occurred. Impacts (determined following a review of patient's medical records) were awarded a useful result (UR) and these were assigned objectively for each TID, according to criteria outlined in an earlier study (Takemura et al., 1999, see Other Publications of Related Interest). Patients could have several URs arising from effective tests in different disease sub-categories.
Effectiveness results The five inflammation-related test items contributed to 147 URs in 123 patients with infection or inflammatory diagnosis at the TID stage.
Contribution rates to URs for the five items were led by CRP (0.53), followed by sialic acid (0.51), protein fractionation (0.44), WBC (0.43) and ESR (0.41).
The results showed that the tests (alone or together) contributed to the generation of a UR when combined, with the result that WBC often generated different URs from other tests.
Subsequent correlation results revealed that WBC possessed a pathophysiological function (described in terms of time delay to response) that was different to other inflammatory markers.
Clinical conclusions The authors concluded that the most effective single test, measured by its contribution to UR generation, was CRP.
Measure of benefits used in the economic analysis The measure of benefit used was the number of URs gained. This was obtained directly from the effectiveness results.
Direct costs Discounting was not relevant because of the short timeframe of the analysis. The total health service costs included related to test reagents, consumables required for the operation of the analysers, computers and printers, disposables needed for specimen collection and analysis, equipment amortisation, and staff costs relating to medical technologists. The costs for tests measured by the same automated multichannel analyser (CRP plus sialic acid) were derived by adding each test-specific cost to a common baseline value (to reflect the cost of operating the analyser). The baseline cost included equipment amortisation, staffing, and common usage with tests performed simultaneously. The source of the cost data and method of costing was not stated, but it might have been described in other studies cited by the authors (Takemura et al., 1999 and 2002, see Other Publications of Related Interest). The resource quantities and the costs were reported separately, but the price year was not given.
Statistical analysis of costs The cost data were deterministic.
Indirect Costs The indirect costs were not reported.
Sensitivity analysis A sensitivity analysis was not reported.
Estimated benefits used in the economic analysis The incremental benefits used in the economic analysis were derived from the combination of tests beyond the CRP marker. The number of URs generated was:
93 with CRP;
127 with CRP + WBC;
103 with CRP + ESR;
106 with CRP + sialic acid;
106 with CRP + protein fractionation;
135 with CRP + WBC + ESR;
138 with CRP + sialic acid + WBC;
138 with CRP + WBC + protein fractionation; and
147 with (CRP + sialic acid) + WBC + ESR + protein fractionation.
Cost results The total cost for a single test was:
Y59,195 for CRP;
Y39,743 for WBC;
Y17,373 for ESR;
Y67,861 for sialic acid; and
Y31,265 for protein fractionation.
The total costs for combined testing were:
Y98,938 with CRP + WBC;
Y76,568 with CRP + ESR;
Y74,679 with CRP + sialic acid;
Y90,460 with CRP + protein fractionation;
Y116,311 with CRP + WBC + ESR;
Y114,422 with CRP + sialic acid + WBC;
Y130,203 with CRP + WBC + protein fractionation; and
Y163,060 with (CRP + sialic acid) + WBC + ESR + protein fractionation.
Synthesis of costs and benefits When the costs and benefits were synthesised for single tests, the average costs per effective test were:
Y248 with ESR;
Y423 with protein fractionation;
Y530 with WBC;
Y637 with CRP; and
Y848 with sialic acid.
When the costs and benefits were synthesised for combined tests, the results were given as average and incremental cost-effectiveness ratios (CERs):
CRP + WBC, average CER Y779, incremental CER Y1,169;
CRP + ESR, average CER Y743, incremental CER Y1,737;
CRP + sialic acid, average CER Y705, incremental CER Y1,191;
CRP + protein fractionation, average CER Y853, incremental CER Y2,405;
CRP + WBC + ESR, average CER Y862, incremental CER Y1,360;
CRP + sialic acid + WBC, average CER Y829, incremental CER Y1,227;
CRP + WBC + protein fractionation, average CER Y950, incremental CER Y1,578; and
CRP + sialic acid + WBC + ESR + protein fractionation, average CER Y1,109, incremental CER Y1,923.
Authors' conclusions When single tests were performed, the inflammatory marker with the highest contribution to useful result (UR) generation was C-reactive protein (CRP), although its cost was relatively high compared with the others. However, as this test was effective in only half of patients with the target condition, the exploration of combined testing became relevant. In combination testing, CRP plus white blood cell count (WBC) produced the most favourable incremental cost-effectiveness ratio (CER). In addition, as sialic acid and CRP can be measured simultaneously on an automated multichannel analyser, these tests might also be combined (resulting in a decreased total cost), but only in hospital settings where this equipment is available.
CRD COMMENTARY - Selection of comparators Although no explicit justification was provided for the diagnostic tests (or order of tests) chosen for infection and inflammation in this study, it would appear to represent current practice in the authors' setting. The authors referred to a preliminary study, in which test accuracy measures were analysed (Takemura et al., 1996, see Other Publications of Related Interest). You should decide if these represent widely used technologies in your own setting.
Validity of estimate of measure of effectiveness The study referred to patients with infection or inflammatory symptoms, and those patients selected for analysis were diagnosed as such at the TID stage. The study design was associated with several limitations. For example, the uncertainty surrounding the comparability of the patients, and potential confounding effects of different disease and severity status. This potentially inhibits the internal reliability of the study and limits the generalisability of the results. The fact that one physician was involved in both the initial clinical evaluation and in the determination of the URs potentially introduces bias to the results.
Validity of estimate of measure of benefit The measure of benefit used was UR (useful result) generation. The robustness of this measure was unclear given that the criteria for allocation were detailed in another study (Takemura et al., 1999, see Other Publications of Related Interest). The determination of a UR appeared to be subjectively derived from medical records, thus potentially limiting the reliability of the results.
Validity of estimate of costs All the relevant costs relating to the health service perspective appear to have been included in the analysis. The separate reporting of the costs (of single measurements) and the quantities (number of tests performed) will usefully aid the reproducibility of the study in other settings. However, the lack of a sensitivity analysis meant that the robustness of the results could not be determined. The resource quantities were taken from the current study, while the source of the price data and price year were not reported in the present paper. This limits any future reflation exercise. However, these detail might have been reported in two other studies cited by the authors (Takemura et al., 1999 and 2002, see Other Publications of Related Interest).
Other issues The authors did not compare their primary outcomes with those from other studies. A substantial limitation of the study (as the authors acknowledged) related to the limited number of patients with inflammatory markers and the consequent lack of ability to sub-categorise patients by specific inflammatory conditions.
Implications of the study The authors suggested that the inclusion of larger patient populations, thus enabling patient sub-categorisation by specific inflammatory conditions, should be a feature of future studies. They also recommended that combined markers of different pathophysiologic function, and those that yield different types of URs to aid decision, would make a further contribution to research. In practice, the combination of sialic acid and CRP is proposed in settings where automated multichannel analysers are available.
Source of funding Supported by grants from the Clinical Pathology Research Foundation of Japan, and from the Kurozumi Medical Foundation.
Bibliographic details Takemura Y, Ishida H, Inoue Y. Utilization of common inflammatory markers in new, symptomatic, primary care outpatients based on their cost-effectiveness. Clinical Chemistry and Laboratory Medicine 2003; 41(5): 668-674 Other publications of related interest Takemura Y, Ishida H, Inoue Y, Beck JR. Common diagnostic test panels for clinical evaluation of new primary care outpatients in Japan: a cost-effectiveness evaluation. Clinical Chemistry 1999;45:1752-61.
Takemura Y, Ishida H, Inoue Y, Beck JR. Yield and cost of individual common diagnostic tests in new primary care outpatients in Japan. Clinical Chemistry 2002;48:42-54.
Takemura Y, Kobayashi, H, Kugai N, Sekiguchi S. The results of the "Essential Laboratory Tests" applied to new outpatients - re-evaluation of diagnostic efficiencies of the test items. Rinsho Byori. The Japanese Journal of Clinical Pathology 1996;44:555-63.
Indexing Status Subject indexing assigned by NLM MeSH Adolescent; Adult; Aged; Aged, 80 and over; Ambulatory Care /economics /methods; Biomarkers /analysis; Child; Clinical Laboratory Techniques /economics /utilization; Cost-Benefit Analysis; Female; Humans; Inflammation /diagnosis; Male; Middle Aged; Outpatients /classification AccessionNumber 22003009568 Date bibliographic record published 30/06/2005 Date abstract record published 30/06/2005 |
|
|
|