Combined Use of Qualitative Method and Metanalysis on the Diagnostic Approhach of Infectious Diarrhea

Abstract

Objective:To compare the diagnostic usefulness of index tests in the identification of invasive diarrhea. Methods: Combination of qualitative and metanalytic methods. Data identification and study selection: Of 2603 MEDLINE references (1970-1994) and correspondence with tield experts, 81 fulfilled preestablished selection criteria and 25 of them had sufficient data for additional metaanalysis. Extraction of data: During the qualitative review, one author extracted extended information on the diagnostic value of index tests and on methodologic characteristics of primary studies. During the metanalytical phase, all data from the selected articles were extracted by one observer whereas the second reviewer checked them for accuracy. True positive (TPR) and false positive rates (FPR) were calculated with the 2 x 2 contingency table. To compare the diagnostic usefulness of index tests, a summary receiver operating characteristic curve (SROC) was fitted to the data with the use of logistic transformation and weighted minimal squares linear regression. Results of data analysis:The non restrictive qualitative approach identified methodologic problems that were additionally assessed with the quantitative approach. ln metanalysis, plots of TPR against FPR demonstrated widely scattered points, indicating heterogeneity. Fecal lactoferrill SROC curve showed the highest performance and fecal leukocytes showed the lowest performance. Conclusions: Fecal lactoferrin appears to be the most useful test ¡n the identitication of invasive diarrhea. The use of a combined qualitative and metanalytic method is proposed as an adequate approach to the assessment of diagnostic tests, ¡n both clinical practice and medical decision analysis.