Applying GRADE for diagnosis 93 2 Appendix 3. GRADE Evidence Profiles Diagnostic accuracy sIgE (D.pteronyssinus) Question: Should sIgE (D.pteronyssinus) be used to diagnose allergic rhinitis in patients highly suspected of having allergic rhinitis? Bibliography: Garcia Robaina, 2003; Haxel, 2016; King, 2008 [1-3] Sensitivity: 0.84 to 1.00 Specificity: 0.54-1.00 Prevalences 30%* Outcome № of studies (№ of patients) Study design Factors that may decrease certainty of evidence Effect per 1.000 patients tested Test accuracy Certainty of evidence Risk of bias Indirectness Inconsistency Imprecision Publication bias pre-test probability of 30% True positives (patients with allergic rhinitis) 3 studies 189 patients crosssectional (cohort type accuracy study) seriousa seriousb not serious seriousc none 252 to 300 ⨁◯◯◯ VERY LOW False negatives (patients incorrectly classified as not having allergic rhinitis) 0 to 48 True negatives (patients without allergic rhinitis) 3 studies 189 patients crosssectional (cohort type accuracy study) seriousa seriousb not serious seriousd none 378 to 700 ⨁◯◯◯ VERY LOW False positives (patients incorrectly classified as having allergic rhinitis) 0 to 322 * This prevalence was chosen, based on the Dutch clinical guideline on allergic and non-allergic rhinitis for general practitioners [4], and was confirmed in the study of King et al [3] a. By far the largest study (Haxel 2016) has high risk of selection bias (QUADAS domain patient selection). One study (Garcia Robaina, 2003) has high risk of bias in flow & timing, one study (King, 2008) has high risk of bias in interpreting results of the reference test. b. Two studies have been performed in tertiary (university) care, one study in secondary care. This review focusses on primary care. Diagnostic accuracy might vary between primary, secondary and tertiary care, because of variation in pre-test probabilities. c. Only 108 patients with TP or FN results d. Only 81 patients with TN or FP results
RkJQdWJsaXNoZXIy MTk4NDMw