Background: Systematic reviews of diagnostic test accuracy commonly report heterogeneity in sensitivity and specificity across studies. Such heterogeneity can be partially explained by methodological heterogeneity (study characteristics, selection, verification and test interpretation) and clinical heterogeneity (variation in study populations, tests and reference tests). If a test’s accuracy (sensitivity and specificity) is associated with particular patient characteristics, patient level information could be used to further reduce unexplained heterogeneity in test performance. Using individual patient data from studies included such meta-analyses, patient characteristics can be used to statistically explain heterogeneity in sensitivity and specificity. Patient characteristics associated with disease prevalence could be used to differentiate patients in terms of prior probability of disease (predictive value). Patient level factors could also be associated with both accuracy and prevalence, yet this need not be the case. Objectives: To illustrate the relation between heterogeneity in diagnostic accuracy, prevalence and patient characteristics using individual patient data from a meta-analysis (IPDMA) of diagnostic test accuracy. Methods: Data available from two IPDMA of diagnostic tests (Endometrial thickness measured with transvaginal ultrasonography (TVS) in women presenting with postmenopausal bleeding for detecting endometrial carcinoma and hysterosalpingography (HSG) in subfertile women for diagnosing tubal pathology (TP)) were used. Diagnostic accuracy (sensitivity and specificity) and predictive values of the test are differentiated for several patient characteristics. Descriptive analyses were used to explore associations; multivariable non-linear mixed modeling analyses are used to account for the stratified data structure. Results: Patient characteristics age, obesity, DM and parity were associated with the predictive value of TVS for endometrial carcinoma, but only obesity was associated with a reduced specificity. The diagnostic accuracy of HSG for TP varied with duration of subfertility and PID, but only duration affected sensitivity and specificity. Conclusions: Implications from an evaluation perspective, a clinical perspective and assumptions regarding the (in)dependence of test accuracy and prevalence, will be discussed.