LATENT VARIABLE MODELING OF DIAGNOSTIC-ACCURACY

Latent class analysis has been applied in medical research to assessin g the sensitivity and specificity of diagnostic tests/diagnosticians. In these applications, a dichotomous latent variable corresponding to the unobserved true disease status of the patients is assumed. Associa tions among multiple diagnostic tests are attributed to the unobserved heterogeneity induced by the latent variable, and inferences for the sensitivities and specificities of the diagnostic tests are made possi ble even though the true disease status is unknown. However, a shortco ming of this approach to analyses of diagnostic tests is that the stan dard assumption of conditional independence among the diagnostic tests given a latent class is contraindicated by the data-in some applicati ons. In the present paper, models incorporating dependence among the d iagnostic tests given a latent class are proposed. The models are para meterized so that the sensitivities and specificities of the diagnosti c tests are simple functions of model parameters, and the usual latent class model obtains as a special case. Marginal models are used to ac count for the dependencies within each latent class. An accelerated EM gradient algorithm is demonstrated to obtain maximum likelihood estim ates of the parameters of interest, as well as estimates of the precis ion of the estimates.