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.