Main Article Content
Latent class is a method for classifying subjects, originally based on binary outcome data but now extended to other data types. A major difficulty with the use of latent class models is the presence of heterogeneity of the outcome probabilities within the true classes, which violates the assumption of conditional independence, and will require a large number of classes to model the association in the data resulting in difficulties in interpretation. A solution is to include a normally distributed subject level random effect in the model so that the outcomes are now conditionally independent given both the class and random effect. A further extension is to incorporate an additional period level random effect when subjects are observed over time. The use of the randomLCA R package is demonstrated on three latent class examples: classification of subjects based on myocardial infarction symptoms, a diagnostic testing approach to comparing dentists in the diagnosis of dental caries and classification of infants based on respiratory and allergy symptoms over time.