Premium
Latent class analysis of persistent disturbing behaviour patients by using longitudinal profiles
Author(s) -
Bruckers Liesbeth,
Serroyen Jan,
Molenberghs Geert,
Slaets Herman,
Goeyvaerts Willem
Publication year - 2010
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/j.1467-9876.2009.00704.x
Subject(s) - latent class model , scope (computer science) , longitudinal data , cluster (spacecraft) , psychology , linear discriminant analysis , longitudinal study , class (philosophy) , variety (cybernetics) , psychiatry , medicine , computer science , artificial intelligence , data mining , machine learning , pathology , programming language
Summary. Persistent disturbing behaviour refers to a chronic condition in highly unstable, therapy resistant psychiatric patients. Because these patients are difficult to maintain in their natural living environment and even in hospital wards, purposely designed residential psychiatric facilities need to be established. Therefore, it is important to define and circumscribe the group carefully. Serroyen and co‐workers, starting from the longitudinal analysis of a score based on data from the Belgian national psychiatric registry, undertook a discriminant analysis to distinguish persistent disturbing behaviour patients from a control group. They also indicated that there is scope for further subdividing the persistent disturbing behaviour patients into two subgroups, using conventional cluster analysis techniques. We employ a variety of novel longitudinal‐data‐based cluster analysis techniques. These are based on either conventional growth models, growth–mixture models or latent class growth models. Unlike in earlier analyses, where some evidence for two groups was found, there now is an indication of three groups, which is a finding with high practical and organizational relevance.