Statistical approaches to explore clinical heterogeneity in psychosis
PhD ceremony: | Mr A. (Atiqul) Islam |
When: | July 05, 2017 |
Start: | 09:00 |
Supervisor: | prof. dr. E.R. van den Heuvel |
Co-supervisors: | R. (Richard) Bruggeman, dr. B.Z. (Behrooz) Alizadeh |
Where: | Academy building RUG |
Faculty: | Medical Sciences / UMCG |
Psychotic disorders display a very heterogeneous presentation which is often overlooked. The main aim of this thesis is to explore the heterogeneity, stability and familial liability in psychotic patients and their unaffected siblings. Classical clustering, linear, generalized linear mixed effects and group-based trajectory modeling (GBTM) techniques were applied to dissect heterogeneity and stability.On heterogeneity, we confirmed previous work on cognitive subtypes within the GROUP-project, by demonstrating that Duda and Hart was indeed the best performing index to identify homogenous clusters. Heterogeneity of cognition and negative symptoms was validated in the course of the disease by using GBTM, yielding clinical relevant subtypes. Stability of these subtypes turned out to be a key feature for cognition, but not for negative symptoms. Familial liability was reflected in cognition (siblings performing between healthy controls and patients), in somatic diseases (the risk for siblings being in between likewise) and in psychotic experiences (to be more prevalent in siblings than in controls).Within the framework of this thesis, various predictive factors were identified. The cognitive profiles of patients predicted siblings’ cognitive performance. Also, good Theory of Mind in siblings predicted milder psychotic experiences, 3 years later. For negative symptoms, subtypes were strong predictors of outcomes over time. For multimorbidity, familial liability is the major determinants. Additionally, the highest risk for long duration on being untreated was migration.In conclusion, heterogeneity in psychosis is a clinical relevant concept. Subtyping patients provide new avenues to better understanding and more effectively treating people with psychosis in a personalized manner.