Treatment Response in Psychotic Patients in a Naturalistic Setting : Classification, Genes, Drugs, Insight and Social Networks

Many patients with psychotic symptoms respond poorly to treatment. Various approaches have been made to classify these patients according to treatment response. However, existing classifications have been criticized for various reasons and a new classification system is needed. Further, no satisfactory explanation of the poor treatment response has been apparent. The general aim of this thesis was therefore to develop and validate a new classification method of functional remission in a naturalistic population of patients with psychosis and to utilize this classification to investigate the population from genetic, drug treatment, insight and social network points of view. Data for this cross-sectional study of patients (n=123) attending the Psychosis Outpatient Care clinic in the county of Jönköping, Sweden, were obtained from patient interviews, blood samples and information from patient files. The new classification method CANSEPT, which combines the CAN rating scale (CAN), the UKU side effect rating scale (SE) and the patient’s previous treatment history (PT), showed validity in discriminating the patients and was accepted well by the patients. CANSEPT was used to group the patients in the other studies in this thesis. The results indicated that the gene polymorphism ABCB1 3435T, was related to worse significant social and clinical needs for patients on olanzapine, while the polymorphism DRD2 Taq1 A1 was related to a greater risk of significant side effects; especially if male, or taking strong dopamine D2-receptor antagonistic drugs. Drug treatment factors were also related to treatment response; longer duration of untreated prodromal and early psychosis was seen for patients with current significant social and clinical needs and non-adherence to treatment was associated with worse significant side effects. Worse treatment outcomes also appeared to be associated with smaller social network groups, worse insight into illness, poorer knowledge of warning signs and worse coping strategies. In summary, CANSEPT was shown to be a useful valid, multidimensional tool for classification of treatment response. Gene polymorphisms, duration of untreated illness, non-adherence to treatment, social networks and knowledge should be taken into consideration when investigating inadequate treatment response