TY - JOUR
T1 - The latent structure of psychiatric symptoms across mental disorders as measured with the PANSS and BPRS-18
AU - Van Dorn, Richard A.
AU - Desmarais, Sarah L.
AU - Grimm, Kevin
AU - Tueller, Stephen J.
AU - Johnson, Kiersten L.
AU - Sellers, Brian G.
AU - Swartz, Marvin S.
N1 - Publisher Copyright:
© 2016
PY - 2016/11/30
Y1 - 2016/11/30
N2 - Raw data were used from five studies of adults with mental illnesses (N=4,480) in an attempt to identify a psychiatric symptoms factor structure, as measured by the Positive and Negative Syndrome Scale or the Brief Psychiatric Rating Scale, that was generalizable across participant characteristics. First, the fit of four extant models was tested via confirmatory factor analysis (CFA), then exploratory factor analyses (EFA) were conducted with a 50% random sample, followed by a CFA with the remaining 50% to confirm the EFA factor structure. Measurement invariance of the factor structure was also examined across diagnosis, sex, race, age, and hospitalization status. The extant models were not generalizable to these data. However, a 4-factor (Affective, Positive, Negative, Disorganized Cognitive Processing) model was identified that retained all items and showed invariance across participant characteristics. It is possible to obtain a psychiatric symptoms factor structure that is generalizable across patient characteristics, which has clinical and research implications. Specifically, future research examining the impact of various interventions on psychiatric symptoms among adults with mental illnesses should confirm, and assuming good model-data fit, use the 4-factor model identified in this study.
AB - Raw data were used from five studies of adults with mental illnesses (N=4,480) in an attempt to identify a psychiatric symptoms factor structure, as measured by the Positive and Negative Syndrome Scale or the Brief Psychiatric Rating Scale, that was generalizable across participant characteristics. First, the fit of four extant models was tested via confirmatory factor analysis (CFA), then exploratory factor analyses (EFA) were conducted with a 50% random sample, followed by a CFA with the remaining 50% to confirm the EFA factor structure. Measurement invariance of the factor structure was also examined across diagnosis, sex, race, age, and hospitalization status. The extant models were not generalizable to these data. However, a 4-factor (Affective, Positive, Negative, Disorganized Cognitive Processing) model was identified that retained all items and showed invariance across participant characteristics. It is possible to obtain a psychiatric symptoms factor structure that is generalizable across patient characteristics, which has clinical and research implications. Specifically, future research examining the impact of various interventions on psychiatric symptoms among adults with mental illnesses should confirm, and assuming good model-data fit, use the 4-factor model identified in this study.
KW - Between-diagnoses
KW - Factor structure
KW - Invariance
KW - Mental illness
UR - http://www.scopus.com/inward/record.url?scp=84981526574&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84981526574&partnerID=8YFLogxK
U2 - 10.1016/j.psychres.2016.08.029
DO - 10.1016/j.psychres.2016.08.029
M3 - Article
C2 - 27529666
AN - SCOPUS:84981526574
SN - 0165-1781
VL - 245
SP - 83
EP - 90
JO - Psychiatry Research
JF - Psychiatry Research
ER -