Vol.:(0123456789) 1 3 Qual Life Res (2017) 26:1777–1784 DOI 10.1007/s11136-017-1516-6 Demonstration of two types of fatigue in subjects with chronic liver disease using factor analysis Ali A. Weinstein 1,2,3  · Guoqing Diao 4  · Heibatollah Baghi 3  · Carey Escheik 1  · Lynn H. Gerber 1,2,5  · Zobair M. Younossi 1,5   Accepted: 1 February 2017 / Published online: 21 February 2017 © Springer International Publishing Switzerland 2017 Results 106 participants with CLD were included (50% female; age: 51 ± 10). Two factors were identiied. The factors included one that clustered around questions addressing fatigue related to physical activity (peripheral fatigue) and the other to the questions addressing general- ized fatigue that did not require physical tasks to produce the fatigue (central fatigue). The standardized factor load- ings of all items were greater than 0.6 on their underlying constructs. Moreover, all factor loadings are signiicant at p < 0.01. Peripheral fatigue was related to HAP (r = 0.26, r = 0.24, p < 0.01), as was central fatigue (r = −0.34, r = −0.33, p < 0.01). Central fatigue was related to MCS and CLDQ-EF (r = −0.60; r = −0.63, p < 0.01), whereas peripheral fatigue was not (r = 0.07, p > 0.40). We then tested the original scales to determine if the newly cre- ated factors correlated better with the validity measures. The full FSS did not correlate as well as the newly created central fatigue scale, while the original peripheral fatigue scale (the SF-36v2 physical functioning) was more related to HAP than the newly created scale. Conclusions In individuals with CLD, two separate fac- tors pertaining to fatigue were identiied. This recognition of the multifaceted nature of fatigue may help increase the speciicity of self-reports of fatigue and lead to treatments that can speciically address the underlying factors contrib- uting to fatigue. Keywords Chronic illness · Quantitative methods · Physical activity · Mental health Background and signiicance Fatigue is a common symptom whose origin is poorly understood. One potential reason is that fatigue is an Abstract Purpose The purpose of this investigation was to deter- mine if it was possible to separate fatigue self-reports into two distinct types of fatigue symptom clusters in research subjects with chronic liver disease (CLD). It was hypoth- esized that when items from the Medical Outcomes Study Short-Form (SF-36v2) are combined with items from the Fatigue Severity Scale (FSS), these distinct factors will emerge. Methods Conirmatory and exploratory factor analyses from data collected in a prospective, natural history study of CLD patients were conducted. Items were selected from the SF-36v2 and the FSS for entry into the factor analyses. In order to establish convergent and discriminant validity, derived factor scores were correlated with subscale scores of the Human Activity Proile (HAP), Mental Component Score (MCS) from the SF-36v2, and the Emotional Func- tioning Subscale of the Chronic Liver Disease Question- naire (CLDQ-EF). * Ali A. Weinstein aweinst2@gmu.edu 1 Betty and Guy Beatty Center for Integrated Research, Inova Health System, 3300 Gallows Road, Falls Church, VA 22042, USA 2 Center for the Study of Chronic Illness and Disability, College of Health and Human Services, George Mason University, Fairfax, VA, USA 3 Department of Global and Community Health, College of Health and Human Services, George Mason University, Fairfax, VA, USA 4 Department of Statistics, Volgenau School of Engineering, George Mason University, Fairfax, VA, USA 5 Center for Liver Diseases, Department of Medicine, Inova Fairfax Hospital, Falls Church, VA, USA