Abridging the Human Activity Profile (HAP): Item Reduction for Ambulatory Outpatient Populations

ABSTRACT Introduction The purpose of this study was to develop an abridged Human Activity Profile (HAP-A) for ambulatory patients. Methods Classical test theory item reduction model and exploratory factor analysis (EFA) were used, including maximum likelihood factor extraction with Kaiser varimax rotation, eigenvalues >1 retained, and item loading cutoff of 0.4, followed by confirmatory factor analysis (CFA) and internal consistency reliability analyses. Data were originally collected from an outpatient ambulatory tertiary care clinic and research site of a large nonprofit health system hospital medical campus located in a metropolitan area of the northeastern United States. Four hundred and fifty-five de-identified healthy controls and people with chronic liver diseases or autoimmune or metabolic conditions with basic demographic information and completed HAP were collected across one retrospective and four prospective studies with institutional review board approval over 15 years (2006–2021). Main analyses included maximum and adjusted activity scores (MAS and AAS), EFA and CFA, internal consistency reliability (Cronbach’s α , McDonald’s ω ), and convergent validity. Results HAP and HAP-A MAS and AAS measures were statistically indistinguishable ( P = 1.00, paired t -tests). HAP-A Cronbach’s α was 0.892, and McDonald’s ω was 0.902. CFA revealed three factors (domains) in 29 questions: factor 1, high activity/sport (5.7–10.3 metabolic equivalents (METs)); factor 2, light mobility/leisure (0.9–7.1 METs); and factor 3, chores/activities of daily living (<0.9–6.6 METs). Using CFA, 28 of 29 items loaded as expected; Tucker–Lewis Index, comparative fit index, and root mean square error of approximation were modest (0.716, 0.738, and 0.110, respectively), likely due to cohort composition shifts. Controlling for age, sex, body mass index, hypertension, hyperlipidemia, and diabetes, HAP-A’s AAS score retained its significant correlation with the Fatigue Severity Scale (FSS) (analysis of covariance sum of squares, 6.097; 1 degree of freedom; mean square, 6.097; P = 0.03). Conclusions HAP and HAP-A scores were statistically indistinguishable and preserved a significant correlative relationship with a validated fatigue measure (FSS). HAP-A is a reasonable HAP alternative in ambulatory patients.