Document Type


Publication Date



Maine Medical Center, Medical Education, Maine Medical Center Research Institute, Emergency Medicine

MeSH Headings

Adult, Humans, Psychometrics, Patient Participation, Emergency Service, Hospital, Surveys and Questionnaires


Background: Patient engagement with healthcare is important as engaged patients make more informed healthcare decisions, incur lower expenses, and have improved outcomes. Patient activation, defined as knowledge, skills, and confidence in managing health and healthcare, is closely linked and is measurable using the short form Patient Activation Measure (PAM-13). While evidence supporting the PAM-13s validity has been established in the general and hospitalized populations, its psychometric properties have not been evaluated when used in the emergency setting (ED). Thus, we sought to assess the PAM-13s psychometric properties when used in the ED setting.

Methods: We administered the PAM-13 to a convenience sample of adult patients receiving care in our Level-I tertiary ED. Patients were excluded if they could not read English or if their medical or cognitive condition precluded participation. Analyses included: a) internal consistency reliability using Cronbach’s alpha coefficient and item analysis, b) principle components analysis (PCA) with Varimax rotation and Kaiser normalization and c) internal consistency reliability of the components identified through PCA.

Results: 200 patients participated in the study, 98 (49%) were female, median age 64 (range 21-95). PAM-13 scores ranged from 39.7 to 100, mean 67.2 (SD 13.25, 95% CI: 65.3-69.0). The initial standardized Cronbach’s alpha coefficient for all 13 items was 0.88, indicating high internal consistency reliability. One item displayed a low corrected item-to-total correlation (0.247) but was retained due to limited influence on alpha if deleted. The items were subjected to PCA yielding a two-component solution explaining 51.9% of observed variance. Cronbach’s alpha coefficient for the individual components were 0.87 (Component 1, 7 items) and 0.73 (Component 2, 6 items), indicating that the subscales may be used independently to assess unique aspects of patient activation.

Conclusions: Results suggest that the PAM-13 is psychometrically sound when used with adult ED patients. While the PAM-13 holds promise as a tool for evaluating activation in the ED, additional research in needed to assess the relationships between patient characteristics and scores on the PAM-13. In addition, evaluation using methods grounded in Item Response Theory (IRT) are warranted as the PAM-13 was developed using an IRT-model.


2020 Costas T. Lambrew Research Retreat