Premium
Development of a Measure of Patient Safety Event Learning Responses
Author(s) -
Ginsburg Liane R.,
Chuang YouTa,
Norton Peter G.,
Berta Whitney,
Tregunno Deborah,
Ng Peggy,
Richardson Julia
Publication year - 2009
Publication title -
health services research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.706
H-Index - 121
eISSN - 1475-6773
pISSN - 0017-9124
DOI - 10.1111/j.1475-6773.2009.01021.x
Subject(s) - exploratory factor analysis , patient safety , event (particle physics) , organizational learning , psychology , medicine , knowledge management , applied psychology , computer science , machine learning , structural equation modeling , health care , physics , quantum mechanics , economics , economic growth
Objective. To define patient safety event (PSE) learning response and to provide preliminary validation of a measure of PSE learning response. Data Sources. Ten focus groups with front‐line staff and managers, an expert panel, and cross‐sectional survey data from patient safety officers in 54 general acute hospitals. Study Design. A mixed methods study to define a measure of learning responses to patient safety failures that is rooted in theory, expert knowledge, and organizational practice realities. Extraction Methods. Learning response items developed from the literature were modified and validated in front‐line staff and manager focus groups and by an expert panel and second group of external experts. Actual learning responses gleaned from survey data were examined using exploratory factor analyses and reliability analysis. Principal Findings. Unique learning response items were identified for minor, moderate, major events, and major near misses by an expert panel. A two‐factor model of major event learning response was identified (factor 1=event analysis, factor 2=dissemination/communication of learnings). Organizations engage in greater learning responses following major events than less severe events and, for major events, organizations engage in more factor 1 responses than factor 2 learning responses. Conclusions. Eleven to 13 items can measure learning responses to PSEs of differing severity. The items are feasible, grounded in theory, and reflect expert opinion as well as practice setting realities. The items have the potential for use to assess current practice in organizations and set future improvement goals.