Study Measures
The Maslach Burnout Inventory (MBI) Human Services Survey for Medical Personnel is a 22-item self-reported survey instrument developed to assess the 3 dimensions of burnout using a multidimensional model.1 Whereas previous measures of burnout focused only on the exhaustion dimension,16 the MBI incorporates the central construct of exhaustion and extends measurement to depersonalization and reduced personal accomplishment constructs to articulate interrelationships.17 The MBI is widely used in burnout research and considered the standard tool for measurement.18 MBI items are scored by a 7-point frequency scale from “never” to “daily.”
The MBI was not developed as a clinical diagnostic tool, and therefore does not distinguish between burned-out and not burned-out individuals. However, studies have attempted to validate MBI cutoff scores that reflect both a binary distinction of burnout, and also low, moderate, and high levels of emotional exhaustion, depersonalization, and reduced personal accomplishment. The MBI Manual (4th edition; 2016) removed cut-off scores due to lack of diagnostic validity.17This study reported burnout levels only to compare with previous studies administering the MBI to participants within oncology settings.4,5 Therefore, high level burnout was defined when respondents had 1 of 3 criteria: raw sum score of ≥27 in the emotional exhaustion dimension, ≥10 in the depersonalization dimension, or ≥15 in the reduced personal accomplishment dimension.11,12,19,20
Workplace factors were measured with the National Aeronautics and Space Administration Task Load Index (NASA-TLX),21 which measures 5 aspects of subjective job demand: mental demand, physical demand, temporal demand, effort, and frustration. NASA-TLX scores ranged from 0 to 10; higher scores reflected higher perceived workload. The NASA-TLX has been previously used in healthcare research.14,22
A screener question from the Second Victim Experience and Support Tool (SVEST) was used to assess PSE involvement. The SVEST is a 29-item survey instrument designed to provide healthcare organizations information on PSE involvement support resources most preferred by clinicians.23 SVEST responses are recorded with 5-point Likert-scale agreement items ranging from “strongly disagree” to “strongly agree”.
Participants could also include open-response comments via the item, “Please use this space to include any comments about how we can improve patient safety or the work experience for the St. Jude patient care staff.” These comments were analyzed post-hoc, in response to results demonstrating associations among the NASA-TLX job demand, frustration, and burnout prevalence. Two authors (TD & JDB) reviewed comments independently to identify broad concepts related to potential sources of frustration. Final concept terminology was developed by consensus, and content analysis methods identified relevant comments. Content analysis disagreements were resolved through consensus.