The 12-item ProQOL holds together better than the 30-item original
- 639 helping professionals in Slovakia were split at random into three groups: 300 completed the 30-item ProQOL 5, 169 the 12-item Brief ProQOL-12 and 170 the 9-item Short ProQOL.
- The three-factor model fitted the full ProQOL 5 poorly (CFI = 0.704, RMSEA = 0.097) and fitted both short forms well: Brief ProQOL-12 CFI = 0.957, RMSEA = 0.068; Short ProQOL CFI = 0.976, RMSEA = 0.061.
- The burnout subscale was the weakest part of ProQOL 5 (α = 0.731, ω = 0.661, AVE = 0.296); in Brief ProQOL-12 the same subscale reached α = 0.841, ω = 0.845, AVE = 0.572.
- The authors consider the 12-item form the better balance of brevity and construct coverage, with one more item per subscale than the 9-item form. The paper sets no cut-off scores for any version.
The ProQOL is the questionnaire most often used to measure compassion fatigue in people who help others for a living. Its structure has been questioned for years. A team from Comenius University in Bratislava and the University of Groningen gave three versions to one sample and compared how each held together.
One sample, three versions
Psychologists made up 16% of the sample, nurses 14%, social workers 11%, teachers 9%, paramedics 9%, police officers and clergy 6% each; 78% were women, mean age 38. Every version scores burnout, secondary traumatic stress and compassion satisfaction. ProQOL 5 asks about the last 30 days. Brief ProQOL-12 asks about the last 7 days and has respondents answer in days, from "never (0 days)" to "always (6–7 days)".
The full version failed the factor test. Its burnout subscale explained under a third of item variance, against 0.572 in the 12-item form. The authors trace part of the problem to items that say nothing about work: "I am happy" is scored as a burnout item. No version comes with cut-off scores, so a total cannot be read as burned out or not burned out.
Twelve items, seven days
I would place the Brief ProQOL-12 at the start of one supervision session a month. Twelve items take a few minutes, and the seven-day window ties the score to the week just worked. Without norms, the reference point is the clinician's own previous scores. A burnout line rising across three months, or secondary traumatic stress climbing after a run of trauma referrals, gives the supervisor a reason to look at the caseload while there is still time to change it.
The study did not test retest reliability or sensitivity to change, so a one-point shift between two months carries little weight. Read the trend over a quarter, and discuss it with the clinician who filled in the form.
Without cut-off scores, the 12-item ProQOL is read against the clinician's own previous months, which is what a monthly supervision check needs.
The design was cross-sectional, so retest reliability was not tested. The survey was distributed online and may have drawn self-selected respondents, and the versions used different time frames. The authors limit the findings to Slovak helping professionals.