The FABQ, short for Fear-Avoidance Beliefs Questionnaire, measures the fear-avoidance beliefs of the low back pain patient across two subscales, physical activity and work. The two are not worth the same, and only one predicts anything. The interpreter is right below.
The two subscales are not worth the same: the work one predicts return to work and six-month disability, the physical activity one predicted nothing at all. The detail is below.
Figures taken from the studies listed at the end of this page, every value carrying its source where it is written. The questionnaire itself is not reproduced.
What the FABQ measures
The FABQ was published in 1993 by Waddell et al., from a simple idea: between low back pain and the disability it produces sit beliefs, and those beliefs can be measured.
Principal-components analysis, in 210 patients, yields two factors: work-related beliefs, with an internal consistency of 0.88 and 43.7 % of the variance, and physical-activity ones, at 0.77 and 16.5 %. The first weighs nearly three times the second from the outset.
What follows confirms it on outcomes. In 184 patients, work-related beliefs explain 23 % of the variance of disability in daily activities and 26 % of that of work loss, once pain severity is allowed for. Physical-activity ones add only 9 %.
Two subscales, only one that predicts
This is the point thirty years of literature have established, and that practice still often ignores.
The six-month follow-up study, in 160 patients treated in physical therapy, is unambiguous. The usual threshold of the physical activity subscale, above 14, classifies 69.4 % of patients as elevated, and those show no difference in six-month outcome. Its area under the curve is 0.562, with a confidence interval that contains chance.
The work subscale, by contrast, holds. Its threshold of above 29 classifies only 11.9 % of patients, but those do report higher six-month disability and more often no improvement. In the final model it is the only unique predictor coming from the FABQ, alongside manipulation and exercise.
The two thresholds of the work subscale
Two values circulate for the work subscale. They come from the same study and do not serve the same purpose, which explains the confusion.
In 78 patients with acute work-related low back pain, 29 % of whom keep restrictions at four weeks, a score below 30 gives a negative likelihood ratio of 0.08. A score above 34 gives a positive likelihood ratio of 3.33.
The asymmetry between the two is what to remember. A negative likelihood ratio of 0.08 is powerful: below 30, the risk of prolonged restriction is solidly ruled out. A positive ratio of 3.33 is moderate: above 34, the risk rises without being established. The FABQ work subscale reassures better than it warns, and between 30 and 34 it says nothing decisive.
| What is measured | Physical activity | Work |
|---|---|---|
| Score range | 0 to 24 | 0 to 42 |
| Internal consistency, original publication | 0.77 | 0.88 |
| Share of the questionnaire's variance | 16.5 % | 43.7 % |
| Test-retest reliability, French version | 0.72 | 0.88 |
| Threshold in circulation | 14 | 30 and 34 |
| Patients classified elevated by it | 69.4 % | 11.9 % |
| Area under the curve at six months | 0.562 | 0.694 |
| Smallest detectable change | 9.4 | 12.7 |
| That change, against the range | 39 % | 30 % |
Following a FABQ: what the noise allows
A score can be followed over time, provided you know from when it has really moved.
The study of the smallest detectable change in acute low back pain gives 9.4 points for the physical activity subscale, which runs from 0 to 24, and 12.7 points for the work subscale, which runs from 0 to 42. Against their ranges, that is 39 % and 30 %: a patient must cross nearly half the scale before you can be sure they changed.
The authors draw a clear conclusion: the physical activity subscale has considerable problems detecting improvement as well as deterioration. A Rasch analysis in chronic low back pain reaches the same place by another route, with person separation reliability of 0.69 and 0.79, enough to compare groups but not to decide for a patient.
Why this page does not display the questionnaire
No free licence is established for the FABQ: its items are therefore not reproduced here, nor are its response options. The validated French version is the one by Chaory et al., published in 2004 in Spine; that is the one to obtain, and not a translation picked up online, for which nothing guarantees it keeps the properties quoted on this page.
Three administration pitfalls
Reading a high physical activity score as a risk factor
More than two patients in three cross the threshold of 14, and crossing it predicted no difference in six-month outcome. That figure describes a present belief, it does not announce a course.
Adding the two subscales together
They share neither range, nor weight, nor prognostic value. None of the studies quoted here works on a total, and manufacturing one drowns the subscale that informs in the one that does not.
Deciding for a patient on a few points of change
The smallest detectable change is 9.4 points out of 24 and 12.7 out of 42. Below that, the observed change cannot be told from measurement error, and the Rasch analysis explicitly advises against individual decision-making.
Frequently asked questions
What does FABQ stand for?
FABQ stands for Fear-Avoidance Beliefs Questionnaire. It was published in 1993 by Waddell et al. to measure the fear-avoidance beliefs of the low back pain patient.
How is the FABQ scored?
It yields two subscales: physical activity from 0 to 24 and work from 0 to 42, each item scored across seven options. There is no total score.
At what score does the work subscale warn?
Below 30, the negative likelihood ratio of 0.08 solidly rules out the risk of prolonged work restriction. Above 34, the positive likelihood ratio is 3.33, which raises the risk without establishing it.
Is there a validated French version?
Yes, published in 2004 by Chaory et al., with a test-retest reliability of 0.88 for work and 0.72 for physical activity. Its authors note that responsiveness there is lowest on the work subscale.
Is the FABQ useful for following a patient?
With caution. The smallest detectable change is 9.4 points out of 24 and 12.7 out of 42, and the Rasch analysis concludes its subscales suit group judgement, not individual decision-making.
References
7 sources, PMIDs included
- Waddell G, Newton M, Henderson I, Somerville D, Main CJ. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain 1993;52(2):157-68. PMID 8455963. The original publication. Principal-components analysis, in 210 patients, yields two factors: work, internal consistency 0.88 and 43.7 % of the variance, and physical activity, 0.77 and 16.5 %. In 184 patients, work-related beliefs explain 23 % of the variance of daily disability and 26 % of that of work loss, after allowing for pain.
- Chaory K, Fayad F, Rannou F, Lefèvre-Colau MM, Fermanian J, Revel M, Poiraudeau S. Validation of the French version of the fear avoidance belief questionnaire. Spine (Phila Pa 1976) 2004;29(8):908-13. PMID 15082995. The validated French version. Test-retest reliability of 0.88 for work and 0.72 for physical activity, in 31 patients. Two caveats from the authors: factor analysis extracts four factors in one group and only the two original factors in the other, and the lowest responsiveness falls to the work subscale, with an effect size of 0.30.
- Fritz JM, George SZ. Identifying psychosocial variables in patients with acute work-related low back pain: the importance of fear-avoidance beliefs. Phys Ther 2002;82(10):973-83. PMID 12350212. 78 patients with acute work-related low back pain, 22 of whom keep restrictions at four weeks. The work subscale is the best predictor: negative likelihood ratio of 0.08 below 30, positive likelihood ratio of 3.33 above 34. That is where both thresholds come from.
- George SZ, Fritz JM, McNeil DW. Fear-avoidance beliefs as measured by the fear-avoidance beliefs questionnaire: change in fear-avoidance beliefs questionnaire is predictive of change in self-report of disability and pain intensity for patients with acute low back pain. Clin J Pain 2006;22(2):197-203. PMID 16428956. 63 patients with acute low back pain, reassessed after four weeks. The FABQ items correlate with lumbar flexion but not with pain intensity. Changes in beliefs explain a significant share of changes in disability, once pain is controlled for: that is the argument for using it as an outcome measure.
- George SZ, Fritz JM, Childs JD. Investigation of elevated fear-avoidance beliefs for patients with low back pain: a secondary analysis involving patients enrolled in physical therapy clinical trials. J Orthop Sports Phys Ther 2008;38(2):50-8. PMID 18349490. 160 patients followed for six months. The threshold of 14 on physical activity classifies 69.4 % of patients as elevated, with no difference in six-month outcome, and an area under the curve of 0.562 whose interval contains chance. The threshold of 29 on work classifies only 11.9 % of patients, and those do markedly worse: area under the curve 0.694.
- Ostelo RW, Swinkels-Meewisse IJ, Knol DL, Vlaeyen JW, de Vet HC. Assessing pain and pain-related fear in acute low back pain: what is the smallest detectable change? Int J Behav Med 2007;14(4):242-8. PMID 18001240. The study that gives the ranges and the noise. Physical activity subscale, from 0 to 24, smallest detectable change 9.4; work subscale, from 0 to 42, smallest detectable change 12.7. Its authors write that the physical activity subscale has considerable problems detecting improvement as well as deterioration.
- Franchignoni F, Giordano A, Rocca B, Ferriero G, Monticone M. A further Rasch analysis of the Fear-Avoidance Beliefs Questionnaire in adults with chronic low back pain suggests the revision of its rating scale. Eur J Phys Rehabil Med 2021;57(1):110-19. PMID 33215903. 155 patients with chronic low back pain. Both subscales are properly unidimensional, but the seven response options malfunction and four would suffice. Above all, person separation reliability is 0.69 and 0.79: the authors conclude the subscales suit group judgement, not individual decision-making.
Page written by Anthony Baillon, physiotherapist, co-founder of Physio Learning. The questionnaire itself is by Waddell et al., 1993: this page documents and interprets it, it reproduces none of its items.