Smeets R.J.E.M1,2,3, den Hollander M1,2 de Jong J.R1,2
1.Department of Rehabilitation, Maastricht University Medical Center, P. Debyelaan 25, 6229 HX Maastricht, the Netherlands
2 CAPHRI - School for Public Health and Primary Care, Department of Rehabilitation Medicine, Maastricht University, Universiteitssingel 60, 6229 ER Maastricht, the Netherlands
3 Adelante Centre of Expertise in Rehabilitation, Zandbergsweg 111, 6432 CC, Hoensbroek, The Netherlands

Comment on Plos one DOI:10.1371/journal.pone.0123008

Are pain-related fears mediators for reducing disability and pain in patients with complex regional pain syndrome type 1? An explorative analysis of pain exposure physical therapy by Karlijn J Barnhoorn at al. (1)


The last two decades the fear avoidance model (2, 3) has become increasingly popular to explain the development of chronic pain and the evolving loss of personal relevant activities and quality of life of a person suffering from this condition. According to this model, a patient who has catastrophic appraisals of the experienced pain develops pain related fear (fear to (re)injure himself due to movement, fear for other negative consequences as losing life-roles such as work, becoming wheelchair bound of fear of increasing/uncontrollable pain). The belief that pain means harm or potential damage to the body, implicates that pain should be avoided. This pain-related fear is associated with hypervigilance to any sensation that might indicate (possible) harm further fuelling the pain experience. Likewise, the beliefs about potential new harm and damage can be strongly influenced by peers and health care providers. All these beliefs and fears can result in the avoidance of movements and activities a person used to do or that he performs the movements and activities in an adapted way or only during specific conditions (escape and safety behaviour). This might result in physical deconditioning, social withdrawal and emotional problems such as depression, which further fuel the susceptibility for new pain stimuli and keep the person with chronic pain caught in a self-perpetuating cycle of functional disability.
There is also growing evidence for the consecutive steps within this model (4, 5), although some limitations have been identified (6). Based on this model, graded exposure in vivo (GEXP) treatment has been developed to identify and challenge catastrophic cognitions generating fear and avoidance behaviour. A team of physiatrist, occupational or physical therapist, and psychologist delivers the treatment. The treatment starts with the assessment of the medical as well as psychosocial situation of the patient. Important part is the medical education in which the patient is informed about his medical condition and safety to become active again and general information about pain processing. The psychologist informs the person on the influential role of psychosocial factors using the fear avoidance model (psychoeducation). Next, a hierarchy of perceived harmfulness of activities is established using a series of pictures in which a person performing different types of activities are shown (Photograph series Of Daily Activities; PHODA). This is followed by several sessions in which the actual exposure takes place, inviting the person to perform personal relevant, fear evoking activities, which are first shown by the therapist, and next the therapist invites the patient to first discuss the person’s expectations of what will happen in case the person will perform this acitivty himself and the patient is invited to perform the activity, immediately followed by discussing the person’s beliefs as the feared consequences will not have appeared (behavioural experiments). The aim is to facilitate inhibitory learning, which is currently seen as the mechanism underlying exposure techniques (7). This GEXP has been successfully applied in different pain syndromes (whiplash associated disorder (WAD)(8), low back pain(9)). Using a single case design, this treatment has been also been successfully tested in patients with Complex Regional Pain Syndrome type 1 (CRPS-1)(10), and the very positive results of an RCT comparing GEXP to pain contingent physiotherapy according to the Dutch guideline have just been submitted for publication. In another RCT, GEXP was compared to graded activity in patients with chronic low back pain expressing at least moderate levels of pain-related fears showed that the effect of GEXP relative to Graded Activity on functional disability and main complaints was significantly mediated by decreases in pain catastrophizing and perceived harmfulness of activities as measured with the PHODA (11). Similar results were found in the single case studies in WAD and CRPS-1.

Barnhoorn at all (1) present the results of a secondary analysis of a study in which two interventions, pain exposure physical therapy (PEPT) and conventional (pain contingent) physical therapy, in patients with CPRS type 1 were compared. The title of the publication suggests that the authors have specifically investigated the mediating role of pain-related fears in both treatments. Also in the introduction the authors specifically indicate that they hypothesized that reducing fear is not a prerequisite for reducing disability and pain in patients with CRPS-1, and that treatment does not primarily need to focus on reduction of fear to be effective. In the last sentence of the article they state that this study seems to indicate that pain-related fears do not have to be treated specifically in order to reduce disability and pain effectively in patients with CRPS-1.

However, we specifically have objections to this last statement. In order to justify this statement the authors should have performed a different mediation analysis, namely a mediation analysis for each treatment separately.
In the methods section it becomes clear that the authors have not assessed the mediating role of pain-related fears in each treatment separately, but they only assessed whether the found differences in the change of pain and disability between the two interventions are mediated by changes in pain-related fears. So in our opinion the used method is not appropriate to test their hypothesis and justify their last statement. In fact the authors have only shown that the found difference between the two treatments is not mediated by pain-related fears!
On the other hand they present results showing rather moderate to large and even statistically significant changes of catastrophizing, fear avoidance beliefs and kinesiophobia in both treatments (only change in FABQ-work was not significant for the control group). The authors themselves mention this in the discussion. So it seems plausible, that when they would perform a mediation analysis for each treatment, a mediating role of these factors can be found. Contrary to this, the authors use this change in pain-related fears to point out that without primarily focussing at fear avoidance these fears decreased. However, this statement can be debated. Taking a closer look at the description of the content of the PEPT treatment, they mention that PEPT starts ‘with a thorough education about … and pain-related avoidance behaviour. Therapists explained that pain is not a sign of injury, but rather ‘a rather false warning sign’, pain does hurt, but does not mean harm’, which in our opinion specifically aims to correct misinterpretations (catastrophic cognitions) a person might have about the meaning of pain/CRPS condition. Likewise in the conventional physiotherapy group ‘patients were invited to touch and move their affected limb and therapists convinced patients that moving and touching the affected limb is safe.’ So in both treatments specific attention was payed to, which could have caused the found significant changes in pain-related fears. From our own clinical experience we know that non-evidence based and irrational information from healthcare providers, and especially the internet (e.g. amputation in CRPS-1) often are significant contributors to these irrational beliefs and cognitions. From our single case studies we have shown that providing the medical education already reduces catastrophic cognitions and fear but that the exposure itself is necessary to change the avoidance behaviour (12). The difference with GEXP seems to be the use of the behavioural challenges and using the PHODA. So the suggestions of the authors to compare PEPT with GEXP is very interesting to assess whether the differences in treatment, and more specifically the long term outcome, will result in significant difference in pain and more specifically disability which is the main outcome of GEXP.
Furthermore, Barnhoorn et al used the TSK-11 and as the only study in which moderate to highly fearful patients with CRPS were studies used the TSK-17 (10), it is not clear whether the patients included in Barnhoorn’s study have moderate to high levels of pain-related fears. So, it might be that, despite the changes in pain-related fears as operationalized in the study of Barnhoorn, fear was not the most important factor that needed to be addressed in their population in order to improve disability.
The authors operationalized pain-related fears by choosing the following measures: Fear avoidance Beliefs Questionnaire, Pain Catastrophizing Scale and Tampa Scale for Kinesiophobia-11. Based on the current evidence and debate that we have started in 2010 (13, 14), we prefer the PHODA to assess pain-related fear. This instrument is specifically designed to help constructing the fear hierarchy during the GEXP treatment and the Back version has good clinimetric properties. In the CLBP study this instrument was used to proof the mediating role of perceived harmfulness of activities (9, 11). A version to assess harmfulness activities relevant to leg pain and one for arm pain are also available (15, 16). In a study of de Jong et al. (2011)(17) it was demonstrated that in patients with chronic CRPS-1 perceived harmfulness of activities as measured with the PHODA significantly predicts functional limitations. The study also showed that fear of movement/(re)injury as measured with the TSK was not related to functional limitations. These results support the idea that not general fear of movement/(re)injury, but the perceived harmfulness is a key factor that might be more systematically addressed in patients with CRPS-1.

Finally, we like to point out that the authors state this is a secondary analysis of a RCT but in fact they used the data of the per-protocol analysis, meaning that they only included patients who completed treatment. In the introduction the authors mention that the primary analysis showed that only the active range of motion improved significantly more after PEPT than after conventional therapy. This specific result adds additional concern, as this means that in fact there is no difference regarding pain and disability, the main outcomes, between the treatments, meaning that assessing the mediating role of pain-related fear for the difference between both treatments is in fact useless. Furthermore, the authors mention in the first paragraph of the Data Analysis that they did not abide to the randomization protocol and analysed all patients according to the treatment they received from the beginning of the trial. This is unclear to us; does this mean that some patients received another treatment than they were allocated to? If so, the authors should at least have discussed this and the problem of no difference on pain and disability in the primary analysis as potential biases in the discussion.

So in conclusion, given the limitations of the study of Barnhoorn et al, addressing pain-related fears in patients in CRPS-1 should definitely not be abandoned and further research is highly relevant. Furthermore, consensus on how to measure pain-related fear is necessary (14).

1. Barnhoorn KJ, Staal JB, van Dongen RTM, Frölke JPM, Klomp FP, van de Meent H, et al. Are Pain-Related Fears Mediators for Reducing Disability and Pain in Patients with Complex Regional Pain Syndrome Type 1? An Explorative Analysis on Pain Exposure Physical Therapy. PLoS ONE. 2015;10(4):e0123008.
2. Vlaeyen JWS, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: A state of the art. Pain. 2000;85(3):317-32. PubMed PMID: 2000-05349-001.
3. Leeuw M, Goossens ME, Linton SJ, Crombez G, Boersma K, Vlaeyen JW. The fear-avoidance model of musculoskeletal pain: current state of scientific evidence. J Behav Med. 2007 Feb;30(1):77-94. PubMed PMID: 17180640. Epub 2006/12/21. eng.
4. Hollander Md, de Jong JR, Volders S, Goossens MEJB, Smeets RJEM, Vlaeyen JWS. Fear reduction in patients with chronic pain: a learning theory perspective. Expert Review of Neurotherapeutics. 2010 2010/11/01;10(11):1733-45.
5. Gheldof EL, Crombez G, Van den Bussche E, Vinck J, Van Nieuwenhuyse A, Moens G, et al. Pain-related fear predicts disability, but not pain severity: a path analytic approach of the fear-avoidance model. Eur J Pain. 2010 Sep;14(8):870 e1-9. PubMed PMID: 20189423. Epub 2010/03/02. eng.
6. Wideman TH, Asmundson GGJ, Smeets RJEM, Zautra AJ, Simmonds MJ, Sullivan MJL, et al. Rethinking the fear avoidance model: Toward a multidimensional framework of pain‐related disability. PAIN. 2013;154(11):2262-5. PubMed PMID: 00006396-201311000-00007.
7. Craske MG, Kircanski K, Zelikowsky M, Mystkowski J, Chowdhury N, Baker A. Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy. 2008;46(1):5-27. PubMed PMID: 2008-01013-001.
8. de Jong JR, Vangronsveld K, Peters ML, Goossens ME, Onghena P, Bulte I, et al. Reduction of pain-related fear and disability in post-traumatic neck pain: a replicated single-case experimental study of exposure in vivo. J Pain. 2008 Dec;9(12):1123-34. PubMed PMID: 18722818. Epub 2008/08/30. eng.
9. Leeuw M, Goossens MEJB, van Breukelen GJP, de Jong JR, Heuts PHTG, Smeets RJEM, et al. Exposure in vivo versus operant graded activity in chronic low back pain patients: Results of a randomized controlled trial. Pain. 2008;138(1):192-207.
10. de Jong JR, Vlaeyen JWS, Onghena P, Cuypers C, den Hollander M, Ruijgrok J. Reduction of pain-related fear in complex regional pain syndrome type I: The application of graded exposure in vivo. Pain. 2005 Aug;116(3):264-75. PubMed PMID: WOS:000230937700014.
11. Leeuw M, Goossens ME, van Breukelen GJ, Boersma K, Vlaeyen JW. Measuring perceived harmfulness of physical activities in patients with chronic low back pain: the Photograph Series of Daily Activities--short electronic version. J Pain. 2007 Nov;8(11):840-9. PubMed PMID: 17632038. Epub 2007/07/17. eng.
12. de Jong JR, Vlaeyen JWS, Onghena P, Goossens MlEJB, Geilen M, Mulder H. Fear of Movement/(Re)injury in Chronic Low Back Pain: Education or Exposure In Vivo as Mediator to Fear Reduction? The Clinical Journal of Pain. 2005;21(1):9-17. PubMed PMID: 2004-22093-002.
13. Pincus T, Smeets RJEM, Simmonds MJ, Sullivan MJL. The Fear Avoidance Model Disentangled: Improving the Clinical Utility of the Fear Avoidance Model. The Clinical Journal of Pain. 2010;26(9):739-46. PubMed PMID: 00002508-201011000-00002.
14. Rainville J, Smeets RJEM, Bendix T, Tveito TH, Poiraudeau S, Indahl AJ. Fear-avoidance beliefs and pain avoidance in low back pain—translating research into clinical practice. The Spine Journal. 2011 9//;11(9):895-903.
15. Dubbers AT, Vikström MH, de Jong JR. The Photograph series of Daily Activities (PHODA-UE): cervical spine and shoulder. Heerlen/ Maastricht, the Netherlands: Zuyd University, Institute for Rehabilitation Research (iRv), Maastricht University; 2003.
16. Jelinek S, Germes D, Leyckes N, de Jong JR. The photograph series of Daily Activities (PHODA-LE): lower extremities. Heerlen/Maastricht, the Netherlands: Zuyd University, Institute for Rehabilitation Research (iRv), Maastricht University; 2003.
17. de Jong JR, Vlaeyen JWS, de Gelder JM, Patijn J. Pain-Related Fear, Perceived Harmfulness of Activities, and Functional Limitations in Complex Regional Pain Syndrome Type I. Journal of Pain. 2011 Dec;12(12):1209-18. PubMed PMID: WOS:000298572300002.