Pain catastrophizing measures shown to be invalid

Let’s talk about pain catastrophizing measures: an item content analysis [PeerJ]Mar 2020

This article points out what we’ve been saying all along: chronic pain IS a catastrophe and denying that is “magical thinking”.

Finally, someone is questioning the easy and dismissive beliefs that our pain is simply a mental problem or attitude. This explanation was uncritically embraced by a healthcare system that doesn’t know how (or isn’t allowed) to treat pain effectively.

All the self-report questions used to measure this artificial construct simply represent very normal concerns about a life-changing condition that impacts every aspect of life.

Concerns have been raised about whether self-report measures of pain catastrophizing reflect the construct as defined in the cognitive-behavioral literature.

We investigated the content of these self-report measures; that is, whether items assess the construct ‘pain catastrophizing’ and not other theoretical constructs (i.e., related constructs or pain outcomes) using the discriminant content validity method.  

Method

Items (n = 58) of six pain catastrophizing measures were complemented with items (n = 34) from questionnaires measuring pain-related

  • worrying,
  • vigilance,
  • pain severity,
  • distress, and
  • disability.

Via an online survey, 94 participants rated to what extent each item was relevant for assessing pain catastrophizing, defined as “to view or present pain or pain-related problems as considerably worse than they actually are” and other relevant constructs (pain-related worrying, vigilance, pain severity, distress, and disability).

Results

Data were analyzed using Bayesian hierarchical models. The results revealed that the items from pain-related worrying, vigilance, pain severity, distress, and disability questionnaires were distinctively related to their respective constructs.

This was not observed for the items from the pain catastrophizing questionnaires.

The content of the pain catastrophizing measures was equally well, or even better, captured by pain-related worrying or pain-related distress.

Conclusion

Based upon current findings, a recommendation may be to develop a novel pain catastrophizing questionnaire. However, we argue that pain catastrophizing cannot be assessed by self-report questionnaires.

Pain catastrophizing requires contextual information, and expert judgment, which cannot be provided by self-report questionnaires.

We argue for a person-centered approach, and propose to rename ‘pain catastrophizing’ measures in line with what is better measured: ‘pain-related worrying’.

Above was the abstract – below I get into the free full-text article:

Introduction

In regard to its causal status, studies investigating the processes of change in pain management programs revealed that reductions in pain catastrophizing are key for achieving treatment success (Besen et al., 2017; Burns, Day & Thorn, 2012; Burns et al., 2003; Jensen, Turner & Romano, 2001; Smeets et al., 2006; Turner, Holtzman & Mancl, 2007).

Comforted by the findings that pain catastrophizing ‘makes a difference’ in predicting pain, distress, and disability (Illari & Russo, 2014), researchers seem to have uncritically embraced the view that pain catastrophizing questionnaires measure pain catastrophizing. We argue that this approach is flawed and deserves scrutiny.

First, doubt exists whether the available questionnaires are interchangeable.

Second, catastrophizing is a transdiagnostic construct (Gellatly & Beck, 2016; Linton, 2013), and an essential ingredient in many models of psychopathological disorders (Beck, 1979; Clark, 1986; Salkovskis & Clark, 1993).

Concerns have been raised about whether catastrophizing as discussed in the pain field is faithful to how it is used in the psychopathology literature (Flink, Boersma & Linton, 2013; Neblett, 2017; Turner & Aaron, 2001).

Historically, catastrophizing was introduced by Ellis (1962) and picked up by Beck (1976), who argued that neurosis and emotional disorders are caused by irrational or exaggerated thought patterns, such as catastrophic thinking.

Catastrophizing was considered as repeated thinking that a situation is unbearable or terrible when it is just a nuisance (Ellis, 1962), or as a dwelling on the most extreme negative consequences conceivable (Beck, 1976).

At the core of these definitions is the idea that a person “views or presents a situation as considerably worse than it actually is” (https://en.oxforddictionaries.com/definition/catastrophize, accessed, 30/09/2016).

Third, pain catastrophizing questionnaires should be distinct from questionnaires assessing other theoretical constructs (e.g., pain vigilance, pain-related worry, or fear), and from primary outcomes (e.g., pain severity, pain-related distress, or pain-related disability; (Hirsh et al., 2007).

Otherwise, theory building becomes hazardous (Dixon & Johnston, 2019; Goubert, Crombez & Van Damme, 2004; Wideman, Adams & Sullivan, 2009) and the explanatory power of pain catastrophizing may be inflated.

Yes!

To address these challenges, we investigated to what extent items from pain catastrophizing measures are

(a) relevant for the construct ‘pain catastrophizing’ (content validity), and

(b) distinct from related constructs (i.e., ‘worrying about pain’, ‘pain vigilance’), and primary outcomes (i.e., ‘pain severity’, ‘pain-related distress’, and ‘pain-related disability’) (discriminant content validity).

Construct definitions

For each of the identified constructs, a definition was formulated. Some of the constructs were, however, considered self-explanatory (e.g., ‘pain severity’, ‘pain-related distress’) and no detailed definition was given. For other constructs, there were multiple definitions available.

To avoid the introduction of bias in our findings due to preferring the definition of one theoretical framework over another, we opted to use common language definitions as provided in the Online Oxford Living Dictionaries for English (https://en.oxforddictionaries.com accessed on 30/09/2016).

Where necessary, we adapted these definitions to the context of pain. This resulted in the following definitions:

(1) pain catastrophizing:To view or present pain or pain-related problems as considerably worse than they actually are’;

(2) worrying about pain: ‘To feel troubled or anxious about actual or potential pain or pain-related problems’;

(3) pain vigilance: ‘The action or state of keeping careful watch for possible pain’;

(4) pain-related distress: ‘Distress related to pain or pain-related problems’;

(5) pain-related disability: ‘Being limited in your movements, senses, or activities due to pain’; and

(6) pain severity: ‘The intensity or severity of pain’.

For the ‘other’ category the following description was provided: ‘The item does not measure any of the previous constructs’.

…six measures of ‘pain catastrophizing’ were included in the content analysis:

  1. The catastrophizing subscale (three items, e.g., “When I become aware of my pain, this thought comes through my head: I can’t have a tumour, can I?”) of the Avoidance Endurance Questionnaire
  2. The catastrophizing subscale (10 items1 , e.g., “I begin to worry that something might be seriously wrong with me”) of the Cognitive Coping Strategies Inventory
  3. The catastrophizing subscale (17 items, e.g., “I am disappointed in myself for giving in to the pain”) of the Pain Cognition List
  4. The Pain Catastrophizing Scale (13 items, e.g., “When I feel pain, I feel like I can’t go on”;
  5. The catastrophizing subscale (nine items, e.g., “This pain drives me crazy”) of the Pain-Related Self-Statements Scale
  6. The catastrophizing subscale (six items, e.g., “When I feel pain, I feel I can’t stand it anymore”) of the Coping Strategies Questionnaire

The set of catastrophizing items used can be found in Table S1.

Statistical analysis

First, as a manipulation check we investigated whether the items from questionnaires assessing ‘pain catastrophizing’, ‘worrying about pain, ‘pain vigilance’, ‘pain severity’, ‘pain-related distress’, and ‘pain-related disability’, were indeed most relevant for measuring their respective construct. 

Second, to identify which catastrophizing questionnaires scored highest on ‘pain catastrophizing’, we investigated the effect of measure (AEQ, CSQ, CSSI, PCL, PCS, and PRSS) on the outcome scores for the construct ‘pain catastrophizing’ only. 

Third, we investigated to what extent items of each pain catastrophizing measure were rated to be distinctively associated with the construct of pain catastrophizing, and less to the other constructs. 

In summary,

  • there is variability in the extent to which instruments measure ‘pain catastrophizing’.
  • none of the six instruments of pain catastrophizing distinctively assessed ‘pain catastrophizing’. Most instruments have content that was equally well, or even better captured by the constructs ‘worrying about pain’ or ‘pain-related distress’.
  •  none of the five highest scoring items of the pain catastrophizing instruments distinctively assessed ‘pain catastrophizing’. They have content that was equally well, or even better, captured by the constructs ‘worrying about pain’ or ’pain-related distress’.

Discussion

The results can be readily summarized.

First, there was variability in the extent to which instruments measure ‘pain catastrophizing’.

Second, despite some instruments measure ‘pain catastrophizing’ better than others, none of the instruments purporting to measure pain catastrophizing distinctively assessed ‘pain catastrophizing’. Most instruments had content that was equally well, or was even better, captured by the constructs of ‘worrying about pain’ or ‘pain-related distress’.

Third, this pattern was robust. It did not substantially differ between individuals reporting disabling pain, and it was also observed for the five items assigned the highest values for ‘pain catastrophizing’.

Current findings confirm the doubts raised about whether pain catastrophizing measures actually assess ‘pain catastrophizing’ as defined in the cognitive-behavioral literature

The results of current content analysis are sobering.

None of the six instruments distinctively measured ‘pain catastrophizing’. Similar analyses at the item level corroborated this finding.

Further, instruments purporting to measure ‘pain catastrophizing’ were considered to perform equally well, or even better in their ability to assess ‘worrying about pain’. 

This may not come as a surprise. Many authors consider pain catastrophizing as an extreme instance of worrying

An identical pattern was found for ‘pain catastrophizing’ and ‘pain-related distress’. Such a strong overlap was not expected. Theoretically, pain catastrophizing is a precursor of pain-related distress.

Such studies are designed to make it appear that catastrophizing caused increasing pain, but pain patients claim that pain causes catastrophizing. (see how much research has been corrupted by anti-opioid bias)

My post, Chronic Pain IS a Catastrophe, includes a multitude of reviews and studies showing this concept is seriously flawed. (You can find all my posts on catastrophizing here.)

We anticipated that cause and effect would be distinguishable. This was not observed. Reconsidering the literature, Beck (1976) and Ellis (1962) identified catastrophizing as part of clinical forms of distress and anxiety. Also, Hirsch and colleagues (2007) came to the same conclusion after showing that pain catastrophizing did offer little predictive value beyond negative mood.

We recognize that our conclusion that measures of pain catastrophizing actually do not measure ‘pain catastrophizing’, as understood by our participants, is a major challenge to models that hold pain catastrophizing as a core concept.

Good! Someone needs to push back against this ludicrous idea.

we argue that pain catastrophizing, defined as “to view or present pain or pain-related problems as considerably worse than they actually are”, cannot be assessed by self-report measures.

At the heart is a measurement issue common in psychology. Instruments with discriminative ambitions (Kirshner & Guyatt, 1985), such as pain catastrophizing questionnaires, need a reference standard to substantiate their diagnostic test accuracy (Bossuyt et al., 2015).

For catastrophizing, this means establishing that someone’s belief about pain or pain-related problems is incorrect or exaggerated.

To judge such an error, one needs three things:

  1. an objective measure of the actuality (‘how bad things really are’);
  2. the population standard for worrying about pain (‘how bad everyone else think they are’); and
  3. an expert judgment that the individual’s perception crosses the threshold into extreme (‘it is not as bad as they believe’).

Even when one can measure the extent of a real catastrophe, and if population standards for worrying about pain were established, somebody will always have to decide whether the experience of a person is worse than the situation demands. Such decisions require contextual information, which self-report instruments do not provide.

It’s easy for the person without pain to judge that our pain “isn’t really that bad” or doesn’t warrant “so much worrying” when they don’t have to deal with the loss of jobs and income and daily limitations.

In summary, we do not think it is possible to measure pain catastrophizing using self-report questionnaires. Therefore, we propose to adopt a person-centered approach, and to rename ‘pain catastrophizing’ measures in line with what is better measured: ‘pain-related worrying’

In contrast, the label ‘pain catastrophizing’ may easily elicit inappropriate referral to mental health professionals with expertise in abnormal and extreme cognition and affect, leaving patients feeling less understood and more stigmatized (Amtmann et al., 2018; De Ruddere & Craig, 2016).

Conclusions

Pain catastrophizing, defined as “to view or present pain or pain-related problems as considerably worse than they actually are” is not measured by current self-report questionnaires of pain catastrophizing. It is unlikely that it ever will.

The construct ‘pain catastrophizing’ requires contextual information, and expert judgment, which cannot be provided by self-report questionnaires.

We argue for a person-centered approach, and propose to rename ‘pain catastrophizing’ measures in line with what is better measured: ‘pain-related worrying’.

2 thoughts on “Pain catastrophizing measures shown to be invalid

  1. Kathy C

    Beth Darnell got a lot of mileage out of this. Nearly every article written about pain, discusses pain catastrophization as front and center. It helped spur a level of denial, where the catastrophizaion was more important than the pain, and was interpreted that pain is a psychological issue. The industries that profited were glad to amplify these kinds of terms, or anything that dehumanized and invalidated people with pain.
    They gave healthcare providers a justification for denying care not only for pain, but for painful conditions.

    Like

    Reply
  2. Pingback: “Pain Catastrophizing” Term Correction | EDS and Chronic Pain News & Info

Other thoughts?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.