The importance of measurement content and study design. Comment on Habibi Asgarabad et al. (2023). The relationship of alexithymia to pain and other symptoms in fibromyalgia: A systematic review and meta‐analysis

Did you know that people with multiple sclerosis score higher on alexithymia than a healthy comparison group (Eboni et al., 2018)? Did you know that people with cancer score higher on alexithymia than their healthy counterparts (Veirman et al., 2021)? Also, Habibi Asgarabad et al. (2023) report that alexithymia is highly prevalent in people with fibromyalgia, is substantially elevated compared to healthy people and is correlated with pain intensity. All in all, these findings seem to corroborate that alexithymia is powerful in its contribution to disease, illness and suffering and that interventions are needed to target alexithymia. However, ‘Extraordinary claims require extraordinary evidence’. This is also the case for the authors' claim that therapeutic interventions are recommended to improve emotional awareness, expression and processing in people with fibromyalgia to counteract high levels of alexithymia. We argue that such evidence is currently lacking. First, it can be questioned whether alexithymia was really measured. Indeed, the term ‘alexithymia’ was introduced by Sifneos to describe clinical observations of patients with classic psychosomatic diseases who had difficulty engaging in insight-oriented psychotherapy and later described as ‘the inability to recognize and express emotions’ (Taylor et al., 2016). Following, different questionnaires were developed to capture the key facets of alexithymia, of which the Toronto Alexithymia Scale (TAS) has been most widely employed. In its most recent version, alexithymia is assessed via three key facets, that is (1) difficulty identifying feelings, (2) difficulty describing feelings, and (3) externally oriented thinking. In their review, Habibi Asgarabad et al. (2023) found differences of medium to large effect size for the difficulty describing feelings and the difficulty identifying feelings facets, respectively, but only small and highly heterogeneous differences between people with fibromyalgia and healthy people for the externally oriented thinking facet. No difference was found for the externally oriented thinking facet between people with fibromyalgia and other pain groups. Can we then state that alexithymia is substantially higher in fibromyalgia, if one of the necessary facets is only slightly elevated or indifferent? Of further note, there is reason to believe that some items that allegedly measure alexithymia are in fact measuring something else. For example the item “I have physical sensations that even doctors don't understand” may measure illness perceptions or health anxiety, rather than one's difficulty to identify feelings. Similarly, it may not be surprising that people with fibromyalgia score high on the item “I am often puzzled by sensations in my body”. Yet, whether this is due to heightened alexithymia levels is questionable. For this reason, researchers have called for more caution when using the TAS in people with medical conditions, especially in pain disorders that are ‘medically unexplained’. Veirman et al. (2021), for example, showed that the TAS is only partially valid regarding its item content, and some items measure health anxiety rather than the intended alexithymia facet. We argue to be mindful about the fact that the validity of a questionnaire is not a static characteristic, but is only valid for a particular purpose in a particular group in a particular setting. Therefore, before making firm conclusions about the need for treatment to deal with elevated alexithymia in people with fibromyalgia, there is a further need to assess the content validity of the alexithymia scales in people with fibromyalgia and the setting of interest (e.g., via cognitive interviews). Second, no causal inferences can be made from cross-sectional studies. Indeed, although the authors caution to draw causal inferences, they describe that “…alexithymia may contribute to the presence and severity of fibromyalgia, suggesting that evidence-based interventions for fibromyalgia should target alexithymia…”. This conclusion is premature and potentially misleading because it is based on evidence from cross-sectional correlational and case-control studies. To enable us to draw such causal claims, prospective studies are of the essence. Yet, such studies need to be embedded in conceptual and theoretical models that explicitly and transparently explain how alexithymia may contribute to the development and/or maintenance of fibromyalgia or associated outcomes. The development of these theoretical models will not only boost prospective research but also indicate which confounding variables should be assessed, allowing to research their role in explaining the association between alexithymia and fibromyalgia. High-quality systematic reviews may then be performed using well-designed prospective studies to investigate the causal relationship between fibromyalgia and associated outcomes. Now may be the time to exclude cross-sectional studies from meta-analyses aimed at answering causal relationships. Furthermore, systematic reviews should also incorporate sensitivity analysis providing informative quantitative summaries of evidence strength (Mathur & VanderWeele, 2022). For example, these might then reveal that in the current review differences in alexithymia between people with fibromyalgia and other pain seems to be largely driven by one single study and provide insight into why this may be the case. In sum, we urge for improvement in measurement content and study design in future research on alexithymia in fibromyalgia.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.