What does “moderate pain” mean? Subgroups holding different conceptions of rating scales evaluate experimental pain differently

Background

Pain ratings are almost ubiquitous in pain assessment, but their variability is high. Low correlations of continuous/numerical rating scales with categorical scales suggest that individuals associate different sensations with the same number on a scale, jeopardizing the interpretation of statistical results. We analysed individual conceptions of rating scales and whether these conceptions can be utilized in the analysis of ratings of experimental stimuli in pain‐free healthy individuals and people with reoccurring/persistent pain.
Methods

Using a free positioning task, healthy participants (N = 57) and people with reoccurring/persistent pain (N = 57) ad libitum positioned pain descriptors on lines representing intensity and un‐/pleasantness scales. Furthermore, participants rated experimental thermal stimuli on visual analogue scales with the same end anchors. A latent class regression approach was used to detect subgroups with different response patterns in the free positioning task, indicating different conceptions of pain labels, and tested whether these subgroups differed in their ratings of experimental stimuli.
... mehr
Results

Subgroups representing different conceptions of pain labels could be described for the intensity and the un‐/pleasantness scale with in part opposing response patterns in the free positioning task. Response patterns did not differ between people with and without pain, but in people with pain subgroups showed differential ratings of high intensity experimental stimuli.
Conclusions

Individuals' conceptions of pain labels differ. These conceptions can be quantified and utilized to improve the analysis of ratings of experimental stimuli. Identifying subgroups with different conceptions of pain descriptions could be used to improve predictions of responses to pain in clinical contexts.
Significance

The present results provide a novel approach to incorporate individual conceptualizations of pain descriptors, which can induce large distortions in the analysis of pain ratings, in pain assessment. The approach can be used to achieve better pain estimates, representing individual conceptions of pain and achieving a better comparability between individuals but also between pain‐free persons and patients with chronic pain. Particularly, in clinical settings this could improve quantification of perceived pain and the patient‐clinician communication.