Fibromyalgia patients and healthy volunteers express difficulties and variability in rating experimental pain: a qualitative study

1 Introduction

Pain is a significant public health problem in the United States, affecting more than 50% of all adults and costing the US health care system hundreds of billions of dollars annually [1], [2]. Fibromyalgia in particular (FM) is a rheumatologic disorder characterized by widespread pain affecting nearly 2% of the US population, and is associated with serious comorbid conditions [3], [4]. Furthermore, there are substantial health care costs associated with FM including hospitalization [5], work loss [6], [7], and direct medical costs [8], [9]. Human experimental pain research has enabled researchers to better understand the mechanisms involved in chronic pain conditions such as FM, and has informed clinical research with the goal of developing better therapeutic regimes [10], [11]. For example, studies have found that individuals with FM have lower heat pain thresholds and pain tolerances compared to healthy volunteers (HVs) suggesting the possibility of impaired inhibitory endogenous pain mechanisms in FM patients [12], [13], [14], [15].

Widespread research has demonstrated sound psychometric qualities of the visual analogue scale (VAS) for pain ratings in the clinical and experimental settings, and have supported the validity and reliability of the VAS pain intensity scale [13], [16, 17, 18, 19, 20]. Prior research has also looked at the influences of cognitive processes on experimental pain [21], [22], [23] and found that various thought patterns such as optimism, acceptance coping strategies, and mindful attention can ameliorate pain perception. Significantly fewer studies have used qualitative methods to examine how patients interpret and respond to pain assessments, and these have found considerable variability of interpretation of pain scales and inconsistent methods for rating clinical pain [24], [25], [26]. For example, a previous study asked chronic pain patients to describe how they came up with their pain ratings for the previous week, and found that some patients considered flares of pain, some focused on functioning, and some focused on certain days of the week [24]. Similarly, another study found that individuals with chronic pain had difficulty averaging their pain scores across different time periods such that they rated their pain higher over a 2 week versus 24-h time period [26].

The current study sought to build on these previous studies that have examined how chronic pain patients determine pain scores, by examining how participants rate their pain in the experimental environment. Because experimental pain is often rated by subjects immediately after receiving the pain sensation, rather than having to average pain over days or weeks, one might assume that there is less variability in strategies used to score experimental pain. However, there are several reasons to question this assumption. Many aspects of human pain experiments vary from study to study such as the number and mechanism of pain stimuli, instructions given to participants, purpose of the study, and others. It is possible that individuals rate pain differently under different conditions. Furthermore, these aspects of the experimental setting might have unique influences on strategies used by chronic pain patients and healthy volunteers to score painful stimuli.

We aimed to determine, through qualitative interviews, what participants in pain experiments think about when scoring the stimuli and whether these differ for chronic pain patients versus healthy volunteers. Based on available literature, this is the first study to delineate participants’ thought processes while rating experimental pain using inductive qualitative methods rather than pre-determined questionnaire items. The benefit of inductive qualitative interviews as a methodology is its flexibility to capture unanticipated outcomes, without being restricted to a pre-determined set of questions. In the current study, we use semi-structured interviews to compare and contrast thought processes in those with FM and HVs. We highlight sources of ambiguity in pain ratings, how these differ for individuals with FM versus HVs, and suggest strategies for mitigating difficulties and variability in ratings. Understanding the ways in which the experimental setting influences participant thought processes, and how they vary by pain status could assist pain researchers to better design and interpret pain research.

2 Materials and methods

This was a qualitative descriptive study in which semi-structured interviews were conducted with FM patients and HVs to capture their thought processes while rating experimental pain.

3 Results

Analysis of respondent interviews uncovered six overarching themes encompassing the thought processes of participants as they determined pain scores: (1) comparing with previous or current pain, (2) self-monitoring of one’s ability to endure the pain, (3) focusing on the physical aspects of the pain, (4) knowing the experimental setting is safe, (5) focusing on the pain scale as an anchor, and (6) desire to appear strong (Tables 2 and 3). Categories were not mutually exclusive, meaning that some participants employed multiple strategies to determine pain scores. Similarities and differences between themes seen in FM patients and HVs are described below.

4 Discussion

A first in the available literature, the present study used semi-structured interviews to uncover thought processes of FM patients and HVs while scoring experimental pain. These included: (1) comparing with previous or current pain, (2) self-monitoring of one’s ability to endure the pain, (3) focusing on the physical aspects of the pain, (4) knowing the experimental setting is safe, (5) focusing on the pain scale as an anchor, and (6) desire to appear strong. Additionally, participants described five factors that made it difficult to score the stimuli: (1) falling asleep, (2) mentally fatigued, (3) feeling as though they were guessing, (4) having to make a quick decision, and (5) difficulty in being consistent. Of particular note, several differences were found between FM patients and HVs in factors used to score pain: (1) using current daily pain as a benchmark to judge the experimental pain was seen more in FM participants, (2) wanting to appear strong in front of the study investigators was more common in HVs, and (3) becoming mentally fatigued from rating a long series of stimuli was more common for FM patients. Consistent with previous research in the clinical setting, the current study found considerable variability in strategies employed by participants to rate their pain. For example, a previous study found that chronic pain patients report higher pain over a 2-week versus 24-h period [26]. Another study found inconsistent strategies utilized by chronic pain patients to summarize pain over time [24], such that some considered flares of pain, some thought about functioning, and some focused on certain days of the week. The current study is consistent with these previous studies in that participants used varying strategies to rate the painful stimuli.

The present study found specific differences in strategies for rating the painful stimuli for FM patients versus HVs. One of these differences is that HV participants tried to present themselves as tough for the study investigators. These participants were attuned to the fact that they had an “audience” and wanted to seem “strong.” One possible explanation for why FM patients were less concerned than HVs about the opinions of the study investigators is that chronic pain patients participate in research, in-part, for personal benefit [35] and this might supersede the desire to be perceived favorably. The current findings are consistent with previous clinical research showing the influence of social desirability on pain ratings. For example, a previous study [36] found that some patients did not ask for analgesics after surgery because they didn’t want to be “troublesome,” and another found that patients with chronic lower back or neuropathic pain admitted to not always giving accurate ratings of their pain for concern about what doctors might think [37]. Further research could confirm whether this bias could be mitigated in the experimental setting by including reassurances of non-judgement as part of the instructions.

Another factor that differed for FM patients versus HVs was the way in which previous pain was considered while rating the experimental stimuli. While some participants reported using a previous burn, surgery, or broken bone as a benchmark for rating the experimental pain, FM patients sometimes used their current, daily pain as a comparison point for the experimental pain. Using previous clinical pain as a benchmark for pain ratings has also been found in the clinical setting. Van Dijk et al., 2016 found that postoperative patients used previous surgeries as a benchmark to judge their postoperative pain level and concluded that clinicians need to communicate with patients beyond their pain score when deciding on a treatment plan. Similarly, it is incumbent upon pain researchers to query study participants about the meaning of their scores so that meaningful individual and group comparisons can be made.

Both the current findings and previous research in the clinical setting have found that participants tend to use a constrained part of the scale when scoring pain, but for different reasons. In the clinical setting, chronic pain patients talked about not using the highest pain score because they could not imagine having that much pain [25]. The current findings also showed that some participants reported using a constrained part of the scale, but to preserve room for additional stimuli, rather than hesitance to report extreme scores. The intensity of the pain stimuli in the current study were presented in random order and, therefore, unpredictable to participants who were reluctant to use the entire scale without knowing how far up or down they would need to go. Reluctance to use the entire scale could lead to underreporting of pain in experiments, potentially weakening the strength of tested associations in research studies. Future research might assess whether this constraint can be mitigated by adjusting the number of stimuli or altering procedures.

Unique to the current study is that some factors were tied the experimental setting, such as some participants considering how they appeared to the study investigators and some focusing on their own safety during the study. Participants talked about knowing that there were parameters to the study and that “there was going to be no permanent damage.” Previous experimental pain research has found that participants tolerate painful stimuli better when feeling in control. For example, a previous study found that unpredictable stimuli are perceived as more unpleasant than ascending or anticipated stimuli, providing evidence that feeling in control can impact pain perception [38]. Additionally, studies examining cognitive processes on experimental pain have found that positive thoughts, such as “regarding thoughts as clouds in the sky passing by” can increase tolerance to painful stimuli [21], [22], [23].

Novel findings in the current study are factors reported by participants about what made it difficult for them to rate the painful stimuli during the experiment. All five factors expressed by participants are unique to the experimental setting, and becoming mentally fatigued was seen more in FM participants than in HVs. Several studies have found cognitive impairment in FM patients [39], [40], [41]. The current findings show FM patients report difficulty in staying focused when rating a series of painful stimuli. Patients described becoming mentally fatigued during the 52-min protocol, thereby potentially being less careful in their ratings. The length and complexity of the current experiment is typical in human experimental pain models [10], [11], [42]. Pain researchers should be cognizant that lengthy and complex studies may exacerbate mental fatigue in certain patient groups such as FM patients. In addition to the length of the experiment, several difficulties were related to the specific parameters such as having to make a quick decision. Participants described realizing that they only had a few seconds to score each pain stimulus. These participants talked about trying to “think quicker” and having “no room for error.” Since mental processing rates vary from person to person [43], [44], this difficulty points to additional threats to the validity of the pain scores.

There are several limitations to this study. Because it was adjunct to an existing protocol, certain aspects of the study, such as the limited demographics collected and the specific pain scale used, were predetermined before the qualitative study began. The lack of additional demographic information such as income and education could reduce the generalizability of the findings. Also, very few men were included, although, the gender distribution in our study follows that seen in FM epidemiology [45]. Additionally, the 0–200 VAS pain scale used in the current study is not commonly used outside of certain pain experiments. Despite these limitations, it is reasonable to assume that strategies for determining pain scores apply across pain scales.

5 Conclusions

The current findings show that participants express difficulty in rating experimental heat stimuli and tend to focus on factors other than the pain itself, some of which differ for FM patients versus HVs. FM patients tend to compare their daily pain with the experimental heat pain and become mentally fatigued more easily than HVs. HVs more commonly desire to appear strong in front of the study investigators. Common factors for assessing heat pain include self-monitoring of one’s ability to endure pain, focusing on physical aspects of pain and using the pain scale as an anchor. Difficulties experienced by both FM patients and HVs in rating pain include feelings of having to guess, making quick decisions, and trying to be consistent.

6 Implications

Understanding the ways in which the experimental setting influences patient pain ratings may help pain researchers better design and interpret studies. For instance, researchers can use these findings to mitigate difficulties for participants in experimental research to add to its validity. Examples include reducing the number of pain stimuli and providing more breaks to reduce fatigue, especially in FM patients who are more likely to mention fatigue as a difficulty to scoring the pain stimuli. Other possible modifications include tailoring instructions to reassure participants who might not want to appear afraid in front of the study investigators, and provide clarification about whether to focus on physical or mental aspects of the pain experience, or previous painful events. Additional studies are warranted to expand on the current findings.