In terms of years lived with disability, chronic pain represents the main source of worldwide disability nowadays.1 Chronic widespread pain represents one of the cornerstones in the definition of fibromyalgia (FM), and the recently revised International Classification of Diseases in its 11th revision classifies fibromyalgia as a form of chronic widespread pain.2 With a worldwide prevalence of 2.7%, FM affects mainly women in a 50–60 years range, and it represents a relevant burden at multiple levels, given the impact on the economic, living habits and daily routine spheres.3

In order to measure pain severity, several specific instruments in the framework of chronic pain exist. The Graded Chronic Pain Scale (GCPS) was initially developed as a brief tool to measure pain severity,4 and it has been validated into Spanish in subjects with low-back pain.5 More recently, the instrument has been revised and updated to adhere to the new paradigmatic definition of chronic pain,6 delving into the Graded Chronic Pain Scale-Revised (GCPS-R), which measures the impact of chronic pain. It is based on six simple items on pain in general, through an algorithm that classifies chronic pain into four different categories: absent, mild, bothersome, and high-impact chronic pain. The first two questions of the GCPS-R enquire about the level of pain and limitation in the past three months. The sum of the third, fourth, and fifth questions correspond to the 3-item ultra-brief pain measure Pain, Enjoyment of Life and General Activity (PEG) Scale7 and represent the total score of the questionnaire. The PEG scale specifically assesses pain intensity and interference.7 Item 6, which is not used to grade chronic pain, is a dichotomous question about the impact of paint on work status, and it does not influence the score of the questionnaire whatsoever. The final score of the GCPS-R is not summative, since it follows a tailored algorithm thoroughly explained in the original version of the instrument.6 Globally, subjects reporting, on item 1, no pain or pain on some days are placed at Grade 0 (which stands for “chronic pain absent”). Subjects with no chronic pain, even though they can report non-chronic pain that may be bothersome or may have a relevant impact, may also be included in Grade 0. As for people reporting the presence of chronic pain, they should be placed at Grades 1, 2 or 3: subjects whose chronic pain is not of high impact and with PEG scores below 12 points are placed at Grade 1 (“mild chronic pain”), whilst Grade 2 (i.e., “bothersome chronic pain”) corresponds to scores of and/or beyond 12 points in the PEG subscale. Grade 3 (“high impact chronic pain”) encompasses subjects reporting that pain limits either their life activities or work sphere on most days or every day in the past 3 months. The GCPS-R has been recently adapted and validated in different populations and conditions.8,9

Despite the relevance and clinical impact of the GCPS-R, its validation into Spanish has not been performed yet. The adaptation of the GCPS-R tool into Spanish would provide the clinical community with a useful, time-efficient and specific instrument to measure chronic pain in clinical routines, more specifically in the field of FM, given the prevalence and the subsequent burden of the condition.

The aim of the present study was, therefore, to develop a cross-cultural Spanish-language adaptation of the GCPS-R in a sample of patients diagnosed with FM. Besides, we aimed to carry out an initial psychometric analysis of the questionnaire in this population.

The translation and cross-cultural adaptation of the instrument delved into the final version of the GCPS-R-SP, available in Table 1.

The sample analyzed consisted of 224 subjects overall. A total of 30 different associations from 14 (out of 17) regions in Spain were represented in the sample. Approximately 95% of the sample were women, with mean age of 50.18±11.25 years of age. Most of the sample had either high-school or university studies. Roughly 70% were physically active, and 41.1% were in a full-time job situation. Further data on sociodemographic and clinical data are included in Table 2.

Concerning the CFA performed, the RMSEA scored 0.041, whilst the values for the CFI and the TLI corresponded to 0.984 and 0.960, respectively. The internal consistency of the two factors generated was subsequently tested: Factor 1 displayed a Cronbach's alpha of 0.711 and a McDonald's omega of 0.761, whilst Factor 2 had an alpha value of 0.890 and a McDonald's omega of 0.907. The results confirm the two-factor structure of the GCPS-R-SP, with Factor 1 including two items and Factor 2 encompassing three items. In fact, the second factor integrally corresponds to the PEG scale. Specific data on factor loadings and factor composition are expounded in Table 3.

For the convergent validity of the instrument, Pearson's correlations performed of the specific second factor of the instrument (i.e., the PEG scale) with the NRS produced strong correlation scores (Pearson's r=0.713; p-value<0.001).

Test–retest reliability was calculated from data stemming from a subgroup of 146 subjects. Weighted Cohen's Kappa corresponded to 0.537 (p-value<0.001) for the global GCPS-R-SP instrument. Fig. 1 displays the graphical representation on the levels of chronic pain in the test and retest period.

Fig. 1.

Test–retest and weighted Cohen's Kappa of the GCPS-R-SP (n=146).

To the best of our knowledge, this is the first study to validate and perform an initial psychometric testing of the instrument GCPS-R-SP, a fact that will enable health professionals to apply and assess chronic pain and its impact in clinical routine. The validation has been performed among subjects with fibromyalgia, a widespread and burdensome condition with a complex approach and a recurrent chronic pain assessment, since pain assessment is essential to promote effective care for people with fibromyalgia,25 a fact that highlights the relevance of the study here expounded.

One of the strengths and differential aspects of this article is the validation and subsequent clinical availability of a tool in Spanish for the specific assessment of chronic pain in the field of fibromyalgia. To date, other tools have often been used as “proxies” for measuring chronic pain in FM, such as unidimensional pain intensity tools (NRS) or the FIQ, a specific instrument for measuring the impact of fibromyalgia on functional capacity and quality of life.

The initial psychometric assessment yielded promising results, leading to the general picture of a satisfactory instrument from a statistical perspective, yet some aspects deserve special attention: first, while the scores of the RMSEA, the CFI, and the TLI are adequate and indicate good data-to-model fit,14 the specific analysis of the internal consistency on the first factor yielded a Cronbach's alpha (0.711) evincing a satisfactory level of consistency, yet bordering the cut-off point of 0.70, and lower than that from Liang et al.8 The second factor shows a high level of consistency, and it is, furthermore, an instrument in its own. These differential levels of consistency between both factors could lie on the fact that, as in the original study,6 the first two items were not intended to measure a specific construct, since the first item is used to define whether chronic pain exists, and the second item is used to specifically define limitations due to pain. Both items use a four-category ordinal scale that assesses frequency, which may explain why they are grouped into a single factor.

Moreover, when focusing on the convergent validity of the instrument, the results concerning the pain-intensity subscale of the instrument are statistically significant and they reveal a high extent of correlation with the construct NRS. Further research in this line shall assess other relevant variables (such as pain coping beliefs, psychological distress, health perceptions, or life activity limitations), in order to establish potential hypotheses leading to the evaluation of construct validity through known-groups strategies, since Von Korff et al.6 suggest as a potential further research line the hypothesis that the main differences between subjects with mild and bothersome chronic pain could be predominately characterized by differences in measures of pain coping beliefs, psychological distress and health perceptions, while differences in life activity limitations may ease the identification and differentiation of persons with high impact chronic pain from those with mild chronic pain.

Concerning the test–retest reliability, the results justify the stability of the instrument and its scoring system delving into the ordinal classification of four different levels of chronic-pain, given that the coefficient revealed a moderate to substantial reliability of the two measurements performed. In fact, solely five out of 146 subjects (3.66%) experienced a change in the level of pain impact assigned by the instrument through the algorithm initially provided by the authors.6

The results stemming from our study should be interpreted in the light of its methodological limitations. First, our sample was mainly composed of women: while the figures in our study are somehow consistent with the epidemiologic distribution of the condition – high female predominance, between 80 and 96%,26 the potential exploration of gender-differences in the perception and impact of chronic pain could be of interest within the performance of the instrument, since gender-differences in variables as quality of life have already been stated elsewhere.27 Second, the potential concurrence of other rheumatic diseases (e.g., rheumatoid arthritis, Systemic Lupus Erythematosus, or Sjögren's syndrome) alongside FM was not considered as a potential exclusion criterion, and it could have subsequently limited the pain processing or perception amongst the study subjects. Third, the psychometric properties of the instrument were not tested over time: further studies should explore the sensitivity to change of the measure to assess its performance in capturing true changes through time. Finally, the novelty of the research objective, based on the translation, cultural adaptation and cross-sectional psychometric testing of the instrument specifically in the field of fibromyalgia hinders any potential homogeneity in the comparison with other populations: for instance, the original study by Von Korff et al.6 included subjects with different medical conditions, and with a high percentage of elevated depression/anxiety symptoms, whilst in the study by Liang et al.,8 all subjects stemmed from tertiary care, and a vast majority of the study subjects did not present with chronic conditions. In the present study, by including subjects diagnosed with FM, the fact of recruiting a high percentage of subjects with high-impact chronic pain was foreseeable, and this is confirmed in the characterization of the sample. Further studies should, therefore, focus on the specific field of the outcome-measurement in FM, also psychometrically testing the performance of the instrument over time.

The GCPS-R-SP represents a simple, easy to administrate, and clinically efficient measure with favorable psychometric properties, covering the level and the impact of chronic pain in subjects with FM. The assessment instrument requires little time and resources, a fact that supports and endorses its clinical implementation, since time constraints represent one of the cornerstones of work stress in different professional spheres and healthcare environments.

José Édgar Ferrández-Gómez: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing, Visualization and Supervision.

Mariano Gacto-Sánchez: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing, Visualization and Supervision.

Aitor Baño-Alcaraz: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing, Visualization and Supervision.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

None.