Acute and Perioperative Research Paper Confirmatory factor analysis of the International Pain Outcome questionnaire in surgery Mauricio Polanco-Garcı́aa, Roser Granerob,c, Lluı́s Gallartd, Jaume Garcı́a-Lopezd, Antonio Montesd,* Abstract Background: Choosing perioperative suitable treatments requires reliable and valid outcome measurements. The International Pain Outcome (IPO) questionnaire has been widely used for quality improvement and research purposes within the PAIN-OUT network that has collected more than 550,000 data sets of postoperative patients in 200 hospitals worldwide. Our aim is to confirm psychometric properties of the Spanish version of the IPO questionnaire and its invariance by pain predictors. Method: Sample included 4014 participants within a large age range, who underwent different surgical procedures. Confirmatory factor analysis (CFA) assessed internal structure, considering invariance by sex, age, procedure, smoking, obesity, affective disorder, and chronic pain. Incremental predictive validity of factor scores on question would have liked more pain treatment and opioid requirement was also estimated with logistic binary regression. Results: Confirmatory factor analysis verified original structure in 3 factors measuring pain intensity and interference (F1), adverse effects (F2), and perceptions of care (F3), with good internal consistency. Multigroup CFA analysis confirmed invariance by assessed pain predictors. Good incremental predictive capacity to identify would have liked more pain treatment was achieved. Conclusion: Our study confirms the factor structure, supports reliability, and adds some evidence of convergent validity of the Spanish adaptation of the IPO questionnaire. The sum of scores in its main factors serves a global outcome analysis tool. Low scores in F1 and F2 with high scores in F3 would indicate optimal quality of care. Keywords: Confirmatory factor analysis, Postoperative pain, PAIN OUT, International Pain Outcome questionnaire 1. Introduction related data from 200 hospitals across the world is the PAIN-OUT project, funded by the European Commission’s Seventh Frame- Pain is one of the factors that interferes with the proper recovery of work Programme (ClinicalTrials.gov: NCT02083835).30 It fo- patients after surgery, yet it is one of the most challenging factors cused on 3 areas related to postoperative pain: structure, to quantify because it is a subjective multidimensional experi- process of care, and outcomes, and aims to improve post- ence.17 One recent effort to collect and analyze postoperative operative outcomes through benchmarking, quality indicators, and the best available evidence. Sponsorships or competing interests that may be relevant to content are disclosed As an outcome measurement tool, the International Pain at the end of this article. Outcome (IPO) questionnaire22 was developed in 2 phases by a Department of Anesthesiology, Consorci Sanitari Alt Penedès-Garraf, Barcelona, Rothaug et al. based on the Revised American Pain Society Spain, b Ciber Fisiopatologia Obesidad y Nutrición (CIBERobn), Instituto Salud Patient Outcome Questionnaire.14,22 Using a forward–backward Carlos III, Madrid, Spain, c Department of Psychobiology and Methodology of methodology, the English version was translated to 9 languages, Health Science, Universitat Autònoma de Barcelona, Bellaterra, Spain, d Depart- including Spanish. After phase 1 analysis, they shortened the ment of Anesthesiology, Hospital del Mar. IMIM (Hospital del Mar Medical Research Institute), Universitat Autònoma de Barcelona, Bellaterra, Spain questionnaire to adapt it to the European population. From the *Corresponding author. Address: Department of Anesthesiology, Hospital del Mar, Revised American Pain Society Patient Outcome Questionnaire, IMIM (Hospital del Mar Medical Research Institute), Universitat Autònoma de falls and sleep were combined. They eliminated the items how Barcelona, Passeig Marı´tim de la Barceloneta, 25, 29, 08003 Barcelona, Spain. Tel.: helpful the information was as well as the frightened and de- 134616457451. E-mail address: AMontes@parcdesalutmar.cat (A. Montes). pressed items because they have a high correlation with anxious. Supplemental digital content is available for this article. Direct URL citations appear They included the following 3 additional questions: interference in the printed text and are provided in the HTML and PDF versions of this article on with breathing, would have liked more pain treatment, and the journal’s Web site (www.painrpts.com). presence of previous chronic pain. Using principal component Copyright © 2021 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The International Association for the Study of Pain. This is an open access article analysis, a 3-model factor structure was identified, explaining distributed under the Creative Commons Attribution License 4.0 (CCBY), which 53.8% of the total variance: pain intensity and interference, permits unrestricted use, distribution, and reproduction in any medium, provided the adverse effects, and perception of care.14,22 original work is properly cited. The general objective of this study was to evaluate the PR9 6 (2021) e903 psychometric properties of the Spanish adaptation of the IPO http://dx.doi.org/10.1097/PR9.0000000000000903 questionnaire in a large clinical sample including patients who 6 (2021) e903 www.painreportsonline.com 1 2 · M. Polanco-Garcı́a et al. 6 (2021) e903 PAIN Reports® underwent different types of surgery. The specific objectives were 2.4. Statistical analysis as follows: (1) to obtain empirical evidence confirming the 3-factor Statistical analysis was performed with Stata16 for Windows. structure for the IPO questionnaire found by Rothaug et al.22 Confirmatory factor analysis (CFA) was performed considering (measuring pain intensity and interference, adverse effects, and that the new adapted IPO questionnaire was developed after a perceptions of care); (2) to test invariance and differential item theoretical rationale procedure based on the following steps: (1) functioning of the 3-factor structure by the patients’ sex, adapting a new instrument for covering 3 specific domains chronological age, surgery type, current smoking, obesity, (constructs/areas) related to the pain measure (intensity, adverse affective disorder, and presence of chronic previous pain27; and effects, and self-perception of care); (2) reviewing all the items (3) to estimate the incremental predictive validity of the IPO factor initially assigned to each domain to ensure that the contents were scores on would have liked more pain treatment and total appropriate for the target population(s); and (3) providing a morphine consumption (defined as criteria of poor pain control). meaningful distribution and order to the new version of the questionnaire. Therefore, CFA in our study was performed 2. Methods assuming the existence of 3 latent theoretical factors from items measures on a 0 to 10 scale: factor 1 defined by 9 items 2.1. Participants measuring pain intensity and interference(s); F2 defined by 4 The methodology of the PAIN-OUT, a project funded by the items measuring pain treatment-related adverse effects; and F3 European Commission and supported by the International defined by 3 items assessing the patients’ perception of care: Association for the Study of Pain, has been described else- allowed to participate in pain treatment decision, pain relief, and where.20,29 The Spanish subsample of the European PAIN-OUT satisfaction with pain treatment, with higher scores meaning study was analyzed, which included n 5 4650 patients recruited better perception of care (Fig. 2).22 Maximum likelihood (ML) from 13 hospitals from 7 different regions of Spain. All centers were estimator with missing values was used, and the overall university hospitals (300–1000 beds) with an acute pain care unit. goodness of fit was evaluated through the standard statistical As the number of participants in the sample used for previous measures3: the root mean square error of approximation factorial validation studies was very low (less than 10% of the (RMSEA), Bentler’s comparative fit index (CFI), and the Tucker- sample), we decided to maintain them in the current study to Lewis index (TLI). Adequate model fit was considered for RMSEA increase external validity and generalization capacity. Data were ,0.10, TLI .0.9, and CFI .0.9. The x2 test was not considered collected between February 2010 and December 2013 by trained as a fitting measure because of the strong dependence of this test research assistants, who followed a standard operating procedure with sample sizes (it may fail to reject inappropriate models in provided by the PAIN-OUT project.20 Patients who accepted to small sample sizes because of the lack of statistical power and it participate completed the IPO questionnaire. The registered data may reject appropriate models in large samples sizes because of were entered in a secured multi-institutional web-based database the excess of statistical power). The internal consistency between using a random identifier. As an inclusion criterion, it was required items within each defined factor was estimated by omega that patients were in their first postoperative day and in the ward for coefficient (v, this measure was considered instead of usual at least 6 hours. Exclusion criteria included being asleep, sedated, Cronbach’s alpha because of the low number of items for the not in the ward at the time of data collection, and not able to factors 2 and 3 of the IPO questionnaire) (moderate consistence communicate, including any language barrier, not able to read and was considered for consistency coefficients equal or higher than understand questionnaires, and cognitive impairment (Fig. 1). 0.60 and good consistency for values higher than 0.70). A corrected item-scale correlation was calculated for each item. Because of the strong association between statistical signifi- 2.2. Instruments cance for the coefficients and sample size, a corrected item-scale The IPO questionnaire developed by Rothaug et al.22 and correlation was considered low to poor |R| .0.10, moderate to Process of Care questionnaire. This tool includes sociodemo- medium for |R| .0.24, and large to high for |R| .0.37 (these graphic information, relevant pain treatment comorbid condi- thresholds corresponds to Cohen’s d of 0.20, 0.50, and 0.80, tions, and perioperative anesthetic, surgical, and analgesic data respectively).7,21 documented in medical records. Both questionnaires are avail- Because of a steady association of the pain construct with able on PAIN-OUT website (http://pain-out.med.uni-jena.de/). variables such as sex, age (2 groups were defined based on the median in the own sample), surgery type, current smoking, obesity defined as body mass index greater or equal to 30 kg/m2, 2.3. Procedure history of affective disorder, and presence of chronic previous The study involving human subjects and the use of patient data pain,23,27 the structural configural invariance by these variables for research purposes was approved by the Committee on was analyzed. Given the low frequency of previous consumption Research Ethics of every participating center, and the research of opioid analgesics, this variable was not included in the analysis. was conducted in accordance with the Declaration of the World In this study, structural invariance tested factor loadings Medical Association (Committee on Research Ethics Clı́nica Parc equivalence across the groups. This assumption is supported if de Salut MAR, Reference code: No 2007/2998/I), and signed multigroup CFA (MGCFA) analysis met the following criteria6: (1) informed consent was obtained from all participants. the model specifying the items measuring each latent variable fits The measures analyzed in the study correspond to self-report the data well; (2) all factor loadings are substantial (usually above measures answered by the patients in the first day after surgery at 0.30) and statistically significant; and (3) no large modification the arrival to the surgery ward. Research assistants encouraged indices exist that point to model misspecifications. the patients to complete the questionnaires and could read the Comparison for the raw factor scores between participants’ unanswered questions to patient once, intending that the items sex, groups of age, surgery type, current smoking, obesity, were all answered and that no problems had occurred because of affective disorder, and presence of previous pain was based on a lack of understanding. analysis of variance. Effect size for the mean comparison was 6 (2021) e903 www.painreportsonline.com 3 Figure 1. Flowchart of the study. Inclusion criteria are not mutually exclusive. based on Cohen’s d coefficients (low effect size was considered incremental predictive validity of the IPO scores was estimated for |d| .0.20, mild to moderate for |d| .0.50, and high to large with the change or increase in the Nagelkerke pseudo-R2 (ΔR2) for |d| .0.8016). comparing first and second steps or blocks of the regression and As there are no gold standard measures of postoperative pain the incremental discriminant validity with the increase in the area outcomes, the questions on the survey would have liked more pain under the ROC curve (ΔAUC2). treatment, either pharmacological or nonpharmacological, and highopioid requirements, defined as more than 30 mg per 24 hours 3. Results morphine equivalent consumption, were used as measures of convergent validity. Opioid analgesic doses during the intraoperative 3.1. Participants and the first 24 hours after surgery, obtained from the process Of 4650 patients recruited, 636 did not participate in the study questionnaire, were used to calculate oral morphine equivalents because of being asleep, sedated, not in the ward at the time of consumption, based on published analgesic tables.19 Short-acting data collection (at least 6 hours in their first postoperative day in opioids (fentanyl and remifentanil) were part of the anesthetic the ward), or not able to communicate (patient is deaf or is not protocol and not used to provide postoperative analgesia, so they able to communicate in any of the IPO available languages). were excluded. The value 30 mg per 24 hours was chosen to Therefore, 4014 patients were analyzed. Figure 1 shows the dichotomize the variable morphine equivalent consumption to high flowchart of the study. Descriptive characteristics of the patients opioid consumption, based on previous literature.12,20 are shown in Table 1. The incremental predictive or discriminant validity of the factor scores measured through the IPO questionnaire on the question would have liked more pain treatment and high opioid consump- 3.2. Confirmatory factor analysis for the Spanish adaptation tion was estimated with logistic binary regression in 2 steps or of the International Pain Outcome blocks: (1) the first step or block entered and fixed the patients’ Figure 2 contains the path diagram for the 3-factor model tested sex, age, surgery type, current smoking, obesity, affective in the study, with the standardized coefficients obtained in the disorder, and presence of previous chronic pain; and (2) the single-group CFA (whole sample, n 5 4014). All the coefficients second block added the 3 raw scores in the IPO questionnaire. achieved high loadings with statistically significant results. Good Goodness of fit was valued with the Hosmer–Lemeshow test fitting was obtained for this initial model (RMSEA 5 0.059, 95% CI (adequate fitting was considered for P . 0.05), whereas the 0.056–0.061, CFI 5 0.926, and TLI 5 0.905). Internal 4 · M. Polanco-Garcı́a et al. 6 (2021) e903 PAIN Reports® Figure 2. Path diagram of the CFA for the IPO questionnaire in the study (n 5 4014). CFA, confirmatory factor analysis; CFI, Bentler’s comparative fit index; IPO, International Pain Outcome; RMSEA, root mean square error of approximation; TLI, Tucker-Lewis index; v, omega coefficient. consistency was good for the factor F1 pain intensity and statistical structure for men and women. However, invariance interference (v 5 0.82), moderate for factor F2 adverse effects reported significant results in the joint test for the rest of the (v 5 0.61), and good for F3 perceptions of care (v 5 0.72). These groups tested in the MGCFA. Examining separately the stan- results confirm the structure in three first-order factors for the IPO dardized coefficients in each group, significant high scores questionnaire in the whole sample (Table S1, supplementary (above 0.30) were achieved for all the items (except for the item material, contains the complete results for this CFA, as well as the measuring itching pertaining to the factor adverse effects, which frequency distribution of the raw scores for each item and for the had a score equal to 0.265 for the group defined for other surgery dimension scores in the study, available at http://links.lww.com/ types different to general and orthopedics and 0.271 for the PR9/A96). group of men). Table S2 (supplementary material, available at http://links.lww. com/PR9/A96) contains the results to move from the single- 3.3. Comparison and concordance of the raw group CFA obtained in the whole sample to the MGCFA to cross- dimension scores validate the 3-factor structure across the groups defined by the participants’ sex, age, surgery type, current smoking, obesity, Table 2 includes the comparison of the mean raw factor scores affective disorder, and the presence of previous chronic pain. measured with the IPO questionnaire in the groups considered in Good fitting indexes were obtained in the MGCFA models defined the study. Compared with men, women reported high mean for assessing no differences in the structure for the IPO based on values in the factors assessing pain intensity and interference sex, age, current smoking, affective disorder, obesity, and (P 5 0.001) and adverse effects (P , 0.001). These 2 factors also presence of previous pain, whereas fitting was only moderate registered higher mean scores as lower the patients’ age, for the model measuring invariance by the surgery type. whereas the factor perception of care registered lower mean for Nonsignificant results were found in the tests assessing in- older patients compared with other 2 groups of age (P , 0.001). variance by sex (x2 5 19.7, P 5 0.104), indicating a equally Differences for type of surgery only reported differences for the 6 (2021) e903 www.painreportsonline.com 5 Table 1 5 0.027 and ΔAUC 5 0.031). Significant OR coefficients were Descriptive for the sample (n 5 4014). achieved for the factors measuring pain intensity and interference (OR 5 1.076, P , 0.001) and perception of care (OR 5 0.909, Age (years old); mean 6 SD (n 5 3976) 59.7 616.3 P , 0.001) in would have liked more pain treatment, whereas in Gender; n (%) (n 5 3,990) high opioid consumption, significant OR coefficients were Female 2346 58.8% Male 1644 41.2% achieved for pain intensity and interference (OR 5 1.012, P , 0.001) and adverse effects (OR 5 1.014, P , 0.001), indicating Type of surgery; n (%) (n 5 3872) Orthopedics 2022 52.2% that the probability of liking for more treatment was higher for General Surgery 1713 44.2% patients who perceived higher pain intensity and interference and Others* 137 3.5% lower perception of care, while the probability of high opioid Previous chronic pain; n (%) (n 5 3912) consumption was higher for patients who perceived higher Yes 2081 53.2% intensity and interference and higher adverse effects. No 1831 46.8% Current smoking; n (%) (n 5 3440) Yes 377 11.0% 4. Discussion No 3063 89.0% This study aims to test the psychometric validity of the IPO Affective disorders; n (%) (n 5 3440) questionnaire in a large clinical Spanish sample with patients who Yes 2984 86.7% underwent a broad range of surgical procedures and perioper- No 456 13.3% ative management.20 The main results of this study provide Obesity; n (%) (n 5 3009) evidence about (1) the structure of the IPO questionnaire in 3- Yes 2009 66.8% factors (pain intensity and interference, adverse effects, and No 1000 33.2% perceptions of care); (2) the invariance of the structure by sex, Preoperative opioid use; n (%) (n 5 3873) age, surgery type, current smoking, history of affective disorder, No 3668 94.7% Yes 205 5.3% obesity, and presence of previous pain; and (3) the capability of the factor scores to predict would have liked more pain treatment. Liked more pain treatment; n (%) (n 5 3895) Yes 703 18.0% Similar to the initial IPO exploratory validation study,22 our CFA No 3192 82.0% shows that interference with breathing and coughing was the item with the lowest factor loading (0.33) on F1 (pain intensity and High opioid consumption (.30 mg/24 h oral morphine equivalents); n(%) (n 5 3510) interference). This could be related to the heterogeneity of the Yes 1013 28.8% procedures included in the sample, considering that factor No 2497 71.1% loading on this item increases in general compared with Information about treatment options; n(%) orthopedics procedures because limb procedures usually do (n 5 3867) not affect the respiratory system.20 The moderate adjustment of Yes 2542 65.7% invariance by type of procedure could also be influenced by this No 1325 34.3% item. For F2 (adverse effects), itch (0.32) is the factor with the * Others includes gynecology, urology, ENT, ophthalmology, cardiovascular, and thoracic surgeries. lowest standardized coefficient, which may be due to the low frequency of occurrence and low intensity compared with other factor measuring adverse effects; orthopedic surgery reported adverse effects.20 Itching was introduced in the IPO as one of the lower mean compared with both general (P 5 0.007) and other main adverse effects of intrathecal and epidural opioid treat- (P 5 0.004) surgeries. Current smoking reported less adverse ments, but the frequency of occurrence in large sample studies events (P 5 0.01). has been around 6% to 18%,11 and although it is bothersome, its relevance in the functional status and morbidity of postoperative patients is arguable. Regarding perception of care (F3), the 3.4. Incremental predictive or discriminative capacity of the satisfaction item explains almost all the variability in that factor, in International Pain Outcome contrast to the original exploratory factor analysis by Rothaug,22 where the items loads were more balanced. Also, for F3, our Table 3 includes the final result of the logistic regression study shows moderate internal consistency suggesting that pain measuring the incremental predictive validity of the IPO measures relief, satisfaction, and participation in pain treatment decisions after considering the patients’ sex, age, surgery type, history of measure different aspects of postoperative experience in Spanish affective disorders, current smoking, obesity, and presence of patients and should be treated separately. High participation and previous pain. For would have liked more pain treatment and for information about pain therapy, perceived by the patients, has high opioid consumption, goodness of fit was achieved for the been shown to be a predictor of less pain intensity, restriction with final model for the second step or block (P 5 0.093 and P 5 movement, and dissatisfaction in the German population.18 0.291, respectively). Increase in the pseudo-R2 after including Unlike other countries, such as the United States, where both and fixing the variables defined into the first step or block participation and satisfaction with treatment are high,28 Spanish indicated that the specific incremental predictive capacity of IPO patients, on average, perceive that they participate less in pain factor scores was around 23% (ΔR2 5 0.227; global predictive treatment, but despite this, satisfaction is comparable to other capacity for the final model was R2 5 0.248) and the specific countries.20 This may be due to sociocultural differences, the fact incremental discriminative capacity was also around 23% (ΔAUC that health care is public in Spain, and the degree of involvement 5 0.228; global discriminative accuracy for the final model was that Spanish patients want to have in treatment decisions may AUC 5 0.829) in would have liked more pain treatment. The differ from other countries. predictive capacity of the variables included in the model for high The analysis presented here suggests invariance of the factor opioid consumption was low (R2 5 0.057 and AUC 5 0.626) and structure for sex, age, previous chronic pain, current smoking, did not increase noticeably with the inclusion of factor scores (ΔR2 obesity, history of affective disorders, and type of surgery, which 6 · M. Polanco-Garcı́a et al. 6 (2021) e903 PAIN Reports® Table 2 Comparison for the raw factor scores between groups. Gender → Women (n 5 2346) Men (n 5 1644) Pairwise comparisons Women–men Mean SD Mean SD P |d| F1. Pain intensity/interference 21.4 14.9 20.0 14.1 0.001* 0.09 F2. Adverse effects 7.0 7.3 4.9 5.9 <0.001* 0.31 F3. Perceptions of care 11.6 6.5 11.9 6.2 0.051 0.05 Groups of age → 16–63 y 63–109 y Pairwise comparisons G1 (n 5 2092) G2 (n 5 2083) G1–G2 Mean SD Mean SD P |d| F1. Pain intensity/interference 22.3 15.5 16.3 14.1 <0.001* 0.40 F2. Adverse effects 6.5 7.1 5.1 6.6 <0.001* 0.21 F3. Perceptions of care 11.4 6.8 10.1 7.5 <0.001* 0.18 Previous pain → Without pain With pain Pairwise comparisons n 5 1831 n 5 2081 Without–with Mean SD Mean SD P |d| F1. Pain intensity/interference 20.6 14.3 20.9 14.9 0.755 0.02 F2. Adverse effects 6.1 6.8 6.2 6.9 0.670 0.02 F3. Perceptions of care 11.8 6.2 11.9 6.5 0.784 0.03 Surgery type → Orthopedics General Others Pairwise comparisons G1 (n 5 2022) G2 (n 5 1713) G3 (n 5 279) G1–G2 G1–G3 G2–G3 Mean SD Mean SD Mean SD P |d| P |d| P |d| F1. Pain intensity/interference 20.4 14.0 21.2 15.1 21.3 15.2 0.114 0.05 0.333 0.03 0.878 0.02 F2. Adverse effects 5.8 6.8 6.4 6.9 7.1 7.2 0.007* 0.09 0.004* 0.19 0.138 0.09 F3. Perceptions of care 11.6 6.4 11.8 6.3 11.9 6.4 0.412 0.02 0.476 0.05 0.774 0.01 Current smoking → No Yes Pairwise comparisons n 5 3063 n 5 377 No–Yes Mean SD Mean SD P |d| F1. Pain intensity/interference 21.0 15.0 21.1 15.1 0.569 0.01 F2. Adverse effects 6.3 6.9 5.4 6.9 0.010* 0.13 F3. Perceptions of care 11.6 6.4 10.8 7.5 0.008 0.13 History of affective disorder → No Yes Pairwise comparisons n 5 2984 n 5 456 No–Yes Mean SD Mean SD P |d| F1. Pain intensity/interference 20.9 14.9 22.1 15.6 0.950 0.08 F2. Adverse effects 6.2 6.9 6.2 7.0 0.587 0.01 F3. Perceptions of care 11.6 6.5 11.2 7.1 0.106 0.06 Obesity → No Yes Pairwise comparisons n 5 2009 n 5 1000 No–Yes Mean SD Mean SD P |d| F1. Pain intensity/interference 21.3 14.6 21.7 16.4 0.790 0.03 F2. Adverse effects 6.4 7.0 6.2 7.0 0.206 0.03 F3. Perceptions of care 11.5 6.3 11.8 6.5 0.890 0.05 * Bold: significant comparison (0.05 level). †Bold: effect size into the moderate (|d| . 0.50) to high range (|d| . 0.80). implies that factor scores can be used to compare these groups. the menstrual cycle, that generates a larger painful memory The lesser fit of the model by surgery type is consistent with network and higher pain sensitivity,5 a fact that could explain the previous literature that even lesser surgeries can result in significant increase in pain interference scores. Higher frequency of adverse pain and it it is likely that individual factors have a greater impact effects could be associated to pharmacokinetic differences in the than nociceptiva burden of surgical procedures in the experience metabolism of analgesic medication, as shown by the fact that of postoperative pain and treatment outcomes.12 Raw scores plasma morphine concentrations with the same dose are higher in show differences by sex in intensity and interference and adverse women and exposes them to higher frequency of adverse effects, but not in perception of care. The higher sensitivity to pain effects.10 Interestingly, although men had lower scores in intensity in women compared with men seems to occur after puberty,4 and of pain (F1) than women, in the logistic regression analysis, being a it is believed to be related to increased occurrence of pain due to man is a predictor of would have liked more pain treatment. 6 (2021) e903 www.painreportsonline.com 7 Table 3 Incremental predictive validity of the IPO on the criterion would have liked more pain treatment: logistic regression in 2 steps or blocks. B SE P OR 95% CI (OR) H-L ΔR2 ΔAUC Model 1 (would like more pain treatment) First step/block 0.394 0.021 0.601 Gender (0 5 women; 1 5 men) 0.504 0.121 <0.001* 1.656 1.306 2.099 Age (y) 20.007 0.004 0.052 0.993 0.986 1.000 Previous pain (0 5 no; 1 5 yes) 20.092 0.127 0.472 0.913 0.711 1.171 Surgery orthopedics vs general 0.824 0.133 <0.001* 2.281 1.757 2.961 Others vs general 0.310 0.268 0.247 1.364 0.806 2.307 History of affective disorders (0 5 no; 0.284 0.167 0.089 1.328 0.958 1.841 1 5 yes) Current smoker (0 5 no; 1 5 yes) 0.200 0.184 0.277 1.222 0.851 1.754 Obesity (0 5 BMI ,30; 1 5 BMI $ 30) 20.036 0.122 0.765 0.964 0.760 Second step/block 0.093 0.227 0.228 F1. Pain intensity/interference 0.076 0.004 <0.001* 1.076 1.068 1.083 F2. Adverse effects 0.005 0.008 0.554 1.007 0.993 1.021 F3. Perceptions of care 20.107 0.010 <0.001* 0.909 0.895 0.922 Model 2 (high opioid requirements) First step/block 0.033 0.057 0.626 Gender (0 5 women; 1 5 men) 0.159 0.095 0.092 1.173 0.974 1.412 Age (y) 20.008 0.003 0.008 0.992 0.987 0.998 Previous pain (0 5 no; 1 5 yes) 0.280 0.102 0.006 1.323 1.084 1.615 Surgery orthopedics vs general 0.510 0.104 <0.001 1.665 1.358 2.041 Others vs general 0.360 0.104 0.071 1.434 0.969 2.121 History of affective disorders (0 5 no; 0.007 0.136 0.959 1.007 0.771 1.316 1 5 yes) Current smoker (0 5 no; 1 5 yes) 0.204 0.152 0.180 1.226 0.910 1.652 Obesity (0 5 BMI ,30; 1 5 BMI $ 30) 0.609 0.094 ,0.001 1.838 1.530 2.208 Second step/block 0.291 0.027 0.031 F1. Pain intensity/interference 0.012 0.003 <0.001 1.012 1.006 1.018 F2. Adverse effects 0.027 0.007 <0.001 1.027 1.014 1.041 F3. Perceptions of care 0.002 0.008 0.834 1.002 0.987 1.017 H-L, Hosmer–Lemeshow test (P); ΔR2, change or increase in Nagelkerke pseudo-R; ΔAUC, change or increase in area under ROC curve. * Bold: significant parameter (0.05 level). IPO, International Pain Outcome. When patients are grouped by age, the scores of the younger 3 items: more pain relief, greater participation in the treatment of half of the sample are higher in intensity and interference, adverse pain, and no desire to have received more treatment. As effects, and perception of care. Gerbershagen et al. had already expected, intensity and interference score and adverse effects shown that young patients have greater postoperative pain were also related to high opioid consumption.25 regardless of the type of surgery.13 Although previous studies The main limitation of this study is the analysis of cross- have shown that satisfaction is greater in older patients,26 Jaipaul sectional data, which did not allow to study the temporal and Rosenthal showed that this factor varied with health status, reliability of the IPO measures, such as test–retest analysis. so that older adults (patients older than 65 years) with poor health Other psychometric validations of the IPO, such as acceptabil- status report less satisfaction with treatment compared with older ity, responsiveness, divergent validity, and feasibility, were adults in good physical condition.15 That could explain the higher already proved in Rothaug study, so they were not analyzed perception of care by the younger sample in our results because here. The overlap of the sample with the previous validation our population study came mainly from tertiary care hospitals that study is a limitation, although CFAs are statistical hypothesis treat older people with poor health status. Our raw scores do not tests that evaluate if the data fit a hypothesized structural model differ with the presence or absence of previous chronic pain, (in this case a 3-factor structure model). Thus, the CFA is now which is in agreement with previous studies showing that only sustained as the rational procedure used for the elaboration of previous severe chronic pain is related with worst postoperative the IPO questionnaire. Voluntary participation in the PAIN-OUT outcomes.13,20 Opposite to results in Yang metanalysis,27 study makes the sample representative only of patients who patients with obesity and history of affective disorders had similar attend public university hospitals. Characteristics of patients raw scores in the 3 factors and were not predictors of would have from other populations in Spain could modify the responses to liked more treatment. As expected, current smoking had lower the questionnaire and its factor structure.18 The IPO focus on raw scores in adverse events, mainly by the decreased risk of postoperative pain, which is corroborated by its factor structure nausea,2 but similar raw scores in pain intensity and perception of (the explained variance by intensity and interference factor in the care. Rothaug study is 36%). However, current trends derived from Would have liked more pain treatment is used to validate the fast track and enhanced surgery protocols1,8,9 focus on the predictive capacity of the questionnaire, under the assumption quality of postoperative recovery, especially the patient’s early that wanting more pain treatment is an indirect but specific ability to move, which depends on other factors additionally to measure of poor pain control. Our analysis shows the probability pain outcomes. Studies are needed to test the IPO validity in this of liking more pain treatment increases with the intensity and new paradigm. Another limitation of the study is the absence of interference score and decreases with the perception of care important variables related with the pain construct, such as score. This agrees with Schwenkglenks et al.24 study results that sleep difficulties, degree of catastrophizing, and patient showed that satisfaction with pain treatment was associated with resilience as well as lack of presenting treatment data, such 8 · M. Polanco-Garcı́a et al. 6 (2021) e903 PAIN Reports® as type, doses, and combination of pharmacological and Appendix A. Supplemental digital content nonpharmacological treatment. Items factor loading depend Supplemental digital content associated with this article can be on outcomes distribution, so pain treatments that significantly found online at http://links.lww.com/PR9/A96. influence patient outcomes could have altered the present results. Article history: The main strengths of our study are the large sample size, the Received 13 July 2020 inclusion of heterogeneous patients, and the use of MGCFA Received in revised form 5 January 2021 procedures to assess structure invariance by variables strongly Accepted 15 January 2021 related to pain. It is the first time that the IPO is conceptualized as Available online 5 March 2021 a sum of scores in its main factors, which can serve as a global outcome analysis tool. Low scores in pain interferences and adverse effects with high scores in perception of care would References indicate optimal quality of care.17 [1] Aasvang EK, Luna IE, Kehlet H. Challenges in postdischarge function and The total factor scores also allow simpler comparison between recovery: the case of fast-track hip and knee arthroplasty. Br J Anaesth 2015;115:861–6. centers and procedures and could become an improvement tool [2] Apfel CC, Heidrich FM, Jukar-Rao S, Jalota L, Hornuss C, Whelan RP, in the quality of postoperative pain management. Zhang K, Cakmakkaya OS. Evidence-based analysis of risk factors for Further studies will be needed to increase the convergent and postoperative nausea and vomiting. Br J Anaesth Oxford Univ Press divergent validity of the questionnaire with other measures related 2012;109:742–53. to postoperative pain and recovery from surgery. [3] Barrett P. Structural equation modelling: adjudging model fit. Pers Individ Dif 2007;42:815–24. [4] Boerner KE, Birnie KA, Caes L, Schinkel M, Chambers CT. Sex differences in experimental pain among healthy children: a systematic 5. Conclusion review and meta-analysis. PAIN 2014;155:983–93. In conclusion, the study confirms the 3-factors structure of the [5] Boerner KE, Chambers CT, Gahagan J, Keogh E, Fillingim RB, Mogil JS. Conceptual complexity of gender and its relevance to pain. PAIN 2018; IPO questionnaire in the Spanish population attending public 159:2137–41. university hospitals, its invariance by sex, age, previous chronic [6] Cheung GW, Rensvold RB. Evaluating goodness-of-fit indexes for testing pain, and type of surgery, and serves as a proof that the sum of measurement invariance. Struct Equ Model A Multidiscip J 2002;9: the scores of the factorial structure predicts would have liked 233–55. more pain treatment, a key aspect in patient satisfaction, along [7] Cohen J A power primer. Psychol Bull 1992;112:155–9. [8] Deng QF, Gu HY, Peng W, Zhang Q, Huang ZD, Zhang C, Yu YX. Impact with the ability to participate in treatment decisions and a sense of of enhanced recovery after surgery on postoperative recovery after joint providers caring for them.24 arthroplasty: results from a systematic review and meta-analysis. Postgrad Med J 2018;94:678–93. [9] Feldheiser A, Aziz O, Baldini G, Cox BPBW, Fearon KCH, Feldman LS, Disclosures Gan TJ, Kennedy RH, Ljungqvist O, Lobo DN, Miller T, Radtke FF, Ruiz Garces T, Schricker T, Scott MJ, Thacker JK, Ytrebø LM, Carli F. The authors have no conflicts of interest to declare. Enhanced Recovery after Surgery (ERAS) for gastrointestinal surgery, part 2: consensus statement for anaesthesia practice. Acta Anaesthesiol Scand 2016;60:289–334. Acknowledgements [10] Fillingim RB, Ness TJ, Glover TL, Campbell CM, Hastie BA, Price DD, The authors thank to the PAIN-OUT project, funded by the Staud R. Morphine responses and experimental pain: sex differences in side effects and cardiovascular responses but not analgesia. J Pain 2005; European Commission Seventh Framework Programme, Call 6:116–24. HEALTH- 2007- 3.1 4: Improving Clinical Decision Making, and [11] Fletcher D, Fermanian C, Mardaye A, Aegerter P. A patient-based endorsed by the International Association for the Study of Pain. national survey on postoperative pain management in France reveals The authors thank the members of the PAIN-OUT Spanish significant achievements and persistent challenges. PAIN 2008;137: 441–51. consortium who took part in the study: Dra Puig, IMIM; Drs [12] Gerbershagen HJ, Aduckathil S, van Wijck AJM, Peelen LM, Kalkman CJ, Bogdanovich and Gómez, Hospital Clı́nic, Barcelona; Drs Meissner W, Kalkman CJ, Meissner W. Pain intensity on the first day after Martinez, Pons, Cortell, Femenı́a, and Peláez, Hospital Universi- surgery: a prospective cohort study comparing 179 surgical procedures. tario La Fe, Valencia; Drs Aguilar, Pascual, Nohel, and Carbayo, Anesthesiology 2013;118:934–44. Hospital Son Llatzer, Palma de Mallorca; Dr Pascual, Hospital de [13] Gerbershagen HJ, Pogatzki-Zahn E, Aduckathil S, Peelen LM, Kappen TH, van Wijck AJM, Kalkman CJ, Meissner W, Kappen TH, Wijck Ciudad Real; Drs Martı́nez and Cano, Hospital Virgen de las AJMVan, Kalkman CJ, Meissner W. Procedure-specific risk factor Nieves, Granada; Dr Garcı́a-Filoso, Hospital San Juan, Alicante; analysis for the development of severe postoperative pain. Drs Telletxea, Ortega, and Pérez, Hospital de Galdakao Anesthesiology 2014;120:1237–45. Usansolo; Dr Crespo, Hospital Sant Joan de Deu, Sant Boi de [14] Gordon DB, Polomano RC, Pellino TA, Turk DC, Mccracken LM, Sherwood G, Paice JA, Wallace MS, Strassels SA, Farrar JT. Revised Llobregat; Drs Pérez, Conejo, Cicuéndez, Beltrán, Zorrilla, and American pain society patient outcome questionnaire (APS-POQ-R) for Martı́n, Hospital Clı́nico Universitario, Valladolid; Drs Rodriguez, quality improvement of pain management in hospitalized adults: Hernández, Yánez, Túñez, and Mancuso, Hospital Universitario preliminary psychometric evaluation. J Pain 2010;11:1172–86. Insular, Gran Canaria; Drs Guadalix, Romero, Hinojosa, and [15] Jaipaul CK, Rosenthal GE. Are older patients more satisfied with hospital Medina, Hospital Regional Universitario Carlos Haya, Málaga; care than younger patients? J Gen Intern Med 2003;18:23–30. [16] Kelley K, Preacher KJ. On effect size. Psychol Methods 2012;17:137–52. and Drs Atiénzar, Mansilla, Mercado, Orta, Massó, Sarabia, [17] Meissner W, Huygen F, Neugebauer EAM, Osterbrink J, Benhamou D, Sánchez, Pardo, and Muñoz, Hospital General de Castellón. The Betteridge N, Coluzzi F, De Andres J, Fawcett W, Fletcher D, Kalso E, authors thank Carlos Polanco for the English language revision of Kehlet H, Morlion B, Montes Pérez A, Pergolizzi J, Schäfer M. the article. Management of acute pain in the postoperative setting: the importance of quality indicators. Curr Med Res Opin 2018;34:187–96. No funders had any role in the study design, data collection and [18] Meißner W, Komann M, Erlenwein J, Stamer U, Scherag A. The quality of analysis, decision to publish, or the preparation of the manuscript. postoperative pain therapy in German hospitals - the effect of structural and procedural variables. Dtsch Arztebl Int 2017;114:161–7. 6 (2021) e903 www.painreportsonline.com 9 [19] Pereira J, Lawlor P, Vigano A, Dorgan M, Bruera E, Chb MB. [25] Slappendel R, Weber EWG, Bugter MLT, Dirksen R. The intensity of Equianalgesic dose ratios for opioids. a critical review and proposals for preoperative pain is directly correlated with the amount of morphine long-term dosing. J Pain Symptom Manage 2001;22:672–87. needed for postoperative analgesia. Anesth Analg 1999;88:146–8. [20] Polanco-Garcı́a M, Garcı́a-Lopez J, Fàbregas N, Meissner W, Puig MM. [26] Svensson I, Sjöström B, Haljamäe H. Influence of expectations and actual Postoperative pain management in Spanish hospitals. A cohort study pain experiences on satisfaction with postoperative pain management. using the PAIN-OUT registry. J Pain 2017;18:1237–52. Eur J Pain 2001;5:125–33. [21] Rosnow RL, Rosenthal R. Computing contrasts, effect sizes, and [27] Yang MMHH, Hartley RL, Leung AA, Ronksley PE, Jetté N, Casha S, Riva- counternulls on other people’s published data: general procedures for Cambrin J. Preoperative predictors of poor acute postoperative pain control: research consumers. Psychol Methods 1996;1:331–40. a systematic review and meta-analysis. BMJ Open 2019;9:e025091. [22] Rothaug J, Zaslansky R, Schwenkglenks M, Komann M, Allvin R, Backström [28] Zaslansky R, Meissner W, Chapman CR. Pain after orthopaedic surgery: R, Brill S, Buchholz I, Engel C, Fletcher D, Fodor L, Funk P, Gerbershagen HJ, differences in patient reported outcomes in the United States vs Gordon DB, Konrad C, Kopf A, Leykin Y, Pogatzki-Zahn E, Puig M, Rawal N, internationally. An observational study from the PAIN OUT dataset. Br J Taylor RS, Ullrich K, Volk T, Yahiaoui-Doktor M, Meissner W. Patients’ Anaesth 2018;120:790–7. perception of postoperative pain management: validation of the International [29] Zaslansky R, Rothaug J, Chapman CR, Bäckström R, Brill S, Fletcher D, Pain Outcomes (IPO) questionnaire. J Pain 2013;14:1361–70. Fodor L, Gordon DB, Komann M, Konrad C, Leykin Y, Pogatski-Zahn E, [23] Schnabel A, Yahiaoui-Doktor M, Meissner W, Zahn PK, Pogatzki-Zahn Puig MM, Rawal N, Ullrich K, Volk T, Meissner W. Pain out: the making of EM. Predicting poor postoperative acute pain outcome in adults: an an international acute pain registry. Eur J Pain 2014:1–13. international, multicentre database analysis of risk factors in 50,005 [30] Zaslansky R, Rothaug J, Chapman RC, Backstr öm R, Brill S, Engel C, patients. PAIN Rep 2020;5:e831. Fletcher D, Fodor L, Funk P, Gordon D, Komann M, Konrad C, Kopf [24] Schwenkglenks M, Gerbershagen HJ, Taylor RS, Pogatzki-Zahn E, A, Leykin Y, Pogatzki-Zahn E, Puig M, Rawal N, Schwenkglenks M, Komann M, Rothaug J, Volk T, Yahiaoui-Doktor M, Zaslansky R, Brill S, Taylor RS, Ullrich K, Volk T, Yahiaoui-Doktor M, Meissner W. Pain Ullrich K, Gordon DB, Meissner W. Correlates of satisfaction with pain out: an international acute pain registry supporting clinicians in treatment in the acute postoperative period: results from the international decision making and in quality improvement activities. J Eval Clin PAIN OUT registry. PAIN 2014;155:1401–11. Pract 2014:1–9.