- Email: [email protected]
Measuring acute postoperative pain using the visual analog scale: the minimal clinically important difference and patient acceptable symptom state
British Journal of Anaesthesia, 118 (3): 424–9 (2017) doi: 10.1093/bja/aew466 Pain
PAIN
Measuring acute postoperative pain using the visual analog scale: the minimal clinically important difference and patient acceptable symptom state P. S. Myles1,2,*, D. B. Myles2, W. Galagher1, D. Boyd1, C. Chew1, N. MacDonald3 and A. Dennis3 1
Department of Anaesthesia and Perioperative Medicine, Alfred Hospital and Monash University, Melbourne, Victoria, Australia, 2Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Victoria, Australia and 3Department of Anaesthesia, Royal Women’s Hospital, Parkville, Victoria, Australia *Corresponding author. E-mail: [email protected]
Abstract Background. The 100 mm visual analog scale (VAS) score is widely used to measure pain intensity after surgery. Despite this widespread use, it is unclear what constitutes the minimal clinically important difference (MCID); that is, what minimal change in score would indicate a meaningful change in a patient’s pain status. Methods. We enrolled a sequential, unselected cohort of patients recovering from surgery and used a VAS to quantify pain intensity. We compared changes in the VAS with a global rating-of-change questionnaire using an anchor-based method and three distribution-based methods (0.3 SD, standard error of the measurement, and 5% range). We then averaged the change estimates to determine the MCID for the pain VAS. The patient acceptable symptom state (PASS) was defined as the 25th centile of the VAS corresponding to a positive patient response to having made a good recovery from surgery. Results. We enrolled 224 patients at the first postoperative visit, and 219 of these were available for a second interview. The VAS scores improved significantly between the first two interviews. Triangulation of distribution and anchor-based methods resulted in an MCID of 9.9 for the pain VAS, and a PASS of 33. Conclusions. Analgesic interventions that provide a change of 10 for the 100 mm pain VAS signify a clinically important improvement or deterioration, and a VAS of 33 or less signifies acceptable pain control (i.e. a responder), after surgery. Key words: analgesia; pain measurement; surgery
Pain scales are useful for the assessment of postoperative pain and for monitoring the effectiveness of treatment.1 Most are based on self-reporting of a unidimensional scale aiming to represent subjective pain intensity.2–6 The 100 mm visual analog scale (VAS) and the 11-point numerical rating scale (NRS) are the most commonly used. But a reduction in a pain score of itself may not equate to an improvement in the patient’s experience.5–9
The VAS is frequently used as a measure of pain intensity, and authors and readers infer that a statistically significant difference in the VAS score equates to a clinically important reduction in pain. This is not necessarily correct.9 10 Previous studies have indicated that reductions in pain scores of around 30–40% are needed in order to reflect clinically useful improvements in pain.6 8 11–14 But it is unclear what is the minimal clinically
Editorial decision: December 10, 2016; Accepted: December 19, 2016 C The Author 2017. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. V
For Permissions, please email: [email protected]
424
Measurement of acute postoperative pain
Editor’s key points
|
425
perioperative data were collected on a case report form and later de-identified and transcribed onto an electronic database.
• The 100 mm visual analog scale (VAS) score is widely
used to measure pain intensity after surgery, but the minimal clinically important difference in the VAS is not clear. • A change of 10 for the 100 mm pain VAS would be the minimal clinically important difference, and the VAS of 33 or less signifies acceptable pain control after surgery.
Determination of the MCID for pain
important difference (MCID) of the pain VAS;15 16 that is, what minimal change in a pain VAS score would indicate a real change in a patient’s pain intensity.16 17 Several studies have attempted to define the MCID or clinically useful effect in the postoperative setting,18–20 but the methods used did not comply with existing standards nor did they include patient evaluation.15 The clinically important difference of the numerical rating scale (NRS) has been estimated for various chronic pain states,13 15 21 22 as has the MCID of the VAS in chronic pain23 and in the emergency department setting,24 25 but it is unclear whether these results can be applied in the acute postsurgical pain setting. One study has determined the MCID of the pain VAS in patients after shoulder rotator cuff surgery.26 The Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) recently reviewed and recommended specific methods that can be used for interpreting the clinical importance of treatment outcomes in chronic pain trials,14 but there are currently no recommendations for acute postoperative pain. The patient acceptable symptom state (PASS) is the value beyond which patients consider themselves well.15 27 28 The PASS can therefore be used to define responders and nonresponders to analgesic treatment in postoperative pain studies.21 The PASS of the pain VAS score in this setting is often assumed, but has not been determined according to current recommendations.15 The aim of this study was to determine the MCID and PASS for the pain VAS in patients recovering from surgery.
There is currently no consensus on the optimal method for MCID estimation. As such, we chose to include a triangulation (average) of several methodologies,15 16 as we have done previously when evaluating quality-of-recovery scales.29 Previous methods used to determine MCID have included the SD/2 rule,30 and 0.2 SD,31 0.3 SD,32 and the standard error of measurement (SEM);33 others have used 5–10% of the instrument range.34 We chose to include three distribution-based measures: the 0.3 SD, 35 SEM, and 5% range. In addition, we used an anchor-based method with a global rating-of-change questionnaire.36 This uses a 15-point Likert scale ranging from –7 (a very great deal worse) to þ7 (a very great deal better).15 16 36 A 100 mm VAS, ranging from 0 (no pain) to 100 (very severe pain), was used to measure pain intensity throughout the previous 24 h on two occasions in the days after surgery. At the second visit, patients were asked to assess the following (adapted from Tubach and colleagues).15 ‘Think only about your pain you have felt over the past 24 hours. Compared with yesterday, is your pain’: 7, a very great deal worse; 6, a great deal worse; 5, a good deal worse; 4, moderately worse; 3, somewhat worse; 2, a little worse; 1, almost the same, hardly any worse at all; 0, no change; 1, almost the same, hardly any better at all; 2, a little better; 3, somewhat better; 4, moderately better; 5, a good deal better; 6, great deal better; or 7, a very great deal better? Patients whose score on the global rating-of-change questionnaire was 0, 1, or 1 were classified as unchanged.36 Patients whose score was 2, 3, 2, or 3 were considered to have experienced a small change equivalent to the MCID; those with scores of 4, 5, 4, and 5 were considered to have experienced moderate change, and those with scores of 6, 7, 6, and 7 were considered to have experienced large change.36 Absolute (i.e. we changed the sign of the scores for those who deteriorated) mean changes in pain VAS scores according to patient-rated change in postoperative recovery health status were then calculated. All four estimates (0.3 SD, SEM, 5% range, and global rating of change) were then averaged.
Methods
Patients’ opinion of their improvement
This prospective observational study evaluated adult patients recovering from surgery, using the 100 mm VAS to measure pain on two occasions, along with a generic Likert scale of overall recovery (see next subsection). Most patients (n¼204) enrolled in this study participated in a concurrent study evaluating quality-of-recovery scales.29 The study settings were the surgical wards at three hospitals in Australia (Alfred, Royal Women’s, and Shepparton Hospitals) representing tertiary adult, tertiary obstetric/gynaecology, and rural/regional hospitals. Ethics approval was obtained from the institutional ethics committee at each hospital, and patient consent was obtained in all instances. Patients were eligible to participate in the study if they were >18 yr of age and recovering from a surgical procedure requiring general or major neuraxial block anaesthesia. Patients were excluded if they had poor English comprehension, drug or alcohol dependence, psychiatric disorder, uncontrolled pain, or a concurrent serious medical disorder impairing completion of the VAS and questionnaire. Baseline patient characteristics and
The PASS was determined using the direct opinion-based approach,15 37 in which patients were asked to define any improvement: ‘In your opinion, have you made a good recovery from your operation?’, with response options of yes, no, or unsure. Patients responding in the affirmative were classified as having made a good recovery, and those who responded negatively or were unsure were classified as having a poor recovery.
Statistical analysis We could not reliably estimate a required sample size for this study. Given that previous relevant studies had enrolled 40–100 subjects,20 36 we planned to enrol at least 150 subjects to provide an adequate number for subgroup testing. Data are presented as mean (SD) or number (%) unless otherwise specified. Selected results are reported with a 95% confidence interval (CI), with the mean (95% CI) VAS score for minimal change calculated using 1000 bootstrap samples. The SEM was calculated as the SD multiplied by the square root of one minus the intraclass correlation coefficient.33 Changes in pain
426
|
Myles et al.
scores at each interview were compared with the Wilcoxon signed rank test, and according to extent of surgery using the Kruskal–Wallis test. Both Student’s unpaired t-test and independent-samples median test were used to compare pain scores in those with a good or poor recovery. We defined the PASS as the 25th centile of the pain VAS in those who rated their recovery as good at the second postoperative interview. The predictive utility of the pain VAS in achieving a good recovery was assessed using a receiver operating characteristic curve (ROC).38 The association between the change in pain VAS score and the global rating-of-change score was quantified by the Spearman rank correlation coefficient (r). We determined test–retest reliability in those who had rated their pain state as unchanged or almost the same (global rating scores of 1, 0 or 1) using the intraclass correlation coefficient. Responsiveness was measured in those with global rating-of-change scores of at least 4 using standardized response means, calculated as the mean change divided by its SD.39 40 All statistical analyses were performed using SPSS for Windows V23.0 (SPSS Australasia Ltd, Sydney, NSW, Australia). A P-value of <0.05 was considered significant; no correction was made for multiple comparisons.
Results We enrolled 224 patients with varying degrees of postoperative pain and recovering from a broad range of surgical procedures; there were no refusals, but five were unavailable at the second postoperative visit (Table 1). Patients undergoing more extensive surgery tended to have higher VAS scores (Table 2). Most patients improved their pain score ratings at the second postoperative visit (Table 3). The median pain VAS scores reduced from 26 (13–47) to 20 (11–36), P¼0.002. The distributionbased estimates of the MCID for the VAS score were 6.9 (0.3 SD), 4.7 (5% range), and 12.4 (SEM). The association between the change in pain VAS score and the global rating-of-change score was r ¼ 0.48, P<0.0005. The absolute mean changes (95% CI) in VAS scores according to patient-rated change in postoperative pain intensity are reported in Table 4. The anchor-based estimate of the MCID for the pain VAS was 15.7. This results in an averaged MCID of 9.9. Those who rated themselves as having had a poor postoperative recovery [n¼22 (11%)] had significantly lower pain VAS scores at the second postoperative visit compared with those who reported a good postoperative recovery [n¼177 (89%)]; see Table 5 and Fig. 1. The 75th centile for the pain VAS score in those with a good recovery was 33; the 25th centile for the pain VAS score in those with a poor recovery was 20. The area under the ROC curve was 0.72 (95% CI: 0.60–0.85). Test–retest reliability (n¼22) of the pain VAS was high, intraclass correlation coefficient 0.79 (95% CI: 0.49–0.91). The standardized response mean (n¼125) of the pain VAS was 0.36, indicating a mild-to-moderate degree of responsiveness to change.
Discussion We studied a broad range of surgical and obstetric patients to quantify the MCID and PASS for the pain VAS as a measure of pain intensity in the postoperative setting. These findings can guide the conduct and interpretation of studies evaluating analgesic therapies in surgical patients. Each can be used to define a ‘responder’ (effective analgesia) as a standard metric to be used across acute pain studies.
Table 1 Patient and surgical characteristics (n¼224). Number (%) or median (interquartile range), unless otherwise stated. *New Zealand, South Africa, UK, or USA Variable Age (yr) Mean (SD) Range Female sex White-collar worker Country of birth Australia Other English-language country* Non-English-speaking country ASA physical status I II III IV Extent of surgery Ambulatory or other minor Intermediate Major Type of surgery General Orthopaedic Gynaecology Caesarean section Urological Plastics Major vascular Ear, nose, throat, or faciomaxillary Cardiac Other Duration of surgery (min) Timing of assessment after surgery (h) Visit 1 Visit 2 (n¼218)
53 (17) 19–86 148 (66) 95 (42) 160 (71) 30 (13) 34 (15) 47 (21) 112 (50) 56 (25) 9 (4.0) 17 (7.6) 103 (46) 104 (46) 51 (23) 44 (20) 36 (16) 37 (17) 11 (4.9) 11 (4.9) 6 (2.7) 5 (2.2) 5 (2.2) 18 (8.1) 112 (58–161) 22 (9.5–26) 43 (31–48)
It is generally accepted that a pain VAS score of 30, 70, and 100 indicates the upper boundaries of mild, moderate, and severe pain intensity. Several studies have found that a pain scale PASS value of 40 can be used as an outcome criterion for osteoarthritis and various chronic rheumatic diseases with pain.41 27 Pooled data from two postoperative analgesic studies found that pain VAS scores of up to 44 are consistent with a patient rating of pain being mild.42 Current guidelines recommend titrating analgesia to achieve a pain VAS score of 40 or less,43 and postoperative pain studies using patient-controlled analgesia typically titrate to a VAS or NRS score of 30.44 Our PASS estimate of 33 for the pain VAS is consistent with these results. Clinical trials ought to focus more strongly on the patient’s response to pain treatments; that is, the proportion of patients who are responders to treatment.14 Clearly, giving sufficient analgesia to eliminate all pain for all patients is a wrong target,45 but so is undertreatment. We thus recommend a pain VAS score of 33 be used as a suitable target for optimal analgesic titration. As outlined by Moore and colleagues,18 a dichotomous pain outcome (‘responder’ and ‘non-responder’) allows for the estimation of clinically useful statistics, such as absolute and relative risk reduction, and number needed to treat. This can then provide more useful information for clinicians and patients.
Measurement of acute postoperative pain
Table 2 Summary statistics for pain visual analog scale scores at the first postoperative visit, expressed as mean (SD). VAS, visual analog scale. *Kruskal–Wallis test Extent of surgery Pain VAS scale (range of scores)
Minor Intermediate Major (n¼17) (n¼103) (n¼104)
VAS score (0–100) 19 (24)
29 (23)
34 (22)
|
427
Table 5 Comparison of the pain visual analog scale scores in those who rated their recovery at the second postoperative visit, expressed as mean (SD) and median (interquartile range). CI, confidence interval; IQR, interquartile range; VAS, visual analog scale. *Levene’s test P¼0.01, so equal variances not assumed. †Mann–Whitney U-test
P-value Patient-rated good recovery 0.008*
Pain intensity Yes (n¼177) Pain VAS scale 24 (20) Median (IQR) 20 (10–33)
No/Unsure (n¼22) 45 (27) 51 (20–63)
Difference (95% CI) P-value 21 (8.2–33) —
0.002* 0.001†
Table 3 The frequency of each level of patient-reported change in pain intensity between the first and second postoperative visits (n¼219) n (%)
A very great deal worse A great deal worse A good deal worse Moderately worse Somewhat worse A little worse Almost the same, hardly any worse at all No change Almost the same, hardly any better at all A little better Somewhat better Moderately better A good deal better A great deal better A very great deal better
4 (1.8) 4 (1.8) 2 (0.9) 4 (1.8) 4 (1.8) 11 (5.0) 8 (3.7) 23 (11) 5 (2.3) 23 (11) 20 (9.1) 30 (14) 41 (19) 28 (13) 12 (5.5)
100 80 VAS score
Extent of change
60 40 20 0 Yes
No/unsure Good recovery
Fig 1 Pain visual analog scale (VAS) scores in those who rated their recovery at the second postoperative visit. Box plots represent median (interquartile range), with the whiskers representing 10th and 90th centiles.
Table 4 Mean change in pain visual analog scale scores according to patient-rated change in postoperative pain intensity (n¼219). CI, confidence interval; VAS, visual analog scale. *Used to define the anchor-based estimate of the minimal clinically important change (see main text for details) Change categories
Mean change
95% CI
(A) According to change categories of improvement or deterioration Large improvement (n¼40) Moderate improvement (n¼71) Minimal improvement (n¼43) No change (n¼36) Minimal deterioration (n¼15) Moderate deterioration (n¼6) Large deterioration (n¼8)
14.8 11.0 6.7 0.8 19.1 10.3 34.5
8.0 to 21.6 6.8 to 15.1 1.3 to 12.0 5.8 to 4.1 26.4 to 11.7 20.2 to 0.5 56.9 to 12.1
(B) Absolute change categories No change (n¼36) Minimal change (n¼58)* Moderate change (n¼77) Large change (n¼48)
0.8 15.7 15.1 20.1
5.8 to 4.1 12.9 to 18.8 12.3 to 18.1 14.5 to 26.4
Anchor-based methods have been used to determine the MCID for scales used in many medical conditions,16 20 26 32 35 36 and we have recently done this for several quality-of-recovery scales.29 Our approach, using a combination of distribution- and anchor-based methods, suggests that the MCID for pain in a surgical population is smaller than previously assumed. Studies of patients presenting to emergency departments with various pain states found an MCID for the VAS score of 1224 and 13,25 and a change of 18 correlated with the patient being a ‘little bit better’.19 Slightly higher values have been found for the pain NRS in chronic pain conditions, with an MCID of 2 (comparable to VAS 20 mm).13 21 22 The MCID for the pain VAS was 14 in patients treated for shoulder rotator cuff disease,26 and ranged from 4 to 21 in patients with degenerative cervical spine disease undergoing surgery, but in the latter study the evaluation was done by the clinician (not the patient).20 For patients recovering from minimally invasive spine surgery, an MCID for the VAS was 12 for back pain and 16 for leg pain.46 Others have found that a change in VAS of 20 represented satisfactory pain relief in patients after surgery.5 11 Some authors have suggested that a 33% decrease,42 or a change in VAS score of 15–20,5 6 47 would indicate a clinically important change in pain intensity. Taken
428
|
Myles et al.
together, these results largely support our estimate of the MCID for the pain VAS being 10 in the postoperative setting. Although the VAS and NRS are incomplete representations of the pain experience and cannot fully reflect the multidimensional aspects of pain, they remain the most widely used metrics of pain after surgery. In view of the high correlations between the VAS and NRS reported in many studies, it is reasonable to accept that the MCID can be applied to either instrument. The MCID can be used to guide the non-inferiority margin for clinical trials48 and for determining the sample size in a clinical trial.49 This study has limitations. Pain is both subjective and multidimensional and so the VAS (and NRS) cannot capture the complete pain experience. But clinical decisions are made on the basis of existing pain scales, and so it is important to know how much reduction in a VAS score is likely to be clinically meaningful from the patient’s perspective. The extremes of pain indicated on a VAS, typically ‘no pain’ and ‘worst pain ever’, may not truly represent absolute limits of perception. Our data (Table 4A) suggest that there is a difference in the MCID for those who have improvement or deterioration in their pain status, in that a larger reduction in a VAS score might be needed to indicate true improvement. A similar finding has been previously reported50 and is consistent with studies of loss aversion. It is possible that the experience of change/improvement in pain is non-linear. A reduction of pain from 95 to 75 could indicate greater relief than a reduction of 30 to 10 (particularly as this is below the identified PASS). We determined the PASS by asking patients to rate their overall recovery after surgery; although pain is an important component of this, other factors will, of course, influence recovery. In conclusion, we have shown that in patients with acute pain after surgery, including women recovering from Caesarean section, the MCID and PASS for the pain VAS score is 9.9 (rounded to 10) and 33, respectively.
Authors’ contributions Study concept and protocol development, analysis of results, first draft and revision of manuscript: P.M., D.M. Patient consent and enrolment, data collection: Ethics committee application, liaison with obstetric staff, and oversight of the conduct of the study at the Royal Women’s Hospital: A.D., N.M. Critical review and revisions of the manuscript: P.M., D.M., W.G., D.B., C.C., N.M., A.D.
Funding Australian National Health and Medical Research Council Practitioner Fellowship (APP1042462; Canberra, ACT, Australia; to P.M.).
References 1.
2.
3.
4.
5.
6. 7.
8.
9. 10.
11.
12.
13.
14.
Acknowledgements We thank the study participants willing to contribute to a better understanding of the measurement of pain, and the numerous clinicians providing postoperative care and pain relief in these settings.
15.
16.
Declaration of interest P.M. is an editor of the BJA; none of the other authors report any competing interests relating to the topic of this paper.
17.
Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain 1986; 27: 117–26 Revill SI, Robinson JO, Rosen M, Hogg MI. The reliability of a linear analogue for evaluating pain. Anaesthesia 1976; 31: 1191–8 Downie WW, Leatham PA, Rhind VM, Wright V, Branco JA, Anderson JA. Studies with pain rating scales. Ann Rheum Dis 1978; 37: 378–81 Price DD, Bush FM, Long S, Harkins SW. A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 1994; 56: 217–26 DeLoach LJ, Higgins MS, Caplan AB, Stiff JL. The visual analog scale in the immediate postoperative period: intrasubject variability and correlation with a numeric scale. Anesth Analg 1998; 86: 102–6 Campbell WI, Patterson CC. Quantifying meaningful changes in pain. Anaesthesia 1998; 53: 121–5 Myles PS, Urquhart N. The linearity of the visual analogue scale in patients with severe acute pain. Anaesth Intensive Care 2005; 33: 54–8 Bodian CA, Freedman G, Hossain S, Eisenkraft JB, Beilin Y. The visual analog scale for pain: clinical significance in postoperative patients. Anesthesiology 2001; 95: 1356–61 Myles PS, Christelis N. Measuring pain and analgesic response. Eur J Anaesthesiol 2011; 28: 399–400 Weibel S, Jokinen J, Pace NL, et al. Efficacy and safety of intravenous lidocaine for postoperative analgesia and recovery after surgery: a systematic review with trial sequential analysis. Br J Anaesth 2016; 116: 770–83 Aubrun F, Langeron O, Quesnel C, Coriat P, Riou B. Relationships between measurement of pain using visual analog score and morphine requirements during postoperative intravenous morphine titration. Anesthesiology 2003; 98: 1415–21 Cepeda MS, Africano JM, Polo R, Alcala R, Carr DB. What decline in pain intensity is meaningful to patients with acute pain? Pain 2003; 105: 151–7 Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Defining the clinically important difference in pain outcome measures. Pain 2000; 88: 287–94 Dworkin RH, Turk DC, McDermott MP, et al. Interpreting the clinical importance of group differences in chronic pain clinical trials: IMMPACT recommendations. Pain 2009; 146: 238–44 Tubach F, Wells GA, Ravaud P, Dougados M. Minimal clinically important difference, low disease activity state, and patient acceptable symptom state: methodological issues. J Rheumatol 2005; 32: 2025–9 Wells G, Beaton D, Shea B, et al. Minimal clinically important differences: review of methods. J Rheumatol 2001; 28: 406–12 Guyatt GH, Osoba D, Wu AW, Wyrwich KW, Norman GR; Clinical Significance Consensus Meeting Group. Methods to explain the clinical significance of health status measures. Mayo Clin Proc 2002; 77: 371–83
Measurement of acute postoperative pain
18. Moore A, Moore O, McQuay H, Gavaghan D. Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesics: use of pain intensity and visual analogue scales. Pain 1997; 69: 311–5 19. Todd KH, Funk JP. The minimum clinically important difference in physician-assigned visual analog pain scores. Acad Emerg Med 1996; 3: 142–6 20. Auffinger B, Lam S, Shen J, Roitberg BZ. Measuring surgical outcomes in subaxial degenerative cervical spine disease patients: minimum clinically important difference as a tool for determining meaningful clinical improvement. Neurosurgery 2014; 74: 206–13 21. Farrar JT, Young JP, Jr, LaMoreaux L, Werth JL, Poole RM. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 2001; 94: 149–58 22. Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine (Phila Pa 1976) 2005; 30: 1331–4 23. Emshoff R, Bertram S, Emshoff I. Clinically important difference thresholds of the visual analog scale: a conceptual model for identifying meaningful intraindividual changes for pain intensity. Pain 2011; 152: 2277–82 24. Kelly AM. The minimum clinically significant difference in visual analogue scale pain score does not differ with severity of pain. Emerg Med J 2001; 18: 205–7 25. Gallagher EJ, Liebman M, Bijur PE. Prospective validation of clinically important changes in pain severity measured on a visual analog scale. Ann Emerg Med 2001; 38: 633–8 26. Tashjian RZ, Deloach J, Porucznik CA, Powell AP. Minimal clinically important differences (MCID) and patient acceptable symptomatic state (PASS) for visual analog scales (VAS) measuring pain in patients treated for rotator cuff disease. J Shoulder Elbow Surg 2009; 18: 927–32 27. Tubach F, Ravaud P, Martin-Mola E, et al. Minimum clinically important improvement and patient acceptable symptom state in pain and function in rheumatoid arthritis, ankylosing spondylitis, chronic back pain, hand osteoarthritis, and hip and knee osteoarthritis: results from a prospective multinational study. Arthritis Care Res (Hoboken) 2012; 64: 1699–707 28. Wariaghli G, Allali F, Berrada K, et al. The patient acceptable symptom state of chronic musculoskeletal pain measured on a visual analog scale in Moroccan patients. Pain Med 2013; 14: 103–9 29. Myles PS, Myles DB, Galagher W, Chew C, MacDonald N, Dennis A. Minimal clinically important difference for three quality of recovery scales. Anesthesiology 2016; 125: 39–45 30. Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation. Med Care 2003; 41: 582–92 31. Samsa G, Edelman D, Rothman ML, Williams GR, Lipscomb J, Matchar D. Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II. Pharmacoeconomics 1999; 15: 141–55 32. Eton DT, Cella D, Yost KJ, et al. A combination of distributionand anchor-based approaches determined minimally
33.
34.
35.
36.
37.
38. 39. 40. 41.
42.
43.
44.
45. 46.
47. 48.
49.
50.
|
429
important differences (MIDs) for four endpoints in a breast cancer scale. J Clin Epidemiol 2004; 57: 898–910 Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life. J Clin Epidemiol 1999; 52: 861–73 Ringash J, O’Sullivan B, Bezjak A, Redelmeier DA. Interpreting clinically significant changes in patientreported outcomes. Cancer 2007; 110: 196–202 Maringwa J, Quinten C, King M, et al. Minimal clinically meaningful differences for the EORTC QLQ-C30 and EORTC QLQ-BN20 scales in brain cancer patients. Ann Oncol 2011; 22: 2107–12 Juniper EF, Guyatt GH, Willan A, Griffith LE. Determining a minimal important change in a disease-specific quality of life questionnaire. J Clin Epidemiol 1994; 47: 81–7 Wells G, Anderson J, Boers M, et al. MCID/Low Disease Activity State Workshop: summary, recommendations, and research agenda. J Rheumatol 2003; 30: 1115–8 Sedgwick P. How to read a receiver operating characteristic curve. Br Med J 2015; 350: h2464 Kirshner B, Guyatt G. A methodological framework for assessing health indices. J Chronic Dis 1985; 38: 27–36 Kazis LE, Anderson JJ, Meenan RF. Effect sizes for intepreting changes in health status. Med Care 1989; 27: S178–89 Perrot S, Bertin P. “Feeling better” or “feeling well” in usual care of hip and knee osteoarthritis pain: determination of cutoff points for patient acceptable symptom state (PASS) and minimal clinically important improvement (MCII) at rest and on movement in a national multicenter cohort study of 2414 patients with painful osteoarthritis. Pain 2013; 154: 248–56 Jensen MP, Chen C, Brugger AM. Interpretation of visual analog scale ratings and change scores: a reanalysis of two clinical trials of postoperative pain. J Pain 2003; 4: 407–14 Hartrick CT, Kovan JP, Shapiro S. The numeric rating scale for clinical pain measurement: a ratio measure? Pain Pract 2003; 3: 310–6 Bainbridge D, Martin JE, Cheng DC. Patient-controlled versus nurse-controlled analgesia after cardiac surgery – a metaanalysis. Can J Anaesth 2006; 53: 492–9 Lee TH. Zero pain is not the goal. JAMA 2016; 315: 1575–7 Park P, Okonkwo DO, Nguyen S, et al. Can a minimal clinically important difference be achieved in elderly patients with adult spinal deformity who undergo minimally invasive spinal surgery? World Neurosurg 2016; 86: 168–72 Myles PS. The pain visual analog scale: linear or nonlinear? Anesthesiology 2004; 100: 744. Committee for Medicinal Products for Human Use: Guideline on the Choice of the Non-Inferiority Margin. London. 2005. Available from http://wwwemaeuropaeu/docs/en_GB/docu ment_library/Scientific_guideline/2009/09/WC500003636pdf (accessed 1 July 2016) Gibbs NM, Weightman WM. Beyond effect size: consideration of the minimum effect size of interest in anesthesia trials. Anesth Analg 2012; 114: 471–5 Sloman R, Wruble AW, Rosen G, Rom M. Determination of clinically meaningful levels of pain reduction in patients experiencing acute postoperative pain. Pain Manag Nurs 2006; 7: 153–8 Handling editor: Takashi Asai
PAIN
Measuring acute postoperative pain using the visual analog scale: the minimal clinically important difference and patient acceptable symptom state P. S. Myles1,2,*, D. B. Myles2, W. Galagher1, D. Boyd1, C. Chew1, N. MacDonald3 and A. Dennis3 1
Department of Anaesthesia and Perioperative Medicine, Alfred Hospital and Monash University, Melbourne, Victoria, Australia, 2Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Victoria, Australia and 3Department of Anaesthesia, Royal Women’s Hospital, Parkville, Victoria, Australia *Corresponding author. E-mail: [email protected]
Abstract Background. The 100 mm visual analog scale (VAS) score is widely used to measure pain intensity after surgery. Despite this widespread use, it is unclear what constitutes the minimal clinically important difference (MCID); that is, what minimal change in score would indicate a meaningful change in a patient’s pain status. Methods. We enrolled a sequential, unselected cohort of patients recovering from surgery and used a VAS to quantify pain intensity. We compared changes in the VAS with a global rating-of-change questionnaire using an anchor-based method and three distribution-based methods (0.3 SD, standard error of the measurement, and 5% range). We then averaged the change estimates to determine the MCID for the pain VAS. The patient acceptable symptom state (PASS) was defined as the 25th centile of the VAS corresponding to a positive patient response to having made a good recovery from surgery. Results. We enrolled 224 patients at the first postoperative visit, and 219 of these were available for a second interview. The VAS scores improved significantly between the first two interviews. Triangulation of distribution and anchor-based methods resulted in an MCID of 9.9 for the pain VAS, and a PASS of 33. Conclusions. Analgesic interventions that provide a change of 10 for the 100 mm pain VAS signify a clinically important improvement or deterioration, and a VAS of 33 or less signifies acceptable pain control (i.e. a responder), after surgery. Key words: analgesia; pain measurement; surgery
Pain scales are useful for the assessment of postoperative pain and for monitoring the effectiveness of treatment.1 Most are based on self-reporting of a unidimensional scale aiming to represent subjective pain intensity.2–6 The 100 mm visual analog scale (VAS) and the 11-point numerical rating scale (NRS) are the most commonly used. But a reduction in a pain score of itself may not equate to an improvement in the patient’s experience.5–9
The VAS is frequently used as a measure of pain intensity, and authors and readers infer that a statistically significant difference in the VAS score equates to a clinically important reduction in pain. This is not necessarily correct.9 10 Previous studies have indicated that reductions in pain scores of around 30–40% are needed in order to reflect clinically useful improvements in pain.6 8 11–14 But it is unclear what is the minimal clinically
Editorial decision: December 10, 2016; Accepted: December 19, 2016 C The Author 2017. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. V
For Permissions, please email: [email protected]
424
Measurement of acute postoperative pain
Editor’s key points
|
425
perioperative data were collected on a case report form and later de-identified and transcribed onto an electronic database.
• The 100 mm visual analog scale (VAS) score is widely
used to measure pain intensity after surgery, but the minimal clinically important difference in the VAS is not clear. • A change of 10 for the 100 mm pain VAS would be the minimal clinically important difference, and the VAS of 33 or less signifies acceptable pain control after surgery.
Determination of the MCID for pain
important difference (MCID) of the pain VAS;15 16 that is, what minimal change in a pain VAS score would indicate a real change in a patient’s pain intensity.16 17 Several studies have attempted to define the MCID or clinically useful effect in the postoperative setting,18–20 but the methods used did not comply with existing standards nor did they include patient evaluation.15 The clinically important difference of the numerical rating scale (NRS) has been estimated for various chronic pain states,13 15 21 22 as has the MCID of the VAS in chronic pain23 and in the emergency department setting,24 25 but it is unclear whether these results can be applied in the acute postsurgical pain setting. One study has determined the MCID of the pain VAS in patients after shoulder rotator cuff surgery.26 The Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) recently reviewed and recommended specific methods that can be used for interpreting the clinical importance of treatment outcomes in chronic pain trials,14 but there are currently no recommendations for acute postoperative pain. The patient acceptable symptom state (PASS) is the value beyond which patients consider themselves well.15 27 28 The PASS can therefore be used to define responders and nonresponders to analgesic treatment in postoperative pain studies.21 The PASS of the pain VAS score in this setting is often assumed, but has not been determined according to current recommendations.15 The aim of this study was to determine the MCID and PASS for the pain VAS in patients recovering from surgery.
There is currently no consensus on the optimal method for MCID estimation. As such, we chose to include a triangulation (average) of several methodologies,15 16 as we have done previously when evaluating quality-of-recovery scales.29 Previous methods used to determine MCID have included the SD/2 rule,30 and 0.2 SD,31 0.3 SD,32 and the standard error of measurement (SEM);33 others have used 5–10% of the instrument range.34 We chose to include three distribution-based measures: the 0.3 SD, 35 SEM, and 5% range. In addition, we used an anchor-based method with a global rating-of-change questionnaire.36 This uses a 15-point Likert scale ranging from –7 (a very great deal worse) to þ7 (a very great deal better).15 16 36 A 100 mm VAS, ranging from 0 (no pain) to 100 (very severe pain), was used to measure pain intensity throughout the previous 24 h on two occasions in the days after surgery. At the second visit, patients were asked to assess the following (adapted from Tubach and colleagues).15 ‘Think only about your pain you have felt over the past 24 hours. Compared with yesterday, is your pain’: 7, a very great deal worse; 6, a great deal worse; 5, a good deal worse; 4, moderately worse; 3, somewhat worse; 2, a little worse; 1, almost the same, hardly any worse at all; 0, no change; 1, almost the same, hardly any better at all; 2, a little better; 3, somewhat better; 4, moderately better; 5, a good deal better; 6, great deal better; or 7, a very great deal better? Patients whose score on the global rating-of-change questionnaire was 0, 1, or 1 were classified as unchanged.36 Patients whose score was 2, 3, 2, or 3 were considered to have experienced a small change equivalent to the MCID; those with scores of 4, 5, 4, and 5 were considered to have experienced moderate change, and those with scores of 6, 7, 6, and 7 were considered to have experienced large change.36 Absolute (i.e. we changed the sign of the scores for those who deteriorated) mean changes in pain VAS scores according to patient-rated change in postoperative recovery health status were then calculated. All four estimates (0.3 SD, SEM, 5% range, and global rating of change) were then averaged.
Methods
Patients’ opinion of their improvement
This prospective observational study evaluated adult patients recovering from surgery, using the 100 mm VAS to measure pain on two occasions, along with a generic Likert scale of overall recovery (see next subsection). Most patients (n¼204) enrolled in this study participated in a concurrent study evaluating quality-of-recovery scales.29 The study settings were the surgical wards at three hospitals in Australia (Alfred, Royal Women’s, and Shepparton Hospitals) representing tertiary adult, tertiary obstetric/gynaecology, and rural/regional hospitals. Ethics approval was obtained from the institutional ethics committee at each hospital, and patient consent was obtained in all instances. Patients were eligible to participate in the study if they were >18 yr of age and recovering from a surgical procedure requiring general or major neuraxial block anaesthesia. Patients were excluded if they had poor English comprehension, drug or alcohol dependence, psychiatric disorder, uncontrolled pain, or a concurrent serious medical disorder impairing completion of the VAS and questionnaire. Baseline patient characteristics and
The PASS was determined using the direct opinion-based approach,15 37 in which patients were asked to define any improvement: ‘In your opinion, have you made a good recovery from your operation?’, with response options of yes, no, or unsure. Patients responding in the affirmative were classified as having made a good recovery, and those who responded negatively or were unsure were classified as having a poor recovery.
Statistical analysis We could not reliably estimate a required sample size for this study. Given that previous relevant studies had enrolled 40–100 subjects,20 36 we planned to enrol at least 150 subjects to provide an adequate number for subgroup testing. Data are presented as mean (SD) or number (%) unless otherwise specified. Selected results are reported with a 95% confidence interval (CI), with the mean (95% CI) VAS score for minimal change calculated using 1000 bootstrap samples. The SEM was calculated as the SD multiplied by the square root of one minus the intraclass correlation coefficient.33 Changes in pain
426
|
Myles et al.
scores at each interview were compared with the Wilcoxon signed rank test, and according to extent of surgery using the Kruskal–Wallis test. Both Student’s unpaired t-test and independent-samples median test were used to compare pain scores in those with a good or poor recovery. We defined the PASS as the 25th centile of the pain VAS in those who rated their recovery as good at the second postoperative interview. The predictive utility of the pain VAS in achieving a good recovery was assessed using a receiver operating characteristic curve (ROC).38 The association between the change in pain VAS score and the global rating-of-change score was quantified by the Spearman rank correlation coefficient (r). We determined test–retest reliability in those who had rated their pain state as unchanged or almost the same (global rating scores of 1, 0 or 1) using the intraclass correlation coefficient. Responsiveness was measured in those with global rating-of-change scores of at least 4 using standardized response means, calculated as the mean change divided by its SD.39 40 All statistical analyses were performed using SPSS for Windows V23.0 (SPSS Australasia Ltd, Sydney, NSW, Australia). A P-value of <0.05 was considered significant; no correction was made for multiple comparisons.
Results We enrolled 224 patients with varying degrees of postoperative pain and recovering from a broad range of surgical procedures; there were no refusals, but five were unavailable at the second postoperative visit (Table 1). Patients undergoing more extensive surgery tended to have higher VAS scores (Table 2). Most patients improved their pain score ratings at the second postoperative visit (Table 3). The median pain VAS scores reduced from 26 (13–47) to 20 (11–36), P¼0.002. The distributionbased estimates of the MCID for the VAS score were 6.9 (0.3 SD), 4.7 (5% range), and 12.4 (SEM). The association between the change in pain VAS score and the global rating-of-change score was r ¼ 0.48, P<0.0005. The absolute mean changes (95% CI) in VAS scores according to patient-rated change in postoperative pain intensity are reported in Table 4. The anchor-based estimate of the MCID for the pain VAS was 15.7. This results in an averaged MCID of 9.9. Those who rated themselves as having had a poor postoperative recovery [n¼22 (11%)] had significantly lower pain VAS scores at the second postoperative visit compared with those who reported a good postoperative recovery [n¼177 (89%)]; see Table 5 and Fig. 1. The 75th centile for the pain VAS score in those with a good recovery was 33; the 25th centile for the pain VAS score in those with a poor recovery was 20. The area under the ROC curve was 0.72 (95% CI: 0.60–0.85). Test–retest reliability (n¼22) of the pain VAS was high, intraclass correlation coefficient 0.79 (95% CI: 0.49–0.91). The standardized response mean (n¼125) of the pain VAS was 0.36, indicating a mild-to-moderate degree of responsiveness to change.
Discussion We studied a broad range of surgical and obstetric patients to quantify the MCID and PASS for the pain VAS as a measure of pain intensity in the postoperative setting. These findings can guide the conduct and interpretation of studies evaluating analgesic therapies in surgical patients. Each can be used to define a ‘responder’ (effective analgesia) as a standard metric to be used across acute pain studies.
Table 1 Patient and surgical characteristics (n¼224). Number (%) or median (interquartile range), unless otherwise stated. *New Zealand, South Africa, UK, or USA Variable Age (yr) Mean (SD) Range Female sex White-collar worker Country of birth Australia Other English-language country* Non-English-speaking country ASA physical status I II III IV Extent of surgery Ambulatory or other minor Intermediate Major Type of surgery General Orthopaedic Gynaecology Caesarean section Urological Plastics Major vascular Ear, nose, throat, or faciomaxillary Cardiac Other Duration of surgery (min) Timing of assessment after surgery (h) Visit 1 Visit 2 (n¼218)
53 (17) 19–86 148 (66) 95 (42) 160 (71) 30 (13) 34 (15) 47 (21) 112 (50) 56 (25) 9 (4.0) 17 (7.6) 103 (46) 104 (46) 51 (23) 44 (20) 36 (16) 37 (17) 11 (4.9) 11 (4.9) 6 (2.7) 5 (2.2) 5 (2.2) 18 (8.1) 112 (58–161) 22 (9.5–26) 43 (31–48)
It is generally accepted that a pain VAS score of 30, 70, and 100 indicates the upper boundaries of mild, moderate, and severe pain intensity. Several studies have found that a pain scale PASS value of 40 can be used as an outcome criterion for osteoarthritis and various chronic rheumatic diseases with pain.41 27 Pooled data from two postoperative analgesic studies found that pain VAS scores of up to 44 are consistent with a patient rating of pain being mild.42 Current guidelines recommend titrating analgesia to achieve a pain VAS score of 40 or less,43 and postoperative pain studies using patient-controlled analgesia typically titrate to a VAS or NRS score of 30.44 Our PASS estimate of 33 for the pain VAS is consistent with these results. Clinical trials ought to focus more strongly on the patient’s response to pain treatments; that is, the proportion of patients who are responders to treatment.14 Clearly, giving sufficient analgesia to eliminate all pain for all patients is a wrong target,45 but so is undertreatment. We thus recommend a pain VAS score of 33 be used as a suitable target for optimal analgesic titration. As outlined by Moore and colleagues,18 a dichotomous pain outcome (‘responder’ and ‘non-responder’) allows for the estimation of clinically useful statistics, such as absolute and relative risk reduction, and number needed to treat. This can then provide more useful information for clinicians and patients.
Measurement of acute postoperative pain
Table 2 Summary statistics for pain visual analog scale scores at the first postoperative visit, expressed as mean (SD). VAS, visual analog scale. *Kruskal–Wallis test Extent of surgery Pain VAS scale (range of scores)
Minor Intermediate Major (n¼17) (n¼103) (n¼104)
VAS score (0–100) 19 (24)
29 (23)
34 (22)
|
427
Table 5 Comparison of the pain visual analog scale scores in those who rated their recovery at the second postoperative visit, expressed as mean (SD) and median (interquartile range). CI, confidence interval; IQR, interquartile range; VAS, visual analog scale. *Levene’s test P¼0.01, so equal variances not assumed. †Mann–Whitney U-test
P-value Patient-rated good recovery 0.008*
Pain intensity Yes (n¼177) Pain VAS scale 24 (20) Median (IQR) 20 (10–33)
No/Unsure (n¼22) 45 (27) 51 (20–63)
Difference (95% CI) P-value 21 (8.2–33) —
0.002* 0.001†
Table 3 The frequency of each level of patient-reported change in pain intensity between the first and second postoperative visits (n¼219) n (%)
A very great deal worse A great deal worse A good deal worse Moderately worse Somewhat worse A little worse Almost the same, hardly any worse at all No change Almost the same, hardly any better at all A little better Somewhat better Moderately better A good deal better A great deal better A very great deal better
4 (1.8) 4 (1.8) 2 (0.9) 4 (1.8) 4 (1.8) 11 (5.0) 8 (3.7) 23 (11) 5 (2.3) 23 (11) 20 (9.1) 30 (14) 41 (19) 28 (13) 12 (5.5)
100 80 VAS score
Extent of change
60 40 20 0 Yes
No/unsure Good recovery
Fig 1 Pain visual analog scale (VAS) scores in those who rated their recovery at the second postoperative visit. Box plots represent median (interquartile range), with the whiskers representing 10th and 90th centiles.
Table 4 Mean change in pain visual analog scale scores according to patient-rated change in postoperative pain intensity (n¼219). CI, confidence interval; VAS, visual analog scale. *Used to define the anchor-based estimate of the minimal clinically important change (see main text for details) Change categories
Mean change
95% CI
(A) According to change categories of improvement or deterioration Large improvement (n¼40) Moderate improvement (n¼71) Minimal improvement (n¼43) No change (n¼36) Minimal deterioration (n¼15) Moderate deterioration (n¼6) Large deterioration (n¼8)
14.8 11.0 6.7 0.8 19.1 10.3 34.5
8.0 to 21.6 6.8 to 15.1 1.3 to 12.0 5.8 to 4.1 26.4 to 11.7 20.2 to 0.5 56.9 to 12.1
(B) Absolute change categories No change (n¼36) Minimal change (n¼58)* Moderate change (n¼77) Large change (n¼48)
0.8 15.7 15.1 20.1
5.8 to 4.1 12.9 to 18.8 12.3 to 18.1 14.5 to 26.4
Anchor-based methods have been used to determine the MCID for scales used in many medical conditions,16 20 26 32 35 36 and we have recently done this for several quality-of-recovery scales.29 Our approach, using a combination of distribution- and anchor-based methods, suggests that the MCID for pain in a surgical population is smaller than previously assumed. Studies of patients presenting to emergency departments with various pain states found an MCID for the VAS score of 1224 and 13,25 and a change of 18 correlated with the patient being a ‘little bit better’.19 Slightly higher values have been found for the pain NRS in chronic pain conditions, with an MCID of 2 (comparable to VAS 20 mm).13 21 22 The MCID for the pain VAS was 14 in patients treated for shoulder rotator cuff disease,26 and ranged from 4 to 21 in patients with degenerative cervical spine disease undergoing surgery, but in the latter study the evaluation was done by the clinician (not the patient).20 For patients recovering from minimally invasive spine surgery, an MCID for the VAS was 12 for back pain and 16 for leg pain.46 Others have found that a change in VAS of 20 represented satisfactory pain relief in patients after surgery.5 11 Some authors have suggested that a 33% decrease,42 or a change in VAS score of 15–20,5 6 47 would indicate a clinically important change in pain intensity. Taken
428
|
Myles et al.
together, these results largely support our estimate of the MCID for the pain VAS being 10 in the postoperative setting. Although the VAS and NRS are incomplete representations of the pain experience and cannot fully reflect the multidimensional aspects of pain, they remain the most widely used metrics of pain after surgery. In view of the high correlations between the VAS and NRS reported in many studies, it is reasonable to accept that the MCID can be applied to either instrument. The MCID can be used to guide the non-inferiority margin for clinical trials48 and for determining the sample size in a clinical trial.49 This study has limitations. Pain is both subjective and multidimensional and so the VAS (and NRS) cannot capture the complete pain experience. But clinical decisions are made on the basis of existing pain scales, and so it is important to know how much reduction in a VAS score is likely to be clinically meaningful from the patient’s perspective. The extremes of pain indicated on a VAS, typically ‘no pain’ and ‘worst pain ever’, may not truly represent absolute limits of perception. Our data (Table 4A) suggest that there is a difference in the MCID for those who have improvement or deterioration in their pain status, in that a larger reduction in a VAS score might be needed to indicate true improvement. A similar finding has been previously reported50 and is consistent with studies of loss aversion. It is possible that the experience of change/improvement in pain is non-linear. A reduction of pain from 95 to 75 could indicate greater relief than a reduction of 30 to 10 (particularly as this is below the identified PASS). We determined the PASS by asking patients to rate their overall recovery after surgery; although pain is an important component of this, other factors will, of course, influence recovery. In conclusion, we have shown that in patients with acute pain after surgery, including women recovering from Caesarean section, the MCID and PASS for the pain VAS score is 9.9 (rounded to 10) and 33, respectively.
Authors’ contributions Study concept and protocol development, analysis of results, first draft and revision of manuscript: P.M., D.M. Patient consent and enrolment, data collection: Ethics committee application, liaison with obstetric staff, and oversight of the conduct of the study at the Royal Women’s Hospital: A.D., N.M. Critical review and revisions of the manuscript: P.M., D.M., W.G., D.B., C.C., N.M., A.D.
Funding Australian National Health and Medical Research Council Practitioner Fellowship (APP1042462; Canberra, ACT, Australia; to P.M.).
References 1.
2.
3.
4.
5.
6. 7.
8.
9. 10.
11.
12.
13.
14.
Acknowledgements We thank the study participants willing to contribute to a better understanding of the measurement of pain, and the numerous clinicians providing postoperative care and pain relief in these settings.
15.
16.
Declaration of interest P.M. is an editor of the BJA; none of the other authors report any competing interests relating to the topic of this paper.
17.
Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain 1986; 27: 117–26 Revill SI, Robinson JO, Rosen M, Hogg MI. The reliability of a linear analogue for evaluating pain. Anaesthesia 1976; 31: 1191–8 Downie WW, Leatham PA, Rhind VM, Wright V, Branco JA, Anderson JA. Studies with pain rating scales. Ann Rheum Dis 1978; 37: 378–81 Price DD, Bush FM, Long S, Harkins SW. A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 1994; 56: 217–26 DeLoach LJ, Higgins MS, Caplan AB, Stiff JL. The visual analog scale in the immediate postoperative period: intrasubject variability and correlation with a numeric scale. Anesth Analg 1998; 86: 102–6 Campbell WI, Patterson CC. Quantifying meaningful changes in pain. Anaesthesia 1998; 53: 121–5 Myles PS, Urquhart N. The linearity of the visual analogue scale in patients with severe acute pain. Anaesth Intensive Care 2005; 33: 54–8 Bodian CA, Freedman G, Hossain S, Eisenkraft JB, Beilin Y. The visual analog scale for pain: clinical significance in postoperative patients. Anesthesiology 2001; 95: 1356–61 Myles PS, Christelis N. Measuring pain and analgesic response. Eur J Anaesthesiol 2011; 28: 399–400 Weibel S, Jokinen J, Pace NL, et al. Efficacy and safety of intravenous lidocaine for postoperative analgesia and recovery after surgery: a systematic review with trial sequential analysis. Br J Anaesth 2016; 116: 770–83 Aubrun F, Langeron O, Quesnel C, Coriat P, Riou B. Relationships between measurement of pain using visual analog score and morphine requirements during postoperative intravenous morphine titration. Anesthesiology 2003; 98: 1415–21 Cepeda MS, Africano JM, Polo R, Alcala R, Carr DB. What decline in pain intensity is meaningful to patients with acute pain? Pain 2003; 105: 151–7 Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Defining the clinically important difference in pain outcome measures. Pain 2000; 88: 287–94 Dworkin RH, Turk DC, McDermott MP, et al. Interpreting the clinical importance of group differences in chronic pain clinical trials: IMMPACT recommendations. Pain 2009; 146: 238–44 Tubach F, Wells GA, Ravaud P, Dougados M. Minimal clinically important difference, low disease activity state, and patient acceptable symptom state: methodological issues. J Rheumatol 2005; 32: 2025–9 Wells G, Beaton D, Shea B, et al. Minimal clinically important differences: review of methods. J Rheumatol 2001; 28: 406–12 Guyatt GH, Osoba D, Wu AW, Wyrwich KW, Norman GR; Clinical Significance Consensus Meeting Group. Methods to explain the clinical significance of health status measures. Mayo Clin Proc 2002; 77: 371–83
Measurement of acute postoperative pain
18. Moore A, Moore O, McQuay H, Gavaghan D. Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesics: use of pain intensity and visual analogue scales. Pain 1997; 69: 311–5 19. Todd KH, Funk JP. The minimum clinically important difference in physician-assigned visual analog pain scores. Acad Emerg Med 1996; 3: 142–6 20. Auffinger B, Lam S, Shen J, Roitberg BZ. Measuring surgical outcomes in subaxial degenerative cervical spine disease patients: minimum clinically important difference as a tool for determining meaningful clinical improvement. Neurosurgery 2014; 74: 206–13 21. Farrar JT, Young JP, Jr, LaMoreaux L, Werth JL, Poole RM. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 2001; 94: 149–58 22. Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine (Phila Pa 1976) 2005; 30: 1331–4 23. Emshoff R, Bertram S, Emshoff I. Clinically important difference thresholds of the visual analog scale: a conceptual model for identifying meaningful intraindividual changes for pain intensity. Pain 2011; 152: 2277–82 24. Kelly AM. The minimum clinically significant difference in visual analogue scale pain score does not differ with severity of pain. Emerg Med J 2001; 18: 205–7 25. Gallagher EJ, Liebman M, Bijur PE. Prospective validation of clinically important changes in pain severity measured on a visual analog scale. Ann Emerg Med 2001; 38: 633–8 26. Tashjian RZ, Deloach J, Porucznik CA, Powell AP. Minimal clinically important differences (MCID) and patient acceptable symptomatic state (PASS) for visual analog scales (VAS) measuring pain in patients treated for rotator cuff disease. J Shoulder Elbow Surg 2009; 18: 927–32 27. Tubach F, Ravaud P, Martin-Mola E, et al. Minimum clinically important improvement and patient acceptable symptom state in pain and function in rheumatoid arthritis, ankylosing spondylitis, chronic back pain, hand osteoarthritis, and hip and knee osteoarthritis: results from a prospective multinational study. Arthritis Care Res (Hoboken) 2012; 64: 1699–707 28. Wariaghli G, Allali F, Berrada K, et al. The patient acceptable symptom state of chronic musculoskeletal pain measured on a visual analog scale in Moroccan patients. Pain Med 2013; 14: 103–9 29. Myles PS, Myles DB, Galagher W, Chew C, MacDonald N, Dennis A. Minimal clinically important difference for three quality of recovery scales. Anesthesiology 2016; 125: 39–45 30. Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation. Med Care 2003; 41: 582–92 31. Samsa G, Edelman D, Rothman ML, Williams GR, Lipscomb J, Matchar D. Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II. Pharmacoeconomics 1999; 15: 141–55 32. Eton DT, Cella D, Yost KJ, et al. A combination of distributionand anchor-based approaches determined minimally
33.
34.
35.
36.
37.
38. 39. 40. 41.
42.
43.
44.
45. 46.
47. 48.
49.
50.
|
429
important differences (MIDs) for four endpoints in a breast cancer scale. J Clin Epidemiol 2004; 57: 898–910 Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life. J Clin Epidemiol 1999; 52: 861–73 Ringash J, O’Sullivan B, Bezjak A, Redelmeier DA. Interpreting clinically significant changes in patientreported outcomes. Cancer 2007; 110: 196–202 Maringwa J, Quinten C, King M, et al. Minimal clinically meaningful differences for the EORTC QLQ-C30 and EORTC QLQ-BN20 scales in brain cancer patients. Ann Oncol 2011; 22: 2107–12 Juniper EF, Guyatt GH, Willan A, Griffith LE. Determining a minimal important change in a disease-specific quality of life questionnaire. J Clin Epidemiol 1994; 47: 81–7 Wells G, Anderson J, Boers M, et al. MCID/Low Disease Activity State Workshop: summary, recommendations, and research agenda. J Rheumatol 2003; 30: 1115–8 Sedgwick P. How to read a receiver operating characteristic curve. Br Med J 2015; 350: h2464 Kirshner B, Guyatt G. A methodological framework for assessing health indices. J Chronic Dis 1985; 38: 27–36 Kazis LE, Anderson JJ, Meenan RF. Effect sizes for intepreting changes in health status. Med Care 1989; 27: S178–89 Perrot S, Bertin P. “Feeling better” or “feeling well” in usual care of hip and knee osteoarthritis pain: determination of cutoff points for patient acceptable symptom state (PASS) and minimal clinically important improvement (MCII) at rest and on movement in a national multicenter cohort study of 2414 patients with painful osteoarthritis. Pain 2013; 154: 248–56 Jensen MP, Chen C, Brugger AM. Interpretation of visual analog scale ratings and change scores: a reanalysis of two clinical trials of postoperative pain. J Pain 2003; 4: 407–14 Hartrick CT, Kovan JP, Shapiro S. The numeric rating scale for clinical pain measurement: a ratio measure? Pain Pract 2003; 3: 310–6 Bainbridge D, Martin JE, Cheng DC. Patient-controlled versus nurse-controlled analgesia after cardiac surgery – a metaanalysis. Can J Anaesth 2006; 53: 492–9 Lee TH. Zero pain is not the goal. JAMA 2016; 315: 1575–7 Park P, Okonkwo DO, Nguyen S, et al. Can a minimal clinically important difference be achieved in elderly patients with adult spinal deformity who undergo minimally invasive spinal surgery? World Neurosurg 2016; 86: 168–72 Myles PS. The pain visual analog scale: linear or nonlinear? Anesthesiology 2004; 100: 744. Committee for Medicinal Products for Human Use: Guideline on the Choice of the Non-Inferiority Margin. London. 2005. Available from http://wwwemaeuropaeu/docs/en_GB/docu ment_library/Scientific_guideline/2009/09/WC500003636pdf (accessed 1 July 2016) Gibbs NM, Weightman WM. Beyond effect size: consideration of the minimum effect size of interest in anesthesia trials. Anesth Analg 2012; 114: 471–5 Sloman R, Wruble AW, Rosen G, Rom M. Determination of clinically meaningful levels of pain reduction in patients experiencing acute postoperative pain. Pain Manag Nurs 2006; 7: 153–8 Handling editor: Takashi Asai