- Email: [email protected]
Determining the minimal clinically important difference for the American Shoulder and Elbow Surgeons score, Simple Shoulder Test, and visual analog scale (VAS) measuring pain after shoulder arthroplasty
ARTICLE IN PRESS J Shoulder Elbow Surg (2016) ■■, ■■–■■
www.elsevier.com/locate/ymse
ORIGINAL ARTICLE
Determining the minimal clinically important difference for the American Shoulder and Elbow Surgeons score, Simple Shoulder Test, and visual analog scale measuring pain after shoulder arthroplasty Robert Z. Tashjian, MDa,*, Man Hung, PhDa,b, Jay D. Keener, MDc, Randy Christopher Bowen, MD, MSa, Jared McAllister, MDc, Wei Chen, PhDb, Gregory Ebersole, MSc, Erin K. Granger, MPHa, Aaron M. Chamberlain, MDc a
Department of Orthopaedics, University of Utah School of Medicine, Salt Lake City, UT, USA Division of Epidemiology, University of Utah School of Medicine, Salt Lake City, UT, USA c Department of Orthopaedics, Washington University, St. Louis, MO, USA b
Background: Minimal clinically important differences (MCIDs) for the American Shoulder and Elbow Surgeons (ASES) score, the Simple Shoulder Test (SST), and a visual analog scale (VAS) measuring pain have not been previously described using an anchor-based method after shoulder arthroplasty. The purpose of this study was to determine the MCIDs for these measures after shoulder arthroplasty for glenohumeral arthritis or advanced rotator cuff disease. Methods: Primary anatomic total shoulder arthroplasty (TSA), primary reverse TSA, or hemiarthroplasty was performed in 326 patients by 1 of 5 shoulder and elbow surgeons. The SST score, ASES score, and VAS pain score were collected preoperatively and at a minimum of 2 years postoperatively (mean, 3.5 years). The MCIDs were calculated for the ASES score, SST score, and VAS pain score using an anchor-based method. Results: The MCIDs for the ASES score, SST score, and VAS pain score were 20.9 (P < .001), 2.4 (P < .0001), and 1.4 (P = .0158), respectively. Duration of follow-up and type of arthroplasty (anatomic TSA vs reverse TSA) did not have a significant effect on the MCIDs (P > .1) except shorter follow-up correlated with a larger MCID for the ASES score (P = .0081). Younger age correlated with larger MCIDs for all scores (P < .024). Female sex correlated with larger MCIDs for the VAS pain score (P = .123) and ASES score (P = .05). Conclusions: Patients treated with a shoulder arthroplasty require a 1.4-point improvement in the VAS pain score, a 2.4-point improvement in the SST score, and a 21-point improvement in the ASES score to achieve a minimal clinical importance difference from the procedure. Level of evidence: Basic Science Study; Validation of Outcomes Instruments © 2016 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved.
Institutional Review Board approval was obtained from the University of Utah School of Medicine (IRB No. 00065636). *Reprint requests: Robert Z. Tashjian, MD, Department of Orthopaedics, University of Utah School of Medicine, 590 Wakara Way, Salt Lake City, UT 84108, USA. E-mail address: [email protected] (R.Z. Tashjian). 1058-2746/$ - see front matter © 2016 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. http://dx.doi.org/10.1016/j.jse.2016.06.007
ARTICLE IN PRESS 2
R.Z. Tashjian et al. Keywords: Shoulder arthroplasty; minimal clinically important difference; outcome scores; pain; function; Simple Shoulder Test; ASES score
Total shoulder arthroplasties (TSAs)—both anatomic TSA and reverse total shoulder arthroplasty (RSA)—are reliable surgical treatment options for patients with shoulder osteoarthritis and advanced rotator cuff disease to reduce pain and restore function. Various patient-reported functional outcome scores have been used to assess outcomes after anatomic TSA and RSA including the American Shoulder and Elbow Surgeons (ASES) score and Simple Shoulder Test (SST).4,5,15 Previous studies have shown the ASES score and SST are valid, reliable, and responsive outcome measures.2,7,13 Responsiveness refers to the ability of a measure to detect a change once it has occurred. The minimal clinically important difference (MCID) is a measure of responsiveness representing the smallest subjective difference in an outcome score that represents a clinically important change to the patient.8 MCID values have an important role in maximizing the utility of outcome measures because they allow interpretation of research comparing different treatments. MCIDs also allow analysis of the amount of improvement that an individual patient achieves after treatment and assist in study design because they provide a measure of effect size. Among the various methods used to determine MCIDs, distribution-based and anchor-based methods are the most commonly used. Distribution-based methods are based on the statistical characteristics of the sample being investigated and include the effect size, standard response mean, and standard error of measurement.9,22 Anchor-based methods use an external question, or “anchor,” to determine if a change that occurs reaches clinical importance.3 Currently, there is no gold standard regarding which method is optimal for determining MCIDs, with each having its own benefits and limitations. Distribution-based methods are easy to calculate and require no further data beyond the questionnaires. Limitations are that the results may not truly reflect the patient’s perception of clinical importance. Anchor-based methods are more difficult to perform and may be limited by the anchor questions created, although they may better capture a true assessment of clinical importance. Limited studies exist on MCIDs for the ASES score and SST. MCIDs have been previously determined for the ASES score and SST in patients treated for global shoulder pain and for each measure in patients undergoing nonoperative treatment of rotator cuff disease.13,18 Both studies used an anchorbased method to determine MCIDs. The MCID for the ASES score was 6.4 points as reported by Michener et al13 and between 12 and 17 points as reported by Tashjian et al.18 The MCID for the SST was approximately 2 points as reported by Tashjian et al.18 Roy et al17 used a distribution-based method to determine the MCID for the SST after shoulder arthroplasty and reported the MCID as 3 points. Angst et al1 used a distribution-based method to determine MCIDs for the ASES
score after shoulder arthroplasty. They also performed an anchor-based assessment of responsiveness, but MCIDs were not reported using the anchor-based data. Finally, Tashjian et al19 reported the MCID for a visual analog scale (VAS) for pain was 1.4 points after nonoperative treatment of rotator cuff disease. No prior studies have determined the MCID for the ASES score or VAS score for pain after shoulder arthroplasty using anchor-based methods. The purpose of this study was to determine the MCID for the ASES score, SST score, and VAS pain score after shoulder arthroplasty using an anchor-based approach. The effect of patient factors (age, sex) and treatment factors (duration of follow-up, arthroplasty type [RSA or anatomic TSA]) on the MCIDs was also evaluated.
Methods Three hundred twenty-six patients who underwent a primary anatomic TSA, RSA, or hemiarthroplasty by 1 of 5 shoulder and elbow surgeons over a 7-year period from 2005 to 2012 were included in the study, retrospectively reviewed, and evaluated at a minimum of 2 years postoperatively. The patients were recruited from 2 medical centers, the University of Utah School of Medicine and Washington University School of Medicine. Patient records were reviewed, and patients were identified as having either a preoperative ASES or SST score available in their records before arthroplasty. Only the patient-reported questionnaire section of the ASES form was recorded. An attempt was made to contact all patients with either a preoperative SST or ASES score through a letter and then a phone call at a minimum of 2 years’ follow-up after surgery. Patients were recruited into the study if they completed the follow-up questionnaire by phone. A total of 1900 shoulder arthroplasties were performed at Washington University and 216 at the University of Utah during the study period. A total of 509 patients met the inclusion criteria of having either a preoperative ASES or SST score available and having undergone an anatomic TSA, RSA, or hemiarthroplasty at least 2 years previously. A total of 363 patients were recruited (72% followup). Only patients with a diagnosis of osteoarthritis, rheumatoid arthritis, rotator cuff arthropathy, or advanced rotator cuff disease were included in the final analysis, resulting in a total of 326 patients in the final dataset. There were 198 anatomic TSAs, 124 RSAs, and 4 hemiarthroplasties. The diagnoses for which arthroplasty was performed included rotator cuff arthropathy (103), irreparable rotator cuff tears with pseudoparalysis (17), osteoarthritis with fullthickness rotator cuff tears (5), rheumatoid arthritis (5), and osteoarthritis (196). There were 184 female and 142 male patients. The mean age of the patients at the time of surgery was 67.4 years (SD, 9.4 years). Patients completed a questionnaire before surgery including the SST, ASES score, and VAS for pain. The SST is a shoulder region– specific outcome measure that focuses on characterizing function in patients with abnormal shoulder pathology. It is composed of 12
ARTICLE IN PRESS MCID after shoulder arthroplasty yes/no questions representing important activities of daily living that can be performed by normally functioning shoulders. The questions specifically assess shoulder comfort and function.12 The ASES score is a patient-assessed region-specific outcome measure that evaluates pain on a VAS and function by 10 Likert-style questions.16 Patients are asked to rate their current shoulder pain on a VAS assessing shoulder pain levels in 1-digit increments from 0, “none,” to 10, “disabling.” The patients were re-evaluated with a follow-up questionnaire by phone at a minimum of 2 years (mean, 3.5 years; SD, 1.6 years) postoperatively. The follow-up questionnaire included the ASES score, the SST, the VAS for pain, and a 4-item anchor question evaluating improvement after treatment. The improvement question was the anchor question used to determine the MCIDs. This anchor question was derived from the anchor question designed by Tubach et al21 and has been previously used to determine MCIDs by Tashjian et al.18,19 Patients were asked the following: “Since your shoulder replacement surgery, please rate your response to treatment: A, none—no good at all, ineffective treatment; B, poor—some effect but unsatisfactory; C, good—satisfactory effect with occasional episodes of pain or stiffness; D, excellent—ideal response, virtually pain free.”
Statistical analysis MCIDs for each outcome variable (SST score, ASES score, VAS pain score) were calculated. First, the change scores for each variable were calculated from preoperative assessment to final postoperative assessment. Patients were classified by the anchor question as having “no change” (A group [none] and B group [poor] combined) or “change” (C group [good]). The D group (excellent) was not included in the analysis because this was considered beyond minimal change. We then performed t tests to compare the means between the unchanged and changed groups. The MCID is the mean difference between the changed and unchanged groups. Correlations between MCIDs and age and between MCIDs and duration of follow-up were determined using Spearman correlation coefficients. Correlations between MCIDs and sex and between MCIDs and implant type (anatomic TSA vs RSA) were determined using univariate regression analysis. Separate analysis of correlations between MCIDs and disease etiology was not performed due to a majority of the TSAs being performed for osteoarthritis without a rotator cuff tear and a majority of the RSAs being performed for rotator cuff arthropathy. Consequently, the correlations between MCIDs and implant type closely reflect the correlations between MCIDs and disease etiology (RSA for rotator cuff arthropathy and TSA for primary osteoarthritis without a rotator cuff tear). P < .05 was considered statistically significant. Spearman correlations were interpreted based on the results as follows: 0.0 to 0.19, very weak; 0.2 to 0.39, weak; 0.4 to 0.59, moderate; 0.6 to 0.79, strong; and 0.8 to 1.0, very strong. All of these analyses were performed with SAS software (version 9.4; SAS Institute, Cary, NC, USA).
Results Of the patients, 112 rated their response as C (good) and were therefore classified as the changed group whereas 21 rated their response as A (none) or B (poor) and were classified as the unchanged group. The MCIDs for the ASES score, SST
3 Table I Average preoperative and postoperative outcome scores and improvements in scores for all patients, RSA patients, and anatomic TSA patients Mean Overall SST Preoperative Postoperative Change VAS Preoperative Postoperative Change ASES Preoperative Postoperative Change TSA SST Preoperative Postoperative Change VAS Preoperative Postoperative Change ASES Preoperative Postoperative Change RSA SST Preoperative Postoperative Change VAS Preoperative Postoperative Change ASES Preoperative Postoperative Change
SD
P value
2.8 8.8 6.0
2.3 2.8 2.7
< .01
7.1 1.4 −5.6
2.0 2.2 2.8
< .01
31.6 79.6 47.9
14.2 20.5 22.3
< .01
3.4 9.5 6.1
2.5 2.3 2.6
< .01
7.1 1.2 −5.9
2.0 2.0 2.7
< .01
33.3 84.3 50.9
14.2 18.7 20.9
< .01
2.0 7.8 5.8
1.6 3.2 3.0
< .01
6.9 1.8 −5.1
2.0 2.2 2.9
< .01
29.8 72.7 43.9
13.7 20.8 23.3
< .01
ASES, American Shoulder and Elbow Surgeons; RSA, reverse total shoulder arthroplasty; SST, Simple Shoulder Test; TSA, total shoulder arthroplasty; VAS, visual analog scale.
score, and VAS score evaluating pain were 20.9 (P < .001), 2.4 (P < .0001), and 1.4 (P = .0158), respectively. Average preoperative and postoperative outcome scores and improvements in scores for all patients, RSA-only patients, and anatomic TSA–only patients are presented in Table I. Correlations between duration of follow-up and MCIDs were evaluated and determined to not have an effect on the MCID for the VAS pain score (r = 0.14, P = .1) or SST score (r = 0.1, P = .21). Shorter duration of follow-up was correlated
ARTICLE IN PRESS 4 with a larger MCID for the ASES score (r = 0.23, P = .01), although this correlation was considered weak. Correlations between age and MCIDs were evaluated, and younger age was correlated with larger MCIDs for all outcome scores (VAS pain score: r = 0.2, P = .02; ASES score: r = 0.28, P = .001; SST score: r = 0.25, P = .004), although all correlations were considered weak. Correlations between sex and MCIDs were evaluated using univariate regressions; no correlation was identified for the SST score (P = .89), whereas female sex was correlated with higher MCIDs for the VAS pain score (P = .01) and ASES score (P = .05). The average MCIDs for male patients for the VAS pain score, ASES score, and SST score were 1.3 (SD, 2.2), 17.9 (SD, 15.9), and 1.2 (SD, 2.5), respectively. The average MCIDs for female patients for the VAS pain score, ASES score, and SST score were 1.3 (SD, 2.5), 22.4 (SD, 17.8), and 3.2 (SD, 2.4), respectively. Correlations between implant type (anatomic TSA vs RSA) and MCIDs were evaluated using univariate regressions, and no correlation was identified for the VAS pain score (P = .18), ASES score (P = .1), or SST score (P = .1). The average MCIDs for anatomic TSA patients for the VAS pain score, ASES score, and SST score were 2.1 (SD, 2.5), 24.5 (SD, 17.2), and 1.5 (SD, 2.3), respectively. The average MCIDs for RSA patients for the VAS pain score, ASES score, and SST score were 0.5 (SD, 2.3), 15.4 (SD, 16.7), and 2.9 (SD, 2.6), respectively.
Discussion MCIDs for the ASES score, SST score, and VAS pain score after shoulder arthroplasty using an anchor-based method have not been previously reported. Our results show that a 1.4point change on the VAS for pain, a 20.9-point change in the ASES score, and a 2.4-point change in the SST score achieve a clinically important improvement after shoulder arthroplasty. Younger age correlated with larger MCIDs for all outcome scores, although correlations were weak. Female patients had higher MCIDs for the VAS pain score and ASES score but not the SST score, although these correlations were also weak. Implant type (anatomic TSA vs RSA) was not correlated; neither was duration of follow-up except for the ASES score, although the correlation was weak. Overall, these MCIDs are very similar to MCIDs previously reported for the nonoperative treatment of rotator cuff disease (VAS pain score, 1.4; ASES score, 12 to 17; SST score, 2 to 2.3).18,19 MCIDs have been reported for a variety of shoulder instruments using a heterogeneous population of patients with a variety of shoulder pathologies.6,11,13,14 These studies determined MCIDs after nonoperative physical therapy for a variety of shoulder diseases in the same study cohort. Leggin et al11 reported that the MCID for the Penn shoulder score for patients undergoing outpatient therapy for a variety of shoulder disorders was 11.4 points. Michener et al13 reported that the MCID for the ASES score for a variety of shoulder pathologies treated after 3 to 4 weeks of physical therapy was 6.4
R.Z. Tashjian et al. points. Mintken et al 14 reported that the MCID for the QuickDASH (shortened Disabilities of the Arm, Shoulder and Hand questionnaire) was 8 points as determined in a group of patients presenting for physical therapy for generalized shoulder pain. Franchignoni et al6 evaluated 255 patients after physical therapy for a variety of upper limb musculoskeletal disorders and determined the MCIDs for the Disabilities of the Arm, Shoulder and Hand questionnaire to be 10.83 points and the QuickDASH to be 15.91 points. Limitations of these studies include that the MCIDs were not specified with regard to the treatment of a specific disease process. Theoretically, the amount of change required for a patient with shoulder instability to achieve a minimal improvement may differ from that for a patient with arthritis. Moreover, these studies were performed after nonoperative treatment; therefore, it is unclear if these MCIDs would be similar if surgery was performed. Theoretically, a patient would expect a larger change in outcome to be considered clinically important after surgery compared with therapy because of the risk involved with the treatment. To address some of the issues of disease heterogeneity and varying MCIDs based on treatment (operative vs nonoperative), various authors have reported on MCIDs of various outcome instruments for specific pathologies and after surgical treatments.10,17-20 Tashjian et al18,19 evaluated a homogeneous group of patients with rotator cuff disease treated with physical therapy and, using an anchor-based approach, determined the MCIDs for a VAS for pain, the ASES score, and the SST score to be 1.4 points, 12 to 17 points, and 2 to 2.3 points, respectively. Kukkonen et al10 used an anchor-based approach to determine the MCID for the Constant score after rotator cuff repair surgery in 802 patients. They determined the MCID to be 10.4 points. Torrens et al20 evaluated 60 patients after RSA for rotator cuff insufficiency and, using an anchor-based approach, determined the MCID for the overall Constant score to be 8 points. Roy et al17 determined MCIDs using a distribution-based approach for the SST in patients undergoing shoulder arthroplasty and reported the MCID to be 3 points. Our results are in fairly close agreement with the published MCIDs for the VAS pain score, ASES score, and SST score for the nonoperative management of rotator cuff disease,17,18 as well as the results of Roy et al19 using the distribution-based approach. Overall, it appears from the results that the MCIDs differ very little between specific diseases, as well as treatments (surgical vs nonoperative), at least for these 3 measures. In general, the MCID is approximately 15% to 20% of the overall maximal score for the outcome measure. This study has several limitations. It was performed retrospectively; therefore, we were not able to include a complete dataset of patients who underwent TSA. Nevertheless, this is the largest dataset investigating MCIDs of any shoulder outcome measure to date. Second, the anchor question used in this study has not been validated. However, it has consistently been used in previous publications on MCIDs, and the values obtained across different datasets and disease processes are remarkably consistent. By analyzing anatomic TSA
ARTICLE IN PRESS MCID after shoulder arthroplasty and RSA, we are examining 2 distinct diseases that may have inherent differences in symptom constellation and response to surgery, thus potentially having different MCID thresholds. However, we found similar MCID values for the chosen scales for each surgical procedure. Finally, the lack of associations of various factors with MCIDs that we report may be a result of a type II error and potentially may be improved with a larger population, although with over 300 patients in our study, it may be very difficult to significantly increase the sample size without compromising the homogeneity of the disease and treatment.
Conclusion Our results show that a 1.4-point change in the VAS pain score, a 20.9-point change in the ASES score, and a 2.4point change in the SST score achieve a clinically important improvement after shoulder arthroplasty. These MCIDs are similar to previously published MCIDs for the nonoperative treatment of rotator cuff disease using an anchor-based technique. These data suggest that MCIDs can probably be translated across various disease pathologies, as well as treatments, with minimal effect. Further study is warranted to evaluate MCIDs for other surgical treatments of shoulder pathology including rotator cuff tearing and instability to confirm a correlation with previously published results.
5
7.
8.
9.
10.
11.
12.
13.
14.
15.
Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References 1. Angst F, Goldhahn J, Drerup S, Aeschlimann A, Schwyzer HK, Simmen BR. Responsiveness of six outcome assessment instruments in total shoulder arthroplasty. Arthritis Rheum 2008;59:391-8. http://dx.doi.org/ 10.1002/art.23318 2. Beaton D, Richards RR. Assessing the reliability and responsiveness of 5 shoulder questionnaires. J Shoulder Elbow Surg 1998;7:565-72. 3. Brozek JL, Guyatt GH, Schünemann HJ. How well-grounded minimal important difference can enhance transparency of labeling claims and improve interpretation of a patient reported outcome measure. Health Qual Life Outcomes 2006;4:69. http://dx.doi.org/10.1186/1477-7525-4-69 4. Churchill RS, Zellmer C, Zimmers HJ, Ruggero R. Clinical and radiographic analysis of a partially cemented glenoid implant: five-year minimum follow-up. J Shoulder Elbow Surg 2010;19:1091-7. http:// dx.doi.org/10.1016/j.jse.2009.12.022 5. Cuff D, Pupello D, Virani N, Levy J, Frankle M. Reverse shoulder arthroplasty for the treatment of rotator cuff deficiency. J Bone Joint Surg Am 2008;90:1244-51. http://dx.doi.org/10.2106/JBJS.G.00775 6. Franchignoni F, Vercelli S, Giordano A, Sartorio F, Bravini E, Ferriero G. Minimal clinically important difference of the Disabilities of the Arm, Shoulder and Hand outcome measure (DASH) and its shortened version
16.
17.
18.
19.
20.
21.
22.
(QuickDASH). J Orthop Sports Phys Ther 2014;44:30-9. http:// dx.doi.org/10.2519/jospt.2014.4893 Godfrey J, Hamman R, Lowenstein S, Briggs K, Kocher M. Reliability, validity, and responsiveness of the simple shoulder test: psychometric properties by age and injury type. J Shoulder Elbow Surg 2007;16:260-7. http://dx.doi.org/10.1016/j.jse.2006.07.003 Guyatt GH, Feeny DH, Patrick D. Proceedings of the International Conference on the Measurement of Quality of Life as an Outcome in Clinical Trials: postscript. Control Clin Trials 1991;12:266S-269S. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989;10:407. Kukkonen J, Kauko T, Vahlberg T, Joukainen A, Aärimaa V. Investigating minimal clinically important difference for Constant score in patients undergoing rotator cuff surgery. J Shoulder Elbow Surg 2013;22:1650-5. http://dx.doi.org/10.1016/j.jse.2013.05.002 Leggin BG, Michener LA, Shaffer MA, Brenneman SK, Iannotti JP, Williams GR Jr. The Penn shoulder score: reliability and validity. J Orthop Sports Phys Ther 2006;36:138-51. http://dx.doi.org/10.2519/ jospt.2006.36.3.138 Lippitt SB, Harryman DT, Matsen FA. A practical tool for evaluating function: the simple shoulder test. In: Matsen FA, Fu FH, Hawkins RJ, editors. The shoulder: A balance of mobility and stability. Rosemont, IL: American Academy of Orthopaedic Surgeons; 1993. p. 501-18. Michener LA, McClure PW, Sennett BJ. American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, patient self-report section: reliability, validity, and responsiveness. J Shoulder Elbow Surg 2002;11:587-94. http://dx.doi.org/10.1067/mse.2002.127096 Mintken PE, Glynn P, Cleland JA. Psychometric properties of the shortened Disabilities of the Arm, Shoulder, and Hand Questionnaire (QuickDASH) and Numeric Pain Rating Scale in patients with shoulder pain. J Shoulder Elbow Surg 2009;18:920-6. http://dx.doi.org/10.1016/ j.jse.2008.12.015 Parks DK, Casagrande DJ, Schrumpf MA, Harmsen SM, Norris TR, Kelly JD II. Radiographic and clinical outcomes of total shoulder arthroplasty with an all-polyethylene pegged ingrowth glenoid component: prospective short- to medium-term follow-up. J Shoulder Elbow Surg 2016;25:246-55. http://dx.doi.org/10.1016/j.jse.2015.07.008 Richards RR, An KN, Bigliani LU, Freidman RJ, Gartsman GM, Gristina AG, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg 1994;3:347-52. Roy JS, Macdermid JC, Faber KJ, Drosdowech DS, Athwal GS. The simple shoulder test is responsive in assessing change following shoulder arthroplasty. J Orthop Sports Phys Ther 2010;40:413-21. http:// dx.doi.org/10.2519/jospt.2010.3209 Tashjian RZ, Deloach J, Green A, Porucznik CA, Powell AP. Minimal clinically important differences in ASES and simple shoulder test scores after nonoperative treatment of rotator cuff disease. J Bone Joint Surg Am 2010;92:296-303. http://dx.doi.org/10.2106/JBJS.H.01296 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. http://dx.doi.org/10.1016/j.jse.2009.03.021 Torrens C, Guirro P, Santana F. The minimal clinically important difference for function and strength in patients undergoing reverse shoulder arthroplasty. J Shoulder Elbow Surg 2016;25:262-8. http:// dx.doi.org/10.1016/j.jse.2015.07.020 Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement. Ann Rheum Dis 2005;64:29-33. http://dx.doi.org/ 10.1136/ard.2004.022905 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;52:86173.
www.elsevier.com/locate/ymse
ORIGINAL ARTICLE
Determining the minimal clinically important difference for the American Shoulder and Elbow Surgeons score, Simple Shoulder Test, and visual analog scale measuring pain after shoulder arthroplasty Robert Z. Tashjian, MDa,*, Man Hung, PhDa,b, Jay D. Keener, MDc, Randy Christopher Bowen, MD, MSa, Jared McAllister, MDc, Wei Chen, PhDb, Gregory Ebersole, MSc, Erin K. Granger, MPHa, Aaron M. Chamberlain, MDc a
Department of Orthopaedics, University of Utah School of Medicine, Salt Lake City, UT, USA Division of Epidemiology, University of Utah School of Medicine, Salt Lake City, UT, USA c Department of Orthopaedics, Washington University, St. Louis, MO, USA b
Background: Minimal clinically important differences (MCIDs) for the American Shoulder and Elbow Surgeons (ASES) score, the Simple Shoulder Test (SST), and a visual analog scale (VAS) measuring pain have not been previously described using an anchor-based method after shoulder arthroplasty. The purpose of this study was to determine the MCIDs for these measures after shoulder arthroplasty for glenohumeral arthritis or advanced rotator cuff disease. Methods: Primary anatomic total shoulder arthroplasty (TSA), primary reverse TSA, or hemiarthroplasty was performed in 326 patients by 1 of 5 shoulder and elbow surgeons. The SST score, ASES score, and VAS pain score were collected preoperatively and at a minimum of 2 years postoperatively (mean, 3.5 years). The MCIDs were calculated for the ASES score, SST score, and VAS pain score using an anchor-based method. Results: The MCIDs for the ASES score, SST score, and VAS pain score were 20.9 (P < .001), 2.4 (P < .0001), and 1.4 (P = .0158), respectively. Duration of follow-up and type of arthroplasty (anatomic TSA vs reverse TSA) did not have a significant effect on the MCIDs (P > .1) except shorter follow-up correlated with a larger MCID for the ASES score (P = .0081). Younger age correlated with larger MCIDs for all scores (P < .024). Female sex correlated with larger MCIDs for the VAS pain score (P = .123) and ASES score (P = .05). Conclusions: Patients treated with a shoulder arthroplasty require a 1.4-point improvement in the VAS pain score, a 2.4-point improvement in the SST score, and a 21-point improvement in the ASES score to achieve a minimal clinical importance difference from the procedure. Level of evidence: Basic Science Study; Validation of Outcomes Instruments © 2016 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved.
Institutional Review Board approval was obtained from the University of Utah School of Medicine (IRB No. 00065636). *Reprint requests: Robert Z. Tashjian, MD, Department of Orthopaedics, University of Utah School of Medicine, 590 Wakara Way, Salt Lake City, UT 84108, USA. E-mail address: [email protected] (R.Z. Tashjian). 1058-2746/$ - see front matter © 2016 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. http://dx.doi.org/10.1016/j.jse.2016.06.007
ARTICLE IN PRESS 2
R.Z. Tashjian et al. Keywords: Shoulder arthroplasty; minimal clinically important difference; outcome scores; pain; function; Simple Shoulder Test; ASES score
Total shoulder arthroplasties (TSAs)—both anatomic TSA and reverse total shoulder arthroplasty (RSA)—are reliable surgical treatment options for patients with shoulder osteoarthritis and advanced rotator cuff disease to reduce pain and restore function. Various patient-reported functional outcome scores have been used to assess outcomes after anatomic TSA and RSA including the American Shoulder and Elbow Surgeons (ASES) score and Simple Shoulder Test (SST).4,5,15 Previous studies have shown the ASES score and SST are valid, reliable, and responsive outcome measures.2,7,13 Responsiveness refers to the ability of a measure to detect a change once it has occurred. The minimal clinically important difference (MCID) is a measure of responsiveness representing the smallest subjective difference in an outcome score that represents a clinically important change to the patient.8 MCID values have an important role in maximizing the utility of outcome measures because they allow interpretation of research comparing different treatments. MCIDs also allow analysis of the amount of improvement that an individual patient achieves after treatment and assist in study design because they provide a measure of effect size. Among the various methods used to determine MCIDs, distribution-based and anchor-based methods are the most commonly used. Distribution-based methods are based on the statistical characteristics of the sample being investigated and include the effect size, standard response mean, and standard error of measurement.9,22 Anchor-based methods use an external question, or “anchor,” to determine if a change that occurs reaches clinical importance.3 Currently, there is no gold standard regarding which method is optimal for determining MCIDs, with each having its own benefits and limitations. Distribution-based methods are easy to calculate and require no further data beyond the questionnaires. Limitations are that the results may not truly reflect the patient’s perception of clinical importance. Anchor-based methods are more difficult to perform and may be limited by the anchor questions created, although they may better capture a true assessment of clinical importance. Limited studies exist on MCIDs for the ASES score and SST. MCIDs have been previously determined for the ASES score and SST in patients treated for global shoulder pain and for each measure in patients undergoing nonoperative treatment of rotator cuff disease.13,18 Both studies used an anchorbased method to determine MCIDs. The MCID for the ASES score was 6.4 points as reported by Michener et al13 and between 12 and 17 points as reported by Tashjian et al.18 The MCID for the SST was approximately 2 points as reported by Tashjian et al.18 Roy et al17 used a distribution-based method to determine the MCID for the SST after shoulder arthroplasty and reported the MCID as 3 points. Angst et al1 used a distribution-based method to determine MCIDs for the ASES
score after shoulder arthroplasty. They also performed an anchor-based assessment of responsiveness, but MCIDs were not reported using the anchor-based data. Finally, Tashjian et al19 reported the MCID for a visual analog scale (VAS) for pain was 1.4 points after nonoperative treatment of rotator cuff disease. No prior studies have determined the MCID for the ASES score or VAS score for pain after shoulder arthroplasty using anchor-based methods. The purpose of this study was to determine the MCID for the ASES score, SST score, and VAS pain score after shoulder arthroplasty using an anchor-based approach. The effect of patient factors (age, sex) and treatment factors (duration of follow-up, arthroplasty type [RSA or anatomic TSA]) on the MCIDs was also evaluated.
Methods Three hundred twenty-six patients who underwent a primary anatomic TSA, RSA, or hemiarthroplasty by 1 of 5 shoulder and elbow surgeons over a 7-year period from 2005 to 2012 were included in the study, retrospectively reviewed, and evaluated at a minimum of 2 years postoperatively. The patients were recruited from 2 medical centers, the University of Utah School of Medicine and Washington University School of Medicine. Patient records were reviewed, and patients were identified as having either a preoperative ASES or SST score available in their records before arthroplasty. Only the patient-reported questionnaire section of the ASES form was recorded. An attempt was made to contact all patients with either a preoperative SST or ASES score through a letter and then a phone call at a minimum of 2 years’ follow-up after surgery. Patients were recruited into the study if they completed the follow-up questionnaire by phone. A total of 1900 shoulder arthroplasties were performed at Washington University and 216 at the University of Utah during the study period. A total of 509 patients met the inclusion criteria of having either a preoperative ASES or SST score available and having undergone an anatomic TSA, RSA, or hemiarthroplasty at least 2 years previously. A total of 363 patients were recruited (72% followup). Only patients with a diagnosis of osteoarthritis, rheumatoid arthritis, rotator cuff arthropathy, or advanced rotator cuff disease were included in the final analysis, resulting in a total of 326 patients in the final dataset. There were 198 anatomic TSAs, 124 RSAs, and 4 hemiarthroplasties. The diagnoses for which arthroplasty was performed included rotator cuff arthropathy (103), irreparable rotator cuff tears with pseudoparalysis (17), osteoarthritis with fullthickness rotator cuff tears (5), rheumatoid arthritis (5), and osteoarthritis (196). There were 184 female and 142 male patients. The mean age of the patients at the time of surgery was 67.4 years (SD, 9.4 years). Patients completed a questionnaire before surgery including the SST, ASES score, and VAS for pain. The SST is a shoulder region– specific outcome measure that focuses on characterizing function in patients with abnormal shoulder pathology. It is composed of 12
ARTICLE IN PRESS MCID after shoulder arthroplasty yes/no questions representing important activities of daily living that can be performed by normally functioning shoulders. The questions specifically assess shoulder comfort and function.12 The ASES score is a patient-assessed region-specific outcome measure that evaluates pain on a VAS and function by 10 Likert-style questions.16 Patients are asked to rate their current shoulder pain on a VAS assessing shoulder pain levels in 1-digit increments from 0, “none,” to 10, “disabling.” The patients were re-evaluated with a follow-up questionnaire by phone at a minimum of 2 years (mean, 3.5 years; SD, 1.6 years) postoperatively. The follow-up questionnaire included the ASES score, the SST, the VAS for pain, and a 4-item anchor question evaluating improvement after treatment. The improvement question was the anchor question used to determine the MCIDs. This anchor question was derived from the anchor question designed by Tubach et al21 and has been previously used to determine MCIDs by Tashjian et al.18,19 Patients were asked the following: “Since your shoulder replacement surgery, please rate your response to treatment: A, none—no good at all, ineffective treatment; B, poor—some effect but unsatisfactory; C, good—satisfactory effect with occasional episodes of pain or stiffness; D, excellent—ideal response, virtually pain free.”
Statistical analysis MCIDs for each outcome variable (SST score, ASES score, VAS pain score) were calculated. First, the change scores for each variable were calculated from preoperative assessment to final postoperative assessment. Patients were classified by the anchor question as having “no change” (A group [none] and B group [poor] combined) or “change” (C group [good]). The D group (excellent) was not included in the analysis because this was considered beyond minimal change. We then performed t tests to compare the means between the unchanged and changed groups. The MCID is the mean difference between the changed and unchanged groups. Correlations between MCIDs and age and between MCIDs and duration of follow-up were determined using Spearman correlation coefficients. Correlations between MCIDs and sex and between MCIDs and implant type (anatomic TSA vs RSA) were determined using univariate regression analysis. Separate analysis of correlations between MCIDs and disease etiology was not performed due to a majority of the TSAs being performed for osteoarthritis without a rotator cuff tear and a majority of the RSAs being performed for rotator cuff arthropathy. Consequently, the correlations between MCIDs and implant type closely reflect the correlations between MCIDs and disease etiology (RSA for rotator cuff arthropathy and TSA for primary osteoarthritis without a rotator cuff tear). P < .05 was considered statistically significant. Spearman correlations were interpreted based on the results as follows: 0.0 to 0.19, very weak; 0.2 to 0.39, weak; 0.4 to 0.59, moderate; 0.6 to 0.79, strong; and 0.8 to 1.0, very strong. All of these analyses were performed with SAS software (version 9.4; SAS Institute, Cary, NC, USA).
Results Of the patients, 112 rated their response as C (good) and were therefore classified as the changed group whereas 21 rated their response as A (none) or B (poor) and were classified as the unchanged group. The MCIDs for the ASES score, SST
3 Table I Average preoperative and postoperative outcome scores and improvements in scores for all patients, RSA patients, and anatomic TSA patients Mean Overall SST Preoperative Postoperative Change VAS Preoperative Postoperative Change ASES Preoperative Postoperative Change TSA SST Preoperative Postoperative Change VAS Preoperative Postoperative Change ASES Preoperative Postoperative Change RSA SST Preoperative Postoperative Change VAS Preoperative Postoperative Change ASES Preoperative Postoperative Change
SD
P value
2.8 8.8 6.0
2.3 2.8 2.7
< .01
7.1 1.4 −5.6
2.0 2.2 2.8
< .01
31.6 79.6 47.9
14.2 20.5 22.3
< .01
3.4 9.5 6.1
2.5 2.3 2.6
< .01
7.1 1.2 −5.9
2.0 2.0 2.7
< .01
33.3 84.3 50.9
14.2 18.7 20.9
< .01
2.0 7.8 5.8
1.6 3.2 3.0
< .01
6.9 1.8 −5.1
2.0 2.2 2.9
< .01
29.8 72.7 43.9
13.7 20.8 23.3
< .01
ASES, American Shoulder and Elbow Surgeons; RSA, reverse total shoulder arthroplasty; SST, Simple Shoulder Test; TSA, total shoulder arthroplasty; VAS, visual analog scale.
score, and VAS score evaluating pain were 20.9 (P < .001), 2.4 (P < .0001), and 1.4 (P = .0158), respectively. Average preoperative and postoperative outcome scores and improvements in scores for all patients, RSA-only patients, and anatomic TSA–only patients are presented in Table I. Correlations between duration of follow-up and MCIDs were evaluated and determined to not have an effect on the MCID for the VAS pain score (r = 0.14, P = .1) or SST score (r = 0.1, P = .21). Shorter duration of follow-up was correlated
ARTICLE IN PRESS 4 with a larger MCID for the ASES score (r = 0.23, P = .01), although this correlation was considered weak. Correlations between age and MCIDs were evaluated, and younger age was correlated with larger MCIDs for all outcome scores (VAS pain score: r = 0.2, P = .02; ASES score: r = 0.28, P = .001; SST score: r = 0.25, P = .004), although all correlations were considered weak. Correlations between sex and MCIDs were evaluated using univariate regressions; no correlation was identified for the SST score (P = .89), whereas female sex was correlated with higher MCIDs for the VAS pain score (P = .01) and ASES score (P = .05). The average MCIDs for male patients for the VAS pain score, ASES score, and SST score were 1.3 (SD, 2.2), 17.9 (SD, 15.9), and 1.2 (SD, 2.5), respectively. The average MCIDs for female patients for the VAS pain score, ASES score, and SST score were 1.3 (SD, 2.5), 22.4 (SD, 17.8), and 3.2 (SD, 2.4), respectively. Correlations between implant type (anatomic TSA vs RSA) and MCIDs were evaluated using univariate regressions, and no correlation was identified for the VAS pain score (P = .18), ASES score (P = .1), or SST score (P = .1). The average MCIDs for anatomic TSA patients for the VAS pain score, ASES score, and SST score were 2.1 (SD, 2.5), 24.5 (SD, 17.2), and 1.5 (SD, 2.3), respectively. The average MCIDs for RSA patients for the VAS pain score, ASES score, and SST score were 0.5 (SD, 2.3), 15.4 (SD, 16.7), and 2.9 (SD, 2.6), respectively.
Discussion MCIDs for the ASES score, SST score, and VAS pain score after shoulder arthroplasty using an anchor-based method have not been previously reported. Our results show that a 1.4point change on the VAS for pain, a 20.9-point change in the ASES score, and a 2.4-point change in the SST score achieve a clinically important improvement after shoulder arthroplasty. Younger age correlated with larger MCIDs for all outcome scores, although correlations were weak. Female patients had higher MCIDs for the VAS pain score and ASES score but not the SST score, although these correlations were also weak. Implant type (anatomic TSA vs RSA) was not correlated; neither was duration of follow-up except for the ASES score, although the correlation was weak. Overall, these MCIDs are very similar to MCIDs previously reported for the nonoperative treatment of rotator cuff disease (VAS pain score, 1.4; ASES score, 12 to 17; SST score, 2 to 2.3).18,19 MCIDs have been reported for a variety of shoulder instruments using a heterogeneous population of patients with a variety of shoulder pathologies.6,11,13,14 These studies determined MCIDs after nonoperative physical therapy for a variety of shoulder diseases in the same study cohort. Leggin et al11 reported that the MCID for the Penn shoulder score for patients undergoing outpatient therapy for a variety of shoulder disorders was 11.4 points. Michener et al13 reported that the MCID for the ASES score for a variety of shoulder pathologies treated after 3 to 4 weeks of physical therapy was 6.4
R.Z. Tashjian et al. points. Mintken et al 14 reported that the MCID for the QuickDASH (shortened Disabilities of the Arm, Shoulder and Hand questionnaire) was 8 points as determined in a group of patients presenting for physical therapy for generalized shoulder pain. Franchignoni et al6 evaluated 255 patients after physical therapy for a variety of upper limb musculoskeletal disorders and determined the MCIDs for the Disabilities of the Arm, Shoulder and Hand questionnaire to be 10.83 points and the QuickDASH to be 15.91 points. Limitations of these studies include that the MCIDs were not specified with regard to the treatment of a specific disease process. Theoretically, the amount of change required for a patient with shoulder instability to achieve a minimal improvement may differ from that for a patient with arthritis. Moreover, these studies were performed after nonoperative treatment; therefore, it is unclear if these MCIDs would be similar if surgery was performed. Theoretically, a patient would expect a larger change in outcome to be considered clinically important after surgery compared with therapy because of the risk involved with the treatment. To address some of the issues of disease heterogeneity and varying MCIDs based on treatment (operative vs nonoperative), various authors have reported on MCIDs of various outcome instruments for specific pathologies and after surgical treatments.10,17-20 Tashjian et al18,19 evaluated a homogeneous group of patients with rotator cuff disease treated with physical therapy and, using an anchor-based approach, determined the MCIDs for a VAS for pain, the ASES score, and the SST score to be 1.4 points, 12 to 17 points, and 2 to 2.3 points, respectively. Kukkonen et al10 used an anchor-based approach to determine the MCID for the Constant score after rotator cuff repair surgery in 802 patients. They determined the MCID to be 10.4 points. Torrens et al20 evaluated 60 patients after RSA for rotator cuff insufficiency and, using an anchor-based approach, determined the MCID for the overall Constant score to be 8 points. Roy et al17 determined MCIDs using a distribution-based approach for the SST in patients undergoing shoulder arthroplasty and reported the MCID to be 3 points. Our results are in fairly close agreement with the published MCIDs for the VAS pain score, ASES score, and SST score for the nonoperative management of rotator cuff disease,17,18 as well as the results of Roy et al19 using the distribution-based approach. Overall, it appears from the results that the MCIDs differ very little between specific diseases, as well as treatments (surgical vs nonoperative), at least for these 3 measures. In general, the MCID is approximately 15% to 20% of the overall maximal score for the outcome measure. This study has several limitations. It was performed retrospectively; therefore, we were not able to include a complete dataset of patients who underwent TSA. Nevertheless, this is the largest dataset investigating MCIDs of any shoulder outcome measure to date. Second, the anchor question used in this study has not been validated. However, it has consistently been used in previous publications on MCIDs, and the values obtained across different datasets and disease processes are remarkably consistent. By analyzing anatomic TSA
ARTICLE IN PRESS MCID after shoulder arthroplasty and RSA, we are examining 2 distinct diseases that may have inherent differences in symptom constellation and response to surgery, thus potentially having different MCID thresholds. However, we found similar MCID values for the chosen scales for each surgical procedure. Finally, the lack of associations of various factors with MCIDs that we report may be a result of a type II error and potentially may be improved with a larger population, although with over 300 patients in our study, it may be very difficult to significantly increase the sample size without compromising the homogeneity of the disease and treatment.
Conclusion Our results show that a 1.4-point change in the VAS pain score, a 20.9-point change in the ASES score, and a 2.4point change in the SST score achieve a clinically important improvement after shoulder arthroplasty. These MCIDs are similar to previously published MCIDs for the nonoperative treatment of rotator cuff disease using an anchor-based technique. These data suggest that MCIDs can probably be translated across various disease pathologies, as well as treatments, with minimal effect. Further study is warranted to evaluate MCIDs for other surgical treatments of shoulder pathology including rotator cuff tearing and instability to confirm a correlation with previously published results.
5
7.
8.
9.
10.
11.
12.
13.
14.
15.
Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References 1. Angst F, Goldhahn J, Drerup S, Aeschlimann A, Schwyzer HK, Simmen BR. Responsiveness of six outcome assessment instruments in total shoulder arthroplasty. Arthritis Rheum 2008;59:391-8. http://dx.doi.org/ 10.1002/art.23318 2. Beaton D, Richards RR. Assessing the reliability and responsiveness of 5 shoulder questionnaires. J Shoulder Elbow Surg 1998;7:565-72. 3. Brozek JL, Guyatt GH, Schünemann HJ. How well-grounded minimal important difference can enhance transparency of labeling claims and improve interpretation of a patient reported outcome measure. Health Qual Life Outcomes 2006;4:69. http://dx.doi.org/10.1186/1477-7525-4-69 4. Churchill RS, Zellmer C, Zimmers HJ, Ruggero R. Clinical and radiographic analysis of a partially cemented glenoid implant: five-year minimum follow-up. J Shoulder Elbow Surg 2010;19:1091-7. http:// dx.doi.org/10.1016/j.jse.2009.12.022 5. Cuff D, Pupello D, Virani N, Levy J, Frankle M. Reverse shoulder arthroplasty for the treatment of rotator cuff deficiency. J Bone Joint Surg Am 2008;90:1244-51. http://dx.doi.org/10.2106/JBJS.G.00775 6. Franchignoni F, Vercelli S, Giordano A, Sartorio F, Bravini E, Ferriero G. Minimal clinically important difference of the Disabilities of the Arm, Shoulder and Hand outcome measure (DASH) and its shortened version
16.
17.
18.
19.
20.
21.
22.
(QuickDASH). J Orthop Sports Phys Ther 2014;44:30-9. http:// dx.doi.org/10.2519/jospt.2014.4893 Godfrey J, Hamman R, Lowenstein S, Briggs K, Kocher M. Reliability, validity, and responsiveness of the simple shoulder test: psychometric properties by age and injury type. J Shoulder Elbow Surg 2007;16:260-7. http://dx.doi.org/10.1016/j.jse.2006.07.003 Guyatt GH, Feeny DH, Patrick D. Proceedings of the International Conference on the Measurement of Quality of Life as an Outcome in Clinical Trials: postscript. Control Clin Trials 1991;12:266S-269S. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989;10:407. Kukkonen J, Kauko T, Vahlberg T, Joukainen A, Aärimaa V. Investigating minimal clinically important difference for Constant score in patients undergoing rotator cuff surgery. J Shoulder Elbow Surg 2013;22:1650-5. http://dx.doi.org/10.1016/j.jse.2013.05.002 Leggin BG, Michener LA, Shaffer MA, Brenneman SK, Iannotti JP, Williams GR Jr. The Penn shoulder score: reliability and validity. J Orthop Sports Phys Ther 2006;36:138-51. http://dx.doi.org/10.2519/ jospt.2006.36.3.138 Lippitt SB, Harryman DT, Matsen FA. A practical tool for evaluating function: the simple shoulder test. In: Matsen FA, Fu FH, Hawkins RJ, editors. The shoulder: A balance of mobility and stability. Rosemont, IL: American Academy of Orthopaedic Surgeons; 1993. p. 501-18. Michener LA, McClure PW, Sennett BJ. American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, patient self-report section: reliability, validity, and responsiveness. J Shoulder Elbow Surg 2002;11:587-94. http://dx.doi.org/10.1067/mse.2002.127096 Mintken PE, Glynn P, Cleland JA. Psychometric properties of the shortened Disabilities of the Arm, Shoulder, and Hand Questionnaire (QuickDASH) and Numeric Pain Rating Scale in patients with shoulder pain. J Shoulder Elbow Surg 2009;18:920-6. http://dx.doi.org/10.1016/ j.jse.2008.12.015 Parks DK, Casagrande DJ, Schrumpf MA, Harmsen SM, Norris TR, Kelly JD II. Radiographic and clinical outcomes of total shoulder arthroplasty with an all-polyethylene pegged ingrowth glenoid component: prospective short- to medium-term follow-up. J Shoulder Elbow Surg 2016;25:246-55. http://dx.doi.org/10.1016/j.jse.2015.07.008 Richards RR, An KN, Bigliani LU, Freidman RJ, Gartsman GM, Gristina AG, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg 1994;3:347-52. Roy JS, Macdermid JC, Faber KJ, Drosdowech DS, Athwal GS. The simple shoulder test is responsive in assessing change following shoulder arthroplasty. J Orthop Sports Phys Ther 2010;40:413-21. http:// dx.doi.org/10.2519/jospt.2010.3209 Tashjian RZ, Deloach J, Green A, Porucznik CA, Powell AP. Minimal clinically important differences in ASES and simple shoulder test scores after nonoperative treatment of rotator cuff disease. J Bone Joint Surg Am 2010;92:296-303. http://dx.doi.org/10.2106/JBJS.H.01296 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. http://dx.doi.org/10.1016/j.jse.2009.03.021 Torrens C, Guirro P, Santana F. The minimal clinically important difference for function and strength in patients undergoing reverse shoulder arthroplasty. J Shoulder Elbow Surg 2016;25:262-8. http:// dx.doi.org/10.1016/j.jse.2015.07.020 Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement. Ann Rheum Dis 2005;64:29-33. http://dx.doi.org/ 10.1136/ard.2004.022905 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;52:86173.