Advertisement

Anatomic shoulder arthroplasty in workers’ compensation patients: predictors of success and return to work

Published:January 22, 2021DOI:https://doi.org/10.1053/j.sart.2020.11.017

      Abstract

      Background

      In patients with shoulder arthritis, workers’ compensation (WC) status presents unique challenges to the clinician because of the socioeconomic and psychosocial aspects attendant to patients covered on this type of insurance. Patients, surgeons, and these insurance programs would be informed by a better understanding of the factors that may impact the results of a shoulder arthroplasty as treatment for glenohumeral arthritis in this population. The objectives of this study were to determine: (1) the extent to which patient-reported outcomes are impacted by WC status in comparison to patients covered by other types of insurance, (2) which factors are predictive of a successful outcome of shoulder arthroplasty as defined by improvement exceeding the minimal clinically important difference (MCID), (3) the ability of WC patients to return to their presurgical occupation after arthroplasty.

      Methods

      This was a retrospective study of a longitudinally maintained institutional database of 677 patients who underwent primary anatomic shoulder arthroplasty with a minimum 2-year follow-up, 39 of whom had WC insurance. These patients were compared to a matched cohort of 78 patients without WC insurance. Primary outcome measures included SST scores, SANE scores, and need for revision surgery. Univariate and multivariate analyses were performed to determine preoperative characteristics associated with success as defined by improvement greater than the MCID of the SST.

      Results

      Success, defined as improvement beyond MCID, occurred in a significantly lower proportion of WC patients compared to non-WC patients (64% vs. 94%, P< .001). Older age (P= .010) and a higher preoperative SF-36 role physical domain score (a measure of the patient's perceived limitations in routine activities;P= .007) were associated with improvement beyond the MCID on univariate analysis. Higher preoperative SF-36 role physical domain scores had the greatest correlation with a successful outcome (OR 1.19, 95% 0.99-1.43, P= .07). A significantly lower percentage of patients with physically-demanding jobs returned to previous occupation compared to patients with non-physically-demanding jobs (13% vs. 73%, P= .001).

      Conclusions

      The challenges in treating patients covered by WC are underscored by the high percentage of these patients that do not improve beyond the MCID and the high percentage of patients with physically-demanding jobs that are unable to return to their presurgical occupation. Those patients who report less disability with routine daily activities (ie, higher SF-36 role physical domain scores) may fare better with anatomic shoulder arthroplasty. Patients with physically-demanding jobs should be counseled that return to their previous occupation is unlikely.

      Level of Evidence

      Level III; Retrospective Cohort Comparative Study

      Keywords

      Treating glenohumeral arthritis in patients under workers’ compensation (WC) status has unique challenges for the shoulder surgeon due to the socioeconomic and psychosocial aspects attendant to patients covered on this type of insurance.
      • Daniels AH
      • Kuris EO
      • Palumbo MA.
      The role of the orthopaedic surgeon in workers' compensation cases.
      The surgeon is often called upon to opine whether the arthritis was caused or aggravated by the patient's work, to determine when the patient can return to work, to define restrictions for work and to report the degree of permanent partial impairment related to the arthritis. While the WC system is designed to protect and provide benefits to injured workers, these associated socioeconomic components can create an environment that shifts the focus away from optimizing the treatment of the patient's arthritis.
      • Daniels AH
      • Kuris EO
      • Palumbo MA.
      The role of the orthopaedic surgeon in workers' compensation cases.
      WC status is known to be associated with poor outcomes after surgery.
      • Harris I
      • Mulford J
      • Solomon M
      • van Gelder JM
      • Young J.
      Association between compensation status and outcome after surgery: a meta-analysis.
      The literature pertaining to the effect of WC status on shoulder arthroplasty outcomes is limited to only a few studies.
      • Cvetanovich GL
      • Savin DD
      • Frank RM
      • Gowd AK
      • Sumner SA
      • Romeo AA
      • et al.
      Inferior outcomes and higher complication rates after shoulder arthroplasty in workers' compensation patients.
      ,
      • Jawa A
      • Dasti UR
      • Fasulo SM
      • Vaickus MH
      • Curtis AS
      • Miller SL.
      Anatomic total shoulder arthroplasty for patients receiving workers' compensation.
      ,
      • Morris BJ
      • Haigler RE
      • Laughlin MS
      • Elkousy HA
      • Gartsman GM
      • Edwards TB.
      Workers' compensation claims and outcomes after reverse shoulder arthroplasty.
      Despite significant improvements in pain and function, patients with WC undergoing shoulder arthroplasty appear to have lower improvement in patient reported outcomes, higher reoperation rates, and higher rates of persistent pain than non-WC patients. Because of the unique challenges associated with WC insurance, it would be helpful to identify the factors that may affect the results of a shoulder arthroplasty in this patient population. We sought to determine: (1) the extent to which patient-reported outcomes are impacted by WC status in comparison to patients covered by other types of insurance, (2) which factors are predictive of a successful outcome of shoulder arthroplasty as defined by improvement exceeding the minimal clinically important difference (MCID), (3) the ability of WC patients to return to their presurgical occupation after arthroplasty. We hypothesized that (1) patient reported outcomes would be significantly worse in patients with WC status than those without, (2) psychosocial factors, such as current work status and involvement of a lawyer, would have a strong correlation with a successful outcome, (3) WC patients with physically-demanding jobs would be unlikely to return to their presurgical occupation.

      Materials and Methods

       Patient selection

      This study was approved by the University of Washington Institution Review Board (IRB #STUDY00007300). This was a retrospective cohort study of a longitudinally-maintained institutional database of patients having primary shoulder arthroplasty between the dates of August 2010 and August 2017. A total of 1,112 patients were enrolled into the database during this time period. Patients were included if (1) they had a primary anatomic shoulder arthroplasty (total shoulder arthroplasty or ream and run arthroplasty) and (2) they had a clinical follow-up of at least 2 years.

       Patient evaluation and outcomes

      Demographic information was recorded including age, sex, body mass index (BMI), history of prior surgery, history of depression or anxiety, diabetes status, tobacco use, alcohol use, marital status, current working status, and involvement of a lawyer. Patient optimism prior to surgery was assessed using the following question: “On a 0-10 scale, where 0 is not at all helpful and 10 is extremely helpful, how helpful do you believe surgery will be for your current shoulder pain and functioning?” Visual Analog Scale pain, Short Form 36 (SF-36), Simple Shoulder Test (SST), and Single Assessment Numeric Evaluation (SANE) scores were collected preoperatively and 2-years postoperatively.

       Retrospective matched cohort

      Patients were divided into two groups based on insurance type: WC and non-WC. Using these criteria, 39 patients with workers compensation were identified. Due to statistically significant differences in patient characteristics (age, sex, ASA class, BMI, history of prior shoulder surgery, active smoking status) between the 39 WC patients and the 638 non-WC patients (Appendix 1), a matched non-WC cohort was created. Two non-WC patients were matched to every WC patient based on similar age, sex, ASA class, history of prior surgery, smoking status, type of procedure, preoperative SST, and preoperative SANE scores.

       Return to previous occupation

      In the WC cohort, the WC claims were reviewed to determine the physical demands of the presurgical occupation as classified by the United States Department of Labor Office of Worker's Compensation Programs. Jobs that required medium (9.1-22.7 kg or 20-50 pounds of force occasionally), heavy (22.7-45.4 kg or 50-100 pounds of force occasionally), or very heavy (>45.4 kg or >100 pounds of force occasionally) physical demands were classified as physically-demanding, while jobs that required sedentary (less than 4.5 kg or 10 pounds of force occasionally) or light (4.5-9.1 kg or 10-20 pounds of force occasionally) physical demands were classified as non-physically-demanding. We classified return to work as either return to previous occupation to some capacity or did not return to previous occupation.

       Statistical analysis

      Descriptive characteristics are presented as means, standard deviations, and ranges for continuous variables and as counts and percentages for categorical variables. Statistical significance of differences in the characteristics between WC and non-WC patients was calculated using the 2-sample t test with unequal variances and either the chi-squared test or Fisher exact test (as appropriate). Similarly, characteristics of WC patients with and without a successful outcome as defined by improvement beyond the defined MCID for each respective cohort, were compared. A successful outcome was defined as one in which the SST improved ≥3 points based on a MCID threshold of 2.4 as previously defined by Tashjian et al.
      • Tashjian RZ
      • Hung M
      • Keener JD
      • Bowen RC
      • McAllister J
      • Chen W
      • et al.
      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.
      A multivariate logistic regression analysis was used to determine independent predictors of a successful outcome. For the multivariate analysis, we included age, sex, work status, preoperative diagnosis, preoperative SST, preoperative SF-36 mental and physical component summaries, and variables found to be significant on univariate analysis. Statistical analyses were performed with the use of SPSS (Version 25.0; IBM Corp., Armonk, NY, USA).

      Results

       Comparison of workers’ compensation and matched nonworkers’ compensation cohorts

      The WC cohort was statistically similar to the matched non-WC cohort for all characteristics with the exception of a heavier BMI (32.2 ± 5.4 vs. 29.2 ± 4.8, P= .003) and more frequent lawyer involvement (21% vs. 1%, P< .001). Preoperative SST score (2.9 ± 2.5 vs. 3.4 ± 2.2, P= .289), preoperative SANE score (32 ± 16 vs. 35 ± 18, P= .381), age (53.9 ± 9.7 vs. 56.4 ± 10.4, P= .215), ASA class (P= 1.000), history of prior surgery (P= .896), and smoking status (P= .242) were similar (Table 1).
      Table 1Patient demographics and preoperative characteristics of workers’ compensation cohort versus matched nonworker's compensation patients (1WC:2 non-WC patient ratio).
      AllN (%) or Mean ± SDWC patients N (%) or Mean ± SDNon-WC patients N (%) or Mean ± SDP value
      Number of patients1173978-
      Age55.5 ± 10.253.9 ± 9.756.4 ± 10.4.215
      Sex1.000
      Male108 (92%)36 (92%)72 (92%)
      Female9 (8%)3 (8%)6 (8%)
      BMI30.2 ± 5.232.2 ± 5.429.2 ± 4.8.003
      p < 0.05
      ASA1.000
      p < 0.05
      113 (11%)5 (13%)8 (10%)
      286 (74%)28 (72%)58 (74%)
      318 (15%)6 (15%)12 (15%)
      40 (0%)0 (0%)0 (0%)
      Smoker10 (9%)5 (13%)5 (6%).242
      Diabetic7 (6%)4 (10%)3 (4%).220
      Depression21 (18%)8 (21%)13 (17%).609
      Anxiety23 (20%)9 (23%)14 (18%).511
      Marital status.882
      Married86 (74%)29 (74%)57 (73%)
      Single19 (16%)4 (10%)15 (19%)
      Divorced / widowed / other12 (10%)6 (15%)6 (8%)
      Currently working65 (56%)20 (51%)45 (58%).511
      Lawyer involved9 (8%)8 (21%)1 (1%)<.001
      p < 0.05
      Diagnosis.584
      Osteoarthritis76 (65%)24 (62%)52 (67%)
      Capsulorrhaphy arthropathy14 (12%)5 (13%)9 (12%)
      Other27 (23%)10 (25%)17 (21%)
      Arthroplasty type1.000
      Total shoulder arthroplasty60 (51%)20 (51%)40 (51%)
      Ream and run arthroplasty45 (39%)15 (39%)30 (39%)
      Hemiarthroplasty12 (10%)4 (10%)8 (10%)
      Surgery on dominant extremity72 (62%)23 (64%)49 (63%).687
      Prior shoulder surgery58 (49%)19 (49%)39 (49%).896
      Pre-op optimism (0 to 10)9.3 ± 1.19.4 ± 1.19.2 ± 1.1.561
      Pre-op pain VAS score7.1 ± 1.97.4 ± 1.36.9 ± 2.1.149
      Pre-op SST3.3 ± 2.32.9 ± 2.53.4 ± 2.2.289
      Pre-op SANE34 ± 1832 ± 1635 ± 18.381
      Preoperative SF-36 domains
      Physical component summary40 ± 2034 ± 1842 ± 20.077
      Mental component summary80 ± 2174 ± 2583 ± 17.098
      Physical function62 ± 2059 ± 2364 ± 18.297
      Role physical43 ± 2736 ± 2846 ± 26.084
      Role emotional82 ± 2772 ± 3686 ± 21.039
      p < 0.05
      Mental health75 ± 1671 ± 1877 ± 14.105
      Bodily pain34 ± 1930 ± 1836 ± 19.136
      Vitality57 ± 2155 ± 1957 ± 22.675
      General health72 ± 1970 ± 2073 ± 19.455
      Social function68 ± 2662 ± 2972 ± 24.090
      BMI, Body Mass Index; ASA, American Society of Anesthesiologists class; VAS, Visual Analogue Scale; SST, Simple Shoulder Test; SANE, Single Assessment Numeric Evaluation; SF-36, Short Form 36; WC, Workers’ Compensation.
      P value: 2-sample t test for parametric continuous variables, Chi-squared or Fisher exact test for categorical variables.
      ASA 1 and 2 compared to ASA 3 and 4 in analysis. Values are presented as frequencies (percentages) and means ± standard deviations.
      low asterisk p < 0.05
      Preoperatively, WC patients reported a similar Visual Analog Scale pain scores (7.4 ± 1.3 vs. 6.9 ± 2.1, P=.149), SST scores (2.9 ± 2.5 vs. 3.4 ± 2.2, P= .289), and SANE score (32 ± 16 vs. 35 ± 18, P= .381) to the matched non-WC patients. For the preoperative SF-36 domains, only role emotional (72 ± 36 vs. 86 ± 21, P= .039) maintained statistical significance between the 2 cohorts.
      Postoperatively, WC patients reported significantly lower SST scores (6.9 ± 3.9 vs. 10.0 ± 2.5, P< .001) and lower improvement in SST (3.9 ± 3.8 vs. 6.6 ± 2.8, P< .001) compared to the matched non-WC cohort (Table 2). Similarly, the percentage of maximum possible improvement (43% ± 45% vs. 78% ± 28%, P< .001) and SANE scores (65 ± 21 vs. 80 ± 18, P< .001) were lower.
      Table 2Patient-reported outcomes in workers’ compensation cohort versus matched nonworker's compensation patients (1WC:2 non-WC patient ratio).
      AllN (%) or Mean ± SDWC patientsN (%) or Mean ± SDNon-WC patientN (%) or Mean ± SDP value
      Number of patients1173978-
      Post-op SST9.0 ± 3.46.9 ± 3.910.0 ± 2.5<.001
      p < 0.05
      SST change5.7 ± 3.43.9 ± 3.86.6 ± 2.8<.001
      p < 0.05
      % MPI66 ± 3843 ± 4578 ± 28<.001
      p < 0.05
      Post-op SANE75 ± 2165 ± 2180 ± 18<.001
      p < 0.05
      Open revision procedure9 (8%)6 (15%)3 (4%).058
      SST, Simple Shoulder Test; SANE, Single Assessment Numeric Evaluation; WC, Workers’ Compensation; % MPI, percentage of maximum possible improvement.
      P value 2-sample t test for parametric continuous variables, Chi-squared for categorical variables. Values are presented as frequencies (percentages) and means ± standard deviations.
      low asterisk p < 0.05
      Six WC patients (15%) required open revision procedures while only 3 non-WC patients (4%) required revision (P= .058). Revision in the WC cohort was performed for subscapularis failure for 2 patients and persistent pain and stiffness in 4 patients. Revision in the non-WC cohort was performed for subscapularis failure in 1 patient and persistent pain and stiffness in 2 patients.

       Factors predictive of success in workers’ compensation patients

      Using improvement greater than the MCID for the SST (≥3) as a metric of success, 25 WC patients (64%) were defined as achieving a successful outcome. This was a significantly smaller percentage compared to the matched non-WC cohort in which 73 patients (94%) were defined as having a successful outcome (P< .001).
      When comparing WC patients that improved more than the MCID of the SST to WC patients that did not, patients that had improvement greater than the MCID were older (56.8 ± 9.5 vs. 48.8 ± 8.2, P= .010) and had higher scores in the SF-36 role physical domain (45 ± 30 vs. 21 ± 19, P= .007; Table 3). Depression (P= .686), anxiety (P= 1.000), marital status (P= .238), current working status (P= .146), involvement of a lawyer (P= .686), and optimism scores (P= .312) were not significantly associated with a successful outcome. On multivariate analysis, a higher SF-36 role physical domain had the strongest correlation with a successful outcome (OR 1.19, 95% 0.99-1.43, P= .07; Table 4), while male sex (OR 10.13, 95% 0.21-482.76, P= .24) and status of currently working (OR 3.06, 95% 0.25-36.71, P= .38) had the highest odds ratios.
      Table 3Patient demographics and preoperative characteristics of all WC patients; WC patients with SST change above MCID; WC patients with SST change below MCID.
      All WC patientsN (%) or Mean ± SDSST change >MCIDN (%) or Mean ± SDSST change <MCIDN (%) or Mean ± SDP value
      Number of patients392514-
      Age53.9 ± 9.756.8 ± 9.548.8 ± 8.2.010
      p < 0.05
      Sex.289
      Male36 (92%)24 (96%)12 (86%)
      Female3 (8%)1 (4%)2 (14%)
      BMI32.2 ± 5.431.4 ± 5.233.7 ± 5.5.217
      ASA
      ASA 1 and 2 compared to ASA 3 and 4 in analysis.
      1.000
      15 (13%)5 (20%)0 (0%)
      228 (72%)16 (64%)12 (86%)
      36 (15%)4 (16%)2 (14%)
      40 (0%)0 (0%)0 (0%)
      Smoker5 (13%)4 (16%)1 (7%).637
      Diabetic4 (10%)3 (12%)1 (7%)1.000
      Depression8 (21%)6 (24%)2 (14%).686
      Anxiety9 (23%)6 (24%)3 (21%)1.000
      Marital status.238
      Married30 (77%)21 (84%)9 (64%)
      Single4 (10%)2 (8%)2 (14%)
      Divorced/other5 (13%)2 (8%)3 (21%)
      Currently working20 (51%)15 (60%)5 (36%).146
      Lawyer involved8 (21%)6 (24%)2 (14%).686
      Diagnosis
      “Capsulorrhaphy arthropathy” and “Other” groups combined in analysis. Values are presented as frequencies (percentages) and means ± standard deviations.
      .268
      Osteoarthritis24 (61%)17 (68%)7 (50%)
      Capsulorrhaphy arthropathy5 (13%)5 (20%)0 (0%)
      Other10 (26%)3 (12%)7 (50%)
      Arthroplasty type.431
      Total shoulder arthroplasty20 (51%)14 (56%)6 (43%)
      Ream and run arthroplasty15 (39%)10 (40%)5 (36%)
      Hemiarthroplasty4 (10%)1 (4%)3 (21%)
      Surgery on dominant extremity23 (59%)15 (60%)8 (57%).862
      Prior shoulder surgery19 (49%)12 (48%)7 (50%)1.000
      Pre-op optimism (0-10)9.4 ± 1.19.5 ± 0.89.1 ± 1.5.312
      Pre-op pain VAS score7.4 ± 1.37.5 ± 1.47.1 ± 1.2.421
      Pre-op SST2.9 ± 2.52.8 ± 2.13.3 ± 3.1.581
      Pre-op SANE32 ± 1633 ± 1629 ± 18.506
      Preoperative SF-36 domains
      Physical component summary34 ± 1840 ± 1727 ± 19.070
      Mental component summary74 ± 2578 ± 2368 ± 27.296
      Physical function59 ± 2364 ± 1650 ± 30.141
      Role physical36 ± 2845 ± 3021 ± 19.007
      p < 0.05
      Role emotional72 ± 3677 ± 3165 ± 43.407
      Mental health71 ± 1877 ± 1663 ± 20.056
      Bodily pain30 ± 1834 ± 1724 ± 20.137
      Vitality55 ± 1955 ± 1856 ± 22.927
      General health70 ± 2073 ± 1564 ± 26.274
      Social function62 ± 2966 ± 2754 ± 33.298
      BMI, Body Mass Index; ASA, American Society of Anesthesiologists class; VAS, Visual Analogue Scale; SST, Simple Shoulder Test; SANE, Single Assessment Numeric Evaluation; SF-36, Short Form 36; MCID, minimal clinically important difference; WC, Workers’ Compensation.
      P value: 2-sample t test for parametric continuous variables, Chi-squared or Fisher exact test for categorical variables.
      low asterisk p < 0.05
      ASA 1 and 2 compared to ASA 3 and 4 in analysis.
      “Capsulorrhaphy arthropathy” and “Other” groups combined in analysis. Values are presented as frequencies (percentages) and means ± standard deviations.
      Table 4Multivariate analysis to determine independent predictors of improvement greater than the SST MCID in WC patients.
      Odds Ratio (95% CI)P value
      Age1.09 (0.94-1.26).23
      Male sex10.13 (0.21-482.76).24
      Currently working3.06 (0.25-36.71).38
      Diagnosis - capsulorrhaphy arthropathy or other0.33 (0.01-9.18).52
      Preoperative SST0.69 (0.39-1.23).21
      Preop SF-36 role physical1.19 (0.99-1.43).07
      Preop SF-36 mental component summary1.01 (0.95-1.08).69
      Preop SF-36 physical component summary1.03 (0.96-1.11).43
      MCID, minimal clinically important difference; SST, Simple Shoulder Test; SF-36, Short Form 36; WC, Workers’ Compensation.
      SST-related variables were removed due to collinearity.
      Odds ratios are presented with 95% confidence intervals (CI).

       Return to previous occupation

      Of the 34 patients with WC claim data available, 11 worked in a job that was classified as sedentary or light physical demands, while 23 worked in a job that was classified as medium, heavy, or very heavy physical demands. In total, 11 of 34 WC patients (32%) were able to return to previous occupation. 8 of 11 patients (73%) working in non-physically-demanding jobs returned to previous occupation while only 3 of 23 patients (13%) working in physically-demanding jobs returned to previous occupation (P= .001).

      Discussion

      The treatment of patients with shoulder arthritis covered by WC carries with it some unique challenges. The data from this study support our hypotheses that patient reported outcomes are significantly worse in patients with WC status than those without and that WC patients with physically demanding jobs are unlikely to return to their presurgical occupation. However, our study did not clearly identify psychosocial factors that were associated with a successful outcome.
      WC status has previously been reported to negatively affect outcomes after shoulder surgery. Cvetanovich et al found that worker compensation patients experience clinical improvements after shoulder arthroplasty but had inferior ASES and SST scores, higher rate of persistent pain at 2-year follow-up, and less forward elevation when matched with a nonworker's compensation population.
      • Cvetanovich GL
      • Savin DD
      • Frank RM
      • Gowd AK
      • Sumner SA
      • Romeo AA
      • et al.
      Inferior outcomes and higher complication rates after shoulder arthroplasty in workers' compensation patients.
      The WC patient group also had a higher re-operation rate. Steinhaus et al performed a systematic review and meta-analysis and reported a return to work rate of 63.6% at a mean of 2.3 months postoperatively after shoulder arthroplasty.
      • Steinhaus ME
      • Gowd AK
      • Hurwit DJ
      • Lieber AC
      • Liu JN.
      Return to work after shoulder arthroplasty: a systematic review and meta-analysis.
      Heavy-intensity occupations, as defined by the United States Department of Labor, resulted in a significantly lower return to work rate. Jawa et al found that worker's compensation patients receiving an anatomic total shoulder appeared to have worse ASES scores when compared to a non-worker's compensation population.
      • Jawa A
      • Dasti UR
      • Fasulo SM
      • Vaickus MH
      • Curtis AS
      • Miller SL.
      Anatomic total shoulder arthroplasty for patients receiving workers' compensation.
      They also reported that less than a third of the patients were able to return to work of any kind, and none were able to return to full duty at their current job. Our results add to this information by defining the characteristics of patients covered by WC in contrast to those covered by other insurance programs, including younger age, a higher percentage of males, a higher percentage of patients who were smoking, had a diagnosis of anxiety, had prior surgery, and had lawyer involvement. In addition, we assessed the factors associated with patients on WC who were able to return to their presurgical occupation after arthroplasty, including a higher levels of presurgical physical activity and a job requiring fewer physical demands.
      Psychosocial factors are often considered to have a profound impact on recovery and outcomes after shoulder arthroplasty. A recent systematic review provided evidence that supports the idea that psychosocial factors may play an important role in patient outcomes after shoulder arthroplasty.
      • Vajapey SP
      • Cvetanovich GL
      • Bishop JY
      • Neviaser AS.
      Psychosocial factors affecting outcomes after shoulder arthroplasty: a systematic review.
      Chen et al also evaluated three patient-reported outcomes measurement information system (PROMIS) domains, physical function, depression, and pain interference scores, and found that preoperative scores were predictive of postoperative achievement of the MCID within each domain.
      • Chen RE
      • Papuga MO
      • Nicandri GT
      • Miller RJ
      • Voloshin I.
      Preoperative Patient-Reported Outcomes Measurement Information System (PROMIS) scores predict postoperative outcome in total shoulder arthroplasty patients.
      Our study did find lower preoperative SF-36 scores in many domains (including social function, mental health, role emotional) for our WC patients compared to non-WC patients, demonstrating baseline differences between the 2 cohorts. This may explain why there was a smaller proportion of patients who exceeded the MCID for the SST within the WC cohort compared to non-WC. However, in our univariate analysis within the WC cohort, we did not find psychosocial factors such as depression, anxiety, status other than married, status of currently working, or lack of lawyer involvement to be associated with a successful outcome, although definitive conclusions cannot be made given our limited sample size. We did find, however, that patients’ preoperative perceived limitations in routine activities (role physical) was associated with failure of improvement past the MCID.
      This study does have limitations. First, we only included those undergoing anatomic shoulder arthroplasty to limit the heterogeneity of the patient population. Therefore, our results may not be generalizable to work-related injuries that involve reverse arthroplasty for cuff-related problems. Second, our sample size may not have been large enough to adequately detect differences in patients with “success” and “failure.” A larger sample size potentially could demonstrate a significant correlation of psychosocial factors to a successful outcome. Third, while we were able to document whether patients returned to previous occupation or not, we were not able to determine the degree to which return to previous occupation was driven by the physical limitations of the patient's shoulder, by other patient psychosocial factors, or by limitations set forth by the surgeon. Fourth, our focus was on patient-reported outcomes and revision rates. We did not collect radiographic or range of motion data.

      Conclusion

      The challenges in treating patients covered by WC are underscored by the high percentage of these patients that did not improve beyond the MCID and the high percentage of patients with physically-demanding jobs that were unable to return to their presurgical occupation. This study helps inform the management of patients with glenohumeral arthritis who are covered by Workmen's Compensation by assessing the factors associated with patients on WC who (1) had improvement greater than the MCID of the SST (older age and less disability with routine activities), and (2) were able to return to their presurgical occupation after arthroplasty (job requiring fewer physical demands).

      Disclaimer

      The author, their immediate family, and any research foundation 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.

      Appendix. Supplementary materials

      References

        • Chen RE
        • Papuga MO
        • Nicandri GT
        • Miller RJ
        • Voloshin I.
        Preoperative Patient-Reported Outcomes Measurement Information System (PROMIS) scores predict postoperative outcome in total shoulder arthroplasty patients.
        J Shoulder Elbow Surg. 2019; 28: 547-554https://doi.org/10.1016/j.jse.2018.08.040
        • Cvetanovich GL
        • Savin DD
        • Frank RM
        • Gowd AK
        • Sumner SA
        • Romeo AA
        • et al.
        Inferior outcomes and higher complication rates after shoulder arthroplasty in workers' compensation patients.
        J Shoulder Elbow Surg. 2019; 28: 875-881
        • Daniels AH
        • Kuris EO
        • Palumbo MA.
        The role of the orthopaedic surgeon in workers' compensation cases.
        J Am Acad Orthop Surg. 2017; 25: e45-e52https://doi.org/10.5435/jaaos-d-16-00499
        • Harris I
        • Mulford J
        • Solomon M
        • van Gelder JM
        • Young J.
        Association between compensation status and outcome after surgery: a meta-analysis.
        Jama. 2005; 293: 1644-1652https://doi.org/10.1001/jama.293.13.1644
        • Jawa A
        • Dasti UR
        • Fasulo SM
        • Vaickus MH
        • Curtis AS
        • Miller SL.
        Anatomic total shoulder arthroplasty for patients receiving workers' compensation.
        J Shoulder Elbow Surg. 2015; 24: 1694-1697https://doi.org/10.1016/j.jse.2015.04.017
        • Morris BJ
        • Haigler RE
        • Laughlin MS
        • Elkousy HA
        • Gartsman GM
        • Edwards TB.
        Workers' compensation claims and outcomes after reverse shoulder arthroplasty.
        J Shoulder Elbow Surg. 2015; 24: 453-459https://doi.org/10.1016/j.jse.2014.07.009
        • Steinhaus ME
        • Gowd AK
        • Hurwit DJ
        • Lieber AC
        • Liu JN.
        Return to work after shoulder arthroplasty: a systematic review and meta-analysis.
        J Shoulder Elbow Surg. 2019; 28: 998-1008https://doi.org/10.1016/j.jse.2018.12.011
        • Tashjian RZ
        • Hung M
        • Keener JD
        • Bowen RC
        • McAllister J
        • Chen W
        • et al.
        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.
        J Shoulder Elbow Surg. 2017; 26: 144-148
        • Vajapey SP
        • Cvetanovich GL
        • Bishop JY
        • Neviaser AS.
        Psychosocial factors affecting outcomes after shoulder arthroplasty: a systematic review.
        J Shoulder Elbow Surg. 2019; https://doi.org/10.1016/j.jse.2019.09.043