Does Resilience Change in Patients Undergoing Shoulder Surgery? A Retrospective Comparative Study Utilizing the Brief Resilience Scale.

BACKGROUND

Resilience refers to the ability to adapt or recover from stress. There is increasing appreciation that it plays an important role in wholistic patient-centered care and may affect patient outcomes, including those of orthopaedic surgery. Despite being a focus of the current orthopaedic evidence, there is no strong understanding yet of whether resilience is a stable patient quality or a dynamic one that may be modified perioperatively to improve patient-reported outcome scores.

QUESTIONS/PURPOSES: (1) Does resilience change postoperatively? (2) How do outcome measures change postoperatively in relation to resilience grouping? (3) For patients who do have resilience instability (change in resilience of ≥ 1 SD between any two follow-up points), how were patient-level factors, surgical characteristics, and outcome measures associated with instability?

METHODS

In this single-surgeon, retrospective, comparative study, we identified all patients who underwent shoulder surgery between March 2021 and March 2023 from the medical records of one US military teaching hospital, resulting in 144 initial patients. Data on resilience (measured by the Brief Resilience Scale) and outcomes (assessed using the Numeric Rating Scale [NRS] and the Single Assessment Numeric Evaluation [SANE]) were collected for all patients and maintained in a longitudinal outcomes score database. Patients younger than 18 years of age (1% [1 of 144]) who underwent surgery for fracture, acute tendon rupture (8% [11 of 144]), or revision surgery (3% [4 of 144]); had concomitant shoulder conditions (such as, instability or rotator cuff tear) (1% [2 of 144]); or had incomplete follow-up data (4% [5 of 144]) were excluded, leaving 84% (121 of 144) of the original sample size for analysis. Among the patients, 12% (15 of 121) were women, the mean age was 41 ± 15 years, and the most common indication for surgery was instability (40% [48 of 121]) followed by rotator cuff repair (29% [35 of 121]). Based on their preoperative Brief Resilience Scale and its deviation from the mean, patients were stratified into low (> 1 SD below mean), intermediate (within 1 SD above and below mean), and high (> 1 SD above mean) resilience groups. Preoperatively, 19% (23 of 121) of patients were classified as low resilience, 62% (75 of 121) as intermediate resilience, and 19% (23 of 121) as high resilience. The mean ± SD preoperative Brief Resilience Scale score was 25 ± 4. The Brief Resilience Scale is a six-item scale with a calculated summary score ranging from 6 to 30. A higher score is suggestive of greater perceived resilience. There were no differences in the preoperative Brief Resilience Scale score with regard to age, gender, type of surgery performed, or outcome measures. Patient resilience was followed during the postoperative period for a minimum of 6 months, and instability in the scale was evaluated. Instability in resilience was defined as change in Brief Resilience Scale score by > 1 SD from one follow-up time point to another. Perioperative NRS and SANE outcomes, in addition to demographic data, were utilized to evaluate the relationship between resilience and patient-level factors.

RESULTS

Brief Resilience Scale groups across all time points remained consistent with no change in grouping or crossover in groups except for patients with low resilience who had an increase in mean ± SD Brief Resilience Scale score by the final follow-up (18 ± 3 versus 20 ± 4; p < 0.05). Regardless of resilience group, there was a decrease in mean ± SD NRS (4.4 ± 2.2 versus 2.4 ± 2.3; p < 0.001) and an improvement in mean ± SD SANE (46 ± 19 versus 69 ± 21; p < 0.001) scores during the postoperative period. At the 1- to 2-month follow-up and the 6- to 10-month follow-up visits, patients with high resilience were more likely to have lower NRS scores than patients with intermediate resilience (1.8 ± 1.0 versus 3.8 ± 2.3; p = 0.003) and low resilience (1.5 ± 1.8 versus 3.3 ± 2.4; p < 0.001), respectively. No relationship was observed between resilience groups and SANE scores, surgical category, and percentage of patients meeting the minimum clinically important difference (MCID) of the NRS or the SANE. Regarding resilience instability, 46% (56 of 121) of patients were categorized as having a Brief Resilience Scale change of ≥ 1 SD from baseline during the postoperative period. Gender (r = 0.03; p = 0.21), age (p = 0.81), and surgical category (r = 0.01; p = 0.88) were not associated with the likelihood of resilience instability. Individuals whose resilience increased had a lower starting Brief Resilience Scale score than those whose resilience stayed the same (22 ± 4 versus 25 ± 4, respectively; p < 0.001) or those whose resilience decreased (22 ± 4 versus 26 ± 3, respectively; p < 0.001).

CONCLUSION

When evaluated by resilience group, the trait appears static; however, at the individual level, resilience appears dynamic and complex. Patients with high resilience may have less postoperative pain. Identification of patients with low resilience may indicate patients who experience more dynamic change in this psychometric property.

LEVEL OF EVIDENCE

Level III, therapeutic study.