Summary

The results of this study are noteworthy as they may suggest that patients who intend to complete the survey- potentially as a result of doing well after surgery- are going to complete it. Although socio-demographic and treatment factors also significantly influenced completion rate, the reality may be that these surveys are only burdensome to individuals who perceive surveys as a burden.

Abstract

Background

Patient-reported outcome measures (PROMs) are important for tracking outcomes following shoulder surgery. However, PROMs can be limited by time constraints and patient compliance. The aim of this study was to evaluate the time to complete commonly administered PROMs as well as determine how it affects survey completion rates following shoulder surgery. We hypothesized that longer PROMs completion times will negatively affect completion rates.

Methods

Utilizing our institutions prospective outcomes registry, we identified patients who underwent shoulder surgery, including rotator cuff repair (RCR), and (anatomic and reverse) total shoulder arthroplasty (TSA). Patients were invited to complete a survey consisting of 6 PROMS pre- and postoperatively at 3, 6, 12, and 24 months. A series of univariate multiple logistic regression analyses were performed to determine all possible predictors of completion rate at each time interval.

Results

An overall total of 742 patients (male, 58.9%; mean age, 60.9±10.8 years) who underwent shoulder surgery were enrolled (RCR, n=467; TSA, n=275). An overall average completion time of 10.60 (SD, 19.20) minutes was observed. Six variables significantly increased the likelihood of completion at each follow-up interval: (1) Doubling the time per question of the most recent survey taken (6-month, OR: 2.04, p=0.011); (2) The most recent survey being taken through email (3-month, OR: 2.41, p=0.015; 1-year, OR: 9.48, p<0.000001); (3) Each additional access attempt (averaged over all previous surveys) (6-month, OR: 1.84, p=0.016; 2-year, OR: 2.09, p=0.017); (4) Obtaining the MCID in the most recent survey taken (6-month, OR: 3.27, p=0.011); (5) Undergoing RCR (vs. TSA) (pre-op, OR: 2.32, p=0.00001; 2-year, OR: 2.32, p=0.015); and (6) Being over the age of 66 (vs. 58 to 66) years (2-year, OR: 2.40; p=0.048). Two variables significantly decreased the likelihood of completion at each follow-up interval: (1) Each additional (reminder) email sent decreases (pre-op, OR: 0.85, p<0.0000001); 3-month, OR: 0.40, p<0.000001; 6-month, OR: 0.58, p<0.000001); 1-year, OR: 0.85, p=0.021); and (2) Receiving care at the hospital clinic (vs. the satellite clinic) (pre-op, OR: 0.467, p=0.00033).

Conclusions

The results of this study are noteworthy as they may suggest that patients who intend to complete the survey – potentially as a result of doing well after surgery – are going to complete it. This study demonstrated that doubling the time per question of the most recent survey taken significantly increases the likelihood of 6-month survey completion after shoulder surgery. Those who previously completed the most recent survey via email, attempted to access the survey more, or obtained the MCID in the most recent survey taken are more likely to complete the survey, while those who received additional reminder emails are less likely to complete the survey. Although socio-demographic and treatment factors also significantly influenced completion rate, the reality may be that these surveys are only burdensome to individuals who perceive surveys as a burden.