Modifiable and Nonmodifiable Factors Associated With Patient Satisfaction in Spine Surgery and Other Orthopaedic Subspecialties

A Retrospective Survey Analysis

Leah Steinmetz, BA; Dennis Vasquez-Montes, MS; Bradley C. Johnson, MD; Aaron J. Buckland, MBBS, FRACS; Jeffrey A. Goldstein, MD, FACS; John A. Bendo, MD; Thomas J. Errico, MD; Charla R. Fischer, MD


Curr Orthop Pract. 2019;30(6):555-560. 

In This Article


Overall, 16,175 orthopaedic patient surveys were analyzed, including 1400 in spine surgery (8.7%) and the following nonspine orthopaedic subspecialties: adult reconstructive surgery, shoulder and elbow surgery, sports medicine, pediatric surgery, and hand surgery. The mean age for all respondents was 55.4 ± 16.88, and 6666 (41.2%) of survey respondents were female.

Patient Demographic Factors (Nonmodifiable)

Patient age, gender, and education were significant predictors for PR and LTR in the nonspine group (P<0.001). In nonspine orthopaedic patients, white (P<0.001) and Asian (P<0.001) ethnicities were significant predictors for PR and LTR, while Hispanic and African American ethnicities were not significant for PR or LTR (P>0.05). In spine patients, age (P<0.001) and white ethnicity (P= 0.01) were the only factors significantly associated with PR. However, gender (P=0.03), education (P=0.03), and white ethnicity (P=0.01) were significantly associated with LTR (Table 1). Female physicians (PR OR=1.320 P=0.07, LTR OR=1.529 P=0.04) and a high school education or lower (PR OR=1.488 P=0.05, LTR OR=1.968 P=0.01) were significant predictors of satisfaction in nonspine orthopaedic subspecialties but not in spine patients. Satisfaction scores were inversely correlated with education level in both spine and nonspine orthopaedic specialties (Table 2).

Provider-related Questions on CG-CAHPS (Modifiable)

Overall satisfaction correlated more closely with provider communication (Spine: PR R=0.853, LTR R=0.792; Orthopaedics: PR R= 0.813, LTR R=0.763) than staff communication (Spine: PR R=0.279, LTR R=0.312; Orthopaedics: PR R=0.267, LTR R=0.309). By question, top significant correlations with satisfaction were as follows: "Provider explained in a way you understand" (Spine: PR R=0.743, LTR R=0.687; Orthopaedics: PR R=0.098, LTR 0.100), "Provider listened carefully to you" (Spine: PR R=0.836, LTR R=0.784; Orthopedics: PR R=0.711, LTR R=0.661), "Shows respect for what you say" (Spine: PR R=0.807, LTR R=0.753; Orthopaedics: PR R=0.226, LTR R=0.257), "Spends enough time with you" (Spine: PR R=0.753, LTR R=0.713; Orthopaedics PR R=0.689, LTR R=0.667). Staff-related questions were found to have smaller correlation coefficients: "Clerks and receptionists helpful" (Spine PR R=0.283, LTR R=0.346), "Clerks treat with courtesy and respect" (Spine PR R=0.218, LTR R=0.201). Correlations were all significant with P<0.001 and are shown in Table 3.

Combination of Factors

Independent predictor analysis showed provider communication to be the strongest independent predictor of high PR and LTR ratings for both spine patients (PR R2= 0.728, LTR R[2] =0.627) and nonspine orthopaedic patients (PR R[2] =0.662, LTR R2=0.583), followed by staff communication, ethnicity, and age. Multivariate stepwise regression analysis using these predictors showed that the combination of provider communication, staff communication, same ethnicity as provider, and age bias significantly predicted scores for PR (Spine R[2] =0.735, Orthopaedics R2= 0.665) and LTR (Spine R2:0.642, Orthopaedics R2= 0.599) in both spine and nonspine orthopaedic groups (Table 4).