No Correlation Between Press Ganey Survey Responses and Outcomes in Post–Total Hip Arthroplasty Patients

In the era of the online orthopedic market, patients tend to equate publicly available online satisfaction surveys with what they presume their ultimate surgical outcome will be. Therefore, the purpose of this study was to assess whether there is a correlation between Press Ganey (PG) scores and (1) Hip Western Ontario and McMaster Universities Osteoarthritis Index and Harris Hip Score; (2) Short Form-12 and Short Form-36 scores; (3) University of California Los Angeles and Visual Analog Scale scores assessed at a mean of 3 years (range, 1 to 6 years) after surgery. In addition, we assessed whether (4) these correlations persist in patients who were evaluated under 2 years and 3 or more years after surgery.

Six-hundred ninety-two patients from November 2009 to January 2015 were identified from our institutional PG database. One-hundred ninety (27%) responded to the survey. One-hundred forty-nine (78%) patients were given the total hip arthroplasty assessment tools at a minimum of 2-year follow-up, and 33 patients (17%) completed their survey before 2 years after surgery. We assessed whether overall hospital rating scores correlated with the above assessment tools.

Pearson correlation analysis revealed no correlation between the PG survey score and the assessment tools. HHS had the highest correlation coefficient (r = .120; P = .316); however, this was not significant. After removing the patients who had their follow-up survey administered under 2 years after surgery (33 patients), there was still no statistically significant correlation between the above-mentioned outcome scores and PG overall hospital rating (P > .05).

No statistically significant relationship was found between commonly used total hip arthroplasty assessment tools and the PG overall hospital rating. Based on these results, PG surveys may not be a suitable implementation of the Center for Medicare and Medicaid services. A set of measures that can be widely collected and reported by hospitals for patients to use in order to evaluate hip arthroplasty outcomes needs to be developed. These results are of paramount importance, indicating a necessary reevaluation of PG surveys as a major determinant for reimbursements rendered by orthopedists and their use by patients.