Procedure Type and Preoperative Patient-Reported Outcome Metrics Predict Variation in the Value of Hip Arthroscopy for Femoroacetabular Impingement

Abstract

Purpose

To characterize variation in the value of hip arthroscopy for femoroacetabular impingement and explore associations between value and patient-specific demographic characteristics, comorbidities, preoperative patient-reported outcome measures (PROMs), and intraoperative variables.

Methods

We included all patients aged 18 years or older who underwent primary arthroscopic acetabular labral repair or debridement between 2015 and 2020 with minimum 2-year follow-up. The exclusion criteria were hip dysplasia, advanced hip osteoarthritis (TÖnnis grade >1), or unreconcilable documenting errors. Value was calculated by dividing 2-year postoperative International Hip Outcome Tool 33 scores by time-driven activity-based costs. To protect the confidentiality of internal hospital cost data, the study average for value was normalized to 100. Multivariable linear mixed-effects models were used to identify factors underlying variation in value.

Results

This study included 161 patients. There were 76 women (47.2%) and 85 men, with a mean age of 36.0 years (standard deviation [SD], 10.9 years) and mean body mass index (BMI) of 25.8 (SD, 4.3). Most patients were white (92.5%), were not Hispanic (93.8%), and were commercially insured (92.5%). Preoperatively, 57.1% of hips were classified as Tönnis grade 1 (57.1%) whereas the remainder were grade 0. The normalized value of hip arthroscopy ranged from 25.4 to 216.4 (mean ± SD, 100 ± 38.4), with a 3.0-fold variation between patients in the 10th and 90th percentiles. Higher value was significantly associated with Tönnis grade 0 (12.2-point increase, P = .025), no prior contralateral hip arthroscopy (17.3-point increase, P = .039), higher preoperative PROMs (0.52-point increase per 1-unit increase, P < .001), and no bone marrow aspirate concentrate or microfracture (33.8-point increase, P < .001). Value was also significantly associated with osteoplasty type and labral treatment technique (P < .05 for both). In contrast, operative year, age, sex, BMI, race, ethnicity, Outerbridge grade, and American Society of Anesthesiologists score were not independently associated with value. A model incorporating these factors as fixed effects and the surgery center as a random effect explained 42.3% of the observed variation in value. Sensitivity analyses revealed that value drivers may vary slightly across PROMs.

Conclusions

This study revealed wide variation in the value of hip arthroscopy that was most strongly explained by osteoplasty type, labral management technique, and preoperative PROMs. In contrast, patient demographic characteristics such as age, sex, and BMI contributed minimal independent variability.

Level of Evidence

Level IV, economic and decision analysis.