Table 1 Statistical analysis plan summary
Objective | Outcome | Hypothesis | Method of analysisa | |
|---|---|---|---|---|
Name | Type | |||
Primary objective | ||||
To compare pain levels at 1 year | Pain (VAS) | Continuous | Osteochondroplasty will reduce pain compared to lavage | Multiple linear regression |
Secondary objectives 1 | ||||
To compare patient-reported health-related quality of life | Hip function (HOS) | Continuous | Osteochondroplasty will improve health-related quality of life, function, and utility compared to lavage | Multiple linear regression |
Hip-specific disease on hip function (iHOT-12) | Continuous | |||
Physical health (SF-12 PCS) | Continuous | |||
Mental health (SF-12 MCS) | Continuous | |||
Health Utility (EQ-5D) | Continuous | |||
Secondary objective 2 | ||||
To compare hip complications | Hip-related complications (e.g. re-operation) | Binary | Osteochondroplasty will reduce rate of re-operations compared to lavage | Multiple logistic regression |
Subgroup analysis | ||||
Hip impingement severity: mild (alpha angle < 60 – > 50 degrees), moderate (alpha angle > 60 – < 83°), severe (alpha angle > 83°) | Pain (VAS) | Continuous | Patients with severe impingement at baseline will have the greatest improvement with the osteochondroplasty procedure compared with those with moderate to mild impingement | Multiple linear regression |
Gender: male, female | Pain (VAS) | Continuous | The osteochondroplasty procedure will perform better in males | Multiple linear regression |
Cartilage status (based on Tonnis and Heinecke classification): grades 3 and 4, grades 1 and 2 | Pain (VAS) | Continuous | Osteochondroplasty will perform worse in patients with worse cartilage status (i.e. grades 3 and 4) | Multiple linear regression |
Treatment of the labrum: labral repair, resection | Pain (VAS) | Continuous | Patients receiving a labral repair will perform better than those receiving a resection as part of the osteochondroplasty procedure | Multiple linear regression |
Sensitivity analysis | ||||
Trial site (centre-effects) | Pain (VAS) | Continuous | We do not expect the effect to change substantially when centre-effects are removed from the primary analysis | Multiple linear regression with centre-effects removed |
Missing data effect | Pain (VAS) | Continuous | We do not expect the effect to change substantially without imputation for missing data | Multiple linear regression with complete cases only |
Potential baseline imbalance | Pain (VAS) | Continuous | Results will remain robust after adjusting for potential baseline imbalance on age, any comorbidities, onset of symptoms, and presence of labral tears at initial surgery | Multiple linear regression with complete cases only |