Skip to main content

Table 2 Predictive ability of HHS: Association of HHS scores at 2- and 5-years with the risk of revision in patients with THA

From: Clinically important improvement thresholds for Harris Hip Score and its ability to predict revision risk after primary total hip arthroplasty

 

Hazards of THA Revision (95 % CI)

P-value*

Hazards of THA Revision (95 % CI)

P-value*

 

2-year data

 

5-year data

 
 

N = 3151 with 88 events

 

N = 2461 with 71 events

 

Harris Hip Score categories

    

  ≤ 55 vs. 81–100

4.34 (2.14, 7.95)

< 0.001

3.08 (1.45, 5.84)

0.002

 56–63 vs. 81–100

1.31 (0.36, 3.32)

0.62

1.25 (0.34, 3.20)

0.69

 64–71 vs. 81–100

1.80 (0.59, 4.18)

0.23

0.64 (0.07, 2.36)

0.60

 72–80 vs. 81–100

0 (0, 0.5)

0.006

1.13 (0.37, 2.66)

0.80

Harris Hip Score clinical categories

    

  < 70 vs. 90–100

2.32 (1.32, 3.85)

0.002

1.60 (0.84, 2.85)

0.14

 70–79 vs. 90–100

0 (0, 0.5)

0.005

1.18 (I0.38, 2.79)

0.74

 80–89 vs. 90–100

0.67 (0.33, 1.23)

0.23

0.66 (0.30, 1.30)

0.27

 

N = 1011 with 13 events**

 

N = 583 with 5 events**

 

Improvements in Harris Hip Score

    

 ≤ 0 vs. 51–75

18.10 (1.41, 234.8)

0.02

2.20 (0.02, 29.01)

0.68

 1–20 vs. 51–75

6.21 (0.81, 60.73)

0.10

0.71 (0.005, 8.78)

0.85

 21–50 vs. 51–75

2.19 (0.50, 20.55)

0.41

0.62 (0.12, 3.92)

0.64

  1. *P-value was calculated using a Cox proportional hazards regression model
  2. **The main reason for decrease in the number of patients between absolute score (top) and change HHS score analyses (bottom) was that a lower number completed both surveys (See Fig. 1 for details)