Table 3 Baseline ePIRS quartile ranking predicts likelihood of advancing HY status.

From: The emerging postural instability phenotype in idiopathic Parkinson disease

 

ePIRS quartile

Number in quartile

Quartile value (Score, Baseline)

Within group hazard ratio (HR) compared to PD quartile I (95% CI for HR)

z

Pr (>|z|)

Hazard ratio (HR) compared to controls (95% CI for HR)

 

Controls

186

Median = 2

Controls I derivation reference group

   

A: Derivation Sample (N = 380)

I

111

≤2

   

5.3 (1.7–16.3)***

II

85

3–6

2.03 (1.00–4.11)

1.97

0.049

10.7 (3.7–31.5)****

III

78

7–10

3.34 (1.71–6.54)***

3.52

0.00043

17.7 (6.2–51.0)****

IV

76

≥11

8.28 (4.48–15.29)****

6.75

<0.00001

44.1 (15.9–122.2)****

B: IPD Sample (N = 79)

I (Ref)

86

Derivation sample de-novo quartile I reference group

(Modified)

I

25

≤2

1.90 (0.51–7.02)

0.96

0.34

9.7 (1.9–48.4)**

II

18

3–6

2.52 (0.68–9.29)

1.38

0.17

12.7 (2.6–63.6)***

III

21

7–10

2.53 (0.68–9.47)

1.38

0.17

13.2 (2.6–67.3)***

IV

14

≥11

16.18 (6.34–41.23)****

5.83

<0.00001

88.0 (23.2–333.5)****

C: GPD Sample (N = 141)

I (Ref)

111

Derivation sample de-novo quartile I reference group

5.3 (1.7–16.3)***

I

20

≤2

2.81 (0.56–14.06)

1.26

0.21

21.6 (2.5–185.3)***

II

29

3–6

0.94 (0.11–7.87)

−0.056

0.96

7.4 (0.6–95.1)

III

43

7–10

3.84 (1.15–12.83)

2.19

0.029

28.5 (4.7–174.1)***

IV

60

≥11

13.26 (5.13–34.28)****

5.34

<0.00001

103.2 (18.7–569.9)****

  1. Progression to HY ≥ 3, by quartile, compared to healthy controls (right-most columns) and compared to Quartile I. Reference Quartile I values in all cases are taken from the de-novo sample. A: Significant, and progressive impact of quartile on risk development of HY ≥ 3 is noted in the de-novo PD derivation sample. B: In the IPD validation sample, ePIRS scores in the fourth quartile (≥11) were associated with a significantly higher likelihood of progression to HY ≥ 3. C: In the GPD cohort, once again membership in Quartile 4 confers a higher risk of progression to HY ≥ 3 compared to de-novo subjects with ePIRS scores ≤ 2.
  2. **p ≤ 0.01, ***p ≤ 0.005, ****p ≤ 0.0001.