Table 4

Predicting PIA in EAC patients: results of backward stepwise regression, excluding MSUS variables

Variable

Coding if categorical

B

SE

Wald

P-value

OR (95% CI)


Age

(years)

<50:

0

≥50:

1

-0.80

0.32

6.27

0.012

0.45 (0.24 to 0.84

Sx Durn

(weeks)

n/a

-0.030

0.01

12.64

<0.001

0.97 (0.96 to 0.99)

SJC /72

<1:

0

≥1:

1

1.53

0.31

23.94

<0.001

4.61 (2.50 to 8.43)

JPS

n/a

0.42

0.26

2.66

0.100

1.52 (0.92 to 2.52)

CRP

(g/l)

<5:

0

5 to 14:

1

0.77

0.21

13.48

<0.001

2.16 (1.43 to 3.27)

≥14:

2

ESR

(mm/h)

<24

0

≥24

1

0.79

0.34

5.33

0.021

2.21 (1.13 to 4.32)

ACPA

Neg:

0

Pos:

1

4.89

1.10

21.36

<0.001

132.4 (17 to 1052)

Constant

-2.12

0.52

16.63

<0.001

-


Results of backward stepwise logistic regression analysis to identify independent clinical (non-MSUS) predictors of PIA amongst EA clinic attendees. B values are regression coefficients. CI, confidence interval; OR, odds ratio; SE, standard error of B. Additional abbreviations as per Table 3. See text.

Pratt et al. Arthritis Research & Therapy 2013 15:R118   doi:10.1186/ar4298

Open Data