A prospective cohort study of 14-3-3η in ACPA and/or RF-positive patients with arthralgia

Table 2 Univariate and multivariable association of 14-3-3η, ACPA and RF with arthritis development

Univariate logistic regression
Variable	RR (95 % CI)	p value
14-3-3η
Cut-off ≥0.19	2.5 (1.2–5.6)	0.02
Cut-off ≥0.40	1.7 (1.0–2.8)	0.04
Cut-off ≥0.80	2.2 (1.3–3.5)	<0.01
Levels	1.04 (1.01–1.07)	0.01
Multivariable logistic regression
Variable	RR (95 % CI)	p value
Categorical variable:
RF	Reference
ACPA	7.9 (1.9–32.4)	<0.01
RF and ACPA	15.0 (3.8–59.7)	<0.01
Adding 14-3-3η to the above categorical variable
14-3-3η ≥0.19	1.6 (0.7–3.5)*	0.25
14-3-3η ≥0.40	1.2 (0.7–1.8)	0.52
14-3-3η ≥0.80	1.3 (0.8–2.1)	0.36
14-3-3η levels	1.01 (0.98–1.04)**	0.50

Abbreviations: ACPA anti-citrullinated protein antibodies, RF rheumatoid factor, RR relative risk, CI confidence interval
*Interpretation of the relative risk, example: subjects that are 14-3-3η positive at the 0.19 cut-off point have a 1.6 times higher risk of developing arthritis, given the knowledge that these subjects must be either RF or ACPA positive
**For continuous variables the relative risk conveys the higher risk of developing arthritis per ascending unit of the independent variable, in this case 14-3-3η

ISSN: 1478-6362