Table 2.

Independent impact of daily fever burden (> 37.9 °C) within the study period of 16 days on poor 3-month functional outcome in 361 patients

Variables	Adjusted OR	95%-CI	P-value
Fever burden > 37.9 °C, °C × h	1.07	1.01–1.32	0.027
Admission Hunt & Hess score	1.64	1.28–2.09	< 0.001
Age, years	1.06	1.04–1.09	< 0.001
Delayed cerebral ischemia	2.79	1.33–5.85	0.007
Sepsis/bacteremia	3.07	1.33–7.10	0.009
Pneumonia	1.42	0.78–2.57	0.253
Urinary tract infection	1.33	0.67–2.64	0.418
Ventriculitis	0.87	0.41–1.87	0.722
Diclofenac, daily sum (mg)	0.995	0.991–0.999	0.010
Naproxen, daily sum (mg)	0.999	0.998–1.001	0.332
Paracetamol, daily sum (mg)	1.00	1.00–1.00	0.163
Metamizole, daily sum (mg)	1.00	0.99–1.00	0.211
Pethidin, daily sum (mg)	0.999	0.995–1.003	0.621
Treatment with feedback device, per day	1.60	0.80–3.19	0.180

Multivariable logistic regression analysis was done with a generalized estimating equation model with an independent correlation matrix to account for repeated measures; the dependent variable (poor 3-month outcome, modified Rankin Scale score 3–6) was used as binary variable. The model was corrected for known outcome predictors (Hunt & Hess score, age, and delayed cerebral ischemia), infections as well as daily sums of used pharmacologic or physical fever treatment. Respective OR (95%-CI) are given

CI, confidence interval, OR, odds ratio