Table 5. Multivariate analysis.

Variables (n = 573)	β (SE)	OR (95% CI)	P-value
Risk factor(s) identified (≥1)
Gender (female versus male*)	1.1 (0.6)	2.9 (0.8-10.2)	0.098
Residence area (rural versus urban*)	−1.5 (0.6)	0.2 (0.060-0.697)	0.011
Employment status (employed versus unemployed*)	1.2 (0.6)	3.4 (0.9-12.6)	0.062
Early symptom(s) identified (≥1)
Educational level (university versus scholar*)	1.2 (0.5)	3.4 (1.1-9.8)	0.023
Residence area (rural versus urban*)	−0.8 (0.4)	0.4 (0.1-1.04)	0.063
Diabetes (yes versus no*)	−1.4 (0.5)	0.2 (0.07-0.68)	0.008
Obesity (yes versus no*)	1.3 (0.7)	3.7 (0.8-17.2)	0.093
Consequence(s) identified (≥1)
Gender (female versus male*)	1.8 (0.7)	6.6 (1.6-26.9)	0.008
Residence area (rural versus urban*)	−1.8 (0.6)	0.1 (0.04-0.5)	0.005
Income level (medium versus low*)	1.4 (0.6)	4.1 (1.04-15.7)	0.043
Income level high versus low*)	1.7 (1.1)	5.5 (0.58-52.03)	0.137
Taking a patient to a hospital
Educational level (university versus school*)	0.9 (0.4)	2.5 (1.1-5.5)	0.030
Employment status (employed versus unemployed*)	0.6 (0.2)	1.8 (1.1-3.1)	0.028
Diabetes (yes versus no*)	−0.9 (−0.3)	0.4 (0.18-0.85)	0.018

β, Beta; SE, standard error; OR; adjusted ratio; CI, confidence interval.

Logistic regression taking identification of stroke risk factors, stroke early symptoms, stroke consequences, response if faced with stroke as the dependent variables and sociodemographic factors (gender, residence area, educational level, employment status, and income level) as independent variables.