Skip to main content

Table 3 Association of acute seizures with binary behavioural and emotional problems as the outcome/dependent variable

From: Prevalence, risk factors and behavioural and emotional comorbidity of acute seizures in young Kenyan children: a population-based study

Behavioural and emotional problems Number without acute seizures (N = 3073) Acute seizures (200) Adjusted RR (95% CI)a AICa Adjusted RR (95% CI)b AICb AICb < AICa
Total problems 366 (12%) 54 (27%) 2.24 (1.75–2.88) 0.78 2.40 (1.61–3.58) 0.41 Yes
Externalising problems 302 (10%) 36 (18%) 1.76 (1.28–2.41) 0.67 2.09 (1.40–3.13) 0.41 Yes
Internalising problems 648 (21%) 80 (40%) 1.92 (1.60–2.31) 1.11 1.99 (1.40–2.91) 0.41 Yes
CBCL syndromes
 Emotionally reactive 299 (10%) 32 (16%) 1.67 (1.19–2.33) 0.67 1.38 (0.86–2.21) 0.41 Yes
 Anxious or depressed 362 (12%) 55 (28%) 2.38 (1.86–3.05) 0.77 2.18 (1.50–3.18) 0.41 Yes
 Somatic problems 622 (20%) 73 (37%) 1.77 (1.45–2.15) 1.08 2.15 (1.52–3.05) 0.41 Yes
 Withdrawn 327 (11%) 33 (17%) 1.58 (1.14–2.20) 0.71 1.27 (0.78–2.05) 0.41 Yes
 Sleep problems 100 (3%) 14 (7%) 2.06 (1.19–3.57) 0.30 1.82 (1.00–3.35) 0.41 Yes
 Attention problems 461 (15%) 41 (21%) 1.33 (1.00–1.76) 0.89 1.52 (1.04–2.23) 0.41 Yes
 Aggressive behaviour 166 (5%) 23 (12%) 2.04 (1.36–3.07) 0.44 1.87 (1.19–2.94) 0.41 Yes
DSM-oriented scales
 Affective problems 163 (5%) 26 (13%) 2.41 (1.65–3.53) 0.44 2.05 (1.18–3.54) 0.39 Yes
 Anxiety problems 363 (12%) 45 (23%) 1.91 (1.45–2.52) 0.77 1.64 (1.14–2.35) 0.67 Yes
 Pervasive developmental problems 157 (5%) 16 (8%) 1.59 (0.98–2.60) 0.42 0.85 (0.42–1.73) 0.37 Yes
 Attention deficit or hyperactivity problems 146 (5%) 17 (8%) 1.72 (1.06–2.77) 0.40 1.38 (0.69–2.75) 0.34 Yes
 Oppositional defiant problems 71 (2%) 5 (3%) 1.02 (0.42–2.51) 0.22 1.01 (0.31–3.30) 0.18 Yes
  1. AIC Akaike’s information criterion, CBCL Child Behaviour Checklist, CI confidence interval, RR risk ratio
  2. aModel one, in which RR is adjusted for child’s age and sex and region of residence. bModel two, in which RR is adjusted for known non-seizure risk factors for behavioural and emotional problems namely pregnancy problems, perinatal complications, maternal age, maternal employment, house status and head injury, in addition to those adjusted in model one. The italicised RR reached a cut-off p-value of 0.05 or less, or their lower CI did not cross over 1 
  3. Behavioural and emotional problems were entered into the model as binary variables
\