Table 3.

Effectiveness of interventions on parents’ knowledge

Author (date)	Parent's knowledge (significant results in bold)	Quality
Baker et al 200918	Significant reduction in knowledge scores: 54% reduction in responses that fever was dangerous (p<0.0001, 95% CI 0.43 to 0.65) 28% reduction in responses that child with fever should be woken (p<0.0001, 95% CI 0.19 to 0.39) 30% increase in responses identifying aspirin as inappropriate (p<0.0001, 95% CI −0.42 to 0.16)	***
Broome et al 200319	Knowledge increased significantly more in both groups than in control group at 24–72 h and 1,3 and 6 months p<0.03. No information on the size of the effect provided. Those given individual instruction reported to have higher scores—no p value provided	*
Isaacman et al 199227	Parent recall of medication data higher in all groups than other items but with no significant differences between groups. Recall of signs of improvement increased significantly for both interventions groups compared to controls at exit interview, day 1 and 3 (mean correct responses Exit int. Control 0.9, Verbal 25.3, Verbal and Written 56.9; Day 1 C 33.3, V 54.5, V&W 61.0; Day 3 C 44, V 60, V&W 73.2; all p<0.05). Recall of worrying signs improved significantly compared to controls at exit and on day 1 (Exit int. C 5.5, V 32, V&W 38.1 ; Day 1 C 19.1, V 37.5, V&W 44.5; Both p<0.5). The written and verbal intervention groups performed better than the verbal group at exit interview only for signs of improvement and recall of worrisome signs (p<0.05)	**
Kelly et al 199636	Indirect measurement of knowledge: No significant difference in level of fever at which antipyretics were administered (p=0.91). A significant difference was found in accuracy of antipyretic dose (n=30 incorrect dose preintervention, 18/30 (60%) accurate doses postintervention p=0.04)	**
McCarthy et al 199023	Indirect measurement of knowledge: Reliability of mother's judgements: intervention group were more likely to agree with clinician than control group: 91.7% vs 72.4% (κ 0.50 vs 0.26) Specificity of mother's judgements: Mothers in the intervention group were less likely to score the child's illness as more severe than the paediatrician than those in the control group (Intervention 90% vs 59% control group p<0.0001) Sensitivity of mother's judgements: Serious illness was the outcome used to measure sensitivity. No difference found between intervention and control group (80% vs 90% respectively)	*
Robbins et al 200312	Non-significant reduction in knowledge at 7 months in intervention group	***
Steelman et al 199922	Significantly fewer incorrect responses in intervention group at 2 months (Intervention 10.4 vs Control 11.8; p=0.006) and at 4 months (Intervention 8.5 vs Control 10.3; p=0.002)
Stockwell et al 201031	Significant increase in knowledge/attitude health literacy score (61% p<0.05)	**
Wassmer and Hanlon 199928	Significant increase in parental knowledge of febrile convulsion in the intervention group p<0.05 but these parents children had already had a febrile convulsion. See the original paper for details on size of the effect as these are reported per question asked of parents
Summary	8/9 showed significant increase in knowledge, although implied in 2 studies and 1 study had high risk of bias. 1 paper showed reduction in knowledge at 7 months. 1 qualitative paper