Ten RCTs were included (1800 children) comprising 7 RCTs of continuous supplementation (1502 children) and 3 RCTs of short-course supplementation (298 children).
Methodological scores ranged from 76 to 95 (maximum 100).
Continuous supplementation:
Diarrheal incidence (7 RCTs): zinc supplementation was associated with significantly lower incidence. OR = 0.82 (95% CI: 0.72, 0.93).
Diarrheal prevalence (7 RCTs): zinc supplementation was associated with significantly lower incidence. OR = 0.75 (95% CI: 0.63, 0.88).
Persistent diarrhea (5 RCTs): no significant effect of zinc supplementation was found. OR = 0.67 (95% CI: 0.42, 1.06).
Dysentery (3 RCTs): no significant effect of zinc supplementation was found. OR = 0.87 (95% CI: 0.64, 1.19).
Pneumonia (4 RCTs): zinc supplementation was associated with significantly lower incidence. OR = 0.59 (95% CI: 0.41, 0.83).
Incidence of diarrheal illness or pneumonia did not vary significantly between any of the subgroups analyses (age, zinc concentration, nutritional status, and sex). Results displayed in graphical format only.
Overall there was no statistically significant interaction between methodological quality and any of the outcomes apart from diarrheal prevalence for which a negative correlation was found (Pearson r = -0.768, P = 0.04).
No significant effect was seen between dose and effect size for diarrheal incidence (Pearson r = 0.655, P = 0.11), diarrheal prevalence (Pearson r = 0.298, P = 0.52); or pneumonia incidence (Pearson r = -0.340, P = 0.66).
Short-course trials:
Diarrheal incidence (3 RCTs): no significant effect of zinc supplementation was found. OR = 0.89 (95% CI: 0.62, 1.28).
Diarrheal prevalence (3 RCTs): zinc supplementation was associated with significantly lower incidence. OR = 0.66 (95% CI: 0.52, 0.83).
Pneumonia (2 RCTs): no significant effect of zinc supplementation zinc was found. OR = 0.74 (95% CI: 0.40, 1.37).