Summary of findings 1. Summary of findings ‐ initiation of penicillin prophylaxis versus placebo.

Penicillin prophylaxis compared with placebo for pneumococcal infection in SCD
Patient or population: children with SCD Settings: outpatients Intervention: initiation of penicillin prophylaxis Comparison: placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	placebo	penicillin
Incidence of S pneumoniae infection Isolated bacterial infection Follow‐up: up to 5 years	90 per 1000	33 per 1000 (14 to 77)	OR 0.37 (95% CI 0.16 to 0.86)	457 (2)	⊕⊕⊝⊝ low1,2	The John trial reported that no pneumococcal isolations had occurred while the children were actually receiving penicillin. However, this was stopped after the participants reached age 3 years and was not continued for the 5‐year duration of the trial.
Deaths Follow‐up: up to 5 years	40 per 1000	4 per 1000 (0.4 to 84)	OR 0.11 (95% CI 0.01 to 2.11)	457 (2)	⊕⊕⊝⊝ low1,2	One of the trials reported no deaths in either group for the duration of the prophylaxis (John 1984). In addition, one child in the placebo group died as a result of fulminant H influenzae, OR 0.11 (95% CI 0.01 to 2.11) (PROPS 1986).
Adverse effects Follow‐up: up to 5 years	See comment	See comment	N/A	457 (2)	⊕⊕⊝⊝ low1,2	No adverse effects were seen in the John trial after penicillin injections (John 1984). The penicillin was well‐tolerated and no confirmed allergic reactions occurred in the PROPS trial (PROPS 1986).
Antibiotic‐resistant organisms isolated Follow‐up: N/A	Outcome not reported				N/A
Requirement of other courses of antibiotics Follow‐up: N/A	Outcome not reported				N/A
Compliance to treatment Follow‐up: average 15 months	See comment	See comment	N/A	215 (1)	⊕⊕⊝⊝ low2,3	An attempt was made to measure compliance via pill counts and urine analysis, but only 66% of appointments were kept and only 31% of the expected numbers of urine samples were obtained (PROPS 1986). The John trial did not measure compliance but attempted to minimise non‐compliance by giving monthly injections.
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; H influenzae : Haemophilus influenzae; N/A: not applicable; OR: odds ratio; SCD: sickle cell disease; Spneumoniae: Streptococcus pneumoniae
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.

1. Downgraded once for risk of bias as the randomisation methodology was unclear in one of the trials and both trials were at risk of bias due to incomplete outcome data.
2. Downgraded once due to imprecision as there were low event rates.
3. Downgraded once due to risk of bias from incomplete outcome data.