Skin-to-skin contact for the prevention of neonatal hypoglycaemia: a systematic review and meta-analysis

Table 3 GRADE assessment

Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	№ of participants (studies) Follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with standard care	Risk difference with skin-to-skin contact
Neonatal hypoglycaemia (study-defined)	922 (7 RCTs)	⨁⨁◯◯ Low^a	RR 0.32 (0.13 to 0.76)	163 per 1,000	111 fewer per 1,000 (141 fewer to 39 fewer)
Special care nursery or neonatal intensive care nursery admission for hypoglycaemia	816 (1 observational study)	⨁◯◯◯ Very low^b	OR 0.50 (0.25 to 1.00)	83 per 1,000	40 fewer per 1,000 (61 fewer to 0 fewer)
Duration of initial hospital stay after birth	3437 (31 RCTs)	⨁◯◯◯ Very low^c,d,e	-	Comparator	MD 2.37 days fewer (3.66 fewer to 1.08 fewer)
Exclusive breastmilk feeding from birth to discharge	1250 (1 observational study)	⨁◯◯◯ Very low^f	OR 4.30 (3.19 to 5.81)	465 per 1,000	324 more per 1,000(270 more to 370 more)

CI Confidence interval, MD Mean difference, OR Odds ratio, RR Risk ratio
Explanations
^aDowngraded two levels of risk of bias as four of the seven included studies were at unclear risk of selection bias, one study was at high risk of selection bias, none of the seven studies provided a detailed protocol for measuring blood glucose or explained how blood glucose was measured, and only four of the studies provided a definition for neonatal hypoglycaemia
^bDowngraded one level for imprecision due to insufficient sample size
^cDowngraded two levels for risk of bias as 25 out of 31 studies were at high risk of bias for at least one domain
^dDowngraded two levels for inconsistency due to unexplained substantial heterogeneity (I² = 90%)
^eDowngraded one level for publication bias due to asymmetry in the funnel plot (p = 0.0159 Egger's test)
^fDowngraded two levels for risk of bias as overall study quality was weak

ISSN: 1471-2393