Table 3.

Summary of findings and overall assessment of quality of evidence, by outcome

Quality Assessment						Summary of Findings
No. studies and study design	Heterogeneity of results?	Consistent size of effect?	Generalizable to intervention of interest?	Other sources of bias (e.g., major limitations in study design)	N	Statistical method	Effect estimate [95% CI]
Small for gestational age, Overall quality of evidence grade=Low
2 RCTs27,9 (both in HIV positive women)	Low (I2=0%, p=0.51); Neither study showed a significant effect	Both studies null	Both studies provided the same supplement (5,000 IU Vit. A+30 mg β-C+postpartum 200,000 IU.	None	1387	RR (IV, Fixed)	0.89 [0.68, 1.17]
Low birthweight <2.5 kg, Overall quality of evidence grade=Low
5 RCTs9, 27, 29, 49, 87 (3 in HIV+ women)	Moderate (I2=20%, p=0.29); 1 study showed significant effect	All but 1 study had protective direction of effect	Different supplements: VA+ βC, VA, Palm oil. Iron-folate, malaria prophylaxis varied across trials.	All but one study were of high or moderate grade quality	2254	Risk Ratio (IV, Fixed, 95% CI)	0.83 [0.67, 1.01] HIV+: 0.79 [0.64, 0.99] HIV−: 1.11 [0.61, 2.01]
Very low birthweight <2.0 kg, HIV-positive: Overall quality of evidence grade=Low
2 RCTs27,9 (both in HIV-positive women)	Low (I2=0%, p=0.49); Neither study showed a significant effect	Both studies null, effect estimates towards protective	Both used the same supplement (5,000 IU Vit. A+30 mg β-C).	None, all were high grade quality	1486	Risk Ratio (IV, Fixed, 95% CI)	0.73 [0.41, 1.29]
Mean birthweight (kg): Overall quality of evidence grade=Moderate
7 RCT’sb, 9, 27, 29, 49, 87, 92 (3 in HIV positive women)	High (I2=53%, p=0.05); 2 studies showed significant effects in opposite directions	5 were null, 2 showed opposite effects.	Supplements used included VA, VA+βC, βC, and Palm oil as a source of βC.	All but one study were of high or moderate grade quality	2417	MD IV, Random	0.03 [−0.04, 0.10]kg HIV+: 0.04 [−0.01, 0.08] HIV−: 0.02 [−0.17, 0.21]
Lycopene supplementation on mean birthweight (kg): Overall quality of evidence grade= Very low
2 RCTs31, 32	High (I2=65%, p=0.09);	Both studies null, RR’s in opposite directions	One study supplemented with 2 mg/day lycopene, the other with 4 mg/day	Both studies had low grade quality, neither used intent-to treat analysis	410	MD IV Random	0.03 kg [−0.12, 0.18]
Preterm birth <37 weeks: Overall quality of evidence grade=High
7 RCT’sb, 9, 24, 27, 87, 92, 93	Low, (I2=8%, p=0.37)	One study protective, 6 null	Two studies in HIV+ women; 5 in HIV-negative; Differences in supplement type/dosage	All but one study were of high or moderate quality	19,799a	Risk Ratio (IV, Fixed, 95% CI)	Overall 0.99 [0.88, 1.10] HIV+: 0.93 [0.75, 1.14] HIV−: 1.01 [0.89, 1.15]
Early preterm birth <34 weeks: Overall quality of evidence grade=Low
2 RCT’s27,9 (both in HIV-positive women)	High, (I2=78%, p=0.03);	No, one study protective, one null	Both studies used the same supplement (VA+βC)	Both studies of high GRADE quality	1513	RR IV, Random	HIV+ 0.65 [0.20, 2.11]
Lycopene, mean gestational age at delivery: Overall quality of evidence grade=Very low
2 RCTs31, 32	(I2=94%, P<0.0001);	1 null; one positive	1 with 2 mg/day lycopene, the other with 4 mg/day	Both low GRADE quality, neither used intent-to treat analysis	410	MD IV Random	0.40 weeks [−1.1, 1.9]
Stillbirth: Overall quality of evidence grade=High
5 RCT’s (2 HIV+) 9, 25, 29, 88, 94	Low (I2=0%, p=0.98); All studies null	Yes, all studies null	Variation in supplements used; VA, VA+ βC,	All studies of high GRADE quality	106,894a	RR IV Fixed	1.03 [0.97, 1.10] HIV+ 1.07 [0.66, 1.74] HIV− 1.03 [0.97, 1.10]
Fetal loss: Overall quality of evidence grade=High
7 RCT’s (3 HIV+) 9, 25, 27, 29, 54, 88, 94	Low (I2=0%, p=0.78); All studies null	Yes, all studies null	Variation in supplements used;	All studies of high GRADE quality	113,207a	RR IV Fixed	0.99 [0.95, 1.04] HIV+ 0.92 [0.50, 1.67] HIV− 0.99 [0.95, 1.04]
Miscarriage: Overall quality of evidence grade=Moderate
4 RCT’s (2 HIV+) 9, 24, 29, 88	Low (I2=0%, p=0.58); All studies null	Yes, all studies null	Variation in supplements used; VA, βC, VA+ βC	All studies of high GRADE quality	62,138a	RR IV Fixed	0.99 [0.95, 1.04] HIV+ 0.92 [0.51, 1.67] HIV− 0.99 [0.95, 1.04]
Maternal pregnancy related mortality: Overall quality of evidence grade=High
3 RCT’s25, 54, 93	High, (I2=74%, p=0.02); 1 study showed protective effect, 2 were null	1 large study protective; others null	Variation in supplements used; VA, VA+ βC,	All studies of high GRADE quality	160,690a	RR IV Random	0.86 [0.60, 1.24]
Maternal mortality due to sepsis: Overall quality of evidence grade=Low
2 RCT’s54, 93	Low: (I2=0%, P=0.72)	All studies null	Both were pooled estimates of VA and βC arms	All studies of high GRADE quality	81,885a	RR IV Fixed	0.67 [0.31, 1.46]
Maternal mortality due to hemorrhage: Overall quality of evidence grade=Low
2 RCT’s54, 93	Moderate: (I2=44%, p=0.18)	All studies null	Both were pooled estimates of VA and βC arms	All studies of high GRADE quality	81,885a	RR IV Fixed	1.33 [0.61, 2.91]
Maternal mortality due to eclampsia/pre-eclampsia: Overall quality of evidence grade=Low
2 RCT’s54, 93	High: (I2=77%, p=0.04);	All studies null	Both were pooled estimates of VA and βC arms	All studies of high GRADE quality	81,885a	RR IV Random	1.09 [0.24, 4.88]
Maternal mortality due to infection related causes: Overall quality of evidence grade=Low
2 RCT’s54, 93	Low: (I2=9%, p=0.29)	All studies null	Both were pooled estimates of VA and βC arms	All studies of high GRADE quality	81,885a	RR IV Fixed	0.93 [0.53, 1.61]
Lycopene supplementation on pre-eclampsia: Overall quality of evidence grade=Very low
2 RCTs31, 32	(I2=54%, p=0.14);	1 protective, 1 null	2 mg/day lycopene vs. 4 mg/day	Both low GRADE quality, neither intent-to treat	410	RR IV Fixed	0.71 [0.44, 1.14]
Neonatal and early infant mortality: Overall quality of evidence grade= High
4 RCT’s25, 27, 29, 88 (2 in HIV-positive women)	Moderate: overall (I2=40.3%. p=0.31); in HIV-negative I2=47%, p=0.17), in HIV− I2=0%, p=0.86).	All studies null.	Variation in supplements used; VA, βC, VA+ βC,	All studies of high GRADE quality	91,022a	RR IV Fixed	0.97 [0.90, 1.05] HIV+: 0.68 [0.39, 1.17] HIV−: 0.98 [0.90, 1.06]
Infant and neonatal mortality: Overall quality of evidence grade=High
5 RCT’s25, 27, 29, 47, 88	Low; (I2=0%);	All studies null	Variation in supplements used; VA, βC, VA+ βC; variation in postpartum supplements.	All studies of high GRADE quality	132,903a	RR IV Fixed	0.99 [0.94, 1.04] HIV+: 1.03 [0.79, 1.35] HIV−: 0.99 [0.94, 1.04]
HIV transmission from mother to child: Overall quality of evidence grade=Moderate
3 RCT’s9, 27, 29	High; (I2=75%, p=0.02);	2 studies null, 1 study showed significant increase in HIV transmission	Variation in supplements used; two had VA+ βC, 1 had VA	All studies of high GRADE quality	2022	RR IV Random	HIV+: 1.05 [0.78, 1.41]
Maternal anemia (Hb<11 g/dL) during follow-up in pregnancy: Overall quality of evidence grade=High
8 RCT’sb, 21, 22, 24, 64, 86, 87	High: overall (I2=54%, p=0.03%), (Anemic women I2=74%, p=0.004); (Non anemic women I2=0%)	All studies had effect estimates in a prot. direction 3/8 sig.	Variation in supplements used; VA, Differences in provision of iron-folate supplements	All studies but 2 of high GRADE quality	1587	RR IV Random	0.81 [0.69, 0.94] Anemic: 0.72 [0.53, 0.99] Unscreened for anemia: 0.83 [0.69, 0.95]
Maternal severe anemia <8.0 or 8.5 g/dL) during pregnancy: Overall quality of evidence grade=Low
3 RCT’sb, 21, 24	Low; (I2=0%, p=0.03)	All studies null	Both trials used vitamin A	Both studies of high GRADE quality	961	RR IV Fixed	0.93 [0.60, 1.46]
Mean maternal hemoglobin during follow-up in pregnancy: Overall quality of evidence grade=Moderate
8 RCT’sb22–24, 49, 86, 87	Some; (I2=17%, p=0.29)	All studies null but one factorial trial null where both comparisons showed significantly positive effects	Five studies used vitamin A; one used palm oil. All women except those in placebo arms of two factorial trials received iron folate	All but one study of high grade quality	1034	MD IV Fixed	0.35 [0.24, 0.45]
Change in maternal hemoglobin during pregnancy: Overall quality of evidence grade=Moderate
8 RCT’s21, 24, 64, 86, 87	Some; (I2=15%)	All studies but the factorial trial null	Variation in supplements used; all had iron-folate	All but one study of high grade quality	1131	MD IV Random	0.19 [0.06, 0.33]
Mean postpartum maternal hemoglobin (4–6 weeks postpartum): Overall quality of evidence grade=Moderate
4 RCT’s49, 86, 92	Low; (I2=0%, p=0.31)	All studies null,	1 study with 2 arms used βC, 2 studies used VA	All but one study of high grade quality	365	MD IV FIXED	0.01 [−0.24, 0.25]
Infant anemia: Overall quality of evidence grade=Low
2 RCT’s29, 33 (Both in HIV+ women)	Yes; (I2=74%, p=0.05);	1 study protective, 1 null	Variation in supplements used	All studies of high GRADE quality	894	RR IV Random	HIV+ 0.71 [0.49, 1.03], p=0.07
Infant hemoglobin at 4–6 months (g/dL): Overall quality of evidence grade=Low
2 RCT’s92, 95 (1 factorial)	No; (I2=0%, p=0.96)	All three studies null	Variation in supplements used;	All studies of high GRADE	284	MD IV Fixed	−0.13 [−2.29, 2.03]
Infant weight-for-age z score at 6 months: Overall quality of evidence grade=Low
2 RCT’s40, 92 (2 factorial)	Some; (I2=22%)	Yes, all studies null	Variation in supplements used	All studies of high GRADE	1022	MD IV Fixed	0.07 [−0.07, 0.22]
Infant height-for-age z score at 6 months: Overall quality of evidence grade=Low
2 RCT’s40, 92 (2 factorial)	No; (I2=0%)	Yes, all studies null	Variation in supplements used	All studies of high GRADE	1022	MD IV Fixed	0.01 [−0.12, 0.13]

^aEffective sample size for analysis is actually smaller due to inclusion of cluster randomized trials in the estimation of variance

^bDifferent arms of factorial trials counted as separate trials, prot=protective