Quality assessmentSummary of findingsImportance
No. of patientsEffectQuality
No. of studiesDesignLimitationsInconsistencyIndirectnessImprecisionOther considerationsDiet and nutrition counsellingControlRelative (95% CI)Absolute
Gestational weight gain (kg) (better indicated by lower values)
9Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriousNo serious imprecisionNone12211215MD 3.36 lower (4.73 to 1.99 lower)⊕⊕⊕⊖

Moderate
Important
Birthweight (kg) (better indicated by lower values)
9Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriouscSeriousdReporting biase13651372MD 0.07 lower (0.21 lower to 0.07 higher)⊕⊖⊖⊖

Very low
Important
LGA
5Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriouscNo serious imprecisionReporting biase134/1196 (11.2%)203/1182 (17.2%)RR 0.78 (0.51 to 1.19)38 fewer per 1000 (from 84 fewer to 33 more)⊕⊕⊖⊖

Low
Critical
SGA
3Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriouscNo serious imprecisiondNone80/1124 (7.1%)79/1128 (7%)RR 1.02 (0.75 to 1.37)1 more per 1000 (from 18 fewer to 26 more)⊕⊕⊕⊖

Moderate
Critical
Pre-eclampsia
6Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriouscNo serious imprecisionNone99/1309 (7.6%)150/1315 (11.4%)RR 0.67 (0.53 to 0.85)38 fewer per 1000 (from 17 fewer to 54 fewer)⊕⊕⊕⊖

Moderate
Critical
Gestational hypertension
2Randomised trialsNo serious limitationsa,bNo serious inconsistencyNo serious indirectnessNo serious imprecisionNone4/139 (2.9%)14/143 (9.8%)RR 0.3 (0.1 to 0.88)69 fewer per 1000 (from 12 fewer to 88 fewer)⊕⊕⊕⊕

High
Critical
GDM
2Randomised trialsNo serious limitationsa,bNo serious inconsistencyNo serious indirectnessSeriousfReporting biasg11/139 (7.9%)22/146 (15.1%)RR 0.52 (0.27 to 1.03)72 fewer per 1000 (from 110 fewer to 5 more)⊕⊕⊖⊖

Low
Critical
Preterm delivery
4Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriouscNo serious imprecisionReporting biasg49/744 (6.6%)70/730 (9.6%)RR 0.68 (0.48 to 96)31 fewer per 1000 (from 50 fewer to 9110 more)⊕⊕⊖⊖

Low
Critical
Caesarean section
5Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriouscNo serious imprecisionNonee387/1135 (34.1%)416/1138 (36.6%)RR 0.93 (0.84 to 1.04)26 fewer per 1000 (from 58 fewer to 15 more)⊕⊕⊕⊖

Moderate
Critical
Induction of labour
4Randomised trialsNo serious limitationsa,bNo serious inconsistencyhSeriouscNo serious imprecisionReporting biasg370/1145 (32.3%)326/1132 (28.8%)RR 1.12 (0.99 to 1.27)35 more per 1000 (from 3 fewer to 78 more)⊕⊕⊖⊖

Low
Critical
Post-partum haemorrhage
2Randomised trialsNo serious limitationsNo serious inconsistencySeriouscNo serious imprecisionReporting biasi32/606 (5.3%)37/626 (5.9%)RR 0.80 (0.57 to 1.42)6 fewer per 1000 (from 26 fewer to 25 more)⊕⊕⊖⊖

Low
Critical
Intrauterine death
2Randomised trialsNo serious limitationsaNo serious inconsistencySeriousSeriousdNone0/647 (0%)6/673 (0.9%)RR 0.15 (0.02 to 1.2)8 fewer per 1000 (from 9 fewer to 2 more)⊕⊕⊖⊖

Low
Critical
Admission to NICU
2Randomised trialsNo serious limitationsa,bSeriousjSeriouscSeriousdNone400/983 (40.7%)374/979 (38.2%)RR 0.98 (0.66 to 1.47)8 fewer per 1000 (from 130 fewer to 180 more)⊕⊖⊖⊖

Very low
Critical
Shoulder dystocia
3Randomised trialsNo serious limitationsa,bNo serious inconsistencySeriousNo serious imprecisionNone14/1049 (1.3%)37/1033 (3.6%)RR 0.38 (0.21 to 0.69)22 fewer per 1000 (from 11 fewer to 28 fewer)⊕⊕⊕⊖

Moderate
Critical
Birth trauma
2Randomised trialsNo serious limitationsaNo serious inconsistencySeriousSeriousdNone3/982 (0.3%)9/979 (0.9%)RR 0.36 (0.11 to 1.23)6 fewer per 1000 (from 8 fewer to 2 more)⊕⊕⊖⊖

Low
Critical
Neonatal hypoglycaemia
3Randomised trialsNo serious limitationsNo serious inconsistencySeriouscNo serious imprecisionReporting biasi119/946 (12.6%)107/931 (11.5%)RR 1.05 (0.83 to 1.33)12 more per 1000 (from 38 more to 20 fewer)⊕⊕⊖⊖

Low
Critical
a

Poor information about allocation concealment, which was assessed as not strongly significant.

b

Poor information about blinding of subjective outcomes, which was assessed as not strongly significant.

c

Women with GDM.

d

Wide CI.

e

Slight skew in funnel plot for given outcome.

f

Non-significant RR result, with large disproportion between study groups.

g

Meaningful skew in funnel plot graph.

h

High heterogeneity (> 50%), which can be explained by the diversity in the health of the study populations.

i

Difficult to interpret as only two studies.

j

Qualitative difference in effect.

From: Appendix 13, Grading the quality of evidence for the primary and clinically important outcomes for the effectiveness of dietary interventions in pregnancy

Cover of Interventions to Reduce or Prevent Obesity in Pregnant Women: A Systematic Review
Interventions to Reduce or Prevent Obesity in Pregnant Women: A Systematic Review.
Health Technology Assessment, No. 16.31.
Thangaratinam S, Rogozińska E, Jolly K, et al.
Southampton (UK): NIHR Journals Library; 2012 Jul.
© 2012, Crown Copyright.

Included under terms of UK Non-commercial Government License.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.