Skip to main content

Table 1 Summary of meta-analysis estimates of effects, confidence intervals, and certainty of evidence

From: Pregnancy-related complications and perinatal outcomes following progesterone supplementation before 20 weeks of pregnancy in spontaneously achieved singleton pregnancies: a systematic review and meta-analysis

 

Relative effect odds ratio (95% CI)*

Anticipated absolute effect, per 1000 patients† (95% CI)

No. of participants (studies)

Certainty of evidence

Number needed to treat (95% CI)

Progestogen

Placebo/no treatment

Difference

Maternal outcome

 Pre-eclampsia

0.64 (0.42, 0.98)

14

22

8 fewer (15 fewer to 1 fewer)

3982 (3 RCTs)

, moderate4

125 (67 to 1000)

 Gestational diabetes mellitus

1.02 (0.79, 1.32)

61

58

3 more (12 fewer to 16 more)

3438 (2 RCTs)

, low1, 4

Not calculated (non-statistically significant)

 Placenta previa

0.67 (0.32, 1.39)

6

8

2 fewer (7 fewer to 2 more)

3590 (3 RCTs)

, vey low1, 2, 4

Not calculated (non-statistically significant)

 Placental abruption

0.58 (0.30, 1.13)

7

12

5 fewer (10 fewer to 1 more)

3219 (2 RCTs)

, very low1, 2, 3, 4

Not calculated (non-statistically significant)

Perinatal outcome

 Preterm birth

1.02 (0.86, 1.20)

113

107

4 more (14 fewer to 15 more)

4782 (9 RCTs)

, low3, 4

Not calculated (non-statistically significant)

 Low birth weight

0.57 (0.34, 0.95)

52

81

29 fewer (53 fewer to 1 fewer)

935 (4 RCTs)

, moderate4

34 (19 to 1000)

 Small for gestational age

0.86 (0.58, 1.27)

52

58

6 fewer (29 fewer to 9 more)

4145 (4 RCTs)

, very low1, 3, 4

Not calculated (non-statistically significant)

 Perinatal mortality

1.05 (0.58, 1.90)

9

8

1 more (2 fewer to 5 more)

4632 (7 RCTs)

, very low1, 2, 3, 4

Not calculated (non-statistically significant)

  1. *Significant results are in bold
  2. †Data obtained directly from study sample (studies reporting outcome data)
  3. 1 Wide confidence intervals (imprecision)
  4. 2 Small number of events (imprecision)
  5. 3 High percent of patients were lost to follow-up in Elgergawy, 2019 (assumed risk of bia)
  6. 4 Possibility of publication bias