Summary of findings 1. Live birth.

Patient or population: women with threatened miscarriage or a history of recurrent miscarriage Interventions: multiple progestogens (vaginal micronized progesterone, oral micronized progesterone, dydrogesterone and 17‐α‐hydroxyprogesterone) Comparison: placebo and dydrogesterone Outcome: live birth Settings: hospitals
Treatment	Direct evidence		Indirect evidence		Anticipated absolute effects for direct estimate
	RR (95% CI)	Certainty	RR (95% CI)	Certainty	Risk with intervention	Risk with comparator	Risk difference with intervention
Threatened miscarriage
Vaginal micronized progesterone versus placebo	1.03 [1.00, 1.07]	⊕⊕⊕⊕ HIGH	Unavailable	‐	761 per 1000 (vaginal micronized progesterone)	725 per 1000 (placebo)	36 more per 1000 (from 36 fewer to 123 more)
Subgroup analysis: number of previous miscarriages
No previous miscarriages and early pregnancy bleeding	0.99 [0.95, 1.04]	⊕⊕⊕⊕ HIGH	Unavailable	‐	739 per 1000 (vaginal micronized progesterone)	747 per 1000 (placebo)	7 fewer per 1000 (from 37 fewer to 30 more)
One or more previous miscarriages and early pregnancy bleeding	1.08 [1.02, 1.14]	⊕⊕⊕⊕ HIGH	Unavailable	‐	755 per 1000 (vaginal micronized progesterone)	699 per 1000 (placebo)	56 more per 1000 (from 14 more to 105 more)
Dydrogesterone versus placebo	0.98 [0.89, 1.07]	⊕⊕⊕⊝ MODERATEa	Unavailable	‐	816 per 1000 (dydrogesterone)	833 per 1000 (placebo)	17 fewer per 1000 (from 92 fewer to 58 more)
17‐α‐hydroxyprogesterone versus placebo	Unavailable	‐	Unavailable	‐	See comment*	See comment**	See comment***
Oral micronized progesterone versus placebo	Unavailable	‐	Unavailable	‐	See comment*	See comment**	See comment***
Vaginal micronized progesterone versus dydrogesterone	Unavailable	‐	1.07 [0.93, 1.23]	⊕⊕⊕⊝ MODERATEb	See comment*	See comment**	See comment***
Recurrent miscarriage
Vaginal micronized progesterone versus placebo	1.04 [0.94, 1.15]	⊕⊕⊕⊕ HIGH	Unavailable	‐	659 per 1000 (vaginal micronized progesterone)	633 per 1000 (placebo)	25 more per 1000 (from 38 fewer to 95 more)
Dydrogesterone versus placebo	1.00 [0.23, 4.37]	⊕⊝⊝⊝ VERY LOWc	Unavailable	‐	850 per 1000 (dydrogesterone)	850 per 1000 (placebo)	0 fewer per 1000 (from 195 fewer to 255 more)
Vaginal micronized progesterone versus dydrogesterone	Unavailable	‐	1.04 [0.79, 1.38]	⊕⊝⊝⊝ VERY LOWd	See comment*	See comment**	See comment***
17‐α‐hydroxyprogesterone versus placebo	Unavailable	‐	Unavailable	‐	See comment*	See comment**	See comment***
Oral micronized progesterone versus dydrogesterone	Unavailable	‐	Unavailable	‐	See comment*	See comment**	See comment***
*No included studies or there are no events in included studies to estimate the baseline risk. **Absolute risk with intervention cannot be estimated in the absence of absolute risk with the comparator. ***Risk difference cannot be estimated in the absence of absolute risks with intervention and the comparator. CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

^a Direct evidence downgraded ‐1 due to serious limitations in study design.

^b Indirect evidence ‐1 due to serious limitations in study design.

^c Direct evidence downgraded ‐1 due to serious limitations in study design (unclear random sequence generation and allocation concealment) and ‐2 due to and severe imprecision (wide 95% CIs and small number of events).

^d Indirect evidence downgraded ‐1 due to serious limitations in study design (unclear random sequence generation and allocation concealment) and ‐2 due to and severe imprecision (wide 95% CIs and small number of events).