Who decides, and what a partner is actually for
The profession's own consent guidance does not mention the partner once. That is the right answer, and it defines the job.
This guide is written for a couple. So it should be honest about something at the start.
ACOG's committee opinion on informed consent and shared decision making in obstetrics and gynecology runs to more than forty thousand characters. Searched end to end, the word "partner" appears zero times. So does "support person". So does "companion". "Family" appears, and where it does, it appears like this:
A patient who is pregnant is fully capable of making medical care decisions during pregnancy and during labor and delivery, even if those decisions are in disagreement with obstetrician-gynecologists or family members, involve withdrawal of life-sustaining treatment, or may adversely affect the health of the fetus.
That is not an oversight, and it is not a criticism to point it out. It is the correct ethical position. The pregnant patient is the person whose body it is and the person who decides. Her decision stands even against her family's wishes, and a consent process that let anyone else answer for her would not be consent.
The World Health Organization's recommendation is phrased with the same care: a companion of choice is recommended for all women throughout labour and childbirth.
Of choice. Which may be her partner, and may deliberately not be. The presence is hers to grant, not the partner's to assume.
Help her ask. CDC's own guidance to partners says it in those words:
Offer to go with her to visit a health care provider and help her ask questions.
Bridge the gap with the staff. A synthesis of 51 studies found that companions do informational work, translating between the woman and the clinical team.
Speak up for her. The same synthesis:
companions were advocates, which means they spoke up in support of the woman.
Watch for what she cannot. The warning signs topic later in this part is written for the partner for exactly this reason.
The distinction that runs through every partner topic in this guide is between a partner who helps her get the information, and a partner who supplies the answer. The first is the whole job. The second is not available.
The reason that job is worth doing is measurable. In the Giving Voice to Mothers survey, a large self-selected US sample of 2,700 respondents, a companion analysis of 2,490 of them found that those who identified as Black reported more procedures done without consent during perinatal care (adjusted odds ratio 1.89, 95% CI 1.35 to 2.64) and during vaginal births (1.87, 95% CI 1.23 to 2.83) than white participants. People of other minoritised identities reported more pressure to accept procedures (1.55, 95% CI 1.08 to 2.20). Those differences held after accounting for where the birth happened, who provided the care, and the prenatal context. The consent process itself fails unequally.
And the one study that examined a partner characteristic found something else entirely. In the main analysis of that same survey, covering the 2,138 women who completed every section, 17.3% reported at least one form of mistreatment, rising to 28.1% among those who gave birth in hospital. Its finding on partners, quoted exactly:
Regardless of maternal race, having a partner who was Black also increased reported mistreatment.
That is not the sentence anyone wants to write in a partner section. It is what the study found, and softening it would be the opposite of what this guide is for. Both analyses come from the same self-selected online survey rather than from a representative sample of US births.
One limitation, stated plainly: nearly all of the research quoted in these partner topics studied fathers and male partners. Same-sex partners, non-birthing mothers and non-binary partners are almost absent from the quantitative evidence. Where this guide says "partner", the underlying study usually measured fathers, and that gap is the researchers' rather than yours.