306 new fathers in Shanghai rated how sure they felt about helping with breastfeeding. Their weakest point: telling whether the baby had had enough

Breastfeeding advice is usually aimed at mothers, but the person standing next to her in the first weeks matters too. Studies have linked a partner's support with whether a mother starts breastfeeding and how long she keeps going. What sits underneath that support is confidence: a father who believes he can help is more likely to step in. A study published in Frontiers in Nutrition on 27 October looked at how confident new fathers actually feel, and whether they fall into distinct groups.

Researchers at Shanghai First Maternity and Infant Hospital, part of Tongji University, approached fathers on the postnatal wards between September 2024 and February 2025. Most filled in the questionnaire on the second or third day after the birth, before their partner went home. Of 355 questionnaires, 49 were thrown out for being rushed (under two minutes) or answered in a straight line, leaving 306 fathers with an average age of 31.7. Seven in ten had a bachelor's degree or higher and three quarters were first-time fathers. They answered the 14-item Father Support Breastfeeding Self-Efficacy Scale, a short form developed in Canada, where each item runs from 1 (not at all confident) to 5 (very confident), plus a 17-question knowledge quiz on the benefits and practical skills of breastfeeding.

On paper, the fathers were on board: 95.5% said breastfeeding was important and 51.3% hoped their baby would be exclusively breastfed after discharge. Only 44.8% had gone to a breastfeeding class with their partner before the birth. The average confidence score was 53 out of 70, and the item that scored lowest was "I can help my spouse determine whether the baby is fully fed", at 3.51 out of 5. The authors point to earlier research showing that fathers often miss out on exactly this kind of practical teaching.

A statistical method called latent profile analysis sorted the men into three groups. About two thirds (64.7%) sat in the middle, with an average score of 54. Close to one in five (18.6%) were in a low-confidence group averaging 37, and one in six (16.7%) were in a high group averaging 67. The groups ran largely parallel across the 14 questions, so the authors read them mainly as low, medium and high levels of the same confidence rather than different kinds of father. The low group scored lowest on the knowledge quiz (10.5 out of 17, against 12.5 and 13.6), and only 26.3% of them had been to a class, against 66.7% of the high group. They also more often said their partner's milk supply was low, that she was in pain or that she was worn out.

When all eight factors went into one model, three things held up. Each extra point on the knowledge quiz went with about 10% lower odds of being in the low group rather than the middle one. Fathers who said breastfeeding wasn't important had more than nine times the odds of being in the low group, though only 14 men gave that answer and the range around the figure is very wide. And fathers who called breastfeeding only "generally" rather than "very" important, or who expected anything other than exclusive breastfeeding, were less likely to be in the high group. Class attendance, the mother's pain, tiredness and milk supply didn't stand out once the rest were accounted for, and age, education and income changed nothing.

The study has clear limits, which the authors spell out. It was a one-off survey, so it can't say whether knowledge builds confidence or confident fathers simply learn more, or whether low confidence colours how a father sees his partner's milk supply. Everyone came from one well-resourced hospital in Shanghai, the mothers' pain, tiredness and milk supply were the fathers' own impressions, and nobody checked what the fathers actually did once the family got home or how long the babies were breastfed. The authors suggest the three groups could help nurses spot the fathers who need more help, and say whether teaching fathers alongside mothers changes anything is a question for future trials.