Under-twos in Ghana who ate one more food group the day before scored slightly higher on development. Mothers' play and talk went with higher scores at every age checked; fathers' showed no clear link

Few studies have set a young child's food and play side by side, and fathers' part in this is little studied in Ghana. Researchers at McGill University and colleagues, including in Ghana, looked at both in the same households and asked both parents. The paper appeared in BMJ Public Health on 26 October.

The data come from the Mental Health in Ghana study, which visited two-parent households in the Ga South district of Greater Accra between December 2023 and February 2024. This analysis covers the 460 children aged 6 to 36 months for whom development data were collected: 296 aged 6 to 23 months and 164 aged 24 to 36 months.

Mothers said which of seven food groups the child had eaten in the past 24 hours. Each parent separately said which of six activities they had done with the child in the same period: books or picture books, an outing, stories, songs, play, and talking to the child during housework. Mothers then answered a 108-item checklist of milestones, which gives a score where 0 is the average of the tool's reference population.

The children's diets were narrow. The younger group averaged 3.0 food groups out of seven and the older group 3.73, with grains, roots and tubers, and flesh foods the most common. Mothers averaged 2.87 activities out of six and fathers 1.81. Overall development averaged -0.14, a little below the reference population, and 61% of the children had been ill in the previous two weeks.

In the younger group, each extra food group went with a development score 0.11 of a standard deviation higher, after adjusting for things like income, parents' education and recent illness. The confidence interval ran from 0.01 to 0.22, so it only just clears zero. By domain, the link showed in motor and language development, not in cognitive or social-emotional development. In the older group the figure was 0.08, and its interval crossed zero.

Each extra activity reported by a mother went with a score 0.13 higher in the younger group and 0.15 higher in the older group. Fathers' activities showed nothing clear in either group: 0.04 and -0.03, both with intervals spanning zero. Food variety wasn't clearly linked to either parent's activities.

Who answered matters here. Mothers reported the diet, their own activities and the child's milestones, so a mother who sees her child's progress in a good light might also describe her own efforts that way. The authors flag this shared reporting as a possible bias. Fathers reported their own activities but not the milestones, so their answers had no such overlap. To our reading, that could account for some of the gap between mothers and fathers, though the study can't say how much.

The maternal link also depended on the adjustments. Before them the figures were 0.08 and 0.10, and both intervals crossed zero. The authors didn't correct for multiple testing and ask that results close to zero be read cautiously. They also say plainly that their findings don't mean fathers' involvement doesn't matter, pointing to other studies showing it does, and suggest that local ideas of fathers as providers may shape how much of this care they give.

A single survey can't show which came first. A child who is developing quickly might get more talk and play, rather than the other way round. Breastfeeding and growth weren't recorded, and the food and activity answers describe just one day. The authors want longitudinal studies and trials next, including of feeding and play support offered together.