I just looked at the study which is the source for this article. It is published in the Lancet. It also seems to avoid correct sex terms. "Assigned female at birth" occurs a few times, and they generally refer to 'adolescents' or 'participants rather than 'girls'.
However, the term 'girl' slips through twice and (slightly hilariously) the word 'women' occurs 15 times. I think we can probably credit shoddy use of Word's 'find and replace' tool for researchers' acknowledgement of sex in the article.
Interestingly, the research is based on data gathered in the Avon Longitudinal Study of Parents and Children which, judging from its web pages, doesn't seem reticent about correct-sex terms.
Conceivably, the Guardian might justify its language choices by saying that they were honouring the language used by the research published in the Lancet. But that begs the question as to why these researchers felt able to distort the language which appears to have been used in the Avon longitudinal study. (Disclaimer: I haven't read text from the Avon study -- i'm just going by the manner in which its own website speaks of it.)