Link to the original study below, free to access [0]. The key data appear to be this:

>At baseline, 184 (14%) of 1280 children with successful FEV1 measurement in London had a clinically low FEV1 (pre-bronchodilator FEV1 <80% of predicted value) compared with 126 (9%) of 1339 children in Luton (appendix p 8). At the end of the study, these values had dropped to 51 (9%) of 585 children in London and 41 (7%) of 606 children in Luton (appendix p 8).

So the proportion of children with low lung function dropped from 14% to 9% in London, and from 9% to 7% at the comparator site in Luton. It's not that much different, though any improvement is obviously welcome.

https://www.thelancet.com/journals/lanpub/article/PIIS2468-2...