It is worth noting that the 0.5% receiving any breastmilk at 12 months is from the Lancet 2016 meta-analysis which relied on estimates, because most high income countries only gather stats on breastfeeding rates at 6 OR 12 months, rather than both. For the UK data, they used the Infant Feeding Survey from 2010, so it is quite out of date now (and did only collect info up to 9 months and I can't find the 0.5% figure anywhere in it, so I assume it is an estimate). Due to a lack of funding, the survey did not run in 2015 or 2020, despite running every 5 years since 1975, but there was one comissioned in 2023 based on babies born in December 2023 and this has not yet been published - I expect we might see it by the end of this year. They are collecting data up to 8-10 months so we should get a good estimate of BF rates at 8-10 months.
Based on what I've noticed on parenting forums between my 2008 and 2018/2021 babies, in 2010 most people thought 6 months was the "recommended" length of breastfeeding, with 12 months (NHS actual recommendation at the time) being seen more as a deadline to have stopped by. I think that has shifted and it is now seen as 12 months is the usual, and then people's views seem to vary but somewhere around 18m-2.5 tends to be the cut off for being "weird" if you haven't stopped by then. People used to refer to "walking and talking" and now talk about "saying full sentences". So I think that the 0.5% may be reflective of this - with breastfeeding prevalence (how long people feed for regardless of other foods/milks) you tend to see a very sharp drop off in the first few weeks (I predict this will have got worse - I don't think breastfeeding support these days is any better) but then it's fairly stable until weaning takes over at ~6-9 months and then it starts to drop off more sharply again - I think it's likely this curve will soften, due to slower weaning and more acceptance of BF past 12 months.
You can see a similar trend if you look at the older surveys too. BF rates often drop off in the first few weeks because it can be difficult to establish BF. This seems to be higher when more mothers initiate BF in the first place, which shows very clearly IMO the consequences of heavily promoting BF without investing in adequate support. But then things stay fairly stable until you start to reach an age where you can see (and is often separately measured in the survey) solids start to overtake milk in the baby's diet, or when mums go back to work if this happens before 6 months of age. Statistically, once you pass 6 months maternity leave, return to work doesn't significantly affect breastfeeding rates. But then past the point where solids are established, you tend to get a fairly stable rate again until you get to an age where BF is considered culturally unacceptable, at which point they dwindle to almost nothing. You can see that pattern in every country that takes stats.
IME it is quite a specific demographic who continue to breastfeed as well, there is often some kind of driver which overrides the cultural norm because the most influential aspect of how long you breastfeed for is generally how long you see people breastfeeding around you, including how long you were breastfed yourself.
You tend to get really 3-4 rough groups which of course overlap and are not exclusive - but they are mothers who are highly educated/intelligent/possibly slight overthinkers and tend to read up on everything and follow this more so than following what is the "norm" (overrepresented on MN, or at least they were 15 years ago!) mothers who lean towards a more "natural" or eco-friendly/hippy/alternative type lifestyle, mothers influenced by a different cultural norm where breastfeeding is more accepted, and then mothers who have extremely difficult to settle babies and/or clingy, sensitive toddlers, who perhaps would not have sought out attachment parenting or continued breastfeeding otherwise, but their baby/toddler seems to need that more intense level of nurturing, and it's somewhat of a relief when they find that other people are doing this and it works for them too and nothing terrible has happened to their older children. Often those babies/toddlers later turn out to be neurodivergent but not always.
So I don't think it's hugely surprising that it would be a minority, because all of these things are quite niche, although I do also think the 0.5% figure is suspect. The figure at 9 months (in 2010) is 24% so it seems like an extremely sharp drop off.