You are missing information there. Lee could see the evidence was not quite right. So he agreed that he would assemble a team to examine the medical records and share the results whether or not they were favourable to Letby.
If the science had shown murder was likely, so be it. But leaving her locked up without sufficient evidence was obviously wrong.
Anyone working in science or academia or medical research will recognise what you call "collating other people's work" as a review paper. This is where you collate, review and synthesise all relevant published works on a topic. It is a standard method in science, and every medical journal is full of such papers. They are particularly important in studying rare and dangerous phenomena, since you can't usefully hang around watching on the off-chance one will happen, and you can't inflict them as an experiment.
So if you meant that to imply there was anything wrong with Lee's paper for using this method, you are wrong there.
If new research on such an area is wanted, the way to do it is to look at papers published since the last comprehensive review paper. Gathering the new data added since 1989 would be a very obvious and necessary step - you refer to it as an update, and yes, we would expect such updates from a scientist focusing on the data. It would be odd for Lee to ignore 35 years worth of data.
You seem to be criticising Lee here for following standard scientific methods. I have seen that reaction online a lot, and I think people reproducing it might want to pause and think. Why are people so critical of someone following standard scientific processes - is it because they are unaware of these processes? Because they don't like the results?
There's nothing at all wrong with Lee's words or the approach that you describe there. I can't over-emphasise how standard and appropriate Lee's approach is, in his scientific field.