We are repeatedly told that we have more data about our practices and patients than at any time in the past. So why do we have more difficulty in producing reports for the QOF QP indicators than ever? Gavin Jamie explores
I admit that I am a data nerd but what I really want is the raw data. Like cooking with raw ingredients it is much easier to control which ingredients go in and the recipe that is used.
We are repeatedly told that we have more data about our practices and patients than at any time in the past. So why do we have more difficulty in producing reports for the QOF QP indicators than ever? Gavin Jamie explores
I admit that I am a data nerd but what I really want is the raw data. Like cooking with raw ingredients it is much easier to control which ingredients go in and the recipe that is used.
Much of the provided data sources report only at a high level. A recent example has been the primary care website (primarycare.nhs.uk). This is a data 'ready meal'. Most of the recipes are obscure and the ingredient list is less than reassuring. It is devoid of context and personally I find it almost impossible to get any useful information out of it.
It is by no means unique (NHS Comparators has been bewildering for a number of years now) but it is one of the most recent. These services say much more about the organisations that have set them up than they do about GPs and practices. There is normally a fairly specific agenda behind these services.
It is not just about control – the fact that you've put together the data yourself means you can have a much better idea of what the results actually mean. Context is everything in these cases and nobody knows the context of your patients better than you.
Most GPs think in term of what happens to individual patients and having the raw data allows you to drill down easily to that level. Large sites with public data deliver very little meaning to the majority of clinicians. It is much easier to understand what is happening if you can give a few examples, and rather easier to establish causes.
It would not be so bad if I could get at the raw data in a usable form from another source. In fact I cannot even easily extract data from my own practice system. Data about hospital use, the basis of QP indicators, is delivered as a PDF file. As one informatics email list user wrote, PDF is where data goes to die. The files are almost unreadable with anything except your eyes and are impossible to combine with other data. A few years ago I could download a customised CSV file. The problems I had with the formatting of that file seems trivial now.
Of course the situation is not unique to GPs. CCGs are having their own problems getting hold of patient data. In theory it should be simpler for GPs who already hold personal data about the patients in question.
Access to raw data to GPs is very limited. Public data sets, particularly QOF and prescribing data, are hugely better than it used to be. However when it comes to analysis of patient data we are being encouraged to believe that our data needs to be taken out of the practice and then served back up after factory processing. Looking at the data has become outsourced, and practices risk becoming deskilled in this area.
I don't believe that is true and that is what I hope that my data site says about me. There is alway some agenda in how the data is displayed. Mine is about avoiding league tables of QOF scores, but you can download all of the data and do what you want with it. Please do.
Home cooking can be fun and nobody knows your data like you do. You the tricky bit is finding all of the ingredients.
