Wanted - The Good Hospital Guide

Making hospital data more user friendly

The NHS must make better use of its data, a hugely valuable resource that needs to be shared for the benefit of research scientists and patients. It has been said so often it’s a cliche - I have banged on in this newsletter about the ridiculous barriers put on place to stymie NHS data sharing schemes and it seems a new data strategy is a key recommendation in Lord Darzi’s report on the NHS coming out today.

It is only fair to say that some progress has been made in recent years in giving patients more control of their own data. The NHS app, while still pretty clunky, has been central to this mission and I hope it will become the key interface between patients and the health service. As someone who has interacted with three different hospitals and a GP practice over the last year I have found it invaluable as a place to read letters from my consultants to my GP, to order repeat prescriptions and to remind myself about upcoming appointments. To my great annoyance, one of the hospitals used an alternative platform for a while, but thankfully they have now seen sense.

But if it is getting easier to interact with your own data, it is a very different picture when it comes to patients interrogating NHS data to make decisions about their treatment - there we seem to be going backwards. This came home to me during a conversation with a man who has been immersed in the world of health data for nearly 20 years. These days Alex Kafetz is among other things a member of the Human Fertilisation and Embryology Authority but 13 years ago he was a witness at the inquiry into the Stafford Hospital scandal.

That was as a result of his work for a private company Dr Foster, whose Good Hospital Guide had highlighted high mortality rates at the hospital. The idea of the guide was ti give patients a better idea if how different hospitals and their doctors compared:

“I’d describe it as an experiment where we put data in the public domain and measured how that affected doctors, particularly, and how that affected patients.”

Alex explained that the company did have access to sensitive data but was very conscious of the need to handle it securely:”We were the first and at the time, only company that had access to the patient level data, some of which was identifiable….in the office we had a locked caged room where the data scientists that had access to that bit of data could go in and no one else.”

Dr Foster was a collaboration between journalists and academics and as the man running the Good Hospital Guide, a key part of his role was being a middleman “so we could get data out there that the journalists felt would hit the mark, but the academics felt would be academically rigorous.”

But they soon found that neither the hospitals nor the doctors were too keen on being assessed in this very public way. “We got sued a lot.. We got injuncted. We got faxes coming in at 6:30pm on a Friday, saying, if you publish this on Sunday, all hell is going to break loose.”

In the case of Stafford Hospital the team had spotted a very high risk adjusted mortality rate “then underneath that you had alerts per specialty and per operation type, and it was just alerting, alerting, alerting…”

They contacted the hospital trust privately but it responded by insisting the Dr Foster data was wrong and it was commissioning Birmingham University to show that was the case.

In the wake of the Stafford scandal the government agreed to publish mortality rates for every NHS hospital. But when Dr Foster was bought by an Australian telecoms company it decided that the Good Hospital Guide did not fit with the rest of the business and it was discontinued.

These days it is difficult to find and compare hospital mortality data - it is“buried in unintelligible spreadsheets,” according to Alex Kafetz. No private company has stepped in to replicate the work done by the Good Hospital Guide. Alex suggests that the government may need to make it easier for companies to access the data, encouraging them to create a market of apps which can help patients become informed consumers of healthcare.

When I googled “hospital mortality rates” I did find a government site but it was forbiddingly complex. I did find a good chart comparing mortality rates across the NHS - but it was a BBC story from 2011 and it labelled its source “Dr Foster”. There have been plenty more medical scandals since Stafford, the latest involving a surgeon at Great Ormond Street Hospital. In all of them the data provided an early warning - so making that data more accessible and transparent can only be a good thing.