Fusing Randomised Trials with Big Data

The CLAHRC WM host organisation, University Hospitals Birmingham NHS Foundation Trust (UHBFT), houses the UK’s (arguably Europe’s) most extensive and sophisticated Patient Electronic Record. It has long been a dream for UHBFT to use the electronic record as a platform to automate RCTs – to determine eligibility, automate randomisation of willing patients, collect outcome data and co-variates, and perhaps incorporate adaptive designs (where the randomisation ratio varies according to emerging results – a kind of “play the winner” algorithm). Under such a scenario a single patient may be cost-effectively offered a set of randomisation opportunities as their care plan unfolds – so called ‘platform trials’. Well, the Americans have caught up, and a well-written article in JAMA argues for just such a process.[1] But barriers are likely on both sides of the Atlantic, as the CLAHRC WM Director encountered when trying to generate a coalition for change – the incentives of people interested in research and those battling to maintain the service are imperfectly aligned. Also, there is a real issue in getting access to source code to specify or implement the necessary sub-routines for automated, clinical trials. And there is a genuine concern that programmes that interact with electronic record systems in real time will slow computer response times. However, the rewards for those who succeed will be enormous. The CLAHRC WM Director wonders how long it will be before ‘point-of-care’ trials become widespread.

— Richard Lilford, CLAHRC WM Director

References:

  1. Angus DC. Fusing Randomized Trials With Big Data. The Key to Self-learning Health Care Systems? JAMA. 2015;314(8): 767-8.
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s