This weeks paper of the week focuses on ‘Temporal trends in use of tests in UK primary care, 2000-15: retrospective analysis of 250 million tests’.

Listen to the podcast to hear author Jack O’Sullivan talk about this paper with 3V’s Dr Tim Wilson

For privacy reasons SoundCloud needs your permission to be loaded.
I Accept

Reference: O’Sullivan Jack W, Stevens Sarah, Hobbs F D Richard, Salisbury Chris, Little Paul, Goldacre Benet al. Temporal trends in use of tests in UK primary care, 2000-15: retrospective analysis of 250 million tests BMJ 2018; 363 :k4666

3V’s Bottom Line

There has been an 8% per annum increase in testing in UK primary care. Some of this increase will represent better monitoring or attempts to make earlier diagnoses. But some may represent overuse of testing, and certainly the impact of the increase is not just in the cost of testing, and the time taken for doctors to process results, but also the knock-on effects from positive results (both true and false).

Implications for value improvement

We have highlighted previously David Eddy’s seminal paper from JAMA (1) where he outlined the three causes of increase in demand- ageing and demographic change; new interventions and innovation in healthcare; and the “relentless increase in the volume and intensity of medical practice”. The latter is often missed but is perhaps the most important. It contributes to 50% of the rise in demand, and yet seems to go unnoticed. In some instances the increase is good; but equally it may represent overuse and harm to individuals.

At a time when health systems around the world, including the NHS, are requesting more resources, it is important to understand why they are need these additional resources, and question- are the resources being used wisely? Estimates suggest that every year around 2% of low value care is added to health systems budgets. In England this equates to £2.5bn of low value care.

This paper from Oxford shines a window on one element of where the 2% increase in demand creeps in- testing. This paper shows that there has been an 8% increase in testing in UK Primary Care, year on year since 2000. The clustering of testing has grown; so, where 9% of patients had 10 or more tests a year, it is now over 30% of patients who have over ten test per annum. Some of the increase represents better care. But some will be overuse; the continued use of lumbar X-rays being an example.

It is tempting to jump to conclusions when a paper like this is published- but this is a complex issue, with complex causes. Jack O’Sullivan and colleagues have shone a useful spotlight on an area that Finance Directors should be concerned about. It is not the costs of tests, but the 1.5 hours of GP time, the (we estimate) 3 million extra referrals due to false negative results, and subsequent retesting that Finance Directors are having to fund. Jack O’Sullivan has at least made us pause to think, is this a wise investment?


(1) Eddy DM. Three Battles to Watch in the 1990s. JAMA. 1993;270(4):520–526. doi:10.1001/jama.1993.03510040124050

OpenPrescribing – one of a range of projects built by the EBM DataLab to help make complex medical and scientific data more accessible and more impactful in the real world. It provides a search interface onto the raw prescribing data files published by NHS Digital. Every month, the NHS in England publishes anonymised data about the drugs prescribed by GPs. But the raw data files are large and unwieldy, with more than 700 million rows. OpenPrescribing are trying to make it easier for GPs, managers and everyone to explore – supporting safer, more efficient prescribing. Link to