Pfizer closure shows UK's weakening commitment to research
To lose 6,000 skilled jobs in science R&D may be regarded as a misfortune for the coalition; to lose another 2,400 and a world-class research facility looks like carelessness.
This adds up to a huge number of personal catastrophes, but it also hits at one of the UK’s strongest research areas. Biomedicine is an area where the ‘economic impact’ of UK research, the interaction between the public research base and industry, is particularly strong. The loss in capacity and competency will be serious.
What is the significance of biomedicine and pharmaceuticals within the UK research base?
Let’s take a quick look at specifics, using data from Thomson Reuters Web of KnowledgeSM and Evidence’s address database for the UK. Sandwich has a research history that goes back to 1954. It has recently focussed on Discovery High-Throughput Screening, the process that fuels research efforts by providing target compounds to pursue. The bulk of published work from Sandwich is in pharmacology but there is also core chemistry and a spread of articles across many other biomedical disciplines.
Sandwich produces around one-tenth of Pfizer’s annual global output of around 1,400 articles and 500 conference proceedings. Its output in pharmacology is similar in volume to that of the London School of Pharmacy (see diagram below). There is tangible evidence of the ‘economic impact’ of public-sector research because many of Pfizer’s papers are co-authored with high-end UK universities, hospitals, and Research Council and charity-sponsored laboratories. There is also ‘academic impact’ because the average citation impact of those papers is above the UK benchmark.
Taking a broader view, Thomson Reuters data on corporate activity shows the extent of the inter-relationship between the university research base and pharmaceutical companies. Over the last 10 years the UK research base has been involved in more biomedical sector deals, involving such things as drugs and devices, than any other country outside the USA. The UK has a similar number of clinical trials to Germany and Canada and more than Japan. Thomson Reuters Derwent Innovation Index shows that the UK has more pharmaceutical patent collaborations between universities and companies than any nation except the USA and Japan.
Powerful stuff for economic innovation and, equally significantly, the UK corporate sector accounts for around 10 per cent of national biomedical research output. Around half of these papers have an academic co-author and this makes up a high quality contribution to one of the highest quality parts of UK research. The impact of UK biological research makes it the world leader on performance in the G8, as our annual reports to BIS have consistently shown. In biomedicine, the relative citation impact of the UK is around 1.5 times world average in the academic sector and twice world average in the corporate sector. The collaborative papers get higher citation counts than either sector manages to acquire alone.
These companies are major players in the research base. GSK has a diverse output of papers across a wide range of biomedical areas, covering around a third of the journal categories in the Web of Science with about one-sixth of its output in pharmacology. Its total output would make it comparable to one of the larger 1994 Group universities in any national research assessment. AstraZeneca is a little smaller but would stand alongside Bangor or Keele for productivity.
The companies have even more significance when we look at pharmacology in isolation. The bubble diagram shows ten-year output (2000-2009) for GSK, AstraZeneca and Pfizer and for UK universities, many with a reputation for teaching as well as research in this area. Data cover the Thomson Reuters Essential Science Indicators category of pharmacology & toxicology. Now we can see that the private sector is publishing as much as our leading universities and that the academic impact of those publications is absolutely top notch.
These enterprises individually contribute to the substance – the quantity and quality – of the UK research base in one of the country’s most successful areas. They are productive; they are excellent; and they collaborate pervasively with the public sector. Collectively, that may represent a critical part of the platform on which the UK builds its continuing research reputation.
Why have these companies chosen to work in Britain? Pfizer originally opened its research facility because it needed a base inside the UK to enable it to import regulated bulk materials from the USA. But R&D-intensive companies can draw on knowledge from all over the world. GSK collaborates with many leading UK groups, but its UK operation also collaborates with researchers in Australia, California and Japan. The reason why pharmaceutical research operations flourished in the UK has been that the companies could draw on a pool of skilled and talented people produced by the research base growing in UK universities through the 1960s and later.
There is a global war for talent. What keeps companies in one place is good access to the right kind of talent. Talent thrives in an environment of challenge and stimulus, which means the well-found research laboratory. If UK universities are no longer able to provide that environment then the risk must be that Pfizer will only be the first in a line queuing up at Heathrow.
Talent generation, the environment for innovation and their impact on research competiveness, are all weakening. The credibility of the UK’s commitment to research is under question. The importance of being earnest about science and innovation policy is a growing challenge for this Government.
The figure is based on an analysis of research articles and reviews published in journals assigned to the Thomson Reuters Essential Science Indicators category of Pharmacology & toxicology. Publications were collated for the period 2000-2009 and citation counts were made to the end of 2009 for each item and then normalised against the year of publication. A mean normalised citation impact was calculated for each organisation (word average impact will be 1.0). The bubbles are placed in order of descending publication volume and the size of each bubble is scaled by that volume.