Babel - and the power of measurement
Metrics are the common tongue for science. Could vets now be finding their voice?
One of the world’s underrated forces is the power of metrics and measurement. Agreeing and inventing common ways of measuring things allow tests, proof, progress and revolution - in medicine, science and (often) policy. I am an honorary member of the Worshipful Company of Scientific Instrument Makers, one of the old London guilds full of people who are passionate about the importance, power and potential of measurement. Bringing on young people to master the science and the art. Now and again, the world of metrics sees a breakthrough. One may be happening in the world of veterinary medicine.
When you compare surgical infection rates across different veterinary hospitals, you find wild variation: not by a factor of two or three, but sometimes by an order of magnitude. Are some hospitals dramatically cleaner than others? Perhaps. But there is another explanation, one that reveals something important about how science actually works. The hospitals have been measuring different things.
A study published in the American Journal of Veterinary Research (pdf) has attempted to fix this problem. Researchers asked 32 veterinary specialists from around the world to agree on measurement definitions: what counts as a “superficial” versus “deep” infection, for example. When a “surgery” begins and ends. What temperature constitutes a fever in a horse (38.3°C) versus a goat (39.7°C). The result is a sheet of definitions that run to 14 pages, complete with tables and footnotes.
This is not bureaucratic housekeeping: it may prove to be one of the most consequential interventions in veterinary medicine for years. The reason lies in a phenomenon that has been demonstrated repeatedly in human healthcare: standardised measurement itself improves outcomes.
Ernest Codman - and measurement in hospitals
Ernest Codman is one of the most important medics you’ve never heard of: a Boston surgeon who in the early 1900s developed what he called the “End Result Idea.” His proposal was simple: every hospital should follow every patient long enough to determine whether the treatment had been successful then ask “if not, why not?” with a view to preventing similar failures. He kept meticulous records on index cards, publishing his own complication rates (123 errors among 337 patients) at his own expense.
He proposed something that would save lives, but was hated because it meant serious accountability. His colleagues were horrified. Massachusetts General Hospital refused to implement his system. He turned out to be a kind of metric martyr: ostracised, stripped of his staff privileges, he died in 1940 with little appreciation for his work. Today he is acknowledged as the father of outcomes measurement and the Joint Commission that accredits American hospitals gives an annual award in his name.
In 1980, a landmark 10-year study of nearly 63,000 surgical wounds found large differences in infection rates between individual surgeons. When each surgeon was privately informed of their own rate - and shown how they compared anonymously to their peers - the overall infection rate in clean surgeries fell by half. Nobody mandated new protocols. Nobody forced anyone to change. The information alone was enough.
The pattern repeats across medicine. When New York State began publishing risk-adjusted mortality rates for cardiac surgeons in 1989, death rates after bypass surgery fell from 3.52 per cent to 2.78 per cent within three years. The hospitals with the worst initial records improved the most. When the World Health Organization introduced a simple 19-item surgical safety checklist in 2008 - requiring teams to pause and confirm basic information before anaesthesia, before incision, and before leaving the operating room - complication rates fell by 36 per cent and deaths by more than 40 per cent across hospitals on four continents. A checklist. A shared language. That was all it took.
Psychologists recognise it as a variant of the Hawthorne effect, named after a 1920s factory where workers’ productivity improved whenever they knew they were being observed. Whatever you call it, the mechanism appears robust: when people know they are being measured against a standard, and when they can see how they compare to others, they get better. This can have negative effects: a journalist judged on clicks, for example, would game the system by churning out clickbait. This - Campbell’s Law - has had a seriously corrosive effect on newspapers. So much depends on deploying the right KPI in the right way - a basic life-saved metric has worked wonders in medicine.
Why vets have been flying blind
Until now, veterinary medicine could not benefit from the Hawthorne effect. The problem was not a lack of will but a lack of common language. Researchers attempting to compare infection rates across hospitals - or to pool data for meta-analysis - found themselves stymied. Some borrowed criteria from the US CDC, designed for human patients. Others invented their own. A systematic review in human healthcare found that 41 per cent of studies could not combine their outcome data because underlying research defined success and failure differently. In veterinary medicine, the situation was worse: no universally accepted definitions existed at all.
The consequence was that nobody could say with confidence whether one surgical technique was superior to another, or whether a particular antibiotic protocol actually worked. The evidence base, such as it was, rested on sand. Veterinary surgery has been flying blind.
The consensus, led by Denis Verwilghen of the University of Sydney and Augusta Pelosi of the Veterinary Heart Institute in Florida, is designed to fix that. Initially, the authors warn, adopting standardised definitions may actually cause reported infection rates to rise—not because more animals are getting infected, but because more infections are being properly counted. That will be uncomfortable. Veterinary hospitals do not like to see their numbers go up. But the experience from human medicine suggests that this is precisely how improvement begins. In the human healthcare sector, surveillance programmes have produced infection reductions of 25 to 69 per cent over four to eight years. The hospitals that measure honestly are the hospitals that get better.
A gap in global surveillance
There may be a further benefit beyond animal welfare. Healthcare-associated infections have become a global crisis: of the 421 million patients hospitalised worldwide each year, approximately 43 million acquire at least one infection during their stay. In the European Union alone, more than 3.5 million such infections occur annually, causing over 90,000 deaths. In England, the figure is 834,000 infections and 28,500 deaths. Much of this is driven by antibiotic-resistant bacteria and 71 per cent of all infections caused by antibiotic-resistant bacteria are associated with healthcare settings. The hospitals meant to heal us have become breeding grounds for the superbugs that may one day defeat us.
Global efforts to track and combat this resistance operate under the principle of “One Health”: the recognition that humans and animals share susceptibility to the same organisms. We cohabit. We exchange pathogens. The MRSA that colonises a surgical wound in a dog is the same MRSA that terrifies infection-control specialists in human hospitals. To track resistance properly, we need data from veterinary settings as well as human ones. Until now, veterinary medicine has been unable to contribute because it lacked the consistent definitions that make surveillance possible. This new paper changes that. It won’t solve the superbug crisis, but it closes a gap.
The unglamorous path to progress
There is a deeper lesson here about how improvement happens. We celebrate the lone genius with the breakthrough insight: the new drug, the revolutionary technique. But much of the real work is collaborative, iterative and unglamorous. It involves getting people to agree on what words mean, then counting things carefully, then showing people their own results. Ernest Codman understood this a century ago. He was right, and it cost him his career.
It’s hard to overstate just how true that remains, and for how many fields. Politics is in the pre-Codman stage; policies almost never judged against results and ministers never held accountable.
Veterinary medicine has now taken the same step that human surgery took decades ago. The 18 definitions now agreed by this international panel may not seem like much. They are merely the foundation - the common language without which no evidence can accumulate, no comparisons can be made, no progress can be measured. The Tower of Babel was never finished because its builders could not understand one another. Veterinary surgery, at least, has now agreed on a common tongue.