Impact is not an old word, unlike its close relative effect. The OED credits Chaucer with the first use of effect, at the end of the 14^th century. Impact is four centuries later, and is a particular type of effect: a physical one. The first uses of impact were in the field of dynamics, in relation to momentum, and it came from the verb impact, which is much older. But after only 30 years as a noun, impact became just another word for effect. And in the last 30 years, both in general English and in medical English, impact has been gradually replacing effect (and impact on is replacing affect). As impact has a special meaning in dentistry and engineering relating to its literal meaning, its metaphorical meaning as just effect (the OED notes “fig.”) may cause a bit of confusion in those areas, but otherwise I suppose that—much though I dislike it—it makes little real difference that impact has usurped effect in “the impact of social network characteristics” or “to reduce the impact of these missing values”.

Where the originally metaphorical impact has no place is in the awful phrase “have a negative impact.” It was almost unknown in English before 1960, but has become steadily more common since 1970. Positive impact has followed the same trajectory but is less common. Medical English mirrors this. In 2015, 0.2% of all PubMed® articles contained negative impact; 0.1% contained positive impact. The phrases are unknown in PubMed before 1973, when it first appeared as “could have a negative impact on fertility.” This means could reduce fertility but is four words longer, none of those words adding anything.

Overall, effect is nearly four times more common than impact in PubMed, but this hides that in 1975 there were four effects to each impact whereas in 2016 there were only two. Oddly, though, this pattern is not followed for “negative effect” and “positive effect”, which, like their impact analogues, were almost unknown before 1970 and have more or less paralleled their rise. For what reason has decrease been replaced by have a negative impact (or effect) and increase by have a positive impact? Is there anyone who thinks these changes are for the better?

I’ve wandered from my metaphorical remit but, as my underlying aim is to make journals easier to read, I make no apology. Surely, make it easier to select is better than “should have a positive impact on the selection of proper antiviral medication for patients”? Would anyone say, “TENS appeared to have a positive effect on cognition” instead of “TENS appeared to improve cognition”? How easy is it to work out that “a strong negative effect on the growth recovery” means growth recovered far less well? Overactive bladder syndrome is a nuisance, but how many patients would agree that it “has a negative impact on quality of life” rather than reducing quality of life? Except that the patients are more likely to say that it makes life less worth living, a phrase that encapsulates the misery of incontinence far better than having a negative impact on the quality of life.

Neville Goodman is a retired consultant anaesthetist and a writer. He is co-author of a book on medical English.

Competing interests: I have read and understood BMJ policy on declaration of interests and declare that my only competing interest is my co-authorship of a book about medical English.