When I was a house officer, the department implemented a new policy – All intravenous catheters were to be replaced every 3 days to prevent phlebitis and bloodstream infections. Our group read aghast the memo. That meant a lot of extra work.
That evening, I wrote an e-mail to Angela (I was trying to date her back then), explaining why the new policy was not evidence-based and detailing the study design required to answer this question.
Imagine how I felt when I read Lancet last Saturday. Claire Rickard and colleagues from 3 Australian hospitals randomly assigned 3283 patients who required intravenous catheters for several days to change catheters every 3 days or only when clinically indicated (Lancet 2012;380:1066-74). Phlebitis occurred in 7% of patients in both groups. Nine patients in the routine replacement group and 4 in the clinically indicated group developed bloodstream infection. The authors concluded that both groups were equivalent and the practice of routine catheter replacement should be abandoned.
The study was strikingly simple and could be designed by a house officer like me. Why then, you may ask, was the study not published by me 10 years ago instead?
During a casual discussion on study planning, I mentioned a friend with my Mentor and remarked that he had loads of brilliant ideas but surprisingly few publications. He agreed with my observation and thought for a while. Then he said, “We do not know how he does things. I suspect he lacks the ability to complete a project. You can have a wonderful idea every day, but nothing will happen if you do not do anything.”
P.S. The dating request in the same e-mail was, as you can guess, successful. The intravenous catheter policy, on the other hand, was surprisingly short-lived.