On surgeon error and slipped cups

On the very useful and comprehensive Surface Hippy web site, there have been recent posts about slipped cups becoming a more common problem in hip resurfacing, with this problem being strictly surgeon error. Since in my academic research analysis of error is something I think about a lot, I feel the need to add a comment because of the way technical and everyday uses of the word "error" can cause confusion.

Nothing is definite and exact: all measurements and procedures will have error. Good engineering is about using the latest technology to construct something such that the error tolerances are way outside our day-to-day experience. So error is part of life in this world. Used in this way, 'error' is a value-neutral term.

If you think about all that's involved in a process like setting an acetabular cup (drilling, shaping, pounding, testing, bone quality....), some "error" is quite possible and can be expected. Experienced people will have lower error rates, but the complexity of the process and of the human body means that this rate can never be completely zero. To say it again: this rate can never be zero.

If I gave you 100 nails, and I just told you to pound them into a board, I bet you would get most of them into place, but I also bet you'd blow a few of your attempts at this very simple procedure: you'd make a few errors. A skilled person would make fewer mistakes. Anyone would in general make more mistakes if they were tired, rushed, or momentarily distracted, and fewer mistakes if they were really focused and practiced. All these are shades of gray between degrees of error.

My concern is that when people hear "surgeon error" they will tend to make the leap towards thinking "malpractice" or "incompetence" or "neglect" when they should just be hearing the value-neutral word, error. Clearly, if a surgeon has a complication rate way outside the norm, something might be up (assuming that the issue is in the operating room and not in patient selection, and that the numbers are high enough to be statistically valid). You really have to watch out for the latter - low to moderate-n statistics involving random error are really prone to runs and streaks: look at baseball as an example: with enough players randomly batting 250, a few players are always going to have a hitting streak going.

All surgeons go through training, and if hip resurfacing is to spread and help lots of people live better lives, lots of those people will be seeing surgeons who have done relatively fewer procedures. Instead, should they all, every one of them, go visit the handful of superstars who have done many hundreds of operations?

The tough thing for the prospective resurfacing patient is that you can't easily tell if you're surgeon is just riding an unlucky or a lucky streak, or if they are truly inept (or gifted). That is where high numbers do tend to help out. But you have to keep things in perspective, and beware of the human tendency to seek out zero risk - I'm sorry: but you cannot have that, because it does not exist. I think some realism and perspective is important, and at a time when you are already under stress and making lots of decisions, feeling pressed to hook up with a superstar might not be the most helpful thing, all other things considered.

I'm sure others might disagree.