I just got back from the annual meeting of the Society of General Internal Medicine. You should check out all the tweets.

Below is a poster that I presented there on how doctors and patients communicate in the hospital. Though the findings are limited and preliminary, I would welcome your comments!

Number 1 in a series (because all 10 can't fit in one post)

Treating bacteria in the urine in diabetic women without symptoms of urinary infection: DOES NOT HELP

Details:
In contrast to European groups, many American societies recommended using antibiotics to treat such women. This randomized trial found increased antibiotic use associated with such a practice but without any improvement in complications or reduction in the time of symptomatic infection.

Take-home:
Antibiotics don't help in women without urinary infection symptoms, even if their urine test shows bacteria.

Has practice changed?
One question touched only lightly by Prasad et al. in the Mayo Clinic Proceedings piece, below, is whether such reversals in the scientific literature lead to similar reversals in practice. I don't know about current practice regarding asymptomatic bacteriuria, but the gap between bench and bedside (or evidence and practice) is a perennial issue in medicine.

References
1. Medical Reversals come from Prasad et al., Mayo Clinic Proceedings, August 2013 (http://www.mayoclinicproceedings.org/article/PIIS0025619613004059/fulltext)
2. This particular reversal was due to Harding et al., 2002 (http://www.nejm.org/doi/full/10.1056/NEJMoa021042)