SQUIRE Blog

Tracking the Evidence

Susan Mooney

Evidence Based Medicine

All of the prenatal care guidelines that I have looked at recommend screening pregnant women for asymptomatic bacteruria.  They all make the claim that pyelonephritis is a primary cause of preterm labor and delivery with all the attendant morbidity and mortality.  They all claim that asymptomatic bacteruria is a leading cause of pyelonephritis and they all draw the same conclusion that screening for and treating asymptomatic bacteruria is beneficial.  The logic seems irrefutable.  So, as I sit here contemplating the background section of my paper, it seems as though it should be simple to look at the primary literature upon which all these guidelines are based.  But a quick literature search has disavowed me of this thought.  I can find lots of papers that argue that a dipstick is as good as a culture and I can find papers showing that, yes, pregnant women in sub saharan Africa also have asymptomatic bacteruria.  So far, I have not found one prospective, randomized, controlled trial that demonstrates the reduction in preterm births associated with screening for and treating asymptomatic bacteruria.  I wonder if such a trial exists and doubt that it does.  And then I think, do you really need that level of evidence?  Further, when you start a QI project, can you rely on guidelines and expert recommendations or do you need to dig into the primary literature?  Ultimately, when do you stop analyzing and start acting?  It is just a urine culture, after all.

Evidence Based Improvement

Things get even muddier when I try to look at the literature to see if others have undertaken a project like mine.  The short answer is no.  Most QI literature reports on larger initiatives that are driven by academic institutions or QIOs.  It is extraordinarily difficult to find projects that describe how someone like me solved a small problem like mine.  If I branch out, to look at something like IHI posterboards, I am more likely to find work that resembles my own.  But much of that work is not peer reviewed and lacks academic rigor.  If I am looking to publish, can I use such evidence?  The reality is that, as I do my improvement work, I "steal" good ideas from where ever I can find them.  I try them out, adapt them, keep them and discard them based on whether or not the help or hinder the process.  Can I be that honest in a publication?

Balancing Again

It seems that I am back to the balancing act: trying to decide how much time to spend looking through the literature to flesh out the background section of this paper while acknowledging that such work will have very little impact on the front line.  The people that provide prenatal care at APD are very unlikely to question the guidelines or to challenge the evidence.  By the same token, they have very limited time to dedicate to a QI project.  I have not yet convinced them all that improving the work is part of the work.  Perhaps a solid literature review would help me in the endeavor.  Regardless of the ultimate benefit, it is clear to me that I need to do more work at the library before I can start to write this section of the paper.