My head is swimming

Sorry, the formatting is kinda wacky on this post, and I can't seem to correct it.

I'm scheduled to give chief rounds this coming Friday. Although I don't procrastinate as much as, say, I did when writing papers in college, I still feel a lot of pressure to get this thing done. Being in college is one thing, but in residency one looses a bit of control over his schedule. (No skipping work to add finishing touches on the presentation!)

Chief rounds are informal lectures given by residents during the noon hour. Typically around 15 other residents attend as well as several faculty members. Recent topics have included pacemakers, acute respiratory distress syndrome, postoperative nausea and vomiting, and anesthesia in the super-obese. For some reason, I had the creative idea of doing something different: biostatistics. Yes, the whole field of biostatistics, in one 45 minute lecture. Stupid.

For one thing I don't really have a strong background in statistics. My math education was cut abruptly and tragically short after barely pulling an "A" in Dr Tidmore's Calculus III in college. (What was I, a liberal arts major, doing in there anyway?) Secondly, the topic is so broad that's it's hard to hone in on the relevant parts. And thirdly, there's great potential to pour hours into preparing a solid lecture with appropriate scope that will bore people to death.

I've overcome the first obstacle by shear force of will and also with the help of an easy-to-read primer in biostatistics lent to me by my faculty advisor. As far as the second difficulty, I plan to use a couple sample articles that present well-designed studies and use them as a launching point to discuss pertinent stastistical analyses. And for the third hurdle: yes, the topic is bone dry, but with proper preparation and a smattering of creativity, humor, and wit, I think I can hold the crowd's attention for thirty minutes.

What isn't helping, however, is paragraphs like this that I'm wading through: "Baseline medical characteristics were compared statistically (Table 1). The effect of the drugs on outcome was assessed with the use of multivariable logistic regression and propensity-score adjustment. Initially, 97 perioperative risk factors were evaluated for univariate association with outcome (two-tailed P less than 0.20) and then entered stepwise (backward and forward) into multivariable logistic models, blah blah blah..."