|Posted by Jon Chandler on March 4, 2014 at 2:10 PM|
Most of the time, I hate stats. It’s one of my least favorite subjects. Frankly, I think most socially appropriate human beings would agree. But, it’s a necessary evil in this field. So, I put together a few points of logistical relevance so you don’t have to go ballistic on this stuff!
5 takeaway points for evaluating statistics and drug studies:
1-Even in a double-blind study, reported side effects can tip off the clinician as to whether the subject has received the placebo, or the actual treatment.
2-The placebo effect-is shown when a sugar pill is given to the control group and can lead to positive (and less likely negative) symptoms just simply by receiving something from a clinician. This speaks to the power of the mind.
3-Our mind can work against us, too, with the nocebo effect-setting someone up for possible negative side effects by telling them that “you may get all these side effects, or symptoms: lupus, scleroderma, blurred vision, dry mouth, and left foot paralysis.” It never fails that someone will report left foot paralysis!! As you may know there is not a single drug That’s the power of suggestion!
4-Here’s a great tip when deciding whether to read a study, or not. If your confidence interval is <1.0 it IS statistically significant!! If it includes 1.0, don’t read the study because it is NOT statistically relevant.
5-Risk ratio-is the point estimate used for cohort studies.