5 ms·

If they get such classes, they sleep through them and have for decades. There are studies (Eddy 1982; Gigerenzer and Hoffrage 1995) that show only a small perc

If they get such classes, they sleep through them and have for decades. There are studies (Eddy 1982; Gigerenzer and Hoffrage 1995) that show only a small percent of physicians understand even basic statistics. Just Google same and you'll get a zillion links to freshman college homework discussing the problem.

larrys 13y ago
Perhaps it's not as important as the other things they need to learn. While I don't know the answer I wouldn't assume that having a basic understanding of statistics is better than other things they are memorizing and understanding. (It could be but you can't assume that. Similar to feature creep with products or software.)
- rayiner 13y ago
  A basic understanding of statistics is crucial to being a doctor. When my wife was pregnant, we were faced with decisions like whether to get genetic testing and whether to breastfeed, but were frustrated to deal with statistical incompetence from doctors and nurses who couldn't explain the consequences of courses of action except in child-like "do this" and "don't do this" ways. One nurse, when discussing IUDs, quoted a failure rate by itself as a percentage. When we asked about the time frame, not only did she not know it, but she couldn't understand why a failure percentage was meaningless without reference to a time frame. And forget about being able to reason through the research on alcohol exposure in pregnancy or breastfeeding.
  - Someone 13y ago
    I disagree, especially about the nurses. The medical system is a huge machine. Nurses and doctors are but cogs in that machine. The system produces guidelines such as "after looking at tens of studies, for patients above X years old, procedure X doesn't make sense"; it's way more important for doctors to be able to assess symptoms and to have a feeling for how to tweak those guidelines (yes, this guy is over 60, but he ran a marathon last week without problems, has never smoked, and his parents, both in their nineties, do not need care yet) than to know the ins and outs of statistics. After all, all those statistics are kind of meaningless. They rarely are split by sex, age, smoking behaviour, and familial history ("Where was the study of nonsmoking 36-year-old neurosurgeons?") It's a doctor's job to, somehow, find the best approach using incomplete data. More importantly, all guidelines likely are about approaches that have changed since the data was collected. That new anesthetic procedure may have shifted things by a percent, that slightly adjusted monitoring protocol another, the improved CT scanner another, etc. Medicine just isn't that much of an exact science as many of us think it is. In many cases, we haven't managed to make it a science (yet). For example, in IVF, even after years of working together to copy each other's best practices, one hospital might have a 30% success rate, another a 40% one, even after normalizing for every difference in patient population or treatment we can think of.

larrys 13y ago

Perhaps it's not as important as the other things they need to learn. While I don't know the answer I wouldn't assume that having a basic understanding of statistics is better than other things they are memorizing and understanding. (It could be but you can't assume that. Similar to feature creep with products or software.)

rayiner 13y ago
A basic understanding of statistics is crucial to being a doctor. When my wife was pregnant, we were faced with decisions like whether to get genetic testing and whether to breastfeed, but were frustrated to deal with statistical incompetence from doctors and nurses who couldn't explain the consequences of courses of action except in child-like "do this" and "don't do this" ways. One nurse, when discussing IUDs, quoted a failure rate by itself as a percentage. When we asked about the time frame, not only did she not know it, but she couldn't understand why a failure percentage was meaningless without reference to a time frame. And forget about being able to reason through the research on alcohol exposure in pregnancy or breastfeeding.
- Someone 13y ago
  I disagree, especially about the nurses. The medical system is a huge machine. Nurses and doctors are but cogs in that machine. The system produces guidelines such as "after looking at tens of studies, for patients above X years old, procedure X doesn't make sense"; it's way more important for doctors to be able to assess symptoms and to have a feeling for how to tweak those guidelines (yes, this guy is over 60, but he ran a marathon last week without problems, has never smoked, and his parents, both in their nineties, do not need care yet) than to know the ins and outs of statistics. After all, all those statistics are kind of meaningless. They rarely are split by sex, age, smoking behaviour, and familial history ("Where was the study of nonsmoking 36-year-old neurosurgeons?") It's a doctor's job to, somehow, find the best approach using incomplete data. More importantly, all guidelines likely are about approaches that have changed since the data was collected. That new anesthetic procedure may have shifted things by a percent, that slightly adjusted monitoring protocol another, the improved CT scanner another, etc. Medicine just isn't that much of an exact science as many of us think it is. In many cases, we haven't managed to make it a science (yet). For example, in IVF, even after years of working together to copy each other's best practices, one hospital might have a 30% success rate, another a 40% one, even after normalizing for every difference in patient population or treatment we can think of.