How to read a medical study (with notes on Vitamin D studies) – Oct 2019

How to Read a Scientific Study – and why it’s pretty hard – https://examine.com/

Nice PDF
 Download their PDF from VitaminDWiki

  • "Poorly conducted trials can lead to exceptional results. It is usually best to wait and see if those results can be replicated before drawing a conclusion."
  • "One study is just one piece of the puzzle"
    • VitaminDWiki typically provides links to similar studies
  • "Sniffing out the tricks used by the less scrupulous authors is, alas, part of the skills you’ll need to develop to assess published studies"
  • "For scientists, significant doesn’t mean important — it means statistically significant. "
    • ... an effect can be significant yet not clinically relevant (meaning that it has no discernible effect on your health)."
  • "A P-value of 5% (p = 0.05) is no more surprising than getting all heads on 4 coin tosses."

Study Types

Image

Comments by the founder of VitaminDWiki

  1. Meta-analyses can have selection bias - typically picking only about 10% of the potential trials for analysis
  2. Meta-analyses often analyze trial results independent of dose size
    • One Meta-analysis grouped results together for dose sizes ranging from 40 IU to 10,000 IU per day
  3. Meta-analyses often ignore dosing frequency - grouping together daily, weekly, monthly, annual, etc.
  4. Meta-analyses often ignore the type of Vitamin D - D3 is far better than D2
    • Must read many meta-analyses to reduce selection bias and other error
  5. Trials and Meta-analyses may restrict themselves to comparison to a placebo
    • However, some will compare to the intervention to the "standard of care" or a small amount of Vitamin D
  6. Only rarely will a Randomized Controlled Trial or a meta-analysis consider more factors than Vitamin D
    • Frequently Vitamin D is successful only in combination with another item
    • e.g. reduced calories, exercise, Magnesium, Omega-3, Vitamin K, etc.
  7. Sometimes a study will not even measure the Vitamin D before or after an intervention
  8. Trials often error by using a single dose size independent of existing Vitamin D level, obesity,
    • also, the blood response to Vitamin D varies by a range of 4X for the same dose zie
  9. Trials often too short (2 months)
    • Consistent dosing can not result in a good Vitamin D level (and benefits) for 2-6 months
    • Some trials will even document the benefit since the start of the trial (example falls, colds, cancer, etc.)
  10. Trial participants can quickly detect if they are getting "the good stuff"
    • Vitamin D levels needed to get a benefit for one disease often result in the person "feeling better"
  11. Trials often ignore things which will often decease Vitamin D benefits
  12. Trials often "correct" for things which are not independent variables - e.g. obesity
  13. Trials will sometimes "correct" the vitamin D levels based on seasonal statistics
    • But this "correction" is incorrect for perhaps 20% of people - do not vary with season
  14. Some trials are based on Vitamin D measurements made decades before the health event
    • They assume that the vitamin D levels did not change decades later
  15. Trials often ignore Many drugs, such as statins which deplete vitamin D or the Magnesium needed by Vitamin D
  16. Trials will sometimes ignore the season during which a person participated
  17. Studies will often assume some arbitrary Vitamin D threshold (it is unlikely that there is a standard for all health problems
    • Based on national preference: 12 ng, 20ng, 30 ng. but rarely 40-80+ ng as is needed to treat some diseases

Virtually all of the above points could be linked to many actual studies in VitaminDWiki, but there is not enough time to do so

See also VitaminDWiki

12009 visitors, last modified 30 Oct, 2019,
Printer Friendly Follow this page for updates