8. DOSE-RESPONSE:   THE GOLD STANDARD
                                       

  Eight Rules

1. False Precision

2. Base Rate Fallacy

3. Correlation Not Causation

4. Single Variable

5. Substantial Differences

6. Longitudinal Study

7. Differences From Chance 		  If A causes B:
  More of A produces more of B;
Longer exposure to A produces more of B;
Exposure to A, then stopping, produces less of B than continual exposure;  
The amount of B decreases as the amount of time since exposure to A increases;  
  These relationships are not necessarily linear

 
 
 
IF YOU LEARN JUST TWO THINGS FROM THIS COURSE, THIS SHOULD BE ONE OF THEM
  
 
 
SMOKING EXAMPLE
  •   Smoking for 20 years has higher risk than smoking for 10

  •   Smoking 2 packs/day has higher risk than smoking 1/day

          —   If you stop smoking your risk decreases

          —   The longer the time since you stopped, the lower your risk

  •   If you reduce your smoking, the lower your risk

          —   The more you reduce, the lower your risk

          —   The longer the time since you reduced, the lower your risk
    		•  
      TOP
     
     
    MEDICAL STUDY
    		  
      1. Dose x  
  •   Create 5 groups
    		•   
      2. Dose 2x  
  •   Four get the drug
    		•   
      3. Dose 3x  
  •   One does not
    		•   
      4. Dose 4x  
  •   Follow groups for duration of the study
    		•   
      5. Placebo  
  •   See different responses to different doses
    		•   
     
      Disadvantage:   Everyone in study knows their dose does not change throughout study.   How can we fix this?

    What if we change this, so that everyone knows they switch (from placebo to drug or drug to placebo).   How can we fix this?     		 
      TOP Eight Rules