Experimental (Interventional) evidence carries the most weight
1. Meta-analysis of randomized clinical trials
2. Randomized, placebo controlled, double blind human trials
3. Open label (unblinded)
4. Animal studies
5. Petri dish studies
Observational (no attempt to affect outcome)
1. Mendelian randomization
1. Prospective cohort Study
2. Retrospective cohort study
3. Cross-sectional
4. Case control
Argumentative
1. Using animal or petri dish experiments, observational or testimonial data to infer a hypothesis
Testimonial
1. The experience of one person
The primary objective of health research is to determine what is likely to be true for the average individual. Statistically significant means a 95% chance that you will get the same result (p = .05) Highly statistically significant means a 99% change that you will get the same result (p = .01)
Correlation does not prove causation.
Lack of correlation does not prove lack of causation.
Correlations are considered more likely to be cause and effect if there is a plausible biological mechanism.
