Statistics
                                ==========

        Miscellaneous Notes             <StatNotes.TXT>
        Choosing Tests                  <S-Choose.TXT>
        Interpreting Negative Studies   <S-Negat.TXT>
        Confidence Interval             <Confiden.TXT>

þ Quality of Evidence

A                       B                       C
---------------------------------------------------------------------------
1) Experimental         1) retrospective cohort Other published medical
Randomized clinical     studies*                literature directly related
trials*                                         to topic (e.g., case reports,
2) Observational:       2) cross-sectional      small clinical studies,
Prospective cohort      incl point prevalence+  and published panel consensus)
studies*+
3) Meta-analysis:       3) Case control*
of RCTs only*

*negative experimental or observational studies must have 80% power to detect
a clinically significant difference

+for descriptive studies, the precision of the target estimate must be
sufficiently precise that alternative explanations for the findings are
reasonably excluded.

  [ACEP Clinical policy for the initial approach to adolescents and
  adults presenting to the ED with a chief complaint of headache. Ann
  Emerg Med June 1996;27:821-844.]