Statistical Power a beginners' guide

1
Statistical Power a beginners' guide

• Professor Ray Iles

2
Definition

• The power of a statistical test is the
  probability that the test will reject a false
  null hypothesis.
• Type II error
• WHY DO THIS CALCULATIONS
• Are my experiments and conclusions valid
• Often funders see no point in even considering
  your grant proposal if power calculations are not
  done.

3
Some mathematical definitions

• As power increases, the chances of a Type II
  error decrease. The probability of a Type II
  error is referred to as the false negative rate
  (ß). Therefore power is equal to 1 - ß.

4
More maths speak

• Statistical power depends on
• the statistical significance criterion used in
  the test
• the size of the difference or the strength of the
  similarity (that is, the effect size) in the
  population
• the sensitivity of the data.

5
Data and population distributions
6
(No Transcript)
7
Study design

• A priori power analysis is conducted prior to the
  research study, and is typically used to
  determine an appropriate sample size to achieve
  adequate power
• Some high impact factor journals , inparticular
  clinical Journals are signed up to the CONSORT
  agreement and insist on power calculation
  disclosure.

8
Consort Flow Diagram
9
Treatment group

• Received electroacupuncture
• 5 local points (knee) and 1 distal point (foot)
• De Qi Sensation
• Biphasic pulse generator
• 3 Hz, maximum tolerable current
• 30 minutes treatment
• Treatment repeated for 10
  consecutive days

10
Statement in journal article

• Power calculations had indicated that in an
  unpaired control study (of 5 Alpha and 95
  Power) assuming a 20 standard deviation around
  means and for a 30 improvement in symptoms
  compared to sham control, that 13 patients needed
  to be recruited in to each arm of the study.
  However if the variation in SD increased to 30
  28 patients would be required in each group.
  Based on these figures and intermediate minimum
  target recruitment was chosen of 2O in each arm
  of the study.

11
Subjective pain scores
12
Magnitude of change in patients plasma cortisol
(upper panel A) and plasma beta endorphin (lower
Panel B) after sham and elecro-acupuncture
treatment
13
Conclusion

• This study strongly suggests that
  electroacupuncture is effective and safe method
  of relieving pain, stiffness and functional
  disability associated with primary osteoarthritis
  knee.
• These findings were validated from
  disease-specific purpose built high performance
  instrument for evaluative research in clinical
  trials on knee osteoarthritis.
• Stress relief in the form of over all clinical
  improvement in the treatment group was reflected
  by a significant fall in the plasma cortisol
  levels. and changes in the neuro-endocrine
  modulator beta-endorphin.
• The analgesia brought about by electro-acupuncture
  was clearly more than sham acupuncture with
  minimal adverse affects.