Clinical Research: Basic Statistics and Appraising the Literature

1
Clinical ResearchBasic Statistics and
Appraising the Literature
2
Epidemiology and Biostatistics
Epidemiology Study design and interpretation
Biostatistics Methods for analysis
3
Importance of Understanding Basic Statistics in
Medicine

• Research
• Design Studies
• Plan Analyses
• Data Interpretation
• Clinical Medicine
• Understanding the Literature
• Evidence-based practice

4
Learning the Language

• Sampling
• Variable types
• Determine analysis method(s)
• Continuous
• Categorical (nominal, ordinal)
• Independent vs. Correlated Data
• Parametric vs. Non-parametric

5
Sampling Is the study group representative?
CAD caseControl Study n328/group Non-diabetic Mi
ddle-aged Italian Men
Colomba F et al. ATVB 2005 25 1032
6
Sampling Is the study group representative?
Dallas Heart Study Probability-based
sample Over-sampling Minorities
7
Statistical Testing Principles
Question Is blood pressure associated with
stroke?
Study 1
Study 2
Average 136 mm/Hg
Average 136 mm/Hg
Stroke
132 mm/Hg
132 mm/Hg
No Stroke
8
Statistical Testing Principles
Question Is blood pressure associated with
stroke?
Study 1
Study 2
Average 136 mm/Hg
Average 136 mm/Hg
Stroke
132 mm/Hg
132 mm/Hg
No Stroke
9
Statistical Testing
Observed effect (what we see) Expected (under
null)
Test Statistic

Variability of the data
Use test statistic to generate a p-value
10
Learning the Language

• Sampling
• Variable types
• Determine analysis method(s)
• Continuous
• Categorical (nominal, ordinal)
• Independent vs. Correlated Data
• Parametric vs. Non-parametric

11
Categorical Data

• Data where the results are in categories of some
  qualitative trait (yes/no)
• Can be nominal or ordinal

12
Nominal v. Ordinal

• Nominal data (no order to the categories)
• Smoking status (smoker, non-smoker)
• Hair color (blonde, red, black)
• Race (black, white, hispanic, other)

• Ordinal data (order to categories)
• Med school year (1st, 2nd, 3rd, 4th)
• Heart failure class (NYHA 1, 2, 3, or 4)

13
Continuous Data

• Data that are quantitative and measured
• (can perform arithmetic on)
• (can be divided into smaller values)
• Blood pressure
• Age
• Cholesterol levels

14
Variable Types Ordinal, Numerical and Categorical
Svensson AM, et al. Eur Heart J 2005 26 1255
15
Learning the Language

• Sampling
• Variable types
• Determine anlaysis method(s)
• Continuous
• Categorical (nominal, ordinal)
• Independent vs. Correlated Data
• Parametric vs. Non-parametric

16
Data from Independent Samples
3 ?g IP day-1
15 ?g IP day-1
20 ?g IP day-1
40 ?g IP day-1
Diabetic ApoE null mice
Control ApoE null mice
Park L et al. Nat Med 41025
17
Data from Repeated Measures Correlated Data
Control
GIK
2.5
2.5
2
2
1.5
1.5
1
1
0.5
0.5
0
0
Baseline 24 Hours
Baseline 24 Hours
Addo T, et al. Am J Cardiol 2004 94 1288
18
Learning the Language

• Sampling
• Variable types
• Determine anlaysis method(s)
• Continuous
• Categorical (nominal, ordinal)
• Independent vs. Correlated Data
• Parametric vs. Non-parametric

19
Parametric (Gaussian) Distribution
20
Skewed Data
21
Statistical Tests What Type of Data?
22
Power and Sample Size
23
Power What is it

• Power (1-?)
• The probability of rejecting the null hypothesis
  when it is false
• English the probability of detecting a true
  association between an exposure and an outcome
  when there is one

24
Sample Size and Power The assumptions

• Sample size
• To determine sample size, enter three parameters
• Power (80 or 90)
• Effect size
• Control value and variance, or event rate
• dependent on parameter of interest
• best to have pilot data
• Significance level (?) (0.05)
• 1-tailed or 2-tailed testing
• (Confounders)
• Non-compliance, Cross-overs (Drop Ins/Outs), Lost
  to follow up

25
Standards for Effect Size

• Small 20
• Medium 50
• Large 80
• only rough guidelines
• Small, medium and large are subject dependent

26
Adequacy of Sample Size Matters
27
Effect of trial size on results 24 trials of
?-blockade vs. Placebo
28
Ways to Reduce Required Sample Size

• Higher Event Rate
• High risk populations
• Composite Endpoints
• Larger Effect Size
• Lower power
• Larger ?
• 1-tailed or 2
• Change analysis type
• Time dependent

29
Sample size planning

• How much money do you have?
• How much time to you have?
• How many patients/subjects can you expect to
  reasonably get?
• What sample size and study design can I afford?

30
The words to use to describe this

• The study was designed to have gt80 power to
  detect an effect size of gt20 with a 2-tailed
  significance level of 0.05, with a planned sample
  size of 400 participants in each group.

31
Suggested Reading

• Reference texts
• Dawson-Saunders B, Trapp RG. Basic and Clinical
  Biostatistics, Appleton and Lange, Norwalk, CT,
  2nd Edition, 1994.
• Sackett DL. Clinical Epidemiology a basic
  science for clinical medicine. Little Brown,
  Boston, MA, 2nd Edition, 1991.
• Selected papers
• Bias
• Sackett DL. Bias in analytic research. J Chron
  Dis 1979 3251-63
• Power
• Moher D, Dulberg CS, Wells GA. Statistical power,
  sample size, and their reporting in randomized
  controlled trials. JAMA 1994 272 122-4.
• Subgroup analyses
• Assmann SF, Pocock SJ, Enos LE, Kasten LE.
  Subgroup analysis and other (mis)use of baseline
  data in clinical trials. Lancet 2000 355
  1064-1069.
• Yusuf S, Wittes J, Probstfield J, Tyroler HA.
  Analysis and interpretation of treatment effects
  in subgroups of patients in randomized clinical
  trials. JAMA 1991 266 93-98.