Академический Документы
Профессиональный Документы
Культура Документы
RANDOMIZATION METHODS
542-04-#1
RANDOMIZATION
Why randomize
How to randomize
542-04-#2
Randomization (1)
Rationale
542-04-#3
Randomization (2)
Basic Methods
Ref: Zelen JCD 27:365-375, 1974.
Pocock Biometrics 35:183-197, 1979
542-04-#4
Goal: Achieve Comparable Groups to Allow
Unbiased Estimate of Treatment
542-04-#5
Randomization Basis for
Tests of Hypotheses
542-04-#6
Statistical Properties of Randomization
A. Sampling-Based B. Randomization
Population Model Model
Sample at Sample at
Random Random
Randomization
na patients nb patients
542-04-#7
Randomization Model
Under H0, observed responses independent of treatment
Observed group differences depends ONLY on sequence
of n assignments (random).
542-04-#9
Table of Random Numbers
542-04-#10
Allocation Procedures
to Achieve Balance
Simple randomization
Stratified randomization
Minimization method
542-04-#11
Treatment Imbalance
Statistical Properties of Randomization
542-04-#12
Randomization & Balance (1)
n = 100
p=
s = #heads V(s) =
npq = 100 = 25
E(s) = n p = 50
PS 60 PS - np 60 - 50
S np 60 50
P
25 25
10
P
5
P 2 .025
542-04-#13
Randomization & Balance (2)
n = 20
p=
E(s) = 10
V(s) = np = 20/4 = 5
S np 12 10
PS 12 P
V(s) 5
2
P
5
542-04-#14
Simple Random Allocation
A specified probability, usually equal, of patients assigned
to each treatment arm, remains constant or may change
but not a function of covariates or response
542-04-#15
Simple Randomization
Advantage: simple and easy to implement
542-04-#17
Permuted-Block Randomization (1)
Simple randomization does not guarantee balance in numbers
during trial
If patient characteristics change with time, early imbalances
can't be corrected
Need to avoid runs in Trt assignment
Basic Idea
Divide potential patients into B groups or blocks of size 2m
Randomize each block such that m patients are allocated to A and m to B
Total sample size of 2m B
For each block, there are 2mCm possible realizations
(assuming 2 treatments, A & B)
Maximum imbalance at any time = 2m/2 = m
542-04-#18
Permuted-Block Randomization (2)
Method 1: Example
Block size 2m = 4
2 Trts A,B } 4C2 = 6 possible
A 0.07 1 A
A 0.73 3 B
B 0.87 4 A
B 0.31 2 B
542-04-#20
Permuted-Block Randomization (4)
Concerns
- If blocking is not masked, the sequence become
somewhat predictable (e.g. 2m = 4)
ABAB BAB? Must be A.
AA Must be B B.
- This could lead to selection bias
542-04-#21
Biased Coin Design (BCD)
Efron (1971) Biometrika
Allocation probability to Treatment A changes to keep
balance in each group nearly equal
BCD (p)
Assume two treatments A & B
D = nA -nB "running difference" n = nA + nB
Define p = prob of assigning Trt > 1/2
e.g. PA = prob of assigning Trt A
If D = 0, PA = 1/2
D > 0, PA = 1 - p Excess A's
D < 0, PA = p Excess B's
Efron suggests p=2/3
D > 0 PA = 1/3 D < 0 PA = 2/3
542-04-#22
Urn Randomization
Wei & Lachin: Controlled Clinical Trials, 1988
A generalization of Biased Coin Designs
BCD correction probability (e.g. 2/3) remains constant
regardless of the degree of imbalance
Urn design modifies p as a function of the degree of
imbalance
U(, ) & two Trts (A,B)
0. Urn with white, red balls to start
1. Ball is drawn at random & replaced
2. If red, assign B
If white, assign A
3. Add balls of opposite color
(e.g. If red, add white)
4. Go to 1.
Permutational tests are available, but software not as easy.
542-04-#23
Analysis & Inference
Most analyses do not incorporate blocking
Time Trend
Impact of typical time trends (based on ECOG pts)
on nominal p-values likely to be negligible
Stratified Randomization
Covariate Adaptive
Minimization
Pocock & Simon
542-04-#26
Stratified Randomization (1)
May desire to have treatment groups balanced with respect
to prognostic or risk factors (co-variates)
For large studies, randomization tends to give balance
For smaller studies a better guarantee may be needed
Number of strata
i 1
i
f = # risk factors;
li = number of categories in factor i
Randomize within each stratum
H 1 2 Factors
Sex (M,F) X
M L
2 2 Levels in
and each
Risk (H,L)
F H 4 Strata
L 3
542-04-#28
Stratified Randomization (2)
Define strata
542-04-#31
Minimization Method (1)
An attempt to resolve the problem of empty strata when trying
to balance on many factors with a small number of subjects
Balances Trt assignment simultaneously over many strata
Used when the number of strata is large relative to sample
size as stratified randomization would yield sparse strata
A multiple risk factors need to be incorporated into a score
for degree of imbalance
Need to keep a running total of allocation by strata
Also known as the dynamic allocation
Logistically more complicated
Does not balance within cross-classified stratum cells;
balances over the marginal totals of each stratum, separately
542-04-#32
Example: Minimization Method (a)
Three stratification factors: Sex (2 levels),
age (3 levels), and disease stage (3 levels)
Suppose there are 50 patients enrolled and the
51st patient is male, age 63, and stage III
Trt A Trt B
Sex Male 16 14
Female 10 10
Age < 40 13 12
41-60 9 6
> 60 4 6
Disease Stage I 6 4
Stage II 13 16
Stage III 7 4
Total 26 24 542-04-#33
Example: Minimization Method (b)
542-04-#34
Example: Minimization Method (c)
542-04-#35
Minimization Method (2)
These two criteria will usually agree, but not always
Generalization is possible
542-04-#36
Covariate Adaptive Allocation
(Sequential Balanced Stratification)
Pocock & Simon, Biometrics, 1975;
Efron, Biometrika, 1971
In a simple case, if at some point Trt A has more older patients that
Trt B, next few older patients should more likely be given Trt B
until "balance" is achieved
If assigned Treatment 1 (t = 1)
K X2K X12K
Trt Group 1 4 5
2 6 6
Range = |5-6|, = 1
Assign t
t B(t) with probability
2 9 2/3
1 11 1/3
542-04-#42
Response Adaptive
Allocation Procedures
Use outcome data obtained during trial to influence
allocation of patient to treatment
Data-driven
i.e. dependent on outcome of previous patients
Assumes patient response known before next patient
The goal is to allocate as few patients as possible to
a seemingly inferior treatment
Issues of proper analyses quite complicated
Not widely used though much written about
Very controversial
542-04-#43
Play-the-Winner Rule
Zelen (1969)
Treatment assignment depends on the outcome of
previous patients
Response adaptive assignment
When response is determined quickly
1st subject: toss a coin, H = Trt A, T = Trt B
On subsequent subjects, assign previous treatment if it was
successful
Otherwise, switch treatment assignment for next patient
Advantage: Potentially more patients receive the better
treatment
Disadvantage: Investigator knows the next assignment
542-04-#44
Response Adaptive
Randomization
Example
"Play-the-winner Zelen (1969) JASA
Patient 1 2 3 4 5 6 7 8 9 ......
542-04-#45
Two-armed Bandit or
Randomized Play-the-Winner Rule
Treatment assignment probabilities depend on observed
success probabilities at each time point
542-04-#46
ECMO Example
References
Michigan
1a. Bartlett R., Roloff D., et al.; Pediatrics (1985)
1b. Begg C.; Biometrika (1990)
Harvard
2a. ORourke P., Crone R., et al.; Pediatrics (1989)
2b. Ware J.; Statistical Science (1989)
2c. Royall R.; Statistical Science (1991)
542-04-#47
Michigan ECMO Trial
Bartlett Pediatrics (1985)
Modified play-the-winner
Urn model
A ball ECMO
B ball Standard control
If success on A, add another A ball .
Wei & Durham JASA (1978)
Randomized Consent Design
Results
1 2* 3 4 5 6 7 8 9 10
ECMO S S S S S S S S S
CONTROL F
*sickest patient
P-Values, depending on method, values ranged
.001 6 .05 6 .28
542-04-#48
Harvard ECMO Trial (1)
ORourke, et al.; Pediatrics (1989)
Background
Controversy of Michigan Trial
Harvard experience of standard
11/13 died
Two stage
1st Randomization (permuted block) switch to
superior treatment after 4 deaths in worst arm
542-04-#49
Harvard ECMO Trial (2)
Results
Survival
1st 2nd*
ECMO 9/9 19/20
CONTROL 6/10
* less severe patients
P = .054 (Fisher)
542-04-#50
Multi-institutional Trials
Often in multi-institutional trials, there is a marked
institution effect on outcome measures
542-04-#51
Mechanics of Randomization (1)
* Timing
Example
Alprenolol Trial, Ahlmark et al (1976)
393 patients randomized two weeks before therapy
Only 162 patients treated, 69 alprenolol & 93 placebo
542-04-#52
Mechanics of Randomization (2)
* Operational
1. Sequenced sealed envelopes (prone to tampering!)
2. Sequenced bottles/packets
5. Web based