Why randomize? \(\checkmark\)
Standard randomized designs \(\checkmark\)
How to analyze data from randomized trials \(\times\)
Deviations from randomization
Adjusting for covariates
Why randomize? \(\checkmark\)
Standard randomized designs \(\checkmark\)
How to analyze data from randomized trials \(\times\)
Deviations from randomization
Adjusting for covariates
Primary analysis for many trials is often direct comparison of outcome in treatment groups, i.e. t-test, log-rank test, permutation test. Secondary analyses may adjust for specific covariates
Question: Patients enroll and are randomized. Some deviate from assignment and move to the other arm. Do you (i) analyze using randomized assignment (say \(T_R\)) or (ii) actual treatment (\(T\))?
Recall that we estimate \(E[O_i(A)|T_i=A]-E[O_i(B)|T_i=B]\) and implicitly equate to \(E[O_i(A)-O_i(B)]\)
Suppose there is no \(T\)-\(O\) association in truth. In intended DAG, \(T_R\equiv T\). Does it matter which of these we use?
\(E[O_i(A)|T_i=A]-E[O_i(B)|T_i=B]\)
\(E[O_i(A)|T_{iR}=A]-E[O_i(B)|T_{iR}=B]\)
\[ \begin{aligned} &E[O_i(A)|T_i=A]-E[O_i(B)|T_i=B] \\ &= E[U_i|T_i~=~A]-E[U_i|T_i=B] \\ &\neq 0 \end{aligned} \]
(because of indirect path through \(U\)) \[ \begin{aligned} &E[O_i(A)|T_{iR}=A]-E[O_i(B)|T_{iR}=B]\\ &= E[U_i|T_{iR}=A]-E[U_i|T_{iR}=B] \\ &= 0 \end{aligned} \]
'As-Treated' analysis may undo the effects of randomization if there is significant deviation from protocol
'Intent-to-Treat' analysis is appropriate one. Formally, want to test for effect of being randomized to treatment rather than effect of treatment itself
Pre-specified subgroup analyses
Negative public perception of large imbalances
Increased precision (?)
Identify one or two strong prognostic factors
Create separate strata defined by these prognostic factors
Within each stratum, conduct separate, independent permuted block randomization
Block size \(b=4\); after 2 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | A | A | ||
2 | ||||
3 |
Section 4.3, Rosenberger and Lachin (2004)
Block size \(b=4\); after 3 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | A | A | ||
B | ||||
2 | ||||
3 |
Block size \(b=4\); after 4 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | B | A | A | |
B | ||||
2 | ||||
3 |
Block size \(b=4\); after 5 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | B | A | A | |
B | B | |||
2 | ||||
3 |
Block size \(b=4\); after 10 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | B | A | A | A |
B | B | B | B | |
A | B | |||
2 | ||||
3 |
Block size \(b=4\); after 13 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | B | A | A | A |
B | B | B | B | |
A | B | A | ||
A | ||||
2 | A | |||
3 |
Block size \(b=4\); after 30 patients
Block | Stratum 1 | Stratum 2 | Stratum 3 | Stratum 4 |
---|---|---|---|---|
1 | B | A | A | A |
B | B | B | B | |
A | A | B | A | |
A | B | A | B | |
2 | A | A | A | B |
B | A | A | A | |
B | B | A | ||
A | B | |||
3 | B |
Subsequent analysis should then adjust for strata, e.g.
Two potential analysis models:
\(O = \alpha^* + \delta^*1_{[T=A]} + \beta V + U\); \(U|\{T,V\} \sim N(0,\sigma^2)\)
\(O = \alpha + \delta 1_{[T=A]} + U\); \(U|T \sim N(0,\tau^2)\)
Which to use?
Question 1: When can both models hold, i.e. both satisfy all assumptions for linear regression?
Question 2: Assuming both models hold, when are treatment effects the same, i.e. \(\delta^* = \delta\)?
Question 3: When both models hold, when is there gain in precision (smaller variance) by adjusting for \(V\)?
From model 1: \(E[O|T] = \alpha^* + \delta^*1_{[T=A]} + \beta E[V|T]\), which matches regression mean according to model 2 when \(E[V|T] = \mu + \gamma 1_{[T=A]}\), for some \(\mu\) and \(\gamma\), so that \(\alpha = \alpha^* + \beta\mu\) and \(\delta = \delta^* + \beta\gamma\)
And because \(\tau^2=\mathrm{var}[O|T] = \beta^2 \mathrm{var}[V|T] + \sigma^2\), matches constant variance according to model 2 when \(\mathrm{var}[V|T] = \omega^2\), so that \(\tau^2 = \beta^2 \omega^2 + \sigma^2\)
Thus, answer to Q1 is when \(V|T \sim N(\mu + \gamma 1_{[T=A]}, \omega^2)\), i.e. when \(V\) given \(T\) is normal with mean equal to linear function of \(T\) and constant variance
In that case, \[ \begin{aligned} E[O|T] &= \alpha^* + \delta^*1_{[T=A]} + \beta (\mu + \gamma 1_{[T=A]})\\ &= (\alpha^* + \beta\mu) + (\delta^* + \beta\gamma) 1_{[T=A]}\\ \end{aligned} \]
So, answer to Q2 is \(\delta^* = \delta\) when \(\beta=0\) or \(\gamma=0\)
For Q3, we compare \(\mathrm{var}(\hat\delta^*)\) to \(\mathrm{var}(\hat\delta)\)
For brevity, we'll use \(T=1\) as equivalent to \(1_{[T=A]}\). Then, \[ \begin{aligned} \mathrm{var}(\hat\delta) = \tau^2\begin{pmatrix} 1 & ET\\ ET & ET^2 \end{pmatrix}^{-1}_{[2,2]} = \dfrac{\beta^2\omega^2 + \sigma^2}{\mathrm{var}[T]} \end{aligned} \]
\[ \begin{aligned} \mathrm{var}(\hat\delta^*) &= \sigma^2\begin{pmatrix} 1 & ET & EU \\ ET & ET^2 & ETV \\ EU & ETV & EV^2 \end{pmatrix}^{-1}_{[2,2]} \\ & = ... \\ & = \dfrac{\mathrm{var}[V] \sigma^2}{\mathrm{var}[V]\mathrm{var}[T] - \mathrm{cov}^2(T,V)} \end{aligned} \]
\[ \begin{aligned} \dfrac{\mathrm{var}(\hat\delta)}{\mathrm{var}(\hat\delta^*)}& = \dfrac{\beta^2\omega^2 + \sigma^2}{\mathrm{var}[T]}\dfrac{\mathrm{var}[V]\mathrm{var}[T] - \mathrm{cov}^2(T,V)}{\mathrm{var}[V] \sigma^2}\\ & = \dfrac{\beta^2\omega^2 + \sigma^2}{\sigma^2}\dfrac{\mathrm{var}[V]\mathrm{var}[T] - \mathrm{cov}^2(T,V)}{\mathrm{var}[V]\mathrm{var}[T]}\\ & = \dfrac{\beta^2\omega^2 + \sigma^2}{\sigma^2}\left(1- \mathrm{cor}^2(T,V)\right)\\ & = \dfrac{1- \mathrm{cor}^2(T,V)}{1 - \dfrac{\beta^2\omega^2}{\beta^2\omega^2 + \sigma^2}} \end{aligned} \]
\[ \begin{aligned} \dfrac{\beta^2\omega^2}{\beta^2\omega^2 + \sigma^2} = \dfrac{\mathrm{var}(\alpha^* + \delta^*T + \beta V|T)}{\mathrm{var}(\alpha^* + \delta^* T + \beta V + U|T)} \end{aligned} \]
This is proportion of variance of \(O\) explained by \(V\) after conditioning on \(T\)
Often write \(\sqrt{\dfrac{\beta^2\omega^2}{\beta^2\omega^2 + \sigma^2}} = \mathrm{cor}(V,O|T)\), the partial correlation between \(V\) and \(O\) given \(T\)
\[ \begin{aligned} \dfrac{\mathrm{var}(\hat\delta)}{\mathrm{var}(\hat\delta^*)} = \dfrac{1- \mathrm{cor}^2(T,V)}{1 - \mathrm{cor}^2(V,O|T)} \end{aligned} \]
If \(\mathrm{cor}(V,O|T)=0\), meaning \(V\) and \(O\) have no linear association once \(T\) is accounted for, then \(\dfrac{\mathrm{var}(\hat\delta)}{\mathrm{var}(\hat\delta^*)} = 1- \mathrm{cor}^2(T,V) \leq 1\)
Adjusting for \(V\) when has no prognostic ability beyond treatment will tend to increase variance of test statistic
If \(\mathrm{cor}(T,V)=0\), meaning \(T\) is unrelated to \(V\) (encouraged by proper randomization) \(\dfrac{\mathrm{var}(\hat\delta)}{\mathrm{var}(\hat\delta^*)} = (1- \mathrm{cor}^2(V,O|T))^{-1} \geq 1\)
Adjusting for prognostic covariate will decrease variance of test statistic
\(\mathrm{cor}(T,V) = 0\) | \(\mathrm{cor}(T,V) \neq 0\) | |
---|---|---|
\(\text{cor}(V,O\|T)=0\) | \(\delta=\delta^*\); \(\mathrm{var}(\hat\delta)=\mathrm{var}(\hat\delta^*)\) | \(\delta=\delta^*\); \(\mathrm{var}(\hat\delta) < \mathrm{var}(\hat\delta^*)\) |
\(\mathrm{cor}(V,O\|T) \neq 0\) | \(\delta=\delta^*\);\(\mathrm{var}(\hat\delta)> \mathrm{var}(\hat\delta^*)\) | \(\delta\neq \delta^*\); \(\mathrm{var}(\hat\delta)(?)\mathrm{var}(\hat\delta^*)\) |
In logistic regression, adding other covariates always increases variance of \(\hat\delta\). However, \(\hat\delta\) may be less biased (Robinson and Jewell, 1991)
Similar but less definitive findings for time-to-event outcomes (Chastang, Byar and Piantadosi, 1988)
Pre-specify analyses: state what primary analysis, secondary analysis will be before trial starts. Be clear where reported analyses deviate from planned analyses
Stratify only on few, strong prognostic covariates. However, if multi-center trial, center should always be stratification factor
Do not add covariates only to improve precision
Statistically acceptable to ignore imbalances occuring by chance; may not be palatable to broader research community
Randomization key to making causal claims: properly conducted, ensures there is no unmeasured confounding, but cannot address external validity concerns, e.g. Simulation 5 preferentially enrolled patients having larger treatment benefit, over-estimated population-average treatment effect. Under-estimate of treatment effect also possible
Restricted randomization used to encourage balance between treatment groups
Chance of large imbalance decreases with sample size; bias due to unmeasured confounding does not
Chastang, C., Byar, D. and Piantadosi, S. (1988) A quantitative study of the bias in estimating the treatment effect caused by omitting a balanced covariate in survival models. Statistics in Medicine, 7, 1243–1255.
Robinson, L.D. and Jewell, N.P. (1991) Some surprising results about covariate adjustment in logistic regression models. International Statistical Review/Revue Internationale de Statistique, 227–240.
Rosenberger, W.F. and Lachin, J.M. (2004) Randomization in clinical trials: Theory and practice. John Wiley & Sons, New York.