Clinical Trial Designs

Randomized control

1. The standard against which other designs are compared

2. All subjects are equally likely to be assigned to the intervention or control

Non-randomized concurrent

1. Controls are subjects treated without the new intervention at approximately the same time as the intervention group is treated

2. E.g., comparison of survival rates of patients treated at two institutions– one using the control treatment and the other using the intervention

Historical controls

1. A new intervention is used in a series of subjects; results are compared to the outcome in a previous series of comparable subjects

2. The design is non-randomized and non-concurrent

Cross-over design

1. This is a special form of a randomized control trial that has appeal to biomedical researchers

2. Allows each patient to serve as their own control

3. In a typical two period design, each patient will receive either intervention or control (A or B) in the first period and the alternative in the succeeding period

   1. The order in which A or B are given to the patient is randomized

   2. Approximately half of the patients receive the intervention in the sequence AB and the other half in BA

Withdrawal studies

1. Patients are taken off the intervention or have the dose reduced; the objective is to assess the response to discontinuation or reduction

2. May be validly used to evaluate the duration of benefit of an intervention which is already known to be useful

3. Useful in assessing the efficacy of an intervention that has never conclusively been shown to be beneficial

Factorial designs

Attempts to evaluate two interventions compared to control in a single experiment

Group allocation

1. Also called composite randomization design

2. A clinic or community is randomized to a particular intervention or control

3. Has been used where there was difficulty in approaching subjects about the idea of randomization

4. Unit of comparison is groups instead of the individual patient

Hybrid designs

1. Combining two types of designs

2. Common one: historical control + randomized design

3. Not efficient and difficult to conduct