Clinical Trial Basics: Intervention Models in Clinical Trials

What are intervention models in clinical trials?

In clinical trials, intervention model refers to the general structure used for dividing study participants into groups to compare outcomes. These groups are also known as interventional or treatment arms.

What are the different types of intervention models in clinical trials?

Intervention models generally fall under four types: single-group assignments, parallel assignments, cross-over assignments, and factorial assignments.[¹]

The model that is most appropriate for a trial depends on several factors, such as:

The medical condition being tested
The research goals of the trial
The availability of eligible participants

Single group assignment

In single-group assignment, participants are not divided into groups at all. Instead, all participants are assigned to the therapy arm of the trial and receive the same treatment, therapy, or drug, with the same route of administration, dosage, and frequency.

An example of a single-group study is a phase 4 clinical study observing the long-term effects of a newly approved drug in all participants enrolled in the study.

Parallel assignment

Parallel assignment is the most common type of intervention model used in clinical research, wherein trial participants are divided into two or more groups, each receiving a different medical intervention throughout the duration of the study.[²] Participants are given one type of treatment, remaining in the same treatment arm for the entire study, so such studies are also known as non-crossover studies.

An example of a clinical study that uses parallel assignment is a phase III clinical trial comparing the investigational product (drug X) against the standard treatment (drug A) for the condition:

Group 1 (experimental treatment arm) receives drug X
Group 2 (standard treatment control arm) receives drug A

Different dosages of the same drug can also be studied in a parallel group study, for example:

Group 1 receives 50 mg of drug X
Group 2 receives 100 mg of drug X

Cross-over assignment

In a cross-over assignment design, researchers divide trial participants into groups that receive the same experimental treatment(s) but at different times. In other words, participants are switched from one study arm to the other at a given point in time. Sometimes also referred to as a cross-over longitudinal study, this type of intervention model attempts to reduce patient variation for more accurate results.[³] It may also have ethical benefits as all participants are given a chance to benefit from the investigational treatment, which may further encourage patient enrollment and retention.

To reduce any carryover effect from the previous treatments, studies conducted under this model usually include a washout period so the previous treatment can be fully eliminated from the participant's system. Cross-over assignments are generally used when studying chronic conditions, because symptoms are long-term so investigators have enough time to change treatments and study the effects.

A clinical trial employing a cross-over assignment might assign participants to study arms as follows:

Group 1 receives drug X for the first 6 weeks, then drug Y for 6 weeks, with a 6-week washout period in between
Group 2 receives drug Y for the first 6 weeks, then drug X for 6 weeks, with a 6-week washout period in between

Or, another example:

First 2 months: Group 1 receives the experimental intervention while group 2 receives placebo
2 week washout period
Last 2 months: Group 2 receives the experimental intervention while group 1 receives placebo

Factorial assignment

Factorial assignment designs are used when there is more than one intervention to be tested. Trial participants are divided into groups or arms receiving different combinations of two or more interventions/drugs.

The simplest factorial design is a so-called 2x2 factorial assignment, in which two drugs, X and Y, might be tested in four study groups/arms as follows:

Group 1 receives drug X and drug Y together
Group 2 receives drug X and placebo (control)
Group 3 receives drug Y and placebo (control)
Group 4 receives two placebos

The appeal of such a design is that it is essentially similar to conducting two parallel group studies (drug X versus placebo and drug Y versus placebo) on the same study population, allowing comparison of the safety and/or efficacy of drug X versus drug Y. Further, potential interaction (synergy or antagonism) between the two interventions might be elucidated, although this is not the goal of such a study. A general assumption underlying such designs is that the two interventions do not interact with one another, in which case the statistical power of such a study design is greater than that of a multi-arm parallel group trial.[⁴] However, it is often difficult to be sure that there was no interaction, which can lead to difficulty in interpreting results.