Comparison of a phased experimental approach and a single randomized clinical trial for developing multicomponent...

Area of Science:

• Health Sciences
• Behavioral Interventions
• Intervention Development

Background:

• Many health interventions are multicomponent, often including behavioral elements.
• Two empirical approaches exist: classical (a priori intervention design and RCT evaluation) and phased experimental (programmatic research for component discovery).

Purpose of the Study:

• To compare the classical and phased experimental approaches for intervention development.
• To evaluate their relative advantages and disadvantages in selecting potent intervention components and levels.

Main Methods:

• A computer simulation was employed.
• Both classical and phased experimental approaches were applied to identical, randomly generated data.

Main Results:

• The phased experimental approach yielded superior mean intervention outcomes for medium and large effect sizes.
• The phased experimental approach identified correct intervention components and levels more frequently across all simulated conditions.

Conclusions:

• The phased experimental approach warrants serious consideration for developing more efficacious behavioral interventions.
• Limitations include not varying factors like the underlying structural model or the number of components in the simulation.