The Role of Single-Subject Experimental Designs in Evidence-Based Practice Times
This technical brief examines where single-subject experimental designs fit in an era that favors evidence-based practice and treats randomized controlled trials as the gold standard. It compares the two approaches and proposes how they can complement one another in disability and rehabilitation research.
What single-subject designs are
These designs, often called n = 1 designs, establish experimental control within a single unit, which may be a person, classroom, school or community. Most studies actually include several participants. They rely on repeated measurement before and during an intervention, usually shown in graphs and analyzed visually. They differ from medical n-of-1 randomized trials, which use blinding, placebo and random crossovers that rarely suit behavioral treatments.
Place in a five-phase treatment research model
- Phase I: discovering whether an effect exists and estimating its size; single-subject designs suit this stage well.
- Phase II: refining the population, protocol and outcome definitions before a clinical trial; these designs again contribute.
- Phase III: testing efficacy under ideal conditions; between-group trials are called for, though rigorous single-subject protocols may also evaluate efficacy.
- Phase IV: testing whether effects hold in everyday practice; hypothesis-driven single-subject designs fit here.
- Phase V: cost-effectiveness; these designs play no role.
So they contribute to four of the five phases.
Standards for empirically supported treatments
Published quality indicators call for detailed participant and setting descriptions, precisely defined and repeatedly measured outcomes, agreement checks between observers, a documented and actively manipulated intervention, a consistent baseline, and three demonstrations of the experimental effect. For a treatment to be considered empirically supported, the proposal asks for at least five peer-reviewed studies, by three or more investigators in three locations, covering at least twenty participants.
Applying evidence to an individual
Evidence-based practice means informing decisions about an individual client, so practitioners must extrapolate from study participants to that person. Trials draw on samples that may come from subpopulations, report group means rather than individuals, and are conducted under optimal conditions, making transfer to a particular client difficult. Subgroup analyses help but are rare and often underpowered.
Single-subject studies cannot offer statistical generalization from a sample to a population. They can offer logical generality, which builds through repeated replication, and they usually describe each participant, setting and outcome in detail, so practitioners can judge how closely a client resembles those studied. Reporting of participant details is sometimes incomplete and should be checked case by case.
Proposed course of action
The brief recommends continuing well-designed randomized trials for population-level efficacy, with planned subgroup analyses. Rigorous single-subject studies can come before a trial to prepare it, follow a trial to individualize its findings, and be carried out under everyday conditions in Phase IV. Where resources allow, practitioners could also test promising treatments directly with their own clients.
About this summary: this page condenses a longer original publication and does not reproduce it.
This page is a summary. Related material is available from the content areas page.