Evaluating Therapeutic Change through Single Case Experiments

How do we know whether therapy actually works – not just on average across hundreds of participants, but for this specific person, sitting in this specific therapist’s chair? That question sits at the heart of single case experiments (also known as N=1 studies), a research method designed to evaluate therapeutic change at the individual level. Rather than testing a treatment on groups and reporting statistical averages, single case experiments track one client’s progress with rigorous, systematic measurement – before, during, and after an intervention. The result is clinically grounded evidence that is directly relevant to practice.

What is a single case experiment?
The logic of hypothesis testing in a single case
Phases of a single case experiment
Baseline measurement (Phase A)
Intervention and ongoing assessment (Phase B)
Post-therapy evaluation and withdrawal designs
Common design variations
Notable examples from clinical practice
McCann’s application of EMDR for PTSD
Viens and Hranchuk’s work on eating disorders
Why single case experiments matter for counseling research
Strengths and limitations

What is a single case experiment?

Single case experimental designs are a methodological approach for investigating the effectiveness of a treatment with an individual client. Unlike uncontrolled case studies, these experiments involve the systematic manipulation of an independent variable – the intervention – so that scientifically valid inferences can be drawn about whether the treatment produced the observed change. Each participant serves as their own control, meaning their pre-treatment behavior and post-treatment behavior are compared against each other rather than against a separate control group.

This makes single case experiments especially valuable in counseling and psychotherapy, where recruiting large homogeneous samples is often impractical, client problems occur infrequently, or the aim is to deliver personalized solutions. As Lundervold and Belwood (2000) observed, single case designs represent a scientifically credible way to evaluate practice and conduct clinically relevant research – yet they remain surprisingly underused in counseling training programs.

The logic of hypothesis testing in a single case

Single case experiments are not simply observations of one client – they are structured to test hypotheses. The core question is: Did the intervention cause the change, or would the change have happened anyway? To answer this, the design requires careful planning around phases of measurement, each of which serves a distinct purpose in building the evidentiary case.

Repeated measurements of outcome variables are collected across time and across different levels of the intervention. One phase always serves as the baseline, giving the researcher a stable picture of the client’s functioning before any treatment begins. Without this baseline, it is impossible to know how much change – if any – the therapy actually produced.

Phases of a single case experiment

Baseline measurement (Phase A)

The first phase of any single case experiment involves baseline data collection. During this period, the therapist or researcher measures the target behavior or symptom repeatedly, without introducing the intervention. The goal is to establish a stable, predictable pattern. It is recommended that a stable level or rate of change in performance be observed during baseline before moving into the treatment phase, with a minimum of five data points per phase serving as a general guideline.

Baseline data matter because they allow the researcher to predict what would have happened in the absence of treatment. If a client’s anxiety scores are consistently high across ten pre-treatment sessions, and then drop sharply once therapy begins, that pattern is far more informative than a single pre- and post-test comparison.

Intervention and ongoing assessment (Phase B)

Once baseline stability is established, the intervention is introduced. Data collection does not stop – it continues throughout the treatment phase, generating an ongoing record of how the client responds. This continuous monitoring is one of the defining features of single case methodology, and it serves a dual function: it provides real-time feedback to the therapist while simultaneously building the dataset needed to draw conclusions about treatment effectiveness.

Key features examined when comparing baseline and intervention phases include the immediacy of the effect (how quickly change follows the introduction of the treatment), the degree of overlap between data points across phases, and the consistency of data patterns within each phase. The more rapid and non-overlapping the change, the stronger the evidence that the intervention – rather than some external factor – produced it.

Post-therapy evaluation and withdrawal designs

Some single case experiments extend beyond a simple A-B structure by introducing a withdrawal phase. In an ABA design, the intervention is removed after Phase B to examine whether the client’s behavior returns toward baseline. This adds an important layer of evidence: if symptoms worsen when treatment is withdrawn and improve when it is reintroduced, the case for the intervention’s effectiveness becomes considerably stronger.

The most robust version of this approach is the ABAB design, in which the intervention is withdrawn and then reintroduced a second time. ABAB designs provide an additional demonstration of experimental control through the reimplementation of the intervention, and many clinicians prefer them because the study ends with the client receiving treatment rather than its absence. However, there are ethical concerns with withdrawing an effective treatment, particularly when dealing with behaviors that could cause harm if they return to baseline levels.

Common design variations

Beyond reversal designs, single case methodology includes several other structures suited to different clinical questions. The multiple baseline design staggers the introduction of an intervention across different behaviors, settings, or participants, allowing the researcher to demonstrate that change tracks the intervention rather than the passage of time. This is particularly useful when a target behavior is unlikely to reverse once it has been learned, making withdrawal designs impractical or unethical.

The changing criterion design is well-suited to gradually shaping a behavior. Here, the researcher sets a series of progressively higher or lower performance thresholds and documents whether the client’s behavior shifts in line with each new criterion. Each criterion change is pre-specified, and the data must shift accordingly to provide evidence of a treatment effect.

The flexibility of single case methodology allows it to adapt to diverse research and clinical settings, accommodating complex and individualized treatment plans in ways that large-group randomized trials simply cannot.

Notable examples from clinical practice

McCann’s application of EMDR for PTSD

One of the most widely cited early demonstrations of single case methodology in action comes from McCann’s (1992) use of eye movement desensitization and reprocessing (EMDR) to treat post-traumatic stress disorder. McCann’s case study was among the early reports indicating improvement of PTSD symptoms with EMDR, contributing to a growing body of case-based and subsequently controlled evidence that helped establish EMDR as a credible intervention. Today, EMDR is one of the most studied treatments for PTSD and carries the strongest recommendation in most clinical practice guidelines for this condition.

EMDR is structured as an eight-phase treatment that guides the client through history-taking, preparation, target assessment, desensitization, and several consolidation phases. In a single case experiment applying EMDR, baseline measures of distress, intrusive thoughts, and PTSD symptomatology would be collected repeatedly before the intervention begins. Ongoing assessment would then track how these variables shift session by session – a methodology ideally suited to capturing the gradual and sometimes non-linear process of trauma processing.

Viens and Hranchuk’s work on eating disorders

The application of single case methodology is not confined to trauma. Viens and Hranchuk’s work applied the N=1 experimental framework to the treatment of eating disorders, using systematic baseline and follow-up assessments to track therapeutic change in individual clients. This approach reflects a broader recognition that cognitive-behavioral approaches have the strongest evidence base for conditions such as anorexia nervosa, bulimia nervosa, and binge eating disorder, but that the process of change within an individual client – how early it begins, how it progresses, and what maintains it – is best captured through continuous, individualized measurement rather than group-level statistics.

By collecting data before, during, and after their intervention, Viens and Hranchuk were able to test specific hypotheses about what was driving change in their clients, providing clinically grounded evidence that went beyond simple pre-post comparisons. This kind of single case work often forms the foundation from which larger-scale controlled trials are later developed.

Why single case experiments matter for counseling research

The value of single case experiments extends well beyond the individual cases they document. Case studies enable detailed examination of therapeutic processes, relationship dynamics, and developmental trajectories in ways that aggregate data cannot. They are particularly well-suited to investigating exceptions to general trends and to giving voice to experiences that might be statistically rare but clinically significant.

Perhaps most importantly, single case experiments bridge the gap between research and practice. A counselor in an everyday clinical setting cannot easily run a randomized controlled trial, but they can implement systematic baseline and outcome measurement with their own clients. Single case research design offers counseling practitioners a practical and viable method for evaluating the effectiveness of interventions targeting behavior, emotions, personal characteristics, and other relevant constructs – directly within the therapeutic context where those interventions are being delivered.

There is also a cumulative scientific dimension. When multiple single case experiments across different therapists, clients, and settings consistently show the same pattern of change in response to the same intervention, that convergence constitutes meaningful evidence within the evidence-based treatment literature. Single case experiments are therefore not just clinical tools – they are building blocks of a broader scientific understanding of what works in therapy, and for whom.

Strengths and limitations

Single case experiments offer several clear advantages. They are practical to implement in real clinical settings, they generate individualized data directly relevant to the client being treated, and they can document causal relationships between interventions and outcomes without requiring large sample sizes. Advances in statistical methods in the 21st century have further strengthened the scientific credibility of single case methodology, with techniques for quantifying effect sizes and controlling for trends now well-established in the literature.

That said, single case experiments are not without limitations. Generalizability is a genuine concern – results from one client cannot automatically be assumed to apply to others, even with a similar diagnosis. Internal validity can also be threatened by the client’s life circumstances changing during the study, by practice effects on repeated measures, or by the difficulty of maintaining truly stable baseline conditions. The ethical tension around withdrawing effective treatments to demonstrate experimental control remains an ongoing challenge for reversal designs in clinical settings. Researchers and clinicians must weigh the scientific value of demonstrating control against the welfare of the client receiving – or being temporarily denied – the intervention.

What do you think? Given that single case experiments rely on one person’s data to test whether a therapy works, how much weight should clinicians place on this evidence compared to large randomized trials? And should every practicing therapist be trained to conduct systematic N=1 research with their own clients as a routine part of ethical, evidence-based practice?

References

PDF 📄

Table of Contents

What is a single case experiment? 🔗

The logic of hypothesis testing in a single case 🔗

Phases of a single case experiment 🔗

Baseline measurement (Phase A) 🔗

Intervention and ongoing assessment (Phase B) 🔗

Post-therapy evaluation and withdrawal designs 🔗

Common design variations 🔗

Notable examples from clinical practice 🔗

McCann’s application of EMDR for PTSD 🔗

Viens and Hranchuk’s work on eating disorders 🔗

Why single case experiments matter for counseling research 🔗

Strengths and limitations 🔗