Authors: Kerrie Evans, Jane Nikles, Michele Sterling & Andrea Hams
Evidence-informed practice, which combines the best available research and clinical expertise with patient values and preferences, forms the basis of patient-centered physiotherapy interventions. Randomised control trials (RCTs) are considered the gold standard in providing the best available research evidence and are frequently used to investigate physiotherapy treatments for musculoskeletal conditions. Most RCTs are parallel group trials which compare average treatment effects across a group, rather than how a specific individual responded to a specific intervention. Such results may not provide the most useful information for a given individual because, for example, some individuals may have a large improvement with treatment A but be worse after treatment B, while the reverse would be true for other individuals, and some may not respond at all. The average response to the two treatments could therefore be similar, but an individual’s response could differ in both magnitude and direction.
The single-case experimental design (SCED) overcomes this issue by using a patient as their own control and systematically measuring their response to each intervention. In the clinical setting, an application of this design would be to gauge the response of an individual patient to repetitions of one or more interventions, for example, an ABAB or ABBA sequence. The clinician can use the results to be more confident in their decision to continue with treatment A, B or A+B or neither. SCEDs complement RCTs by being more flexible, are suitable for the heterogeneity of musculoskeletal presentations, and can help identify individual responses to a chosen intervention.
Our recent systematic review explored the use, purpose, and outcomes of SCEDs in the field of physiotherapy for musculoskeletal conditions. Several key learnings emerged:
Use accepted and validated tools and guidelines when designing SCEDs such as the RoBiNT Scale and the Single-Case Reporting Guideline In BEhavioural Interventions (SCRIBE). Using these frameworks will help ensure a robust study design.
Provide sufficient detail to allow implementation in clinical practice including study setting, treatment adherence, clarity around when treatment finished and follow-up phase began, adverse responses (if any).
Consider a blinded assessor to measure the effect of an intervention. Blinding can be difficult with patient-therapist interactions. In SCEDs, blinding may be less critical than in RCTs given that the patients and clinicians may be more interested in the net or overall benefits of treatment rather than the effects of a treatment in a given population. Nevertheless, where feasible, using a blinded assessor to measure the effect of the intervention will improve internal validity.
Consider options other than visual inspection for analysis. There are great resources for helping determine which statistical analyses should be used for a SCED. In the studies we evaluated, graphical or visual inspection were the most frequent analyses used which alone may not be sufficient to draw strong conclusions.
At the very least, there needs to be a clear second A phase. Whilst AB designs meet the criteria for a single-case methodology, without a clear second A phase there is no added opportunity to demonstrate the experimental effect. Even with an AB with follow-up phase, causal inferences about the effect of the intervention still cannot be drawn. Thus, when seeking to evaluate a physiotherapy intervention for a musculoskeletal condition, at the very least, an ABA design is recommended.
Evaluate patient (and practitioner) expectations and/or satisfaction. Only one study evaluated patients’ satisfaction, and none reported evaluating patient expectations prior to the intervention. Similarly, no study appeared to include patient preferences in their study design. This was somewhat surprising given the importance of shared decision making and matching care to patient expectations in optimising health outcomes. Incorporating patients in the selection of interventions and analysing their satisfaction with care should be carefully considered when designing a SCED.
Individualising care is at the heart of patient-centred care. However, for the findings from SCEDs of physiotherapy for musculoskeletal conditions to be useful, the design and reporting of these studies need to be more rigorous.
Nikles, J, Evans, K, Hams, A & Sterling, M (2022). A systematic review of N-of 1 trials and single case experimental designs inphysiotherapy for musculoskeletal conditions. Musculoskeletal Science and Practice, 102639.
Tate RL, Rosenkoetter U, Wakim D, et al. (2015). The Risk-of-Bias in N-of-1 Trials (RoBiNT) scale: An expanded manual for the critical appraisal of single-case reports. Robyn Tate.
Tate, RL, Perdices, M, Rosenkoetter, U, et al. (2016). The single-case reporting guideline in behavioural interventions (SCRIBE)2016 statement. Physical Therapy, 96, e1-e10.
Tuttle, N & Evans, K (2015). Clinical practice and randomised controlled trials: There’s room for both in individualising patientcare. International Journal of Therapy and Rehabilitation, 22, 257.
Wendt, O & Rindskopf, D (2020). Exploring new directions in statistical analysis of single-case experimental designs. Evidence-Based Communication Assessment and Intervention, 14, 1-5.
About the authors
Dr Kerrie Evans is a Senior Research Fellow at The University of Sydney and Chief Education and Research Officer for Healthia Limited Australia, a large allied health organisation. She is a Specialist Musculoskeletal Physiotherapist and was awarded this title in 2007 by the Australian College of Physiotherapists. Her area of expertise is the assessment and management of people with back and neck pain. Kerrie’s interest in SCEDs came about via her interest in patient-centred, individualised care.
Associate Professor Jane Nikles is a registered medical practitioner and works at the Recover Injury Research Centre at The University of Queensland. Jane has more than 20 years of experience working in the field of N-of-1 trials and single-case experimental designs. She has an interest in applying single-case experimental designs to work towards better therapies after injury, especially when caused by Road Traffic Crashes. Jane is the co-chair of the International Collaborative Network for N-of-1 Trials and Single-Case Designs.
Professor Michele Sterling is Professor in the Recover Injury Research Centre, Program Lead of the Designing Better Therapies research program and Director of the NHMRC Centre of Research Excellence in Better Health Outcomes for Compensable Injury. She is a Musculoskeletal Physiotherapist and a Fellow of the Australian College of Physiotherapists. Michele’s research focuses on mechanisms underlying the development of chronic pain after injury, predictive algorithms for outcomes and developing effective interventions for musculoskeletal injury and pain.
Dr Andrea Hams is a Titled APA Physiotherapist and Lecturer in the School of Health Sciences and Social Work at Griffith University on the Gold Coast, Australia. Andrea’s academic role focuses on Clinical Education of Physiotherapy students. Andrea is an early career researcher (ECR) whose PhD focused on injury prevention in overhead athletes and her research interests include quality of physiotherapy clinical placements and targeted exercise management for both injury prevention and performance.