Researchers and practitioners often use single-case designs (SCDs), or n-of-1 trials, to develop and validate novel treatments. Standards and guidelines have been published to provide guidance as to how to implement SCDs, but many of their recommendations are not derived from the research literature. For example, one of these recommendations suggests that researchers and practitioners should wait for baseline stability prior to introducing an independent variable. However, this recommendation is not strongly supported by empirical evidence. To address this issue, we used Monte Carlo simulations to generate graphs with fixed, response-guided, and random baseline lengths while manipulating trend and variability. Then, our analyses compared the type I error rate and power produced by two methods of analysis: the conservative dual-criteria method (a structured visual aid) and a support vector classifier (a model derived from machine learning). The conservative dual-criteria method produced fewer errors when using response-guided decision-making (i.e., waiting for stability) and random baseline lengths. In contrast, waiting for stability did not reduce decision-making errors with the support vector classifier. Our findings question the necessity of waiting for baseline stability when using SCDs with machine learning, but the study must be replicated with other designs and graph parameters that change over time to support our results.

Single-case designs (SCDs) are used to evaluate the effects of interventions on individual participants. By repeatedly measuring participants under different conditions, SCD studies focus on individual effects rather than on group summaries. The main limitation of SCDs remains its generalisability to wider populations, reducing the relevance of their findings for practice and policy making. With this limitation in mind, methodological developments for synthesising SCD data from different studies that investigate the same research question have intensified in the past decades (e.g. multilevel modelling). However, these techniques are restricted to comparing two interventions at a time and can only incorporate evidence from studies that directly compare the two treatments of interest. These limitations could be addressed by using network meta-analysis that incorporates both direct and indirect evidence to simultaneously compare multiple interventions. Despite its potential, network meta-analytical techniques have yet to be applied to SCD data. Thus, in this paper, we argue that network meta-analysis can be a valuable tool to synthesise SCD data. We demonstrate the use of network meta-analysis in SCD data using a real dataset, and we conclude by reflecting on the challenges that SCD researchers might face when applying network meta-analysis methods to their data.

Behavior analysts commonly use visual inspection to analyze single-case graphs, but studies on its reliability have produced mixed results. To examine this issue, we compared the Type I error rate and power of visual inspection with a novel approach-machine learning. Five expert visual raters analyzed 1,024 simulated AB graphs, which differed on number of points per phase, autocorrelation, trend, variability, and effect size. The ratings were compared to those obtained by the conservative dual-criteria method and two models derived from machine learning. On average, visual raters agreed with each other on only 75% of graphs. In contrast, both models derived from machine learning showed the best balance between Type I error rate and power while producing more consistent results across different graph characteristics. The results suggest that machine learning may support researchers and practitioners in making fewer errors when analyzing single-case graphs, but replications remain necessary.

BACKGROUND: Whiplash associated disorder (WAD), a common and disabling condition, incurs huge burden and costs to Australia. Yet, current treatments for whiplash are not very effective; improved outcomes are urgently needed. Clinical guidelines recommend simple analgesia (paracetamol and non-steroidal anti-inflammatory drugs) but there have been no trials of guideline-recommended drugs. This study will investigate the effectiveness of evidence-based advice (EBA), paracetamol, naproxen, and both paracetamol and naproxen, in reducing daily neck pain and preventing chronic neck pain after whiplash injury.
METHODS: This study is a pilot series of multi-cycle, double-blinded, randomised N-of-1 trials, nested in a multiple baseline design. The design will comprise three baselines of 5, 8 or 11 days duration. Post enrolment, participants will be randomly assigned to one of the baselines. Fifteen participants with acute (<2 weeks) Grade II WAD, experiencing at least moderate pain (NRS: ≥ 5/10), and at risk of poor recovery will be recruited from hospitals in Queensland, Australia, and through local physiotherapists. Patients will receive EBA plus a randomised sequence of three cycles of ten day treatment triplets (paracetamol designated as a C phase, naproxen, designated as a D phase, and both paracetamol and naproxen, designated as an E phase).
CONCLUSIONS: We will test the effects of different treatments on the primary outcome of average neck pain intensity collected daily and at 4 and 7 months post-injury. Secondary outcomes, including disability, depression, post-traumatic stress symptoms, pain catastrophizing, and feasibility of study procedures, will also be evaluated. The results of this study will inform a larger trial aiming to strengthen the evidence on EBA and simple analgesics for WAD.
BACKGROUND: Clinical Trials Primary Registry: Australian and New Zealand Clinical Trials Registry.
BACKGROUND: ACTRN12618001291279.
UNASSIGNED: 31/07/2018.
UNASSIGNED: The University of Queensland, Brisbane QLD 4072 Australia.
BACKGROUND: The University of Queensland.

Published reports describing interventions featuring a single participant are common in neurorehabilitation. Yet, not all such reports use rigorous single-case methodology and there is mounting evidence to suggest that the design, conduct and report of single-case research in the behavioural sciences (including neurorehabilitation) needs improvement. The first part of this article describes resources that will guide the improved design, conduct and critical appraisal of single-case research, including recently published standards in the field of special education and the Risk of Bias in N-of-1 Trials (RoBiNT) Scale for evaluating internal and external validity of designs using a single participant. The second part of the article reports on work currently in progress in developing a reporting guideline in the CONSORT tradition specifically for single-case experimental designs in the behavioural sciences, entitled the Single-Case Reporting guideline In BEhavioural interventions (SCRIBE). It is anticipated that adoption and use of these resources by authors, reviewers and journal editors will improve the reporting and, potentially, the quality of the single-case literature.