Background

N-of-1 trials are multiple cross-over trials done in individual participants, generating individual treatment effect information. While reporting guidelines for the CONSORT Extension for N-of-1 trials (CENT) and the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) already exist, there is no standardized recommendation for the reporting of N-of-1 trial protocols.

Objective

The objective of this study is to evaluate current literature on N-of-1 design and reporting to identify key elements of rigorous N-of-1 protocol design.

Methods

We will conduct a systematic search for all N-of-1 trial guidelines and protocol-reporting guidelines published in peer-reviewed literature. We will search Medline, Embase, PsycINFO, CINAHL, the Cochrane Methodology Register, CENTRAL, and the NHS Economic Evaluation Database. Eligible articles will contain explicit guidance on N-of-1 protocol construction or reporting. Two reviewers will independently screen all titles and abstracts and then undertake full-text reviews of potential articles to determine eligibility. One reviewer will perform data extraction of selected articles, checked by the second reviewer. Data analysis will ascertain common features of N-of-1 trial protocols and compare them to the SPIRIT and CENT items.

Discussion

This systematic review assesses recommendations on the design and reporting of N-of-1 trial protocols. These findings will inform an international Delphi development process for an N-of-1 trial protocol reporting guideline. The development of this guideline is critical for improving the quality of N-of-1 protocols, leading to improvements in the quality of published N-of-1 trial research.

Objectives

To evaluate the quality of reporting of single-patient (N-of-1) trials published in the medical literature based on the CONSORT Extension for N-of-1 Trials (CENT) statement and to examine factors that influence reporting quality in these trials.

Study design and setting

Through a search of 10 electronic databases, we identified N-of-1 trials in clinical medicine published between January 1, 1985, and December 31, 2013. Two reviewers screened articles for eligibility and independently extracted data. Quality assessment was performed using the CENT statement. Discrepancies were resolved by consensus.

Results

We identified 112 eligible N-of-1 trials published in 87 journals and involving a total of 2,278 patients. Overall, kappa agreement between the two evaluators for compliance with CENT criteria was 0.80 (95% confidence interval: 0.79, 0.82). Trials assessed pharmacology and therapeutics (87%), behavior (11%), or diagnosis (2%). Although 87% of articles described the trial design (including the planned number of subjects and length of treatment period), the median percentage of specific CENT elements reported in the Methods was 41% (range, 16-87%), and the median percentage in the Results was 38% (range, 32-93%). First authors were predominantly from North America (46%), Europe (29%), and Australia (17%). Quality of reporting was higher in articles published in journals with relatively high-impact factors (P = 0.004).

Conclusion

The quality of reporting of published N-of-1 trials is variable and in need of improvement. Because the CENT guidelines were not published until near the end of the period of this review, these results represent a baseline from which improvement may be expected in the future.