Article
Incorporating electronic health record data in N-of-1 trials
Search Medline for
Authors
-
Stefan Konigorski
- Hasso Plattner Institute for Digital Engineering, Potsdam, GermanyHasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, United States
| Published: | February 26, 2021 |
|---|
Outline
Text
Background: N-of-1 trials are multi-crossover randomized controlled trials in a single participant. They allow investigating treatment effects on the level of a single patient and are an efficient study design for treatments that have a short onset and washout as well as for disease endpoints when there is treatment effect heterogeneity [1]. Using established methods, series of N-of-1 trials can be efficiently aggregated to obtain population-level treatment estimates [1]. For personalized medicine approaches in a clinical context, it is helpful to include information from electronic health records (EHR). This has been proposed for screening and recruiting purposes and as secondary outcomes of N-of-1 trials [1], [2], but has not yet been incorporated in the analysis as covariates or as predictors of the treatment effects.
Methods: Here, we present a study design as well as statistical models for the analysis of a planned series of N-of-1 trials at the Icahn School of Medicine at Mount Sinai in New York. The study investigates the effect of coffee on heart rate variability (HRV) and blood pressure (BP), and explores interindividual differences. HRV and BP are two cardiovascular hallmarks associated with chronic disease and are assessed using sensors. For the statistical analysis, we compare (i) different longitudinal/time series models for assessing the short-term and mid-term effect of coffee on HRV and BP and (ii) different models for predicting these effects of coffee on HRV and BP by the individuals' EHR profiles and other clinical and personal covariates.
Results: The results from an evaluation of these models in Monte Carlo simulation studies show which approaches yield valid and efficient estimation as well as inference on the treatment effects. The application of these models in N-of-1 trials can help to gain novel insights into individual treatment effects for personalizing healthcare as well as into interindividual differences of health interventions.
The authors declare that they have no competing interests.
The authors declare that an ethics committee vote is not required.
References
- 1.
- Kravitz RL, Duan N; DEcIDE Methods Center N-of-1 Guidance Panel, editors. Design and implementation of N-of-1 trials: a user's guide. AHRQ Publication No. 13(14)-EHC122-EF. Rockville, MD: Agency for Healthcare Research and Quality; 2014.
- 2.
- Kaplan HC, et al. Evaluating the comparative effectiveness of two diets in pediatric inflammatory bowel disease: a study protocol for a series of N-of-1 trials. Healthcare. 2019;7(4):129.
