N-of-1 Trials in Practice
To demonstrate the methodology in action, I will analyze data collected from my Whoop strap between April 27th, 2018 and October 5th, 2019. The research question for this N-of-1 study is: Does drinking alcohol lead to poor sleep?
As both an athlete and epidemiologist, I am aware of the negative impact alcohol can have on sleep, athletic performance, and overall well-being. Despite being advised against drinking as an athlete, I wanted to see the evidence for myself. After noticing a correlation between alcohol consumption and my sleep score recorded by the Whoop app, I decided to conduct my own analysis.
Data Analysis
The two variables of interest are sleep performance score and alcohol consumption. Sleep performance score ranges from 0 to 100 and is calculated by the Whoop app using biometric data. The alcohol consumption variable is based on the response to the question “Did you have any alcoholic drinks yesterday?” answered by Whoop users daily.
Exploratory Data Analysis
To start, I conducted exploratory data analysis to understand the data distribution. The box plots showed that the average sleep score was higher on nights with no alcohol consumption, with fewer outliers compared to nights with alcohol consumption.
Hypothesis Testing
For hypothesis testing, the null hypothesis was that there is no difference in mean sleep performance between nights with and without alcohol consumption. The alternative hypothesis was that there is a difference in mean sleep performance based on alcohol consumption. A statistical test was conducted to compare the mean sleep scores.
Analysis in R
Using R, the test statistic was calculated, and a null distribution was generated from the sample data. The p-value was found to be 0.017, indicating a significant difference in sleep performance between nights with and without alcohol consumption.
Final Conclusions
The analysis revealed that the average sleep score on nights with no alcohol consumption was significantly higher than on nights with alcohol consumption. This finding supports the idea that alcohol consumption can lead to poor sleep.
Moving Forward
Further analysis and exploration of associations in the data can be done, including using more advanced methods like forecasting and machine learning models. While this study has limitations, such as being observational rather than experimental, it provides valuable insights into the relationship between alcohol consumption and sleep quality.
Source link
