Background and Rationale
Sleep quality and patterns are influenced by various factors, including metabolic health and age. Metabolic disturbances such as obesity are linked with impaired sleep quality, and are known to increase the risk of developing sleep disorders like obstructive sleep apnea (OSA)1. Moreover, ageing also leads to a greater prevalence of sleep disorders, as well as a generalised decrease in sleep quality2,3. These sleep disorders can further exacerbate the underlying metabolic dysregulation, creating a vicious cycle1. While there is a growing body of literature examining the effects of chronic sleep deprivation on various health metrics, relatively few integrate across body mass index (BMI) and age.
A model of chronic sleep deprivation is acknowledged to manifest in individuals observing Ramadan
4. Several studies have explored the impact of Ramadan fasting on sleep quality and patterns, but most have focused on specific clinical populations, such as individuals with type II diabetes
5. There is a paucity of research examining the potential differences in sleep coping mechanisms across various age and BMI brackets during periods such as Ramadan in the general, provisionally healthy population. Building upon our previous research
6 available
here, we looked to see whether primary sleep metrics vary across Ultrahuman Ring AIR users of different ages and metabolic statuses during the month of Ramadan. The objective of this study was to investigate how different people cope with week-on-week lifestyle disruption during this period.
Methods
The overall scientific design and data collection for this subgroup analysis has been described previously
here6. Briefly, the cohort included Ultrahuman Ring AIR users in the Gulf Cooperative Council (GCC) region who were examined for a particular signature for the period ranging from two weeks before the onset of Ramadan (11th March 2024), through the entire month and two weeks after Eid al-Fitr (10th April 2024). Though we had identified two groups using normal and Ramadan mode configuration algorithms, this whitepaper is a pooled analysis of primary sleep metrics from both cohorts. We stratified the combined cohort (N=113) into three age and BMI buckets each. The age buckets included - 1) below 30 years (<30), 2) between 30 & 50 years (30-50) and 3) above 50 years old (>50). The different BMI buckets were 1) normal (<25 kg/m
2), 2) overweight (25-30 kg/m
2) and 3) obese (30 kg/m
2)
7.
We evaluated the following sleep metrics in each of the groups: sleep duration, average resting heart rate (HR), average sleep heart rate variability (HRV), sleep start & end times as well as movements during sleep. Data analyses and aggregation is covered under terms of use of the Ultrahuman Ring AIR and application. Data analysts had no contact with any users during the entire study period and worked with de-identified data. We analysed the data in-house using Python-MATLAB modules. We ran statistical tests (independent t-tests) for comparison within and between groups.
Result
Sleep changes across different BMI groups:
In the normal BMI group, sleep HRV improved significantly from the pre- to post-Ramadan period. Mean HRV increased by ~8 ms for individuals with normal BMI (Figure 1a), with a trend-level increase in HRV between pre- and Ramadan weeks (p=0.068). In both the overweight and obese groups, we found no appreciable changes in HRV throughout the eight week study period. Sleep duration also increased slightly in the normal and obese groups from the period before fasting to the Ramadan weeks, however, these changes were not statistically significant (Figure 1b). Other metrics like sleep movements, average resting heart rate and sleep start & end times displayed almost no differences during or after Ramadan compared to baseline.
Figure 1: Graphical representation of the combined cohort’s HRV and sleep duration mean values across the 3 phases of the study period (pre-Ramadan, during Ramadan & post-Ramadan). Statistical analysis comparing pre-Ramadan to either during or post-Ramadan using independent t-tests ** denotes p<0.01 N=113, error bars represent standard deviation (SD).
Sleep changes across different age groups:
Sleep duration and HR showed almost no differences during or after Ramadan compared to baseline - across all the three age categories. While sleep HRV decreased slightly from the pre- to Ramadan weeks in the >50 group, this change was not statistically significant. Movements during sleep increased consistently across all age groups from the pre- to post-Ramadan period with the > 50 group displaying a statistically significant increase after Ramadan compared to baseline. (Figure 2a). We also found that during Ramadan weeks, there were significant differences in sleep start and end times across age groups. During Ramadan, the <30 group had an average delayed sleep start time of ~2.7 hours compared to baseline, whereas the 30-50 age group had an average delayed sleep start time of ~2 hours compared to baseline. In contrast, the >50 group went to sleep ~22 minutes earlier on average during Ramadan compared to the pre weeks (Figure 2b). Similarly, the <30 age group had an average delayed sleep end time of ~3 hours compared to baseline, while the 30-50 age group had an average delayed sleep end time of ~2.4 hours compared to baseline. Again, the >50 age group woke up ~13 minutes earlier on average during Ramadan compared to the pre weeks (Figure 2b).
Figure 2: Graphical representation of the combined cohort’s movements and sleep start/end time mean values across the 3 phases of the study period (pre-Ramadan, during Ramadan & post-Ramadan). Statistical analysis comparing pre-Ramadan to either during or post-Ramadan using independent t-tests ** denotes p<0.01 N=113, error bars represent standard deviation (SD).
Conclusion and Discussion
In this analysis of the pooled cohort, we found that age and BMI are associated with distinctive changes in various aspects of sleep during periods of chronic sleep deprivation, such as Ramadan. We observed significant improvements in HRV, a marker of cardiovascular health and recovery, in individuals with a normal BMI - suggesting potential benefits of intermittent fasting for this population. Interestingly, age seemed to have minimal impact on HRV during Ramadan. However, sleep movements, an indicator of restlessness, increased significantly in the older age group (above 50 years) after Ramadan, indicating that sleep may be most disturbed in this population during fasting. Every age bracket experienced more sleep movements as the study progressed, even into the post-Ramadan weeks, highlighting that individuals were still recovering two weeks after Eid al-Fitr. This supports our findings from previous studies in our Ramadan series. We also discovered that BMI did not influence sleep movements, suggesting that age may be a more critical factor in determining sleep disturbances during periods of chronic sleep deprivation. Since this is a pooled analysis, we did not investigate day time naps, given the layered weightage of nap detection in the Ramadan configuration mode (RMC).
The limitations of this study include a relatively small sample size and the analysis of only primary sleep metrics. Additionally, we did not measure other metabolic health proxies, which could provide further insights into the interplay between sleep, age, and BMI during Ramadan. To establish more substantial relationships and understand the underlying mechanisms, future studies are required which incorporate a larger cohort, assess a broader range of sleep and metabolic parameters, and consider additional influencing variables such as physical activity levels and dietary quality.
Through this series of whitepapers, we at Ultrahuman, have attempted to better understand the implications of chronic sleep deprivation, using the framework presented by Ramadan as a model for intermittent fasting and sleep disruption. We plan to build upon these findings and continue exploring the complex relationships between sleep, age, BMI, and metabolic health.
Reach out to support@ultrahuman.com for commercial queries and science@ultrahuman.com for scientific queries.
