Association between pre-exercise food ingestion timing and reactive hypoglycemia: Insights from a large database of continuous glucose monitoring data

ABSTRACT

Using a large database of continuous glucose monitoring (CGM) data, this study aimed to gain insights into the association between pre-exercise food ingestion timing and reactive hypoglycemia. A group of 6,761 users self-reported 48,799 pre-exercise food ingestion events and logged minute-by-minute CGM, which was used to detect reactive hypoglycemia (<70 mg/dL) in the first 30 min of exercise. A linear and a non-linear binomial logistic regression model was used to investigate the association between food ingestion timing and the probability of experiencing reactive hypoglycemia. An analysis of variance was conducted to compare the predictive ability of the models. On average, reactive hypoglycemia was detected in 8.34 ± 3.04% of the total events, with <15% of individuals experiencing hypoglycemia in >20% of their events. The majority of the reactive hypoglycemia events were found with pre-exercise food timing between ∼30 and ∼90 min, with a peak at ∼60 min. The superior accuracy (62.05 vs 45.1%) and F-score (0.75 vs 0.59) of the non-linear vs the linear model were statistically superior (P < 0.0001). These results support the notion of an unfavourable 30-to-90 min pre-exercise food ingestion time window which can significantly impact the likelihood of reactive hypoglycemia in some individuals.

HIGHLIGHTS

• Large datasets of self-reported continuous glucose monitoring and food events are used here for the first time to get insights into reactive hypoglycemia, a condition often regarded as negative for endurance performance events

• Using a binomial non-linear logistic regression model, the association between pre-exercise food ingestion timing and reactive hypoglycemia revealed the presence of an unfavourable window, when reactive hypoglycemia is more likely to occur.

• Results confirm an individual predisposition to reactive hypoglycemia and, for 8 in 100 individuals, the pre-exercise food ingestion timing can meaningfully impact the likelihood of experiencing reactive hypoglycemia.

KEYWORDS:

• Observational study
• binomial logistic regression
• real-time glucose tracking system

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability

Continuous glucose monitoring data will not be publicly available. The dataset including the user ID, user characteristics, food timing and reactive hypoglycemia events, might be provided after reasonable request.

Additional information

Funding

This work was supported by Supersapiens.