New Methods Permit a Science of Everyday Functioning in Type 1 Diabetes

Day-to-day functioning is a fundamental consideration when evaluating the health and quality of life of people with type 1 diabetes (1). However, despite many decades of research on the immediate and long-term impacts of hypoglycemia and hyperglycemia, we know relatively little about the functional impact of glucose variability and glycemic excursions over hours and days in real-world environments. New ambulatory technologies for monitoring glucose, activity, and cognitive function allow researchers to fill this gap, which may ultimately lead to better interventions and patient outcomes.

We know that adults with type 1 diabetes have (on average) poorer cognitive performance than those without diabetes (2–4). It is also well established that poorer cognitive performance is linked to worse functional outcomes and more difficulties with diabetes self-management (2,5,6). Factors associated with cognitive impairment in type 1 diabetes include age of onset, HbA_1c, severe hypoglycemic events, and diabetes complications (3,7–9). Hypoglycemic and hyperglycemic clamp studies have demonstrated that acute changes in blood glucose can result in reversible cognitive dysfunction (10–15). However, these general findings obscure a tremendous amount of heterogeneity among individuals and in patient outcomes that leave many questions unanswered (16). Previous research on the cognitive and psychological impacts of hypoglycemia, for example, demonstrates that such impacts can vary between individuals (17) and within individuals over time (18). The factors that influence whether a glycemic excursion impacts short-term cognitive status in a given person on a given day remain largely unknown. Are there glycemic, psychological, and diabetes-related factors that make an individual more or less likely to experience short-term cognitive impairments in everyday environments? As research shifts toward precision medicine approaches, there is increasing acknowledgment that group-level data can obscure meaningful individual differences and that diagnosis and treatment can benefit from approaches that are ecologically sensitive (19).

Using continuous glucose monitoring, actigraphy, and mobile technology, the study by Pyatak et al. (1) begins to fill this gap through a comprehensive investigation of the everyday associations between glucose and functioning and the individual-level factors that may moderate this association. Specifically, Pyatak et al. examined associations between overnight glucose metrics (coefficient of variation, time spent <70 mg/dL, and time spent >250 mg/dL) and next-day functioning. Next-day functioning was estimated based on cognitive performance, self-reported activity, and accelerometer-based actigraphy over 14 days in a sample of individuals with type 1 diabetes. Data were collected using mobile cognitive tests, and surveys were administered every 3 h throughout the day. They found that greater glucose variability overnight (based on coefficient of variation) and percent time with glucose >250 mg/dL was associated with poorer next-day functioning. Higher nocturnal glucose variability was specifically associated with poorer sustained attention and lower engagement in demanding activities the following day. Overnight hyperglycemia was associated with more sedentary time the following day. Some of these effects were at least partially explained by the impact of glycemic fluctuations on sleep. For example, higher overnight glucose variability led to more fragmented sleep, which explained part of the association between glucose variability and lower sustained attention the next day. This suggests that interventions that improve sleep can reduce the impact of glucose variability on next-day cognitive performance.

Importantly, Pyatak et al. (1) also found that the association between overnight hypoglycemia and next-day attention was linked to patient-reported outcomes like global illness intrusiveness and diabetes-related quality of life. This may represent a specific mechanistic determinant of quality of life among people with type 1 diabetes that could be addressed by behavioral interventions.

There are many strengths to this study, such as the sample size and the comprehensive approach to measuring next-day function through both subjective (self-reported activities) and objective (cognitive tests and actigraphy) measures. The sample was also racially and ethnically diverse, which is an important characteristic given the large race-related disparities in type 1 diabetes health outcomes (20).

One potential weakness of the study is that the effect sizes reported were quite small. However, accumulated day by day over a lifetime, these effects are likely important for those living with type 1 diabetes (i.e., “death by a thousand paper cuts”), and for interventions that aim to improve daily functioning. Indeed, the science of everyday life is typically a science of small effects. For instance, the impact of sleep and stress on next-day functioning (21,22) and the impact of circadian rhythms on cognitive performance tend to be small (23) but meaningful enough that even healthy individuals notice the impact. For individuals, knowing how, when, and in what way their diabetes impacts their everyday functioning (and subjective experience of their everyday functioning) is a critical part of patient-centered science. The work by Pyatak et al. demonstrates that, as a field, our science and technology tools are finally poised to meet this need.

The findings from this study, while important in their own right, also provide a framework that can be used to answer many other questions about the underlying factors that predict daily functioning in individuals with type 1 diabetes. This knowledge is key to developing interventions that can make meaningful change in group outcomes that also matter to individuals. Ultimately, work like this will lead to tools that let patients self-monitor and identify the factors that impact their own daily functioning using continuous glucose monitoring coupled with other mobile technology prediction tools. This personalized data could be used clinically or by the individual to improve day-to-day functioning (e.g., for academic and work performance). For example, even subtle and temporary cognitive impairments can lead to accidents as well as financial or self-management mistakes, which can have disastrous impacts on well-being, health, and independence. Finally, results like these will help generate hypotheses about the brain mechanisms that underlie everyday functional impairments and resiliency in individuals with type 1 diabetes. It is our hope that studies like this one will lead to rational interventions that can minimize the impact of glycemic excursions on cognitive and functional status, allowing patients to maximize their functioning and quality of life in everyday environments.