Cross-Sectional and Individual Relationships between Physical Activity and Glycemic Variability : Translational Journal of the American College of Sports Medicine

Secondary Logo

Journal Logo

Original Investigation

Cross-Sectional and Individual Relationships between Physical Activity and Glycemic Variability

Sparks, Joshua R.1; Sarzynski, Mark A.2; Davis, J. Mark2; Grandjean, Peter W.3; Wang, Xuewen2

Author Information
Translational Journal of the ACSM: Fall 2022 - Volume 7 - Issue 4 - p 1-12
doi: 10.1249/TJX.0000000000000207

Abstract

INTRODUCTION

The prevalence of adults in the United States classified as overweight or obese continues to increase and is widely considered a major public health crisis of the current generation (1). Adults who are overweight or obese are at an increased risk for the development of impaired glucose metabolism and regulation, as well as type 2 diabetes mellitus (2,3). Impaired glucose metabolism and regulation often results in increased production of reactive oxygen species (4). In turn, oxidative stress and inflammation occur, which are principal risk factors for type 2 diabetes mellitus and cardiovascular disease (5). Interestingly, ~38% of adults in the United States have prediabetes, and ~8.5 million adults are living with undiagnosed type 2 diabetes mellitus (6). One of the primary drivers for these undiagnosed conditions is the use of fasting blood glucose as a global measure of glycemic health without relying on glucose dynamics, such as periods of undulating hypoglycemia and hyperglycemia, captured by glycemic variability (7).

Glycemic variability accounts for glucose excursions, has been utilized to aid in assessment of glycemic health, and potentially acts as a more sensitive analysis of glycemic health compared with clinical assessments, such as a fasted or glucose-challenged state (8). Frequent or exacerbated fluctuations of glucose concentrations are prevalent in overweight or obese adults, which may additively contribute to complications linked to impaired glucose metabolism (9,10). The ability to monitor those at risk for the development of type 2 diabetes (i.e., adults with overweight or obesity) and evaluate their potential risk based on recurrent oscillations in their glucose concentrations presents an opportunity to integrate a novel and therapeutic tool before disease presents itself (11–13). As technology continues to advance, glycemic variability assessment has become a minimally invasive and cost-effective procedure utilizing continuous glucose monitoring (CGM), which uses a device worn during everyday life. The real-time analysis of glucose dynamics and glycemic variability may prompt behavior change, such as physical activity (PA), through immediate feedback (14,15). Practitioners or clinicians may also utilize CGM devices to provide adaptive feedback to improve glucose metabolism and regulation (16). Enhanced utilization of CGM, specifically in adults with overweight or obesity, may aid in the prevention of future development of type 2 diabetes mellitus.

PA can be either structured (e.g., exercise) or incidental, such as daily activities at work, home, or during transport (17). As such, PA is a multifaceted health behavior that may be free-living or planned, which benefits from 24-h objective monitoring to understand patterns, including PA time and intensity (18). However, overweight or obese adults tend to spend more time sedentary and less time performing PA of light-, moderate-, and vigorous intensity compared with adults with normal weight (19,20). Evidence indicates that decreasing sedentary time and increasing the amount of PA of any intensity, including light intensity and steps per day, benefits cardiometabolic health (e.g., decreases body mass index (BMI), dyslipidemia, and fasting glucose concentration) (21,22). Objectively measured time spent sedentary and performing light-intensity PA (LPA) and moderate-to-vigorous-intensity PA (MVPA) are independently associated with 2-h glucose concentration during oral glucose tolerance tests in overweight adults without diagnosed diabetes (23). However, limited evidence exists to suggest that PA variability throughout the week (specifically MVPA), as opposed to PA consistency, may influence cardiometabolic health outcomes, with most previous studies focused on children and youth (24,25).

To date, few studies have evaluated the relationship between objectively measured sedentary and PA time with CGM-assessed glucose concentrations and glycemic variability. Therefore, the purpose of this study was to examine the associations between sedentary time and PA with glucose concentrations and glycemic variability in overweight or obese adults. The overarching hypothesis was that less time being sedentary and more time participating in PA would be associated with lower glycemic variability.

METHODS

This was a cross-sectional study using baseline data from two independent clinical trials, which recruited between January 2015 and November 2016 (NCT02413866) (26,27) and November 2017 and April 2019 (NCT03162991) (28). There were no repeat participants between the two studies. The present study includes 28 participants with valid CGM and accelerometry data before either study’s intervention. All participant visits and testing were completed by the same trained research staff in our research center at the University of South Carolina.

Participants

Both studies had similar inclusion/exclusion criteria. Briefly, inclusion criteria were reporting <120 min of resistance or endurance exercise (structured PA) per week during the previous 3 months, overweight or obese (25 ≤ BMI ≤ 40 kg·m−2), age 35–55 yr, weight stable (±2%) during the previous 3 months, and, for females, eumenorrheic. Exclusion criteria included any self-reported medical conditions such as diabetes, cardiovascular diseases, chronic or recurrent respiratory conditions (e.g., uncontrolled asthma or chronic obstructive pulmonary disease), active cancer, and eating or neurological disorders; medications that affect metabolism (e.g., thyroid medications, statins); psychological issues including but not limited to untreated depression and attention deficit disorder; excessive caffeine use (>500 mg·d−1); smoking during the past year; pregnant or lactating females; and/or unwillingness to provide informed consent. Study protocols were approved by the University of South Carolina Institutional Review Board (Pro00036519 and Pro00067271), and all participants signed an informed consent form.

Measurements

Height, Body Weight, and BMI

Height and body weight were measured at the first visit using a stadiometer and an electronic scale that was calibrated annually (CC Vaughan & Sons Inc., Columbia, SC). BMI was calculated using the following equation: BMI (kg·m−2) = body weight (kg) ÷ [height (m)]2.

Sedentary Time and PA Measures

At the first visit, participants were trained on the use of the Sensewear Mini Armband device (BodyMedia®, Pittsburgh, PA), which was initially placed by trained research staff on the posterior aspect of the left arm at the midpoint between the olecranon and acromion processes. The Sensewear device objectively measures sedentary time and PA and has been demonstrated to be reliable and valid (29,30). The Sensewear device has sensors that measure skin temperature, galvanic skin response, heat flux from the body, and movement, which are physiologic data processed to calculate and report total and intensity-specific PA energy expenditure (EE), as well as sleep duration in a free-living environment (31). Participants wore the device for 7 consecutive days and were instructed to maintain their normal daily routine. Participants were instructed to only remove the device for water-based activities, such as showering/bathing or swimming, but to reposition the device immediately and wear it continuously thereafter, including during sleep. On the final day of the 7-d monitoring period, participants reported back for their second visit and returned the Sensewear device. Data were considered valid for analysis if participants wore the monitor for ≥5 d including one weekend day, with a minimum wear time of 20 h each day.

Sedentary time and PA measures were assessed utilizing manufacturer provided software (BodyMedia® Sensewear version 7.0). Sedentary time was established as <1.5 metabolic equivalents (METs; excluding sleep time), and time spent performing and intensity-specific EE for LPA as 1.5 to <3.0 METs, for MVPA as ≥3.0 METs, and for total PA intensity as ≥1.5 METs were instituted. Because MET values were defined for assessment of intensity-specific PA time and EE, MET-minutes per day was also calculated for LPA, MVPA, and total PA. Sedentary time and time spent performing PA of varying intensity, as well as associated EE and MET-minutes per day, were obtained for each valid day of wear time, and the average of those days were calculated. In addition, the standard deviation (SD) of sedentary and PA measures across all valid days was calculated as a measure of day-to-day variability. Estimated PA information during nonwear times was excluded from analysis.

Continuous Glucose Monitoring

A CGM device (Dexcom G4 Platinum Professional, San Diego, CA) was used to assess interstitial glucose concentrations throughout the same 7 consecutive days as Sensewear Mini Armband monitoring. At the first visit, participants reported to our research center for placement and instruction for use of the CGM device by trained research staff. Participants had a catheter inserted under the skin on the preferred side of the abdomen with an attached sensor and transmitter, approximately 2 cm to the side of the umbilicus. They were instructed to carry a recording device, which received and stored interstitial glucose concentration readings every 5 min over the 7 consecutive days. The CGM device was blinded so that participants could not observe the live readings, to deter any alterations in diet, PA, or general lifestyle, and participants were requested to maintain their normal daily routine during the monitoring period. On the final day of the 7-d monitoring period, participants reported back for their second visit and the CGM device was removed.

This specific Dexcom model’s transmitter and sensor have been approved for up to 7 consecutive days of wear. However, there is a possibility that the transmitter and its sensor can become dislodged or fall off during wear and therefore decrease the number of days worn. To combat this, we ensured proper adhesion before the participant left rather than leaving our research center, as well as placed a piece of Tegaderm™ transparent film (3M Company, Saint Paul, MN) over the transmitter and sensor. Lastly, if the receiver’s battery runs out and is not recharged, or if the transmitter and sensor are further than 50 ft away from the receiver, there is a possibility of missing data. To limit such events, we ensured that the receiver was fully charged before the participant left rather than leaving our research center, provided a charger and instructed participants to charge the device overnight while asleep in their room, and supplied a carrying case to limit the distance between the receiver and the transmitter. Data were considered valid for analysis if participants wore the device for ≥5 d including one weekend day, with a minimum available glucose measure over 20 h each day. Manufacturer-provided software (Dexcom Studio 12.0.4.6) was used to download and export CGM data to Excel (Microsoft, Redmond, WA) datafiles.

Glycemia states were established utilizing two different criteria: adults with diagnosed type 2 diabetes mellitus (32) and healthy nondiabetic adults (33). Although none of our participants reported being diagnosed with type 2 diabetes mellitus, there is a large proportion of the US adult population that may have undiagnosed prediabetes or type 2 diabetes mellitus (6). The criteria for type 2 diabetic adults were established as percent of time spent in range (TIR; 70–180 mg·dL−1) and hyperglycemia (>180 mg·dL−1) for each 24-h period and averaged (32), whereas the criteria for healthy nondiabetic adults were established as percent of TIR (70–140 mg·dL−1) and hyperglycemia (>140 mg·dL−1) for each 24-h period and averaged (33). Mean 24-h glucose concentration was calculated for each 24-h period and averaged. Percent of TIR and hyperglycemia, and 24-h mean glucose concentration were assessed from midnight to midnight for each valid day. Because PA is a waking behavior, we also assessed waking mean glucose concentration and percentage of time spent in varying glycemia states through self-reported time in and out of bed.

The 1-, 2-, and 4-h continuous overlapping net glycemic actions (CONGA-1, CONGA-2, and CONGA-4, respectively) were calculated manually in Excel, whereas the mean amplitude of glycemic excursion (MAGE) and mean of daily difference (MODD) measures of glycemic variability were calculated in EasyGV (version 9.0.R2; University of Oxford, Oxford, UK), which is an Excel-enabled workbook that utilizes macros. MAGE was calculated for each participant by taking the arithmetic mean of increased or decreased glucose concentrations (nadirs and peaks or vice versa) when both ascending and descending concentration exceeds 1 SD from the 24-h mean glucose concentration for the same 24-h monitoring period (34). CONGA-1, CONGA-2, and CONGA-4 were calculated as the SD of the differences between each observation and the previous 1-, 2-, and 4-h observations (35). CONGA-1, CONGA-2, and CONGA-4 were chosen for this study because the 1-, 2-, and 4-h time periods approximate the time intervals between different activities (CONGA-1), between snacks (CONGA-2), and between meals (CONGA-4) (35). MODD accounts for the mean of absolute differences between glucose concentrations obtained at the same time of day on consecutive days and has been highly correlated with the SD for between-day variability (36). MAGE and CONGA-1, CONGA-2, and CONGA-4 were utilized as measurements of intraday glycemic variability for each valid day of wear time and the averages of those days calculated, whereas MODD was utilized as a measurement of interday glycemic variability for all valid days combined.

Statistical Analysis

Statistical analyses were performed using SAS (version 9.4; SAS Institute, Cary, NC), with P < 0.05 considered significant. Participant characteristics were reported as mean ± SD. Descriptive statistics were calculated for accelerometer measures, including minutes per day sedentary and engaging in PA as well as their associated MET-minutes per day and EE, and CGM measures, including 24-h and waking glycemia states, mean glucose concentration, and glycemic variability. Linear regression analyses were performed to examine the associations between accelerometer and CGM measures (unadjusted; model 1). Adjustments for age (model 2) or BMI (model 3) independently and combined (model 4) were performed. Further adjustments for race/ethnicity (model 5) or sex (model 6) independently and combined (model 7) were also performed. Lastly, an adjusted model was performed for all covariates combined, including age, BMI, race/ethnicity, and sex (model 8). A single CGM measure was placed as the dependent variable (DV), and a single accelerometer measure was placed as the independent variable (IV) for unadjusted and adjusted models. Because interindividual and intraindividual variability was anticipated in our sample for accelerometer and CGM measures, we further performed regression analyses to examine the relationships between accelerometer and CGM measures across days within each individual.

RESULTS

Participants were, on average (mean ± SD), 46.0 ± 6.1 yr of age, 1.7 ± 0.1 m tall, and weighed 92.6 ± 21.4 kg; approximately 70% were female, 57% African American, and 54% classified as obese (BMI ≥30 kg·m−2; Supplemental Content 1, Table, https://links.lww.com/TJACSM/A185). Overall, participants spent 12 h·d−1 sedentary and a total of 5.2 h·d−1 performing PA, with ~1 h being MVPA (Table 1). When examined as a global measure of PA time, accounting for all activities within a 24-h time frame, 23 (82%) participants met the minimum guidelines for PA engagement, established as 150 min of MVPA per week (34). In addition, over 83% of the day measured TIR for glucose concentrations between 70 and 140 mg·dL−1 and over 90% of the day for 70–180 mg·dL−1, which are considered normal reference ranges for nondiabetic adults and type 2 diabetic adults, respectively (Table 2).

TABLE 1 - Sensewear Mini Armband Accelerometer Measurements.
Wear time
 No. days 5.9 ± 0.7
 h·d−1 23.4 ± 0.5
Accelerometer measurements
 Sedentary time, min·d−1 717.2 ± 124.0
 LPA time, min·d−1 247.0 ± 94.7
 MVPA time, min·d−1 65.5 ± 42.2
 Total PA time, min·d−1 312.5 ± 121.8
 PA EE, kcal·d−1
  LPA EE 1052.3 ± 412.9
  MVPA EE 383.6 ± 293.8
  Total PA EE 1435.9 ± 682.8
 PA, MET-min·d−1
  LPA MET-minutes 559.8 ± 240.4
  MVPA MET-minutes 255.5 ± 162.2
  Total PA MET-minutes 960.2 ± 370.9
Day-to-day variability across accelerometer wear days
 SD, min·d−1
  Sedentary time 113.8 ± 65.8
  LPA time 61.4 ± 29.1
  MVPA time 32.5 ± 24.3
  Total PA time 74.5 ± 36.6
 PA EE SD, kcal·d−1
  LPA EE 298.1 ± 179.5
  MVPA EE 854.0 ± 750.9
  Total PA EE 502.1 ± 337.6
 PA SD, MET-min·d−1
  LPA MET-minutes 12.8 ± 8.7
  MVPA MET-minutes 14.8 ± 16.7
  Total PA MET-minutes 24.3 ± 18.0
Data presented as mean ± SD.

TABLE 2 - CGM Measurements.
CGM observations
 No. days 6.0 ± 0.5
 Percentage of maximum observations 96.0 ± 0.1
24-h measurements
 Glycemia states, % time of day
  TIR (70–140 mg·dL−1) 83.0 ± 11.7
  TIR (70–180 mg·dL−1) 91.1 ± 9.8
  Hyperglycemia (>140 mg·dL−1) 10.2 ± 4.6
  Hyperglycemia (>180 mg·dL−1) 2.1 ± 5.3
 Glucose concentrations, mg·dL−1
  24-h mean 101.2 ± 16.7
 Glycemic variability, mg·dL−1
  MAGE 43.0 ± 12.1
  CONGA-1 19.3 ± 5.0
  CONGA-2 23.3 ± 6.3
  CONGA-4 25.5 ± 6.6
  MODD 19.9 ± 4.5
Waking measurements
 Waking hours per day 16.3 ± 1.0
 Glycemia states, % time of waking
  TIR (70–140 mg·dL−1) 76.0 ± 9.8
  TIR (70–180 mg·dL−1) 83.7 ± 8.8
  Hyperglycemia (>140 mg·dL−1) 24.7 ± 16.9
  Hyperglycemia (>180 mg·dL−1) 15.1 ± 7.6
 Glucose concentrations, mg·dL−1
  Waking, mean 101.8 ± 16.0
 Glycemic variability, mg·dL−1
  MAGE 44.6 ± 11.9
  CONGA-1 21.0 ± 5.4
  CONGA-2 25.2 ± 6.6
  CONGA-4 26.6 ± 6.5
  MODD 22.0 ± 6.5
Data presented as mean ± SD or proportion. Number of observations per day (n; % of max) = (number of CGM observations per day/maximum observations per day) × 100.

Relationship between Accelerometer and 24-h CGM Measures

Significant positive associations were found between LPA MET-minutes per day and percentage of TIR based on nondiabetic (model 1; β = 0.00, P = 0.02) and type 2 diabetic (model 1; β = 0.00, P = 0.01) criteria (Table 3). In addition, significant negative associations were observed between LPA MET-minutes per day and percentage of time in hyperglycemia based on nondiabetic (model 1; β = −0.00, P = 0.03) and type 2 diabetic (model 1; β = −0.00, P = 0.03) criteria. Furthermore, a negative association was observed between total PA time and 24-h mean glucose concentration (model 1; β = −0.06 P = 0.04), such that every 30-minute increase in total PA related to a 1.8-mg·dL−1 decrease in 24-h mean glucose concentration. After adjustment for age and BMI combined (model 4), race/ethnicity and sex combined (model 7), and all covariates combined (model 8), this relationship was lost (P > 0.05 for all).

TABLE 3 - Linear Regression Analysis between Average Daily Sedentary Time and PA Measures with 24-h and Waking Glycemia Measurements, Glucose Concentrations, and Measures of Glycemic Variability.
min·d−1 PA EE, kcal·d−1 MET-min·d−1
Sedentary LPA MVPA Total PA LPA MVPA Total PA LPA MVPA Total PA
24-h glycemia, % of day
 TIR (70–140 mg·dL−1) 0.00 (0.00) 0.85 −0.00 (0.00) 0.20 −0.00 (0.00) 0.19 −0.00 (0.00) 0.16 −0.00 (0.00) 0.27 −0.00 (0.00) 0.25 −0.00 (0.00) 0.24 0.00 (0.00) 0.02 *,**,*** −0.00 (0.00) 0.39 −0.00 (0.00) 0.18
 TIR (70–180 mg·dL−1) 0.00 (0.00) 0.62 −0.00 (0.00) 0.16 −0.00 (0.00) 0.22 −0.00 (0.00) 0.14 −0.00 (0.00) 0.29 −0.00 (0.00) 0.31 −0.00 (0.00) 0.28 0.00 (0.00) 0.01 *,**,*** −0.00 (0.00) 0.43 −0.00 (0.00) 0.15
 Hyper (>140 mg·dL−1) −0.00 (0.00) 0.96 0.00 (0.00) 0.21 0.00 (0.00) 0.54 0.00 (0.00) 0.24 0.00 (0.00) 0.15 0.00 (0.00) 0.82 0.00 (0.00) 0.32 −0.00 (0.00) 0.03 * 0.00 (0.00) 0.70 0.00 (0.00) 0.17
 Hyper (>180 mg·dL−1) −0.00 (0.00) 0.96 0.00 (0.00) 0.21 0.00 (0.00) 0.55 0.00 (0.00) 0.25 0.00 (0.00) 0.15 0.00 (0.00) 0.83 0.00 (0.00) 0.32 −0.00 (0.00) 0.03 * 0.00 (0.00) 0.70 0.00 (0.00) 0.18
Glucose concentration, mg·dL−1
 24-h mean 0.04 (0.02) 0.16 −0.06 (0.03) 0.07 −0.17 (0.08) 0.05 * −0.06 (0.03) 0.04 *,**,*** **** −0.00 (0.00) 0.06 −0.01 (0.00) 0.07 −0.00 (0.00) 0.06 −0.02 (0.01) 0.07 −0.04 (0.02) 0.10 −0.02 (0.01) 0.07
24-h glycemic variability, mg·dL−1
 MAGE 0.01 (0.02) 0.56 −0.04 (0.02) 0.10 −0.06 (0.06) 0.39 −0.03 (0.02) 0.12 −0.00 (0.00) 0.14 −0.00 (0.00) 0.42 −0.00 (0.00) 0.21 −0.02 (0.01) 0.12 −0.01 (0.02) 0.60 −0.01 (0.01) 0.19
 CONGA-1 0.00 (0.01) 0.87 −0.01 (0.01) 0.26 −0.01 (0.03) 0.62 −0.01 (0.01) 0.30 −0.00 (0.00) 0.19 −0.00 (0.00) 0.54 −0.00 (0.00) 0.29 −0.00 (0.00) 0.28 −0.00 (0.00) 0.88 −0.00 (0.00) 0.46
 CONGA-2 0.01 (0.00) 0.56 −0.02 (0.01) 0.15 −0.03 (0.03) 0.40 −0.01 (0.01) 0.16 −0.00 (0.00) 0.14 −0.00 (0.00) 0.41 −0.00 (0.00) 0.21 −0.01 (0.00) 0.24 −0.00 (0.01) 0.61 −0.00 (0.00) 0.29
 CONGA-4 0.01 (0.01) 0.34 −0.02 (0.01) 0.10 −0.03 (0.03) 0.35 −0.02 (0.01) 0.12 −0.00 (0.00) 0.19 −0.00 (0.00) 0.48 −0.00 (0.00) 0.26 −0.01 (0.01) 0.14 −0.00 (0.01) 0.66 −0.00 (0.00) 0.25
 MODD 0.00 (0.00) 0.73 −0.01 (0.01) 0.59 −0.02 (0.02) 0.36 −0.01 (0.01) 0.48 −0.00 (0.00) 0.53 −0.00 (0.00) 0.43 −0.00 (0.00) 0.47 −0.00 (0.00) 0.81 −0.00 (0.01) 0.59 −0.00 (0.00) 0.82
Waking glycemia, % of day
 TIR (70–140 mg·dL−1) 0.00 (0.00) 0.57 −0.00 (0.00) 0.14 −0.00 (0.00) 0.11 −0.00 (0.00) 0.10 −0.00 (0.00) 0.11 −0.00 (0.00) 0.16 −0.00 (0.00) 0.12 0.00 (0.00) 0.02 −0.00 (0.00) 0.35 −0.00 (0.00) 0.16
 TIR (70–180 mg·dL−1) 0.00 (0.00) 0.45 −0.00 (0.00) 0.11 −0.00 (0.00) 0.18 −0.00 (0.00) 0.10 −0.00 (0.00) 0.13 −0.00 (0.00) 0.19 −0.00 (0.00) 0.14 0.00 (0.00) 0.01 −0.00 (0.00) 0.38 −0.00 (0.00) 0.13
 Hyper (>140 mg·dL−1) −0.00 (0.00) 0.61 0.00 (0.00) 0.08 0.00 (0.00) 0.31 0.00 (0.00) 0.09 0.00 (0.00) 0.06 0.00 (0.00) 0.57 0.00 (0.00) 0.15 −0.00 (0.00) 0.01 0.00 (0.00) 0.44 0.00 (0.00) 0.06
 Hyper (>180 mg·dL−1) −0.00 (0.00) 0.61 0.00 (0.00) 0.09 0.00 (0.00) 0.31 0.00 (0.00) 0.09 0.00 (0.00) 0.06 0.00 (0.00) 0.56 0.00 (0.00) 0.16 −0.00 (0.00) 0.01 0.00 (0.00) 0.44 0.00 (0.00) 0.06
Glucose concentration, mg·dL−1
 Waking, mean 0.03 (0.03) 0.22 −0.05 (0.03) 0.13 −0.18 (0.08) 0.03 −0.04 (0.02) 0.07 −0.00 (0.00) 0.03 −0.01 (0.00) 0.04 −0.00 (0.00) 0.03 −0.02 (0.01) 0.09 −0.04 (0.02) 0.06 −0.02 (0.01) 0.08
Waking glycemic variability, mg·dL−1
 MAGE 0.01 (0.02) 0.50 −0.04 (0.02) 0.09 −0.06 (0.06) 0.37 −0.03 (0.02) 0.11 −0.00 (0.00) 0.13 −0.00 (0.00) 0.40 −0.00 (0.00) 0.20 −0.02 (0.01) 0.13 −0.01 (0.02) 0.57 −0.01 (0.01) 0.18
 CONGA-1 0.00 (0.00) 0.97 −0.01 (0.01) 0.26 −0.01 (0.03) 0.75 −0.01 (0.01) 0.33 −0.00 (0.00) 0.33 −0.00 (0.00) 0.85 −0.00 (0.00) 0.49 −0.01 (0.00) 0.23 0.00 (0.01) 0.89 −0.00 (0.00) 0.52
 CONGA-2 0.01 (0.01) 0.51 −0.02 (0.01) 0.10 −0.05 (0.03) 0.18 −0.02 (0.01) 0.09 −0.00 (0.00) 0.05 −0.00 (0.00) 0.20 −0.00 (0.00) 0.08 −0.01 (0.01) 0.14 −0.01 (0.01) 0.31 −0.01 (0.00) 0.15
 CONGA-4 0.01 (0.01) 0.35 −0.02 (0.01) 0.10 −0.03 (0.03) 0.34 −0.02 (0.01) 0.11 −0.00 (0.00) 0.19 −0.00 (0.00) 0.48 −0.00 (0.00) 0.27 −0.01 (0.00) 0.12 −0.00 (0.01) 0.64 −0.00 (0.00) 0.23
 MODD 0.01 (0.01) 0.57 −0.02 (0.01) 0.28 −0.05 (0.03) 0.13 −0.01 (0.01) 0.19 −0.00 (0.00) 0.26 −0.00 (0.00) 0.20 −0.00 (0.00) 0.22 −0.01 (0.01) 0.33 −0.01 (0.01) 0.22 −0.00 (0.00) 0.29
Data represented as unadjusted β estimate, standard error, and P value for unadjusted linear regression model, such that the DVs were set as CGM measurements, whereas IVs were set as accelerometer measurements. A single accelerometer measure (IV) was input at a time for analysis with no model, including multiple or all accelerometer measures. Bold indicates P < 0.05, and italics indicate 0.05 ≤ P ≤ 0.10 for unadjusted linear regression model (model 1).
*P < 0.05 adjustment for age only (model 2).
**P < 0.05 adjustment for BMI only (model 3).
***P < 0.05 adjustment for race/ethnicity only (model 5).
****P < 0.05 adjustment for sex only (model 6).

Significant negative associations between the SD of total PA time with percent of time spent in hyperglycemia per type 2 diabetic criteria (model 1; β = −0.00, P = 0.04) and 24-h mean glucose concentration were also observed (model 1; β = −0.06, P = 0.04; Table 4). Lastly, the SD of LPA, MVPA, and total PA EE expressed a significantly negative relationship with 24-h mean glucose concentration (model 1; β = −0.01 for all, 0.01 ≤ P ≤ 0.03). This indicates greater day-to-day variability for total PA time, PA EE, and PA MET-minutes per day corresponding to less time each day spent in hyperglycemia, as well as lower 24-h mean glucose concentration. Many of these relationships persisted after adjustment for age and/or BMI (models 2–4; β estimate ≥0.01, P ≤ 0.04 for all), but were lost when adjusting for race/ethnicity and/or sex (models 5–7) and all covariates combined (model 8; P > 0.05 for all).

TABLE 4 - Linear Regression Analysis between Day-to-Day Variability across Average Daily Sedentary Time and PA Measures (Expressed as SD) and 24-h and Waking Glycemia Measurements, Glucose Concentrations, and Measures of Glycemic Variability.
SD, min·d−1 PA EE SD, kcal·d−1 SD, MET-min·d−1
Sedentary LPA MVPA Total PA LPA MVPA Total PA LPA MVPA Total PA
24-h glycemia, % of day
 TIR (70–140 mg·dL−1) −0.00 (0.00) 0.91 −0.00 (0.00) 0.84 −0.00 (0.00) 0.72 −0.00 (0.00) 0.18 −0.00 (0.00) 0.82 −0.00 (0.00) 0.21 −0.00 (0.00) 0.44 −0.00 (0.00) 0.15 −0.00 (0.00) 0.14 −0.00 (0.00) 0.35
 TIR (70–180 mg·dL−1) 0.00 (0.00) 0.62 −0.00 (0.00) 0.16 −0.00 (0.00) 0.22 −0.00 (0.00) 0.14 −0.00 (0.00) 0.76 −0.00 (0.00) 0.24 −0.00 (0.00) 0.44 −0.00 (0.00) 0.09 −0.00 (0.00) 0.14 −0.00 (0.00) 0.22
 Hyper (>140 mg·dL−1) −0.00 (0.00) 0.61 −0.00 (0.00) 0.30 −0.00 (0.00) 0.46 −0.00 (0.00) 0.04 −0.00 (0.00) 0.07 −0.00 (0.00) 0.45 −0.00 (0.00) 0.16 −0.00 (0.00) 0.22 −0.00 (0.00) 0.39 −0.00 (0.00) 0.15
 Hyper (>180 mg·dL−1) −0.00 (0.00) 0.96 0.00 (0.00) 0.21 0.00 (0.00) 0.55 0.00 (0.00) 0.25 −0.00 (0.00) 0.07 −0.00 (0.00) 0.44 −0.00 (0.00) 0.16 −0.00 (0.00) 0.23 −0.00 (0.00) 0.40 −0.00 (0.00) 0.14
Glucose concentration, mg·dL−1
 24-h mean 0.04 (0.02) 0.16 −0.06 (0.03) 0.07 −0.17 (0.08) 0.05 * −0.06 (0.03) 0.04 −0.01 (0.00) 0.01 *,**,*** −0.01 (0.00) 0.03 *,**,*** −0.01 (0.00) 0.01 *,**,*** −0.61 (0.37) 0.11 −0.44 (0.21) 0.05 −0.31 (0.18) 0.10
24-h glycemic variability, mg·dL−1
 MAGE 0.01 (0.02) 0.56 −0.04 (0.02) 0.10 −0.06 (0.06) 0.39 −0.03 (0.02) 0.12 −0.00 (0.00) 0.27 −0.00 (0.00) 0.39 −0.00 (0.00) 0.29 0.02 (0.29) 0.94 −0.10 (0.17) 0.54 0.03 (0.14) 0.86
 CONGA-1 0.00 (0.01) 0.87 −0.01 (0.01) 0.26 −0.01 (0.03) 0.62 −0.01 (0.01) 0.30 −0.00 (0.00) 0.32 −0.00 (0.00) 0.44 −0.00 (0.00) 0.34 0.06 (0.11) 0.59 −0.01 (0.07) 0.79 0.04 (0.06) 0.53
 CONGA-2 0.01 (0.00) 0.56 −0.02 (0.01) 0.15 −0.03 (0.03) 0.40 −0.01 (0.01) 0.16 −0.00 (0.00) 0.36 −0.00 (0.00) 0.51 −0.00 (0.00) 0.39 0.08 (0.14) 0.58 −0.02 (0.08) 0.85 0.05 (0.07) 0.50
 CONGA-4 0.01 (0.01) 0.34 −0.02 (0.01) 0.10 −0.03 (0.03) 0.35 −0.02 (0.01) 0.12 −0.00 (0.00) 0.34 −0.00 (0.00) 0.54 −0.00 (0.00) 0.40 −0.00 (0.15) 1.00 −0.04 (0.09) 0.63 0.01 (0.07) 0.89
 MODD 0.00 (0.00) 0.73 −0.01 (0.01) 0.59 −0.02 (0.02) 0.36 −0.01 (0.01) 0.48 −0.00 (0.00) 0.06 −0.00 (0.00) 0.11 −0.00 (0.00) 0.06 −0.13 (0.10) 0.19 −0.10 (0.06) 0.11 −0.08 (0.05) 0.11
Waking glycemia, % of day
 TIR (70–140 mg·dL−1) 0.00 (0.00) 0.77 0.00 (0.00) 0.99 −0.00 (0.00) 0.14 0.00 (0.00) 0.49 −0.00 (0.00) 0.90 −0.00 (0.00) 0.09 −0.00 (0.00) 0.35 −0.00 (0.00) 0.43 −0.00 (0.00) 0.06 −0.00 (0.00) 0.68
 TIR (70–180 mg·dL−1) 0.00 (0.00) 0.86 −0.00 (0.00) 0.62 −0.00 (0.00) 0.32 0.00 (0.00) 0.74 −0.00 (0.00) 0.89 −0.00 (0.00) 0.16 −0.00 (0.00) 0.43 −0.00 (0.00) 0.22 −0.00 (0.00) 0.12 −0.00 (0.00) 0.46
 Hyper (>140 mg·dL−1) −0.00 (0.00) 0.61 −0.00 (0.00) 0.34 −0.00 (0.00) 0.61 −0.00 (0.00) 0.09 −0.00 (0.00) 0.13 −0.00 (0.00) 0.56 −0.00 (0.00) 0.25 −0.00 (0.00) 0.29 −0.00 (0.00) 0.48 −0.00 (0.00) 0.21
 Hyper (>180 mg·dL−1) −0.00 (0.00) 0.62 −0.00 (0.00) 0.34 −0.00 (0.00) 0.61 −0.00 (0.00) 0.10 −0.00 (0.00) 0.14 −0.00 (0.00) 0.56 −0.00 (0.00) 0.25 −0.00 (0.00) 0.29 −0.00 (0.00) 0.47 −0.00 (0.00) 0.20
Glucose concentration, mg·dL−1
 Waking, mean −0.02 (0.05) 0.64 −0.20 (0.11) 0.08 −0.39 (0.13) 0.01 *,**,****,***** −0.19 (0.08) 0.03 −0.01 (0.00) 0.01 *,**,****,***** −0.01 (0.00) 0.01 *,**,****,***** −0.01 (0.00) <0.01 *,**,****,***** −0.69 (0.35) 0.06 −0.49 (0.20) 0.02 *,**,****,***** −0.36 (0.17) 0.04
Waking glycemic variability, mg·dL−1
 MAGE 0.17 (0.04) 0.65 −0.10 (0.08) 0.25 −0.11 (0.11) 0.33 −0.06 (0.07) 0.36 −0.00 (0.00) 0.31 −0.00 (0.00) 0.40 −0.00 (0.00) 0.32 0.03 (0.28) 0.91 −0.09 (0.16) 0.58 0.03 (0.14) 0.81
 CONGA-1 0.00 (0.02) 0.92 −0.03 (0.04) 0.37 −0.03 (0.05) 0.51 −0.02 (0.02) 0.57 −0.00 (0.00) 0.48 −0.00 (0.00) 0.69 −0.00 (0.00) 0.55 0.07 (0.12) 0.60 −0.01 (0.07) 0.93 0.04 (0.06) 0.54
 CONGA-2 0.00 (0.02) 0.65 −0.05 (0.05) 0.33 −0.07 (0.06) 0.26 −0.03 (0.04) 0.35 −0.00 (0.00) 0.18 −0.00 (0.00) 0.36 −0.00 (0.00) 0.22 0.05 (0.15) 0.74 −0.03 (0.09) 0.70 0.03 (0.07) 0.66
 CONGA-4 0.00 (0.02) 0.87 −0.05 (0.04) 0.26 −0.06 (0.06) 0.35 −0.03 (0.03) 0.43 −0.00 (0.00) 0.37 −0.00 (0.00) 0.51 −0.00 (0.00) 0.40 −0.01 (0.15) 0.96 −0.05 (0.08) 0.56 0.00 (0.07) 0.90
 MODD −0.01 (0.02) 0.79 −0.04 (0.05) 0.37 −0.13 (0.06) 0.03 −0.05 (0.04) 0.21 −0.00 (0.00) 0.08 −0.00 (0.00) 0.05 −0.00 (0.00) 0.05 −0.10 (0.15) 0.52 −0.15 (0.09) 0.10 −0.07 (0.08) 0.37
Data represented as unadjusted β estimate, standard error, and P value for unadjusted linear regression model, such that the DVs were set as continuous glucose monitor measurements, while IVs were set as accelerometer measurements. A single accelerometer measure (IV) was input at a time for analysis with no model, including multiple or all accelerometer measures. Bold indicates P < 0.05, and italics indicate 0.05 ≤ P ≤ 0.10 for unadjusted linear regression model (model 1).
*P < 0.05 adjustment for age only (model 2).
**P < 0.05 adjustment for BMI only (model 3).
***P < 0.05 adjustment for age and BMI (model 4).
****P < 0.05 adjustment for race/ethnicity only (model 5).
*****P < 0.05 adjustment for sex only (model 6).

No significant relationships were observed between sedentary time and PA measures or the day-to-day variability in sedentary time and PA measures with 24-h glycemic variability indices, including MAGE, CONGA-1, CONGA-2, CONGA-4, and MODD (unadjusted and adjusted).

Relationship Between Accelerometer and Waking CGM Measures

Because PA is a waking behavior, we further examined the relationship between sedentary time and PA measures with waking CGM measures, established from self-reported time in and out of bed. The results were similar to the 24-h analysis; LPA MET-minutes per day was found to be significantly positively associated with percent of waking TIR (model 1; β = 0.00 for both, 0.01 ≤ P ≤ 0.02) and significantly negatively associated with percent of waking time spent in hyperglycemia (model 1; β = −0.00 for both, P = 0.03 for both) based on nondiabetic and type 2 diabetic criteria, respectively. Furthermore, a significant negative association was observed between MVPA time and waking mean glucose concentration (model 1; β = −0.18, P = 0.03), suggesting that for every 30-minute increase in MVPA time, waking mean glucose concentration decreases 5.4 mg·dL−1. LPA, MVPA, and total PA EE were also negatively associated with waking mean glucose concentration (model 1; −0.01 ≤ β ≤ −0.00, 0.03 ≤ P ≤ 0.04). Many of these associations were lost after adjusting for age, BMI, race/ethnicity, and/or sex (models 2–8; P > 0.05 for all).

Of note, the SD of MVPA and total PA time, LPA, MVPA, and total PA EE, as well as MVPA and total PA MET-minutes per day were all found to be negatively associated with waking mean glucose concentration (−0.39 ≤ β ≤ −0.01, 0.001 ≤ P ≤ 0.04). Many of these relationships persisted after adjustment for age (model 2), BMI (model 3), race/ethnicity (model 5), or sex (model 6) independently (P ≤ 0.04 for all), but were lost when adjusted for age and BMI (model 4), race/ethnicity and sex (model 7), and all covariates (model 8).

Similar to the 24-h analyses, no significant relationships were observed between sedentary time and PA measures or the day-to-day variability in sedentary time and PA measures with waking glycemic variability indices, including MAGE, CONGA-1, CONGA-2, CONGA-4, and MODD (unadjusted and adjusted).

Individual-Level Relationships between Sedentary, LPA, and MVPA Time with 24-h Mean Glucose Concentration, MAGE, and CONGA-4

Because we found high levels of interindividual variability between our participants for PA measures specifically, we further analyzed intraindividual level relationships between time spent sedentary and LPA and MVPA time with 24-h mean glucose concentration, MAGE, and CONGA-4 (Fig. 1). In 15 participants (54%), increased sedentary time was associated with increased 24-h mean glucose concentration, whereas increased LPA (n = 20; 71%) and MVPA (n = 15; 54%) time was associated with decreased 24-h mean glucose concentration. In 14 participants (50%), increased sedentary time and decreased LPA time were associated with increased MAGE, whereas in 16 participants (57%), increased MVPA time was associated with decreased MAGE. Similarly, 16 participants (57%) expressed increased CONGA-4 with increased sedentary time, whereas 15 (54%) and 17 (61%) participants expressed decreased CONGA-4 with increased LPA and MVPA time, respectively.

F1
Figure 1:
Individual-level regression lines between time spent sedentary and performing LPA and MVPA with 24-h mean glucose concentration, MAGE, and CONGA-4. Each individual line on the respective graphs represents an individual participant’s relationship between sedentary, LPA, and MVPA time with 24-h mean glucose concentration, MAGE, and CONGA-4.

DISCUSSION

To our knowledge, this is among the first studies to examine the relationship between objectively assessed sedentary and PA outcomes with CGM-assessed measures in overweight or obese nondiabetic adults. Notably, multiple PA measures and day-to-day variability in these measures were associated with CGM measures of interest, including percent of 24-h and waking TIR and time spent in hyperglycemia based on nondiabetic and type 2 diabetic criteria, as well as waking and 24-h mean glucose concentration; however, no significant relationships were found between accelerometer measures and CGM-accessed glycemic variability indices. Interestingly, individual-level analyses found that 50%–71% of participants expressed anticipated relationships between sedentary and PA time with 24-h mean glucose concentration, MAGE, and CONGA-4. These findings suggest that leveraging data from objective accelerometers and CGM technology could help identify contributors to impaired glucose metabolism. These findings also suggest that individual-level evaluation allows for consideration of potential individual-level PA prescription as a therapeutic for those at greater risk of development of type 2 diabetes, such as those with overweight or obesity.

Previous evidence suggests that increased time spent sedentary is associated with increases in time spent in hyperglycemia and negatively associated with TIR in type 2 diabetic adults (37,38). In addition, studies examining cardiometabolic disease risk biomarkers in adults with or without diagnosed type 2 diabetes found that objectively measured minutes and intensity of PA per day are favorably associated with fasting and 2-h oral glucose tolerance test glucose concentrations (23,39). These studies support our findings that total PA time is significantly and favorably associated with 24-h mean glucose concentration, as well as glycemia TIR and time in hyperglycemia. Our study further expanded the evidence to support the existence of this relationship with previously unexplored CGM-assessed glucose concentrations, as well as valid and reliable measures of PA.

Limited evidence exists regarding the relationship between objectively measured sedentary time and PA with glycemic variability. Our study found no associations between sedentary time and PA with intraday and interday glycemic variability measurements. These findings could be a consequence of sample size in our study; however, previous findings suggest similar results. Gude et al. (40) found that glycemic variability indices, which included MAGE, CONGA-1, and MODD, were not associated with subjective assessment of time spent sedentary or performing PA in adults without diagnosed diabetes. Martyn-Nemeth et al. (41) found that objectively assessed total PA time did not significantly correlate with glycemic variability assessed as the SD of 24-h mean glucose concentration in type 1 diabetic adults. These studies are in line with our findings that increases in PA time may not necessarily be related to improvement in glycemic variability. However, the inclusion of older adults, utilization of adults with diagnosed type 1 and 2 diabetes, or the use of subjective assessment of time spent sedentary and performing PA does not allow for direct translation of their findings to ours.

Unexpectedly, increased day-to-day PA variability was associated with decreases in percent of time in hyperglycemia and 24-h mean glucose concentration. When examining individual level data, the interindividual differences in time spent sedentary and performing LPA and MVPA potentially explain some of our conflicting or null findings. Previous research suggests that it is difficult to account for the day-to-day variability in sedentary time and PA in adults but that stability in daily PA and associations with aspects of physical health, such as chronic disease and mortality, are equivocal if achieving and adhering to prevailing PA guidelines (42). Because the participants in our study were achieving an average of ~65.5 min of MVPA per day during the 7 d of Sensewear Mini Armband wear time, it can be generally accepted that they were not only meeting but exceeding current PA guidelines regardless of the extent of their day-to-day PA variability (43). Therefore, general participation in PA may potentially be more important than PA stability for glycemic health. PA is a multifaceted health behavior and accelerometers capture multiple PA modalities in a free-living setting, including structured (e.g., exercise) or incidental PA such as daily activities at work, home, or during transport (17,18). Although our participants reported engaging in <120 min of resistance or endurance exercise (structured PA) per week during the previous 3 months, this report did not capture other types of PA, which would undoubtedly increase overall daily PA. This study helps provide evidence that when examining PA with accelerometer data, consideration for the modality of PA may be critical (44).

To account for multilevel data, which has been previously explored in adults with type 2 diabetes, individual-level relationship analyses were conducted between sedentary, LPA, and MVPA time with 24-h glucose concentration, MAGE, and CONGA-4. Individual-level relationships aid in elucidating that individual-level variability may impact our findings. These findings were similar in the study performed by McMillan et al. (45) in adults with type 2 diabetes, which noted individual-level differences when examining the relationship between sedentary time and glycemic variability assessed as the SD of 24-h mean glucose concentrations. This suggests that individual-level factors should be considered and evaluated when attempting to understand the relationship between sedentary and PA time with free-living measures of glycemic health, including CGM-assessed glucose concentrations and glycemic variability.

Strengths and Limitations

The primary strengths of this study include the use of the Sensewear Mini Armband accelerometer to objectively measure sedentary time and PA levels, as well as CGM technology to assess glycemic variability, which allows for the observation of a free-living condition as opposed to standard clinical measures. The primary limitation is that participants were from a convenience sample, and statistical power was not calculated a priori. In addition, the results are only generalizable to overweight or obese adults (male or female), 35–55 yr of age, without overt diabetes, and not taking diabetic medications. Another limitation was that diet was not fully considered; only self-reported dietary intake was provided (Supplemental Content 2, Figure, https://links.lww.com/TJACSM/A186) and was not required for study involvement. However, we recommended that each participant attempt to maintain their daily dietary routines during the 7-d monitoring period. This was also considered a strength of the study because not controlling for diet provides insight into each participant’s free-living glucose concentrations and glycemic variability. Because the sample size was small, we were unable to perform analyses between sexes and races that may be pertinent for addressing relationships between outcomes of interest related to sedentary time and PA and glucose concentrations and glycemic variability. Lastly, the Sensewear Mini Armband device is no longer in production, making reproducibility of our findings limited to technological availability. Although we are not able to account for this limitation, the Sensewear Mini Armband device continues to be utilized in research settings as a reliable and valid measure of PA.

Conclusions

Total PA time and LPA MET-minutes per day, as well as the SD of daily MVPA and total PA time, were favorably associated with 24-h and waking mean glucose concentration and glycemia states. After adjustment for age and/or BMI, many of these relationships remained significant. There were also interesting and differential individual-level relationships. This supports our claims that sedentary time and general participation in PA, whether structured, spontaneous, purposeful, or un-purposeful, play a vital role in free-living glycemic health, but further examination into the individual-level characteristics of participants is required to determine the extent of these relationships.

The future of precision medicine to target glycemic variability warrants individual-level prescription. To effectively improve this measure of glycemic health, clinicians need to understand those who are most at risk for impaired glycemic health (i.e., increased glycemic variability) and whether PA may be the most adequate therapeutic to target and implement in overweight or obese adults. Therefore, based on these considerations and the evidence provided in our presented findings, researchers need to better understand how to incorporate sensitive measurement techniques, such as glycemic variability assessment using CGM, in conjunction with appropriate approaches to enhance PA.

The authors thank Kimberly Bowyer for their hard work and dedication. The results of this study do not constitute endorsement by the American College of Sports Medicine.

The authors disclose no pertinent conflicts of interest. The authors received funding from a University of South Carolina SPARC graduate research grant (no. 11530-17-43917) and an American Heart Association grant (14BGIA20380706).

REFERENCES

1. National Center for Chronic Disease Prevention and Health Promotion. Obesity & Overweight [Internet]. Atlanta (GA): Centers for Disease Control and Prevention; [cited 2022 June 7]. Available from: https://www.cdc.gov/obesity/index.html.
2. NHLBI Obesity Education Initiative Expert Panel on the Identification, Evaluation, and Treatment of Obesity in Adults (US). Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. Bethesda (MD): National Heart, Lung, and Blood Institute; 1998. 262 p.
3. National Heart, Lung, and Blood Institute Overweight and Obesity Expert Panel (US). Managing Overweight and Obesity in Adults: Systematic Evidence Review from the Obesity Expert Panel, 2013. [Internet]. Washington (DC): National Heart, Lung, and Blood Institute; 2013 [cited 2020 June 4]. Available from: https://www.nhlbi.nih.gov/health-topics/managing-overweight-obesity-in-adults.
4. Yu T, Robotham JL, Yoon Y. Increased production of reactive oxygen species in hyperglycemic conditions requires dynamic change of mitochondrial morphology. Proc Natl Acad Sci USA. 2006;103(8):2653–8.
5. Ceriello A, Motz E. Is oxidative stress the pathogenic mechanism underlying insulin resistance, diabetes, and cardiovascular disease? The common soil hypothesis revisited. Arterioscler Thromb Vasc Biol. 2004;24(5):816–23.
6. Centers for Disease Control and Prevention. National Diabetes Statistics Report [Internet]. Atlanta (GA): Centers for Disease Control and Prevention; [cited 2022 April 20]. Available from: https://www.cdc.gov/diabetes/data/statistics-report/index.html.
7. Hall H, Perelman D, Breschi A, et al. Glucotypes reveal new patterns of glucose dysregulation. PLoS Biol. 2018;16(7):e2005143.
8. Monnier L, Colette C, Owens DR. Glycemic variability: the third component of the dysglycemia in diabetes. Is it important? How to measure it?J Diabetes Sci Technol. 2008;2(6):1094–100.
9. Wang C, Lv L, Yang Y, et al. Glucose fluctuations in subjects with normal glucose tolerance, impaired glucose regulation and newly diagnosed type 2 diabetes mellitus. Clin Endocrinol (Oxf). 2012;76(6):810–5.
10. Salkind SJ, Huizenga R, Fonda SJ, et al. Glycemic variability in nondiabetic morbidly obese persons: results of an observational study and review of the literature. J Diabetes Sci Technol. 2014;8(5):1042–7.
11. Freckmann G, Hagenlocher S, Baumstark A, et al. Continuous glucose profiles in healthy subjects under everyday life conditions and after different meals. J Diabetes Sci Technol. 2007;1(5):695–703.
12. Soliman A, DeSanctis V, Yassin M, et al. Continuous glucose monitoring system and new era of early diagnosis of diabetes in high risk groups. Indian J Endocrinol Metab. 2014;18(3):274–82.
13. Kim J, Lam W, Wang Q, et al. In a free-living setting, obesity is associated with greater food intake in response to a similar pre-meal glucose nadir. J Clin Endocrinol Metab. 2019;104(9):3911–9.
14. Zhu X, Zhao L, Chen J, et al. The effect of physical activity on glycemic variability in patients with diabetes: a systematic review and meta-analysis of randomized controlled trials. Front Endocrinol (Lausanne). 2021;12:767152.
15. Sparks JR, Kishman EE, Sarzynski MA, et al. Glycemic variability: Importance, relationship with physical activity, and the influence of exercise. Sports Med Health Sci. 2021;3(4):183–93.
16. Reddy N, Verma N, Dungan K. Monitoring technologies—continuous glucose monitoring, mobile technology, biomarkers of glycemic control. 2020 Aug 16 In: Feingold KR, Anawalt B, Boyce A, et al., editors. Endotext [Internet]. South Dartmouth (MA): MDText.com, Inc.; 2000.
17. Strath SJ, Kaminsky LA, Ainsworth BE, et al. Guide to the assessment of physical activity: Clinical and research applications: a scientific statement from the American Heart Association. Circulation. 2013;128(20):2259–79.
18. Silfee VJ, Haughton CF, Jake-Schoffman DE, et al. Objective measurement of physical activity outcomes in lifestyle interventions among adults: a systematic review. Prev Med Rep. 2018;11:74–80.
19. Tudor-Locke C, Brashear MM, Johnson WD, et al. Accelerometer profiles of physical activity and inactivity in normal weight, overweight, and obese U.S. men and women. Int J Behav Nutr Phys Act. 2010;7:60.
20. Ortega FB, Artero EG, Jiménez-Pavón D, et al. Role of physical activity and fitness in the promotion of metabolic and overall health. Eur J Hum Mov. 2018;41:6–16.
21. Chan CB, Ryan DAJ, Tudor-Locke CT. Health benefits of a pedometer-based physical activity intervention in sedentary workers. Prev Med. 2004;39(6):1215–22.
22. Wing RR, Bolin P, Brancati FL, et al. Cardiovascular effects of intensive lifestyle intervention in type 2 diabetes. N Engl J Med. 2013;369(2):145–54.
23. Healy GN, Dunstan DW, Salmon J, et al. Objectively measured light-intensity physical activity is independently associated with 2-h plasma glucose. Diabetes Care. 2007;30(6):1384–9.
24. Holman RM, Carson V, Janssen I. Does the fractionalization of daily physical activity (sporadic vs. bouts) impact cardiometabolic risk factors in children and youth?PLoS One. 2011;6(10):e25733.
25. Janssen I, Wong SL, Colley R, et al. The fractionalization of physical activity throughout the week is associated with the cardiometabolic health of children and youth. BMC Public Health. 2013;13(1):554.
26. Wang X, Sparks JR, Bowyer KP, et al. Influence of sleep restriction on weight loss outcomes associated with caloric restriction. Sleep. 2018;41(5). doi:10.1093/sleep/zsy027.
27. Sparks JR, Porter RR, Youngstedt SD, et al. Effects of moderate sleep restriction during 8-week calorie restriction on lipoprotein particles and glucose metabolism. Sleep Adv. 2020;1(1). doi:10.1093/sleepadvances/zpab001.
28. Sparks JR, Sarzynski MA, Davis JM, et al. Alterations in glycemic variability, vascular health, and oxidative stress following a 12-week aerobic exercise intervention—a pilot study. Int J Exerc Sci. 2021;14(3):1334–53.
29. Johannsen DL, Calabro MA, Stewart J, et al. Accuracy of armband monitors for measuring daily energy expenditure in healthy adults. Med Sci Sports Exerc. 2010;42(11):2134–40.
30. Laeremans M, Dons E, Avila-Palencia I, et al. Physical activity and sedentary behaviour in daily life: a comparative analysis of the Global Physical Activity Questionnaire (GPAQ) and the SenseWear armband. PLoS One. 2017;12(5):e0177765.
31. Koehler K, Drenowatz C. Monitoring energy expenditure using a multi-sensor device—applications and limitations of the SenseWear armband in athletic populations. Front Physiol. 2017;8:983.
32. Battelino T, Danne T, Bergenstal RM, et al. Clinical targets for continuous glucose monitoring data interpretation: Recommendations from the International Consensus on Time in Range. Diabetes Care. 2019;42(8):1593–603.
33. Shah VN, DuBose SN, Li Z, et al. Continuous glucose monitoring profile in health nondiabetic participants: a multicenter prospective study. J Clin Endocrinol Metab. 2019;104(10):4356–64.
34. Molnar GD, Rosevear JW, et alService FJ. Mean amplitude of glycemic excursions, a measure of diabetic instability. Diabetes. 1970;19(9):644–55.
35. McDonnell CM, Donath SM, Vidmar SI, et al. A novel approach to continuous glucose analysis utilizing glycemic variation. Diabetes Technol Ther. 2005;7(2):253–63.
36. Rodbard D, Bailey T, Jovanovic L, et al. Improved quality of glycemic control and reduced glycemic variability with use of continuous glucose monitoring. Diabetes Technol Ther. 2008;11(11):717–23.
37. Fritschi C, Park H, Richardson A, et al. Association between daily time spent in sedentary behavior and duration of hyperglycemia in type 2 diabetes. Biol Res Nurs. 2016;18(2):160–6.
38. Paing AC, McMillan KA, Kirk AF, et al. The associations of sedentary time and breaks in sedentary time with 24-hour glycaemic control in type 2 diabetes. Prev Med Rep. 2018;12:94–100.
39. Hamasaki H, Noda M, Moriyama S, et al. Daily physical activity assessed by a triaxial accelerometer is beneficially associated with waist circumference, serum triglycerides, and insulin resistance in Japanese patients with prediabetes or untreated early type 2 diabetes. J Diabetes Res. 2015;2015:526201.
40. Gude F, Díaz-Vidal P, Rúa-Pérez C, et al. Glycemic variability and its association with demographics and lifestyles in a general adult population. J Diabetes Sci Technol. 2017;11(4):780–90.
41. Martyn-Nemeth P, Quinn L, Penckofer S, et al. Fear of hypoglycemia: Influence on glycemic variability and self-management behavior in young adults with type 1 diabetes. J Diabetes Complications. 2017;31(4):735–41.
42. Maher JP, Huh J, Intille S, et al. Greater variability in daily physical activity is associated with poorer mental health profiles among obese adults. Ment Health Phys Act. 2018;14:74–81.
43. Piercy KL, Troiano RP, Ballard RM, et al. The physical activity guidelines for Americans. JAMA. 2018;320(19):2020–8.
44. Ross R, Chaput JP, Giangregorio LM, et al. Canadian 24-hour movement guidelines for adults aged 18-64 years and adults aged 65 years or older: an integration of physical activity, sedentary behaviour, and sleep. Appl Physiol Nutr Metab. 2020;45(10 Suppl. 2):S57–102.
45. McMillan KA, Paing AC, Kirk AF, et al. Measuring group and individual relationship between patterns in sedentary behaviour and glucose in type 2 diabetes adults. Pract Diabetes. 2020;37(1):13–18c.

Supplemental Digital Content

Copyright © 2022 by the American College of Sports Medicine