Abstract
Purpose
To examine the impact of a community-based exercise training intervention on cardiometabolic outcomes in African American men who have a family history of type 2 diabetes.
Methods
The Aerobic Plus Resistance Training and Insulin Sensitivity in African American Men (ARTIIS) study randomized participants into either an exercise training intervention or an information only control group for 5 months. The exercise training intervention consisted of 150 minutes of moderate intensity aerobic activity and two days of resistance training per week, consistent with the current federal physical activity guidelines. Participants in the control group received monthly newsletters featuring topics focused heavily on type 2 diabetes education and prevention. Outcome data were analyzed using repeated measures analysis of covariance (ANCOVA) models and incorporating both intention-to-treat and per-protocol principles.
Results
Adherence to the aerobic and resistance training prescriptions were between 77% and 79%. Despite significant within group improvements in glucose and insulin levels (fasting, 2-hour, 2-hour minus baseline) and Homeostatic Model 2-Insulin Resistance, there were not significant between group differences. There was a marginally significant between group difference for Homeostatic Model 2-Beta (p < 0.06), and significant between group differences in peak cardiorespiratory fitness (p < 0.001) and waist circumference (p = 0.03).
Conclusion
These findings suggest that exercise training in accordance with the current national recommendations is effective in improving some health parameters in middle-aged African American men who have a family history of type 2 diabetes, but did not have a significant impact on glycemic status.
Trial registration
The study is registered in clinicaltrials.gov [].
Keywords: ethnic minority, male, physical activity, fitness, diabetes risk
Introduction
It is estimated that 12.2% of African American men have type 2 diabetes (1). African American men diagnosed with type 2 diabetes have a greater prevalence of cardiovascular disease compared to men without the condition (2). There is a need for prevention strategies that can reduce African American men’s risk of developing both type 2 diabetes and cardiovascular disease.
Regular physical activity is associated with reduced risk of developing diabetes (3) and cardiovascular disease (4). Exercise training trials in non-African American samples have shown that physical activity can positively influence many diabetes and cardiovascular disease risk factors, including fitness (5), insulin resistance (6), and many others (7, 8). Exercise training trials conducted to date including African American men were largely short term (<8 weeks), uncontrolled, limited to small sample sizes (<50 men), and many important diabetes-related outcomes (i.e. glucose, insulin, HbA1c (hemoglobin A1c), insulin resistance) were not measured (9–14). To our knowledge, no controlled, properly powered exercise training trial in African American men exists, leaving a large void in the prevention research literature.
The primary objective of the current study, Aerobic Plus Resistance Training and Insulin Sensitivity in African American Men (ARTIIS), was to determine the effectiveness of a community-based combination training intervention on cardiometabolic outcomes in African American men with an elevated risk for developing type 2 diabetes. The exercise training program was specifically designed to achieve the 2008 physical activity guidelines which were applicable at the time of study conception and initiation (15). Understanding the impact of the physical activity recommendations on the cardiometabolic health of African American men has significant clinical and public health importance.
Research Design and Methods
Study design
ARTIIS was a 5-month trial in which African American men were randomly assigned to either an aerobic plus resistance training intervention conducted in community-based facilities or a healthy living control group (Figure 1. Study Flow Chart). Detailed information about the methods can be found elsewhere (16). The study is registered in clinicaltrials.gov [].
Figure 1.
Study Flowchart
Study Participants
Participants were recruited between November 5, 2012 and January 23, 2015. In order to be eligible for the study, participants had to 1) self-identify as an African American male; 2) be between the ages of 35 to 70 years inclusive; 3) have a BMI ≥ 25.0 kg/m2 and ≤ 45 kg/m2; 4) be insufficiently active (≤ 3 d·wk−1 for 20 min each bout during the previous 6 months) and not be currently participating in regular resistance exercise; and 5) have a nuclear family member who had been diagnosed with type 2 diabetes. The exclusion criteria included men 1) with a history of significant cardiovascular disease (e.g. arrhythmias, cardiomyopathy, congestive heart failure); 2) with other serious medical conditions (e.g. chronic respiratory, gastrointestinal, neuromuscular disorders); 3) with clinically significant lab values outside of the acceptable range or showed abnormalities on assessments of blood pressure or cardiac function (at rest and under stress); or 4) taking any antihyperglycemic medication (e.g. metformin, insulin, liraglutide, etc.). Men diagnosed with diabetes, defined as a) having a previous diagnosis of type 1 or type 2 diabetes mellitus; b) currently taking medications for type 2 diabetes (e.g. metformin); or c) having baseline lab values clinically diagnostic of type 2 diabetes (fasting glucose values > 125 mg/dL or 120-minute oral glucose tolerance test glucose values ≥ 200 mg/dL) (17), were excluded. The study protocol was approved and annually reviewed by the Pennington Biomedical Institutional Review Board, and all participants provided written informed consent.
Interventions
Healthy living control (HL) group
Participants received monthly newsletters featuring topics focused heavily on type 2 diabetes education and prevention. After the participants completed their follow-up clinic testing, they were given a five-month membership to the Young Men’s Christian Association (YMCA).
Aerobic plus resistance training (A+R) group
Participants were prescribed 150 minutes of moderate to vigorous aerobic training and two days of 20–30 minutes of resistance training per week in accordance with 2008 Physical Activity Guidelines for Americans (15). The target training intensity (treadmill speed and grade) was self-selected within a range of the heart rate associated with 65% to 85% of VO2 max, which is considered to be within the moderate to vigorous intensity range (18). The aerobic training component was completed on a treadmill and heart rate was assessed continuously with a polar heart rate monitor (Polar FT1, Lake Success, NY) during the treadmill sessions. The resistance training regimen remained constant at 2 days per week, for 20–30 minutes, regardless of the number of aerobic training days. Each resistance training session consisted of 2 sets of 9 exercises with each set consisting of 12 repetitions to failure. Once the participant was able to lift two sets of 12 repetitions for a given exercise for two consecutive sessions, the load was increased. The circuit resistance exercises included the abdominal crunch, vertical chest press, shoulder press, leg press, leg extension, leg curl, triceps extension, lat pull-down, and seated row. Participants self-monitored all exercise training session activities on specifically designed prescription cards. All exercise sessions were conducted at the Young Men’s Christian Association location of the participant’s choice.
Each participant met with their Young Men’s Christian Association fitness instructor for three sessions during the first week and once per week thereafter. The fitness instructors were responsible for monitoring sessions weekly, problem-solving any difficulties, adjusting the aerobic and/or resistance training regimen as appropriate, and entering the participants exercise data from the participant-completed prescription cards into the Research Electronic Data Capture (REDCap) database (http://www.project-redcap.org/). Each fitness instructor undertook eight hours of study-specific training prior to working with a study participant. In addition, each instructor was monitored during their first two training sessions. Refresher trainings were provided every six months or as needed.
Measurements
Height and weight were measured to the nearest 0.1 cm and 0.1 kg, respectively, and body mass index (BMI) was calculated as kg·m−2. Waist circumference was measured to the nearest 0.1 cm. A blood draw was taken to test fasting blood glucose, insulin, cholesterol, lipoproteins and triglyceride levels. Fasting glucose and insulin were used to calculate the Homeostatic Model Assessment 2 (HOMA2). An oral glucose tolerance test (OGTT) was performed by providing the participant with a 75 g glucose-equivalent oral glucose challenge over the course of 5 minutes. Blood samples were taken at 30, 60, and 120 minutes for glucose and insulin measurement. Blood pressure was measured at rest using a standard mercury sphygmomanometer and during the cardiorespiratory fitness test using a Tango Automated Blood Pressure Monitor (SunTech Medical, Morrisville, NC). Body composition was assessed using the General Electric (General Electric; Milwaukee, WI) Lunar i-Dual energy X-ray absorptiometry. All cardiorespiratory fitness maximal exercise tests were conducted using a standardized graded exercise testing protocol administered on a treadmill (Trackmaster 425, Newton, KS). Muscular strength and endurance were measured via isokinetic testing on a Biodex System 4 Isokinetic Dynamometer (Biodex Medical Systems, Shirley, NY). The Center for Epidemiology Studies Depression Scale (CES-D) (19) and Health Related Quality of Life (SF-36) (20) questionnaires assessed mood and quality of life, respectively. The Block Food Frequency Questionnaire (21) was used to estimate daily intake of selected nutrients (macronutrients (e.g. percent calories from fat), and micronutrients), food group servings, and alcohol intake. The Aerobics Center Longitudinal Study Physical Activity questionnaire was used to collect self-reported data on physical activity (22). Data were collected at Pennington Biomedical.
Participants received $100 for successfully completing baseline assessments, $75 for the week 10 OGTT, and $75 upon completing the follow-up assessments, and therefore, could receive up to $250 for completing the study.
Randomization
Potential participants meeting all eligibility criteria and who provided written informed consent, were randomly assigned in a 1:1 ratio to either the aerobic plus resistance training intervention or the healthy living control group. Block randomization was used to generate the allocation sequence, with each block size equal to four. The randomization sequence was computer generated by Statistical Analysis Software. The study coordinator enrolled participants as directed by the next assignment letter contained in a randomized sequence of sealed and numbered envelopes.
Power analysis
The primary outcome variable for estimating sample size was two-hour insulin response to the OGTT. The primary hypothesis was tested against two-directional alternatives with a 5% significance level. Estimates of variability were based on the results of Tokmakidis (23). The objective was to achieve at least 80% power to detect a minimum final differential in mean effect = –1.5μU·ml−1 (an exercise effect resulting in a 10% reduction) with a maximum standard deviation (SD) = 2.7μU·ml−1. This resulted in a net sample size of at least 52 per group with a projected 95% CI = – 2.5 to –0.5. Drop out was estimated at 10%, resulting in a sample size of 58 African American men for each study group and a total sample size of 116.
Statistical Analysis
The two-hour insulin response to the OGTT was the primary outcome variable. Secondary outcomes included the remaining metabolic, fitness, and strength variables. Tertiary outcomes were the remaining cardiovascular (i.e. blood pressure, blood draw) and body composition factors. All measurements were assessed at baseline and five months, with the exception of the OGTT, blood draw variables, and blood pressure, which were also assessed at 10 weeks. All clinic visits were conducted at least 48 hours after the last exercise bout. Research assessment staff were blinded from participant randomization. Data are expressed as means and standard deviations (Mean ±SD). The analytic objective was to assess the statistical significance of on-study changes in participants who underwent aerobic plus resistance training against on-study changes in participants in the healthy living control group. Wilcoxon’s signed-rank test was used to analyze self-reported questionnaire data because the relevant outcomes were not normally distributed. All other variables were analyzed in terms of linear mixed statistical models as described elsewhere (16). Changes in insulin, glucose and homeostatic model assessment 2 (HOMA2) (https://www.dtu.ox.ac.uk/homacalculator/) from baseline to week 10 and week 20 were analyzed employing repeated measures analysis of covariance (ANCOVA) models. All other outcomes were analyzed using ANCOVA models based on changes from baseline to week 20. Covariates, including age, BMI, and baseline values of outcome measures, were included in all models to account for pre-specified sources of extraneous concomitant variability that may affect the analytical outcomes. Outcome data were first analyzed using the intention-to-treat principle and then by invoking the per-protocol principle to assess the sensitivity of analytical interpretation to the statistical approach employed in the analysis. The per-protocol approach included only the participants who were at least 70% adherent to the aerobic treatment plan. Careful consideration was given to the validity of assumptions underlying the statistical methods by employing model diagnostics available in quartile plots of studentized residuals, component-plus residual plots, and other plots that facilitate identification of leveraged points and outliers. Statistical analyses were performed using SAS® version 9.4 software.
Results
One hundred thirteen men were randomized into the study. Recruitment ended because the requisite sample size was obtained. Table 1 summarizes the baseline characteristics of the study participants. The mean age was 51.8±9.0 years, with a mean BMI in the obese category at 31.6±5.3 and generally unfit with a mean maximal metabolic equivalent value of 8.2±1.4. Men reported being prescribed anti-hypertensive and cholesterol medications at baseline (Table 1). Only one participant in the HL was placed on diabetes medication during the course of the study. There were 6.1% and 22.7% of men in the HL and A+R group, respectively, who reported smoking (data not shown). The average adherence (attendance) and compliance (prescription) to the aerobic training regimen was 77.3% and 76.5%, respectively. The average adherence and compliance to the resistance training regimen was 79.2% and 80.3%, respectively.
Table 1.
Baseline participant characteristics (M±SD).
| All (n = 103) | HL (n = 49) | A+R (n = 54) | |
|---|---|---|---|
| Age, mean ± SD, y | 51.8 ± 9.0 | 53.7 ± 8.0 | 50.1 ± 9.6 |
| Metabolic factors | |||
| Fasting glucose, mg·dL−1 | 94.3 ± 8.8 | 94.2 ± 7.9 | 94.4 ± 9.6 |
| Fasting insulin, uU·mL−1 | 12.3 ± 7.2 | 11.6 ± 6.4 | 13.0 ± 7.8 |
| Anthropometrics | |||
| Weight, lbs. | 99.2 ± 18.2 | 99.9 ± 19.9 | 98.5 ± 16.6 |
| BMI | 31.6 ± 5.3 | 31.6 ± 5.5 | 31.6 ± 5.0 |
| Waist circumference, cm | 105.8 ± 14.0 | 106.6 ± 15.2 | 105.0 ± 12.9 |
| Body fat, % (DEXA) | 32.0 ± 6.4 | 32.5 ± 6.5 | 31.5 ± 6.3 |
| Fat body mass, kg (DEXA) | 32.8 ± 11.6 | 33.8 ± 13.1 | 31.9 ± 10.1 |
| Lean body mass, kg (DEXA) | 63.7 ± 8.7 | 63.5 ± 8.4 | 63.9 ± 9.1 |
| Exercise test variables | |||
| Peak relative VO2, mL·min−1·kg−1 | 24.5 ± 5.5 | 24.4 ± 5.2 | 24.7 ± 5.7 |
| Peak absolute VO2, L·min−1 | 2.4 ± 0.5 | 2.4 ± 0.5 | 2.4 ± 0.6 |
| Time on treadmill, s | 782.0 ± 165.9 | 806.5 ± 191.6 | 759.7 ± 136.6 |
| Speed/grade estimated METs | 8.2 ± 1.4 | 8.2 ± 1.4 | 8.2 ± 1.5 |
| Muscular work, N m | 3995.5 ± 908.7 | 3937.5 ± 830.5 | 4048.1 ± 979.0 |
| Muscular torque, N m·kg−1 | 2.9 ± 0.6 | 2.8 ± 0.5 | 3.0 ± 0.6 |
| Cardiovascular disease factors | |||
| Low-density lipoprotein, mg·dL−1 | 107.3 ± 30.2 | 104.9 ± 28.4 | 109.5 ± 31.9 |
| High-density lipoprotein, mg·dL−1 | 47.2 ± 11.5 | 48.9 ± 10.5 | 45.6 ± 12.2 |
| Triglycerides, mg·dL−1 | 90.0 ± 45.7 | 83.1 ± 35.9 | 96.1 ± 52.5 |
| Systolic blood pressure, mmHg | 123.6 ± 11.7 | 122.3 ± 11.3 | 124.8 ± 12.1 |
| Diastolic blood pressure, mmHg | 81.7 ± 8.0 | 81.3 ± 7.1 | 82.0 ± 8.8 |
| Comorbidities | |||
| SBP ≥ 130 mmHg | 33.0 (%) | 30.6 (%) | 35.2 (%) |
| DBP ≥ 80 mmHg | 58.3 (%) | 57.1 (%) | 59.3 (%) |
| SBP ≥ 130 or DBP ≥ 80 mmHg | 63.1 (%) | 63.3 (%) | 63.0 (%) |
| Triglycerides ≥ 150 mg·dL−1 | 9.7 (%) | 4.1 (%) | 14.8 (%) |
| LDL ≥ 130 mg·dL−1 | 21.4 (%) | 16.3 (%) | 25.9 (%) |
| HDL ≤ 40 mg·dL−1 | 29.1 (%) | 16.3 (%) | 40.7 (%) |
| Medications | |||
| Antidiabetic | 0 | 0 | 0 |
| Antihypertensive | 43 | 19 | 24 |
| Cholesterol lowering | 16 | 6 | 10 |
Metabolic measures
The intent-to-treat analysis (Table 2) did not show a significant group X time interaction or a between group difference for fasting, 2-hour, or the 2-hour minus baseline insulin values (p values > 0.060). The per-protocol results (see Table, Supplemental Digital Content 1, Per-protocol baseline, 10-, and 20-week metabolic and cardiovascular values) showed a significant difference in fasting insulin at Week 10 (p = 0.024).
Table 2.
Baseline, 10-, and 20-week metabolic values (Mean±SD).
| HL Least-Squares Means |
A+R Least-Squares Means |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Change from baseline (95% CI) | Change from baseline (95% CI) | ||||||||||
| P | P | ||||||||||
| Within (20 week) | Within (20 week) | Group | Time | Group X Time | |||||||
| Baseline (n = 49) | 10-weeks (n = 47) | 20-weeks (n = 46) | Baseline (n = 54) | 10-weeks (n = 52) | 20-weeks (n = 48) | ||||||
| Fasting Insulin, uU·mL−1 | 11.5 ± 0.9 | 2.4 (0.4,4.4) | 0.09 (−1.4,1.6) | 0.980 | 13.0 ± 1.1 | −0.33 (−2.3,1.6) | −0.80 (−2.5,0.9) | 0.393 | 0.062 | 0.112 | 0.301 |
| Fasting Glucose, mg·dL−1 | 94.1 ± 1.2 | 3.7 (1.3,6.1) | −0.87 (−2.8,1.0) | 0.481 | 94.2 ± 1.4 | 2.88 (0.5,5.3) | 1.9 (−0.1,3.9) | 0.095 | 0.438 | 0.012 | 0.108 |
| 2h insulin, uU·mL−1 | 108.7 ± 9.4 | −18.0 (−30.2,−5.8) | −22.8 (−39.6,−6.1) | <0.001 | 104.8 ± 8.2 | −24.6 (−37.4,−11.8) | −19.3 (−30.9,−7.8) | <0.001 | 0.810 | 0.926 | 0.468 |
| 2h glucose, mg·dL−1 | 144.2 ± 4.2 | −8.0 (−15.6,−0.5) | −7.4 (−15.8,0.9) | 0.135 | 133.8 ± 3.7 | −8.0 (−15.6,−0.5) | −7.7 (−14.8,−0.6) | 0.012 | 0.353 | 0.854 | 0.957 |
| Delta insulin, uU·mL−1 | 97.2 ± 9.0 | −20.4 (−32.6,−8.3) | −22.9 (−39.7,−6.1) | <0.001 | 91.8 ± 7.3 | −24.3 (−36.4,−12.1) | −18.5 (−29.5,−7.5) | <0.001 | 0.890 | 0.906 | 0.542 |
| Delta glucose, mg·dL−1 | 50.1 ± 3.9 | −11.7 (−19.3,−4.2) | −6.6 (−14.0,0.9) | 0.210 | 39.6 ± 3.4 | −10.9 (−18.8,−3.0) | −9.6 (−16.4,−2.8) | 0.002 | 0.178 | 0.323 | 0.577 |
| HOMA2-IR | 1.7 ± 0.1 | 0.36 (0.1; 0.7) | 0.00 (−0.2; 0.2) | 0.623 | 1.9 ± 0.2 | −0.04 (−0.3; 0.2) | −0.10 (−0.3; 0.1) | 0.745 | 0.217 | 0.268 | 0.152 |
| HOMA2-β | 121.1 ± 6.9 | 4.8 (−5.9; 15.5) | 3.1 (−8.7; 14.9) | 0.991 | 129.9 ± 7.3 | −11.1 (−21.7; −0.5) | −10.0 (−21.1; 1.1) | 0.227 | 0.062 | 0.800 | 0.475 |
HOMA P-values based on log transformed data.
P = 0.027 in per protocol analysis.
The intent-to-treat analysis did not show a significant group X time interaction or a between group difference for fasting, 2-hour, and the 2-hour minus baseline glucose values (p values > 0.107). These results were the same in the per-protocol analysis (p values > 0.199).
There was no significant group X time interaction for HOMA2-insulin resistance or HOMA2-beta (p values > 0.151) nor a between group difference for HOMA2-IR (p = 0.217). There was a marginally significant between group difference for HOMA2-beta (p = 0.062) in the intent-to-treat analysis which was significant (p = 0.005) in the per-protocol analysis. HOMA2-beta values increased in the HL group (+2.0%) and decreased in the aerobic plus resistance group (−9.3%).
Cardiovascular fitness/strength
The intent-to-treat analysis (Table 3) showed significant between group differences in peak and relative VO2, estimated maximal metabolic equivalents, and time on the treadmill (p values < 0.001). There were no between group differences for resistance parameters including torque endurance, strength, or muscular work. These results were similar in the per-protocol analysis (see Table, Supplemental Digital Content 2, Per-protocol change in fitness, body composition, and strength variables).
Table 3.
Change in fitness, body composition, and strength variables.
| HL (n = 44–47)a | A+R (n = 43–48)a | Between Group Change | Effect size (95% CI) | |
|---|---|---|---|---|
| Peak VO2, mL·min−1·kg−1 | −0.74 (−2.0;0.5) | 2.84 (1.7;4.0) | <0.001 | 3.6 (1.8, 5.3) |
| Estimated MET | 0.04 (−0.29;0.37) | 0.98 (0.64;1.3) | <0.001 | 0.94 (0.46, 1.4) |
| Strength, N m | 0.08 (−0.05;0.20) | 0.20 (0.07;0.32) | 0.189 | 0.12 (−0.06, 0.30) |
| Muscular Work, N m | 1.8 (−0.43;4.1) | 4.6 (2.3;7.0) | 0.090 | 2.8 (−0.45, 6.1) |
| Weight, kg | −0.37 (−1.3; 0.54) | −0.53 (−1.4; 0.37) | 0.808 | −0.16 (−1.5, 1.1) |
| Fat Mass Total, kg | −843.6 (−1567.7;−119.6) | −1393.1 (−2117.2;−669.1) | 0.296 | −549.5 (−1588.0, 489.0) |
| Lean Mass Total, kg | 351.7 (−115.6;819.1) | 307.8 (−159.6;775.2) | 0.896 | −43.9 (−711.3, 623.4) |
| Waist Circumference, cm | −0.27 (−1.4;0.87) | −2.1 (−3.2;−1.0) | 0.025 | −1.9 (−3.5, −0.24) |
The sample sizes vary across the outcome measures due to missing data.
All means are adjusted for covariates.
Body composition
There was a significant between group difference in waist circumference (p = 0.025). The HL participants maintained their waist circumference (−0.28%) whereas those in the A+R reduced their circumference by almost 2%. No other body composition between group differences in the intent-to-treat analysis existed. The waist circumference findings were similar in the per-protocol analysis (p = 0.014). In addition, fat mass was significantly reduced in the A+R compared to the HL group (p = 0.017). This analysis demonstrated a reduction of total fat of −2.1% in the healthy living group and −4.9% in the A+R group.
Physiological measures
There were no significant group by time interactions or between group differences in blood pressure, total cholesterol, triglyceride, or lipoprotein levels in either the intent-to-treat (Table 4) or per-protocol analyses. The per-protocol analysis (see Table, Supplemental Digital Content 3, Per-protocol baseline, 10-, and 20-week metabolic and cardiovascular values) showed a marginally significant group difference in diastolic blood pressure (p = 0.085), in which blood pressure decreased 2.7% in the HL group and 4.2% in the aerobic plus resistance group.
Table 4.
Baseline, 10-, and 20-week cardiovascular values (mean±SD).
| HL Least-Squares Means |
A+R Least-Squares Means |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Change from baseline (95% CI) | Change from baseline (95% CI) | ||||||||||
| P | P | ||||||||||
| Within (20 week) | Within (20 week) | Group | Time | Group × Time | |||||||
| Baseline (n = 49) | 10-weeks (n = 47) | 20-weeks (n = 46) | Baseline (n = 54) | 10-weeks (n = 52) | 20-weeks (n = 48) | ||||||
| SBP, mmHg | 122.6 ± 11.4 | −7.2 (−9.8, −4.6) | −3.3 (−5.9, −0.7) | 0.014 | 124.5 ± 12.1 | −9.0 (−11.5, −6.5) | −1.4 (−4.0, 1.2) | 0.280 | 0.979 | <0.001 | 0.157 |
| DBP, mmHg | 81.5 ± 7.2 | −9.5 (−11.5, −7.6) | −2.5 (−4.5, −0.5) | 0.013 | 82.2 ± 8.9 | −11.2 (−13.0, −9.3) | −2.1 (−4.0, −0.1) | 0.036 | 0.548 | <0.001 | 0.295 |
| Total Cholesterol | 170.6 ± 34.7 | −4.7 (−10.6, 1.2) | −5.8 (−11.8, 0.2) | 0.403 | 173.6 ± 40.9 | −7.5 (−13.2, −1.9) | −2.5 (−8.3, 3.4) | 0.056 | 0.931 | 0.505 | 0.294 |
| Triglycerides, mg·dL−1 | 84.0 ± 36.3 | 0.01 (−0.09, 0.11) | −0.05 (−0.14, 0.05) | 0.348 | 97.9 ± 52.8 | −0.06 (−0.16, 0.03) | −0.02 (−0.12, 0.07) | 0.610 | 0.591 | 0.835 | 0.315 |
| HDL, mg·dL−1 | 48.5 ± 10.4 | −0.63 (−2.4, 1.1) | −0.32 (−2.1, 1.4) | 0.715 | 45.4 ± 12.3 | −0.77 (−2.4, 0.89) | 0.72 (−1.0, 2.4) | 0.408 | 0.609 | 0.300 | 0.494 |
| LDL, mg·dL−1 | 105.5 ± 29.2 | −3.5 (−8.2, 1.2) | −4.9 (−9.6, −0.07) | 0.047 | 108.6 ± 32.1 | −6.3 (−10.8, −1.8) | −2.9 (−7.5, 1.7) | 0.214 | 0.856 | 0.666 | 0.310 |
Physical activity
The Wilcoxon signed rank test showed a between group difference in self-reported physical activity at the end of the study (p < 0.001). Participants in the A+R group reported engaging in more physical activity compared to participants in the HL group, which was a difference of 22 maximal metabolic equivalent hrs/week (Table 5). There were no significant effects in the intent-to-treat or per-protocol analyses for any dietary variable.
Table 5.
Change in self-reported physical activity (Mean±SD).
| HL (n = 45) | A+R (n = 42) | |
|---|---|---|
| Baseline METs | 25.0±20.2 | 24.0±24.9 |
| 20-weeks | 28.1±23.5 | 45.9±22.2 |
| Difference | 3.1±26.6 | 21.9±23.2* |
p =< 0.001
Adverse events
There were 21 adverse events in the study and one serious adverse event (Table 6). Seventeen of the 21 adverse events were in participants randomized to the A+R group. Eleven of the seventeen were clearly unrelated to the study, five were either unlikely, possibly, or probably related to the interveniton (e.g. shoulder pain, back pain, nausea during exercise), and one was deemed definitely related to the study (participant fell while walking on the treadmill).
Table 6.
Adverse events.
| Groups | |||
|---|---|---|---|
| HL | A+R | Total | |
| Categories | |||
| Abnormal lab values | 1 | 1 | 2 |
| GI | 0 | 1 | 1 |
| Hypoglycemia | 0 | 1 | 1 |
| Illness | 0 | 2 | 2 |
| Muscle/Joint/Injury | 3 | 9 | 12 |
| Surgery/hospitalization | 0 | 3 | 3 |
| Relatedness to Intervention | |||
| Unrelated | 4 | 11 | 15 |
| Unlikely | 0 | 1 | 1 |
| Possible | 0 | 3 | 3 |
| Probable | 0 | 1 | 1 |
| Definite | 0 | 1 | 1 |
Discussion
This community-based combination exercise training program in African American men at risk for developing type 2 diabetes resulted in significant cardiometabolic changes compared to a control group. The ARTIIS study is one of the first investigations to 1) assess metabolic changes and 2) to test the effect of a replicable model of exercise training, i.e. one that utilizes community resources, in this population. ARTIIS also provides some of the first evidence of the physiological effects of the 2008 physical activity guidelines (15) in African American men. This level of activity represents that minimum amount prescribed in the 2018 physical activity guidelines (24), and future studies can assess the effect of higher levels of activity, up to 300 minutes, which is consistent with current recommendation. This novel study provides important cardiometabolic findings of exercise training in African American men.
The study did not result in between group changes in two-hour insulin response, which was the primary outcome variable. The main cause of this lack of between group differences is that the HL participants showed similar improvements in 2-hr insulin response (as measured by 2-hr insulin levels and 2-hr – baseline) to the A+R participants. It is unclear what led to these positive changes in the HL group. Although HL participants were provided a monthly newsletter, these participants did not show significant changes in objectively measured body composition or fitness, nor were there changes in self-reported dietary intake or physical activity. These results suggest that the participants did not make substantial changes to lifestyle behaviors that would influence insulin resistance. Furthermore, the results cannot be attributed to medication alterations as only one HL participant reported being placed on anti-diabetic medications after enrollment. Changes in the HL participants were also not a function of aberrant participant treatment responses as there were few outliers. It is also possible that participants in the HL group underreported changes in dietary intake, physical activity, and/or medications, although there does not seem to be a plausible explanation for doing so on a group-wide basis.
The HOMA2-beta findings suggest that there were changes in beta cell functioning for men randomized to the exercise training intervention. The lower amount of insulin secretion in exercising men was able to dispose of the same amount of glucose (improved insulin sensitivity), suggesting that they became more metabolically efficient compared to men who maintain activity below the recommended levels. In this way our findings are consistent with other combination exercise training programs in participants without type 2 diabetes (6, 25–28). African American men in the HERITAGE study (29), the only other study assessing the effect of exercise training on metabolic function in this population, also demonstrated similar findings of metabolic efficiency. It is important to note that the ARTIIS participants were relatively “healthy”, meaning that most men did not meet criteria for insulin resistance, hypertension, or hypercholesterolemia, although nearly half the sample were men with pre-diabetes (30). These findings, along with the clinically significant fitness and waist circumference results, suggest that future research should focus on developing maintenance strategies in order to reduce this population’s risk of becoming diagnosed with pre-diabetes or type 2 diabetes.
The participant’s average baseline VO2 max of ~24 mL·min−1·kg−1 is in the 10th percentile of maximal aerobic fitness, (31) placing the men at high risk of developing cardiovascular disease. (32) Fitness increased 11.7% when expressed as mL·min−1·kg−1, indicating that the intervention was delivered. This increase is within the range (9% – 25%) demonstrated in other supervised, aerobic exercise training studies involving African American men (9–14). The fitness changes were also clinically meaningful because the 1 maximal metabolic equivalent increase has been associated with a 15% reduced risk of developing type 2 diabetes (33), and other chronic diseases (34–40).
The findings from the ARTIIS study need to be interpreted in the context of its strengths and limitations. Study strengths include utilizing an aerobic plus resistance training intervention, prescribing an intervention in accordance with the 2008 (15) and 2018 national physical activity guidelines (24), recruiting the largest number of African American men into an exercise training program to date, a community-based setting, and a randomized controlled trial design. One of the major limitations is that the control group experienced positive changes from baseline in several metabolic and cardiovascular variables. It is unclear what led to these positive changes as there were no significant changes in body composition, fitness, or medications. The men in the control group received information on healthy living that may have resulted in small changes in dietary intake. The fact that we used an OGTT and calculated HOMA2-IR and HOMA2-beta instead of using a more sensitive measure of insulin resistance, such as a frequently sampled intravenous glucose tolerance test or the euglycemic-hyperinsulinemic clamp, may have limited our ability to detect changes in glucose regulation and insulin resistance. The fact that only African American men were enrolled may limit the generalizability of the findings, although the advantage is the increase in scientific knowledge on the effects of exercise training in a population in which there was previously limited information.
The ARTIIS study showed that an exercise training program resulted in significant changes in some metabolic, cardiovascular, and body composition outcomes in African American men. The exercise program was in accordance with the current physical activity recommendations of 150 minutes of moderate intensity exercise and two days of strength training and was conducted in a community setting with relatively limited supervision. ARTIIS is one of the first studies to show that exercise training results in metabolic changes in African American men. Future studies should develop strategies to maintain these levels of physical activity in order to have a long-term impact on health disparities.
Supplementary Material
SDC 1: Per-protocol baseline, 10-, and 20-week metabolic and cardiovascular values (Mean±SD).
SDC 2: Per-protocol change in fitness, body composition, and strength variables.
SDC 3: Per-protocol baseline, 10-, and 20-week metabolic and cardiovascular values (Mean±SD).
Acknowledgements
Conflicts of Interest and Source of Funding. This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases [R01 DK09523301]. The work was also supported in part by 1 U54 GM104940 from the National Institute of General Medical Sciences of the National Institutes of Health, which funds the Louisiana Clinical and Translational Science Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The study sponsor was not involved in the design of the study; the collection, analysis, and interpretation of data; writing the report; or the decision to submit the report for publication. No conflicts of interest are declared.
This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases [R01 DK09523301]. The work was also supported in part by 1 U54 GM104940 from the National Institute of General Medical Sciences of the National Institutes of Health, which funds the Louisiana Clinical and Translational Science Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The study sponsor was not involved in the design of the study; the collection, analysis, and interpretation of data; writing the report; or the decision to submit the report for publication. No conflicts of interest are declared. The results of the study do not constitute endorsement by the American College of Sports Medicine.
We would like to thank the staff at Pennington Biomedical, including Johanna Veal, Arwen Marker, Tance Sonnier, Ashunti Pearson, Ronald Monce, Megan Duhe, Sheletta Donatto, Corbin Lemon, and Matthew Abboud, for their assistance with this project. We would also like express our sincere appreciation to Melvin ‘Kip’ Holden, Rani Whitfield, Rudy Macklin, and Dawson Odoms for serving as spokespersons for the study. In addition, we would like to recognize the staff members and directors at the YMCA of Greater Baton Rouge, including Don Clark, Donna Chustz, Ron Smith, Josh Landry, and Randy Brown for their collaborative efforts. Finally, we would like to thank Emanuel Andrews III for his tireless efforts in recruiting the participants for the trial.
The results of the study do not constitute endorsement by the American College of Sports Medicine.
Footnotes
Data Availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
- 1.Centers for Disease Control and Prevention. National Diabetes Statistics Report, 2017. Centers for Disease Control and Prevention, 2017. [Google Scholar]
- 2.Vimalananda VG, Biggs ML, Rosenzweig JL, et al. The influence of sex on cardiovascular outcomes associated with diabetes among older black and white adults. J Diabetes Complications 2014;28(3):316–22. Epub 2014/01/28. doi: 10.1016/j.jdiacomp.2013.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bird SR, Hawley JA. Update on the effects of physical activity on insulin sensitivity in humans. BMJ open sport & exercise medicine. 2016;2(1):e000143 Epub 2017/09/08. doi: 10.1136/bmjsem-2016-000143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nystoriak MA, Bhatnagar A. Cardiovascular Effects and Benefits of Exercise. Frontiers in cardiovascular medicine. 2018;5:135 Epub 2018/10/17. doi: 10.3389/fcvm.2018.00135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Church TS, Blair SN, Cocreham S, et al. Effects of aerobic and resistance training on hemoglobin A1c levels in patients with type 2 diabetes: a randomized controlled trial. JAMA. 2010;304(20):2253–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.AbouAssi H, Slentz CA, Mikus CR, et al. The effects of aerobic, resistance, and combination training on insulin sensitivity and secretion in overweight adults from STRRIDE AT/RT: a randomized trial. J Appl Physiol (1985). 2015;118(12):1474–82. Epub 2015/04/18. doi: 10.1152/japplphysiol.00509.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Albarrati AM, Alghamdi MSM, Nazer RI, Alkorashy MM, Alshowier N, Gale N. Effectiveness of Low to Moderate Physical Exercise Training on the Level of Low-Density Lipoproteins: A Systematic Review. Biomed Res Int. 2018;2018:5982980 Epub 2018/12/06. doi: 10.1155/2018/5982980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ross R, Hudson R, Stotz PJ, Lam M. Effects of exercise amount and intensity on abdominal obesity and glucose tolerance in obese adults: a randomized trial. Ann Intern Med. 2015;162(5):325–34. Epub 2015/03/04. doi: 10.7326/m14-1189. [DOI] [PubMed] [Google Scholar]
- 9.Akinpelu AO. Responses of the African hypertensive to exercise training: preliminary observations. J Hum Hypertens. 1990;4(2):74–6. [PubMed] [Google Scholar]
- 10.Kokkinos PF, Narayan P, Colleran JA, et al. Effects of regular exercise on blood pressure and left ventricular hypertrophy in African-American men with severe hypertension. N Engl J Med. 1995;333(22):1462–7. [DOI] [PubMed] [Google Scholar]
- 11.Bond V, Bartels MN, Sloan RP, et al. Exercise training favourably affects autonomic and blood pressure responses during mental and physical stressors in African-American men. J Hum Hypertens. 2009;23(4):267–73. Epub 2008/10/10. doi: 10.1038/jhh.2008.125. [DOI] [PubMed] [Google Scholar]
- 12.Bond V, Stephens Q, Adams RG, et al. Aerobic exercise attenuates an exaggerated exercise blood pressure response in normotensive young adult African-American men. Blood Press. 2002;11(4):229–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hanson ED, Sheaff AK, Sood S, et al. Strength training induces muscle hypertrophy and functional gains in black prostate cancer patients despite androgen deprivation therapy. J Gerontol A Biol Sci Med Sci. 2013;68(4):490–8. Epub 2012/10/24. doi: 10.1093/gerona/gls206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Taylor TR, Makambi K, Sween J, Roltsch M, Adams-Campbell LL. The effect of a supervised exercise trial on exercise adherence among African American Men: a pilot study. J Natl Med Assoc. 2011;103(6):488–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.United States Department of Health and Human Services. 2008 Physical Activity Guidelines for Americans. United States Department of Health and Human Services, 2008. [Google Scholar]
- 16.Newton RL Jr., Johnson WD, Hendrick C, et al. A randomized controlled exercise training trial on insulin sensitivity in African American men: The ARTIIS study: Major category: study design, statistical design, study protocols. Contemp Clin Trials. 2015;43:75–82. Epub 2015/05/17. doi: 10.1016/j.cct.2015.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Diagnosis and classification of diabetes mellitus. Diabetes Care. 2014;37 Suppl 1:S81–90. Epub 2013/12/21. doi: 10.2337/dc14-S081. [DOI] [PubMed] [Google Scholar]
- 18.Pescatello LS, Arena R, Riebe D, Thompson PD. ACSM’s Guidelines for Exercise Testing and Prescription. 9 ed. Philadelphia: Walters Kluwer/Lippincott Williams & Wilkins; 2014. 456 p. [Google Scholar]
- 19.Nguyen HT, Kitner-Triolo M, Evans MK, Zonderman AB. Factorial invariance of the CES-D in low socioeconomic status African Americans compared with a nationally representative sample. Psychiatr Res. 2004;126(2):177–87. [DOI] [PubMed] [Google Scholar]
- 20.Wolinsky FD, Miller DK, Andresen EM, Malmstrom TK, Miller JP. Health-related quality of life in middle-aged African Americans. J GerontolB PsycholSci SocSci. 2004;59(2):S118–S23. [DOI] [PubMed] [Google Scholar]
- 21.Harlan LC, Block G. Use of adjustment factors with a brief food frequency questionnaire to obtain nutrient values. Epidemiology. 1990;1(3):224–31. [DOI] [PubMed] [Google Scholar]
- 22.Kohl HW, Blair SN, Paffenbarger RS Jr., Macera CA, Kronenfeld JJ. A mail survey of physical activity habits as related to measured physical fitness. Am J Epidemiol. 1988;127(6):1228–39. [DOI] [PubMed] [Google Scholar]
- 23.Tokmakidis SP, Zois CE, Volaklis KA, Kotsa K, Touvra AM. The effects of a combined strength and aerobic exercise program on glucose control and insulin action in women with type 2 diabetes. Eur J Appl Physiol. 2004;92(4–5):437–42. [DOI] [PubMed] [Google Scholar]
- 24.United States Department of Health and Human Services. Physical Activity Guidelines for Americans: 2nd Edition. United States Department of Health and Human Services, 2018. [Google Scholar]
- 25.Donges CE, Duffield R, Guelfi KJ, Smith GC, Adams DR, Edge JA. Comparative effects of single-mode vs. duration-matched concurrent exercise training on body composition, low-grade inflammation, and glucose regulation in sedentary, overweight, middle-aged men. Appl Physiol Nutr Metab. 2013;38(7):779–88. Epub 2013/08/29. doi: 10.1139/apnm-2012-0443. [DOI] [PubMed] [Google Scholar]
- 26.Stefanov T, Vekova A, Bonova I, et al. Effects of supervised vs non-supervised combined aerobic and resistance exercise programme on cardiometabolic risk factors. Central European journal of public health. 2013;21(1):8–16. Epub 2013/06/08. [DOI] [PubMed] [Google Scholar]
- 27.Krishnan S, Gustafson MB, Campbell C, Gaikwad NW, Keim NL. Association between circulating endogenous androgens and insulin sensitivity changes with exercise training in midlife women. Menopause. 2014;21(9):967–74. Epub 2014/02/06. doi: 10.1097/gme.0000000000000198. [DOI] [PubMed] [Google Scholar]
- 28.Glynn EL, Piner LW, Huffman KM, et al. Impact of combined resistance and aerobic exercise training on branched-chain amino acid turnover, glycine metabolism and insulin sensitivity in overweight humans. Diabetologia. 2015;58(10):2324–35. Epub 2015/08/10. doi: 10.1007/s00125-015-3705-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Boule NG, Weisnagel SJ, Lakka TA, et al. Effects of exercise training on glucose homeostasis: the HERITAGE Family Study. Diabetes Care. 2005;28(1):108–14. [DOI] [PubMed] [Google Scholar]
- 30.Grundy SM, Brewer HB Jr., Cleeman JI, Smith SC Jr., Lenfant C. Definition of metabolic syndrome: Report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation. 2004;109(3):433–8. [DOI] [PubMed] [Google Scholar]
- 31.Blair SN, Kohl HW III, Barlow CE, Paffenbarger RS Jr., Gibbons LW, Macera CA. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. JAMA. 1995;273(14):1093–8. [PubMed] [Google Scholar]
- 32.Swift DL, Staiano AE, Johannsen NM, et al. Low cardiorespiratory fitness in African Americans: a health disparity risk factor? Sports Med. 2013;43(12):1301–13. Epub 2013/08/29. doi: 10.1007/s40279-013-0092-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Zaccardi F, O’Donovan G, Webb DR, et al. Cardiorespiratory fitness and risk of type 2 diabetes mellitus: A 23-year cohort study and a meta-analysis of prospective studies. Atherosclerosis. 2015;243(1):131–7. Epub 2015/09/20. doi: 10.1016/j.atherosclerosis.2015.09.016. [DOI] [PubMed] [Google Scholar]
- 34.Hayes C, Kriska A. Role of physical activity in diabetes management and prevention. J Am Diet Assoc. 2008;108(4 Suppl 1):S19–S23. [DOI] [PubMed] [Google Scholar]
- 35.Sieverdes JC, Sui X, Lee DC, et al. Physical activity, cardiorespiratory fitness and the incidence of type 2 diabetes in a prospective study of men. Br J Sports Med. 2010;44(4):238–44. [DOI] [PubMed] [Google Scholar]
- 36.Williams PT. Vigorous exercise, fitness and incident hypertension, high cholesterol, and diabetes. Med Sci Sports Exerc. 2008;40(6):998–1006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lee IM, Wolin KY, Freeman SE, Sattlemair J, Sesso HD. Physical activity and survival after cancer diagnosis in men. J Phys Act Health. 2014;11(1):85–90. Epub 2012/12/20. doi: 10.1123/jpah.2011-0257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Lidegaard LP, Hansen AL, Johansen NB, et al. Physical activity energy expenditure vs cardiorespiratory fitness level in impaired glucose metabolism. Diabetologia. 2015;58(12):2709–17. Epub 2015/09/08. doi: 10.1007/s00125-015-3738-x. [DOI] [PubMed] [Google Scholar]
- 39.Juraschek SP, Blaha MJ, Blumenthal RS, et al. Cardiorespiratory fitness and incident diabetes: the FIT (Henry Ford ExercIse Testing) project. Diabetes Care. 2015;38(6):1075–81. Epub 2015/03/15. doi: 10.2337/dc14-2714. [DOI] [PubMed] [Google Scholar]
- 40.Chow LS, Odegaard AO, Bosch TA, et al. Twenty year fitness trends in young adults and incidence of prediabetes and diabetes: the CARDIA study. Diabetologia. 2016;59(8):1659–65. Epub 2016/05/18. doi: 10.1007/s00125-016-3969-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
SDC 1: Per-protocol baseline, 10-, and 20-week metabolic and cardiovascular values (Mean±SD).
SDC 2: Per-protocol change in fitness, body composition, and strength variables.
SDC 3: Per-protocol baseline, 10-, and 20-week metabolic and cardiovascular values (Mean±SD).