Physical Activity Levels and Cardiometabolic Risks in Obese African American Adults: A Pilot Intervention Study

Abstract

Limited information is available regarding the effects of physical activity on risks of cardiometabolic diseases among obese African American adults. We conducted a church-based 12-week weight control and cardiometabolic risk reduction intervention (n=30, 22 females, 56.7±11.4 years old, BMI 37.4±6.7 kg/m²), after which body weight was slightly reduced (98.3±18.4 and 97.3±19 kg, p=.052); body fat percentage was significantly decreased among males (34.7±8.9 to 28.5±8.4 %; p=.049); and walking steps were increased, but not significantly. Among measured cardiometabolic risk biomarkers, hemoglobin A1c (HbA1c) was decreased significantly (6.8±1.1 to 6.1±1.1%; p=.0004) while time spent in sedentary behaviors was associated with less favorable change in total cholesterol (β=11.49, SE=3.55, p=.003) and tumor necrosis factor (TNF-α, β=0.3, SE=0.13, p=.038). Our study shows that adiposity reduction was feasible through a short-term healthy lifestyle program for obese African American adults, and suggests that reducing sedentary behaviors through light physical activity might lead to a decrease in cardiovascular risks.

Healthy aging is becoming a critical health issue as the older population increases. Age-related diseases such as type 2 diabetes, cancer, cardiovascular diseases, and physical disability are linked to lifestyle, and disproportionately affect African Americans.¹ In 2014, age-specific mortality for African Americans and non-Hispanic Whites per 100,000 was 211 and 170 for heart disease, 38 and 19 for diabetes, and 194 and 171 for all cancers, respectively (all p<.05).¹ Physical inactivity is considered an independent risk factor for type 2 diabetes,^2,3 cardiovascular disease and mortality,^3,4 certain cancers (including colon cancer, breast cancer, endometrial cancer),⁵ and increased risk of premature death.⁶ Age-adjusted leisure time inactivity is more prevalent among African Americans than among non-Hispanic Whites: 19% and 25% of African American males and females compared with 10% and 12% for non-Hispanic White males and females, respectively.⁷

Emerging evidence has shown the anti-inflammatory effects of physical activity,⁸ providing more understanding of its protective role in age-related disease prevention. Inflammation is a complex biochemical response to infection or trauma that involves secreting a large family of cytokines.⁹ A prolonged inflammatory state is characterized by elevated levels of inflammatory markers such as C Reactive Protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor (TNF-α), and reduced levels of anti-inflammatory biomarkers such as soluble interleukin-1 receptor α (IL-1Rα), which can result in various adverse health consequences.⁹ The elevated inflammatory state can be attributed to obesity since adipose tissue is an active organ that produces a variety of inflammatory markers;¹⁰ it can also be driven by aging.¹¹ For the most part, observational studies among middle-aged and older people provide promising findings on the inverse associations of the inflammatory biomarkers with physical activity at different levels of intensity and frequencies, independent of adiposity levels.¹² In contrast, the intervention studies—including randomized controlled trials (RCTs)—provide much less consistent findings on the associations of physical activity with inflammatory marker levels. The latter were often implemented among patients with various clinical conditions,^12–15 using heterogeneous physical activity protocols in terms of type, duration, or intensity of exercise training,^15,16 and combined with other intervention approaches such as a hypocaloric diet.^16,17 Most studies were implemented in non-Hispanic White participants and targeted at moderate to vigorous aerobic exercise training; furthermore they often only collected CRP and/or IL-6 and/or TNF-α, and often collected questionnaire-based physical activity measures (although objectively measured physical activity has been increasingly applied in recent studies). Far fewer studies have been implemented in a community setting and among minority groups.

The goal of the present pilot study was to explore the feasibility of a 12-week church-based health promotion behavioral intervention on the biomarkers of cardio-metabolic diseases, with objectively measured physical activity levels among African Americans at risk of such disorders. The design of this pilot study, conducted in collaboration with St. James Missionary Baptist Church in Nashville, Tennessee, has been described previously.¹⁸ Data were collected at baseline and after the 12-week intervention. A primary outcome was the change in body weight; secondary outcomes were changes in the amount of physical activity and biomarkers (lipids, hemoglobin A1c [HbA1c], and inflammatory markers) after the intervention. The exploratory objective was to evaluate whether the changes in cardiometabolic risks were associated with changes in the objectively measured amounts and patterns of physical activity. The length of the 12-week intervention was primarily based on previous similar community–based or clinical-based studies. In addition, such a period was necessary to detect changes in the biomarker levels caused by lifestyle behavior changes.^19,20 For example, the average lifespan of erythrocytes is approximately 115 to 120 days when hemoglobin glycation measured by HbA1c occurs.²¹ Half-life of other biomarkers ranges from a few minutes for cytokines²² to two days for CRP,²³ three days for LDL-cholesterol,²⁴ and five days for HDL-cholesterol.²⁴

Methods

Overview of study design.

The design of this study has been described previously.¹⁸ Briefly, the participants were recruited from participants in a trial (The Obesity Reduction Trial: Faith, Activity, Nutrition [FAN]) established at a local church.¹⁸ The intervention program consisted of 12 weekly education sessions promoting healthy eating, physical activity, and meditation to reduce stress, with half-an-hour devoted to each component. The program goals were to help the participants: (1) increase the daily consumption of wholesome food and a pescetarian diet (i.e., a vegetarian diet that include seafood); (2) improve daily physical activity levels; and (3) learn meditation practice skills. The 1^st round was implemented in February to May of 2013. The 2^nd round was conducted in March to early June of 2015. This pilot study is the first step in the development of larger-scale interventions targeting African American communities in the Southeastern U.S. Data were collected at baseline and after the 12-week intervention. The study was approved by the Institutional Review Board (IRB) at Meharry Medical College, and a written informed consent was signed by all participants before they began the study. Participants received a $25 gift card at both baseline and after the 12-week visit when measurements were obtained.

Recruitment.

The eligibility for this study included being 21 years of age or older, self-defined as African American, English speaking, having a body mass index (BMI) above 25 kg/m², having telephone access, and being willing and able to provide informed consent, and to participate in the weekly intervention sessions. Exclusion criteria were a malignant cancer diagnosis, and/or any condition that would preclude participation in the physical activity component of the intervention, unintentional weight loss in excess of five pounds in the previous three months, and pregnancy or lactation. Prospective participants with self-reported inflammation-related conditions such as rheumatoid arthritis, Crohn’s disease, ulcerative colitis, and cancer, and any psychiatric illness were asked to provide a doctor’s approval for study participation.

Procedure.

Participants were assessed at the baseline visit and after the 12-week intervention. The primary outcome was a change in body weight from baseline to the 12-week visit. Secondary outcomes were changes in physical activity measures, blood lipids, HbA1c, body fat percentage, and the inflammatory biomarkers.

Baseline visit.

During the visit, trained study personnel collected demographic and health information from the participants. Height was measured using a portable SECA 213 stadiometer (SECA, Chino, USA) and recorded to the nearest 0.1 cm. Body weight was measured to the nearest 0.1 kg and body composition was assessed using a portable body composition analyzer (Tanita Corporation, Tokyo, Japan). Systolic and diastolic blood pressures were measured using a digital OMRON HEM-907XL blood pressure monitor (OMRON, Kyoto, Japan), and recorded to the nearest 0.1 mmHg. Blood was drawn from each participant and collected in EDTA vacutainers and one plain tube. An aliquot of whole blood was stored at 4°C for analysis of HbA1c and serum specimens were stored at 4°C and tested for blood lipids (total cholesterol, LDL-cholesterol, HDL-cholesterol) at QUEST Diagnostics (QUEST Diagnostics, Madison, USA). Plasma specimen were frozen at −80°C until being analyzed for inflammatory markers (CRP, IL-6, TNF-α, Interferon γ (IFN-γ), and IL-1Rα) at the Vanderbilt Survey and Biospecimen Shared Resource (Vanderbilt University Medical Center, Nashville, U.S.).

Participants were instructed to wear a BodyMedia Fit Armband (BodyMedia, Pittsburgh, U.S.) for at least four days during the week prior to the visit (two weekdays and two weekend days) except during water-related activities such as showering and swimming.

The 12-week visit.

One week after the completion of the 12 weekly sessions, the anthropometric and physical activity measures and blood specimen were collected.

Data Analyses

Measures.

Physical activity.

Daily physical activity and energy expenditure were assessed using a SenseWear Armband (SWA) monitor. The monitor combines information from various sensors such as biaxial accelerometers,^25,26 heat flux (heat being dissipated by the body), galvanic skin response (onset, peak, and recovery of maximal sweat rates), and skin temperature. The information is integrated and processed by software (Professional 8.1 software (BodyMedia, Pittsburgh, U.S.) using proprietary algorithms utilizing an individual’s demographic characteristics (sex, age, height, and weight). Collected data were used to assess minute-by-minute daily energy expenditure (kcal) and time (min) spent in sedentary behaviors, physical activities classified as light, moderate, and vigorous intensity categories, and to predict number of steps walked.

The BodyMedia monitor has good psychometric properties compared with other portable monitors^27,28 and has the advantage of quantifying physical activity at low intensities as well as for unstructured or intermittent physical activities. These characteristics are important because obese adults may perform most many of their daily activities at low intensities with minimal ambulation, and may perform work in very short bouts.

Inflammatory biomarkers.

Plasma CRP level was measured using a Millipore Human C-Reactive Protein (CRP) ELISA kit (EMD Catalog # CYT298).²⁹ Plasma IL-1Rα level was measured by using a Human IL-1Rα/IL-1F3 Quantikine ELISA Kit (R&D Systems Inc., Catalog # DRA00B); IL-6, TNF-α, and IFN-γ were measured by using a MILLIPLEX MAP Human High Sensitivity T Cell Panel (EMD Catalog # HSTCMAG-28SK) on Luminex 200™ analyzer, following the manufacturer’s protocols. The samples collected at baseline and after the 12-week intervention from the same person were analyzed in the same plate. All samples were measured in duplicate and the values were averaged. The lower limits of detection of the assays were 0.20ng/ml for CRP, 6.3pg/ml for IL-1Rα, 0.18pg/ml for IL-6, 0.45pg/ml for TNF-α, and 0.61pg/ml for IFN-γ. The inter-assay coefficients of variation for the two controls were 3.08% (low) and 6.89% (high), and the coefficients of variation for duplicate samples were <8.69%.

Statistical analyses.

Descriptive characteristics were summarized as mean and standard deviation for continuous variables, and frequency for categorical variables. The paired t-test was conducted to examine pre- and post-intervention measures of adiposity, physical activity, and cardiometabolic biomarkers. As exploratory analyses, regressions were computed for the associations between cardiometabolic risk measure changes (dependent variables) and changes in time spent in sedentary behaviors, moderate activity, and daily walking steps (independent variables). Age and body weight change were the covariates. Statistical analyses were conducted using Statistical Analysis System software, SAS 9.3 (SAS Institute, Inc., Cary, NC), and a statistical significance was determined at p<.05.

Results

Protocol adherence.

Thirty-seven participants started the study protocol. Three individuals dropped out during the intervention due to surgery (n=1), conflict with work time (n=1), and household responsibilities (n=1). In total, 34 individuals completed the program, and 30 participants completed physical activity measures, gave blood specimen, and were included in the final analysis.

Characteristics of the participants.

Among 30 individuals (22 females) who completed the intervention (aged 56.7±11.4 years, BMI 37.4±6.7 kg/m²), about half were currently employed, two-thirds had at least some college education, and only one individual was a current smoker (Table 1). Reported medication use was 43.3% for hypertension, 29.4% for high cholesterol and 17.7% for type 2 diabetes medication; this did not change during the 12-week intervention (Table 1).

Table 1.

DEMOGRAPHIC CHARACTERISTICS

Participants’ characteristics
Age at baseline (years)a	56.7±11.4
Currently Employed (%)b
Yes	53.6
No	10.7
Retired	35.7
Education (%)b
Less than high school	3.6
High school	28.6
Some college (including associate degree)	39.3
College and above	28.6
BMI (kg/m2)	37.4±6.7
Obesity (BMI≥30 kg/m2) (%)	86.7
Body fat percentage
Female (n=22)	45.8±8.2
Male (n=8)	34.7±8.9
Systolic blood pressure (mmHg)a	138.1±22.3
Diastolic blood pressure (mmHg)a	83±10.3
Alcohol use (%)b
Never	28.6
Former	21.4
Currentc	50
Smoking status (%)
Never	66.7
Former	30
Currentc	3.3
Perceived Stress Scale (PSS)a	13.5±8
Medication use
Hypertension (%)	43.3
Hyperlipidemia (%)	29.4
Diabetes (%)	17.7

Notes

^amean± STD

^bn=28

^cCurrent refers to currently consuming alcoholic beverages and drinking alcoholic beverages at least once a week.

Changes in physical activity, adiposity, and metabolic and inflammatory biomarkers (Table 2).

Table 2.

CHANGES IN PHYSICAL ACTIVITY LEVELS, METABOLIC AND INFLAMMATORY MEASURES AFTER THE 12-WEEK INTERVENTION

	All (n=30)
Measures	Baseline	After 12-week intervention	p value
Physical activity
Total energy expenditure (kcal)	2514.0±613.4	2472.1±596.9	0.412
Sedentary behaviors (h/day)a	15.4±1.9	15.5±2.7	0.981
Moderate activity time (h/day)	1.1±1.4	1.2±1.8	0.378
Steps (step/day)	4548±2605	4996±2561	0.279
Adiposity
Weight (kg)	98.3±18.4	97.3±19	0.052
BMI (kg/m2)	37.4±6.7	37±6.9	0.073
Body fat percentage
Female	45.8±8.2	44.9±5	0.322
Male	34.7±8.9	28.5±8.4	0.049
Biochemical measures
Blood HbA1C (%)a	6.8±1.1	6.1±1.1	0.0004
Serum lipids
Total cholesterol (mg/dL)	175.7±35.8	166.8±46.8	0.177
LDL-cholesterol (mg/dL)	116.3±42.4	118.5±46	0.665
HDL-cholesterol (mg/dL)	52.7±11.6	43.2±14	0.0008
Plasma Inflammatory markersb
CRP (mg/L)c	8.9±10.1	7.9±6.7	0.487
IL-6 (pg/mL)c	1.6±1.1	1.8±0.8	0.307
TNF-α (pg/mL)c	3.2±1.7	3.7±1.7	0.011
IFN-γ (pg/mL)c	8.1±7.5	9.5±7.1	0.02
IL-1Rα (pg/mL)d	551.2±366.8	476.4±209.5	0.087

Notes

^an=29 (sedentary behaviors refers to the sum of minutes with metabolic equivalent of task (MET)≤1.5 during waking time.²⁵ Moderate activity refers to the sum of minutes with MET >3 and <6 during waking time)

^bn=23

^cpro-inflammatory marker

^danti-inflammatory marker

After the 12-week intervention, body weight was reduced from 98.3±18.4 to 97.3±19 kg (p=.052). Body fat percentage was decreased significantly among men (from 34.7±8.9 to 28.5±8.4 %, p=.049), and was reduced slightly among women (from 45.8±8.2 to 44.9±5 %, p=.322). Daily number of walking steps increased slightly from 4548±2605 to 4996±2561 steps (p=.279). No significant changes were noted in other physical activity measures (Table 2). Significant changes in biochemical biomarkers were decreased HbA1c (from 6.8±1.1 to 6.1±1.1 %; p=.0004) and HDL-c (from 52.7±11.6 to 43.2±14 mg/dL; p=.0008). Among the inflammatory markers, TNF-α increased from 3.2±1.7 to 3.7±1.7 pg/mL (p=.011), IFN-γ increased from 8.1±7.5 to 9.5±7.1 pg/mL (p=.02), and IL-1Rα decreased from 551.2±366.8 to 476.4±209.5 pg/mL(p=.087) (Table 2).

Associations between changes in armband measures with changes in biomarker concentrations.

With adjustment for age and weight change after the 12-week intervention, the increase in time spent in sedentary behaviors was positively associated with an increase in total cholesterol (β=11.49, SE=3.55, p=.003), and TNF-α (β=0.3, SE=0.13, p=.038), but a decrease in IL-1Rα (β= −66.73, SE=29.78, p=.037) (Table 3). In contrast, the increase of time spent in moderate activity was associated with a decrease in total cholesterol (β=−25.63, SE=8.65, p=.006) and LDL-cholesterol (β=−17.59, SE=6.87, p=.017, independent of age and weight change (Table 3). Similar patterns were noted in the associations of daily walking steps (light physical activity) with total cholesterol (β=−7.72, SE=2.81, p=.011) and LDL-cholesterol (β=−6.24, SE=2.12, p=.007) (Table 3).

Table 3.

ASSOCIATIONS OF THE CHANGES IN PHYSICAL ACTIVITY MEASURES WITH THE CHANGES IN BIOMARKER CONCENTRATIONS AFTER THE 12-WEEK INTERVENTION^a.

	ΔSedentary activity time (h)			ΔModerate activity time (h)			ΔSteps (per 1000)
	β	SE	P	β	SE	P	β	SE	P
ΔCholesterol (mg/dL)	11.49	3.55	0.003	−25.63	8.65	0.006	−7.72	2.81	0.011
ΔLDL-cholesterol (mg/dL)	6.14	3.02	0.053	−17.59	6.87	0.017	−6.24	2.12	0.007
ΔHDL-c (mg/dL)	3.83	1.39	0.011	−4.3	3.81	0.269	−0.86	1.23	0.493
ΔHbA1c (mg/dL)b	−0.11	0.12	0.346	0.049	0.28	0.861	0.037	0.1	0.711
ΔCRP (mg/dL)c	1.48	1.23	0.245	−0.1	5.16	0.984	0.32	0.92	0.734
ΔIL-6 (mg/dL)c	0.14	0.12	0.239	−0.18	0.49	0.716	0.02	0.09	0.812
ΔIFN-γ (mg/dL)c	0.65	0.43	0.145	−2.51	1.75	0.168	−0.24	0.33	0.476
ΔTNF-α(mg/dL)c	0.3	0.13	0.038	−0.18	0.61	0.769	0.07	0.11	0.52
ΔIL-1Rα(mg/dL)c	−66.73	29.78	0.037	31.38	135.27	0.819	−7.44	24.22	0.762

Notes

^alinear regression adjusted for age and change in weight after the 12-week intervention

^bn=29 (n=28 for sedentary activity time)

^cn=23

Discussion

This pilot study investigated the effect of a 12-week healthy living behavioral intervention in African American adults at risk of chronic diseases. As a result of the intervention, body weight was reduced slightly, body fat percentage was decreased significantly among men, and the number of walking steps per day increased somewhat. Changes in other measures of physical activity were not significant.

In the exploratory analyses, after adjustment for age and weight change, the difference in time spent in sedentary behaviors was positively associated with changes in total-cholesterol and TNF-α, and was inversely associated with the change in the anti-inflammatory marker IL-1Rα levels. The opposite association patterns were observed between moderate activity time/daily walking steps and total cholesterol/LDL-cholesterol. Our study adds to existing literature by documenting positive association of sedentary behaviors and negative association of a light intensity physical activity (e.g., walking) with cardiometabolic risk biomarkers in African American adults. The findings from this pilot study will help with the design and development of studies in other African American populations.

Another finding of the current study is that the relationship of cardiometabolic risk biomarkers, including multiple inflammatory cytokine measures, with physical activity at different levels were independent of weight and age. Compared with moderate or vigorous physical activity well documented in the literature, less information is available on how sedentary behaviors, also termed physical inactivity, affects cardiometabolic risks in adult African Americans. A recent meta-analysis of randomized controlled trials examined the effects of leisure time physical activities (e.g., yoga, walking) on glycemic control, and found a pooled effect size of 0.6% reduction in HbA1c among adults with type 2 diabetes who performed light physical activities for more than eight weeks at least three times per week.³⁰ Previous studies also reported that light-intensity physical activity was independently associated with reduced insulin resistance,^31,32 whereas sedentary time was a significant predictor of increased insulin resistance.³³ However, the latter association was not shown in a similar study with shorter follow-up (one year).³⁴ Moreover, the increase of sedentary time during a six-year follow-up was associated with an increased triacylglycerol and a clustered cardiometabolic risk index.³⁵ For inflammatory markers, sedentary time showed a positive cross-sectional association with IL-6, which was independent of confounders including moderate to vigorous physical activity, fatness, and medications.³⁶ In a cohort with six-month follow-up, a one-hour decrease in sedentary time during a six-month follow-up was associated with a 24% CRP reduction after the adjustment for age, adiposity, and other confounding factors.³⁷ In the current study, no significant associations of sedentary time with CRP and IL-6 were observed. More data are needed in future studies to test the role of sedentary behaviors and physical activity in regulating CRP and IL-6 levels.

The participants in the current study were middle-aged and older adults, predominantly overweight and obese. Physical limitations were reported by the participants as one of the barriers to engaging in physical exercises,¹⁸ which may have explained the relatively low time spent in moderate-to-vigorous activities during the intervention. We did not find significant associations of time spent in moderate activity with inflammatory markers. Additionally, a previous study suggests that weight loss of approximately 5%−7% is necessary to induce a decrease in the circulating inflammatory marker levels,³⁸ whereas weight reduction in the current study did not reach a similar level. It is possible that replacing sedentary behaviors with light physical activity would be a more feasible goal for this population.

Compared with CRP, IL-6 and TNF-α, less information is available for IFN-γ and IL-1Rα regarding how they can be modulated by physical activity. In the exploratory analysis, we observed a positive association of IFN-γ and an inverse relationship of IL-1Rα with sedentary behaviors. These results are consistent with similar evidence in the literature. For example, after participating in moderate to vigorous exercises for three months or more, individuals with various health conditions had reduced circulating IFN-γ.^39–41

Physical exercise could contribute to the mitigation of cardiometabolic risks through multiple pathways.

Reduced body fatness may mediate the beneficial effects of physical exercises on mitigating cardiometabolic risks.⁴² Adipose tissue is a well-recognized endocrine organ.¹⁰ The adipose tissue of obese individuals shows an increased infiltration of inflammatory cells which in turn generate more pro-inflammatory cytokines such as TNF-α and IL-6.¹⁰ This obesity-associated chronic inflammation state is suggested to be driven by the endothelial dysfunction in the enlarged adipose tissue, i.e., adipose tissue hypoxia (characterized as a decreased angiogenesis, an increased vasoconstriction and a decreased blood flow) and vascular endothelial cell damage.⁴³ Exercise, on the other hand, helps to improve blood flow, reduce vasoconstriction, and attenuate adipose tissue hypoxia, thus to reduce chronic-inflammation, the effects of which are independent of the exercise-induced weight loss.⁴³ Physically active individuals likely have lower adiposity and therefore lower inflammatory marker levels.⁴⁴ Moreover, in obesity excessive free fatty acids are released from adipose tissue, and free fatty acids can inhibit insulin signaling, thus inducing insulin resistance.⁴⁵

It is increasingly acknowledged that skeletal muscle is also an endocrine organ.⁴⁶ For example, in a clinical trial, a single bout of exercise can lead to up to a 100-fold transitory increase in IL-6 plasma level⁴⁷ which in turn stimulates the release of anti-inflammatory cytokines such as IL-1Rα and inhibits the production of pro-inflammatory cytokine TNF-α in muscle fibers.⁴²

Available evidence suggests weight-loss independent mechanisms via which physical activity can attenuate inflammatory responses and favor an anti-inflammatory environment systematically: (1) exercise increases blood flow which helps to attenuate the expression of adhesion molecules in endothelial cells and reduce leukocyte migration into the vessel wall and mitigate local inflammation responses;⁴³ (2) long-term exercise can upregulate the activity of the peroxisome proliferator-activated receptor γ coactivators 1 (PGC-1s) in human skeletal muscles. PGC-1s are involved in the regulation of metabolic and inflammatory processes in skeletal muscle. They facilitate the increase of muscle insulin sensitivity, the induction of anti-inflammatory cytokines, including IL-1Rα and the inhibition of the pro-inflammatory signaling pathway;⁴⁸ (3) for innate immune cells, regular exercise could reduce the size of the monocyte and pro-inflammatory cytokines released by peripheral mononuclear cells.⁴³ In a clinical trial, sedentary individuals showed a higher percentage of circulating pro-inflammatory monocytes than their active counterparts, and they showed a reduced percentage of these monocytes by exercise training.⁴⁹

In the current study, no significant associations were observed between HbA1c and physical activity measures. The modest change of physical activity in the current study may be lack of power to show significant associations between physical activity measures and HbA1c. In a previous systematic review, controlled clinical trials that only provided recommendations of physical activity (which presumably had less effects on participants’ physical activity behavior) did not result in significantly decreased HbA1c. In contrast, trials that provided structured exercise did result in significantly decreased HbAlc and trials in which participants spent more time on exercise (e.g., >150 minutes per week) reported more reduction in HbA1c.⁵⁰ More studies are needed to investigate the effects of low intensity physical activity on insulin sensitivity and glycemic control.

This pilot study has several strengths.

The study objectively measured physical activity and cardiometabolic risk biomarkers including circulating lipids and several inflammatory markers. The objectively measured physical activity helps to reduce the recall bias and to increase accuracy compared with the questionnaire-based physical activity measures.⁵¹ The current study also provided novel findings in an African American population at high risk for cardiovascular disease. Our findings suggest positive associations of sedentary behaviors with elevated plasma IL-6, CRP, TNF-α, and IFN-γ, decreased IL-1Rα, and increased total cholesterol and LDL-cholesterol, and positive associations of light physical activity (i.e., walking steps) with decreased total and LDL-cholesterol levels.

Furthermore, our measures included body fat percentage, less frequently reported in the literature. After the 12-week intervention, we observed reduced weight by approximately 1 kg in both men and women and decreased body fat percentage by approximately 0.8% in women and 6% in men. The magnitude of changes in weight loss and body fat percentage reduction was comparable to the results in the previous lifestyle interventions with 12-week follow-up⁵² or with longer follow-up duration.^53–55

This pilot study has some limitations.

First, since it was a feasibility pilot study, a sample size calculation was not performed before the study. Given the small sample size (8 males, 22 females) a subgroup analysis could not be conducted. Second, data were adjusted for age and weight but not for other lifestyle confounding factors, such as stress or dietary intake. However, no change was noted in self-reported use of medications for elevated blood pressure, lipids, and blood glucose levels throughout the study among the participants who reported such information at baseline. Third, we were not able to test the influence of potential confounding factors such as participant attendance and program implementation fidelity. It has been documented that in addition to being underpowered, studies with small sample size could run into the risk of having inflated effect size.⁵⁶ However, the major goal of this feasibility pilot study was to provide us and other investigators with guidance for refining future interventions and to help us to design larger-scale interventions that can be conducted in various populations.

In the current study, self-reported dietary intake was collected at baseline and after the 12-week intervention, but these self-reported dietary intake data were not included in the analysis since they were not associated with inflammatory markers in our preliminary analyses (data not shown). In previous studies, exercise without concomitant dietary change improved cardiometabolic risk profile by changes such as reduced CRP^57–59 and IL-6,⁵⁹ reduced HbA1c, decreased insulin resistance measure HOMA-IR,⁵⁹ and increased HDL-cholesterol.⁵⁷ In our previous report, we attributed the decreased HDL-cholesterol to a decrease in meat consumption among the participants after the 12 week intervention.¹⁸ Given the important role of dietary intake in regulating cardiometabolic risk factors such as blood lipids, insulin resistance levels, and body weight, poor nutritional status has the potential to exacerbate the cardiometabolic risk biomarker measures among physically inactive individuals, which should be tested in future studies with a larger sample size.

Although we found an increase of TNF-α and IFN-γ and a decrease of IL-1Rα during the study while the participants had reduced adiposity, we could not determine whether these changes were associated with the intervention or other factors since there was no control group. For example, the intervention started in February and ended in May/June when seasonal allergies more likely occur. Both pro-inflammatory cytokines TNF-α and IFN-γ are involved in allergic inflammation,⁶⁰ and could increase due to the environmental factors. Additionally, we cannot rule out that the inflammation status of the individuals recruited was getting worse due to the development of their inflammatory-related conditions over the 12 weeks.

Clinical and public health implications.

Physical activity promotion intervention should be tailored to participants whose health conditions and needs might vary. In the current study, the participants were retired, employed, and unemployed; living alone and living with young children/family; were apparently healthy, taking medications for diabetes/high cholesterol/hypertension, and having conditions such as arthritis. We found that they had distinct obstacles to engaging in the physical activities recommended for chronic disease prevention.¹⁸

Findings from the current study suggest that the reduction of sedentary behaviors and increase of light-intensity activities might be a feasible approach to reducing risk of chronic disease in African American adults.

Our study adds to the existing evidence that the goals of the physical activity component can be modified (1) to facilitate inactive individuals to become more active, with approaches such as practicing a light level of physical activity (e.g., purposeful walking) whenever possible, or integrating frequent breaks and/or light activity while engaging in sedentary behaviors, such as engaging in various types of housework during TV viewing; and (2) to guide younger and more able individuals to adopt more intensive physical exercises.

Future directions.

Emerging evidence has demonstrated positive effects of an increase in intermittent breaks in prolonged sedentary activities on cardiometabolic risk biomarkers.³¹ Clinical trials have shown that frequent breaks in prolonged sitting by short bouts of light levels of activity (e.g., two minutes walking for every 20 minutes of sitting) resulted in improved postprandial glycemia compared with sitting and standing still.⁶¹ Similarly, regular breaks during prolonged sitting were associated with greater improvement in postprandial glycemia and insulinemia than sitting for the same amount of time followed by half an hour of light activity.⁶² Moreover, the inverse associations of breaks in sedentary time with the inflammatory marker IL-6 became stronger among inactive people,³⁶ indicating that a reduction of prolonged sedentary time would bring more health benefits to inactive individuals. Finally, frequent interruption of prolonged sedentary time has been shown to relate independently to attenuated cardiometabolic risk measures including adiposity, triglycerides, and two-hour plasma glucose.⁶³

Given the promising but limited evidence regarding positive effects of an increase of breaks in prolonged sedentary activities, future interventions should be designed to explore further and test: (1) to what extent breaks (e.g., how often, how long, what types of movement) during prolonged sedentary time would be able to provide detectable disease protection effects;^61–63 and (2) whether the beneficial effects of breaks during prolonged sedentary time is homogeneous across population sub-groups or not.⁶⁴ Such interventions will require objective measures of physical activity that can differentiate various types of sedentary behaviors such as sitting time, TV viewing, and other screen-time behaviors, or count the breaks during sedentary time. Findings from such interventions will be helpful to refine the physical activity recommendations for various populations and age groups.

Conclusion.

The results from the current study showed that a relatively short-term (12–week) church-based behavioral intervention may lead to weight and adiposity reduction and raises the possibility that replacing sedentary behaviors with light intensity physical activity may lead to positive changes in cardiovascular risk biomarkers in obese African American adults. These findings could help in designing an effective lifestyle intervention that includes a physical activity component for this population.