Abstract
Background:
Obesity is highly prevalent in African American women, especially those in the rural southern U.S., resulting in persistent health disparities.
Objective:
To test the effectiveness of an evidence-based behavioral weight loss intervention delivered by community health advisors to African American women in the rural south.
Design and Methods:
Overweight or obese African American women (30–70 years) from eight counties in Mississippi and Alabama participated in a 24-month randomized controlled trial of an evidence-based behavioral weight loss program augmented with community strategies to support healthy lifestyles (Weight Loss Plus, N=154) compared to the weight loss program alone (Weight Loss Only, N=255). This study reports on 6-month outcomes on primary (weight change) and secondary (waist circumference, blood pressure, lipids, fasting blood glucose) outcomes, coinciding with the completion of the intensive weight loss phase.
Results:
Weight Loss Only participants lost an average of 2.2 kg (p <.001). Weight Loss Plus participants lost an average of 3.2 kg (p <.001). The proportion of the total sample that lost at least 5% of their body weight was 27.1% with no difference between treatment groups. Similarly, we observed statistically significant reductions in blood pressure, waist circumference, and triglycerides in each treatment group, with no statistical differences between groups.
Conclusion:
Trained lay health staff and volunteers from the rural southern U.S. were able to deliver a translation of a high intensity behavioral intervention targeted to African American women, resulting in clinically meaningful weight loss and improvement in other metabolic outcomes in a significant proportion of participants.
Keywords: weight loss, metabolic outcomes, community-based, randomized trial, African American, rural communities
Introduction
African Americans suffer from a disproportionate prevalence of obesity, especially those in the rural southern U.S. In the U.S., the prevalence of obesity, defined as a body mass index (BMI) of ≥ 30 kg/m2, increased from 30.5% in 1999–2000 to 37.7% in 2013–14 based on data from the National Health and Nutrition Examination Survey (NHANES).[1] While trends in obesity have been generally stable for men since 2003–2004, over the past decade, a positive linear trend of increasing obesity is noted among women even after adjusting for age, race, ethnicity, education and smoking status.[1] This is particularly noted among African American women. [2] African American women have a prevalence of obesity that is greater than 1 in 2 (57.2%) compared to a prevalence of 38.2% in non-Hispanic white women.[1] African American women living in rural settings have a higher prevalence of obesity compared to others within the same race/sex subgroups in urban settings.[3] In Southern states like Alabama and Mississippi, where there are large rural areas and up to one-third of the population is African American, state-wide estimates suggest that African American women have a greater burden of excess weight than white women in the region, and the prevalence of obesity in African American women is higher than the national average.[4]
Obesity has a direct, negative effect on health and quality of life, and it contributes to a large number of health risks and metabolic and cardiovascular diseases like diabetes, hypertension, and stroke. [5–7][8] In addition, the International Agency for Research on Cancer (IARC) recently updated its list of cancers with sufficient evidence of a link between excess body fat and increased cancer risk to include colon and rectum, gastric cardia, liver, gallbladder, pancreas, kidney, esophageal adenocarcinoma, corpus uteri, ovary, postmenopausal breast, thyroid, meningioma, and multiple myeloma.[9] Racial/ethnic disparities in cardiometabolic diseases and cancer are well-established, with African American populations at greater risk of morbidity and mortality.[10, 11] As a result of the higher prevalence of obesity in African American populations living in the rural south, many are suffering from higher rates of diseases that are linked to obesity. These disparities are driven in part by a confluence of social and environmental factors that support weight gain (e.g., limited access to supermarkets and facilities for physical activity) and lead to lower access to healthcare and health promoting interventions. [12, 13]
The Deep South Network for Cancer Control (DSN) is a 15-year old academic-community partnership established with funding from the National Cancer Institute (NCI) to eliminate cancer health disparities in the Deep South.[14, 15] The cornerstone of this work has been identifying and training an extensive network of lay staff and Community Health Advisors from communities in Alabama and Mississippi.[16] Based on a Community-Based Participatory Research (CBPR) approach [17], this network has been successful in deploying interventions for cancer screening and control [14, 15], particularly among a broad range of rural residents that are typically hard to reach or lack access to traditional healthcare systems.
Because obesity is a major contributing factor to cancer disparities, the investigators of the DSN worked with community partners and volunteers to design a translational research project for weight reduction in African American residents from rural Alabama and Mississippi. Due to recent calls to explore macro-environment interventions for obesity prevention,[18, 19] this study tested the additional benefits of community-level strategies to support healthier weight on weight loss outcomes of African American women living in rural Alabama and Mississippi. The Journey to Better Health (JTBH) project involved a network of paid lay staff and Community Health Advisors as Research Partners (CHARPs) who deliver a translation of the phase I behavioral intervention of the Weight Loss Maintenance (WLM) trial with or without a community strategy intervention to support healthier weight [20]. CHARPs were key to the delivery of the intervention because of their unique knowledge of the community context for participants, stature as a trusted member of the community, and noted success in prior trials of healthy lifestyle interventions.[21, 22] This study reports the impact at 6 months of the lay staff and CHARP delivered translation of the WLM phase I on changes in body weight and key cardiometabolic risk factors, including blood pressure, blood lipids, blood glucose, and waist circumference.
Materials and Methods
Brief Overview
A two-group cluster randomized controlled trial design was utilized to test whether an evidence-based behavioral weight loss program augmented with community strategies (Weight Loss Plus) would produce greater weight loss and improvements in metabolic outcomes than the evidence-based weight loss program alone (Weight Loss Only). The comparison of two interventions rather than an intervention vs. control condition was a direct result of our commitment to a CBPR approach in which members of the target community were active in all aspects of the study including the study design. Feedback from our long-term partners indicated a lack of desire to be randomized to a non-intervention condition given their keen awareness of cancer health disparities among African Americans. As such, for this study, participants were clustered within 8 rural DSN research counties (4 AL; 4 MS) and randomized to either intervention condition. The intervention is a 24-month program. This report presents 6-month study findings which coincide with completion of the intensive weight loss phase. The study was approved by the University of Alabama at Birmingham’s Institutional Review Board for Human Use, and each participant provided written informed consent.
Participants
Overweight and obese (BMI≥25 kg/m2) self-identified black women residing in one of 8 rural DSN counties (4 AL; 4 MS) aged 30 to 70 years were recruited between January 2011 and September 2013. Local DSN paid lay staff and CHARPs assisted with recruitment by identifying potentially eligible women primarily through word-of-mouth, personal contact, or ongoing cancer awareness/outreach activities. Exclusion criteria included the following: (1) pregnancy or plans to become pregnant in the next year, (2) diagnosed major medical or psychological condition known to influence body weight loss (e.g., medicated or poorly controlled diabetes (glucose ≥ 126 mg/dL), uncontrolled hypertension (systolic blood pressure (SBP) > 160 mm Hg or diastolic blood pressure (DBP) > 100 mm Hg, cardiovascular event (e.g., myocardial infarction) in the preceding 12 months, history of gastric bypass or other bariatric surgery), (3) history of psychiatric hospitalization in the past 2 years, (4) history of substance abuse or eating disorder, or (5) any other condition for which a medical professional has suggested diet modification, physical activity, and/or weight reduction would be contraindicated.
Intervention conditions
The 24-month group weight loss program for both conditions was adapted from evidence-based behavioral weight loss interventions that have shown efficacy in large clinical trials including racially/ethnically diverse populations.[20, 23] Based on our prior research on weight-related beliefs and practices among African American women[24–27] and formative work with our rural community partners, we edited the intervention curriculum to reflect recipes, sources of physical activity, and other content that would more likely resonate with the target audience. In addition, we developed or modified participant handouts and materials to include pictures and graphics of African American women and rural community settings.
Group weight loss intervention sessions were delivered across both groups by trained, paid DSN lay staff with assistance from CHARPs. The program focused on weight loss through moderate energy restriction (~500 kcal/d less than usual mean intake) and regular moderate-intensity physical activity (180 minutes per week). During the initial 6-months (intensive weight loss phase), intervention sessions (20 total; 1.5 hours each) were designed to be participant-centered and interactive and include guided physical activity or food demonstration [20]. Group sessions targeted behavioral strategies known to support weight loss including goal-setting, problem-solving, self-monitoring, and personalized plans for diet and physical activity. Weekly group sessions were delivered consecutively for 6 months, followed by a 6-month follow-up period (3 months bi-monthly sessions followed by 3 months of monthly sessions). Key targets in the second part of the intervention included continued tailored reinforcement messages, continued self-monitoring, relapse prevention, social support, and problem solving. After completing group sessions, CHARPs maintained monthly contact with participants via telephone. Follow-up phone calls were 10–20 minutes in duration and included personal guidance and support in maintaining or continuing weight loss based on motivational interviewing (MI) techniques.
Weight loss plus
Four of the eight intervention counties were randomized to receive support for implementing strategies to promote healthy eating and/or physical activity in the local community. A request for applications was sent to key stakeholders in each of the counties to solicit applications for mini-grants to be provided to support local initiatives. Grant applications were reviewed by a team of investigators, staff, and community advisory group members and scored and ranked based on predetermined criteria (agency experience, impact, capacity, and sustainability). UAB investigators and research staff also provided technical support to aide local communities in achieving their goals. Community strategies selected included a community garden, enhancement of a walking trail, incentives for the purchase of fresh produce from the local farmers’ market, and a dance class.
Key Measurements/Clinical Assessments
Clinical assessments were conducted in the local communities where the intervention took place. Consistent with CBPR principles [17], UAB research staff traveled to each of the 8 intervention counties to complete data collection with local paid staff and CHARPs. Participants were weighed (to the nearest 0.1 kg) while wearing light clothing without shoes using a professional digital scale regularly calibrated. Height (to the nearest 0.1 cm) was measured at baseline without shoes using a portable and calibrated stadiometer. BMI was calculated using measured height and weight as weight (kg)/height (m2). Waist circumference was measured using a constant-tension spring-loaded tape device on bare skin at the end of a normal expiration at the natural indentation between the 10th rib and the iliac crest to the nearest 0.5 cm. Dietary intake was conducted by interview using the NCI Automated Self-administered 24-hour Dietary Recall (ASA24). The interviewer-administered approach was used to minimize literacy challenges in our target population. Two diet recalls were collected on site, one on a Saturday and the other on a Monday, to capture the previous day’s food representing weekday and weekend intake, respectively. Total intake, macronutrient composition, nutrient and food group estimates, and a global score for diet quality were computed. Physical activity was measured using the physical activity module of the Behavioral Risk Factor Surveillance Survey. The 7 items in this module assess days per week and total time per day engaged in moderate and vigorous physical activity. After a 5 minute rest period, SBP and DBP were measured twice per assessment by trained staff using a calibrated automatic sphygmomanometer and were averaged. A fasting blood sample was collected by fingerstick to assess glucose and lipids [total cholesterol, HDL cholesterol (HDL-C), triglycerides, and LDL cholesterol (LDL-C)]. The Alere Cholestech LDX™ Analyzer (San Diego, CA) was used.
Statistical Analysis
This was a cluster randomized study in which 8 rural counties were randomized to either a culturally-tailored, evidence-based weight loss only program or the same program plus support for community strategies to promote healthy eating and/or physical activity. A total of 409 participants enrolled from 8 DSN counties with an average of 51 participants in each cluster, which provided 89% power to detect a 1.5 kg difference in weight between two groups assuming an intraclass correlation coefficient (ICC) of .015, with standard deviation of 3 kg and two-sided alpha of .05. A multilevel model was used to estimate the ICC with the method of the generalized linear mixed model [28] Descriptive analysis is presented for all variables of interest by intervention assignment as mean and standard deviation for baseline and 6-month data. A waterfall plot was created to visualize individual weight change from baseline to 6 months. We conducted correlation analyses between participant characteristics and outcome measures using Spearman correlation coefficients. We used generalized linear mixed model with PROC MIXED procedure in SAS, version 9.4 (SAS Institute, Cary, NC) to assess the association between the primary endpoints of BMI/body weight and intervention assignment with covariance structure accounting for variability between clusters (county). The model was adjusted for baseline measurement of weight, demographics and biomedical covariates, including HDL-C, LDL-C, SBP, and DBP. We performed an intent-to-treat analysis for all subjects enrolled in the study. We treated weight as a normal distribution, and missing weight data in 6 month follow-up was imputed using multiple imputation procedures.[29, 30] Our primary comparison was between the two intervention groups at the alpha level of 5%, thus the p-value was not adjusted for the primary comparison of weight difference between two groups at 6 months. In addition, a sensitivity analysis (data not presented here) was conducted on persons who completed the study. The findings remained the same. Results are presented as mean (standard deviation).
Results
Participants
A total of 409 participants were enrolled in the study (Table 1) from 8 counties with comparable demographic characteristics. Weight Loss Only counties (2 per state) had a total of 154 participants. Weight Loss Plus counties (2 per state) had a total of 255 participants at baseline. Figure 1 describes the cluster randomization and 6-month retention. Study participants were on average in their mid-forties. This sample included a high percentage of high school and college graduates, with approximately 92% of participants completing at least high school. Nearly 40% of the entire sample had at least a college education. Twenty-two percent of the intervention group reported household incomes of $40,000 or more, while only 11% of the comparison group reported income in this range. Unemployment rates were slightly higher in the comparison group (11%) compared to those in the intervention counties (7.8%). With the exception of slightly higher educational attainment (92% vs 74% > high school degree), demographic characteristics of our sample mirror that of residents of the counties in which they live.[31] There were no statistical differences between the groups at baseline.
Table 1:
Journey to Better Health study population demographics
| Demographics | Weight Loss Only N=154 |
Weight Loss Plus N=255 |
Total N=409 |
|---|---|---|---|
| Age, Mean ± SD | 44.8 (10.4) | 47.5 (9.5) | 46.5 (9.9) |
| Income, N (%) | |||
| $10,000 or less | 37 (24.0) | 44 (17.3) | 81 (19.8) |
| $10,000 to $19,000 | 33 (21.4) | 58 (22.7) | 91 (22.2) |
| $20,000 to $29,000 | 33 (21.4) | 51 (20.0) | 84 (20.5) |
| $30,000 to $39,000 | 23 (14.9) | 38 (14.9) | 61 (14.9) |
| $40,000 to $49,000 | 8 (5.2) | 26 (10.2) | 34 (8.3) |
| $50,000 or over | 9 (5.8) | 30 (11.8) | 39 (9.5) |
| Don’t know/unsure | 11 (7.1) | 8 (3.1) | 19 (4.6) |
| Education, N (%) | |||
| Don’t know/unsure | 3 (1.9) | 5 (2.0) | 8 (2.0) |
| Less than high school | 8 (5.2) | 15 (5.9) | 23 (5.6) |
| High school graduate/GED | 63 (40.9) | 78 (30.6) | 141 (34.5) |
| Some post high school | 26 (16.9) | 48 (18.8) | 74 (18.1) |
| College graduate or more | 54 (35.1) | 109 (42.7) | 163 (39.9) |
| Marital Status, N (%) | |||
| Married | 47 (30.5) | 115 (45.1) | 162 (39.6) |
| Living with intimate partner | 11 (7.1) | 8 (3.1) | 19 (4.6) |
| Separated | 11 (7.1) | 10 (3.9) | 21 (5.1) |
| Divorced | 18 (11.7) | 36 (14.1) | 54 (13.2) |
| Widowed | 6 (3.9) | 10 (3.9) | 16 (3.9) |
| Single | 59 (38.3) | 74 (29.0) | 133 (32.5) |
| Don’t know/unsure | 2 (1.3) | 1 (0.4) | 3 (0.7) |
| Employment, N (%) | |||
| Employed | 107 (69.5) | 172 (67.5 ) | 279 (68.2) |
| Self-employed | 3 (1.9) | 7 (2.7) | 10 (2.4) |
| Retired | 7 (4.5) | 17 (6.7) | 24 (5.9) |
| Disabled | 12 (7.8) | 18 (7.1) | 30 (7.3) |
| Homemaker | 6 (3.9) | 5 (2.0) | 11 (2.7) |
| Student | 0 (0.0) | 6 (2.4) | 6 (1.5) |
| Unemployed | 17 (11.0) | 20 (7.8) | 37 (9.0) |
| Don’t know/unsure | 2 (1.3) | 10 (3.9) | 12 (2.9) |
Figure 1.
CONSORT diagram
Weight changes
At baseline, the Weight Loss Only and Weight Loss Plus groups had average BMI’s of 38.0 and 39.0 kg/m2, respectively (Table 2). ICC was 0.001 for body weight and 0 for BMI, suggesting independence within the counties within the same treatment groups. Overall, the mean weight loss for the combined sample was 2.4 (4.4) kg. During the 6-month intervention, 57.1% of Weight Loss Only participants had any weight loss (Figure 2) while 60.4% of the Weight Loss Plus participants had any weight loss. There was no significant difference between groups in the proportion of participants who lost any weight. The average weight change for those with any weight loss was 4.3 (3.2) kg for the Weight Loss Only group and 5.1 (4.3) kg for the Weight Loss Plus group, corresponding to an average 4.8 (4.0) kg reduction in initial body weight for both groups.
Table 2:
Changes in risk factors for all participants by treatment group
| Total (N= 409) | Weight Loss Only (N= 154) | Weight Loss Plus (N= 255) | Group Difference | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Risk Factor | Baseline | Follow up |
Changea | Baseline | Follow up | Changea | Baseline | Follow up |
Changea | Follow up difference |
P valueb |
| Weight (kg) | 103.4 (22.7) |
101.0 (22.7) |
−2.4 (4.4)** |
100.2 (21.7) |
98.3 (22.0) | −1.9 (3.9)** | 105.3 (23.1) |
102.6 (23.0) |
−2.7 (4.6)** |
−4.3 (22.7) |
0.4826 |
| BMI (kg/m2) | 38.6 (8.0) |
37.7 (8.1) |
−0.9 (1.6)** |
38.0 (7.7) | 37.3 (7.9) | −0.7 (1.5)** | 39.0 (8.2) |
38.0 (8.2) |
−1.0 (1.7)** |
−0.7 (8.1) | 0.477 |
|
Waist Circumference (cm) |
110.6 (16.3) |
108.4 (15.4) |
−2.2 (7.8)** |
109.4 (14.8) |
108.3 (15.4) |
−1.2 (6.2)* | 111.3 (17.2) |
108.4 (15.4) |
−2.9 (8.5)** |
−0.1 (15.4) |
0.1166 |
| SBP (mmHG) | 125.1 (16.5) |
121.8 (17.1) |
−3.2 (13.3)** |
123.9 (16.0) |
121.9 (17.9) |
−1.9 (11.0)* | 125.8 (16.2) |
121.8 (16.5) |
−4.0 (14.5)** |
0.1 (17.1) | 0.4492 |
| DBP (mmHG) | 79.5 (10.0) |
77.0 (10.8) |
−2.4 (8.9)** |
79 (9.6) | 77.3 (10.5) | −1.7 (8.3)* | 79.7 (10.3) |
76.9 (10.9) |
−2.9 (9.1)** |
0.5 (10.8) | 0.6135 |
|
Total Cholesterol (mg/dL) |
179.5 (34.9) |
177.3 (34.0) |
−2.2 (18.9)* |
179.3 (32.7) |
176.8 (33.0) |
−2.5 (17.7) | 179.7 (36.3) |
177.6 (34.7) |
−2.0 (19.6) |
−0.9 (34.0) |
0.9345 |
| Triglycerides (mg/dL) | 122.1 (91.9) |
104.3 (68.2) |
−17.8 (84.1)** |
112.4 (81.3) |
101.7 (64.5) |
−10.7 (62.0)* |
127.9 (97.3) |
105.9 (70.4) |
−22.1 (95.0)** |
−4.2 (68.2) |
0.8202 |
| Fasting Glucose (mg/dL) | 94.0 (11.9) |
93.9 (15.2) |
−0.1 (13.4) |
93.5 (12.1) | 95.0 (15.7) | 1.5 (13.5) | 94.3 (11.9) |
93.3 (15.0) |
−1 (13.3) | 1.7 (15.2) | 0.1344 |
| HDL | 51.7 (14.3) |
51.8 (13.9) |
0.1 (8.7) | 50.1 (14) | 49.8 (13.1) | −0.4 (8.6) | 52.7 (14.5) |
53.1 (14.2) |
0.4 (8.7) | −3.3 (13.9) |
0.3466 |
| LDL | 105.8 (33.0) |
106.2 (31.4) |
0.4 (19.4) |
108.9 (30.7) |
107.4 (31.2) |
−1.4 (18.0) | 104.0 (34.2) |
105.5 (31.6) |
1.5 (20.1) |
1.9 (31.5) | 0.592 |
| Ratio | 3.7 (1.1) | 3.6 (1.1) |
−0.05 (0.6) |
3.8 (1.2) | 3.7 (1.2) | −0.1 (0.7) | 3.6 (1.1) | 3.6 (1.1) | −0.03 (0.6) |
0.1 (1.1) | 0.881 |
pre-post (baseline and 6 –month) comparison
two-group (Weight loss only and Weight Loss Plus) comparison
p<0.04
p<0.001
Figure 2.
Percent of participants reporting any, 3% or 5% weight loss by treatment group
Because sustained weight loss of 3–5% has been associated with improvements in risk factors for cardiovascular and metabolic disease [8], we assessed the proportion of participants who achieved weight loss in this range. Over forty percent of all participants lost at least 3% of their initial body weight. The average weight loss for participants who lost at least 3% was 6.1 (2.9) kg in the Weight Loss Only group and 7.3 (4.1) kg in the Weight Loss Plus group, with no significant difference between groups. At the higher target of 5% weight loss, 23% of the total sample met this target with no difference between treatment assignments. The average weight loss in this subset was 7.3 (2.8) kg compared to 9.1 (4.2) kg for Weight Loss Only and Weight Loss Plus, respectively.
Risk factor changes
Results of changes in risk factors associated with body weight, waist circumference, blood pressure, blood glucose, cholesterol, and triglycerides are shown in Table 2. ICC for these risk variables ranged from 0.013 to 0.029, indicating limited clustering effect among counties within the same treatment assignment. At 6 months, waist circumference decreased on average for both groups; however, the Weight Loss Plus group had a greater mean reduction of 2.9 (8.5) cm compared to the reduction in waist circumference for the Weight Loss Only group of 1.2 (6.2) cm (p<0.001). Both the Weight Loss Only and Weight Loss Plus groups had significant reductions in blood pressure. For the Weight Loss Only group, there was a mean reduction in SBP (1.9 mm Hg; p=0.03) and DBP (1.7 mm Hg; p = 0.01). For the Weight Loss Plus group, a mean reduction in SBP of 4.0 mm Hg (p<.001) and DBP of 2.9 mmHg (p<0.001) were noted. Neither group experienced statistically significant reductions in blood glucose, total cholesterol, HDL-C, LDL-C, or ratio by the 6-month follow up. On the other hand, we observed significant reductions in serum triglycerides of 10.7 (62.0) mg/dL for the Weight Loss Only group (p = .03) and 22.1 (95.0) mg/dL for the Weight Loss Plus group (p<0.001).
Process measures-Attendance
The primary measure of intervention engagement was attendance at the 20 group behavioral sessions. The average number of sessions attended by those in the Weight Loss Plus group was slightly higher (12.0 sessions) and significantly different (p=0.046) from the Weight Loss Only group (10.6 sessions) (Table 3). There was a significant correlation coefficient of 0.36 (p<0.001) between weight loss at 6 months and attendance in the overall sample; the Pearson correlation coefficients were 0.24 (p=0.006) for the Weight Loss Only group and 0.46 (p<0.001) for the Weight Loss Plus group. Consistent with this correlation, those who achieved any weight loss attended 4.4 more sessions than those who did not. Those who lost at least 3% or 5% of their initial weight attended an average of approximately 5 more sessions than those who did not reach these weight change benchmarks. Using multivariate analysis to control for demographic and biomedical covariates, we found that attendance was significantly associated with 6-month weight and weight change.
Table 3.
Correlation between Weight Loss and Attendance
| Weight Loss Only |
Weight Loss Plus |
Total | Between Groups |
|
|---|---|---|---|---|
|
Correlation between Weight loss and Attendance R (p-value) |
0.27 (0.006) | 0.46 (<0.001) | 0.36 (<0.001) | NA |
|
Overall attendance Number who attended at least one session (%) Mean number of sessions attended (STD) |
134 (87.0%) 10.59 (6.7) |
215 (84.3%) 12.01 (6.2) |
349 (85.3) 11.46 (6.47) |
0.0459 |
| Sessions attended | ||||
| No weight loss Mean (STD) |
8.21 (6.04) |
8.61 (5.87) |
8.44 (5.91) |
NS |
| Any weight loss Mean (STD) |
11.82 (6.80) |
13.34 (5.87) |
12.80 (6.28) |
NS |
| ≥3%Weight Loss Mean (STD) <3%Weight Loss Mean (STD) |
13.95 (8.01) 8.30 (6.40) |
15.34 (4.90) 9.98 (5.86) |
14.83 (5.36) 9.36 (6.70) |
0.0502 |
| ≥5% Weight Loss Mean (STD) <5% Weight Loss Mean (STD) |
14.91 (5.38) 9.69 (6.87) |
16.47 (4.23) 11.37 (5.92) |
15.89 (4.73) 10.77 (6.32) |
NS |
Pearson Correlation Coefficient
Discussion and Conclusion
In this study of African American women from rural areas of Alabama and Mississippi, our first objective was to assess whether a network of paid lay staff and CHARPs could disseminate an evidence-based behavioral weight loss intervention. The CHARPs successfully recruited a sample of 409 African American women from eight Southern counties in Alabama and Mississippi to participate in the intervention based on the phase I intervention of the WLM trial. They also achieved strong engagement as measured by attendance. Among all participants, over thirty-six percent lost at least 3% of their initial body weight, and these reductions in body weight were associated with significant reductions in blood pressure, serum triglycerides, and waist circumference. These results signify successful translation of the evidence-based intervention by trained lay interventionists into a high risk population that has traditionally been difficult to reach and engage in weight loss interventions.
From a comparative standpoint, the observed results of this first phase are some of the strongest translational outcomes seen to date for African Americans. Even when comparing this effectiveness study to major efficacy clinical trials of behavioral weight loss interventions conducted under ideal conditions and designed to modify risk factors, including the Diabetes Prevention Program (DPP) and PREMIER, the mean weight losses for African American women are within a similar range. [32, 33] For example, African American women in the DPP who received the intensive lifestyle intervention lost 4.7 kg at 6 months. [33] In the PREMIER trial, African American women in the Established and Established + DASH treatment groups lost 3.2 kg at 6 months. [32] The intervention upon which JTBH was based, the phase I WLM trial, reported mean weight loss of 4.1 kg at 20 weeks for African American women. [34] When comparing this study to other effectiveness trials for which one might expect diminished results due to less than ideal intervention conditions, our results are consistent with other trials led by both health professionals as well as individuals who were not trained as health professionals or psychologists. For example, a recent review of evidence-based behavioral weight loss translational studies with an exclusive African American population, noted an average of 2.9 kg weight loss at 6-months. [35] Notably, only one of the seven studies in this category [36] involved a community setting using interventionists who were not health professionals or persons with backgrounds in psychology. That study saw weight loss of 2.3 kg or 2.7% of baseline weight.
Previous reports of behavioral interventions targeting African American women have demonstrated a range of results. Those based in community settings have reported weight losses of 2.0–4.1 kg at 6 months on average. [37] The majority of these studies have utilized well-trained professionals to deliver the interventions, typically with ongoing support from academic experts. JTBH is one of the first examples of the use of CBPR methods to engage the target audience and then use lay health staff from the community to deliver the translated intervention. The JTBH interventionists were not health professionals but did complete an extensive training program to learn the key counseling skills and knowledge required to facilitate the behavioral group sessions. Using this approach, the amount of weight loss achieved in JTBH was similar to weight losses observed in other studies that utilized lay heath staff. For example, a 2011 pilot study that utilized lay health staff to implement a faith-based intervention for a rural black population reported 2.3 kg of weight loss at 16 weeks.[36] A larger randomized control trial testing the effectiveness of utilizing lay health staff to promote weight loss among a mostly white sample of older individuals at senior centers achieved weight loss of 3.7 kg 4 months after completing a 12-week intervention.[38] Faucher et al[39] also reported similar results (−2.9 kg) at 20 weeks when using promotoras de salud to deliver a behavioral weight loss intervention to a group of Latinas. The lay staff and CHARPs were also effective at engaging participants with good levels of session attendance (57%) and study retention (99.5%) through 6 months. Session attendance in JTBH is consistent with several DPP-translation studies in community settings that ranged in duration from 16–26 weeks and reported attendance ranging from roughly 50–70% of sessions. However, the study retention for JTBH exceeds that of these other studies which reported attrition rates from 7–41%.[35]
Because we did not design this study to isolate predictors of effectiveness, we can only speculate as to drivers of the overall outcomes observed. Studies of behavioral interventions for weight reduction targeted to African Americans have generally included a range of cultural adaptations and enhancements that might better engage the audience through increased salience and relevance. In general, these adaptations have not produced greater weight losses for African Americans. When specific strategies have been isolated in experimental designs, such as including extended family members [40], all-African American groups [41], or race-concordant interventionists [42], the effects of these strategies have been modest. The use of lay local interventionists may have provided a level of inherent cultural specificity that cannot be gained through didactic sessions or training workshops. The cultural specificity is not only related to ethnic and/or cultural identity but also regional identity as a rural Southerner. These innate characteristics may have been critical to engaging these participants, and in turn, the ability of CHARPs to engage their community neighbors may have been the key to driving the weight and biomarker changes observed with this intervention. We believe it is reasonable to conclude that engagement is a primary driver of the observed results because a large body of research has shown that engagement in behavioral lifestyle modification interventions through attendance of group and individual counseling sessions is highly predictive of weight loss outcomes. [34, 43, 44] In addition to inherent cultural specificity and participant engagement, the success of the lay health staff may be also attributable to the notion that people are more likely to relate to messages and initiate change if they can connect to the messenger through perceived similarities.[45, 46]
All studies are limited by a function of design and context. The generalizability of this study is limited to higher educated African American women in a rural community setting in the southern U.S. However, given noted racial/ethnic and geographic health disparities, this demographic represents an important target group in the effort to alter the impact of the obesity epidemic in the U.S. and address many of the health disparities that plague minority communities. Another limitation of this specific report is the length of follow up. The 6-month outcomes are of interest as they demonstrate the effectiveness of delivering the intervention using lay staff and CHARPs. We recognize though that longer term outcomes beyond one year of follow up will be of primary interest as maintenance of weight loss is also a critical challenge for many people.
In conclusion, this study demonstrates that trained lay staff and volunteers were able to effectively deliver a translation of an evidence-based behavioral weight loss intervention to African American women who reside in rural Southern communities. The weight loss achieved was associated with clinically meaningful changes in risk factors. Further work is needed to understand the longer term effectiveness in this setting, but the early results are clearly promising. These results are comparable to many intensive behavioral interventions performed at some of the best academic medical centers in the U.S. If this model can be further validated, it may be a primary tool to provide weight loss interventions for groups of people at high risk of obesity related cardiovascular disease and cancer outcomes who have limited access to any other qualified providers in low resource, rural settings.
Figure 3:
Waterfall plot of individual participant change in weight (kg) by treatment group
Acknowledgements
We gratefully acknowledge and appreciate the support provided by all of the Deep South Network for Cancer Control staff. We especially thank the community health advisors trained as research partners and the study participants who helped make this research possible.
Source of Funding
This study was made possible by grant number U54CA153719 from the Center to Reduce Cancer Health Disparities (CRCHD) of the National Cancer Institute. The contents of this paper are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health (NIH).
Footnotes
Conflict of Interest Statement
There are no conflicts of interest among any of the authors.
References
- 1.Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL: Trends in obesity among adults in the united states, 2005 to 2014. JAMA 2016, 315(21):2284–2291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ogden CL, Carroll MD, Kit BK, Flegal KM: Prevalence of childhood and adult obesity in the united states, 2011–2012. JAMA 2014, 311(8):806–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Jackson JE, Doescher MP, Jerant AF, Hart LG: A National Study of Obesity Prevalence and Trends by Type of Rural County. The Journal of Rural Health 2005, 21(2):140–148. [DOI] [PubMed] [Google Scholar]
- 4.Centers for Disease Control and Prevention: Behavioral Risk Factor Surveillance System Survey Data Edited by US Department of Health and Human Services; 2007. [Google Scholar]
- 5.Harris MI, Flegal KM, Cowie CC, Eberhardt MS, Goldstein DE, Little RR, Wiedmeyer HM, Byrd-Holt DD: Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. The Third National Health and Nutrition Examination Survey, 1988–1994. Diabetes Care 1998, 21(4):518–524. [DOI] [PubMed] [Google Scholar]
- 6.Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ: Cancer Statistics, 2008. CA Cancer J Clin 2008, 58(2):71–96. [DOI] [PubMed] [Google Scholar]
- 7.Mensah GA, Mokdad AH, Ford ES, Greenlund KJ, Croft JB: State of Disparities in Cardiovascular Health in the United States. Circulation 2005, 111(10):1233–1241. [DOI] [PubMed] [Google Scholar]
- 8.Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG, Donato KA, Hu FB, Hubbard VS, Jakicic JM, Kushner RF et al. : 2013 AHA/ACC/TOS Guideline for the Management of Overweight and Obesity in Adults. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and The Obesity Society 2014, 129(25 suppl 2):S102–S138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lauby-Secretan B, Scoccianti C, Loomis D, Grosse Y, Bianchini F, Straif K: Body Fatness and Cancer — Viewpoint of the IARC Working Group. New England Journal of Medicine 2016, 375(8):794–798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Garth G: Disparities in Cardiovascular Disease Risk in the United States. Current Cardiology Reviews 2015, 11(3):238–245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.O’Keefe EB, Meltzer JP, Bethea TN: Health Disparities and Cancer: Racial Disparities in Cancer Mortality in the United States, 2000–2010. Frontiers in Public Health 2015, 3(51). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liese AD, Weis KE, Pluto D, Smith E, Lawson A: Food Store Types, Availability, and Cost of Foods in a Rural Environment. Journal of the American Dietetic Association 2007, 107(11):1916–1923. [DOI] [PubMed] [Google Scholar]
- 13.Wilcox S, Castro C, King AC, Housemann R, Brownson RC: Determinants of leisure time physical activity in rural compared with urban older and ethnically diverse women in the United States. J Epidemiol Community Health 2000, 54(9):667–672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lisovicz N, Johnson RE, Higginbotham J, Downey JA, Hardy CM, Fouad MN, Hinton AW, Partridge EE: The Deep South Network for cancer control. Cancer 2006, 107(S8):1971–1979. [DOI] [PubMed] [Google Scholar]
- 15.Partridge EE, Fouad MN, Hinton AW, Hardy CM, Liscovicz N, White-Johnson F, Higginbotham JC: The Deep South Network for Cancer Control: Eliminating Cancer Disparities Through Community–Academic Collaboration. Family Community Health 2005, 28(1):6–19. [DOI] [PubMed] [Google Scholar]
- 16.Hardy CM, Wynn TA, Huckaby F, Lisovicz N, White-Johnson F: African American community health advisors trained as research partners: recruitment and training. Family & Community Health 2005, 28(1):28–40. [DOI] [PubMed] [Google Scholar]
- 17.Israel BA, Schulz AJ, Parker EA, Becker AB: Review of community-based research: Assessing partnership approaches to improve public health. Annual Reviews of Public Health 1998, 19:173–202. [DOI] [PubMed] [Google Scholar]
- 18.Calancie L, Leeman J, Jilcott Pitts SB, Khan LK, Fleischhacker S, Evenson KR: Nutrition-Related Policy and Environmental Strategies to Prevent Obesity in Rural Communities: A Systematic Review of the Literature, 2002–2013. Preventing Chronic Disease 2015, 12:E57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kirk S, Penney T, McHugh T-LF: Characterizing the obesogenic environment: the state of the evidence with directions for future research. Obesity Reviews 2010, 11(2):109–117. [DOI] [PubMed] [Google Scholar]
- 20.Brantley P, Appel L, Hollis J, Stevens V, Ard J, Champagne C, Elmer P, Harsha D, Myers V, Proschan M et al. : Design considerations and rationale of a multi-center trial to sustain weight loss: the weight loss maintenance trial. Clinical Trials 2008, 5(5):546–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cummings DM, Lutes LD, Littlewood K, DiNatale E, Hambidge B, Schulman K: EMPOWER: A randomized trial using community health workers to deliver a lifestyle intervention program in African American women with Type 2 diabetes: Design, rationale, and baseline characteristics. Contemporary Clinical Trials 2013, 36(1):147–153. [DOI] [PubMed] [Google Scholar]
- 22.Nelson D, Harris A, Horners-Ibler B, Harris K, Burns E: Hearing the Community: Evolution of a Nutrition and Physical Activity Program for African American Women to Improve Weight. Journal of Health Care for the Poor and Underserved 2016, 27(2):560–567. [DOI] [PubMed] [Google Scholar]
- 23.The Diabetes Prevention Program Research Group: The Diabetes Prevention Program (DPP). Description of lifestyle intervention 2002, 25(12):2165–2171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ard JD, Zunker C, Qu H, Cox T, Wingo B, Jefferson W, Shewchuk R: Cultural Perceptions of Weight in African American and Caucasian Women. American Journal of Health Behavior 2013, 37(1):3–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Jefferson WK, Zunker C, Feucht JC, Fitzpatrick SL, Greene LF, Shewchuk RM, Baskin ML, Walton NW, Phillips B, Ard JD: Use of the Nominal Group Technique (NGT) to understand the perceptions of the healthiness of foods associated with African Americans. Evaluation and Program Planning 2010, 33(4):343–348. [DOI] [PubMed] [Google Scholar]
- 26.Malpede CZ, Faulk LE, Fitzpatrick SL, Jefferson WK, Shewchuk RM, Baskin ML, Ard JD: Racial influences associated with weight related beliefs in African American and Caucasian women. Ethnicity and Disease 2007, 17(1):1–5. [PubMed] [Google Scholar]
- 27.Pekmezi D, Marcus B, Meneses K, Baskin ML, Ard JD, Martin MY, Adams N, Robinson C, Demark-Wahnefried W: Developing an intervention to address physical activity barriers for African–American women in the deep south (USA). Women’s Health 2013, 9(3):301–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hox J: Multilevel analysis: Techniques and applications Mahwah, NJ: Lawrence Erlbaum; 2010. [Google Scholar]
- 29.Collins LM, Schafer JL, Kam C-M: A comparison of inclusive and restrictive strategies in modern missing data procedures. Psychological Methods 2001, 6(4):330–351. [PubMed] [Google Scholar]
- 30.Little R, Rubin D (eds.): Statistical Analysis with Missing Data, 2nd ed.: Wiley: Hoboken, NJ; 2002. [Google Scholar]
- 31.US Census Bureau: State and County Quick Facts. Population Estimates. In.: Available at: http://quickfacts.census.gov. Accessed February 21, 2017; 2013.
- 32.Svetkey LP, Erlinger TP, Vollmer WM, Feldstein A, Cooper LS, Appel LJ, Ard JD, Elmer PJ, Harsha D, Stevens VJ: Effect of lifestyle modifications on blood pressure by race, sex, hypertension status, and age. J Hum Hypertens 2005, 19(1):21–31. [DOI] [PubMed] [Google Scholar]
- 33.West DS, Elaine Prewitt T, Bursac Z, Felix HC: Weight Loss of Black, White, and Hispanic Men and Women in the Diabetes Prevention Program. Obesity 2008, 16(6):1413. [DOI] [PubMed] [Google Scholar]
- 34.Hollis JF, Gullion CM, Stevens VJ, Brantley PJ, Appel LJ, Ard JD, Champagne CM, Dalcin A, Erlinger TP, Funk K et al. : Weight Loss During the Intensive Intervention Phase of the Weight-Loss Maintenance Trial. American journal of preventive medicine 2008, 35(2):118–126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Samuel-Hodge CD, Johnson CM, Braxton DF, Lackey M: Effectiveness of Diabetes Prevention Program translations among African Americans. Obesity Reviews 2014, 15:107–124. [DOI] [PubMed] [Google Scholar]
- 36.Yeary KH-cK, Cornell CE, Moore P, Bursac Z, Prewitt TE, West DS, Turner J: Feasibility of an Evidence-Based Weight Loss Intervention for a Faith-Based, Rural, African American Population. Preventing Chronic Disease 2011, 8(6):A146. [PMC free article] [PubMed] [Google Scholar]
- 37.Fitzgibbon ML, Tussing-Humphreys LM, Porter JS, Martin IK, Odoms-Young A, Sharp LK: Weight loss and African–American women: a systematic review of the behavioural weight loss intervention literature. Obesity Reviews 2012, 13(3):193–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.West DS, Bursac Z, Cornell CE, Felix HC, Fausett JK, Krukowski RA, Lensing S, Love SJ, Prewitt TE, Beck C: Lay Health Educators Translate a Weight-Loss Intervention in Senior Centers: A Randomized Controlled Trial. American journal of preventive medicine 2011, 41(4):385–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Faucher MA, Mobley J: A Community Intervention on Portion Control Aimed at Weight Loss in Low-Income Mexican American Women. Journal of Midwifery & Women’s Health 2010, 55(1):60–64. [DOI] [PubMed] [Google Scholar]
- 40.Kumanyika SK, Wadden TA, Shults J, Fassbender JE, Brown SD, Bowman MA, Brake V, West W, Frazier J, Whitt-Glover MC et al. : Trial of Family and Friend Support for Weight Loss in African American Adults. Arch Intern Med 2009, 169(19):1795–1804. [DOI] [PubMed] [Google Scholar]
- 41.Ard JD, Kumanyika S, Stevens VJ, Vollmer WM, Samuel-Hodge C, Kennedy B, Gayles D, Appel LJ, Brantley PJ, Champagne C et al. : Effect of Group Racial Composition on Weight Loss in African Americans. Obesity 2008, 16(2):306–310. [DOI] [PubMed] [Google Scholar]
- 42.Batch BC, Ard JD, Vollmer WM, Funk K, Appel LJ, Stevens VJ, Samuel-Hodge C, Loria CM, Hollis JF, Svetkey LP: Impact of participant and interventionist race concordance on weight loss outcomes. Obesity 2013, 21(4):712–717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Fitzpatrick SL, Bandeen-Roche K, Stevens VJ, Coughlin JW, Rubin RR, Brantley PJ, Funk KL, Svetkey LP, Jerome GJ, Dalcin A et al. : Examining behavioral processes through which lifestyle interventions promote weight loss: Results from PREMIER. Obesity 2014, 22(4):1002–1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Orth WS, Madan AK, Taddeucci RJ, Coday M, Tichansky DS: Support Group Meeting Attendance is Associated with Better Weight Loss. Obesity Surgery 2008, 18(4):391–394. [DOI] [PubMed] [Google Scholar]
- 45.Artinian NT, Fletcher GF, Mozaffarian D, Kris-Etherton P, Van Horn L, Lichtenstein AH, Kumanyika S, Kraus WE, Fleg JL, Redeker NS et al. : Interventions to Promote Physical Activity and Dietary Lifestyle Changes for Cardiovascular Risk Factor Reduction in Adults. A Scientific Statement From the American Heart Association 2010, 122(4):406–441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Sloane B, Zimmer C: The power of peer health education. J Am Coll Health 1993, 41(6):241–245. [DOI] [PubMed] [Google Scholar]