If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address for correspondence: Michelle C. Kegler, DrPH, Department of Behavioral Sciences and Health Education, Emory Prevention Research Center, Rollins School of Public Health, Emory University, 1518 Clifton Rd NE, Atlanta, GA 30322; Phone: (404) 712-9957; Fax: (404) 727-1369
Department of Behavioral Sciences and Health Education, Emory Prevention Research Center, Rollins School of Public Health, Emory University, Atlanta, GA
Department of Behavioral Sciences and Health Education, Emory Prevention Research Center, Rollins School of Public Health, Emory University, Atlanta, GA
To describe home food environments and examine which aspects are associated with fruit and vegetable intake and percent calories from fat among overweight and obese women.
Design
Baseline data from a weight gain prevention trial collected through telephone interviews.
Setting
Participants were recruited from 3 federally qualified health centers in rural Georgia.
Participants
Overweight and obese patients (n = 319) were referred by their providers if they had a body mass index (BMI) > 25 and lived with at least 1 other person. Participants were primarily African American (83.7%), with a mean BMI of 38.4.
Main Outcome Measures
Fruit and vegetable intake and percent calories from fat.
Analysis
Descriptive statistics and multiple regression.
Results
Participants reported a large variety of both fruits and vegetables and unhealthy foods in their homes, and an average of 2.6 family meals from non-home sources per week. Eating family meals with the television on was common. Availability of fruits and vegetables in the home (P < .001) and frequency of fruit shopping (P = .01) were associated with fruit and vegetable intake. The number of unhealthy foods in the home (P = .01) and food preparation methods (P = .01) were associated with percent calories from fat.
Conclusions and Implications
Home food environments may be effective intervention targets for nutrition programs designed for overweight and obese women.
By providing both social and physical cues for eating, exercise, and sedentary activity, the home can trigger behaviors both positively and negatively associated with weight status. Until recently, much of the research on home environments was focused on children because of their dependence on caregivers for eating and activity opportunities.
found that overweight adults had fewer low-fat snacks and fruit and vegetables in the home and more high-fat snacks than did normal weight adults.
The quality of foods available in the home is largely influenced by grocery shopping behavior, use of non-home sources for family meals, and food preparation methods. Studies have shown that households with overweight individuals purchased foods higher in fat and calories than those with normal weight adults
Socio-economic, behavioural and environmental factors predicted body weights and household food insecurity scores in the Early Childhood Longitudinal Study-Kindergarten.
The frequency with which family meals are obtained from non-home sources, such as fast food, takeout, and delivery, also has implications for weight management and healthfulness of the home food environment.
Away-from-home family dinner sources and associations with weight status, body composition, and related biomarkers of chronic disease among adolescents and their parents.
One study found that parents who reported at least 1 family dinner per week from 1 of these sources were significantly more likely to be overweight or obese than those who did not report these food sources.
Away-from-home family dinner sources and associations with weight status, body composition, and related biomarkers of chronic disease among adolescents and their parents.
Differential associations of fast food and restaurant food consumption with 3-y change in body mass index: the Coronary Artery Risk Development in Young Adults Study.
Eating at fast-food restaurants is associated with dietary intake, demographic, psychosocial and behavioural factors among African Americans in North Carolina.
unless accompanied by healthy cooking methods. In the same study, healthier food preparation was associated with reduced adolescent obesity. In a sample of mid-life women, spending more time on home meal preparation resulted in meals with greater caloric content.
Family norms around eating with the television on are another dimension of the home food environment. Research suggests that television may contribute to obesity through 2 mechanisms: increased food intake and decreased physical activity. Studies have documented that television watching is associated with increased consumption of energy-dense foods and decreased fruit and vegetable intake for both adolescents and adults.
Huffman FG, Vaccaro JA, Exebio JC, Zarini GG, Katz T, Dixon Z. Television watching, diet quality, and physical activity and diabetes among three ethnicities in the United States [published online ahead of print July 17, 2012]. J Environ Public Health. doi:10.1155/2012/191465.
Finally, family support for healthy eating, or lack thereof, is associated with dietary behavior. Although findings have been inconsistent, family social support, defined as encouragement or sabotage from household members, has been shown to predict dietary behaviors.
Sociodemographic and psychological predictors of changes in dietary fat consumption in adults with high blood cholesterol following counseling in primary care.
The purposes of the current article were to describe the home food environment and examine which aspects are associated with healthy eating in overweight and obese women. The study was guided by a conceptual model based on a social ecologic perspective (Figure).
The guiding hypothesis was that food inventories, food placement, grocery shopping, food preparation, meal serving practices, family meals from non-home sources, television watching while eating, and family support for healthy eating would be associated with dietary behavior, operationalized as percent calories from fat and fruit and vegetable intake.
Data were from the Healthy Homes/Healthy Families study, a randomized, controlled trial testing the effectiveness of a coach-based intervention to prevent weight gain over time. Community input into study design was obtained through a community advisory board. Study participants were asked to complete baseline and 6- and 12-month follow-up telephone interviews. Data reported here are from baseline only. The study was approved by the Emory Institutional Review Board and verbal consent was obtained from all participants.
Study Population
Recruitment for the trial was done in partnership with 3 federally qualified health centers that serve low-income residents in southwest Georgia, with providers from 9 clinic sites referring overweight and obese women to the study. These federally qualified health centers are represented on the community advisory board. Participants were age 35–65 years, lived with at least 1 other person, and lived within 30 miles of the referring clinic. Recruitment occurred from February, 2011 to December, 2012, and baseline data were collected across all seasons. Of the 510 women enrolled, 355 (69.6%) completed baseline data collection, which involved 3 telephone interviews. Each interview lasted approximately 45 minutes, and participants received a $40 gift card for completion of baseline data survey. Interviews were conducted by trained research staff and Master of Public Health–level graduate research assistants.
Measures
Most of the measures were used in a pilot test of the intervention and exhibited adequate reliability in a similar population from the same region.
Survey instruments for the current study were pilot-tested (n = 16) and refined based on feedback.
Food inventory
Participants were asked about the presence of 13 fresh or frozen fruits, 17 fresh or frozen vegetables, 8 unhealthy foods, and 3 unhealthy drinks in the home over the past week.
A comprehensive list of items was developed from other measures, and then items that were common from earlier studies in the same region were added (eg, okra, greens and squash), and items that were ambigous in terms of their nutritional value (eg, crackers) were eliminated.
The inventory scores were calculated by summing the number of items present in the household. Prior studies have documented test-tetest reliability for individual food items ranging from 0.63 to 0.90.
Participants were asked 3 questions about household placement of fruits, vegetables, and unhealthy snacks in their home. In yes/no format, participants were asked whether they kept these food items in the home where they were easy to see and easy to reach.
The food placement score was calculated by reverse-coding the snack item and then summing all 3 items such that a score of 3 meant that fruit and vegetables were kept in a place where they were easy to see and easy to reach, and snacks were not. A score of 1 meant the converse. Test-retest reliability ranged from 0.50 to 0.60 for items in the original measure.
Participants were asked how often they or the primary food shopper bought fresh vegetables and fresh fruit when they went grocery shopping in the past month. For the regression model, the authors calculated the maximum shopping frequency for the 2 items.
Food preparation and meal serving practices
Food preparation was assessed through 15 items that asked how often in the past month participants served healthier food options or prepared foods using healthy cooking methods.
Differences in fat-related dietary patterns between black, Hispanic and White women: results from the Women's Health Trial Feasibility Study in Minority Populations.
Differences in fat-related dietary patterns between black, Hispanic and White women: results from the Women's Health Trial Feasibility Study in Minority Populations.
; it was .64 in the current study. Meal-serving practices focused on strategies with potential to decrease portion size, including using smaller plates, serving smaller amounts of food, and not serving food family style. Both scores are the means of the items.
Non-home food source
Participants were asked to assess the number of days in the past week they purchased family meals from a fast-food restaurant, full-service or sit-down restaurant, takeout, and delivery.
and 2 new items. Participants were asked how often in the past month others in their household encouraged them to avoid unhealthy foods, reminded them to eat fruits and vegetables, and ate high-fat or high-salt foods in front of them, among other behaviors. The score was calculated as the mean of the items after reverse-coding negative items. Cronbach alpha was .70. Prior studies have documented Cronbach alpha ranging from .45 to .52.
Usual fruit and vegetable intake was assessed using 6 items from the 2005 Behavioral Risk Factor Surveillance System and scored according to the Behavioral Risk Factor Surveillance System guidelines.
The National Cancer Institute Quick Food Scan (Fat Screener) was used to assess percentage of calories from fat based on eating habits over the past year.
For each participant, a percentage of daily calories from fat was calculated using the National Cancer Institute–provided scoring guidelines. This item has adequate validity, with correlations ranging from r = 0.36 to r = 0.77 compared with 24-hour dietary recall data.
Participants were asked questions regarding their age, gender, race/ethnicity, marital status, annual household income, education, employment status, number of adults and children in the home, and neighborhood type (eg, in town or a house in the country with few neighbors close by).
Participants were also asked their height and weight for calculation of BMI.
Data Analysis
Normality of outcome variables was assessed. Both fruit and vegetable intake and percentage of fat intake were distributed normally. Univariate and bivariate descriptives for all variables were investigated, as well as multicollinearity. The researchers conducted multivariate regression models with fruit and vegetable intake as well as percentage of calories from fat intake as continuous outcomes. Variable inclusion was based on the conceptual model. Demographic variables were included as control variables because of their potential associations with the dependent variables of interest. Data were analyzed using SAS 9.3 (SAS Institute, Cary, NC, 2011).
Results
Description of Study Participants
The vast majority of participants were African American (83.7%), almost half the sample was unemployed, and two thirds reported an annual household income < $25,000 (Table 1). More than half of the women were not married (59.6%). Household sizes varied, but most women lived with 1 other adult (52.8%) and at least 1 child (50.3%). The average BMI was 38.4 (SD, 8.63). The women consumed an average of 2.8 servings of fruits and vegetables per day (SD, 1.81) and 33.1% of their caloric intake came from fat (SD, 4.22).
Table 1Demographic and Nutrition Behaviors of Study Participants (n = 319)
n
%
Age, y
35–44
93
29.2
45–54
122
38.2
55–65
104
32.6
Race
White
49
15.4
African American/black
267
83.7
Other
3
0.9
Employment
Unemployed
158
49.5
Retired
21
6.6
Part-time
32
10.0
Full-time
108
33.9
Income
≤ $10,000
105
32.9
$10,001–$25,000
108
33.9
$25,001–$50,000
74
23.2
> $50,000
26
8.2
Marital status
Married
129
40.4
Not married, but living with partner
23
7.2
Widowed
15
4.7
Separated
32
10.0
Divorced
48
15.1
Not married
72
22.6
Neighborhood type
In town
160
50.2
In country or rural area, with neighbors close by
136
42.6
House in country with few neighbors close by
23
7.2
Additional adults in household
None
33
10.4
1 person
168
52.8
2 people
79
24.8
3 people
28
8.8
≥ 4 people
10
3.1
Children in household
None
158
49.7
1 child
85
26.7
2 children
48
15.1
3 children
16
5.0
≥ 4 children
11
3.5
Body mass index and eating behaviors (mean [SD])
Body mass index
38.4
8.63
Fruit and vegetable intake (servings/d)
2.8
1.81
Fat intake (% calories from fat)
33.1
4.22
Note: Percentages might not add up to 100% owing to rounding or refusal to answer.
When asked about household food inventories within the past week, women reported an average of 14 types of fruits and vegetables in their home (SD, 5.00) (Table 2). Most commonly, people had bananas (64.9%), oranges or tangerines (55.8%), and apples (53.3%) in their fruit inventories. For vegetables, onions (90.9%) and peas, beans, or green beans (89.0%) were mentioned most often. Participants reported 4.6 unhealthy foods in the home during the past week (SD, 1.77) on average and 1.8 unhealthy beverage items (SD, 0.95). Unhealthy drinks most often reported were sugared drinks other than soft drinks such as sweet tea (68.3%), followed by regular soft drinks (64.3%). A large majority reported keeping fruits (92.2%) as well vegetables (88.7%) in a place where they were easy to see and easy to reach. Almost two thirds of women reported having high-calorie snack foods in visible locations.
Table 2Home Food Inventories and Food Placement in Study Participant Households (n = 319)
Fruit and Vegetable Inventory
Fruit and vegetables score (out of 32)
14.0
(5.00)
Fruit, n (%)
Bananas
207
(64.9)
Oranges or tangerines
178
(55.8)
Apples
170
(53.3)
Grapes
157
(49.2)
Peaches
126
(39.5)
Strawberries
119
(37.3)
Pineapples
80
(25.1)
Grapefruit
73
(22.9)
Cantaloupes or melons
64
(20.1)
Watermelons
60
(18.8)
Pears
54
(16.9)
Blueberries
49
(15.4)
Mangos
32
(10.1)
Vegetables, n (%)
Onions
290
(90.9)
Peas, beans, or green beans
284
(89.0)
Potatoes, not french fries
251
(78.7)
Corn
243
(76.2)
Greens
237
(74.3)
Lettuce
235
(73.7)
Tomatoes
233
(73.0)
Bell peppers
188
(58.9)
Carrots
174
(54.6)
Broccoli
161
(50.5)
Cabbage
155
(49.0)
Okra
155
(48.6)
Celery
138
(43.3)
Cucumbers
135
(42.3)
Squash
87
(27.3)
Cauliflower
50
(15.7)
Asparagus
14
(4.4)
Unhealthy foods inventory, mean (SD)
Unhealthy foods score (out of 8)
4.6
(1.77)
Unhealthy foods, n (%)
Bacon/sausage
277
(86.8)
Cookies, cakes
248
(77.7)
Hot dogs
210
(65.8)
Candy bars
204
(64.0)
Regular potato chips
197
(61.8)
Ice cream
152
(47.7)
Salted nuts
115
(36.1)
Pork rinds
63
(19.8)
Unhealthy drinks inventory, mean (SD)
Unhealthy drinks score (out of 3)
1.8
(0.95)
Unhealthy drinks, n (%)
Sugared drinks other than soft drinks
218
(68.3)
Regular soft drinks
205
(64.3)
Regular whole milk
157
(49.2)
Food placement, mean (SD)
Food placement score (out of 3)
2.2
(0.63)
Items placed easy to see/easy to reach, n (%)
Fruit
294
(92.2)
Vegetables
283
(88.7)
Unhealthy snacks
200
(62.7)
Note: Food placement refers to whether fruits, vegetables, and unhealthy snacks were kept where they were easy to see and easy to reach.
Table 3 lists household food practices. Participants reported grocery shopping for fruits slightly more frequently (mean, 1.0; SD, 1.14) than for vegetables (mean, 0.9; SD, 1.06). The healthy food preparation score was on average 2.2 (SD, 0.45), indicating that they used healthy food preparation methods occasionally; the meal serving practices score was on average 2.5 (SD, 0.65) which meant that they used healthy meal serving practices midway between occasionally and often. Participants reported serving family meals from non-home sources 2.6 days per week on average, with large variation in behavior (SD, 3.14). Most often, those meals came from fast-food restaurants (mean, 1.3; SD, 1.58) or takeout (mean, 0.7; SD, 1.25). When asked how regularly their families ate in front of the television with the television on, the average score for evening meals was 2.8, for other meals, 2.6 (SD, 1.14), and for snacks, 2.7 (SD, 1.06). A score of 2 means occasionally and a score of 3 means often. Family support for healthy eating had a mean of 2.5 (SD, 0.55), with both negative and positive support items scoring relatively low.
Table 3Household Food Practices and Family Social Support Among Study Participants (n = 319)
Table 4 reports results from the multivariate regression analysis investigating the impact of the home food environment on fruit and vegetable intake and percentage of calories from fat intake. Having a higher number of fruits and vegetables in the home was predictive of higher fruit and vegetable intake (β = 0.11; SE, 0.02; P < .001). Similarly, a high frequency of fruit and vegetable shopping per week was associated with a higher fruit and vegetable intake (β = 0.29; SE, 0.08; P < .001). There was no significant association between fruit and vegetable intake and the unhealthy food and drinks inventories, food placement, meal preparation, and serving methods, or with the number of family meals from outside the home, eating with the television on, or social support for healthy eating. The model for fruit and vegetable intake was statistically significant (P < .001) and explained 27.9% of variance in fruit and vegetable intake.
Table 4Multivariate Models of Dimensions of Home Food Environment Associated With Fruit and Vegetable Intake and Percent Calories From Fat
Variable
Fruit and Vegetable Intake
Calories From Fat Intake, %
β
SE
P
β
SE
P
Intercept
1.04
1.14
.36
40.58
2.85
< .001
Fruit and vegetable inventory
0.11
0.02
< .001
−0.08
0.06
.15
Unhealthy food inventory
0.10
0.06
.11
0.45
0.15
.01
Unhealthy drinks inventory
−0.19
0.11
.09
−0.11
0.28
.70
Food placement
−0.10
0.17
.56
−0.63
0.42
.14
Fruit or vegetable shopping
0.29
0.08
.01
−0.10
0.21
.63
Meal preparation
0.43
0.26
.10
−2.06
0.65
.01
Meal serving
0.21
0.15
.16
−0.51
0.38
.19
Family meal purchased from non-home sources, times/wk
0.02
0.03
.60
−0.03
0.08
.66
Eating with television on
−0.16
0.10
.13
0.24
0.26
.36
Family support healthy eating
0.28
0.18
.12
0.14
0.46
.76
Model
r2
27.9%
17.9%
P
< .001
< .001
Note: Regression analysis was adjusted for demographic differences in race, age, employment status, income, marital status, having children in the home or not, and having other adults in the home or not.
The number of unhealthy food items in the home was positively associated with fat intake (β = 0.45; SE, 0.15; P < .001). Healthy meal preparation was negatively associated with fat intake (β = −2.06; SE, 0.65; P < .01), indicating the people who used healthier methods to prepare meals had lower intake of fat. None of the other home environment domains were associated with fat intake. The fat intake model explained almost 18% of the variance in percentage of calories from fat.
Discussion
This study is 1 of the first to examine the influence of the home food environment on healthy eating among overweight and obese low-income women. Study participants were largely African American and lived in the rural south. Prior research on home food environments has tended to focus on youth and their parents, often in urban or suburban settings with less poverty than is present in rural Georgia. Thus, the current study provides insight into the association between the home food environment and healthy eating in a high-risk population for obesity and associated chronic diseases.
Of note, among the physical aspects of the home environment, participants reported a wide range of fruits and vegetables available in their homes. Interestingly, a study of the home environment among a largely white population in Rhode Island documented 7.2 fruits and vegetables, whereas the current study documented 14 types.
This difference may be regional, cultural, or an artifact of measurement approaches, because the current study asked about a larger number of fruits and vegetables. Similar to other studies, the current study found that having more fruits and vegetables in the home was associated with higher levels of fruit and vegetable intake.
Although participants reported a variety of fruits and vegetables available in their home, they also had a significant number of unhealthy snack and drink options. This study found higher levels of high-fat snacks in the home than did the study of Gorin et al
(4.6 vs 3.7), although the current study asked about a larger number of items. Similar to other studies, the current study observed that the availability of high-fat snacks in the home was associated with fat intake.
Only 1 other study has assessed whether food accessibility in the home (eg, ease of access) is associated with eating behavior among adults. Gattshall et al
found that fruit and vegetable accessibility was correlated with fruit consumption for youth. Participants in the current study reported that both healthy foods and unhealthy foods were easily accessible in their homes; however, accessibility was not associated with eating behavior in study participants.
The high reporting of produce shopping per week was encouraging. Participants reported shopping for fruits and vegetables about once per week. A study of food purchasing in the Mississippi Delta reported that participants shopped for fruits and vegetables about 2 times a week,
whereas a study of Supplemental Nutrition Assistance Program recipients in North Carolina found that only 30.6% reported shopping at a supermarket once a week or more.
Association between travel times and food procurement practices among female supplemental nutrition assistance program participants in eastern North Carolina.
Grocery shopping is clearly an important decision point for healthy eating, although it is affected by a broad range of factors such as household income, price of foods, and location of food stores.
Because non-home food sources tend to be more caloric than foods prepared at home, a healthier home food environment is characterized by relatively few family meals from non-home sources.
Study participants indicated that they frequently used non-home food sources for family meals, about 2.6 d/wk, with 69.3% purchasing a family meal at least once per week. This is comparable to, or even higher than, finding from other studies, for example, Fulkerson et al
Away-from-home family dinner sources and associations with weight status, body composition, and related biomarkers of chronic disease among adolescents and their parents.
reported that 50% of families purchased family meals from non-home sources at least once per week. Home cooking, although often healthier than non-home sources, must be accompanied by healthy food preparation methods.
In the current study, healthier food preparation methods were associated with lower fat intake.
Another finding was that eating with the television on was common, with the majority reporting that their families occasionally or often eat meals and snacks in front of the television. Numerous studies have documented that television watching is associated with unhealthy diets.
Huffman FG, Vaccaro JA, Exebio JC, Zarini GG, Katz T, Dixon Z. Television watching, diet quality, and physical activity and diabetes among three ethnicities in the United States [published online ahead of print July 17, 2012]. J Environ Public Health. doi:10.1155/2012/191465.
Contrary to expectations, there was no association between eating in front of the television and either of the dietary behaviors assessed. Similarly, there was no association between family support for healthy eating and dietary behavior.
Despite the strengths of this study, a few limitations should be noted. First, most of the data were self-reported via telephone interviews and could be subject to recall or social desirability bias. Second, the ability to detect associations was dampened by the lack of variability in socioeconomic and weight status among participants. Indeed, this lack of variation in weight status as well as eating behavior may explain the lack of associations observed for many of the home environment variables. Only 3 dimensions of the home environment were associating with eating behaviors in this study: food inventories, grocery shopping, and meal preparation. With a more diverse study population, more associations among the constructs in the conceptual model would be anticipated. There was, however, considerable variability in some of the home environment domains. Fourth, data were cross-sectional and causality could not be examined. Finally, several of the measures have had modest reliability and/or validity in prior studies, which suggests that additional measurement research in this area may be warranted.
Differences in fat-related dietary patterns between black, Hispanic and White women: results from the Women's Health Trial Feasibility Study in Minority Populations.
The study highlights some common conditions that likely contribute to obesity in South Georgia: plentiful high-sugar beverages and high-fat snacks, family norms for eating in front of the television, and a substantial proportion of family meals from non-home sources. Although availability of fruits and vegetables was good, local traditions involve preparing these foods in non-healthful ways. Eating in front of the television was common, and changing that behavior is particularly challenging when working with individuals who are unemployed and who consequently spend a lot of time at home.
The women in this study were primarily obese, African American, unemployed, and low-income. There are likely additional psychosocial and environmental factors that promote and inhibit healthy eating in this population. Future research should question the intensity of social support needed to change the home environment and explore who is most effective in providing such support. Future studies should also consider use of a social ecologic perspective to examine multiple behavior settings simultaneously, such as neighborhoods, workplaces, and faith-based organizations, in addition to homes. The ways in which Supplemental Nutrition Assistance Program participation and food insecurity may affect home food environments and eating behaviors may also be an important line of inquiry. Finally, studies should examine the home food environment among a range of communities and population groups to more deeply understand its role in promoting healthy eating among adults.
Acknowledgments
This publication was supported by Cooperative Agreement 5U48DP001909 from the Centers for Disease Control and Prevention. The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. The authors thank members of the Emory Prevention Research Center's Community Advisory Board for guidance in the design and implementation of this research. They also thank the interviewers, study participants, and Community Health Center partners for their valuable contributions to this research.
References
Booth S.L.
Sallis J.F.
Ritenbaugh C.
et al.
Environmental and societal factors affect food choice and physical activity: rationale, influences, and leverage points.
Socio-economic, behavioural and environmental factors predicted body weights and household food insecurity scores in the Early Childhood Longitudinal Study-Kindergarten.
Away-from-home family dinner sources and associations with weight status, body composition, and related biomarkers of chronic disease among adolescents and their parents.
Differential associations of fast food and restaurant food consumption with 3-y change in body mass index: the Coronary Artery Risk Development in Young Adults Study.
Eating at fast-food restaurants is associated with dietary intake, demographic, psychosocial and behavioural factors among African Americans in North Carolina.
Huffman FG, Vaccaro JA, Exebio JC, Zarini GG, Katz T, Dixon Z. Television watching, diet quality, and physical activity and diabetes among three ethnicities in the United States [published online ahead of print July 17, 2012]. J Environ Public Health. doi:10.1155/2012/191465.
Sociodemographic and psychological predictors of changes in dietary fat consumption in adults with high blood cholesterol following counseling in primary care.
Differences in fat-related dietary patterns between black, Hispanic and White women: results from the Women's Health Trial Feasibility Study in Minority Populations.
Association between travel times and food procurement practices among female supplemental nutrition assistance program participants in eastern North Carolina.
00003-7&id=gr1.jpg){kind=link}