| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communication |
1 From the Division of Epidemiology and Clinical Applications, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD; the Endocrine-Hypertension Division, Brigham & Women's Hospital, Harvard Medical School, Boston; Kaiser Permanente, Center for Health Research, Portland, OR; the Pennington Biomedical Research Center, Louisiana State University, Baton Rouge; the Welch Center for Prevention, Johns Hopkins University, Baltimore; the Sarah W Stedman Center for Nutritional Studies, Duke University School of Medicine, Durham, NC; Merck & Co, Boston; and the Geriatric Research, Education and Clinical Center, VA Medical Center, Durham, NC.
2 Supported by NHLBI Cooperative Agreements U01-HL50968, U01-HL50972, U01-HL50977, U01-HL50981, and U01-HL50982; by RR00722 and RR02635; and by a supplement from the NIH Office of Research and Minority Health.
3 Address correspondence to E Obarzanek, Division of Epidemiology and Clinical Applications, National, Heart, Lung, and Blood Institute, 6701 Rockledge Drive, Room 8136, MSC 7936, Bethesda, MD 20892-7936. E-mail: obarzane{at}nhlbi.nih.gov.
See corresponding editorial on page 1.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: Our goal was to determine the effects of diet on plasma lipids, focusing on subgroups by sex, race, and baseline lipid concentrations.
Design: This was a randomized controlled outpatient feeding trial conducted in 4 field centers. The subjects were 436 participants of the Dietary Approaches to Stop Hypertension (DASH) Trial [mean age: 44.6 y; 60% African American; baseline total cholesterol: 
6.7 mmol/L (260 mg/dL)]. The intervention consisted of 8 wk of a control diet, a diet increased in fruit and vegetables, or a diet increased in fruit, vegetables, and low-fat dairy products and reduced in saturated fat, total fat, and cholesterol (DASH diet), during which time subjects remained weight stable. The main outcome measures were fasting total cholesterol, LDL cholesterol, HDL cholesterol, and triacylglycerol.
Results: Relative to the control diet, the DASH diet resulted in lower total (-0.35 mmol/L, or -13.7 mg/dL), LDL- (-0.28 mmol/L, or -10.7 mg/dL), and HDL- (-0.09 mmol/L, or -3.7 mg/dL) cholesterol concentrations (all P < 0.0001), without significant effects on triacylglycerol. The net reductions in total and LDL cholesterol in men were greater than those in women by 0.27 mmol/L, or 10.3 mg/dL (P = 0.052), and by 0.29 mmol/L, or 11.2 mg/dL (P < 0.02), respectively. Changes in lipids did not differ significantly by race or baseline lipid concentrations, except for HDL, which decreased more in participants with higher baseline HDL-cholesterol concentrations than in those with lower baseline HDL-cholesterol concentrations. The fruit and vegetable diet produced few significant lipid changes.
Conclusions: The DASH diet is likely to reduce coronary heart disease risk. The possible opposing effect on coronary heart disease risk of HDL reduction needs further study.
Key Words: Diet • plasma total cholesterol • LDL cholesterol • HDL cholesterol • triacylglycerol • feeding study • Dietary Approaches to Stop Hypertension Trial • DASH
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite extensive information, however, several important issues remain to be addressed. First, the effects of diet on blood lipids are known best in men. Several studies (3–6), but not all (7, 8), noted that blood lipids in women may be less responsive to diet than are blood lipids in men. Most studies had small numbers of women, and statistical comparisons of lipid responses between men and women were not always made. Second, there is even less information in African Americans because most studies enrolled whites or did not report results separately for different racial groups. Third, the effects of dietary carbohydrate on blood triacylglycerol may vary according to type of carbohydrate (9) as well as energy balance (10). It is well established that simple carbohydrates or mixtures of simple and complex carbohydrates raise triacylglycerol when substituted for fats (1). What is not fully established is whether carbohydrate-rich foods that are rich in fiber, or that are digested slowly and produce a relatively mild increase in blood glucose and insulin, also raise triacylglycerol concentrations when substituted for dietary fat.
The Dietary Approaches to Stop Hypertension (DASH) Trial provided the opportunity to study each of these topics (11, 12). DASH was an outpatient controlled feeding study that tested the effects on blood pressure of 2 experimental dietary patterns compared with a control dietary pattern similar to what many Americans eat. Both diets differed from the control diet in the type of carbohydrates they contained. Relative to the control diet, each experimental diet contained less refined grains and sweets and more whole grains, fruit, and vegetables. The last 3 foods contain carbohydrates that have been proposed to have little effect on triacylglycerol (13). Because one-half of the 459 DASH participants were women and 60% were African American, the DASH data can be used to determine the effects of a reduced-fat, increased-carbohydrate diet on blood lipids under conditions of stable weight in these large but understudied segments of the US population. Because LDL and HDL cholesterol are independent risk factors for coronary heart disease (CHD) (14), and because triacylglycerol may be a risk factor for cardiovascular disease independent of HDL (15), this information is relevant for disease prevention programs.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Participants and methods
Participants were eligible for the trial if they were 
22 y old and had mean diastolic blood pressure between 80 and 95 mm Hg and systolic blood pressure <160 mm Hg. Lipid exclusion criteria included fasting or nonfasting plasma total cholesterol >6.7 mmol/L (>260 mg/dL) or, on repeat testing, LDL cholesterol 4.1 mmol/L (160 mg/dL) or 3.4–4.1 mmol/L (130–159 mg/dL) with 2 other CHD risk factors. Other major exclusion criteria included poorly controlled diabetes mellitus, a cardiovascular event within the previous 6 mo, body mass index (in kg/m2) >35, use of medications affecting blood pressure or nutrient metabolism, unwillingness to stop taking vitamin and mineral supplements or antacids containing magnesium or calcium, and an alcohol intake of >14 drinks/wk. Because of the disproportionate incidence of hypertension in minority populations, the study placed an emphasis on recruiting minority participants. Participants were recruited in 4–5 cohorts per center.
Dietary patterns
Shown in Table 1
are the nutrient and food group content of the menus for the 3 dietary patterns used in the study, based on an energy intake of 9.9 MJ (2100 kcal). Actual mean (±SD) energy intakes of the participants during the 8-wk intervention were 11.1 ± 2.4 MJ (2649 ± 584 kcal), 11.0 ± 2.3 MJ (2624 ± 552 kcal), and 10.9 ± 2.3 MJ (2610 ± 550 kcal) for the control, fruit and vegetable (FV), and DASH dietary patterns, respectively.
|
Study procedures
During the 3 wk of run-in and 8 wk of intervention feeding, the clinical centers provided participants with all of their meals and snacks according to a 7-d menu cycle. All clinics followed a common protocol to prepare the standardized menus in a research kitchen (19). Energy intakes were adjusted as needed to keep body weight constant. Participants attended the clinic each weekday to be weighed, to consume one meal on-site (lunch or dinner), and to pick up the balance of their meals and snacks. The participants picked up their weekend meals and snacks on Fridays. Participants were allowed to consume 3 nonalcoholic diet beverages (coffee, tea, and soft drinks) and 2 alcoholic beverages daily. Lists of permissible beverages were provided. Participants were requested not to change their habitual levels of physical activity for the duration of the study.
Measurements
All measurement staff were blinded to the diet assignments. Twelve-hour fasting blood samples were drawn toward the end of the run-in period and again at the end of the 8-wk intervention period. Blood was collected into tubes containing 0.1% EDTA and was centrifuged at 4°C and 500 x g for 10 min to separate the plasma. Frozen plasma was shipped to a central laboratory, the Oregon Health Sciences University Lipid-Atherosclerosis Laboratory, for analysis of total cholesterol, HDL cholesterol, and triacylglycerol. Lipoproteins were separated by preparative ultracentrifugation (100000 x g and 4°C for 12 h) at a density of 1006 g/L and were precipitated with use of standardized techniques (20, 21). Enzymatic colorimetric methods (20) were used to determine the amounts of total cholesterol, HDL cholesterol, and triacylglycerol in plasma by using an automated analyzer (model 704; Hitachi, Indianapolis). No participant had a triacylglycerol concentration >4.5 mmol/L (400 mg/dL). All lipid determinations were standardized with samples of known composition obtained from the Centers for Disease Control and Prevention–National Heart, Lung, and Blood Institute Lipid Standardization Program (22). The interassay CVs of the control plasma samples were 2.5% for total cholesterol and 3.5% for triacylglycerol. The ratios of total to HDL cholesterol (TC:HDL) and of LDL to HDL cholesterol (LDL:HDL) were calculated.
During the screening, height and weight were measured while the participants wore light indoor clothing and no shoes. Body mass index was calculated as kg/m2. Participants completed a food-frequency questionnaire (23, 24) to provide information on their usual food intake over the year before their entry into the trial; from this, we calculated a Keys score (25), which reflects the atherogenicity of the diet (26). Participants also provided information during the screening on their smoking habits, alcohol intake, and medication use. Physical activity information was obtained from a 7-d activity recall (27) completed during the screening and at the end of the intervention period. During the entire 11 wk of feeding, participants were instructed to keep a daily diary that was reviewed by study staff to keep track of the allowed beverages and diet deviations as well as consumption of standardized muffins and cookies (unit foods) that were provided as needed to maintain body weight (12, 28).
Moore's Extended Nutrient (MENu) database (version 1.0, 1994; Pennington Biomedical Research Foundation, Baton Rouge, LA) was used to calculate the energy, macronutrient, cholesterol, and mineral contents of the menus. The Nutrition Data System (NDS) database (version 2.8, 1995; University of Minnesota, Minneapolis) was used to calculate total dietary fiber, water soluble and insoluble fiber, and sucrose. Energy and nutrient intakes of individuals were obtained on the basis of the calculated nutrient content of the assigned diet's energy level and on the number of unit foods consumed. The energy value of alcoholic beverages was included.
Statistical methods
We used one-way analysis of variance and chi-square tests to determine the significance of any baseline differences between diet groups and between population subgroups defined by sex and race. Although we had no real hypothesis of interest regarding baseline differences between diet groups, these tests provide the reader with a quantitative measure of the homogeneity of the samples.
The primary outcome measures for this report were the changes in lipids from run-in to the end of the intervention. We used standard linear regression techniques to model these changes, adjusted for cohort, as a function of treatment. The differences in changes produced by the FV and DASH diets in comparison with the control diet were tested for significance by calculating the appropriate contrasts from the regression models. That is, the changes were expressed "net of control."
Because triacylglycerol concentrations were highly skewed, we used nonparametric statistics to summarize the baseline concentrations. The distribution of change in triacylglycerol appeared symmetrical, so we did not transform these change data in the regression analyses.
The primary analyses used linear regression models that tested differences in lipid changes between diets, adjusted only for cohort effects, and that did not include subgroup indicators. Subgroup-specific estimates of treatment effects were simultaneously examined by using models that included main effects of treatment, cohort, race (African Americans and non-African Americans), sex, and baseline lipid concentrations (dichotomized as above or below a cutoff, as defined below, for total, LDL, and HDL cholesterol and triacylglycerol) and interactions of treatment with race, sex, and baseline lipid concentration. (Only the specific baseline lipid concentration corresponding to the outcome variable was used in any given model.) These models thus allowed us to estimate treatment effects within any one subgroup variable (eg, effects in men and women) and to test whether these effects differed (eg, men compared with women) after adjustment for the effects of the other 2 subgroup variables.
Baseline total and LDL-cholesterol cutoffs indicating a high risk of CHD were chosen (14): a baseline total cholesterol concentration 5.2 mmol/L (200 mg/dL), comprising 41% of the study sample, and a baseline LDL-cholesterol concentration 3.4 mmol/L (130 mg/dL), comprising 42% of the study sample. Only 13% had HDL cholesterol <0.9 mmol/L (<35 mg/dL) and 7.1% had triacylglycerol 2.3 mmol/L (200 mg/dL), so cutoffs for HDL and triacylglycerol were based on study sample medians [1.2 mmol/L (45 mg/dL) for HDL and 1.2 mmol/L (94 mg/dL) for triacylglycerol].
We used the equations of Keys et al (29) and of Mensink and Katan (1) to calculate predicted changes in lipid concentrations on the basis of calculated changes in macronutrient and cholesterol intakes of the participants and compared these with observed changes. To compare differences between observed and predicted lipid values net of control, we subtracted from each of the 2 experimental groups the mean observed change in the control group. All statistical analyses were conducted with SAS (version 6.12; SAS Institute Inc, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
60% were African American, 6% were other minorities, and 34% were white. Baseline characteristics of these participants did not differ significantly across the 3 diet groups (Table 2). The participants were overweight and inactive (30). At baseline, only 3 persons, 1 in each diet group, were taking cholesterol-lowering medication and 30 women, 10 in each diet group, were taking estrogen replacement therapy.
|
). Participants reported consuming 9.2 MJ (2190 kcal) as their habitual energy intake, but required 11.0 MJ (2630 kcal) for weight maintenance during the study. The participants' usual intakes of total fat, saturated fat, and polyunsaturated fat as a percentage of energy were within one percentage point of the content of the control diet, and their usual intake of carbohydrate as a percentage of energy was 4 percentage points lower than the content of the control diet.
|
). African Americans had a significantly lower TC:HDL and significantly lower triacylglycerol concentrations at baseline than did non-African Americans (Table 5). Additionally, women had significantly lower concentrations of LDL cholesterol and triacylglycerol, a lower TC:HDL and LDL:HDL, and higher concentrations of HDL cholesterol at baseline than did men (Table 5).
|
|
). These changes represented net reductions of 7.3%, 9.0%, and 7.5% in mean concentrations of total, LDL, and HDL cholesterol, respectively. No significant changes were observed for triacylglycerol [0.04 (-0.06, 0.13) mmol/L, or 3.2 (-5.1, 11.6) mg/dL], TC:HDL [-0.03 (-0.19, 0.13)], or LDL:HDL [-0.08 (-0.22, 0.06)].
|
). However, the reductions in total and LDL cholesterol and the changes in the TC:HDL and LDL:HDL differed significantly between men and women, with reductions in values being larger in men than in women. Adjusted for cohort, race, and baseline concentrations, men had 0.27-mmol/L (10.3 mg/dL) and 0.29-mmol/L (11.2 mg/dL) greater net reductions in total and LDL cholesterol, respectively, and greater net reductions in TC:HDL and LDL:HDL than did women. Changes in total cholesterol, LDL cholesterol, HDL cholesterol, cholesterol ratios, and triacylglycerol did not differ significantly by race. Changes in total cholesterol, LDL cholesterol, and triacylglycerol also did not differ significantly by baseline lipid concentrations. However, adjusted for cohort, race, and sex, reductions in HDL were greater by 0.11 mmol/L (4.1 mg/dL) in those whose baseline HDL concentrations were 1.2 mmol/L (45 mg/dL) than in those whose baseline concentrations were <1.2 mmol/L (<45 mg/dL). This significant interaction with baseline HDL concentrations persisted after we omitted several outlying observations. Shown in Figure 2 are the scatter plots of the changes in LDL and HDL in relation to their respective baseline concentrations.
|
|
). Mean triacylglycerol concentrations tended to be reduced by -0.09 (-0.19, 0.002) mmol/L [-8.2 (-16.5, 0.2) mg/dL]. Changes in the TC:HDL of –0.14 (-0.29, 0.02) and in the LDL:HDL of –0.10 (-0.23, 0.04) also were not significant.
Compared with their counterparts consuming the control diet, men consuming the FV diet had a significantly reduced LDL:HDL, both men and non-African Americans consuming the FV diet had a significantly reduced TC:HDL, and participants with higher baseline triacylglycerol concentrations had significantly reduced triacylglycerol concentrations (Table 6). Although there were no other significant changes in lipids in subgroups of participants consuming the FV diet, the following trends (ranging from P = 0.053 to P = 0.076) were noted: a trend for reduced total cholesterol in men (P = 0.053) and trends for reduced triacylglycerol (P = 0.076) and reduced LDL:HDL (P = 0.069) in non-African Americans. None of the changes differed significantly between the subgroups (interaction terms NS).
Predicted versus observed lipid changes
For the DASH diet, the Keys equation generally predicted a greater total cholesterol reduction than was observed, whereas the Mensink and Katan equation predicted a smaller reduction than was observed (Table 7). The observed rise in triacylglycerol was less than predicted on the basis of the Mensink and Katan equation. The observed reductions in LDL and HDL cholesterol were not significantly different from the predictions of the Mensink and Katan equation.
|
). However, the observed change in triacylglycerol was less than predicted.
Alcohol and physical activity
Mean changes in alcohol consumption and physical activity were small and did not differ significantly among the 3 diet groups (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Previous controlled feeding studies tended to have insufficient numbers of men and women to make comparisons with adequate power to detect a differential sex response of lipids to diet. In 2 controlled feeding studies that specifically tested for sex differences, no significant differences in lipid response were detected between 57 women and 46 men (7) or between 33 women and 30 men (8). In contrast, Schaefer et al (6) pooled the results of 5 controlled feeding studies having a total of 48 women and 72 men to compare their lipid response to a National Cholesterol Education Program Step II diet and found that, similar to the present study, men had greater mean net reductions in total cholesterol (by –0.24 mmol/L, or -9 mg/dL; P < 0.03) and in LDL cholesterol (by –0.17 mmol/L, or -6.6 mg/dL; P = 0.058) than did women. Unlike in the present study, however, Schaefer et al (6) reported a trend for a 0.05-mmol/L (1.9 mg/dL) greater reduction in HDL in men than in women (P = 0.09). The results of the present study, which used a single protocol with 223 men and 213 women and controlled for race and baseline lipid concentrations, suggest that women exhibit a smaller response in total and LDL cholesterol, but not in HDL cholesterol or triacylglycerol, to reduced-fat diets than do men of similar age and body mass index and under conditions of stable weight.
Connett and Stamler (31) reported that 416 black men in the special intervention group of the Multiple Risk Factor Intervention Trial study had total cholesterol, LDL-cholesterol, and triacylglycerol responses to the intervention similar to those of 5338 white men. Similarly, Ginsberg et al (7), examining 26 black participants, and Howard et al (8), examining 34 black participants, reported no significant differences in lipid response to diet between black and nonblack participants in their controlled feeding studies. The present study, with 261 African Americans, found no significant differences between African Americans and non-African Americans.
We found no significant differential reduction in plasma total cholesterol, LDL cholesterol, or triacylglycerol according to baseline concentrations of these measures. As shown in Figure 2, trends in LDL change were similar after both the control and DASH diets, indicating regression to the mean or some other nonspecific effect, rather than a specific diet effect. This finding suggests that appropriate statistical comparisons with the control group should be made before drawing conclusions regarding effects of baseline concentrations. In contrast, HDL cholesterol was reduced more by the DASH diet in individuals with higher baseline HDL-cholesterol concentrations than in those with lower baseline concentrations. Thus, the DASH diet has a smaller effect on HDL cholesterol in those whose low HDL-cholesterol concentrations might place them at increased CHD risk.
The Keys equation appeared to overpredict the degree of total cholesterol reduction from the DASH diet, perhaps because it was based primarily on studies of hyperlipidemic men (29). In the present study, persons with high total cholesterol were excluded and men had a greater total cholesterol response to the DASH diet than did women. Because the FV diet did not alter dietary fat, changes in total or LDL cholesterol were not expected and were not significantly different from those predicted. However, data from several studies indicated that increasing dietary soluble fiber reduces blood cholesterol concentrations independent of dietary fat intake (32–35). The present study suggests that changes in total and LDL cholesterol are minimal without concomitant reductions in fat intake when dietary fiber is increased by 20 g/d, with no particular emphasis on soluble or insoluble fiber.
Intriguing findings in this report were the –0.09-mmol/L (-8.2-mg/dL; P = 0.055) lowering of triacylglycerol produced by the FV diet when no significant change was expected, and the nonsignificant 0.04-mmol/L (3.2-mg/dL; P = 0.45) increase in triacylglycerol produced by the DASH diet, when a much greater rise of 0.21 mmol/L (18.9 mg/dL) was expected on the basis of the prediction equations. Some investigators suggested that triacylglycerol concentrations will not rise if the carbohydrates consumed are low in sucrose and high in fiber (5, 35, 36). However, other investigators observed increases in triacylglycerol despite the use of foods high in fiber and complex carbohydrates (37). It is also unclear whether low-fat, high-carbohydrate diets improve insulin resistance (38, 39). The sucrose content of the FV and DASH diets was about one-third less, and the soluble, insoluble, and total dietary fiber contents were 2.5–3.0 times higher, than in the control diet. However, the increase in carbohydrate with the DASH diet (58% compared with 50%) was only moderate. The present study was not designed to test the specific dietary components responsible for the observed response in lipids. However, these results raise the intriguing possibility that lower-fat diets based on plants and whole grains prevent the predicted rise in triacylglycerol (40).
As did the diets in other studies in which dietary fat intakes were lowered, the DASH diet lowered HDL cholesterol in addition to total and LDL cholesterol. This reduction occurred similarly in each race and sex and was more pronounced in those with higher baseline HDL-cholesterol concentrations. The observation that reduced-fat diets lower HDL has led to controversy over the advisability of lower-fat, higher-carbohydrate diets and whether CHD risk is reduced when HDL is lowered along with LDL (41, 42). Several studies showed that populations with low total cholesterol and low HDL cholesterol do not have increased CHD risk (43–45), and that plant-based diets similar to the DASH diet are associated with decreased CHD risk (35, 46, 47). It is also recognized that overweight persons who lose weight while consuming reduced-fat diets have favorable changes in all lipids, including HDL and triacylglycerol (26, 48).
The DASH diet substantially reduced systolic and diastolic blood pressure (12), caused favorable changes in total and LDL cholesterol, resulted in no changes in triacylglycerol, and reduced HDL cholesterol. Efforts have been made to estimate the overall effect of various risk factors on CHD risk (49, 50). Applying the Framingham risk equation to estimate 10-y CHD risk on the basis of baseline and postintervention values of systolic blood pressure and total cholesterol, 10-y risk was decreased by 12.1% in participants consuming the DASH diet compared with a 0.9% increase in risk in participants consuming the control diet (P < 0.01). The respective changes in 10-y risk when LDL values were used in place of total cholesterol were –12.2% and –2.2% (P = 0.057). Thus, on balance, the DASH diet affects CHD risk favorably. The possible opposing effect on CHD risk of HDL reduction needs further elaboration in population studies, particularly in women who have smaller reductions in total cholesterol and in experimental models of HDL metabolism and atherosclerosis.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. L. Cugnetto, P. G. Saab, M. M. Llabre, R. Goldberg, J. R. McCalla, and N. Schneiderman Lifestyle Factors, Body Mass Index, and Lipid Profile in Adolescents J. Pediatr. Psychol., August 1, 2008; 33(7): 761 - 771. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J de Souza, J. F Swain, L. J Appel, and F. M Sacks Alternatives for macronutrient intake and chronic disease: a comparison of the OmniHeart diets with popular diets and with dietary recommendations Am. J. Clinical Nutrition, July 1, 2008; 88(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T. Fung, S. E. Chiuve, M. L. McCullough, K. M. Rexrode, G. Logroscino, and F. B. Hu Adherence to a DASH-Style Diet and Risk of Coronary Heart Disease and Stroke in Women Arch Intern Med, April 14, 2008; 168(7): 713 - 720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. B. Mellen, S. K. Gao, M. Z. Vitolins, and D. C. Goff Jr Deteriorating Dietary Habits Among Adults With Hypertension: DASH Dietary Accordance, NHANES 1988-1994 and 1999-2004 Arch Intern Med, February 11, 2008; 168(3): 308 - 314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. O'Neil and T. A. Nicklas State of the Art Reviews: Relationship Between Diet/ Physical Activity and Health American Journal of Lifestyle Medicine, December 1, 2007; 1(6): 457 - 481. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Mejean, P. Traissac, S. Eymard-Duvernay, J. El Ati, F. Delpeuch, and B. Maire Diet Quality of North African Migrants in France Partly Explains Their Lower Prevalence of Diet-Related Chronic Conditions Relative to Their Native French Peers J. Nutr., September 1, 2007; 137(9): 2106 - 2113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Esmaillzadeh, M. Kimiagar, Y. Mehrabi, L. Azadbakht, F. B Hu, and W. C Willett Dietary patterns, insulin resistance, and prevalence of the metabolic syndrome in women Am. J. Clinical Nutrition, March 1, 2007; 85(3): 910 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Griel, E. H. Ruder, and P. M. Kris-Etherton The Changing Roles of Dietary Carbohydrates: From Simple to Complex Arterioscler. Thromb. Vasc. Biol., September 1, 2006; 26(9): 1958 - 1965. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B Michels The women's health initiative--curse or blessing? Int. J. Epidemiol., August 1, 2006; 35(4): 814 - 816. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Estruch, M. A. Martinez-Gonzalez, D. Corella, J. Salas-Salvado, V. Ruiz-Gutierrez, M. I. Covas, M. Fiol, E. Gomez-Gracia, M. C. Lopez-Sabater, E. Vinyoles, et al. Effects of a Mediterranean-Style Diet on Cardiovascular Risk Factors: A Randomized Trial Ann Intern Med, July 4, 2006; 145(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Lichtenstein, L. J. Appel, M. Brands, M. Carnethon, S. Daniels, H. A. Franch, B. Franklin, P. Kris-Etherton, W. S. Harris, B. Howard, et al. Diet and Lifestyle Recommendations Revision 2006: A Scientific Statement From the American Heart Association Nutrition Committee Circulation, July 4, 2006; 114(1): 82 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Cakir and R. Pinar Randomized controlled trial on lifestyle modification in hypertensive patients. West J Nurs Res, March 1, 2006; 28(2): 190 - 209. [Abstract] [PDF]
|
|
|
|
 |
|
 L. Azadbakht, P. Mirmiran, A. Esmaillzadeh, T. Azizi, and F. Azizi Beneficial Effects of a Dietary Approaches to Stop Hypertension Eating Plan on Features of the Metabolic Syndrome Diabetes Care, December 1, 2005; 28(12): 2823 - 2831. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Appel, F. M. Sacks, V. J. Carey, E. Obarzanek, J. F. Swain, E. R. Miller III, P. R. Conlin, T. P. Erlinger, B. A. Rosner, N. M. Laranjo, et al. Effects of Protein, Monounsaturated Fat, and Carbohydrate Intake on Blood Pressure and Serum Lipids: Results of the OmniHeart Randomized Trial JAMA, November 16, 2005; 294(19): 2455 - 2464. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
