Journal of Human Hypertension (2015), 1–6 © 2015 Macmillan Publishers Limited All rights reserved 0950-9240/15 www.nature.com/jhh

ORIGINAL ARTICLE

Comparison of three DASH scoring paradigms and prevalent hypertension among older Hispanics C Tangney1, D Sarkar1,3 and BA Staffileno2 Older Hispanics are less likely to be aware of their hypertension or adopt lifestyle modifications for hypertension control than non-Hispanic whites. Few reports exist concerning Dietary Approaches to Stop Hypertension (DASH) accordance among Hispanics. This study was designed to: (1) assess accordance to a DASH pattern using three widely used DASH scoring paradigms; and (2) determine which DASH paradigm was most strongly associated with hypertension in 169 older Hispanics (mean age, 66 years and 73% female). Food frequency questionnaires were used to calculate DASH scores. Logistic regression modeling was performed for prevalent hypertension with the DASH scores, age, gender and acculturation. Using the Folsom et al. DASH scoring paradigm, 55% of adults were deemed DASH accordant compared with 17% using Fung et al. scores and 13% using the Toledo et al. Folsom et al. scores were predictive of prevalent hypertension (odds ratio = 1.35, 95% confidence interval (1.04, 1.77) in this older Hispanic sample; the remaining two scoring systems were not associated with hypertension in this sample. Journal of Human Hypertension advance online publication, 28 May 2015; doi:10.1038/jhh.2015.50

INTRODUCTION Hypertension is a major public health concern in the United States population. Approximately one out of every three adults or 77.9 million people have high blood pressure (BP).1 Although the prevalence of hypertension is lower among Hispanics compared with Caucasians (22 vs 28%), Hispanics are less likely to be aware of their hypertension or adopt healthy lifestyle modifications for controlling their hypertension.2 The Dietary Approaches to Stop Hypertension (DASH) eating pattern has been widely promoted and was developed as a treatment recommended for hypertension (http://www.nhlbi.nih.gov/health/health-topics/topics/dash). Lifestyle changes, such as the DASH eating plan along with weight loss and increased physical activity have been shown to reduce BP.3–6 More recently, there is renewed attention to this pattern with the release of the American Heart Association/ American College of Cardiology guidelines for lifestyle management of cardiovascular risk.7 The DASH eating pattern is high in fruits and vegetables, moderate in low-fat dairy products, low in animal protein such as red or processed meats and moderate in legumes, nuts and seeds. There are more than 10 different DASH scoring paradigms to assess accordance or adherence. (Accordance is used8–10 to assess the level of conformity to a dietary recommendation/pattern. Adherence is used compliance to a plan is assessed once the individual has been counseled on this plan.5,6) These paradigms differ in many ways including, the tools used to collect food intake, the number of food items that constitute a key food group to be scored, which food items comprise the food group to be scored or finally whether intakes of nutrients or food items/groups (or a combination of both) are the basis for scoring. Although food frequency questionnaires (FFQ), food records or 24-hour recalls may all be used to measure usual dietary intakes, the FFQ is 1

easiest to use with large population samples. Then, the scoring is standardized for food items on the FFQ. However, if there are no specific food items on the FFQ to score (such as a whole grain food item), assumptions are made that could contribute to misclassification. Scoring paradigms selected include those of Toledo et al.,11 Fung et al.12 and Folsom et al.13 because: (1) these investigators predominately used food groups to assess DASH accordance; (2) FFQs were used to measure usual intake; and (3) there is evidence for lower risk of hypertension8,10,11,14 and cardiovascular disease12,13 among those deemed accordant to one of these DASH scoring systems. The emphasis on food items or groups as a feature of the scoring tool was particularly key because we wanted to identify a scoring approach which could be applied directly on the completed FFQ, so the clinician might provide immediate feedback to clients. Scoring intake of foods or food groups would also be useful for behavioral targets with clients. In our previous work15 we tested a scoring paradigm by Mellen et al.10 that is exclusively nutrient-driven. Because many nutrients are found in the key foods advocated for the DASH food pattern, it was important to ascertain whether DASH accordance with more food-based scoring paradigms would also be associated with hypertension in this sample of adults. Because we had observed that acculturation scores were predictive of hypertension status,15 these were also examined in relation to DASH scores. Thus, the overall purpose of this study was to apply three commonly available DASH scoring paradigms11–13 to FFQs completed by a sample of Hispanic individuals currently living in the Chicago area. The primary objective was to determine whether DASH scores computed from the three tools were associated with the presence or absence of hypertension, and measured systolic BP (SBP) and diastolic BP or (DBP). In addition,

Department of Clinical Nutrition, Rush University Medical Center, Chicago, IL, USA and 2Department of Adult Health and Gerontological Nursing, College of Nursing, Rush University Medical Center, Chicago, IL, USA. Correspondence: Dr C Tangney, Department of Clinical Nutrition, Rush University Medical Center, 1700 West Van Buren Street, Suite 425, Chicago, IL 60612, USA. E-mail: [email protected] 3 This work is in partial fulfillment of a Master’s of Science degree in Clinical Nutrition. Received 9 February 2015; revised 4 April 2015; accepted 9 April 2015

Comparison of DASH scores C Tangney et al

2 we wished to examine whether acculturation or other demographic characteristics differed according to DASH accordance as defined by these three scoring approaches.

This is a secondary analysis of a cross-sectional study known as the Cognition and Physical Activity Community Elderly Study (CAPACES), in which the health and physical activity behaviors of Hispanic older adults currently residing in the metropolitan Chicago area were examined. Details on the design of CAPACES study have been reported.16

from 0 to 11. Daily total calories of 2000 kcal were used to adjust food and nutrient intakes. The scoring system has been more recently applied to overweight, hypertensive men and women in a year-long diet (DASH) and exercise trial.23 One point was assigned to adults who ate ⩾ 4 servings of vegetables and 0.5 point was awarded to those who had an intake of 2–3 daily servings of vegetables, and 0 points for participants consuming o2 servings per day. Reverse scoring was applied to those food groups where a lower intake was desired. For items for which a lower intake is desired (sweets), 1 point was assigned to participants who had ⩽ 5 servings/week, 0.5 point was given for 6–7 servings/week and 0 points were given to those who had an intake of ⩾ 8 servings/week of sweets. A score of 6 and higher was considered DASH accordant, whereas o 6 was considered DASH non-accordant.

Measures

Statistical analysis

Demographic and clinical characteristics included age, BP and heights and weights. Weight and height were measured with shoes removed. Body mass index (BMI) was calculated as weight (kilograms)/height (meters) squared. BP was measured in triplicate (Omron HEM-907XL, Omron Health Care, Inc; Vernon Hills, IL, USA), following the American Heart Association recommendations.17 Hypertension was defined as having a mean SBP of ⩾ 140 mm Hg, and/or a mean DBP of ⩾ 90 mm Hg, or if it was confirmed that they were taking hypertension medications.10 Acculturation was measured using the multidimensional scale, the Acculturation Rating Scale for Mexican Americans-II (ARISMA-II) as reported by Cuellar and colleagues.18 The CAPACES investigators replaced the term ‘Mexican’ or ‘Mexicano’ with ‘Latinos/Hispanics’ in an effort to be more inclusive of all the Hispanic groups, such as, Columbians or Puerto Ricans. Scores were computed by subtracting the mean score for the Latino subscale from that of the Anglo subscale. Lower scores reflect low acculturation and higher scores reflect higher acculturation.16

Demographic variables of the CAPACES participants were summarized using descriptive statistics using PASW statistics version 18.0 (SPSS, Inc., Chicago, IL, USA). For every participant, three DASH scores were calculated; these include scores based on those by Toledo et al.,11 Fung et al.12 and Folsom et al.13 Categorical variables such as sex, age, acculturation and education were reported as frequencies, whereas continuous variables such as BMI, SBP and DBP were reported as means and s.d. or medians and interquartile ranges. In effort to describe participants who are DASH accordant with each of the selected scoring paradigms, demographic characteristics were compared between those who are DASH accordant and those who are not (for each scoring paradigm). In addition, daily intakes of fruits, vegetables and dairy of all men and women who were deemed accordant to each of the three DASH scoring paradigms were reported these intakes were represented graphically. Analysis of variance was used in an effort to determine the differences between groups (DASH score grouping) for demographic characteristics and dietary intakes with post hoc Bonferroni corrections. For variables that were not normally distributed, Mann– Whitney and Kruskal–Wallis tests with Bonferroni corrections were used. Intakes of sodium, glycemic load and dietary fiber were adjusted for total energy intake using the residual method.24 Cross-tabulations, Χ2-tests, and kappa tests of agreement were used to assess agreement or association between DASH scoring pairs. Additionally, crude DASH scores were correlated using Pearson or Spearman correlation tests depending on normality. In effort to determine whether any of the three DASH scores were associated with presence or absence of hypertension, logistic regression modeling was performed with age, sex, acculturation and hypertension. Stepwise regression models were also run for the three chosen scores with the outcome of SBP (and DBP), controlling for potential confounders such as age, sex, acculturation and education completed. Significance level was set at Po0.05.

METHODS Design and participants

Dietary assessment The Chicago Health and Aging Project (CHAP) FFQ is an adapted version of the Harvard FFQ and consists of 139 food items, and vitamin and mineral supplements19,20 in which usual portion sizes and frequency of foods over the previous 12 months were selected. FFQs were administered by a single bilingual interviewer. Nutrient intakes were estimated using the Harvard nutrient database, an adaptation of the United States Department of Agriculture nutrient database of foods and supplements. Validity and reproducibility of the FFQ has been assessed previously.21,22

DASH scores Toledo et al. DASH score. Toledo et al.11 applied a scoring paradigm to completed FFQ responses from 10 800 people from the SUN cohort in Spain. A score of 0 or 1 was assigned based on whether food group targets were achieved (in servings per day). The intake of ⩾ 5 servings per day of fruit, ⩾ 4 servings of vegetables, 2–3 servings of low-fat or non-fat dairy, ⩽ ½ serving of sweets and ⩾ 1 serving of whole grains, and 1–3 servings of lean meat, poultry or fish was given a score of 1 point. Intakes not meeting these targets were assigned a score of 0. Total scores ranged from 0 to 6; the higher the value, the more DASH accordant. By using this scoring scheme for CHAP FFQ responses, CAPACES participants assigned a total score of 3 or higher were deemed DASH accordant and those with a score of less than 3 were deemed DASH non-accordant. Fung et al. DASH scores. Similar to the approach used by Fung et al.12 in the Nurses’ Health Study, food group intakes based on the FFQ responses of CAPACES adults were stratified into quintiles. For the first five component scores (fruits, vegetables, nuts/legumes, whole grains and low-fat dairy), higher intakes are deemed desirable. Thus, intakes in quintile 1 (low consumption) were assigned 1 point and those in quintile 5 (high consumption) was assigned 5 points. For the last three components (sodium, red and processed meats and sweets and beverages), where low intakes are preferred, the lowest quintile was assigned a score of 5, and the highest quintile, 1 point. Scores for all components were summed for a maximum possible score of 40 and a lowest possible score of 8. Participants with scores of 29 or higher were deemed DASH accordant, whereas participants who scored o29 were deemed DASH non-accordant. Folsom et al. DASH score. Folsom et al.13 computed DASH scores for Iowa Women’s Health Study participants, aged 55–69 years. Scores could range Journal of Human Hypertension (2015), 1 – 6

RESULTS In Table 1, characteristics of the entire sample are shown in the first column and the next six columns, demographic characteristics of the top and bottom strata of the three chosen scoring paradigms are presented. Participants were, on average, 66 years of age, predominately female, 57% had less than a high school degree, and obese (BMI, 30.2 ± 5.0, mean ± s.d.). Almost a third were hypertensive, and of these half were on anti-hypertensive medications. In the next three pairs of columns in Table 1, the characteristics of participants who fall in the least accordant group and higher groups (most DASH accordant) of each of the DASH scoring paradigms are presented. According to scores by Toledo et al.,11 13% were DASH accordant, while based on Fung et al.12 scores, 16.5% of the adults were accordant. With the Folsom et al.,13 about 55% of the population were DASH accordant. When the demographics of participants were compared across groups defined by low and high accordance to any of the DASH tools, there were few notable differences. Although education did not differ across accordance groups, acculturation scores were different (P = 0.012) across Folsom et al.13 DASH groups. Those with higher Folsom et al. DASH accordance were significantly more Hispanic oriented than those in the lower tertile. There were no differences in SBP or DBP when comparing low and high © 2015 Macmillan Publishers Limited

Comparison of DASH scores C Tangney et al

3 Table 1.

Demographic and clinical characteristics of all CAPACES participants and as stratified reported dietary behaviors based on three different DASH scores Demographic characteristics

Age, years Female, % Education, % Less than high school Some college College graduate MMSE Acculturation scorec Clinical characteristics Systolic BP, mm Hg Diastolic BP, mm Hg Hypertension, n (%)f Anti-HTN meds, n (%) BMI, kg m − 2

All N = 169

66.0 ± 9.0a 73.4 56.8 32 11.2 12.4 ± 2.1 1.8 − 2.3, − 0.9 130 ± 18 70 ± 11 49 (29) 89 (53) 30.2 ± 5.0

Toledo et al. DASH

Fung et al. DASH

Folsom et al. DASH

Score 0 N = 42

Score 3–6 N = 22

Score 8–20 N = 35

Score 29–38 N = 28

Score 0–5 N = 40

Score 6–10 N = 93

65 58, 71b 71.4

71 65.3, 76.0 77.3

66.9 ± 7.0

66.5 ± 8.4

60.0

75.0

67.0 61.3, 74.5 17.2

67.0 58.0, 72.0 43.2

57.1 35.7 7.1 13 11, 14 − 1.4 − 2.4, − 0.9

68.2 22.7 9.1 12 11,13 − 2.0 − 2.3, − 1.3

57.1 37.1 6.0 12.3 ± 1.5

61.0 32.1 7.1 12.2 ± 2.8

− 1.4 − 2.2, − 0.1

− 1.9 − 2.4, − 1.3

52.5 37.5 10.0 13 11, 14 − 1.0 − 1.8, − .0.04d

53.8 32 14.0 12 12, 13 − 2.0 − 2.3, − 1.2e

127 119, 138 66 61, 77 9 (21) 23 (55) 29.1 26.8, 32.5

130 121, 138 66 59, 76 4 (18) 14 (64) 30 29.5, 32.6

128 ± 18

132 ± 20

70 ± 11

69 ± 12

8 (23) 19 (54) 29.9 ± 5.7

10 (36) 16 (57) 31.0 ± 5.4

130 118, 139 68 61, 76 9 (22) 16 (61) 30.1 27.9, 32.8

131 119, 144 71 64, 78 32 (34) 59 (52) 29.8 26.5, 32.5

Abbreviations: BMI, body mass index; BP, blood pressure; CAPACES, Cognition and Physical Activity Community Elderly Study; DASH, Dietary approaches to Stop Hypertension; MMSE, mini-mental state examination. AntiHTN meds= anti-hypertensive medications. aValues represent mean ± s.d. bValues represent median and Interquartile range. cAs defined by the Acculturation Rating Scale for Mexican Americans (ARISMA II) reported by Cuellar et al.18 d,eFor each scoring paradigm, those bearing different superscripts are significantly different from one another. dHypertension is defined as systolic BP ⩾ 140 mm Hg and/or diastolic BP ⩾ 90 mm Hg, or if it was confirmed that they were taking hypertension medications.

accordance strata for each of the three scoring paradigms, nor were there differences in hypertension status. Overall, 53% of the participants reported taking medications for hypertension, but no differences were observed across DASH accordance strata with respect to anti-hypertensive medication use. Dietary intakes The dietary characteristics of the sample as defined by the lowest and highest quartiles of Toledo et al.11 DASH scores are shown in Table 2. Toledo et al.11 DASH scores ranged from 0 to 3.3 with the median and interquartile range for the entire sample (n = 169) being 1.0 (0.5, 2.0) (data not shown). Glycemic load, total calories, fiber and sodium increased with an increase in scores (P o0.001 for all). Total calories differed across quartiles increasing with higher DASH score (P o 0.001), as did dietary fiber. Participants in top quartile of saturated fat intakes ate 8.2 ± 1.8 g per day and these amounts were significantly less than that consumed by those in quartiles 1, 2 or 3 (P = 0.04 for post hoc analysis). For food components, no differences across quartiles were observed, except for sweets servings/day. Dietary intakes of CAPACES participants stratified by Fung et al.12 DASH score quintiles are shown in the middle columns of Table 2, specifically the lowest and top quintiles. Saturated fat intake was significantly different across quintiles, with a gradual decrease in intake with higher DASH scores (P o 0.001). As expected, significant differences in fruit servings, vegetable servings and the nuts, seeds and legumes scoring component as well as energy-adjusted sodium intakes were observed. Mean sodium intake for entire sample was 2614 ± 886 mg day − 1 (data not shown in table) with participants in quintile 5 having an intake of 2997 ± 704 mg day − 1. © 2015 Macmillan Publishers Limited

Dietary and food intakes are shown for two of the three tertiles of Folsom et al.13 DASH scores in last two columns of Table 2. Although there were no differences in energy across tertiles, saturated fat, polyunsaturated fat, monounsaturated fats and trans-fats intakes differed across tertiles (P o0.001), with a gradual decrease in intakes with increasing Folsom et al.13 DASH scores (P o 0.001). DASH scores ranged from 3.5 to 7.5, and no one received a perfect score of 11. Agreement between different DASH scores Several positive relationships were observed between (1) the DASH scores by Fung et al.12 with those according to Toledo et al.11 (r = 0.45, P o 0.001); (2) between scores by Fung et al.12 and those by Folsom et al.13 (r = 0.73, P o 0.001); and (3) between scores by Toledo et al.11 and those by Folsom et al.13 (rs = 0.43, P o0.001).The corresponding kappas are as follows: Fung et al.12 -Toledo et al.11 (Κ = 0.25, P o 0.001), Toledo et al.11 and Folsom et al.13 (Κ = 0.10 P o0.01) and Fung et al.12 and those by Folsom et al..13 (Κ = 0.16; P o0.0001). Diet accordance and blood pressure When examining the ability of the three DASH tools to predict hypertension (SBP of 140 or higher and/or DBP of 90 or higher), only DASH scores according to Folsom et al.13 were predictive of hypertension (odds ratio (OR) = 1.35, 95% confidence interval (CI) (1.04, 1.77), along with acculturation score (OR = 1.59 (1.16, 2.17), P = 0.004 for overall. Variables not included were Fung et al.12 DASH scores, Toledo et al.11 DASH scores, BMI, sex and antihypertensive medications. When we excluded those participants reporting use of anti-hypertensive medications (n = 79), only acculturation (OR = 2.3, 95% CI (1.37-3.94), P = 0.02), sex (OR = 0.20, Journal of Human Hypertension (2015), 1 – 6

Comparison of DASH scores C Tangney et al

4 Table 2.

Comparison of dietary intakes and food group intakes between low and high strata of three DASH scores.

Daily dietary Intakes

Toledo et al. DASH

Fung et al. DASH

Folsom et al. DASH

Low Score 0 n = 42 (25%)

High Score, 3–6 n = 22(13%)

Q1 (8–20) n = 35 (21%)

Q5 (29–38) n = 28 (16%)

Scores 0–5 n = 40 (24%)

Scores 6–10 N = 93 (55%)

116 101, 150b 0 0, 1.1 1662 1455, 2140 21.8 20.1, 21.8 2093 1846, 2667 NA

198 165, 231 0 0, 2.3 2820 2342, 3276 31.6 27.6, 35.7 3488 2909, 4032 NA

143 ± 56c

170 ± 42

3.4 ± 6.8

1.9 ± 4.0

2042 ± 793

2415 ± 588

25.2 ± 7.0

28.2 ± 5.0

2551 ± 971

2997 ± 704

NA

NA

9.9 8.3, 12.1 11 9.8, 11.9 1.3 1.0, 1.6 0.5 0.5, 0.7

8.2 7.1, 9.1 8.3 6.8, 10.2 0.8 0.6, 1.1 0.6 0.5, 0.7

11.1 ± 2.0

8.2 ± 2.4

11.2 ± 2.1

8.2 ± 1.2

1.4 ± 0.4

0.9 ± 0.3

0.6 ± 0.2

0.5 ± 0.1

135 109, 165 0 0, 1.1 1928 1565, 2353 24.2 21.0, 27.7 2412 1974, 2921 33 30.5, 35.3 11.2 10.3, 12.7 11.8 11.1, 12.9 1.5 1.3, 1.7 0.6 0.5, 0.8

135 104, 170 0 0, 1.1 1927 1488, 2413 24.1 20.3, 28.3 2412 1879, 2999 25.5 21.6, 27.1 8.4 7.0, 9.6 8.8 7.3, 9.7 0.9 0.7, 1.1 0.5 0.5, 0.6

Components of DASH Score, servings/day Fruits 2 1.2, 3.1 Vegetables 2.7 1.8, 3.6 Nuts & legumes NA

5.4 3.3, 6.0 4.9 3.5, 6.6 NA

1.9 ± 1.3

4.3 ± 1.4

1.8 ± 0.8

5.3 ± 1.3

NA

NA

0.9 0.4, 1.1 NA 0.5 0.3, 0.7 NA

2.9 2.4, 3.3 NA 1.2 0.9, 1.4 NA

NA

NA

1.1 ± 1.1 NA

2.2 ± 1.0 NA

1.6 0.8, 2.4 2.3 1.6, 2.8 6.0 2.0, 7.5 1.6 0.9, 2.8 NA NA

3.4 2.3, 4.5 4.0 2.7, 5.1 7.0 4.8, 9.8 2.2 1.3, 3.4 NA NA

1.1 ± 0.5

0.7 ± 0.4

NA

NA

NA

NA

0.4 0.1, 0.6 6.8 3.0, 10.8 NA 0.0 0, 0

1.5 1, 1.9 6.2 2.6, 10.3 NA 3.0 3.0, 3.3

0.9 ± 0.9

1.9 ± 1.0

NA

NA

1.0 ± 1.1 17.5 ± 2.2

0.3 ± 0.6 30.4 ±1.4

1.8 1.1, 2.2 2.5 1.7, 3.3 0.6 0.3, 1.0 9.3 6.1, 16.9 NA 4.0 3.5, 4.0

1.6 1.1, 2.1 2.6 2.0, 3.6 1.0 0.6, 1.5 3.5 1.5, 6.5 NA 7.0 6.5, 7.5

Glycemic load, ga Alcohol, ga Energy, kcal Dietary fiber, ga a,d

Sodium, mg

Total fat, % of energyd d

Saturated Fat, % energy MUFA, % energy Trans fat, % energy Omega 3, % energy

Dairy Low-fat dairy Lean meats, poultry, fish Red & processed meats Total grains Whole grains Sweets Sweetened Beverages Total DASH SCORE

Abbreviations: DASH, Dietary approaches to Stop Hypertension; NA, food component or dietary intake not included in this scoring paradigm. aValues adjusted for energy using the residual method of Willett et al.24 bValues represent median and interquartile range. cValues represent mean ± s.d. dDietary intake contributed to only some of the scoring paradigms: Sodium intake was scored as part of the DASH score by Fung et al.,12 and that by Folsom et al. Total fat and saturated fat as % of energy constitutes the DASH score component used by Folsom et al.13

95% CI (0.05–0.71), P = 0.01) and Folsom et al.13 DASH (OR = 1.6, 95% CI (1.01–2.57), P = 0.05) were predictive of hypertension. When DASH scores were regressed on SBP, no DASH score was associated with SBP; acculturation scores were associated with SBP (ß = 2.90 (1.18), adjusted r2 = 0.03, P = 0.02). Similar findings were observed for DBP. DISCUSSION There are few reports of DASH accordance/adherence in Hispanics. In this study, we examined the differences and similarities in the three DASH accordance scores and assessed whether scores were related to hypertension in a sample of older Hispanics. Because there are reports that suggest fewer Hispanics are treated for their hypertension as compared with Journal of Human Hypertension (2015), 1 – 6

Caucasians,2,25 assessment of those lifestyle behaviors known to reduce BP must be performed. It is critical to have tools that are sensitive to the disease condition as well being one that is feasible to engage commitment by the patient. In our cross-sectional sample, acculturation scores differed between accordant and non-accordant groups using the Folsom et al.13 DASH scores (Table 1). It is not clear which component(s) of the Folsom et al.13 paradigm is/are sensitive to acculturation. Interestingly, SBP and DBP were not different between the groups for all three scoring paradigms, nor were there differences in anti-hypertensive medication use across DASH score strata as shown in Table 1. The Hispanic health paradox suggests the original culture may be protective among the new immigrants but with acculturation, that protection is lost.26,27 Higher acculturation © 2015 Macmillan Publishers Limited

Comparison of DASH scores C Tangney et al

(more Anglo-oriented) may lead to higher BMIs among Hispanics.28,29 Among our CAPACES participants, lower acculturation scores (more Hispanic orientation) were observed in those with higher DASH accordance. It is possible that there is an element in the Folsom et al.13 paradigm that captures acculturation levels in this sample. Further testing in larger samples is warranted. The accordance to the DASH eating plan was relatively low for the sample when assessed according to the Fung et al.12 and Toledo et al.11 DASH scores with only 16.5 and 13%, respectively of the sample being classified as DASH accordant. The Toledo et al.11 DASH score ranged from 0 to 3.3 for the entire sample, where a score if 6 is considered perfect accordance. The participants in the ‘high’ category with a score of 3–6 achieved a higher median intake of fruits and vegetables when compared with Fung et al.12 DASH and Folsom et al.13 DASH (Table 2) scores. As described in Table 2, daily servings of fruits, vegetables, nuts, legumes and whole grains increased with increased Fung et al.12 DASH adherence while servings of red and processed meats decreased. Recommendations of fruit and vegetable intake for a 2000 calorie plan is 4–5 servings of fruits and 4–5 servings of vegetables, which is comparable to the quintile 5 of CAPACES sample as they achieved with target for fruits and vegetables. Interestingly, sodium also increased with an increase in Fung et al.12 DASH accordance, which might suggest that this tool is not capturing the sodium intake appropriately for our sample. The DASH plan recommends consuming (for a 2000 calorie diet) 4–5 servings of nuts, seeds and legumes. Median (interquartile range) intake of nuts, and legumes for the sample was 7.0 (4.5, 9.5) when the Folsom et al.13 DASH scores are applied. The Toledo et al.11 DASH scores do not ‘count’ nuts and legumes as part of their DASH score. Thus, what food items are included in these scoring systems may differ markedly. Most of the evidence for DASH accordance among Hispanic Americans relies upon contrast between this ethnic group and others. For example, in the Multi-Ethnic Study of Atherosclerosis cohort, Hispanic participants (n = 415) had significantly higher DASH accordance scores (mean ± s.d., 1.8 ± 1.2) when compared to those of whites (1.5 ± 1.4) and African Americans, 1.4 ± 1.3.9 In the scoring tool used in Multi-Ethnic Study of Atherosclerosis, a score of 8 is perfect DASH accordance; it is based on nutrient analysis (total fat, saturated fat, protein, cholesterol, fiber, potassium, magnesium, and calcium).8 Approximately 44% of the sample was hypertensive and they had a mean acculturation score of 2.4 ± 1.1 (measured on a 5-point scale where 0 is least acculturated and 5 is most acculturated).9 As described in another cross-sectional study involving a diverse sample of older (n = 711, mean age of 66 years) Hispanics and non-Hispanic whites, researchers found that less acculturated Hispanics report lower intakes of simple sugars when compared with more acculturated Hispanics and non-Hispanic whites (P o 0.05 for both groups).30 In the current study, participants who were more acculturated (more Anglicized) had a higher intake of sweets servings/day of 9.3 (6.1, 16.9) (median (interquartile ranges)) as compared with those who were less acculturated with a median intake of 3.5 (1.5, 6.5) (P o0.001). Comparison of the three DASH scoring paradigms Clinicians and nutrition researchers must be able to compare different DASH accordance tools. In contrast to the lack of differences in demographic characteristics across DASH accordance groups for each of the scoring paradigms, there are several differences in actual food and/or nutrient intakes among the three DASH scores. Although all of the research groups who originally developed the three scoring tools used FFQs to measure usual intake of their participants, the FFQs differed considerably. In particular, the number of items that address each food varied. For © 2015 Macmillan Publishers Limited

example, 23 items comprised the fruit group for the Fung et al.12 DASH scores and Folsom et al.13 DASH scores whereas there were only 18 fruit items in the Toledo et al.11 FFQ. In the current study where the CHAP FFQ was used, there were only 11 fruit items that matched those for the Fung et al.12 DASH score and Toledo et al.11 DASH scores, but there were 13 items for Folsom et al.13 DASH scores. Another difference between scoring paradigms was the choice of food and nutrients to score DASH accordance. Folsom et al.13 included meats, poultry and fish as a single component, whereas Fung et al.12 included red and processed meats as a component and do not include fish or poultry as score components. The categories/subcomponents common to both Fung et al.12 DASH scores and Folsom et al.13 DASH scores are fruits, vegetables, whole grains and sodium. Thus, it is quite possible that intake differences may be attributable to the differences in scoring as well as in the foods included in the key food groups. For example, Folsom and others13 include corn bread, popcorn, cold cereal, oatmeal, dark bread and others grains like bulgur as whole grains. Fung et al. choose to include cooked cereal, brown rice, dark bread and whole grain cereal as whole grain foods. To the best of our knowledge there are no other studies that compare DASH scoring paradigms with one another among Hispanics. There are many differences in the individual components of the DASH scoring paradigms. For example, sodium was not a part of the original DASH trial3 but Fung et al.12 and Folsom and others13 chose to include sodium as part of their scoring components. DASH scores as a predictor of prevalent hypertension We found that Folsom et al.13 DASH scores and acculturation scores were predictive of hypertension; however, the Folsom et al.13 DASH scores explains only 3% of the variance, which may not be clinically significant. In a secondary analysis where DASH scores were regressed to predict prevalent hypertension in participants who were not on medications, age, acculturation and Folsom et al.13 DASH scores were once again predictive. This secondary analysis was done because only 68.7% of Hispanics report taking their anti-hypertensive medications as compared to the national average of 79.2% in 2009.25 There may be several reasons why the Folsom et al.13 scores were predictive of hypertension. It is possible that it is because this score includes calories from fat, saturated fat in the scoring paradigm. The saturated fat score is positively correlated with Hispanic orientation (rs = 0.18, P = 0.02) and the total fat score was correlated with the acculturation score (rs = − 0.23, P = 0.003). The effect sizes of these correlations are relatively small, however. This score also differentiates whole grains from total grains and awards points for both food groups and perhaps, this feature is important. This sample of Hispanics had a median acculturation score of − 1.8 (−2.3, − 0.9) and thus, mostly Latino/Hispanic oriented. This sample did not meet targets but had relatively good fruit and vegetable intakes (6.4 ± 3.3 servings per day). This is consistent with studies that report that Hispanics who are less acculturated tends to have diets that are higher in fruits and vegetables.26,27 These scores may be helpful for clinicians to assess DASH accordance because they are easy to use. The Toledo et al.11 DASH scores are likely the easiest to use because the tool just scores food groups (total score of 1–6). Limitations One of the limitations of this study is the cross-sectional design; scores are used to predict hypertension at one point in time. We cannot infer that continued accordance to the DASH plan, as defined by Folsom et al.13 may improve hypertension status. A further limitation of this study is the small sample size of 169 participants. Because only older Hispanic adults were studied, the Journal of Human Hypertension (2015), 1 – 6

5

Comparison of DASH scores C Tangney et al

6 observed comparisons may not be generalizable to other groups. While the primary objective of this effort was to identify which of the three DASH scoring paradigms would be most useful/ predictive of prevalent hypertension in this sample of adults, it would be of interest to apply these DASH scores to other groups such as African Americans, Caucasians and Asians. CONCLUSIONS The DASH score by Folsom et al.13 was predictive of hypertension status in this sample of older Hispanics; acculturation was also important. There are numerous differences among the DASH scores and it is possible that these scores are not defining the intake of our Hispanic sample appropriately because the original DASH scoring paradigm were not created for use in Hispanics. These present findings suggest that the use of DASH score by Folsom et al.13 needs to be tested in a larger, more age-diverse Hispanic sample. What is known about the topic ● These are extensive evidence for the role that the DASH diet pattern plays in lowering blood pressure, but the challenge is facilitating dietary behavior change. ● Clinicians need a quick tool to assess how well their patients are doing with respect to such a diet plan. ● There are at least 10 different ways to score adherence or accordance to the DASH pattern. What this study adds ● In this effort we selected 3 common tools and described the level of DASH accordance using each tool for a community sample of Hispanic older adults. ● Of the three tools, only one was associated with prevalent hypertension.

CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGEMENTS We want to thank all the participants of CAPACES for their time. This study has been supported by the Rush University Medical Center Translational Sciences Consortium.

REFERENCES 1 Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ et al. Executive Summary. Heart Disease and Stroke Statistics—2014 Update. Areport from the American Heart Association. Circulation 2014; 129(3): 399–410. 2 Gu Q, Burt V, Dillon C, Yoon S. Trends in antihypertensive medication use and blood pressure control among United States adults with hypertension: the National Health and Nutrition Examination Survey, 2001 to 2010. Circulation 2012; 126(17): 2105–2114. 3 Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM et al. A clinical trial of the effects of dietary patterns on blood pressure: DASH collaborative research group. N Engl J Med 1997; 336(16): 1117–1124. 4 Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D et al. Effects on blood pressure of reduced dietary sodium and the dietary approaches to stop hypertension (DASH) diet. DASH-sodium collaborative research group. N Engl J Med 2001; 344(1): 3–10. 5 Obarzanek E, Vollmer W, Lin P, Cooper L, Young D, Ard J et al. Effects of individual components of multiple behavior changes: The PREMIER trial. Am J Health Behav 2007; 31(5): 545–560. 6 Blumenthal J, Babyak M, Hinderliter A, Watkins L, Craighead L, Lin P et al. Effects of the DASH diet alone and in combination with exercise and weight loss on blood pressure and cardiovascular biomarkers in men and women with high blood pressure: The ENCORE study. Arch Intern Med 2010; 170(2): 126–135.

Journal of Human Hypertension (2015), 1 – 6

7 Eckel RH, Jakicic JM, Ard JD, de Jesus JM, Houston Miller N, Hubbard VS et al. 2013 AHA/ACC Guideline on Lifestyle Management to Reduce Cardiovascular Risk. Circulation 2014; 129: S76–S99. 8 Gao S, Fitzpatrick A, Psaty B, Jiang R, Post W, Cutler J et al. Suboptimal nutritional intake for hypertension control in 4 ethnic groups. Arch Intern Med 2009; 169(7): 702–707. 9 Mujahid M, Diez Roux A, Cooper R, Shea S, Williams D. Neighborhood stressors and race/ethnic differences in hypertension prevalence (the multi-ethnic study of Atherosclerosis). Am J Hypertens 2011; 24(2): 187–193. 10 Mellen PB, Gao SK, Vitolins MZ, Goff DC Jr. Deteriorating dietary habits among adults with hypertension: DASH dietary accordance, NHANES 1988-1994 and 1999-2004. Arch Intern Med 2008; 168(3): 308–314. 11 Toledo E, de A Carmona-Torre F, Alonso A, Puchau B, Zulet M, Martinez JA et al. Hypothesis-oriented food patterns and incidence of hypertension: 6-year followup of the SUN (Seguimiento Universidad de Navarra) prospective cohort. Public Health Nutr 2010; 13(3): 338–349. 12 Fung T, Chiuve S, McCullough M, Rexrode K, Logroscino G, Hu F. Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women. Arch Intern Med 2008; 168(7): 713–720. 13 Folsom A, Parker E, Harnack L. Degree of concordance with DASH diet guidelines and incidence of hypertension and fatal cardiovascular disease. Am J Hypertens 2007; 20(3): 225–232. 14 Günther ALB, Liese A, Bell R, Dabelea D, Lawrence J, Rodriguez B et al. Association between the dietary approaches to hypertension diet and hypertension in youth with diabetes mellitus. Hypertension 2009; 53(1): 6–12. 15 Staffileno B, Tangney C, Wilbur J, Marquez D, Fogg L, Manning A et al. Dietary approaches to stop hypertension patterns in older Latinos with or at risk for hypertension. J Cardiovasc Nurs 2013; 28(4): 338–347. 16 Wilbur J, Marquez DX, Fogg L, Wilson RS, Staffileno BA, Hoyem RL et al. The relationship between physical activity and cognition in older Latinos. J Gerontol B-Psychol Sci Soc Sci 2012; 67(5): 525–534. 17 Pickering T, Hall J, Appel L, Falkner B, Graves J, Hill M et al. Recommendations for blood pressure measurement in humans and experimental animals: Part 1: Blood pressure measurement in humans: a statement for professionals from the subcommittee of professional and public education of the American Heart Association Council on High Blood Pressure Research. Circulation 2005; 111(5): 697–716. 18 Cuellar I, Arnold B, Maldonado R. Acculturation rating scale for Mexican Americans-II: A revision of the original ARISMA scale. Hispanic J Behav Sci 1995; 17(3): 275–304. 19 Rockett HR, Wolf AM, Colditz GA. Development and reproducibility of a food frequency questionnaire to assess diets of older children and adolescents. J Am Diet Assoc 1995; 95: 336–340. 20 Morris MC, Colditz GA, Evans DA. Response to a mail nutritional survey in an older bi-racial community population. Ann Epidemiol 1998; 8: 342–346. 21 Morris M, Tangney C, Bienias J, Evans D, Wilson R. Validity and reproducibility of a food frequency questionnaire by cognition in an older biracial sample. Am J Epidemiol 2003; 158(12): 1213–1217. 22 Tangney C, Bienias J, Evans D, Morris M. Reasonable estimates of serum vitamin E, vitamin C, and beta-cryptoxanthin are obtained with a food frequency questionnaire in older black and white adults. J Nutr 2004; 134(4): 927–934. 23 Epstein DE, Sherwood A, Smith PJ, Craighead L, Caccia C, Lin PH et al. Determinants and consequences of adherence to the dietary approaches to stop hypertension diet in African-American and white adults with high blood pressure: results from the ENCORE trial. J Acad Nutr Diet 2012; 112(11): 1763–1773. 24 Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr 1997; 65(4 Suppl): 1220S–1228S. 25 Liao Y, Bang D, Cosgrove S, Dulin R, Harris Z, Taylor A et al. Surveillance of health status in minority communities - Racial and Ethnic Approaches to Community Health across the U.S. (REACH U.S.) risk factor survey, United States, 2009. MMWR Surveill Summ 2011; 60(6): 1–44. 26 Morales L, Lara M, Kington R, Valdez R, Escarce J. Socioeconomic, cultural, and behavioral factors affecting Hispanic health outcomes. J Health Care Poor Underserved 2002; 13(4): 477–503. 27 Barcenas C, Wilkinson A, Strom S, Cao Y, Saunders K, Mahabir S et al. Birthplace, years of residence in the United States, and obesity among MexicanAmerican adults. Obesity 2007; 15(4): 1043–1052. 28 Popkin BM, Udry JR. Adolescent obesity increases significantly in second and third generation U.S. immigrants: The national longitudinal study of adolescent health. J Nutr 1998; 128(4): 701–706. 29 Yeh M, Viladrich A, Bruning N, Roye C. Determinants of Latina obesity in the United States: The role of selective acculturation. J Transcult Nurs 2009; 20(1): 105–115. 30 Bermudez OI, Falcon LM, Tucker KL. Intake and food sources of macronutrients among older Hispanic adults: association with ethnicity, acculturation, and length of residence in the United States. J Am Diet Assoc. 2000; 100(6): 665–673.

© 2015 Macmillan Publishers Limited