Short Communication | Published:

Dairy intake and changes in blood pressure over 9 years: the ARIC study

European Journal of Clinical Nutrition volume 63, pages 12721275 (2009) | Download Citation


Dairy product intake could contribute to preventing hypertension, but information linking intake of these foods with changes in blood pressure over long periods of time, particularly in non-whites, is scarce. We analyzed the Atherosclerosis Risk in Communities study, a prospective cohort in the United States, to assess whether different types of dairy products were associated with changes in blood pressure over time. The analysis included 6912 white and 1296 African-American nonhypertensive men and women, aged 45–64 at baseline. After 9 years of follow-up, systolic blood pressure of whites consuming three or more daily servings of low-fat milk increased 2.7 mm Hg less than in those consuming less than one serving per week (P for trend=0.01). Among African Americans, dairy products intake was not associated with changes in blood pressure over time. In conclusion, higher low-fat milk intake was associated with lower increases in blood pressure in whites but not in African Americans.


An elevated intake of dairy products has been repeatedly associated with reduced risk of high blood pressure (BP) (Pereira et al., 2002; Alonso et al., 2005; Wang et al., 2008; Toledo et al., 2009). However, information on the relation between these foods and BP levels over long periods of time, particularly in non-whites, is limited. We assessed the association between intake of dairy products and changes in BP over 9 years in the Atherosclerosis Risk in Communities (ARIC) study, a biracial cohort in the United States (The ARIC Investigators, 1989).

The ARIC study recruited a population-based sample of 15 792 men and women (27% African American), aged 45–64, from four US communities in 1987–1989. At baseline, participants underwent a physical examination, including measurements of BP and a dietary assessment. Sitting BP was measured by trained technicians after a 5 min rest using a random zero sphygmomanometer. Systolic and diastolic BP were calculated as the average of the second and third of three consecutive measurements. Diet was assessed using a 66-item food-frequency questionnaire based on the Willett 61-item questionnaire (Willett et al., 1985). Six items corresponded to dairy products (low-fat milk, whole-fat milk, yoghurt, ice cream, ricotta or cottage cheese, and other cheese). A previous study found moderate correlations between dairy product intake assessed by the food-frequency questionnaire and a diet record (up to 0.62 and 0.81 for whole-fat milk and low-fat milk, respectively) (Salvini et al., 1989). ARIC participants were examined three more times during the follow-up at approximately 3-year intervals. Response rates were 93, 86 and 80% at the follow-up exams. BP was measured in each exam using the same methodology.

For this analysis, we excluded participants who had diabetes, hypertension (systolic BP140, diastolic BP90 or using antihypertensive medication), or cardiovascular disease at baseline, and those with missing variables for any baseline covariate or implausible energy intakes (<500 or >3500 kcal per day in women, <700 or >4500 kcal per day in men). Table 1 includes selected characteristics of the 8208 eligible participants at baseline.

Table 1: Selected characteristics of non-hypertensive study participants at baseline, ARIC study (1987–1989)

Associations between intake of dairy products and changes in BP over time were assessed using general linear models for repeated measures (Proc Mixed in the SAS software, SAS Institute Inc., Cary, NC, USA), with BP at each visit as the dependent variable, running separate models for whites and African Americans. We included the following independent variables: age, sex, study center, body mass index, waist-to-hip ratio, physical activity, smoking, and intake of total energy, alcohol, sodium, potassium, fruits and vegetables, dairy products, study visit, and an interaction term between study visit and dairy intake. The coefficient for this interaction term estimated the average change in BP from visit to visit (3-year intervals) by category of dairy intake. Use of antihypertensive medication at the follow-up visits was taken into account by adding a constant to BP values at the relevant visits (10 mm Hg for systolic BP, 7 mm Hg for diastolic BP) as recommended elsewhere (Tobin et al., 2005). Other approaches to deal with antihypertensive medication use, such as introducing an indicator variable in multivariable analysis or excluding antihypertensive medication users, could cause bias (Tobin et al., 2005). Prevalence of antihypertensive medication use in successive follow-up visits was 4.5, 10.0 and 17.0%.

Table 2 reports average changes in systolic and diastolic BP between study visits across categories of total dairy, low-fat milk and whole-fat milk intake, for the entire sample, and separately for whites and African Americans. Only low-fat milk was inversely associated with systolic BP longitudinally, and this association was restricted to whites. Systolic BP between visits 1 and 4 (9 years) increased 2.7 mm Hg (95% CI −0.3, 6.0) less in whites consuming three or more daily servings of low-fat milk than in those consuming less than one serving per week (P for trend=0.01). Neither intake of whole-fat milk in whites nor any dairy food in African Americans was associated with BP changes over time (Table 2). Intake of other individual dairy products did not have enough variability in the population to permit study of associations with BP change (data not shown). The same pattern of results was observed when we used different values for the effect of antihypertensive medications, when we ignored BP measurements from visit 1, or when we included individuals with prevalent diabetes or cardiovascular disease. We observed similar associations when incidence of hypertension and not BP change was the main outcome variable (data not shown).

Table 2: Changes in systolic and diastolic BP per 3 years (period between study visits) by categories of servings of total dairy and specific dairy products (ARIC study, 1987–1999)a

Our study has valuable strengths. The ARIC cohort includes a racially diverse, representative sample of four different communities in the United States, increasing the generalizability of our results. BP was measured repeatedly in each participant, reducing within-person variability. The results were robust to different analytical assumptions. In addition, information on other risk factors for elevated BP, including other dietary factors, was available, allowing control of confounding.

Weaknesses of the present study include the limited range of exposure to some dairy products, particularly in African Americans; the suboptimal sample size in the race-specific analysis; nondifferential measurement error in the dietary assessment, which could obscure existing associations; potential for selection bias due to selective losses to follow-up and presence of unmeasured confounding that could explain the observed inverse associations. These limitations suggest that our results should be interpreted cautiously.

Intake of dairy foods, particularly low-fat products, has been consistently associated with lower BP levels and reduced risk of hypertension in observational studies (Pereira et al., 2002; Alonso et al., 2005; Wang et al., 2008; Toledo et al., 2009). In addition, the Dietary Approaches to Stop Hypertension trial showed that a dietary pattern rich in fruits and vegetables, low-fat dairy products and low in total fat was more effective than a control diet or fruits and vegetables alone in reducing BP levels (Appel et al., 1997). This effect was similar in whites and African Americans (Appel et al., 1997). Less information is available, however, from epidemiologic studies including non-white populations. To date, most studies addressing the relationship between dairy products intake and the risk of hypertension have been limited to samples of Europeans or European-descent individuals (Alonso et al., 2005; Wang et al., 2008; Toledo et al., 2009). In the CARDIA study, which included a sample of 3157 blacks and whites from the United States, aged 18–30, and followed up for 10 years, dairy products intake (particularly low-fat dairy products) was associated with lower incidence of metabolic syndrome, including high BP, in both racial groups (Pereira et al., 2002). The different results in CARDIA and ARIC could be consequence of the different age structure of the population (younger in CARDIA) or differences in dietary assessment. A major limitation in addressing the association between dairy products intake and health outcomes is the high prevalence of lactose intolerance in African-Americans adults and other non-white racial groups (Jackson and Savaiano, 2001), reducing their range of dairy products intake and, therefore, limiting the study of statistical associations. In addition, the present study included only a limited number of African Americans.

In conclusion, our results offer some evidence that higher low-fat milk intake might prevent BP elevations associated with age, at least in whites. Though the differences in BP change across categories of low-fat milk intake were small, these results are meaningful from a public health perspective: even average reductions of 2–3 mm Hg in average systolic BP could lead to dramatic drops in the incidence of cardiovascular disease at a population level (Appel et al., 2006). Further research in African Americans, selecting larger populations and paying attention to the role of lactose intolerance on dairy consumption, is required.

Conflict of interest

The authors declare no conflict of interest.


  1. , , , , (2005). Low-fat dairy consumption and reduced risk of hypertension: the Seguimiento Universidad de Navarra (SUN) cohort. Am J Clin Nutr 82, 972–979.

  2. , , , , , (2006). Dietary approaches to prevent and treat hypertension: a scientific statement from the American Heart Association. Hypertension 47, 296–308.

  3. , , , , , et al. (1997). A clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med 336, 1117–1124.

  4. , (2001). Lactose maldigestion, calcium intake and osteoporosis in African-, Asian-, and Hispanic-Americans. J Am Coll Nutr 20, 198S–1207S.

  5. , , , , , (2002). Dairy consumption, obesity, and the insulin resistance syndrome in young adults. The CARDIA Study. JAMA 287, 2081–2089.

  6. , , , , , et al. (1989). Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int J Epidemiol 18, 858–867.

  7. The ARIC Investigators (1989). The Atherosclerosis Risk in Communities (ARIC) study: design and objectives. Am J Epidemiol 129, 687–702.

  8. , , , (2005). Adjusting for treatment effects in studies of quantitative traits: antihypertensive therapy and systolic blood pressure. Stat Med 24, 2911–2935.

  9. , , , , , et al. (2009). Low-fat dairy products and blood pressure: follow-up of 2290 older persons at high cardiovascular risk participating in the PREDIMED study. Br J Nutr 101, 59–67.

  10. , , , , (2008). Dietary intake of dairy products, calcium, and vitamin D and the risk of hypertension in middle-aged and older women. Hypertension 51, 1–7.

  11. , , , , , et al. (1985). Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 122, 51–65.

Download references


The Atherosclerosis Risk in Communities study is carried out as a collaborative study supported by National Heart, Lung, and Blood Institute contracts N01-HC-55015, N01-HC-55016, N01-HC-55018, N01-HC-55019, N01-HC-55020, N01-HC-55021 and N01-HC-55022. We thank the staff and participants of the ARIC study for their important contributions.

Author information


  1. Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, MN, USA

    • A Alonso
    • , L M Steffen
    •  & A R Folsom


  1. Search for A Alonso in:

  2. Search for L M Steffen in:

  3. Search for A R Folsom in:

Corresponding author

Correspondence to A Alonso.

About this article

Publication history






Further reading