Vol. 20, No. 4 Printed in Great Britain
International Journal of Epidemiology © International Epidemiological Association 1991
Nutrient Intakes and Blood Pressure in Normotensive Males ALBERTO ASCHERIO*, MEIR J STAMPFER***. GRAHAM A COLDITZ***. WALTER C WILLETT**»t AND JOHN McKINLAYt
intake of an individual.22 Furthermore, some of these studies may have included hypertensive subjects who changed their diet after the diagnosis. In the present cross-sectional study, we examine nutrient intake in relation to blood pressure in a community-based sample of US males.
International comparisons,' and results from studies of migrants2 and religious groups,3-4 suggest that differences in diet and body habitus are important determinants of variability in blood pressure. While alcohol intake5'6 and obesity7-8 have consistently been shown to be associated with elevated blood pressure, the role of other dietary factors is controversial. In a recent multicentre study large variations in salt intake had only a minor relation with blood pressure levels,9 but re-analysis of observational,10'" and experimental12 data have suggested a stronger association of salt intake with blood pressure. Dietary potassium has been inversely related to blood pressure in some13"15 but not all8-16 cross-sectional studies, and supplementation trials have yielded conflicting results.17-18 Evidence on the protective effects of calcium13-15'19 magnesium,20 and fibre,21 is still inadequate. Most epidemiological studies investigating the association between nutritional factors and blood pressure measured nutrient intakes by 24-hour recall8'415-21 which, because of intra-individual variability, is generally a poor reflection of the usual
METHODS Between 1987 and 1989 a cross-sectional survey of men aged 40-69 years was conducted in the area of Boston, Massachusetts, as part of a study on health and aging.23 A letter of invitation to participate in the study was sent to a sample of 2300 men, randomly selected from the computerized Massachusetts city and town street lists. An appointment was arranged by telephone, and respondents were interviewed at home, between 8 and 10 am. A first blood pressure measurement was taken (after the subject had been sitting quietly for five minutes) from the right arm positioned at heart level (with the kitchen or dining table as support) using a standard portable mercury manometer and following the Pawtucket Heart Health Program (PHHP) protocol.24 A second measurement was taken in the same way at the end of the interview. Height was measured with a hinged, wooden set square and wooden carpenter's rule and weight with a portable battery operated scale, following the National Center for Health Statistics (NCHS) and PHHP protocol.24 Hip and waist circumferences were taken by a technician following the Health and Nutrition Examination
•Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston MA 02115, USA. ••Charming Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston MA, USA. t Department of Nutrition, Harvard School of Public Health, Boston MA, USA. {New England Research Institute, Watertown MA, USA.
886
Downloaded from http://ije.oxfordjournals.org/ at University of California, San Francisco on December 13, 2014
Ascherio A (Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston MA 02115, USA), Stampfer M J , Colditz G A, Willett W C and McKinlay J . Nutrient intakes and blood pressure in normotensive males. International Journal of Epidemiology 1991; 20: 886-891. The association of dietary intakes with blood pressure levels was examined in a cross-sectional sample of 805 men aged 40-69 years free from clinical hypertension, coronary heart disease or cancer. After controlling for age, body habitus, and alcohol consumption, blood pressure levels were inversely associated with the intake of fruit (r= —0.09, p
International Journal of Epidemiology © International Epidemiological Association 1991
Nutrient Intakes and Blood Pressure in Normotensive Males ALBERTO ASCHERIO*, MEIR J STAMPFER***. GRAHAM A COLDITZ***. WALTER C WILLETT**»t AND JOHN McKINLAYt
intake of an individual.22 Furthermore, some of these studies may have included hypertensive subjects who changed their diet after the diagnosis. In the present cross-sectional study, we examine nutrient intake in relation to blood pressure in a community-based sample of US males.
International comparisons,' and results from studies of migrants2 and religious groups,3-4 suggest that differences in diet and body habitus are important determinants of variability in blood pressure. While alcohol intake5'6 and obesity7-8 have consistently been shown to be associated with elevated blood pressure, the role of other dietary factors is controversial. In a recent multicentre study large variations in salt intake had only a minor relation with blood pressure levels,9 but re-analysis of observational,10'" and experimental12 data have suggested a stronger association of salt intake with blood pressure. Dietary potassium has been inversely related to blood pressure in some13"15 but not all8-16 cross-sectional studies, and supplementation trials have yielded conflicting results.17-18 Evidence on the protective effects of calcium13-15'19 magnesium,20 and fibre,21 is still inadequate. Most epidemiological studies investigating the association between nutritional factors and blood pressure measured nutrient intakes by 24-hour recall8'415-21 which, because of intra-individual variability, is generally a poor reflection of the usual
METHODS Between 1987 and 1989 a cross-sectional survey of men aged 40-69 years was conducted in the area of Boston, Massachusetts, as part of a study on health and aging.23 A letter of invitation to participate in the study was sent to a sample of 2300 men, randomly selected from the computerized Massachusetts city and town street lists. An appointment was arranged by telephone, and respondents were interviewed at home, between 8 and 10 am. A first blood pressure measurement was taken (after the subject had been sitting quietly for five minutes) from the right arm positioned at heart level (with the kitchen or dining table as support) using a standard portable mercury manometer and following the Pawtucket Heart Health Program (PHHP) protocol.24 A second measurement was taken in the same way at the end of the interview. Height was measured with a hinged, wooden set square and wooden carpenter's rule and weight with a portable battery operated scale, following the National Center for Health Statistics (NCHS) and PHHP protocol.24 Hip and waist circumferences were taken by a technician following the Health and Nutrition Examination
•Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston MA 02115, USA. ••Charming Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston MA, USA. t Department of Nutrition, Harvard School of Public Health, Boston MA, USA. {New England Research Institute, Watertown MA, USA.
886
Downloaded from http://ije.oxfordjournals.org/ at University of California, San Francisco on December 13, 2014
Ascherio A (Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston MA 02115, USA), Stampfer M J , Colditz G A, Willett W C and McKinlay J . Nutrient intakes and blood pressure in normotensive males. International Journal of Epidemiology 1991; 20: 886-891. The association of dietary intakes with blood pressure levels was examined in a cross-sectional sample of 805 men aged 40-69 years free from clinical hypertension, coronary heart disease or cancer. After controlling for age, body habitus, and alcohol consumption, blood pressure levels were inversely associated with the intake of fruit (r= —0.09, p
