INTERMAP - Nature

Journal of Human Hypertension (2003) 17, 591–608 & 2003 Nature Publishing Group All rights reserved 0950-9240/03 $25.00 www.nature.com/jhh

ORIGINAL ARTICLE

INTERMAP: background, aims, design, methods, and descriptive statistics (nondietary) J Stamler1*, P Elliott2*, B Dennis3*, AR Dyer1*, H Kesteloot4*, K Liu1*, H Ueshima5* and BF Zhou6* for the INTERMAP Research Group 1

Department of Preventive Medicine, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA; 2Department of Epidemiology and Public Health, Faculty of Medicine, St Mary’s Campus, Imperial College, London, UK; 3Department of Biostatistics, Collaborative Studies Coordinating Center, University of North Carolina, Chapel Hill, NC, USA; 4Central Laboratory, Akademisch Ziekenhuis St Rafael, Leuven, Belgium; 5Department of Health Science, Shiga University of Medical Science, Otsu, Japan; 6Department of Epidemiology, Fu Wai Hospital and Cardiovascular Institute, Chinese Academy of Medical Sciences, Beijing, People’s Republic of China

Blood pressure (BP) above optimal (p120/p80 mmHg) is established as a major cardiovascular disease (CVD) risk factor. Prevalence of adverse BP is high in most adult populations; until recently research has been sparse on reasons for this. Since the 1980s, epidemiologic studies confirmed that salt, alcohol intake, and body mass relate directly to BP; dietary potassium, inversely. Several other nutrients also probably influence BP. The DASH feeding trials demonstrated that with the multiple modifications in the DASH combination diet, SBP/DBP (SBP: systolic blood pressure, DBP: diastolic blood pressure) was sizably reduced, independent of calorie balance, alcohol intake, and BP reduction with decreased dietary salt. A key challenge for research is to elucidate specific nutrients accounting for this effect. The general aim of the study was to clarify influences of multiple nutrients on SBP/DBP of individuals over and above effects of Na, K, alcohol, and body mass. Specific aims were, in a cross-sectional epidemiologic study of 4680 men and women aged 40–59 years from 17 diverse population samples in China, Japan, UK, and USA, test 10 prior hypotheses on relations of macronutrients to SBP/DBP and on role of dietary factors in inverse associations of education with

BP; test four related subgroup hypotheses; explore associations with SBP/DBP of multiple other nutrients, urinary metabolites, and foods. For these purposes, for all 4680 participants, with standardized high-quality methods, assess individual intake of 76 nutrients from four 24-h dietary recalls/person; measure in two timed 24-h urine collections/person 24-h excretion of Na, K, Ca, Mg, creatinine, amino acids; microalbuminuria; multiple nutrients and metabolites by nuclear magnetic resonance and high-pressure liquid chromatography. Based on eight SBP/DBP measurements/person, and data on multiple possible confounders, utilize mainly multiple linear regression and quantile analyses to test prior hypotheses and explore relations of multiple dietary and urinary variables to SBP/DBP of individuals. The 4680 INTERMAP participants are equally divided across four age/gender strata: diverse in ethnicity, education, occupation, physical activity; use of cigarettes, alcohol; diagnosed high BP, CVD, diabetes; CVD family history; women vary in parity, use of contraceptive medication and hormone replacement therapy. Journal of Human Hypertension (2003) 17, 591–608. doi:10.1038/sj.jhh.1001603

Keywords: population study; nutrients and blood pressure; diet and blood pressure; international epidemiologic

research

Background Population-wide adverse blood pressure and serum cholesterol levels

Correspondence: Dr J Stamler, Department of Preventive Medicine, Feinberg School of Medicine, Northwestern University, Suite 1102(D335), 680 North Lake Shore Drive, Chicago, IL, USA. E-mail: [email protected] *Member, INTERMAP Steering and Editorial Committee.

Blood pressure (BP) is an established major risk factor for coronary heart disease (CHD), stroke, all cardiovascular diseases (CVD), end-stage renal disease (ESRD), and impaired longevity for adult men and women of all ethnic, socioeconomic, and geographic backgrounds. Its relations to risk are

INTERMAP J Stamler et al 592

continuous, graded (curvilinear), strong, consistent, independent, predictive, and aetiologically significant. For many populations worldwide, adverse BP levels are the rule from age 35 years on and optimal levels (p120/p80 mmHg) the rare exception, due to rise in SBP/DBP (SBP ¼ systolic blood pressure, DBP ¼ diastolic blood pressure) from youth through middle age. Inordinate prevalence rates of highnormal BP (X130 SBP and/or X85 DBP) and high BP (X140 SBP and/or X90 DBP) produce large increases in CHD–CVD relative risk, absolute risk, and absolute excess risk.1–4 In all these respects, BP closely resembles serum cholesterol, also an established major CHD–CVD risk factor (Table 1).5,6

Influences of diet on blood pressure and on serum cholesterol

In regard to serum cholesterol, such findings decades ago stimulated extensive research on lifestyle factors influencing this trait, particularly nutritional factors. By the 1960s, it was clear that several dietary components had sizable effects: higher intake of cholesterol and saturated fatty acids (SFA) raises serum cholesterol on average; calorie imbalance (on ‘Western’ fare) with weight gain also raises serum cholesterol; in contrast, higher intake of polyunsaturated fatty acids (PFA) and watersoluble fibre have moderate cholesterol-lowering effects, as does weight loss by obese people

consuming fat-modified fare.7–11 Extensive evidence demonstrating these dietary influences has been from the early 1960s on the scientific foundation for repeated recommendations by expert groups that the general population improve eating patterns to lower serum cholesterol.12–19 More recently, data on adverse effects of dietary trans fatty acids on serum cholesterol led to expansion of recommendations to encompass avoidance of foods with these components.11,13,18,19 Broad sectors of the population of ‘Western’ countries have acted on these advisories by modifying their eating patterns accordingly, with resultant sizable declines in adult average serum cholesterol levels.20 It is a reasonable inference that these trends contributedFalong with reduced prevalence of cigarette smoking and treatment of hypertensionFto sizable declines in CHD–CVD mortality. In contrast to the extensive research effort on nutrition influencing serum cholesterol, little investigative work was carried out until the 1980s on dietary factors impacting blood pressure. During the first half of the 20th century, clinical studies in France and the USA showed that marked reduction in salt intake lowered SBP/DBP of people with severe hypertension.21 Confirmatory evidence on the salt–BP relationship soon emerged from animalexperimental studies. Clinical and epidemiologic data showed a direct relationship of body mass and of alcohol intakeFparticularly heavy drinkingFto BP.22–24 It was also reported that vegetarian peoples

Table 1 Relationship of SBP and of serum cholesterol, considered singly, to risk of CHD death, 72 144 men baseline aged 35–39 years screened for the Multiple Risk Factor Intervention Trial (MRFIT) Baseline level of risk factor

Number of CHD deaths

Death rate per 10 000 p-y

Multivariate adjusted hazard ratio (HR)a

Systolic blood pressure (mmHg) p120 121–129 130–139 140–159 X160 Cox multivariate adjusted coefficient and HR, 1 s.d. higherb

25 589 19 581 16 050 9645 1279 F

4.6 6.6 9.5 13.7 27.4 F

1.00 1.33** 1.81*** 2.36*** 4.17*** 0.0235*** 1.37

Serum cholesterol (mg/dl) o160 160–179 180–199 200–219 220–239 240–259 260–279 X280 o200 200–239 X240 Cox multivariate adjusted coefficient and HR, 1 s.d. higherb

6847 11 548 15 098 15 228 10 854 6447 3525 2597 33 493 26 082 12 569 F

3.0 3.8 5.3 6.7 9.7 12.2 16.5 30.2 4.3 7.9 17.1 F

1.00 1.20 1.57w 1.92** 2.58*** 3.11*** 3.97*** 6.70*** 1.00 1.64*** 3.07*** 0.0094*** 1.44

a

Hazard ratio adjusted for baseline age, race, serum cholesterol or SBP, and cigarettes smoked per day. 1 s.d. higher: for SBPF13.3 mmHg; for serum cholesterolF38.8 mg/dl. w Po0.10, **Po0.01, ***Po0.001. p-y=person-years. b

Journal of Human Hypertension


had lower SBP/DBP than omnivorous ‘Westerners’, but these papers lacked data on specific ingested nutrients.25 In the late 1970s and early 1980s, Japanese researchers published data from population studies and animal experiments suggesting an inverse relation of dietary protein (particularly animal protein) to BP26–29Fin contrast to judgements of ‘Western’ researchers who hypothesized a direct association of protein with BP, or concluded in reviews that the limited available data indicated no relationship.30,31

INTERSALT Neglect of the nutrition–BP issue began to change in the latter 1980s, in part as a result of the INTERSALT Study. The history of INTERSALT was set down in a special number of this journal 15 years ago, along with detailed data from the study.32 In brief, INTERSALT was an international cooperative cross-sectional epidemiologic study. Its main original aim was to testFin a large standardized highquality international epidemiologic studyFtwo sets of prior hypotheses on the relation of dietary sodium (also potassium), to SBP and DBP in 10 000+ men and women aged 20–59 years from 52 diverse population samples in 32 countries worldwide. The two sets of prior hypotheses were: (1) for individuals (ie within-populationFN ¼ 10 000+), and (2) for samples (ie cross-population or ecologicFN ¼ 52).21,32 The specific prior hypotheses for individuals were: For the 10 000+ participants, SBP and DBP would be directly and independently related to 24-h urinary sodium excretion, also to Na/ K excretion, body mass index (BMI), and alcohol use, and inversely related to 24-h potassium excretion. The specific cross-population (ecologic) hypotheses were: For the 52 population samples, five blood pressure end points would be directly related to sample median 24-h sodium and Na/K excretion,

sample median BMI, and sample alcohol intake, and inversely related to sample median 24-h potassium excretion; the five BP variables would be sample median SBP, sample median DBP, sample slope of SBP with age (20–59 years), sample slope of DBP with age, and sample prevalence of high BP (SBP X140 and/or DBP X90 mmHg, or receiving antihypertensive drugs). For tests of both sets of hypotheses, data would be collected on multiple other variables to control for possible confounders. The main INTERSALT results on these prior hypotheses for individuals (N ¼ 10 074) were: a consistent significant independent relation between 24-h urinary sodium excretion and SBP. With multivariate correction for reliability, and BMI included or excluded in analyses, estimates of the size of the sodium–SBP/DBP cross-sectional relation were 3.1–6.0/0.1–2.5 mmHg lower on average with 100 mmol/day lower Na intake (Table 2).21 BMI, 24-h Na/K excretion, and heavy alcohol intake (X300 ml/week) were directly and independently related to SBP and DBP; 24-h potassium excretion was inversely related.21 Results on sodium and blood pressure in INTERSALT prevailed for younger (ages 20–39 years) and older (ages 40–49 years) participants, with coefficients about two to three times larger for older than younger; results also prevailed for men and women, and for nonoverweight and overweight persons.32,33 Analyses were also performed on best fit for the sodium–SBP relation. Linear and exponential fits were virtually identical in their multivariate adjusted Z score, measuring goodness of fit. In contrast, log Na and ONa were poorer fits.21 These findings cast doubt on conjectures that sodium intake 4100 or 4150 mmol/day has little further influence in raising blood pressure. INTERSALT within-population analyses on individuals also showed an inverse relation between education and blood pressure for many samples worldwide,34 and for the first time elucidated

Table 2 Coefficients for relations of lifestyle traits of individuals to their blood pressures in the INTERSALT study: multivariate analyses with five or four variables to test the INTERSALT prior within-population hypothesesa SBP

24-h urinary Na excretion 100 mmol lower (mmHg) 24-h urinary K excretion 50 mmol higher (mmHg) Alcohol intake 0 or 1–299 ml/week (mmHg) Alcohol intake 0 or X300 ml/week (mmHg) BMI 3 units lower (mmHg)

DBP

5-factor analyses

4-factor analyses

5-factor analyses

4-factor analyses

3.1* 3.4* 0.5 3.5* 2.2*

6.0* 3.3* 0.6 3.2* F

0.1 1.9 0.1 2.1* 1.8*

2.5* 1.6w 0.2 1.8* F

a N=10 074; of the total (10 079 persons), five were not included because of missing data on alcohol intake. Multivariate coefficients were adjusted for regression-dilution bias. SBP=systolic blood pressure; DBP=diastolic blood pressure. Analyses were performed with and without inclusion of BMI for two reasons: (1) BMI inclusion may be overadjustment, since persons with higher BMI excreted more NaFage–sex-sample-adjusted correlation coefficient for 24-h urinary Na and BMI=0.202, with adjustment also for withinperson variation=0.330; (2) BMI, a variable with reliability of measurement approaching 1.00, can produce distortion of coefficients for variables measured with much lower reliability (eg Na, K). *Po0.001, wZ=1.933, Po0.10 >0.05.



possible reasons for this finding: the less the education of individuals, the greater their 24-h sodium excretion, alcohol intake, and BMI, and the lower their 24-h potassium excretion. These four lifestyle variables accounted significantly for the inverse association between education and BP.34 To explore whether the sodium–BP relation was due largely to salt sensitivity among hypertensive participants, analyses were repeated involving only nonhypertensive persons.21 Coefficients for the Na– SBP and Na–DBP relations were similar for 8344 nonhypertensive individuals and all 10 079 participants. Thus, salt sensitivity is common throughout the population. The main INTERSALT results on its prior crosspopulation (ecologic) hypotheses (N ¼ 52) were: with standardization for age and sex, all five coefficients significant (Po0.01 or o0.001) and strong; with control also for BMI and alcohol intake, four of five coefficients significant and strong (Table 3).21 Four methods were used to assess BP differences of persons aged 55 years compared with 25 years for each of the 52 samples, and sample median (and mean) 24-h sodium excretions were related to these differences.21,35 All methods gave similar significant estimates (Po0.001): SBP/DBP higher by 10–11/ 6 mmHg over 30 years (eg age 55 years compared to age 25 years) with sodium excretion higher by 100 mmol/day. These findings indicate that much of the total overall upward slope of SBP/DBP with age, for example, age 25–55 years, averaging 15/ 11 mmHg for the 52 samples, could be attributable to high-salt intake. ImplicationsFboth theoretical (basic science) and practical (public health/medical care)Fof these findings have been presented.21,32,35,36 After the first reports,32,37 the INTERSALT leadership posed the question: do these data on relations of Na, K, Na/K, BMI, alcohol to BP encompass all possible influences of diet on BP? The judgement was: NoFand availability at the Central Laboratory of deep-frozen urine specimens for 10 000+ participants enabled a phase-2 investigation, on relations to BP of individuals of their 24-h urinary nitrogen excretion (total and urea N) as indices of individual dietary total

protein intake, and of sulphate excretion as index of dietary sulphur-containing amino acids. With control for multiple possible confounders, there was for 24-h urinary excretion of total nitrogen and urea a significant inverse relationship to SBP/DBP.38 Both total N and urea in 24-h urine specimens were assessed for methodologic reasons: it is far simpler and cheaper to measure urinary urea than total N as an index of total protein intake by individuals. For all 10 020 individuals in the analyses, the age–sex–sample-adjusted correlation between these two variables was 0.984; with correction for within-person variation, it was 1.000.38,39 This correlation was consistently high for persons of both genders from all 52 samples.

Other studies on diet and blood pressure

During the mid-1990s, the foregoing results served to stimulate corresponding explorations of other data sets. A significant independent inverse relation of dietary total protein to BP was reported from analyses of a British national nutrition survey, and for 11 000+ US men randomized into the Multiple Risk Factor Intervention Trial (MRFIT).40–42 These concordant results led to renewed attention to earlier Japanese findings, indicating an independent inverse relation of dietary protein to BP.26–29 Chinese epidemiologic studies also indicated such an influence on BP of dietary protein, and of individual amino acids.43–45 In the late 1980s and early 1990s, reports were published on randomized trials on amount and type of protein, especially vegetable vs animal protein, and BP. These were short-term studies of small sample size, hence their findings were limited.46–51 Among population-based epidemiologic investigations on diet and BP, the Western Electric Study was unusual in reporting prospective data. With use of an in-depth nutritional survey at examination years 1 and 2, to assess usual eating pattern during the previous month by 1714 employed middle-aged Chicago men, it related average nutrient intakes (years 1 and 2) of each man to his average annual change in SBP/DBP through examination year 9.52 In

Table 3 Multiple regression coefficients for sample 24-h median sodium excretion and sample blood pressure: tests of the INTERSALT prior cross-population (ecologic) hypothesesa Dependent variable

SBP slope with age (mmHg over 30 years with 100 mmol/day greater Na intake) DBP slope with age (mmHg over 30 years with 100 mmol/day greater Na intake) Median SBP (mmHg with 100 mmol/day greater Na intake) Median DBP (mmHg with 100 mmol/day greater Na intake) Hypertension prevalence with 100 mmol/day greater Na intakeb a

Adjusted for age and sex

Adjusted for age, sex, BMI, and alcohol

9.0*** 6.3*** 7.1*** 3.8** 6.2**

10.2*** 6.3*** 4.5*** 2.3w 4.8*

N=all 52 INTERSALT population samples. Hypertension was defined as SBP X140 mmHg, DBP X90 mmHg, or receiving antihypertensive drugs. Units of prevalence are percentage points. SBP=systolic pressure; DBP=diastolic pressure. *Po0.05, **Po0.01, ***Po0.001, wP=0.08. b



analyses of single dietary variables, total and animal protein; total, saturated, monounsaturated, and polyunsaturated fatty acids; cholesterol; Keys dietary lipid score; calcium; alcohol; and average annual change in weight were positively and significantly related to average annual SBP change; vegetable protein, total carbohydrate, beta-carotene, and an antioxidant vitamin score (vitamin C and betacarotene) were inversely and significantly related to SBP change. With combinations of dietary factors, cholesterol, Keys score, and alcohol were positively related to SBP change; vegetable protein and antioxidant index were inversely related to SBP/DBP change; change in weight was directly related to SBP/DBP change. MRFIT has also published on cross-sectional relations of multiple nutrients to SBP and DBP.41,42 For each of 11 342 men, use was made of average nutrient values from 4–5 24-h dietary recalls and average values for BP at six annual visits, with adjustment for multiple possible confounders. BP was directly related to BMI and to intake of sodium, Na/K, alcohol, and inversely related to dietary potassium.41 In analyses of single macronutrients, there was a significant direct relation of dietary cholesterol (mg/day) and starch (% kcal) to SBP/ DBP, and a significant inverse relation of total protein (% kcal) to SBP/DBP. There was a significant direct relation to DBP of saturated fatty acids (% kcal), dietary cholesterol (mg/1000 kcal), Keys dietary lipid score, and a significant inverse relation to DBP of PFA (% kcal), PFA/SFA ratio, and simple

carbohydrates other than refined sucrose.41,42 Zscore for the relation of omega-3 PFA to DBP was 1.94 (Po0.10 40.05). With several macronutrients considered together, there was a significant independent direct relation of dietary cholesterol (mg/ 1000 kcal) and starch (% kcal) to SBP/DBP, of SFA (% kcal) to DBP, and a significant independent inverse relation of total protein (% kcal) and of PFA/ SFA to DBP (Table 4).42 In summary, several recent sets of epidemiologic data on diet–BP relations support the concept that multiple macro- and micronutrientsFin addition to dietary NaCl, K, alcohol intake, calorie imbalanceFinfluence SBP/DBP. They also indicate thatFat least in cross-sectional and short-term prospective studiesFinfluences of each individual nutrient are apparently ‘small’, but independent and additive, so that in combination they can be sizable and important in accounting for the common considerable rise in SBP/DBP from youth through middle age, resultant population-wide adverse average SBP/ DBP from age 35 years on, high prevalence rates of high-normal and high SBP/DBP, and low prevalence rates of optimal SBP/DBP.

The DASH feeding trials

Data on macronutrient–BP relations amassed in the mid-1990s were made available before publication to the DASH research group, and served as a basis for development and formulation of the DASH

Table 4 Relationship of combinations of macronutrients to BP (SBP and DBP) for 11 342 men, years 1–6 of MRFIT: multiple linear regression analyses Variable

Linear regression coefficient (Z-score) SBP

DBP

Model 1 Total protein (% kcal) Cholesterol (mg/1000 kcal) Saturated fatty acids (% kcal) Polyunsaturated fatty acids (% kcal) Starch (% kcal) Other simple carbohydrates (% kcal)

0.0346 0.0039 0.0755 0.0100 0.1366 0.0327

(1.10) (2.46) (1.45) (0.24) (4.98) (1.35)

0.0568 0.0032 0.0848 0.0284 0.0675 0.0006

(3.17) (3.51) (2.86) (1.22) (4.34) (0.04)

Model 2 Total protein (% kcal) Cholesterol (mg/1000 kcal) Saturated fatty acids (% kcal) Polyunsaturated fatty acids (% kcal) Starch (% kcal)

0.0344 0.0034 0.0786 0.0029 0.1149

(1.10) (2.14) (1.73) (0.08) (4.65)

0.0489 0.0029 0.1051 0.0230 0.0608

(2.77) (3.19) (4.08) (1.07) (4.35)

Model 3 Total protein (% kcal) Cholesterol (mg/1000 kcal) P/S (% kcal/% kcal) Starch (% kcal)

0.0385 0.0036 0.0393 0.0920

(1.25) (2.38) (0.15) (4.13)

0.0580 0.0034 0.5586 0.0442

(3.31) (3.93) (3.93) (3.51)

Cholesterol indicates dietary cholesterol. All three models controlled for baseline age, race (African American, non-African American), education, smoking, serum cholesterol, status with regard to smoking at year 6, antihypertensive drug treatment status at year 6, BMI, and reported alcohol intake (% kcal). Models 2 and 3 also controlled for intake of caffeine (mg/day) and of sodium and potassium (mmol/day, Model 2; Na/K, Model 3). Coefficients are unadjusted for regression–dilution bias. Journal of Human Hypertension


‘combination’ diet. The two DASH feeding trials demonstrated that an eating pattern modified in several respects from usual US intakeFthe DASH combination dietFsubstantially reduced SBP/DBP of both nonhypertensive and hypertensive adults, independent of and additive to the sizable reduction in BP that results from lower salt intake.53,54 The DASH combination dietFhigh in fruits, vegetables, low-fat and fat-free dairy products, and reduced in red meats, other fat-containing animal products, total fats, and sweetsFinvolved multiple modifications in food and nutrient intake. This design precluded assessment of influences on BP of changes in specific nutrients. Although data from observational studies indicate that several of these nutrients may affect BP, their limitations preclude definitive conclusions. Given the substantial reduction in SBP/DBP with the DASH combination diet, and consequent implications for basic science, medical care, and public health, clarification of BP effects of its individual nutrient alterations is now a major challenge.

rise of adult BP with age, and on CVD risks attributable to high-normal and high BP, in the adult population overall, and in less favoured socioeconomic and ethnic strata (eg US African Americans) with especially adverse BP patterns. Basic underlying concept and framework of reference

Recent investigative advances have created preconditions for major research progress on the longstanding problem of relations of multiple nutrients to BP. These advances encompass: 1. 2.

3.

INTERMAP General aim

INTERMAP is a basic epidemiological investigation aiming to clarify unanswered questions on the role of multiple dietary factors in the aetiology of unfavourable BP patterns prevailing for most middle-aged and older individuals. Its general aim isFby means of an international cooperative 17sample population study of 4680 men and women aged 40–59 years in four countries (China, Japan, UK, US)Fto advance knowledge on influences of dietary factors on BP of individuals, and on their role in the aetiology of epidemic high-normal and high BP (HBP) in men and women of varied ethnic– racial and socioeconomic backgrounds. This is to be done through high-quality standardized data collection on the 4680 people in the 17 samples, East Asian and Western, consuming diverse diets. With knowledge available on adverse effects on BP of overweight, heavy alcohol use, high salt (NaCl) and suboptimal potassium (K) intake, the focus is on elucidating influences of other dietary factors on BP of individuals: amount and type of protein (including specific amino acids), lipids (including specific fatty acids), carbohydrates (including fibre), also dietary calcium, magnesium, iron, selenium, vitamins, caffeine, and the role of these factors in the higher BP of less-educated compared to more-educated adults. A long-term aim is also to explore relations of food groups (eg fish, lean red meat, low-fat dairy products, fruits and vegetables) to BP. Based on relations of multiple dietary components to BP, the general aim also is to estimate favourable impact of improved nutritional patterns on population average SBP/DBP in middle age, on Journal of Human Hypertension

4.

enhanced understanding of the nature of the population-wide BP problem and of optimal research designs for its study; clarification of the role of high dietary salt, high Na/K, inadequate K, high body mass, and heavy alcohol use in the population-wide BP problem; accrual of new data on possible relations of other dietary factors to BP, including data from population-based observational studies and from the DASH feeding trials; sharpened definition of methodological considerations to be met by population-based epidemiologic investigations to address unresolved aspects of the diet–BP problem, that is, recognition that single dietary factors have apparently ‘small’ effects on SBP/DBP in cross-sectional studies, hence detection of these effects requires: a. large population-based samples, preferably with diverse lifestyles; b. collection of high-quality dietary data by methods taking into account high ratios of intra- to interindividual variances of nutrient intake by individuals, and enabling estimation of and adjustment for resultant regression–dilution bias;55–58 c. control for multiple possible confounding variables; d. standardized quality-controlled data collection methodology; e. modern methods for data entry, transmission, processing, review, edit, and analysis.

Recognition of and focus on these advances are the conceptual framework of reference of the INTERMAP research endeavour, and the rationale for its design and methods (set down below, following the section Specific aims). Specific aims

Specific aims include testing the following a priori primary hypotheses on diet–BP relationsFindependent of age, sex, body mass, alcohol use, Na and K intake, and other possible confoundersFin the 4680 individual participants (men and women aged


Figure 1 Locations of the INTERMAP field centres in Japan, People’s Republic of China, UK, and USA; maps reproduced courtesy of Rand McNally & Company, Skokie, IL, USA, from the Rand McNally Classroom Atlas.

40–59 years) from 17 diverse population samples in four countries, East Asian and Western (ChinaF three samples north to south; JapanFfour samples north to south; UKFtwo samples; and USFeight

samples, diverse in geographic, socioeconomic, and ethnic composition) (Figure 1): 1. Dietary total proteinFassessed both by four 24-h diet recalls/person and by urea nitrogen in two Journal of Human Hypertension


24-h urine collections/personFis inversely related to SBP/DBP of individuals. 2. The inverse relation between education and BP is accounted for to an important degree by interindividual differences in dietary protein intake, as well as in BMI, and in intake of alcohol, Na, and K, all related to education. 3 and 4. There are direct relations of dietary SFA and of dietary cholesterol (CHOL) intake to SBP/ DBP of individuals. 5–8. There are inverse relations of total dietary polyunsaturates (PFA), omega-3 PFA, omega-6 PFA, and the PFA/SFA ratio to SBP/DBP of individuals. 9. There is a direct relation of Keys dietary lipid score (and Hegsted score) of individuals to their SBP/DBP. 10. There is a direct relation of dietary starch to SBP/DBP of individuals. Specific aims also entail testing the following a priori subgroup hypotheses: 1 and 2. In individuals from East Asian samples (China, Japan) and Western samples (UK, US), dietary total protein is inversely related to SBP/ DBP, and dietary lipidFthat is, Keys (and Hegsted) scoreFis directly related to SBP/DBP, independent of possible confounders. 3 and 4. In less-educated and more-educated participants, dietary total protein is inversely related to SBP/DBP, and dietary lipid (Keys, Hegsted score) is directly related to SBP/DBP, independent of confounders. Specific aims also encompass exploratory analyses of relations of other dietary factors to SBP/DBP of individuals, for example, animal and vegetable protein, specific amino acids, calcium and magnesium, simple carbohydrates other than sucrose, dietary fibre; antioxidants (vitamins, etc), caffeineFall suggested to be inversely related to BP; also, iron, total fat, individual fatty acids, sucroseFall suggested to be directly related to BP. Specific aims also include exploratory analyses on relations to SBP/DBP of microalbuminuria, urinary amino acids, other nutrients and metabolites in urine identifiable by nuclear magnetic resonance (NMR) and high-pressure liquid chromatography (HPLC). Specific aims also encompass planningFbased on nutrient–BP findingsFexploratory analyses on relations of intake of food groups to BP of individuals.

Design

INTERMAP isFas notedFan international crosssectional basic epidemiologic study of 4680 individual men and women aged 40–59 years from 17 diverse population samples (Figure 1). To assess Journal of Human Hypertension

relations of dietary variables to SBP/DBP, it utilizes average values for each person of nutrients from four 24-h dietary recalls and two timed 24-h urine collections and average values for each person of eight carefully measured SBP/DBP (two per visit), plus data on multiple other variables possibly confounding nutrient–BP relations. Quantile and multiple linear regression analyses are its main statistical procedures to assess nutrient–BP relations.

Methods

Each sample was representative of a defined population; both general population and workforce samples were included. Individuals were selected randomly from population lists, stratified by age/ gender to give approximately equal numbersF65 personsFin each of four 10-year age/gender groups. Personnel for data collection and processing were trained and certified in study methods at national training sessions led by senior staff (national and international) based on detailed protocols set out in Manuals of Operations. Each participant attended the local research centre four times: two visits on consecutive days, two such visits again 3–6 weeks later. Wherever possible, one visit included a weekend day (or an equivalent rest day) according to work schedule. BP of the seated participant was measured twice/ visit (right arm where possible) with a random zero sphygmomanometer after at least 5 min rest;59 pulse was measured three times/visit. Dietary data were collected at each visit with the 24-h recall method.60 All foods and drinks consumed in the previous 24 h, including dietary supplements, were recorded in an interview by a trained and certified dietary interviewer. To aid accurate recall, food and drink models, measuring devices, and photographs were used. Interviewers used neutral probing techniques to check completeness. Interviews were recorded; a random sample was later independently reviewed for quality control. In the US, dietary information was directly computerized, with use of a program (the Nutrition Data System (NDS), Nutrition Coordinating Center, University of Minnesota) to guide on-screen coding. This enabled computerization of detailed information on all reported foods and beverages and a selected list of supplements (mainly vitamins and minerals) in the NDS; other supplements were entered onto standard forms and later computerized centrally. In the other three countries, dietary data were first entered onto standard forms, then coded and computerized. A random 10% of recalls were recoded and re-entered, with staff blinded to original entries. Daily alcohol consumption over the previous 7 daysFand, for abstainers, information on previous drinkingFwas obtained by interview twice (first


and third visits). Consumption during the previous 24 h was also obtained at 24-h dietary interviews. Two timed 24-h urine specimens were collected for measurement of urinary sodium, potassium, creatinine, urea, calcium, magnesium, micro- and macroalbumin, amino acids, and multiple metabolites. Urine was collected in standard 1-l plastic jars containing boric acid (for preservation). Timed collections were started at the research centre (first and third visits), and completed at the centre the following day, with instruction on collection.59 Specimens were rejected if collection time fell outside the range 22–26 h, if the participant responded that collection was incomplete, or he/she had lost ‘more than a few drops’ of urine, or if total volume was less than 250 ml. The participant was then asked to repeat the collection. Height and weight without shoes were measured at first and third visits.59 Questionnaire data were obtained by interview on demographic and other possible confounding factors, including education, occupation, physical activity, smoking, medical history, current special diet, medication use, andFfor womenFmenopause, parity, use of contraceptive or hormone replacement medication. Exclusions and supplementary participants: Participants were excluded if: they did not attend all four visits (110 people); diet data were considered unreliable by the diet interviewer and the Site Nutritionist (seven people); energy intake from any 24-h dietary recall was below 500 kcal/day or greater than 5000 kcal/day for women and 8000 kcal for men (37 people); two complete urine collections were not available (37 people); data on other variables were incomplete, missing, or indicated violation of study protocol (24 people). When a participant was excluded, a supplementary participant was recruited from the sample age/sex group. Urine specimen preparation and biochemical analyses: Height of urine in each jar was obtained with use of a specially designed measuring scale; height was later converted by computer into volume with an empiric formula based on repeated measurements of volume in like jars. All urine from a 24h collection was then combined, mixed thoroughly by vigorous stirring, and several urinary aliquots taken and stored locally at 201C. Aliquots were periodically sent by airplane to the Central Laboratory, Leuven, Belgium. Urinary sodium and potassium concentrations were measured by emission flame photometry. Methods for analyses of urinary creatinine, urea, calcium and magnesium (atomic absorption flame photometry), and microalbuminuria are given in references.61–64 Work is in progress on measurement of urinary amino acids,65 and multiple nutrients and metabolites by NMR and HPLC.66–71 Pre-prepared reference samples and commercial samples were included daily in analyses to check laboratory variation. As part of quality control, 10% of samples were split at the clinical centre and sent

to the laboratory with different ID numbers. Results were compared to P obtain an estimate of technical error, defined as ( d2/2N), where d is the difference between a pair of measurements and N the number of split pairs. Percent technical error (defined as 100 times technical error divided by mean value of split samples) was calculated and averaged over samples, weighted by N. Average per cent technical errors are 1.87% for sodium; 1.61%, potassium; 2.46%, creatinine; 3.28%, urea; 4.20%, calcium; 3.98%, magnesium. Study organization and leadership: There are three International Coordinating Centers, two for the study overall, in Chicago and London, and the thirdFfor nutritionFin Chapel Hill. Chicago and London have been jointly responsible for development of the Protocol, field and International Coordinating Center Manuals of Operations, training, data collection, computerization, editing, analysis; Chapel Hill was responsible for the Nutrition Manual of Operations, four national supplements to this manual, and nutritional aspects of the study, in collaboration with the International Coordinating Centers in Chicago and London, and the Nutrition Coordinating Center (NCC), Minneapolis.60 NCC provided nutrient content of foods (76 nutrients) not included in national nutrient databases, and for checking and updating data on other foods.60,72,73 The international Central Laboratory (Leuven, Belgium) was responsible for its Manual of Operations and work, and the NMR-HPLC Metabonomics Laboratory (London, England) was responsible for its Manual of Operations and work. Study leadership nationally was accomplished by country centres in Shiga (Japan), Beijing (PRC), London (UK), and Chicago (USA). Members of the Steering and Editorial Committee, Advisory Committees, local, national, and international centres, are listed at the end of this paper. Statistical methods: BMI was calculated as weight (kg)/height2 (m2). Dietary data were converted into nutrients with use of country-specific food tables enhanced by NCC.60,72,73 Total energy intake was estimated from conversion factorsFfat: 9 kcal/g; protein: 4 kcal/g; available carbohydrate: 4 kcal/g; alcohol: 7 kcal/g. Nutrient densities were calculated as follows: for nutrients supplying energy, as per cent of total kilocalories, that is, (kcal from nutrient/ total kcal) 100; for other nutrients, per 1000 kcal, that is, (amount per day/total kcal) 1000. Dietary protein was partitioned into animal and vegetable. Urinary values/24 h were calculated as products of urinary concentrations and timed volumes standardized to 24 h. Urinary urea (g/24 h) was converted into urinary urea nitrogen (g/24 h) with the multiplier 0.4667 and then multiplied by (1.21787 6.25 ¼ 7.61) to estimate dietary total protein.38 The multiplier 1.21787 is to convert urinary urea N to urinary total N (from all dietary protein);38 the multiplier 6.25 is to convert urinary total N to estimated dietary total protein.38,60 Journal of Human Hypertension


For descriptive statistics, means, standard deviations, and medians were computed for nutrients and other variables for the 17 samples, by gender and age, for men and women, from average of all measurements for each individual, and then averaged to give data by sample, country, and region (East Asian, Western).74 For analyses on nutrient–BP relations for individuals, BP of each individual was the average of eight measurements from the four visits, nutrients were the average of four measurements from four 24-h dietary recalls and urinary variables were the average of two measurements from two 24-h urine collections. To examine relationships of dietary and other variables to BP, the INTERMAP Protocol and Analysis Plan stipulated use of quantile and multiple regression analyses, with control for age and gender, then control for other variables. Multiple regression analyses were to be performed for each sample individually, with pooling of resulting coefficients across samples and countries by weighting each regression coefficient by inverse of its variance. Owing to possible over-correction of BP– nutrient relationships due to associations of nutrient intake with education and with body mass,21 the multiple regression analyses were to be run without and with inclusion of height, weight, and education. The four pooled country-specific coefficients were to be tested for heterogeneity.57,58 Where significant (Po0.05) heterogeneity was detected, the standard error of the coefficient was to be recalculated to allow for a random effects component. Linearity assumptions of the model were to be checked by examining mean SBP and DBP data, by country, for country-specific quantiles of the dietary variable. Sensitivity analyses were stipulated, with various exclusions: 1. regression analyses excluding all people (a) on special diets, (b) with high BP, or (c) only people receiving antihypertensive or other CVD medication, or combinations of (a), (b), and (c); 2. given the well-known high ratio of intra- to interindividual variability in nutrient intake assessed by 24-h dietary recalls, exclusion from regression analyses of individuals with inordinately high variability, that is, coefficient of variation (CV%) 480% for dietary variables (in nutrient densities) calculated from comparing the mean of the first two 24-h recall values with the mean of the second two (for energy, protein, SFA, PFA, PFA/SFA, cholesterol, starch, fibre, sodium, potassium, magnesium, calcium); coefficient of variation 480% for urinary variables, based on the two measurements (24-h urinary Na, K, urea, creatinine); coefficient of variation 414 mmHg for SBP or 413 mmHg for DBP; absolute values for the ratio dietary sodium/ urinary sodium outside 0.4–2.4; for the ratio dietary potassium/urinary potassium outside 0.7–2.5; for the ratio dietary total protein/estimated dietary total protein from urinary urea outside 0.7–2.5. Analyses Journal of Human Hypertension

were also stipulated with exclusion of individuals identified by both (1) and (2) above.

Descriptive statistics (nondietary)

The 4680 participants were by design equally divided approximately by age and gender, overall and in each country (Table 5). They were ethnically diverse across the four countries, as were the 2195 US participants (54% self-identified as non-Hispanic White Americans; 17% African Americans; 13% Hispanic Americans, including the all-HispanicAmerican sample in Corpus Christi, Texas; 12% Japanese Americans, including the all JapaneseAmerican sample in Honolulu, Hawaii) (Table 5). (See also Appendix tables in this issue of the Journal of Human Hypertension for age–sex-specific data on nondietary and dietary variables by sample, country, region, and overall.)74 Higher percentages of Chinese and Japanese participants were married (92 and 93%) compared to UK and US participants (76 and 69%) (Table 5). Correspondingly, proportion divorced or separated was lower for East Asians than Westerners (0.5–2.4 vs 9–18%), as was percentage never married (0.4–3.6 vs 7–9%). Chinese participants, from rural samples, were on average much less educated (5.4 years) than persons from the other countries (Japan 12.0 years, UK 12.7 years, US 15.0 years) (Table 6). Of PRC participants, none reported 16 or more years of education; of Japanese, 8%; of UK, 19% (15% 17+ years, that is, probable university graduates); of US, 47%. Most of these people, aged 40–59 years at survey, reported themselves to be employed or (a small proportion) self-employed (70–90%, 85% overall) (Table 6). Only a small percentage described their occupation as homemakerF12% overall, 26% of the Chinese, 7–9% of others. On average, Chinese participants described themselves as engaged in almost 2 h per day of heavy physical activity; others, less than an hour (Table 6). Japanese, UK, and US individuals reported themselves to be sedentary 7–8 h per day, and watching television an additional 2.3–2.4 h; for Chinese, these low levels of activity involved 3.5 and 1.8 h. A majority of East Asian men were current cigarette smokers; only 17–19% of UK and US men (Table 7). A small percentage of East Asian men had never smoked (16–23%); but nearly one-half (44–46%) of Western men. For East Asian women, per cent of current smokers was lower (5–9%) than for Western women (14–18%). Most Asian women (89–94%) had never smoked; 60–61% of Western women. Proportion with a diagnosis of high BPF24% overallFwas about twice as high for Western (27– 32%) than for East Asian participants (14–15%) (Table 8). Similarly, percentage reporting drug treatment for hypertension/CVD was considerably higher for UK and US individuals (16–24%) than for Japanese or PRC (7–8%).


Table 5 INTERMAP participants by country, gender, age, ethnicity, marital status Variable Number % Gender Age, years Age group (years) Ethnicity

Marital status

Japan Total Male Female 40–59 50–59 White, non-Hispanic African American African Caribbean Chinese Japanese Indian (subcontinent) Other Asian Hispanic, non-White Hispanic, White Native American Other Now married Divorced or separated Widowed Never married Cohabiting

1145a 574 571 49.4c 579 566

1145

F F F F F F F F F F

PRC

100.0b 50.1 49.9 (5.3)d 50.6 49.4

839 416 423 49.0 425 414

839 100.0

1054 28 19 41 3

92.0 2.4 1.7 3.6 0.3

UK

780 4 52 3

100.0 49.6 50.4 (5.8) 50.7 49.3 F F F 100.0 F F F F F F F 93.0 0.5 6.2 0.4 F

501 266 235 49.2 263 238 466 9 23 2

1 383 47 9 37 25

USA

100.0 53.1 46.9 (5.6) 52.5 47.5 93.0 F 1.8 F F 4.6 0.4 F F F 0.2 76.4 9.4 1.8 7.4 5.0

2195 1103 1092 49.1 1098 1097 1190 369 5 16 269 18 14 175 113 10 16 1520 399 52 188 36

All 100.0 50.2 49.8 (5.4) 50.0 50.0 54.2 16.8 0.2 0.7 12.3 0.8 0.6 8.0 5.2 0.5 0.7 69.2 18.2 2.4 8.6 1.6

4680 2359 2321 49.2 2365 2315 1656 369 14 855 1414 41 16 175 113 10 17 3737 478 132 269 64

100.0 50.4 49.6 (5.5) 50.5 49.5 35.4 7.9 0.3 18.3 30.2 0.9 0.3 3.7 2.4 0.2 0.4 79.8 10.2 2.8 5.8 1.4

a

Number. Per cent. c Average. d s.d. b

Table 6 INTERMAP participants by country, education, occupational status, physical activity of work and leisure Variable

Japan

PRC

UK

USA

All

Education (years)

12.0a

(2.1)b

5.4

(2.9)

12.7

(3.1)

15.0

(3.0)

12.3

(4.4)

Education 0–6 7–8 9–11 12 13 14–15 16 17 X18

0c 0 273 621 27 131 62 11 20

0.0d 0.0 23.8 54.2 2.4 11.4 5.4 1.0 1.7

580 132 114 11 1 1 0 0 0

69.1 15.7 13.6 1.3 0.1 0.1 0.0 0.0 0.0

6 4 209 57 58 74 16 38 39

1.2 0.8 41.7 11.4 11.6 14.8 3.2 7.6 7.8

17 21 72 429 200 424 489 121 422

0.8 1.0 3.4 19.5 9.1 19.3 22.3 5.5 19.2

603 157 668 1118 286 630 567 170 481

12.9 3.4 14.3 23.9 6.1 13.5 12.1 3.6 10.3

Occupation Employed Self-employed Homemaker Other

948 81 107 9

82.8 7.1 9.3 0.8

528 67 220 24

62.9 8.0 26.2 2.9

412 13 37 39

82.2 2.6 7.4 7.8

1778 141 180 96

81.0 6.4 8.2 4.4

3666 302 544 168

78.3 6.5 11.6 3.6

2.5d 8.3 16.2 9.5 34.0 29.5 100.0

1.9 (3.4) 4.1 (3.2) 3.9 (2.5) 1.8 (1.2) 3.5 (2.4) 8.8 (1.2) 24.0

7.9 17.1 16.2 7.5 14.6 36.7 100.0

0.4 (1.0) 1.8 (2.2) 3.5 (2.6) 2.4 (1.6) 8.2 (3.2) 7.6 (1.1) 23.9

1.7 7.5 14.6 10.0 34.3 31.8 99.9

0.8 (1.6) 2.4 (2.5) 3.9 (2.7) 2.3 (1.6) 7.4 (4.0) 7.2 (1.3) 24.0

3.3 10.0 16.2 9.6 30.8 30.0 99.9

0.9 (2.1) 2.5 (2.9) 3.8 (3.0) 2.2 (1.5) 7.0 (4.1) 7.5 (1.4) 23.9

3.8 10.5 15.9 9.2 29.3 31.4 100.1

Hours/day of activity Heavy Moderate Slight Watching TV Sedentary Inactive (sleep) Sums

0.6a 2.0 3.9 2.3 8.2 7.1 24.1

(1.8b) (3.2) (3.9) (1.3) (4.4) (1.1)

a

Average. s.d. c Number. d Per cent. b



Table 7 INTERMAP participants by country, gender, and cigarette smoking status Variable

Japan

PRC

UK

USA

All

Men and women Never smokers Ex-smokers Cigs/day when smoking Current smokers Cigs/day 1–19/day 20+/day All participants

642a 157 22.8c 346 21.1 113 233 1145

56.1b 13.7 (14.1)d 30.2 (9.9) 9.9 20.3 100.0

465 69 20.4 305 21.7 89 216 839

55.4 8.2 (10.2) 36.4 (11.2) 10.6 25.7 100.0

267 147 22.9 87 14.1 59 28 501

53.3 29.3 (15.6) 17.4 (9.0) 11.8 5.6 100.0

1149 677 22.8 369 16.2 193 176 2195

52.3 30.8 (16.4) 16.8 (10.7) 8.8 8.0 100.0

2523 1050 22.6 1107 19.1 454 652 4680

53.9 22.4 (15.6) 23.7 (10.8) 9.7 13.9 100.0

Men Never smokers Ex-smokers Cigs/day when smoking Current smokers Cigs/day 1–19/day 20+/day All participants

133 144 23.8 297 22.7 76 221 574

23.2 25.1 (14.2) 51.7 (9.4) 13.2 38.5 100.0

67 65 21.1 284 22.3 77 207 416

16.1 15.6 (10.0) 68.3 (11.2) 18.5 49.8 100.0

123 98 24.9 45 15.4 27 18 266

46.2 36.8 (16.2) 16.9 (11.2) 10.2 6.8 100.0

491 401 24.4 211 17.2 104 107 1103

44.5 36.4 (16.2) 19.1 (9.5) 9.4 9.7 100.0

814 708 24.1 837 20.8 284 553 2359

34.5 30.0 (15.7) 35.5 (10.9) 12.0 23.4 100.0

Women Never smokers Ex-smokers Cigs/day when smoking Current smokers Cigs/day 1–19/day 20+/day All participants

509 13 11.2 49 11.5 37 12 571

89.1 2.3 (5.2) 8.6 (7.0) 6.5 2.1 100.0

398 4 8.0 21 13.3 12 9 423

94.1 1.0 (2.5) 5.0 (7.3) 2.8 2.1 100.0

144 49 19.0 42 12.6 32 10 235

61.3 20.9 (13.7) 17.9 (8.4) 13.6 4.3 100.0

658 276 20.4 158 14.9 89 69 1092

60.3 25.3 (15.6) 14.5 (9.7) 8.2 6.3 100.0

1709 342 19.7 270 13.8 170 100 2321

73.6 14.7 (15.1) 11.6 (9.0) 7.3 4.3 100.0

a

Number; bPer cent; cAverage; ds.d.

Table 8 INTERMAP participants by country, personal medical diagnosis, drug treatment for high BP/CVD, family history of high BP Variable Diagnoses High BP Heart attack Heart disease Stroke Diabetes Drug treatment For HBP/CVD For CVD only Family history of HBP Father Yes No Do not know Mother Yes No Do not know Sister Yes No Do not know No sister Brother Yes No Do not know No brother All participants a

Japan

PRC

UK

All

167a 8 91 1 39

14.6b 0.7 8.0 0.1 3.4

121 0 33 12 14

14.4 0.0 3.9 1.4 1.7

136 9 34 4 9

27.2 1.8 6.8 0.8 1.8

707 40 136 34 179

32.2 1.8 6.2 1.6 8.2

1131 57 294 51 241

24.2 1.2 6.3 1.1 5.2

85 12

7.4 1.0

66 3

7.9 0.4

82 16

16.4 3.2

517 34

23.6 1.6

750 65

16.0 1.4

259 673 213

22.6 58.8 18.6

113 623 103

13.5 74.3 12.3

93 282 126

18.6 56.3 25.2

761 931 503

34.7 42.4 22.9

1226 2509 945

26.2 53.6 20.2

322 693 130

28.1 60.5 11.4

159 578 102

19.0 68.9 12.2

148 247 106

29.5 49.3 21.2

964 929 302

43.9 42.3 13.8

1593 2447 640

34.0 52.3 13.7

80 620 160 285

7.0 54.2 14.0 24.9

73 613 62 91

8.7 73.1 7.4 10.8

54 237 59 151

10.8 47.3 11.8 30.1

387 992 330 486

17.6 45.2 15.0 22.1

594 2462 611 1013

12.7 52.6 13.1 21.6

89 622 169 265

7.8 54.3 14.8 23.1

56 643 56 84

6.7 76.6 6.7 10.0

57 255 72 117

11.4 50.9 14.4 23.4

427 950 369 449

19.4 43.3 16.8 20.5

629 2470 666 915

13.4 52.8 14.2 19.6

1145

100.0

839

100.0

501

100.0

2195

100.0

4680

100.0

Number; bPer cent; BP=blood pressure; CVD=cardiovascular disease.


USA


Table 9 INTERMAP female participants by country, live births, use of contraceptive pill, menopausal status, use of hormone replacement treatment Variable Number of live births (%) 0 1 2 3 4+ Current users of birth control pills Menopausal status Premenopause Menopause now Postmenopause Users of hormone replacement therapy All women

Japan

42a 49 326 133 21 2 254 64 252 14 571

7.4b 8.6 57.1 23.3 3.7 0.4 44.5 11.2 44.1 2.4 100.0

PRC

UK

USA

All

4 16 127 113 163 6

1.0 3.8 30.0 26.7 38.5 1.4

51c 41 86 36 20 12c

21.7 17.4 36.6 15.3 8.5 5.1

188c 180 359 220 144 32c

17.2 16.5 32.9 20.2 13.2 2.9

285d 286 898 502 348 52d

12.3 12.3 38.7 21.6 15.0 2.2

200 25 199 1 423

47.3 5.9 47.0 0.2 100.0

106e 72 53 48 235

45.1 30.6 22.6 20.4 100.0

413c 294 384 343 1092

37.8 26.9 35.2 31.4 100.0

973f 455 888 406 2321

41.9 19.6 38.3 17.5 100.0

a

Number. Per cent. c Data missing for one woman. d Data missing for two women. e Data missing for four women. f Data missing for five women. b

A majority of Chinese rural women (65%) reported having three or more live births; this proportion ranged from 24% (UK) to 33% (US) for other women (Table 9). Of these women aged 40–59 years, only a small percentage reported current use of birth control pills (2% overall, 0.4–5.0% across the four countries). Proportion receiving hormone replacement therapy was considerably higher for UK and US women (20 and 30%) than for Chinese and Japanese women (0.2 and 2.4%).

China, Japan (the Ministry of Education, Science, Sports, and Culture, Grant-in-Aid for Scientific Research [A], No. 090357003), and the UK. We are grateful to Rand McNally & Company, Skokie, IL, USA for permission to reproduce here as Figure 1 three maps from its publication The Rand McNally Classroom Atlas. The INTERMAP Study has been accomplished through the fine work of the staff at the local, national, and international centres. A partial listing of colleagues follows:

Summary

Japan Shiga, Country Coordinating Center Management Committee Hirotsugu Ueshima, Principal Investigator and Chair Akira Okayama, Co-Principal Investigator Sohel R Choudhury, Co-Principal Investigator Yoshikuni Kita, Co-Principal Investigator Nagako Okuda, Country Nutritionist

In summary, the INTERMAP Study is a major new international population-based endeavour providing unique resources for elucidation of nutritional/dietary influences on BP. It draws on a high-quality database for 4680 men and women aged 40–59 years in 17 diverse population samples from four countries. The accompanying papers and Appendix tables in this issue of the Journal of Human Hypertension give further information on methodology (especially for collection and processing of dietary/nutritional data), descriptive analyses, findings on prior hypothesis #2 above, and detailed data tabulations.60,74–77 Results of other analyses on relations of nutritional variables to BP have been presented at national and international meetings, and are to be published.

Acknowledgements This research was supported by Grant 2-RO1HL50490 from the US National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland; by the Chicago Health Research Foundation; and by national agencies in

Aito Town Local Center Akira Okayama, Principal Investigator Sohel R Choudhury, Co-Principal Investigator Yoshikuni Kita, Co-Principal Investigator Harumi Fuse, Research Associate Takashi Kadowaki, Research Associate Atsuko Morino, Research Associate and Site Nutritionist Yoko Sekiya, Research Associate and Site Nutritionist Makoto Watanabe, Research Associate Kazuyo Yamamoto, Research Associate Masako Yoshioka, Research Associate Sapporo Local Center Shigeyuki Saitoh, Principal Investigator Kazuaki Shimamoto, Co-Principal Investigator Shigeyuki Tanaka, Co-Principal Investigator Journal of Human Hypertension


Koko Ishishita, Site Nutritionist Toshio Isomatsu, Research Associate Toshiyuki Takahashi, Research Associate Toyama Local Center Hideaki Nakagawa, Principal Investigator Katsuyuki Miura, Co-Principal Investigator Katsushi Yoshita, Co-Principal Investigator and Site Nutritionist Masako Higashiyama, Research Associate Sadanobu Kagamimori, Research Associate Yuchi Naruse, Research Associate Wakayama Local Center Tsutomu Hashimoto, Principal Investigator Seiji Morioka, Co-Principal Investigator Kiyomi Sakata, Co-Principal Investigator Fujihisa Kinoshita, Research Associate Osamu Mohara, Research Associate Masanori Ohta, Research Associate Kayoko Oki, Research Associate and Site Nutritionist Yoshimi Shibe, Research Associate and Site Nutritionist People’s Republic of China Beijing, Country Coordinating Center Beifan Zhou, Principal Investigator and Country Nutritionist Liancheng Zhao, Country Nutritionist Yangfeng Wu, Co-Principal Investigator Beijing Local Center Yangfeng Wu, Principal Investigator and Coordinator The two other above listed colleagues, plus: Jun Yang, Dietary Interviewer Donghai Lu, Technician for Urine Collection Yao Li, Data Entry Technician Jun Xing, Dietary Interviewer Guangxi Local Center Shuxiong Zhu, Principal Investigator Liguang Zhu, Coordinator and Site Nutritionist Xingsan Li, Quality Control Technician, Non-Dietary Data Collection Xiufan Li, Assistant Coordinator Mengsheng Mo, Dietary Interviewer Shanxi Local Center Ruixiang Yang, Principal Investigator Dongshuang Guo, Coordinator Jinglan Gong, Site Nutritionist Shengying Liang, Assistant Coordinator Peizhu Wang, Quality Control Technician, NonDietary Data Collection Xiuming Li, Data Entry Technician UK London, INTERMAP International and Country Coordinating Center Journal of Human Hypertension

Paul Elliott, Principal Investigator Queenie Chan, Data Coordinator and Statistician Deborah Chee, Coordinator Rana Conway, Country Nutritionist Judith Diamond, Data Coordinator and Statistician Valerie McCormack, Data Coordinator Rob Nichols, Data Coordinator and Statistician Yui Rerkpatima, Dietary Coder Claire Robertson, Country Nutritionist Frankie Robinson, Country Nutritionist Nina Seres, Nutritionist Simon Sheffield, Administrator Caroline Terrill, Data Coordinator and Statistician Michael Tumilty, Nutritionist Jennifer Wells, Special Assistant London, Metabonomics Laboratory Jeremy Nicholson, Co-Principal Investigator Elaine Holmes, Co-Principal Investigator Claire James, PhD Candidate Elaine Taylor, Research Associate Belfast Local Center Alun Evans, Principal Investigator John Yarnell, Co-Principal Investigator Claire Robertson, Coordinator and Site Nutritionist Jackie Bates, Dietary Interviewer Mary Crawford, Dietary Interviewer Elaine Duffy, Interviewer, BP Technician Sally Johnson, Interviewer, BP Technician Julie Laird, Dietary Interviewer Annette McAnnulla, Dietary Interviewer Gillian McGeough, Site Nutritionist Collette Ryan, Dietary Interviewer Margaret Ward, Interviewer, BP Technician West Bromwich Local Center Gareth Beevers, Principal Investigator Gregory Lip, Co-Principal Investigator Koon-Lan Chan, Coordinator Georgina Alderslade, Site Nutritionist Ada Conteh, Dietary Interviewer Melanie Farron, Dietary Coder Anne Gowing, Dietary Interviewer Tina Howe, Dietary Coder Karen Hubbard, Dietary Coder Baljit Sanghera, Site Nutritionist USA Chicago, INTERMAP International and Country Coordinating Center Jeremiah Stamler, Principal Investigator Alan R Dyer, Co-Principal Investigator Kiang Liu, Co-Principal Investigator Rose Stamler (deceased), Co-Principal Investigator Martha Daviglus, Principal Investigator, Study on Urinary Amino Acids Philip Greenland, Co-Investigator


Linda Van Horn, Consultant Nutritionist Sujata Archer, Nutritionist Alicia Moag-Stahlberg, US Coordinator and Country Nutritionist Dan Garside, Chief, Computer Systems Niki Gernhofer, Nutritionist Colleen De Luca, Administrator Joseph Shayka, Chief, Computer Systems for INTERMAP USA Harold Wexler, Secretary

Philip Greenland, Co-Principal Investigator Alan R. Dyer, Co-Investigator Kiang Liu, Co-Investigator Judy Gerber, Coordinator and Nutritionist Niki Gernhofer, Site Nutritionist Cindy Buettgan, Nutritionist Colleen De Luca, Department Administrator Penelope Garrett, Nutritionist Natalie Lopez, Site Administrator Frances Oppenheimer, Nutritionist

Chapel Hill, INTERMAP International Nutrition Coordinating Center Barbara Dennis, International Nutrition Coordinator Susan Blackwell Marshall, Research Associate

Corpus Christi Local Center Darwin R Labarthe, Principal Investigator Milton Nichaman, Principal Investigator Deanna Montgomery Hoelscher, Co-Investigator Lyn Steffen, Project Coordinator Mark Canales, BP Technician Pam Folsom, BP Technician and original Clinic Manager Sarita Garcia, Dietary Interviewer Mimi Guajardo, Administrative Assistant Luanna Ortiz, BP Technician Rose Ramirez, BP Technician Melissa Taylor, BP and Lab Technician Gina Valdez, BP Technician, Dietary Interviewer, Clinic Manager Laurita Yuras, Dietary Interviewer

Minneapolis, Nutrition Coordinating Center Nancy Van Heel, Project Coordinator Marilyn Buzzard, Principal Investigator Lisa Harnack, Principal Investigator Gloria Ray, Dietary Interview Quality Control Sally Schakel, Chief Database Nutritionist Susan Seftick, Dietary Interview Quality Control Alice Shapiro, Project Coordinator Mary Stevens, Dietary Interview and Quality Control Training Coordinator Christine Wold, Database Nutritionist Roberta Zeug, Database Nutritionist Richmond Marilyn Buzzard, Senior Consultant, Dietary Data Collection and Analysis Rhonda Stout, Graduate Assistant, Dietary Interview Quality Control Bethesda, National Heart, Lung, and Blood Institute Eva Obarzanek, Project Officer Jeffrey Cutler, Program Director Baltimore Local Center Paul Whelton, original Principal Investigator Lawrence Appel, Principal Investigator Jeanne Charleston, Director of Research Operations Phyllis McCarron, Site Nutritionist Vicki Shank, Coordinator Sharon Cappelli, Nutritionist Ellen Gold, Nutritionist Charles Harris, Data Technician Dolores Kaidy, Recruiter Shirley Kritt, Recruiter Estelle Levitas, Data Technician Bonnie Peterson, Nutritionist LeeLana Thomas, Nutritionist Letitia Thomas, Nutritionist Bobbie Weiss, Data Technician Chicago Local Center Linda Van Horn, Principal Investigator

Honolulu Local Center Beatriz L Rodriguez, Principal Investigator J David Curb, Co-Principal Investigator Kamal Masaki, Co-Principal Investigator Helen Petrovitch, Co-Investigator Joane Moylan, Site Nutritionist Jackson Local Center Daniel Jones, Principal Investigator Margaret E Miller, Co-Investigator Mary Whitten, Research Administrator Teresa Caruthers, Site Nutritionist Cathy Adair, Research Nurse Mary Cameron, Dietary Interviewer Gertie Carr, Dietary Interviewer Nancy King, Research Assistant Diane Willoughby, Research Nurse Minneapolis Local Center Patricia J Elmer, Principal Investigator David R Jacobs Jr, Co-Principal Investigator Terri Tharp, Project Coordinator and Site Nutritionist Mary Dahlberg Johnson, Dietary Interviewer Kari Elfstrom, Dietary Interviewer Pittsburgh Local Center Arlene Caggiula, Principal Investigator Monica Yamamoto, Co-Principal Investigator and Site Manager Lewis H Kuller, Co-Investigator Lisa Atkins, Site Administrative Assistant and Data Clerk Journal of Human Hypertension


Nancy Epler, Recruiter Karen Givner, Site Administrative Assistant and Data Clerk Refaat Hegasi, Clinical Assessor Deborah Larsen, Dietary Interviewer Alysia Mason Feuer, Dietary Interviewer Rebecca M Meehan, Dietary Interviewer Blanca Nieves, Dietary Interviewer Tony O’Dea, Summer Intern Heather Strickland, Clinical Assesssor Charlene Walter, Dietary Interviewer BelgiumFLeuven, INTERMAP Central Laboratory Hugo Kesteloot, Principal Investigator Peter Declercq, Co-Principal Investigator George Claeys, Co-Principal Investigator Norbert Blanckaert, Head, Central Clinical Laboratory Wim Blom, Adviser Lene Koolen, Technician, INTERMAP Laboratory Kristine Lauwereys, Head Programmer Nadia Vangeel, Technician, INTERMAP Laboratory INTERMAP International Steering and Editorial Committee Jeremiah Stamler, Co-Chair Paul Elliott, Co-Chair Barbara Dennis Alan R Dyer Hugo Kesteloot Kiang Liu Rose Stamler (deceased) Hirotsugu Ueshima Beifan Zhou INTERMAP International Advisory Committee John Chalmers Scott Grundy Mark Hegsted Lisheng Liu Michael Marmot Teruo Omae Kalevi Pyorala Neil Stone INTERMAP International Nutrition Advisory Committee Anna Ferro-Luzzi Daan Kromhout Wen Harn Pan Pirjo Pietinen Wija J van Steveren

2 3

4

5 6

7 8 9

10 11 12 13

14

15

16

17

References 1 Joint National Committee on the Detection, Evaluation, and Treatment of High Blood Pressure (JNC-V). Fifth Report of the Joint National Committee on the Journal of Human Hypertension

Detection, Evaluation, and Treatment of High Blood Pressure (JNC-V). Arch Intern Med 1993; 153: 154–183. National High Blood Pressure Education Program Working Group. Report on the primary prevention of hypertension. Arch Intern Med 1993; 153: 186–208. National High Blood Pressure Education Program. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-VI). NIH Publication No. 984080. National Institutes of Health, National Heart, Lung, and Blood Institute: Bethesda, MD, 1997. Whelton PK et al, for the National High Blood Pressure Education Program Coordinating Committee. Primary prevention of hypertension: clinical and public health advisory from the National High Blood Pressure Education Program. JAMA 2002; 288: 1882–1888. Stamler J, Greenland P, Neaton JD. The established major risk factors underlying epidemic coronary and cardiovascular disease. CVD Prev 1998; 1: 82–97. Stamler J et al. Major risk factors and low risk status in young adulthood: long-term impact on CHD–CVD mortality and longevity. In: Lauer R, Burns TL, Daniels S (eds). Prevention of Adult Cardiovascular Disease Beginning in Childhood and Adolescence, 2003 (to be published). Katz LN, Stamler J, Pick R. Nutrition and Atherosclerosis. Lea and Febiger: Philadelphia, 1958. Stamler J. Lectures on Preventive Cardiology. Grune and Stratton: New York, 1967. Keys A, Anderson JT, Grande F. Serum cholesterol response to changes in the diet. Part I. Iodine value of dietary fat versus 2S-P; Part II. The effect of cholesterol in the diet; Part III. Differences among individuals; Part IV. Particular saturated fatty acids in the diet. Metabolism 1965; 14: 747–787. Hegsted DM, Ausman LM, Johnson JA, Dallal GE. Dietary fat and serum lipids: an evaluation of the experimental data. Am J Clin Nutr 1993; 57: 875–883. Clarke R et al. Dietary lipids and blood cholesterol: quantitative meta-analysis of metabolic ward studies. BMJ 1997; 314: 112–117. Page IH et al. Dietary fat and its relation to heart attacks and strokes. Circulation 1961; 23: 133–136. Krauss RM et al. AHA Dietary Guidelines: Revision 2000: a statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation 2000; 102: 2284–2299. National Research Council. Committee on Diet and Health. Diet and Health: Implications for Reducing Chronic Disease Risk. National Academy Press: Washington, DC, 1989. Department of Health. Nutritional aspects of cardiovascular disease. Report of the Cardiovascular Review Group, Committee on Medical Aspects of Food Policy, London, 1994. Report of Health and Social Subjects #46. WHO Expert Committee on the Prevention of Coronary Heart Disease. Prevention of Coronary Heart Disease. Technical Report Series no. 678. World Health Organization. WHO: Geneva, 1982. WHO Expert Committee on Prevention in Childhood and Youth of Adult Cardiovascular Disease. Prevention in childhood and youth of adult cardiovascular diseasesFtime for action. Report of a WHO Expert Committee. World Health Organization Technical Report Series no. 792. WHO: Geneva, 1990.


18 U.S. Department of Agriculture/U.S. Department of Health and Human Services. Nutrition and Your Health: Dietary Guidelines for Americans, 5th edn. U.S. Department of Agriculture/U.S. Department of Health and Human Services: Washington, DC, 2000. 19 Institute of Medicine of the National Academies. Dietary Reference Intakes for Energy, Carbohydrates, Fiber, Fat, Protein and Amino Acids (Macronutrients). National Academies Press: Washington, DC, 2002. 20 Johnson CL et al. Declining serum total cholesterol levels among U.S. adults: The National Health and Nutrition and Examination Surveys. JAMA 1993; 269: 3002–3008. 21 Stamler J. The INTERSALT Study: background, methods, findings, and implications. Am J Clin Nutr 1997; 65 (2 Suppl): 626S–642S. 22 Stamler J. Epidemiologic findings on body mass and blood pressure in adults. Ann Epidemiol 1991; 1: 347– 362. 23 Klatsky AL, Friedman GD, Abraham BS, Gerard MJ. Alcohol consumption and blood pressure: Kaiser Permanente multiphasic health examination data. N Engl J Med 1977; 296: 1194–1200. 24 Dyer AR et al. Alcohol, cardiovascular risk factors and mortality: the Chicago experience. Circulation 1981; 64 (Suppl III): III-20–III-27. 25 Sacks FM, Rosner B, Kass EH. Blood pressure in vegetarians. Am J Epidemiol 1974; 100: 390–398. 26 Kimura N. Atherosclerosis in Japan. Epidemiology. In: Paoletti R, Gotto AM (eds). Atherosclerosis Reviews. Raven Press: New York, 1977, pp 209–221. 27 Yamori Y et al. Hypertension and diet: multiple regression analysis in a Japanese farming community. Lancet 1981; 1: 1204–1205. 28 Yamori Y et al. Dietary prevention of hypertension in animal models and its applicability to humans. Ann Clin Res 1984; 16 (Suppl 43): 28–31. 29 Kihara M et al. Interrelationships between blood pressure, sodium, potassium, serum cholesterol, and protein intake in Japanese. Hypertension 1984; 6: 736– 742. 30 Meyer TW, Anderson S, Brenner BM. Dietary intake and progressive glomerular sclerosis: the role of capillary hypertension and hyperperfusion in the progression of renal disease. Ann Intern Med 1983; 98: 832–837. 31 Meyer TW, Anderson S, Brenner BM. Dietary protein intake and the course of renal disease: the role of capillary hypertension and hyperperfusion in the pathogenesis of progressive glomerular sclerosis. In: Horan MJ, Blaustein M, Dunbar JB, Kachadorian W, Kaplan NM, Simopoulos AP (eds). NIH Workshop on Nutrition and Hypertension. Proceedings from a Symposium, Bethesda, MD, 1984, pp 217–229. 32 INTERSALT Cooperative Research Group (Elliott P, guest editor). The INTERSALT Study: an international cooperative study of electrolyte excretion and blood pressureFfurther results. J Hum Hypertens 1989; 3: 279–407. 33 Dyer AR et al. Body mass index and associations of sodium and potassium with blood pressure in INTERSALT. Hypertension 1994; 23: 729–736. 34 Stamler R et al, on behalf of the INTERSALT Cooperative Research Group. Higher blood pressure in adults with less education: some explanatory

35

36 37

38

39

40

41

42

43 44

45

46 47 48

49

factors. Findings of the INTERSALT Study. Hypertension 1992; 19: 237–241. Elliott P et al, for the INTERSALT Cooperative Research Group. INTERSALT revisited: further analyses for 24-hour sodium excretion and blood pressure within and across populations. BMJ 1996; 312: 1249– 1253. Stamler J et al. INTERSALT Study findings: public health and medical care implications. Hypertension 1989; 14: 570–577. INTERSALT Cooperative Research Group. INTERSALT: an international study of electrolyte excretion and blood pressure. Results for 24-hour urinary sodium and potassium excretion. BMJ 1988; 297: 319–328. Stamler J et al, for the INTERSALT Cooperative Research Group. Inverse relation of dietary protein markers with blood pressure. Findings for 10,020 men and women in the INTERSALT Study. Circulation 1996; 94: 1629–1634. Dyer AR, Elliott P, Chee D, Stamler J. Urinary biochemical markers of dietary intake in the INTERSALT Study. Am J Clin Nutr 1997; 65 (Suppl): 1246S– 1253S. Elliott P et al. Dietary protein and blood pressure: a report from the dietary and nutritional survey of British adults. Abstracts of the 14th Scientific Meeting of the International Society of Hypertension, Madrid, 1992, p S141. Stamler J, Caggiula AW, Grandits GA. Relation of body mass and alcohol, nutrient, fiber, and caffeine intakes to blood pressure in the special intervention and usual care groups in the Multiple Risk Factor Intervention Trial. Am J Clin Nutr 1997; 65 (1 Suppl): 338S–365S. Stamler J et al, for the MRFIT Research Group. Relationship to blood pressure of combinations of dietary macronutrients. Findings of the Multiple Risk Factor Intervention Trial (MRFIT). Circulation 1996; 94: 2417–2423. Zhou BF et al. Dietary patterns in 10 groups and the relationship with blood pressure. Chin Med J 1988; 102: 257–261. Zhang X, Cai R, Zhou B. The relationships of dietary protein, serum and urine free amino acids and blood pressure in three Chinese populations. Abstract book of the 3rd International Conference on Preventive Cardiology, Oslo, Norway, 27 June–1 July 1993, Abstract 5. Zhou B et al. The relationship of dietary animal protein and electrolytes to blood pressure: a study on three Chinese populations. Int J Epidemiol 1994; 23: 716–722. Sacks FM, Kass EH. Low blood pressure in vegetarians: effects of specific foods and nutrients. Am J Clin Nutr 1988; 48: 795–800. Dwyer JT. Health aspects of vegetarian diets. Am J Clin Nutr 1988; 48: 712–738. Prescott SL et al. A randomized controlled trial of the effect on blood pressure of dietary non-meat protein versus meat protein in normotensive omnivores. Clin Sci 1988; 74: 665–672. Kestin M, Rouse IL, Correll RA, Nestel PJ. Cardiovascular disease risk factors in free-living men: comparison of two prudent diets, one based on lactovegetarianism and the other allowing lean meat. Am J Clin Nutr 1989; 50: 280–287.



50 Beilin LJ, Burke V. Vegetarian diet components, protein and blood pressure: which nutrients are important? Clin Exp Pharmacol Physiol 1995; 22: 195–198. 51 Obarzanek E, Velletri PA, Cutler JA. Dietary protein and blood pressure. JAMA 1996; 275: 1598–1603. 52 Stamler J et al. Eight-year blood pressure change in middle-aged men: relationship to multiple nutrients. Hypertension 2002; 39: 1000–1006. 53 Appel LJ et al, for the DASH Collaborative Research Group. A clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med 1997; 336: 1117–1124. 54 Sacks FM et al, for the DASH-Sodium Collaborative Research Group. Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. N Engl J Med 2001; 344: 3–10. 55 MacMahon S et al. Blood pressure, stroke, and coronary heart disease. I. Prolonged differences in blood pressure: prospective observational studies corrected for the regression–dilution bias. Lancet 1990; 335: 765–774. 56 Grandits GA, Bartsch GE, Stamler J. Methods issues in dietary data analyses in the Multiple Risk Factor Intervention Trial. Am J Clin Nutr 1997; 65 (1 Suppl): 211S–227S. 57 Dyer AR, Shipley M, Elliott P for the INTERSALT Cooperative Research Group. Urinary electrolyte excretion in 24 hours and blood pressure in the INTERSALT Study. I. Estimates of reliability. Am J Epidemiol 1994; 139: 927–939. 58 Dyer AR, Shipley M, Elliott P for the INTERSALT Cooperative Research Group. Urinary electrolyte excretion in 24 hours and blood pressure in the INTERSALT Study. II. Estimates of electrolyte–blood pressure associations corrected for regression dilution bias. Am J Epidemiol 1994; 139: 940–951. 59 Elliott P, Stamler R. Manual of operations for INTERSALT: an international cooperative study on the relation of sodium and potassium to blood pressure. Control Clin Trials 1988; 9: 1S–118S. 60 Dennis B et al, for the INTERMAP Research Group. INTERMAP: the dietary dataFprocess and quality control. J Hum Hypertens 2003; 17: 609–622. 61 Bartels H, Bohmer M. Micro-determination of creatinine. Clin Chem Acta 1971; 32: 81–85. 62 Talke H, Schubert GE. Enzymatische harnstoffbestimmung im blut und serum im optischen test nach Warburg. Klin Wschr 1965; 43: 174–175. 63 Mann CK, Yoe JH. Spectrophotometric determination of magnesium with sodium 1-azo-2-hydroxy-3-(2,4dimethylcarboxanilido)-napthalene-10 -(2-hydroxybenzene-5-sulfonate). Anal Chem 1956; 28: 202–205. 64 Hubbuch A et al. Multicenter study of tina-quant-albumin in urine and b-N-acetylglucosaminidase (b-NAG) in urine. Workshop Munich, 29–30 November 1990. Wien klin Wschr Suppl 1991; 103 (Suppl. 189): 1–66.


65 Piez KA, Morris L. A modified procedure for the automatic analysis of amino acid. Anal Biochem 1960; 1: 187–201. 66 Nicholson JK, Wilson ID. High resolution proton NMR spectroscopy of biological fluids. Prog in NMR Spectrosc 1989; 21: 449–501. 67 Anthony ML, Lindon JC, Beddell CR, Nicholson JK. Pattern recognition classification of the site of nephrotoxicity based on metabolic data derived from high resolution proton nuclear magnetic resonance spectra of urine. Mol Pharmacol 1994; 46: 199–211. 68 Anthony ML et al. An NMR spectroscopic approach to the investigation of the biochemical responses of renal cells to nephrotoxins. Biomarkers 1996; 1: 35–43. 69 Holmes E et al. Automatic data reduction and pattern recognition methods for analysis of 1H nuclear magnetic resonance spectra of human urine from normal and pathological states. Anal Biochem 1994; 220: 284–296. 70 Nicholson JK, Lindon JC, Holmes E. ‘Metabonomics’: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica 1999; 29: 1181–1189. 71 Wilson ID, Nicholson JK, Lindon JC. The development and application of coupled HPLC–NMR spectroscopy. Adv Chromatogr 1995; 36: 315–382. 72 Schakel S, Buzzard M, Gebhardt S. Procedures for estimating nutrient values for food composition databases. J Food Comp Anal 1997; 10: 102–114. 73 Schakel SF et al, for the INTERMAP Research Group. Enhancing data on nutrient composition of foods eaten by participants in the INTERMAP Study in China, Japan, the United Kingdom, and the United States. J Food Comp Anal 2003; 16: 395–408. 74 Stamler J, Elliott P, Chan Q, for the INTERMAP Research Group. Appendix tables. J Hum Hypertens 2003; this issue (appendix tables from p 665–775). 75 Zhou BF et al, for the INTERMAP Research Group. Nutrient intakes of middle-aged men and women in China, Japan, United Kingdom, and United States in the late 1990s: The INTERMAP Study. J Hum Hypertens 2003; 17: 623–630. 76 Ueshima H et al, for the INTERLIPID Research Group. Differences in cardiovascular disease risk factors between Japanese and Japanese-Americans in Hawaii: The INTERLIPID Study. J Hum Hypertens 2003; 17: 631–639. 77 Dyer AR, Elliott P, Chan Q, Stamler J, for the INTERMAP Research Group. Dietary intake in male and female smokers, ex-smokers, and never smokers: The INTERMAP Study. J Hum Hypertens 2003; 17: 641–654.

INTERMAP - Nature

Recommend Documents