Body mass has been consistently associated with blood pressure in accu
lturated populations but not in lean populations with low blood pressu
res. In southwestern China, in 1989, the authors studied the relation
between body mass and blood pressure in three community-based random s
amples: rural Yi farmers (5,023 men, 3,218 women); Yi migrants (1,656
men, 919 women); and Han Chinese living in an urban setting (2,173 men
, 1,516 women). The Yi farmers had virtually no obesity or hypertensio
n and had little rise in blood pressure with age. The Yi migrants and
Han had a steeper slope of blood pressure with age and consequently mo
re hypertension. For men and women, respectively, mean body mass index
(weight (kg)/height (m)(2)) was 20.6 and 21.0 in the Yi farmers, 21.3
and 21.4 in the Yi migrants, and 21.0 and 21.4 in the Han. Both systo
lic and diastolic blood pressure were positively related to body mass
index in all six ethnicity-sex groups, and the association remained st
atistically significant after adjustment for age, heart rate, smoking,
alcohol intake, and physical activity. The change in mean blood press
ure for each kg/m(2) increase in body mass index, after adjustment for
community of residence, was 1.47 mmHg for systolic pressure and 1.13
mmHg for diastolic pressure. The association between body mass index a
nd blood pressure was greater in men than in women and greater in Yi m
igrants and Han than in Yi farmers. The percentages of hypertension at
tributable to overweight (body mass index greater than or equal to 25)
among the Yi farmers, Yi migrants, and Han, respectively, were 4.1%,
34.1%, and 24.0% for men and 0%, 26.2%, and 28.9% for women. Thus, eve
n in this lean Chinese population with a low mean blood pressure, body
mass was positively and independently associated with increased blood
pressure.