Research Article

Secular Trends of the Impact of Overweight and Obesity on Hypertension in Yi People: Yi Migrant Study, 1996–2015

Table 1

Characteristics of participants by body mass index category and time period, Yi Migrant Study, 1996–2015.

VariableBMIa category and time periodb
Underweight (BMI <18.5)Normal weight (BMI 18.5–23.9)Overweight/obesity (BMI ≥24.0)

Yi peopleP1 n = 179P2 n = 313P3 n = 314 for trendcP1 n = 1,084P2 n = 2,511P3 n = 1,696 for trendcP1 n = 218P2 n = 840P3 n = 1,257 for trendc
 Prevalence (%)12.098.549.610.35673.1968.5351.91<0.00114.7222.9338.48<0.001
 BMI (kg/m2)17.5417.5917.270.11720.9421.2221.32<0.00126.0926.6027.04<0.001
 Age, years37.2141.3150.32<0.00133.8238.5245.23<0.00140.6941.3547.16<0.001
 Female (%)44.1351.7666.24<0.00137.9249.7466.80<0.00139.9146.4366.43<0.001
 Hypertension (%)3.352.249.240.0223.695.6911.56<0.00115.1421.6726.730.026

Yi farmersn = 122n = 213n = 242n = 649n = 1,806n = 1,107n = 26n = 283n = 532
 Prevalence (%)15.319.2512.860.37181.4378.4558.85<0.0013.2612.2928.28<0.001
 BMI (kg/m2)17.4517.5517.280.54920.6421.1121.11<0.00126.2125.9526.83<0.001
 Age, years38.8444.2550.59<0.00132.1538.9044.36<0.00133.8138.5544.35<0.001
 Female (%)38.5251.1766.94<0.00134.6750.9465.31<0.00150.0066.7866.540.307
 Hypertension (%)2.462.358.680.0163.084.829.76<0.0017.6912.7220.860.076

Yi migrantsn = 57n = 100n = 72n = 435n = 705n = 589n = 192n = 557n = 725
 Prevalence (%)8.337.345.190.73463.6051.7642.50<0.00128.0740.8952.31<0.001
 BMI (kg/m2)17.7317.6617.220.033321.421.521.730.04726.0726.9227.20<0.001
 Age, years33.7435.0649.44<0.00136.3137.5346.88<0.00141.6242.7749.22<0.001
 Female (%)56.1453.0063.890.11942.7646.6769.61<0.00138.5436.0966.34<0.001
 Hypertension (%)5.262.0011.110.8554.607.9414.940.01316.1526.2131.030.014

Data are presented as mean or percentage, where appropriate. BMI: body mass index; P1: period 1; P2: period 2; P3: period 3. a: weight (kg)/height (m)2. b: period 1, 1996; period 2, 2007–2008; period 3, 2015. c: P-values for trend were adjusted age (except age-trend analysis) and sex (except female-trend analysis) using multiple linear regression, multiple logistic regression, and multinomial logistic regression, where appropriate.