Study population
The participants of the present study came from the prospective survey of China Kadoorie Biobank (CKB) in Qingdao. Details of the study have been previously reported[15-17]. A total of 35,508 residents, aged 30-79 (born in 1930-1970), completed the baseline survey in 2004-2008. Participants that self-reported IHD (n=1827), stroke (n=238), cancer (n=162) at baseline survey was excluded, the final analysis included 33,355 participants.
The ethics board of the University of Oxford, and the National, Shandong Provincial and the Qingdao Centers for Disease Control and Prevention in China all approved this study. All the participants in the survey had signed written informed consent.
Data collection
The laptop-based questionnaire was completed by trained health works, including sociodemographic information (age, education, occupation, household income, marital status), lifestyle (alcohol consumption, smoking status), family history, dairy products, and other diet frequency (rice, wheat, other staple foods, red meat, poultry, fish, eggs, fresh fruit, fresh vegetables, soybean, and preserved vegetables) in last 12 months. The date on the frequency of dairy (e.g. milk, yogurt) consumption included 5 groups (never/rarely, 1-3 d/month, 1-3 d/week, 4-6 d/week, daily).
Physiological measurements include body weight, height, waist circumference (WC), blood pressure, random glucose, etc. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2). For each individual, blood pressure was measured twice and taken as the average. A third measure was required if the blood pressure difference was more than 10 mm Hg between the first two measures, and take the average of the last two blood pressure values recorded. Random blood glucose levels were measured immediately following sample collection using the SureStep Plus System (Johnson & Johnson, New Brunswick, NJ, USA).
Follow-up for IHD and MCE
The primary outcome was the IHD (International Classification of Diseases-10 I20–I25) and MCE(major coronary event) incident, and IHD mortality. Which was ascertained through the Disease Surveillance Point System(DSPs) and the new national health insurance databases[18]. Participants were followed up from baseline until the date of IHD or MCE incidence, IHD mortality, loss to follow-up, or December 31, 2015, whichever came first.
Statistical Analysis
The series of characteristics of the participants were described with frequency (N) and percentages (%) according to categories of dairy consumption, using Student t-test for continuous variables and Chi-square test for categorical variables.
Multivariable Cox proportional hazards model was used to estimate the hazard ratio (HR) and 95% confidence interval (CI) of dairy consumption and IHD risk, which were adjusted for gender (male or female), age (continuous variable), education (below high school, high school and above), occupation, marital status, household income (<20,000 yuan or ≥20,000 yuan), diet frequency (egg, fresh vegetables, red meat, fresh fruit, poultry, soybean), smoking status (non-current smoking or current smoking), alcohol consumption (non-current drinking or current drinking), metabolic equivalent of task (MET), body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), and random glucose (continuous variable), family history of myocardial infarction (MI).
All p values were two-sided, p<0.05 was considered to be statistically significant. All analyses were performed using SPSS (version 25.0). All graphs were plotted using R 4.0.5 (https://www.R-project.org/).