Association between healthy lifestyle and the occurrence of cardiometabolic multimorbidity in hypertensive patients: a prospective cohort study of UK Biobank

Figures

• 

  Fig. 1. Screening flowchart for baseline hypertensive population. Other covariates included missing the occurrence age of cardiometabolic multimorbidity. The only participant enrolled for the year (this occurred in 2006, resulting in a large difference between the groups in the year of enrolment, so it was removed). CHD coronary heart disease, BMI body mass index

• 

  Fig. 2. The relationship between a healthy lifestyle and cardiometabolic multimorbidity. We used four healthy lifestyle factors to estimate the lifestyle score, 0 points: current smoking, excessive drinking, less than two types of a healthy diet, no regular physical activity; 1 point, including non-current smoking, non-excessive drinking, more than 2 types of a healthy diet, regular physical activity. Then, we multiplied the binary lifestyle variables of each participant using the standardised weighted β coefficient and then summed and grouped participants into four ordered categories: very unhealthy (0 ≤ scores < 0.25, reference group), unhealthy (0.25 ≤ scores < 0.50), healthy (0.50 ≤ scores < 0.75), and very healthy (0.75 ≤ scores ≤ 1). CMM was defined as the occurrence of ≥ 1 of the following: CHD (myocardial infarction or angina), stroke (ischaemic stroke, cerebral haemorrhage, or subarachnoid haemorrhage), and diabetes. We adjusted for sex, ethnicity (white or non-white), Townsend deprivation index, sedentary time, antihypertensives drugs, lipid-lowering drugs, aspirin, and the year of attending the assessment centre

• 

  Fig. 3. The relationship between a healthy lifestyle and the survival years difference without cardiometabolic multimorbidity. The difference in CMM-free life expectancy refers to the number of survival years difference between 45 to 100 years that hypertensive participants did not have CMM (CHD, stroke, or diabetes). We used four healthy lifestyle factors to estimate the lifestyle score. 0 points: current smoking, excessive drinking, less than two types of a healthy diet, no regular physical activity; 1 point, including non-current smoking, non-excessive drinking, more than 2 types of a healthy diet, regular physical activity. Then, we multiplied the binary lifestyle variables of each participant using the standardised weighted β coefficient and then summed and grouped participants into four ordered categories: very unhealthy (0 ≤ scores < 0.25, reference group), unhealthy (0.25 ≤ scores < 0.50), healthy (0.50 ≤ scores < 0.75), and very healthy (0.75 ≤ scores ≤ 1). We adjusted for sex, ethnicity (white or non-white), Townsend deprivation index, sedentary time, antihypertensives drugs, lipid-lowering drugs, aspirin, and the year of attending the assessment centre. CMM cardiometabolic…