A Study of Risk Factors for Heart Diseases Focused on Hypertension

Author Name NAWATA Kazumitsu (Hitotsubashi University)
Creation Date/NO. February 2023 23-J-006
Research Project Basic research for exploring the ideal medical intervention after the advent of the new coronavirus
Download / Links


Heart Disease (HD) is a very serious non-communicable disease and is the leading cause of death in the world. In Japan, HD is the second-leading leading cause of death. The prevention of HD is a very important issue and many studies have been done for analyses of risk factors and in particular the relationship with high blood pressure (BP) or hypertension. In 2017, the American College of Cardiology (ACC), the American Heart Association (AHA) and other organizations presented the new 2017 ACC/AHA guideline. Under the 2017 ACC/AHA guideline, the clinical criterion for hypertension was lowered to 130/80 mmHg from the previous criterion of 140/90 mmHg.

This paper analyzes the risk factors of HD with a focus on BP. The results of medical checkups of the JMDC Claims Database are used in the analysis. The database contained 13,157,681 medical checkups obtained from 3,233,271individuals. Individuals who had never been diagnosed with HD at year t and had data concerning HD histories in next year (i.e., year t+1) are analyzed, and the probability of onset of HD by the next year (hereafter, HD risk) is evaluated.

When the two variables, systolic blood pressure (SBP) and HD risk, are compared, the t-values of the probit model becomes very large and a clear positive relation between the two variables is found. However, when various variables such as age and gender are included in covariates, the significant relationship between SBP and HD risk disappears. This implies the possibility that the relationship of SBP and HD risk might be spurious. Moreover, when large-span age dummy variables such as age 40 to 64 and 65 or over are used to evaluate the effects of age, the estimate of SBP becomes significant. This fact indicates that the span of age dummy variables might be too large and that SBP might work as a proxy variable for age.

In the models with many covariates, the coefficients SBP and diastolic BP (DBP) are insignificant even at the 5% level. The estimate of the dummy variable that represents taking antihypertensive drugs is positive and highly significant; that is, taking antihypertensive drugs increases the HD risk. These results do not support the 2017 ACC/AHA guideline and it is necessary to re-evaluate the relationship between HD and hypertension. Especially, the 2017 ACC/AHA guideline heavily depends on the results of the Systolic Blood Pressure Intervention Trial (SPRINT). However, there are many problems in applying the results of SPRINT to the general population.

Other than BP, the estimates of hemoglobin A1c, urine protein, recent wight change, cholesterol, ALT (alanine aminotransferase) and sleep behavior become significant. The signs of most estimates are as expected. However, concerning cholesterol, higher levels reduce HD risk independent of types, so it might also be necessary to reevaluate the relationship between HD risk and cholesterol. Finally, the relationship between COVID-19 and antihypertensive drugs is described.