This retrospective cohort study used data from the National Health Insurance Service (NHIS) database. The NHIS is a single obligatory social insurance system covering health care service costs in Korea. The entire population residing in Korea subscribes to mandatory social insurance, and all medical care institutions claim medical fees. The claims include medical information, such as diagnosis, prescription, medication, and data collected from the NHIS database. Because the NHIS also provides general health checkups biennially for all insured adults (annually for nonoffice workers) and collects health checkup data, the NHIS database is useful for continuously observing the health status of the Korean population. The NHIS database was initiated in 2002 and we followed up data until 2020.

We included older adults who (1) were aged 65 years or older, (2) were newly diagnosed with hypertension (Korean Classification of Disease [KCD] code I10), and (3) underwent the NHIS health checkup in 2003-2004. We excluded patients who (1) had a history of hypertension or CVD, (2) were not prescribed medication for hypertension, (3) had missing blood pressure or any other covariate values, and (4) had fewer than 2 health checkups during the follow-up period. Characteristics of the study population are shown in Figure 1. A total of 68,901 participants were eligible and were followed up until 2020.

Ethics approval was obtained from the institutional review board of Yonsei University Health System (4-2021-1277), and the study complied with the Declaration of Helsinki. The NHIS database was anonymized to ensure confidentiality, and informed consent was not required because this was retrospective study using the NHIS database.

The independent variable in our study was the mean SBP measured during the health checkups. Blood pressure was measured using an automatic blood pressure monitor placed in front of the examiner after the participant had stabilized for at least 5 minutes. If the SBP was >120 mm Hg or the diastolic blood pressure was >80 mm Hg, the blood pressure measurement was repeated after at least 2 minutes. In this study, we categorized the SBP into 6 categories in 10 mm Hg increments from <120 mm Hg to ≥160 mm Hg, and 130-139 mm Hg was used as a reference category.

The dependent variables were all-cause and CVD mortalities. The CVD mortality was defined as death from myocardial infarction (KCD codes I21-23), heart failure (KCD code I50), hemorrhagic stroke (KCD codes I60-62), or ischemic stroke (KCD code I63). Data on the dates and causes of mortality were obtained from Statistics Korea.

Sociodemographic characteristics, health behaviors, blood test results, and hypertension medications were included as covariates. The sociodemographic characteristics included age, sex, and BMI. Age was classified as 65-74 years and 75 years or older. Health behaviors included smoking status, physical activity, and the Charlson Comorbidity Index. Blood test results included fasting glucose, total cholesterol, aspartate transaminase, and alanine transaminase. Hypertension medications included renin-angiotensin system blockers (angiotensin-converting enzyme inhibitors and angiotensin receptor blockers), beta-blockers, dihydropyridine calcium channel blockers, and diuretics.

Descriptive analyses were used to describe sample characteristics. Continuous variables were described as means (SDs), and categorical variables were described as numbers and percentages. Thereafter, Kaplan-Meier curves and Cox proportional hazards models were used to identify the relationship between SBP and all-cause and CVD mortalities after adjusting for other covariates. A subgroup analysis was performed to identify age-related differences in the risk of all-cause and CVD mortalities according to SBP. All statistical analyses were performed using SAS (version 9.4; SAS Institute Inc), and the significance level was set at P<.05.

In this retrospective cohort study, the rates of all-cause and CVD mortalities were significantly lower in older adults with hypertension and SBP within the range of 130-139 mm Hg than in the other SBP categories, except for those with SBP within the range of 120-129 mm Hg. These findings suggest the importance of managing SBP within the range of 120-139 mm Hg to lower all-cause and CVD mortality risks in older adults with hypertension. The optimal target SBP level was 120-139 mm Hg in both age groups (65-74 years and 75 years of age or older), and the impact of SBP on mortality was generally higher in older adults aged 65-74 years, indicating that more stringent SBP control is required in these patients than in their older counterparts.

Our results are consistent with the SBP goal of the 2022 KSH guidelines of <140 mm Hg [14]. The 2022 KSH guidelines have limitations in that the target SBP level for older adults with hypertension was established based on the findings of studies conducted on non-Korean participants [15-18]. Additionally, 1 of the main reasons for the unchanged target SBP level of the 2022 KSH guidelines compared to the 2018 KSH guidelines is that there is still insufficient evidence regarding the optimal target SBP level [14]. Our study provides empirical evidence for the optimal SBP level of older adults with hypertension in Korea to reduce all-cause and CVD mortality rates.

This study showed the presence of a J-curved relationship between SBP and CVD risk, indicating that lowering SBP to below 120 mm Hg increased all-cause and CVD mortalities compared with SBP within the range of 130-139 mm Hg. This result was consistent with an international cohort study on 22,672 patients with coronary artery disease [20] and the data from 2 randomized controlled trials (RCTs) on 30,937 patients with high cardiovascular risk [21], which showed that SBP within the range of 120-139 mm Hg had the lowest risk of cardiovascular events, and the risk increased at SBP <120 mm Hg, with a J-curve pattern. Furthermore, 19,110 older adults aged older than or equal to 60 years in the Singapore Chinese Health Study showed that SBP <120 mm Hg tended to increase CVD mortality risk compared with SBP between 120-139 mm Hg, although the risk estimates did not reach statistical significance [22]. However, most guidelines for hypertension management, including the ACC/AHA and KSH guidelines, do not specify the minimal level of SBP. It may be appropriate to suggest the minimum SBP level when scientists and professionals update the hypertension management guidelines.

Although several studies on the optimal target SBP level have been conducted, the results remain controversial across countries. The Systolic Blood Pressure Intervention Trial (SPRINT) in the United States showed that an SBP target of <120 mm Hg decreased major adverse cardiovascular events and mortality compared to an SBP target of <140 mm Hg in older adults with hypertension aged 75 years or older [15]. The Strategy of Blood Pressure Intervention in the Elderly Hypertensive Patients trial in China showed that an SBP target of 110-130 mm Hg significantly reduced major adverse cardiovascular events compared to an SBP target of 130-150 mm Hg [16]. The Berlin Initiative Study in Germany showed that SBP of <130 mm Hg is associated with increased all-cause mortality compared to SBP within the range of 140-149 mm Hg [23], whereas another cohort study in England showed that 6311 individuals aged 50-79 years and 1299 individuals aged 80-89 years had no significant associations between SBP levels and all-cause and CVD mortalities [24]. A possible explanation for the different results is ethnic disparities. The prevalence of hypertension and CVD mortality differs across countries [6,12,13]. Asian populations are known to be more vulnerable to hypertension because they tend to have an increased aortic pulse and stiffness than European populations [25], and Asians have a stronger association between blood pressure and CVD than Europeans [25]. Therefore, the target SBP level for hypertension may vary according to race and country.

The estimated risk for all-cause and CVD mortalities increased for every 10 mm Hg increase in SBP above 140 mm Hg, compared to within the 130-139 mm Hg range, regardless of age group, although the impact of SBP on mortality was slightly higher in individuals aged 65-74 years. However, the relationship between SBP and mortality in older adults remains controversial. A total of 2 RCTs, involving patients aged 75 years and older in the SPRINT [15] and 80 years or older in the Hypertension in the Very Elderly Trial [26], showed that a lower SBP was associated with decreased cardiovascular events. However, several cohort studies suggested that there was no association in older adults 75 years or older [27,28], and some even showed a negative relationship between SBP and cardiovascular events [29,30]. The discrepancies between the RCTs and cohort studies might be that the participants of RCTs were healthier than the general population of the same age because of the restricted inclusion and exclusion criteria [31]. The reason our study revealed an association between increased SBP and higher mortality, unlike other cohort studies, could be attributed to our deliberate selection of healthy older adults who were capable of undergoing health checkups on their own and had no previous history of hypertension or CVD. Therefore, further studies targeting diverse populations (ie, frail or multimorbid) are needed.

This study had several limitations. Because we included participants who underwent health checkups between 2003 and 2004, older adults who did not undergo health checkups were excluded. We adjusted for many covariates to alleviate bias, but potential age-related confounding variables, such as frailty, activities of daily living, and gait speed, remained due to data availability. In addition, adverse events of decreased blood pressure, such as renal dysfunction or orthostatic hypotension, were not considered in this study due to data limitations; therefore, we suggest that additional studies exploring optimal blood pressure should include adverse events as 1 of the outcomes. Finally, we used a conventional Cox hazard model with mean SBP measured during the health checkups as an independent variable to facilitate interpretation. However, since SBP is a time-varying variable, a time-dependent Cox model could also have been used. The strengths of this study included the use of the NHIS database, which is nationally representative and includes almost all Koreans, and a large sample size. Our results have sufficient statistical power and can be generalized to older Korean adults with hypertension. In addition, through a 17-year follow-up observation period, our study provided evidence that controlling SBP is crucial for long-term CVD-related health outcomes.

Controlling SBP in older adults with hypertension is important for improving overall and CVD-related health outcomes. However, hypertension guidelines regarding optimal SBP levels differ across countries; therefore, it is necessary to identify the appropriate SBP level among older Korean adults with hypertension. Our retrospective cohort study using national data showed that SBP within the range of 120-139 mm Hg was associated with the lowest risk of CVD and all-cause mortality in older Korean adults with hypertension; therefore, SBP should be controlled within this range in this cohort. Our study also suggested that lowering the SBP to below 120 mm Hg could increase all-cause and CVD mortalities, indicating that caution should be taken not to lower the SBP beyond 120 mm Hg. Additionally, SBP had a greater impact on CVD and all-cause mortality in older adults aged 65-74 years than in those aged 75 years and older; more stringent SBP management is thus required in the 65-74-year age group. This study provides critical knowledge that improves our understanding of the optimal SBP in older Korean adults with hypertension to reduce all-cause and CVD mortality rates. The KSH guidelines only recommend lowering SBP to <140 mm Hg; however, we suggest adding 120 mm Hg as the minimum value.