Association between physical activity behaviors and hypertension with comorbid multiple chronic diseases in South Korean middle-aged and elderly: an analysis based on association rule

Abstract

Background

The increasing prevalence of multiple chronic diseases in aging populations presents a serious public health concern. Hypertension, as one of the most common chronic conditions, is frequently observed alongside other chronic diseases such as diabetes, obesity, and dyslipidemia. Physical activity is widely acknowledged to be associated with the prevention and management of chronic diseases. However, few studies have examined how different Physical activity behaviors are associated with specific multimorbidity patterns involving hypertension. This study aimed to identify the patterns of association between physical activity status and co-occurring chronic conditions among hypertensive individuals in middle-aged and older adults in South Korea, using association rule analysis.

Methods

Our study utilized data from the Korean National Health and Nutrition Examination Survey (2016–2021), involving 21,043 participants aged 45 years and older. A total of 25 chronic diseases were included as related variables for the analysis using association rule mining.

Results

In the middle-aged and elderly population in South Korea, hypertension has the highest prevalence among all chronic diseases, with a rate of 45%. Our association rule analysis identified a total of nine chronic conditions as antecedents, with diabetes, obesity, and dyslipidemia being the most frequently observed. Furthermore, a subgroup comparison revealed that the number of association rules identified in the ‘physically inactive’ group (25 rules) was higher than that in the ‘physically active’ group (17 rules), and the overall confidence levels in the ‘inactive’ group were also higher. In terms of the frequency of antecedents, stroke, cardiovascular disease, and arthritis showed the largest increases.

Conclusion

Adequate physical activity is vital for preventing and managing hypertension and reducing its comorbidities, particularly high-mortality conditions like cardiovascular disease and stroke. Promoting lifestyle changes and monitoring metabolic indicators can significantly lower hypertension incidence, improve quality of life, and reduce mortality.

Introduction

World Health Organization (WHO) categorizes chronic or non-communicable diseases as those with “long duration and generally slow progression”. The primary factors contributing to these diseases are identified as smoking, insufficient physical activity, harmful alcohol consumption, unhealthy diets, and exposure to air pollution [1]. Chronic diseases often exhibit complex interconnections, frequently resulting in the co-occurrence of two or more such conditions within the same patient, a phenomenon termed “Multiple Chronic Diseases (MCD)” [2]. The prevalence of MCD increases with age. In the United States, the occurrence rate of MCD is only 18% among individuals aged 18–44, but it rises to 50% in the population aged 45–64. This rate is even higher in the elderly, with 81% of those aged 65 and above suffering from MCD [3]. Individuals with MCD not only frequently face poorer health outcomes [4] but also have higher rates of hospital readmission [5]. Moreover, healthcare expenses tend to rise exponentially with the number of chronic conditions [6]. The healthcare costs incurred by MCD patients are approximately 5.5 times higher than those for patients with a single chronic disease [7]. This substantial financial burden results in 21.2% of MCD patients spending over 10% of their household income on healthcare [8]. High out-of-pocket spending has long-term implications, significantly impacting the overall quality of life of these families [9].In South Korea, the prevalence of MCD among individuals over the age of 45 is 39.1%, with an increasing rate as age advances. The prevalence is higher in females than in males and is more common among vulnerable groups, such as those with lower income and education levels [10]. Therefore, preventing the progression of patients with general chronic diseases to MCD is a crucial component in addressing the broader issue of chronic diseases. This approach not only aims to improve health outcomes but also seeks to mitigate the economic impact on families and the healthcare system.

Essential hypertension, commonly referred to as hypertension or high blood pressure, is a condition characterized by abnormally elevated blood pressure with unclear causes. However, this elevation significantly increases the risk of major events such as cerebral, cardiac, and renal incidents. In industrialized countries, the likelihood of developing symptoms of hypertension (blood pressure exceeding 140/90 mm Hg) exceeds 90% over a person’s lifetime [11]. It is noteworthy that essential hypertension often does not occur in isolation but is closely associated with a range of cardiovascular disease risk factors, including aging, overweight, insulin resistance, diabetes, and hyperlipidemia. Moreover, subtle signs of target-organ damage, such as left ventricular hypertrophy, microalbuminuria, and cognitive dysfunction, often manifest in the early stages of hypertensive cardiovascular disease, indicating disease progression. However, more severe catastrophic events, like strokes, heart attacks, renal failure, and dementia, typically occur only after prolonged periods of uncontrolled hypertension [11]. According to a report published in August 2023, the number of outpatient visits for hypertension in South Korea in 2021 was 7.16 million. This represented a 4.7% increase (320,000 patients) in hypertension cases compared to the previous year. Over the decade from 2011 to 2021, the treatment costs for hypertension escalated by approximately 2 trillion Korean Won [12].

One of the sayings of Hippocrates, the father of modern medicine, is ‘If there is any deficiency in food and exercise the body will fall sick.’ He regarded physical activity (PA) as essential to a healthy life, akin to the necessity of food. This is indeed true, as insufficient exercise is a common risk factor for nearly all chronic diseases [13]. Lack of exercise increases the relative risk of coronary artery disease by 45%, stroke by 60%, hypertension by 30%, colon cancer by 41%, breast cancer by 31%, type 2 diabetes by 50%, and osteoporosis by 59% [14]. Consequently, the WHO recommends at least 150 min of moderate to vigorous physical activity (MVPA) or 75 min of high-intensity PA per week for substantial health benefits [15]. According to the Ministry of Health and Welfare, in 2021, only 47.9% of South Koreans met the benchmark, which is a decrease of approximately 10% from 2014’s 58.3% [16]. This suggests that nearly half of the population in South Korea may be in a state of “physical inactivity,” with this issue being particularly pronounced among the middle-aged and elderly. Given that this group is at high risk for hypertension, promoting lifestyle changes in this population could be an important and effective strategy for controlling and reducing the prevalence of hypertension and MCD in South Korea.

In this study, we aim to explore the characteristics of middle-aged and elderly hypertensive patients in South Korea with MCD under different PA behaviors using association rule mining. Given the rising prevalence of MCD and its significant health and economic impact, it is essential to understand the differences in how PA behaviors influence the progression of hypertension to MCD. Physical inactivity is a major risk factor for chronic diseases, and examining its effects on comorbidities can help identify key intervention strategies. This study not only aims to provide valuable insights into the relationship between PA, hypertension, and MCD but also seeks to support the development of targeted public health interventions to reduce the burden of these chronic conditions on individuals and the healthcare system.

Methods

Research subjects

The data for this study were obtained from the Korean National Health and Nutrition Examination Survey (KNHANES), conducted by the Korea Centers for Disease Control and Prevention (KCDC) from 2016 to 2021 [17]. This survey was approved by the Research Ethics Review Committee of the KCDC, ensuring strict adherence to ethical standards. KNHANES is a nationally representative, cross-sectional survey conducted by South Korea’s Ministry of Health and Welfare. It serves as a critical data source for health policy development in the country, with the primary aim of collecting accurate and representative statistical information on the health status, health-related behaviors, dietary habits, and nutritional intake of the Korean population. Of the 46,828 participants screened during the six years of data collection, a total of 21,043 individuals aged 45 years or older, with complete data on PA and chronic diseases, were included in this analysis.

Our study investigates 25 non-communicable chronic diseases, including hypertension, obesity, diabetes, dyslipidemia (encompassing hypercholesterolemia and hypertriglyceridemia), stroke, Cardiovascular Disease (myocardial infarction and angina), arthritis (including osteoarthritis and rheumatoid arthritis), osteoporosis, asthma, thyroid disorders, cancer (including stomach, liver, colon, breast, cervical, lung, thyroid, and other types of cancer), depression, allergic inflammation (including atopic dermatitis, allergic rhinitis, sinusitis, and otitis media), chronic eye disease (including cataracts, glaucoma, and macular degeneration), kidney disease, and liver cirrhosis. The diagnosis of hypertension, obesity, diabetes, hypercholesterolemia, and hypertriglyceridemia was determined based on objective medical examination criteria. Specifically, hypertension was defined as systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg, or being on antihypertensive medication. Obesity was defined as having a Body Mass Index (BMI) ≥ 25 kg/m². Diabetes was defined by a fasting blood glucose level ≥ 126 mg/dL, a diagnosis by a doctor, or the use of glucose-lowering medication or insulin injections. Hypercholesterolemia was defined as a total cholesterol level ≥ 240 mg/dL or the use of cholesterol-lowering medication, while hypertriglyceridemia was defined as a triglyceride level ≥ 200 mg/dL. For the remaining 20 diseases, diagnoses were based on self-reported data collected through a questionnaire. The questionnaire included two questions: “Have you been diagnosed with this disease by a doctor?” and “Are you currently suffering from this disease?” Participants who answered affirmatively to both questions were considered to have the corresponding chronic disease in our study.

In this study, data on PA behavior were collected using the Global Physical Activity Questionnaire (GPAQ) [18]. Participant responses provided an in-depth understanding of their PA patterns. Participants who did not meet the WHO recommendations for PA were classified as “physically inactive.” Specifically, the WHO recommends at least 2 h and 30 min of moderate-intensity PA per week, or 1 h and 15 min of vigorous-intensity Physical Activity per week, or an equivalent combination of both (with 1 min of vigorous-intensity activity being equivalent to 2 min of moderate-intensity activity). Conversely, participants who met or exceeded these recommendations were classified as “physically active.” Demographic and socioeconomic information, including age, sex, education level, household income, and residential area, was derived from responses to the basic health interview questionnaire administered as part of the KNHANES survey.

Analysis methods

Statistical analysis

Statistical analysis were performed using IBM SPSS Statistics, version 26.0. Graphical representations and figures were generated using OriginPro 2021. Group differences were assessed using the chi-square test, and independent samples t-tests were used for continuous variables, with a significance level set at 0.05 to determine statistical significance. All statistical tests were two-tailed, and p-values less than 0.05 were considered indicative of significant differences.

Association rules analysis

Association rule analysis were employed due to their ability to efficiently uncover complex and non-obvious relationships among multiple chronic diseases, without relying on the restrictive assumptions commonly required by traditional regression-based approaches [19]. In contrast to logistic regression and other parametric models, association rule analysis do not assume linearity, independence, or a predefined model structure, making them particularly suitable for exploratory investigations of multimorbidity and disease co-occurrence patterns. In recent years, this analytical approach has been widely applied in public health and healthcare research to identify hidden disease clusters and support data-driven decision-making [10, 20].

In our study, data mining analysis were conducted using IBM SPSS Modeler, version 18.0. Association Rules analysis derives a set of items with strong rules by utilizing various evaluation scales, and the relationship between these items is expressed as A→ B. Association Rules analysis employs various evaluation scales to derive a set of items with strong relational rules, expressed as A→ B. Here, A serves as the conditional variable, and B is the consequent, with both A and B potentially comprising one or multiple items. Support, confidence, and lift are the three most important evaluation scales that elucidate the relationship between A and B. By leveraging these metrics, our study endeavors to shed light on the intricate patterns of chronic disease co-occurrence, with the potential to inform future medical research and healthcare strategies.

Support refers to the probability of co-occurrence of items A and B. The higher the support, the higher the probability of A and B appearing simultaneously in the association rule. The calculation formula is as follows: Support (A→ B) = P (A, B). In this study, we define “support” as the probability of the co-occurrence of chronic diseases A and B among all subjects.

Confidence refers to the conditional probability of the occurrence of B under the premise that A appears. Confidence measures the reliability of association rules. If the confidence is high, B is more likely to appear after A, and the association rule is more credible. The calculation formula is as follows: Confidence (A→ B) = P (A, B)/P(A). In this study, “confidence” refers to the probability of the co-occurrence of chronic diseases B among patients with chronic diseases A.

Lift refers to the ratio of the probability of the occurrence of B in the presence of A to the probability of the occurrence of B under any condition, explaining the degree of influence of A on B. The calculation formula is as follows: Lift (A→ B) = P (A, B)/P(A)P(B) or lift (A→ B) = confidence (A→ B)/P(B). In this study, “lift” is defined as the confidence value divided by the probability of chronic diseases B. Given that different chronic diseases have widely differing basal prevalence rates, “lift” was the most important evaluation scale value in this study.

Results

Table 1 Demographic and health characteristics by physical activity status

Table 1 shows the demographic and health characteristics of participants stratified by physical activity status. Significant differences were observed between physically active and inactive groups. Physically inactive individuals were significantly older and had a higher proportion of females, lower household income (p < 0.01), and lower educational levels compared to physically active counterparts (p < 0.01). The prevalence of hypertension was significantly higher in the inactive group (p < 0.01), and the average number of chronic diseases was also significantly greater (p < 0.01).

Prevalence of chronic diseases

Table 2 Prevalence of chronic diseases in the study population

In this study, we analyzed the prevalence of various chronic diseases in the overall sample and among hypertensive patients. The results are presented in Table 2. The prevalence of hypertension in the overall sample was 45.2%, making it the most prevalent chronic disease. Dyslipidemia and obesity had prevalence rates of 41.4% and 31.3%, respectively, ranking second and third. The prevalence rates of diabetes and arthritis were 18.7% and 17.9%, respectively. Among hypertensive patients, dyslipidemia had the highest prevalence at 49.8%, followed by obesity at 39.7%, and diabetes and arthritis at 26.9% and 23.0%, respectively.

Except for thyroid disease, cirrhosis, and allergic inflammation, the prevalence of other chronic diseases was significantly higher among hypertensive patients compared to the overall sample (P < 0.05). The prevalence of all chronic diseases, excluding hypertension, was 77.8% in the overall sample and 86.5% among hypertensive patients, indicating that hypertensive patients were significantly more likely to have other chronic diseases (P < 0.01).

Results of association rules analysis

To initiate the association rule analysis, we first divided all study subjects into two subgroups based on whether they were “physically active” or not. Data mining analysis was then conducted separately for each subgroup. In the association rule analysis, we set hypertension as the target variable, with other chronic diseases as the input variables. The minimum conditional support was set at 0.5%, confidence at 10%, lift greater than 1.5, and the maximum number of antecedents was limited to five items. The following two sets of results were obtained from the analysis: one from the “physically active group” (Table 3) and the other from the “physically inactive group” (Table 4).

Table 3 Association rules for hypertension in physically active subgroup

In the ‘physically active’ subgroup, we conducted an association rule analysis for hypertension and other chronic diseases, as shown in Table 3. A total of 17 association rules were identified. Among these rules, the chronic diseases that appeared most frequently as antecedents were as follows: diabetes, which appeared 13 times; dyslipidemia, which appeared 12 times; and obesity, which appeared 8 times. Additionally, chronic eye disease, arthritis, osteoporosis, cardiovascular disease, and stroke also appeared as antecedents.

Among these rules, the rule with the highest support (5.7%) was that “diabetes and dyslipidemia” lead to hypertension. This indicates that, in the ‘physically active’ subgroup, 5.7% of the subjects had both diabetes and dyslipidemia, which was significantly associated with hypertension. The rule with the highest lift (1.91) was that “chronic eye disease, diabetes, and dyslipidemia” lead to hypertension. This indicates that when chronic eye disease, diabetes, and dyslipidemia are present together, the likelihood of hypertension is 1.91 times higher compared to other conditions.

Table 4 Association rules for hypertension in physically inactive subgroup

In the ‘physically inactive’ subgroup, we conducted an association rule analysis for hypertension and other chronic diseases, as shown in Table 4. A total of 25 association rules were identified. Among these rules, the chronic diseases that appeared most frequently as antecedents were as follows: diabetes, which appeared 18 times; obesity, which appeared 13 times; and dyslipidemia, which appeared 10 times. Additionally, arthritis, chronic eye disease, osteoporosis, cardiovascular disease, stroke, and allergic inflammation also appeared as antecedents.

Among these rules, the rule with the highest support (6.3%) was that “diabetes and obesity” lead to hypertension. This indicates that, in the ‘physically inactive’ subgroup, 6.3% of the subjects had both diabetes and obesity, which was significantly associated with hypertension. The rule with the highest lift (1.78) was that " Stroke and Obesity " lead to hypertension. This indicates that when Stroke and Obesity are present together, the likelihood of hypertension is 1.78 times higher compared to other conditions.

Fig. 1

Comparison of Association Rules between Physical Activity Subgroups

Red: Physically Inactive Subgroup Blue: Physically Active Subgroup.

Table 5 Occurrence counts of chronic diseases by physical activity subgroup

By comparing the types and frequencies of chronic diseases appearing as antecedents in Table 5, nine chronic conditions were identified: Diabetes, Obesity, Dyslipidemia, Arthritis, Chronic Eye Disease, Cardiovascular Disease, Stroke, Osteoporosis, and Allergic Inflammation. Notably, Allergic Inflammation did not appear as an antecedent in the “active” subgroup. Furthermore, the frequency of all chronic diseases increased in the “inactive” subgroup. Among these, Stroke, Cardiovascular Disease, and Arthritis showed the highest relative increases, with increases of 300%, 100%, and 100%. Moreover, from Fig. 1, it is evident that the confidence levels of rules in the “physically inactive” subgroup are generally higher. Specifically, all rules in the “physically inactive” subgroup have a confidence level above 70%, whereas the rules in the “physically active” subgroup are mostly distributed between 60% and 70%.

Discussion

In our research, we found that the prevalence of hypertension among individuals aged 45 and above in South Korea is as high as 45.2%, making it the most prevalent chronic disease among the middle-aged and elderly population in the country (Table 2). Moreover, the proportion of hypertension was significantly higher among the “physically inactive” population compared to the “physically active” population (Table 1). Fortunately, in 2019, the treatment rate for hypertension patients in South Korea was leading globally. The treatment rate for men reached 67%, only trailing behind Canada (76%) and Iceland (71%), ranking third; while the treatment rate for women in South Korea was the highest in the world at 77%. Moreover, from 1990 to 2019, South Korea experienced the greatest increase in hypertension treatment rates, with a 46% increase for women and a 50% increase for men [21]. This achievement can be attributed to South Korea’s comprehensive implementation of the HEARTS technical package [22], improvements in treatment protocols, drug and equipment supply, team-based care, patient-centered services, and enhancements in information systems, making significant progress [23]. Despite notable achievements in the treatment of hypertension, South Korea still has room for improvement in hypertension prevention. Prevention, early detection, and effective management of hypertension are among the most cost-effective interventions in healthcare, from both public health and economic cost perspectives, and should be prioritized. South Korea should prioritize the integration of hypertension prevention into its primary healthcare system and national health and welfare strategies. Additionally, greater resources should be allocated to populations with higher prevalence rates when designing targeted policies and interventions. In fact, the economic efficiency and cost ratio of improving hypertension treatment schemes is as high as 18 to 1 [23], further proving the importance of strengthening hypertension prevention and management.

In our data mining results, nine chronic diseases appeared as antecedents, namely Diabetes, Obesity, Dyslipidemia, Arthritis, Chronic Eye Disease, Cardiovascular Disease, Stroke, Osteoporosis, and Allergic Inflammation. This indicates that among the 25 chronic diseases included in our study, these nine diseases have a significant association with hypertension, suggesting that they may be major risk factors or comorbidities of hypertension. Notably, Diabetes, Obesity, and Dyslipidemia were the most frequently occurring chronic conditions in both the physically active and inactive subgroups. This is not only due to their inherently high prevalence in the middle-aged and elderly population but also because they share common pathological mechanisms, such as metabolic syndrome, insulin resistance, and unhealthy lifestyles (e.g., poor diet and lack of PA), which together increase the risk of hypertension. Weight gain is almost invariably associated with elevated blood pressure. The increase in blood pressure is closely related to the extent of weight gain, and even modest weight gain is linked to an increased risk of developing hypertension. However, there is considerable interindividual variability in the blood pressure response to weight gain, and not all obese individuals develop hypertension [24]. In an obese individual, cardiac output increases in proportion to the oxygen and perfusion demands. In individuals with abdominal obesity, elevated blood pressure can also result from increased peripheral vascular resistance. This increase may be related to the activation of the sympathetic nervous system and the renin–angiotensin system [24]. Additionally, high leptin levels observed in obesity may activate the pro-opiomelanocortin pathway and the sympathetic nervous system [25]. Environmental factors play a crucial role in the development of both hypertension and diabetes. Known risk factors for hypertension include obesity, high salt intake, diets rich in saturated fats and low in fruits and vegetables, stress, physical inactivity, smoking, excessive alcohol consumption, and certain medications. Therefore, there is a significant overlap between hypertension and diabetes, particularly when considering elevated blood pressure and abnormal blood glucose levels [26]. Dyslipidemia is a common lipid abnormality in diabetes but is not widely recognized as a typical abnormality in hypertension. However, hypertriglyceridemia can predict the development of hypertension, indicating that hypertension is not merely a vascular disease but is also closely linked to multiple metabolic abnormalities [27]. Furthermore, these conditions (diabetes, hypertension, dyslipidemia, and obesity) are all key components of metabolic syndrome. Metabolic syndrome is defined by several characteristics, including obesity, dyslipidemia (elevated triglycerides and reduced high-density lipoprotein cholesterol), hyperglycemia, and hypertension. Overall, metabolic syndrome is not an independent disease but rather a cluster of factors that increase the risk of cardiovascular disease [28]. Metabolic syndrome not only predicts the development of cardiovascular disease and diabetes but has also been shown to predict the onset of hypertension [29]. Based on the findings, it is evident that hypertension is closely linked to multiple chronic and metabolic conditions, such as diabetes, obesity, and dyslipidemia, as well as to various environmental factors. Therefore, effective prevention and management strategies should focus on maintaining a healthy weight, adopting a balanced diet rich in fruits and vegetables, increasing PA, reducing alcohol intake, quitting smoking, and managing stress. Regular monitoring of blood pressure and metabolic indicators, along with appropriate pharmacological interventions when necessary, are also essential for reducing the risk of hypertension and its associated complications.

By comparing the data mining results between the subgroups, we found that the number of discovered rules in the “physically inactive” subgroup is not only greater than that in the “physically active” subgroup but also that the confidence levels of these rules are generally higher. Specifically, all rules in the “physically inactive” subgroup have a confidence level above 70%, whereas the rules in the “physically active” subgroup are predominantly distributed between 60% and 70% (Fig. 1). This indicates that the associations between chronic diseases are more pronounced in the physically inactive subgroup. The greater number of rules and higher confidence levels suggest that, among individuals with a lack of PA, certain chronic diseases and hypertension are more likely to co-occur, implying a stronger tendency for these conditions to occur together. This could be related to the compounding effects of an inactive lifestyle, which may exacerbate multiple health problems and increase their interrelationship. PA exerts beneficial effects on hypertension and its comorbidities through various physiological mechanisms, including improving endothelial function [30], reducing systemic inflammation [31], enhancing insulin sensitivity, and moderating sympathetic nervous system activity [32]. Regular PA also contributes to favorable metabolic changes, such as improved lipid profiles, reduced adiposity, and better glycemic control, which collectively decrease cardiovascular and stroke risks [33].

Cardiovascular disease and stroke are the first and second leading causes of death globally, respectively. Hypertension is the most common risk factor for stroke, with approximately 64% of stroke patients having hypertension [34]. Additionally, hypertension, smoking, diabetes, and dyslipidemia are the main modifiable risk factors for cardiovascular disease. Among these, hypertension has the strongest etiological association with cardiovascular disease and also has a very high prevalence. Large cohort studies have shown that hypertension is a significant risk factor for heart failure, atrial fibrillation, chronic kidney disease, valvular heart disease, aortic syndromes, dementia, as well as coronary artery disease and stroke [35]. Given that hypertension is pathophysiologically established as a significant precursor and risk factor for cardiovascular diseases and stroke [36], the increased frequency of these conditions as antecedents in physically inactive individuals might represent a progression pathway facilitated by hypertension itself, rather than simply reflecting isolated co-occurrence [37]. Consequently, the stronger associations observed in physically inactive subgroups likely indicate that physical inactivity exacerbates the progression from hypertension to these more severe conditions. Interestingly, our study findings revealed that the frequencies of cardiovascular disease and stroke in the ‘physically active’ subgroup were only half and one-fourth, respectively, of those in the ‘physically inactive’ subgroup. This discrepancy was the largest observed among all chronic diseases analyzed. These results suggest that sufficient PA may effectively disrupt the link between hypertension and cardiovascular disease and stroke. Furthermore, compared to other chronic diseases, the benefits of PA appear to be greatest in reducing the association between hypertension and cardiovascular disease and stroke. Cross-national comparisons also reveal consistent yet context-specific patterns of multimorbidity involving hypertension. For instance, a recent population-based study in Brazil found that combinations of hypertension with diabetes and obesity were the most prevalent among older adults, particularly in those with low PA and socioeconomic disadvantage [38]. Similarly, data from a large-scale Chinese cohort showed that the most frequent multimorbidity dyad was also hypertension and diabetes, with physical inactivity serving as a significant behavioral risk factor [39]. These international findings echo the patterns observed in our South Korean sample, suggesting that the clustering of hypertension with metabolic disorders is a global trend. However, the degree of association and the underlying sociodemographic moderators may vary across regions, reinforcing the need for culturally tailored preventive strategies.

Despite the strengths of this study, several limitations must be acknowledged. First, as a data mining method, association rule analysis are inherently correlational in nature and cannot establish causality between variables. Thus, while meaningful co-occurrence patterns were identified between hypertension and multiple chronic conditions, these relationships should not be interpreted as causal without further longitudinal or interventional research. Second, the association rules derived in this study did not account for potential confounding factors such as age, sex, socioeconomic status, or other demographic characteristics. These factors are known to influence PA levels and chronic disease prevalence, and unlike regression-based models, association rule analysis do not allow for statistical adjustment. As such, the observed associations may be partially influenced by underlying population structure or lifestyle disparities. Third, the generalizability of our findings is limited to the South Korean context. Differences in ethnicity, culture, health systems, and lifestyle behaviors across regions may result in different multimorbidity patterns and PA profiles. Therefore, further studies are needed to validate these findings in other populations and settings.

Conclusions

Despite the significant progress made in hypertension treatment in South Korea, there is still room for improvement in hypertension prevention. Proactive prevention strategies, particularly through the adoption of a healthy lifestyle and timely monitoring and screening of metabolic indicators such as blood glucose, blood lipids, and body weight, can play a crucial role in significantly reducing the incidence of hypertension and its associated comorbidities. By comparing the subgroup data mining results, we found that sufficient PA was significantly associated with a lower prevalence of hypertension and its comorbid multiple chronic diseases. This effect is particularly evident in the prevention of high-mortality and high-disability chronic conditions such as cardiovascular disease and stroke. Therefore, adequate PA should be considered a key intervention for improving the quality of life and reducing mortality among hypertensive patients. Encouraging patients to increase PA through public health interventions can achieve significant success in preventing hypertension and its comorbidities.