Chronic Obstructive Pulmonary Disease

National trend in the prevalence and mortality of COPD in South Korea from 2008 to 2017

Abstract

Background Existing studies on chronic obstructive pulmonary disease (COPD) in Korea lack full population coverage, relying on small sample sizes. Therefore, this study aims to investigate the prevalence and mortality of COPD in the entire Korean population.

Methods This serial cross-sectional study used national databases, linking the National Health Information Database (2008–2017) with Causes of Death Statistics. Identification of individuals with COPD used diagnostic codes (International Classification of Diseases-10: J41–J44) or a history of COPD-related hospitalisation, focusing on adults aged 40 and above. Prevalence and mortality rates, calculated for 2008–2017, encompassed both crude and age-standardised and sex-standardised measures. A multivariate Poisson regression model estimated the association between COPD and all-cause and cause-specific mortality, presenting incidence rate ratios (IRRs) and 95% CIs, using data from the year 2017.

Results Age-adjusted COPD prevalence exhibited a notable increase from 2008 (7.9%) to 2017 (16.7%) in both sexes. The prevalences of diabetes mellitus, hypertension, dyslipidaemia, ischaemic heart disease, cancer, osteoporosis and tuberculosis were higher in the COPD group than in the group without COPD (p for all <0.001). The incidence of stroke and myocardial infarction (p for all <0.001) and overall mortality were higher in the COPD group (adjusted IRR 1.23, 95% CI 1.22 to 1.24, p<0.001). In particular, incidence rate and risk of mortality due to lung cancer were higher than that of those without COPD compared with other cancer types (adjusted IRR 2.51, 95% CI 2.42 to 2.60, p<0.001). It was significantly higher the incidence rate and risk of mortality among group with COPD than those without COPD in lower respiratory disease (adjusted IRR 16.62, 95% CI 15.07 to 18.33, p<0.001), asthma (adjusted IRR 6.41, 95% CI 5.47 to 7.51, p<0.001) and bronchiectasis (adjusted IRR 11.77, 95% CI 7.59 to 18.26, p<0.001), respectively.

Discussion Our study showed that the prevalence of COPD is gradually increasing from 9.2% in 2009 to 16.7% in 2018. Furthermore, in overall (all-cause) mortality, it was significantly higher in group with COPD than in group without COPD. The mortality rate of group with COPD was much higher than the overall mortality rate but is gradually decreasing.

What is already known on this topic

  • The global prevalence of chronic obstructive pulmonary disease (COPD) is reported to be between 8.2% and 12.8%, 5-year average prevalence of COPD in Korea was 12.9%. Globally, COPD was the third leading cause of death in 2019, accounting for 6% of all deaths and trailing only ischaemic heart disease and stroke.

What this study adds

  • The prevalence of COPD in Korea has gradually increased from 9.2% in 2009 to 16.7% in 2018. The overall (all-cause) mortality rate was significantly higher in the COPD group than in the non-COPD group and all population, but is gradually decreasing.

How this study might affect research, practice or policy

  • Because the prevalence and mortality of COPD are high in Korea, physicians should make efforts to diagnose and treat COPD.

Introduction

Chronic obstructive pulmonary disease (COPD) is a significant public health concern that contributes significantly to global morbidity and mortality.1 2 This progressive respiratory condition is primarily characterised by irreversible airway damage and alveolar abnormalities, typically attributed to chronic exposure to harmful particles or gases, particularly tobacco smoke, occupational dusts and air pollution.3 The pathological changes associated with COPD result in persistent respiratory symptoms and progressive airflow obstruction, severely impacting patients’ quality of life.4

Although COPD is both preventable and treatable,3 a combination of underdiagnosis and misdiagnosis often results in many patients receiving no or inappropriate treatment. This issue is critical given that the global prevalence of COPD is reported to be between 8.2% and 12.8%, a figure that is anticipated to rise owing to factors such as increased smoking and an ageing population.5 6 Notably, in a Korean study using data from the Korean National Health and Nutrition Examination Survey (KNHANES) between 2015 and 2019, the 5-year average prevalence of COPD was 12.9%.7

Globally, COPD was the third leading cause of death in 2019, accounting for 6% of all deaths and trailing only ischaemic heart disease (IHD) and stroke.3 The mortality rate for COPD has decreased from 1999 to 2020 in the USA. However, COPD remains a significant cause of death, ranking as the sixth overall leading cause and fifth disease-related cause of death.8 In Korea, a population-based study reported a 5-year mortality rate of 25.4% for COPD, with a notable increase in mortality observed with advancing age.9 Although existing studies on the prevalence and mortality of COPD in Korea provide valuable insights, they are predominantly population-based and do not encompass the entire Korean population. Recognising this significant limitation, this study aimed to bridge this gap by investigating the prevalence and mortality rates of COPD across the broader Korean population, encompassing almost all Koreans.

Methods

Data sources

This serial cross-sectional study was conducted using a population-based database encompassing the entire South Korean population. We established a link between the National Health Information Database (NHID) for the years 2008–2017 and information regarding causes of death acquired from cause-of-death statistics obtained from Statistics Korea. The NHID, provided by the Korean National Health Insurance Service (NHIS), is a public database that includes information on healthcare utilisation, health screening, sociodemographic variables and mortality for the entire population of South Korea.10 In 2013, the NHIS covered 97.2% of the Korean population, and the Medical Aid system covered the remaining 2.8%.11 This system not only provides universal insurance coverage to the entire population of South Korea but also maintains records of individual health information.

The cause-of-death statistics are produced to identify the number and causes of death in the Korean population, providing fundamental data for the establishment of healthcare policies. Because the date of death was already included in the NHID, we only extracted data on the causes of deaths occurring up to the first half of 2018 for eligible participants in the 2008–2017 NHID. The causes of death were selected from the underlying causes of death recorded on death certificates according to the WHO’s International Classification of Diseases (ICD). Then, these causes were further classified according to the Korean Standard Classification of Diseases, 7th revision.12

Study population and definition of COPD

The linked database includes more than 50 million individuals aged 0 years and older annually. However, for this study, we restricted our analysis to adults aged ≥40 years. The NHID does not include clinical parameters, such as spirometry data, necessary for diagnosing COPD. Therefore, we identified individuals with COPD using diagnostic codes. Individuals were classified as having COPD if they had two or more diagnostic codes for COPD (ICD-10 codes J41–J44) separated by a period of time or if they had experienced at least one hospitalisation episode due to COPD.

Mortality

Death cases were tallied from 1 January to 31 December of each year. Mortality rates were calculated for all causes, as well as for nine major causes as follows: neoplasms (C00–D48); endocrine, nutritional and metabolic diseases (E00–E90); mental and behavioural disorders (F00–F99); diseases of the nervous system (G00–G99); diseases of the circulatory system (I00–I99); diseases of the respiratory system (J00–J99); diseases of the digestive system (K00–K93); diseases of the musculoskeletal system and connective tissue (M00–M99) and injury, poisoning and certain other consequences of external causes (S00–T98). Additionally, mortality rates for 13 selected specific causes were calculated as follows: malignant neoplasm of bronchus and lung (C34); other malignant neoplasms (C00–D48 except C34); diabetes mellitus (DM) (E10–E14); hypertension (I10–I15); IHD (I20–I25); heart failure (I50); cerebrovascular disease (I60–I69); lower respiratory diseases (J40–J47); asthma (J45–J46); bronchiectasis (J47); pneumonia (J12-J18); osteoporosis (M80-M82); vertebral or pelvic bone fractures (S22–S29, S30–S39, M48).

Other variables

We included data on factors that could potentially influence the prevalence and mortality of COPD, including age, sex, income level, residence and comorbidities. Income level was classified according to the NHIS insurance contribution, which is, in turn, based on the monthly wage for employees (employee-insured) and metrics of household wealth (eg, income, property and car ownership) for the self-employed (self-employed-insured). The categories comprise Medical Aid recipients as well as the first (lowest 25%), second, third and fourth (highest 25%) insurance contribution quartiles. Residential areas were classified as metropolitan, urban or rural based on the ZIP code. The cumulative burden of comorbidities for each individual was assessed using the Charlson Comorbidity Index (CCI). Individuals were classified into four groups based on their CCI scores: 0, 1–2, 3–4 and ≥5 (indicating the most severe comorbidities). Specific conditions, including DM, hypertension, dyslipidaemia, stroke, IHD, acute myocardial infarction (AMI), cancer, osteoporosis and tuberculosis (TB), were also investigated as comorbidities.

Statistical analyses

For each year from 2008 to 2017, we calculated both the crude and age-standardised and sex-standardised prevalence rates of COPD. Subsequently, all analyses were performed separately for men and women. Sociodemographic characteristics and comorbidities were compared between individuals with and without COPD using a χ2 test for categorical variables and a t-test for continuous variables. This analysis was conducted using data from 2017, the most recent dataset available. Categorical variables were expressed as frequency and percentage (%), while continuous variables were presented as mean and SD. To compare mortality rates between individuals with and without COPD and assess mortality trends over the past decade, we analysed crude and age-standardised and sex-standardised all-cause mortality rates for each year from 2008 to 2017. These analyses were also stratified by sex, and mortality rates were presented as the number of deaths per 100 000 persons. Cause-specific mortality rates for nine major causes, including 13 subcategories, were analysed using data from 2017. The association between COPD and overall or cause-specific mortality was a Poisson regression model adjusted for relevant factors. The presence of overdispersion in the Poisson model was tested and confirmed to be valid. Incidence rate ratios (IRRs) and 95% CIs for overall and cause-specific mortality were adjusted for age, sex, income level, residence and CCI. To compare the prevalence and mortality rates across the years, we standardised the rates for age and sex using the direct standardisation method. The midyear Korean population of 2005 was used as the standard population for standardisation. All analyses were conducted using the SAS statistical software, V.9.4 (SAS Institute, Cary, North Carolina, USA). Two-sided p values of <0.05 were considered significant.

Results

Trends in COPD prevalence from 2008 to 2017

Figure 1 illustrates the trends in the prevalence from 2008 to 2017. The crude prevalence of COPD increased significantly, from 8.2% to 18.8%, indicating a consistent upward trend across both sexes (figure 1A). Similarly, the age-adjusted COPD prevalence also exhibited a notable increase, rising from 7.9% to 16.7% over the same period, with parallel trends observed in both men and women (figure 1B). The total number of participants aged >40 years enrolled from 2008 to 2017 is shown in online supplemental table 1.

Figure 1
Figure 1

Trends in crude and age-standardised prevalence of COPD between 2008 and 2017. COPD, chronic obstructive pulmonary disease.

Comparison of baseline characteristics between individuals with and without COPD

The most recent dataset, from the year 2017, was used to describe the baseline characteristics of the enrolled patients with and without COPD (table 1). In 2017, the prevalence rate of COPD was 18.8%. There were significantly more COPD cases among women than among men (56.7% vs 43.3%, p<0.001). Patients in the COPD group were older (64.5±13.1 vs 56.0±11.4 years, p<0.001) and more likely to report a lower income and live in rural areas (p for all <0.001). Furthermore, the COPD group had a higher proportion of people with disabilities, including both severe and mild disabilities, than the group without COPD (p<0.001). Comparisons of comorbidities between the groups are shown in table 2. The prevalences of DM, hypertension, dyslipidaemia, IHD, cancer, osteoporosis and TB were higher in the COPD group than in the group without COPD (p for all <0.001). The incidence of stroke and AMI was also higher in the COPD group than in the group without COPD (p for all <0.001).

Table 1
|
Basic characteristics of enrolled participants with and without COPD (2017)
Table 2
|
Comparison of comorbidity profiles on enrolled participants with and without COPD (2017)

Trends in COPD mortality rates from 2008 to 2017

Figure 2 shows the mortality rates from 2008 to 2017. The overall mortality rate exhibited a decreasing trend across the entire population during this period. Notably, the COPD group had higher crude and age-standardised mortality rates than both the overall population and the group without COPD (figure 2A,B). This trend was evident in both men and women with COPD, although the mortality rates demonstrated a decreasing pattern over the study period (figure 2C–F).

Figure 2
Figure 2

Trends in COPD mortality between 2008 and 2017 among groups with or without COPD. COPD, chronic obstructive pulmonary disease.

Comparative mortality risks with and without COPD

Table 3 presents the IRR for mortality in the groups with and without COPD. In terms of overall mortality, the rate was higher in the COPD group than in the group without COPD (adjusted IRR 1.23, 95% CI 1.22 to 1.24, p<0.001). The incidence rate and risk of mortality due to neoplasms were higher in the COPD group than in the group without COPD (adjusted IRR 1.11, 95% CI 1.09 to 1.13, p<0.001). In particular, the incidence rate and risk of mortality due to lung cancer were higher in the COPD group than in the group without COPD (adjusted IRR 2.51, 95% CI 2.42 to 2.60, p<0.001). This difference was notable when compared with other cancer types. The incidence rate and risk of mortality due to respiratory diseases were higher in the COPD group (adjusted IRR 2.44, 95% CI 2.38 to 2.50, p<0.001). In particular, the incidence rates and risk of mortality were significantly higher in the COPD group than in the group without COPD for lower respiratory diseases (adjusted IRR 16.62, 95% CI 15.07 to 18.33, p<0.001), asthma (adjusted IRR 6.41, 95% CI 5.47 to 7.51, p<0.001) and bronchiectasis (adjusted IRR 11.77, 95% CI 7.59 to 18.26, p<0.001).

Table 3
|
Incidence rate and risk of mortality among groups with and without COPD (2017)

Discussion

Our study showed that the prevalence of COPD has gradually increased from 7.9% in 2009 to 16.7% in 2018. Trends from 2008 to 2017 revealed a general decrease in mortality rates across the population. However, participants with COPD consistently had higher mortality rates, both crude and age-standardised, than those without COPD. Moreover, the comparative mortality risks indicated that the overall mortality, including specific causes such as neoplasms and respiratory diseases, was significantly higher in the COPD group. The incidence rate and risk of mortality due to lung cancer and lower respiratory diseases, such as asthma and bronchiectasis, were particularly higher in the COPD group, underscoring the severe impact of COPD on health outcomes.

The first significant observation in this study pertained to the prevalence of COPD in Korea and its trends in the entire population. We found a prevalence of 16.7%, which notably exceeds the global prevalence of 10.3% reported by the Global Initiative for Chronic Obstructive Lung Disease.3 Previous research on the prevalence of COPD in Korea, based on the KNHANES, indicated rates ranging from 11.5% to 13.6%.7 13 14 However, these studies were cross-sectional. It is also possible that the COPD prevalence observed in our study might have been somewhat overestimated compared with certain Korean studies that used ICD-10 codes (J43 and J44) employed by the Korea Health Insurance Review and Assessment Service to evaluate appropriate COPD management. Nonetheless, the ICD-10 code range (J41–J44) we employed is not a significant concern, as it is a definition commonly used in other studies, including previous systematic reviews.15 16

The present study, which encompassed the entire Korean population, provides a more comprehensive and detailed reflection of the COPD situation in Korea. A contributing factor to this higher prevalence rate and rising trend could be various reasons, including rapid urbanisation, increased exposure to fine dust particles, cigarette smoking and increased life expectancy.17 Ageing is a recognised risk factor for COPD development,7 18 19 as it is associated with declining lung function and physiological changes, such as cellular senescence, genomic instability and mitochondrial dysfunction.18 20 In fact, the proportion of individuals aged 60 years or older consistently increased in this population, rising from 32.8% in 2008 to 38.5% in 2017 (online supplemental table 1). These demographic shifts were significantly correlated with the onset of COPD, as our findings suggest; approximately 62.4% of the COPD group were aged 60 years or older, compared with 33% in the non-COPD group.

Moreover, since 2013, Korea has been conducting assessments for appropriate COPD treatment and management annually for medical institutions. Efforts have also been made to enhance public awareness and early detection of COPD, which has been under-diagnosed and under-treated compared with other chronic diseases, through collaborative initiatives between academic societies and the government.21 22 These initiatives have led to increased awareness and diagnosis of COPD, which seems to be reflected in prevalence.

A notable finding of this study was the trend in COPD mortality in Korea. The age-standardised mortality rate of COPD was 1830 per 100 000 people in 2008 and has steadily decreased since then to 1046 per 100 000 people in 2017. This downward trend aligns with that reported by previous studies23–25 and likely results from improved COPD management and smoking cessation efforts.3 Particularly, various public health projects aimed at improving the early detection and management of patients with COPD, as discussed in the preceding paragraph, may have positively influenced the mortality rate. These projects involve raising awareness about COPD among the general public, developing educational materials targeting high-risk populations and primary care physicians, emphasising the significance of inhaler treatment and managing COPD according to clinical practice guidelines.22

Additionally, the gradually improving smoking rate in Korea may also correlate with the decrease in mortality among patients with COPD. Cigarette smoking is known as the most common and well-documented risk factor for COPD. Although we could not directly confirm the smoking rate in the current data, national statistics on cigarette smoking rates in Korea indicated a significant decrease, especially from 47.8% to 38.1% among male adults during the study period.26 Smoking cessation is the most effective method to prevent COPD, slow its progression, minimise acute exacerbations and reduce mortality. However, the COPD mortality rate remains two times as high compared with that of the overall population in Korea. Despite being less recognised than common fatal diseases like cancer, heart disease and stroke, COPD poses a significant and persistent public health concern. Therefore, reducing the risk of developing COPD and improving outcomes for those already living with the disease is crucial. This emphasises the need for effective COPD prevention and management strategies.

Another finding of this study was that patients with COPD had a significantly higher mortality risk across various health conditions than those without COPD. Specifically, mortality due to neoplasms and respiratory diseases was notably higher in the COPD group. This finding suggests that COPD may directly exacerbate these conditions or weaken the body’s resilience. Similarly, the TORCH study indicated that 35% of deaths among patients with COPD were due to chronic pulmonary disease.27 Additionally, a Korean study reported a 17.5% mortality rate due to chronic lower respiratory disease in patients with COPD.9 In the present study, the risk of mortality due to respiratory diseases was more than double, and the risk of mortality due to lower respiratory diseases was 16 times higher. This can be linked to mucosal damage from smoking or other environmental irritants, which makes lungs of individuals with COPD more susceptible to infections and exacerbations due to bacterial pathogens or viruses, particularly lower respiratory infections.28 29 Collectively, these findings highlight the severe impact of COPD on health outcomes. They call for a more nuanced understanding of the disease’s progression and its interactions with other health conditions. This, in turn, necessitates a more integrated approach to managing COPD, addressing not only the primary disease but also the comprehensive health of patients. This involves vigilant monitoring and prevention of serious comorbidities.

Despite these clinical implications, this study had several limitations. First, this study did not define COPD based on pulmonary function tests. COPD is defined as airflow limitation in pulmonary lung function, indicated by a post-bronchodilator forced expiratory volume in 1 s/forced vital capacity ratio of <0.70.3 Instead, the study relied on ICD-10 codes (J41–J44) to identify COPD cases. Nonetheless, efforts were made to minimise inaccuracies by including patients with two or more ICD-10 codes or those hospitalised for COPD. However, this approach may not capture the full spectrum of the disease, particularly mild cases or individuals not seeking medical care. Another limitation is the higher prevalence of COPD reported among women in this study. This finding contrasts with most other studies in which the prevalence of COPD was typically higher in men. This discrepancy might be due to differential healthcare utilisation patterns, as reports suggest that women use hospitals more than men.30 31 Similar to this study, previous study also showed a higher prevalence of COPD in women.32 It is believed that these results may have occurred because women live longer than men in the Korean population (online supplemental table 1) and their medical utilisation rate is high. This difference in healthcare-seeking behaviours could have influenced the study’s findings, potentially leading to an overestimation of the prevalence of COPD among women. Finally, our analysis did not fully encompass a range of influential factors, such as socioeconomic status, smoking habits, occupational exposure and lifestyle choices. Specifically, we collected smoking-related variables from individuals who underwent national health checkups in the current or previous year to ascertain information about their smoking habits. However, despite these efforts, the smoking status of more than half of the subjects could not be confirmed based on the 2017 data. These factors play crucial roles in determining the prevalence and severity of COPD. Consequently, the absence of these considerations in our study is a limitation, as they are essential for a comprehensive understanding of the impact of COPD. Therefore, although our findings contribute significantly to the existing body of knowledge, they should be interpreted with an awareness of these unaddressed variables.

In conclusion, our study indicates a gradual increase in COPD prevalence over the past decade. This rise may be attributed not only to an increase in the population exposed to COPD risk factors but also to improvements in diagnosing and treating previously overlooked cases of COPD. Furthermore, we found that both overall (all-cause) and cause-specific mortality rates were significantly higher in the COPD group compared with the non-COPD group. However, the gap in mortality between the COPD and non-COPD groups has been gradually narrowing in Korea, reflecting ongoing efforts to enhance COPD management. However, there is still a need for effective prevention and management strategies to reduce the personal and socioeconomic burdens associated with COPD exacerbations. A further longitudinal cohort study is necessary to further validate our findings and evaluate the actual level of treatment and management after the diagnosis of COPD, considering various potential confounding factors that were not addressed in this study.