INTRODUCTION
Systemic lupus erythematosus (SLE) is a rare, chronic autoimmune disease that can vary in severity from mild to potentially life-threatening. To improve its management, it is important to know the prevalence of the disease, including the clinical and socioeconomic characteristics of the affected population [1-3]. Both the magnitude and severity of SLE differ markedly with respect to sex, age, race, and region [4]. The prevalence and incidence of SLE are 100 to 150/100,000 and > 5/100,000 person-years, respectively [5-9]. A recent review of the epidemiology of SLE in the Asia-Pacific region showed that the prevalence and incidence were in the range of 4.3 to 45.3/100,000 and 0.9 to 3.1/100,000 person-years, respectively [10].
The majority of the epidemiologic data on SLE available in the literature is from single-center or cohort studies, but such studies are costly and time-consuming [11,12]. In contrast, the use of population-based data is more convenient and precise; one such database is the National Health Insurance (NHI) in Korea, which records extensive data of almost all Koreans. Previous epidemiologic studies on SLE in South Korea have focused on identifying a way to precisely estimate the prevalence and incidence of SLE using population-based surveillance systems [13,14].
The trend of peak age of the prevalence and incidence of SLE, sex, region, and socioeconomic status has not been reported thus far in Korea. Therefore, the present study aimed to determine the prevalence and incidence of SLE in South Korea, including temporal trends and variations in age, sex, region, and income from 2005 to 2015 by using the Korean NHI database.
METHODS
Study design and database
We conducted a nationwide cross-sectional study using the Korean NHI database of the Health Insurance Review Agency from 2005 to 2015. This database includes all claims data provided by the NHI and the Medical Aid programs [15]. The Korean NHI program is a compulsory social insurance scheme that covers approximately 97% of the Korean population; the remaining 3% are protected under the Medical Aid program [16]. The database consists of anonymized data that can be used for research purposes and reports the service code (admission or outpatient clinic), year (billing year), age (year old), region (province), sex (male or female), hospital (private clinic, hospital, or university hospital), department (internal medicine, dermatology, etc.), admission (admission or no admission), and billing number included all claimed national insurance data. The NHI database is credible and well-grounded for its use in managing the national statistical resources.
This study was approved by the Institutional Review Board of the National Institute for Bioethics Policy (NHIS-2018-1-109) and the Chonnam National University Hospital (CHUN-EXP-2017-235) in South Korea, and it was conducted according to the principles of the Declaration of Helsinki. Anonymized and de-identified information was used for analyses, and therefore informed consent was not required.
Identification of SLE cases
The NHI statisticians compiled data on all patients from institutes nationwide, who were treated under the International Classification of Diseases, 10th revision (ICD-10), code M32 for 13 years from 2002 to June 2015 (washout period was from 2002 to 2004). The South Korea government has enhanced benefit coverage for four major conditions (cancer, cardiovascular disease, cerebrovascular disease, and rare diseases), and SLE is one of the rare diseases. Within this system, the NHI has established a registration program for rare intractable diseases (code V136). Due to enforced privacy laws, detailed personal information could not be retrieved; instead, each patient was allocated an identification number.
Estimation of prevalence and incidence
For annual prevalence, the year-specific numerator included subjects with prevalent SLE in the specific calendar year, and the denominator included the mid-year population from the Korean National Statistical Office of the year. For annual incidence, the year-specific numerator included subjects with incident SLE in the specific calendar year, and the denominator included the mid-year population from the Korean National Statistical Office of the year. Crude rates, sex- and age-specific rates, standardized rates adjusted for sex and age using the 2015 mid-year population, and their 95% confidence intervals (CIs) were calculated. Incident cases of SLE were defined as those without SLE in a particular year (e.g., 2005) and the preceding 2 years (e.g., 2003 to 2004) that met the algorithm in that year (e.g., 2005) and the following year (e.g., 2006). Subgroup analyses were performed according to age (< 40, 40 to 64, and > 65 years), sex, region, and income. Income was stratified into 20 categories (from P1 [lowest] to P20 [highest]) and a medical aid group (with government-aided medical care).
Statistical analyses
Continuous variables are presented as mean and standard deviation, and categorical variables are presented as number and percentage. To compare the characteristics between the cohorts, the Student’s t test was used for continuous variables and the chi-square test was used for binary and categorical variables. All statistical tests were two-tailed, and the significance level was set at p < 0.05. All statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA) software program.
DISCUSSION
In this study, we determined the nationwide prevalence and incidence of SLE in South Korea by using the NHI claims data that covered almost all Koreans. Our main finding was that the incidence of SLE in South Korea decreased, whereas the prevalence of SLE increased in the last decade.
In our study, the average prevalence of SLE in South Korea was 28.02/100,000 person-years from 2005 to 2015. Overall, the prevalence of SLE is 30 to 50/100,000 people across Asian countries (except Taiwan [97.5/100,000 person-years]) [17-21] and 20 to 70/100,000 people in Europe [22-29]. The prevalence of SLE in Korea reported in our study was lower than that reported in both China (31.1 to 46.5/100,000 person-years) and Japan (7.0/100,000 men, 68.4/100,000 women) [30]. Similarly, our reported prevalence was lower than that reported in America; the prevalence of SLE is highest in America (100 to 150/100,000 people), including USA [5-9,31,32], Canada [33,34], and Puerto Rico [35], compared to all other countries worldwide [30].
The prevalence of SLE in South Korea and Taiwan was highest in the age group of 30 to 39 years. In contrast, the prevalence of SLE in the UK and Norway was highest in the age groups of 50 to 59 years and 60 to 69 years, respectively [20,24,36]. In addition, we confirmed female predominance in the prevalence of SLE, with approximately 10-fold higher prevalence in women than in men. The female-to-male ratio of patients with SLE was similar to that reported previously (~5:1 to ~10:1) [7,8,21,24,26,29,30,37]. Studies have reported that this ratio tends to increase with age and peaks during the childbearing ages, declining slowly thereafter [20,24,29,36].
In this study, we observed regional variation in the prevalence of SLE, with the highest prevalence in Jeju province, highest incidence in Gwangju province, lowest prevalence in Gyeongnam province, and lowest incidence in Incheon province. The incidence rate of most provinces decreased, except Gwangju and Gyeongbuk provinces. Consistent with our findings, the UK has reported regional variations in SLE prevalence and incidence [30]. Such geographical variation could reflect differences in the genetic mixture of the population or variation in environmental exposure. For example, countries near the equator are more exposed to ultraviolet radiation than those located further away from the equator, which may be an environmental trigger for SLE [38]. Korea has a relatively small geographical area and a homogenous race. Although regional variation has not been previously noted in Korea, the present study suggests that there is regional variation in even small areas and within a single race. One of possibility of this regional variation may be distribution of SLE specialists or hometowns of the principle investigators of national SLE cohort studies.
The annual prevalence of SLE showed an increasing trend and the peak age of prevalence increased from 2005 to 2015, which suggests improved treatment of SLE and decreased mortality rate. This is a common phenomenon in developed countries, and similar trends were observed in the UK [30]. Since Korea is an aging society, many diseases including SLE show a similar trend (right shift) in the peak age of prevalence.
In this study, the average incidence was 3.72/100,000 person-years from 2005 to 2015. The incidence of SLE was > 5/100,000 in the USA [6,8,9,20] and 3 to 5/100,000 person-years in Europe [22-26,26] and Asian countries [19-21]. Our result was comparable to the incidence of SLE in Hong Kong and Taiwan. In addition, the SLE incidence in the present study was approximately 6- to 8-fold higher in women than in men, which is similar to the reported difference of 5- to 8-fold in previous studies [19,20,24,26,29,36].
Similar to the case in Taiwan [20], the incidence of SLE increased continuously with age until it reached a peak, after which, it declined slowly, as did the prevalence. However, the incidence of SLE among female Norwegians exhibited a bimodal pattern, with the first and second peaks occurring at 16 to 29 years and 50 to 59 years, respectively. This pattern might be related to the use of contraceptive pills during the reproductive age or hormone therapy after menopause [36]. In other European countries, including the UK, Spain, and Denmark, the incidence of SLE increased until the age of 20 to 39 years and remained high thereafter up to the age of 50 to 69 years, although there are some differences in the age at which the incidence peaks [22,26,27]. In USA, the incidence of SLE appears to increase steadily with age [10,11,32]. In the Caucasian population, late-onset SLE is reported to be more prevalent than early onset SLE [39] and the clinical course is considered to be benign [40]. In contrast, in Asians, SLE is known to be more severe and have a higher mortality rate than in other populations [41]. Our study showed higher death rates in aged persons than in young people and in men than in women. These findings may explain the higher peak age in Korea compared to both Europe and USA and the higher incidence among in the elderly population compared to young people.
We observed an increasing trend in the incidence of SLE during the last decade. Several studies reported a significant increase in the incidence of SLE over several decades [6,29]; however, recent studies conducted over a relatively short period showed almost constant trends in the incidence of SLE [27,36]. The increasing trend reported in the present study might be explained by the increase in social awareness of the disease and wide coverage of the copayment beneficiaries program for rare diseases. In other words, prevalence can be statistically presumed to be a decrease in the number of deaths after diagnosis (prevalence = incidence × disease duration). Therefore, we could cautiously assume SLE management as better than in the past.
This study also showed that the income level did not affect the prevalence and incidence of SLE, except in patients without income who received medical aid. Medical aid group showed that the incidence is 4 to 5 times higher than the upper income, and the prevalence is about 2.5 to 3 times higher than the upper income. As a result, it seems that Korea’s excellent medical care system supports treatment of SLE better than in the past.
Despite our important findings, this study had a few limitations. First, some apparent limitations of using the NHI database stem from the lack of accurate diagnosis, because the original purpose of the data was to manage the NHI rather than use in academic evaluation. Second, the administrative claims data analyzed in this study were obtained when subjects visited healthcare institutions. Therefore, no information was available for SLE patients who did not visit a healthcare institution, which could underestimate the SLE burden. However, this may not have had a substantial impact on our findings, as the NHI claims data cover almost all Koreans and the wide coverage of the NHI system has made access to health institutions easy for the Korean population. Nevertheless, the strength of this study was that the trend of the prevalence and incidence of all Korean SLE patients was evaluated using national big database.
In summary, this study showed that the incidence of SLE in South Korea decreased in the last decade, whereas the prevalence increased. This finding of increased prevalence of SLE suggests that this disease is no longer rare and will have implications for future healthcare planning. Age, sex, and region were found to be risk factors for SLE. Our data confirmed the known predilection of SLE in women. The peak age of diagnosis is middle age, contrary to the generally held belief that lupus mainly targets young people. Furthermore, there is regional variation in the prevalence and incidence in different regions of Korea, and income is not a risk factor for SLE prevalence and incidence in Korea.