This is the first nationwide study in Thailand to evaluate the epidemiology of RA, including the prevalence and incidence, using data from the National Information and Communication Technology Center under the Ministry of Public Health. This dataset includes all healthcare providers and both outpatient and in-patient department visits from 2017 to 2020.
Between 2017 and 2020, the prevalence of RA in Thailand increased from 43.7 per 100,000 (95% CI: 43.2–44.2) to 99.2 per 100,000 (95% CI: 98.4–99.8), about 2.3 times higher over three years. This increase may reflect better laboratory diagnostics and a growing number of rheumatologists in Thailand, which improve case detection and reporting. Despite this upward trend, Thailand’s prevalence remains below the global average of 246.6 per 100,000 (95% CI: 222.4–270.8)4. Similar findings were reported by Almutairi et al., which showed that Asia had the lowest continental prevalence (300.0), while North America had the highest (700.0). Among individual countries, Thailand’s prevalence ranks among the lowest, slightly higher than Taiwan (50.0) and Nigeria (0.0)5, and similar to Saudi Arabia (60.0), Iraq (70.0), and Oman (80.0)12. In contrast, high-income countries, such as Canada (543.1)13Italy (540.0)14France (470.0)15Japan (650.0)16and South Korea (270.0)11report much higher prevalence rates. These discrepancies are due to the differing definitions and diagnostic codes used. Other studies used broader inclusion criteria, such as the ACR1987 classification or inclusion of ICD-9: 714 (RA), and ICD-10: M05 (seropositive RA) combined with M06 (other RA) and M08 (juvenile RA), whereas our study included only ICD-10: M05, specifically representing seropositive RA, resulting in a lower number of identified cases. Additionally, this variation may reflect the role of socioeconomic factors in disease prevalence, as high-income countries benefit from better healthcare access, leading to more frequent diagnoses. Compared to the 1998 study by Chaiamnuay et al., which reported a higher RA prevalence of 120.0 per 100,000 in rural areas of Thailand6, the difference may be explained by variations in study methods and case definitions. The earlier study relied on questionnaire, which may have overestimated prevalence. Furthermore, our study included only seropositive RA patients, excluding seronegative RA patients who account for 15–25% of RA cases13. Overall, the lower observed prevalence in our study may result from the stricter criteria that included only seropositive RA, potentially misclassifying seronegative RA as other types of arthritis, along with contributing factors such as limited diagnostic access, a shortage of rheumatologists, and the possibility that some patients seek treatment at local clinics or private hospitals outside the national database. In addition to socioeconomic factors, ethnic and racial differences may influence the development of RA. However, further research and additional data from Southeast Asian countries, which share similar genetic and environmental backgrounds, are needed to confirm this hypothesis. Currently, epidemiological data from this region remain limited.
Our study showed a slight increase in the incidence rate of RA, rising by 0.7% from 2018 to 2019 (18.6 [95% CI: 18.3–19.0] and 18.8 [95% CI: 18.4–19.1] per 100,000 person-years), consistent with a global annual increase of 0.89% from 1990 to 201914. In 2020, the incidence declined slightly to 18.2 (95% CI: 17.9–18.5) per 100,000 person-years, likely influenced by the COVID-19 pandemic’s impact on healthcare access and led to fewer reported cases. In 2018, the incidence of RA in Thailand was higher than the global incidence reported in 2017 (18.6 vs. 14.9 per 100,000 person-years)4. Furthermore, at the regional level, Thailand’s incidence was three times higher than the Southeast Asian average (18.6 vs. 6.2 per 100,000 person-years). These differences could be explained by differences in the study periods and diagnostic criteria. Previous studies conducted in the 1990 s used the ACR 1987 criteria, which had lower sensitivity for diagnosing RA compared to the ACR 2010 criteria implemented in our study20,21,22.
At the country level, Thailand’s incidence (18.6 per 100,000 person-years) was similar to Taiwan (20.2 per 100,000 person-years), suggesting possible shared ethnic factors23. However, Thailand’s incidence remained lower than that of South Korea, North America, and Italy, which reported rates of 42.0, 38.0, and 33.1 per 100,000 person-years, respectively14,17,24. These differences are likely due to variations in socioeconomic status, as supported by Won et al., who found that the combined RA incidence in both South and North Korea that was lower than the incidence in South Korea alone, a high-income country (28.5 vs. 38.0 per 100,000 person-years)25. These findings suggest that socioeconomic factors, such as urbanization and improved healthcare access, contribute to higher RA incidence rates. However, data on RA incidence in individual Southeast Asian countries remains limited, with most studies providing only broad regional estimates. Given the close similarities in ethnicity, genetics, and environmental characteristics between Thailand and neighboring Southeast Asian countries, additional studies are necessary to improve the understanding of RA incidence in this region.
The majority of RA patients in Thailand were women, consistent with findings from previous studies (Table 3). Our study showed a female-to-male ratio of 3.4:1, which remained stable throughout the 3-year follow-up period (3.5:1). This ratio is comparable to findings from Japan (3.2:1)16Korea (3.6:1)25and Poland (3.5:1)26and aligns with the global ratio range of 2–4:127. This finding suggests that female-specific factors, such as sex hormones (e.g., estrogen and progesterone), may play a role in RA pathogenesis. However, further research is needed to confirm this, as non-hormonal factors like sex chromosomes, microchimerism, and microbiome differences might also influence disease development28,29.
The mean age of RA patients in Thailand was 57.5 ± 11.7 years, which is older than in Korea (53.9 ± 13.5)17but younger than in Western Australia (62.8 ± 16.1)30,France (65.8 ± 16.8)15, and Finland (65.6 ± 14.3)31., The mean age in Thailand was similar to that reported from Spain (56.8 ± 13.9)32. When stratified by age, the highest prevalence and incidence among Thais were observed in the 60–69 age group, aligning with global prevalence, which peaks between 60 and 64 years. These results are also consistent with a previous study from South Korea, which reported the majority of patients in the 60–69 age group17,19. In contrast, Canada reported the highest prevalence in the 75–79 age group13. These findings support that RA is most common in late middle-age, with over 90% of patients being older than 40 years. This age-related trend strengthens the importance of improved management and prevention of RA progression, particularly in older adults as target groups.
Geographically, nearly half of Thailand’s RA patients in 2017 resided in the northeast, which reported the highest prevalence at 59.4 per 100,000. This concentration might be explained by the region’s larger area, higher population density, and potential genetic or environmental factors unique to the northeast. However, by the end of the study in 2020, the highest prevalence shifted to the southern region, with a prevalence of 135.9 per 100,000 (Fig. 1). Similarly, the southern region also recorded the highest incidence rates of RA during the study period, with rates of 30.4, 29.5, and 28.1 per 100,000 person-years in 2018, 2019, and 2020, respectively (Fig. 2). Although the southern region shows the highest prevalence and incidence rates, the actual number of patients is relatively small, only about 5,178 cases, compared to other regions. While other regions have lower prevalence than the southern region, their actual patient counts are much higher: the northeastern region has 27,963 cases, the northern region 13,216, and the central region 12,448. Therefore, these differences in patient numbers should be taken into consideration when formulating healthcare policies. This trend may be influenced by various factors, including environmental triggers, occupational exposures, improved healthcare access, and an increasing number of rheumatologists in the southern region. In contrast, at the provincial level, the northern provinces, including Lampang, Lamphun, and Phayao consistently showed high concentrations of RA patients. This could be influenced by local ethnic characteristics, genetic predisposition, specific environmental exposures, dietary habits, and local healthcare practices (Fig. 1). These findings suggest that geographic factors alone, such as climate differences among the northeastern, southern, and northern regions of Thailand, may not fully explain the development of RA. Despite the distinct climates, dry in the northeast, humid and tropical in the south, and cold in the north, the prevalence remains high across all regions. This indicates the potential influence of non-geographic factors, such as dietary habits, lifestyle, environmental triggers (e.g., infections), occupational exposures, and healthcare infrastructure, which may contribute to RA development. Therefore, further research is needed to investigate these non-geographic factors and their role in RA development, as geographic factors alone may not fully explain the observed patterns in RA prevalence and incidence.
Our study has several strengths. First, it is the first nationwide epidemiological research of RA in Thailand, providing a comprehensive overview of RA prevalence and incidence across diverse regions and population subgroups. Second, we utilized the largest national healthcare database in Thailand, maintained by the Ministry of Public Health. This database covers nearly 90% of hospitals and approximately 80% of the Thai population, including both inpatient and outpatient records, thereby ensuring that the findings are highly representative of the overall Thai population. Its integration into routine national health surveillance and policy highlights its utility for national-level research. Although no formal validation study has yet been published specifically evaluating the accuracy of RA diagnoses using ICD-10 codes within the ICT database in Thailand, this national-level database has been widely used for public health surveillance and epidemiological research. It has supported studies involving a wide range of chronic diseases, including diabetes, hypertension, tuberculosis, and elderly care10. Its widespread application in national health policy and research contexts supports its credibility as a reliable data source. Third, by analyzing demographic factors such as age, gender, and geographic region, this study provides important insights into the distribution and characteristics of RA in Thailand. It helps identify high-risk groups, such as middle-aged women and individuals residing in the southern, northeastern, and certain northern provinces. These findings offer valuable information to guide targeted public health strategies. Finally, our study contributes valuable regional data from Southeast Asia to the global RA research landscape and may offer relevant insights for other countries in Southeast Asia or the Asia-Pacific region with similar ethnic and geographic characteristics.
Despite its strengths, our study has some limitations. First, we included only seropositive RA cases (ICD-10 code M05). While this improves diagnostic specificity and reduces misclassification, it excludes seronegative RA cases (ICD-10 code M06), which represent approximately 15–25% of all RA patients. Since M06 is more variably used in clinical settings and may overlap with other forms of inflammatory arthritis, its exclusion may have resulted in an underestimation of total RA burden18. Second, our dataset did not include information from some private hospitals, local clinics, and certain university hospitals in the central region, which often manage complex cases outside the Ministry of Public Health system. The absence of these data may have led to missed cases and further underestimation of RA prevalence and incidence. Third, our study period included early 2020, when the COVID-19 pandemic disrupted healthcare access and diagnostic coding practices. As a result, some patients may have deferred care or been misclassified, led to under-ascertainment of cases and introduced bias into our incidence and prevalence estimates, thereby altering observed disease patterns. Future analyses comparing pre- and post-pandemic data are needed to quantify and adjust for these biases. Finally, the clinical information from the database was incomplete. While this study emphasizes geographic variations in RA prevalence and incidence, these patterns may not be fully explained by geographic factors alone. Further research on non-geographic factors, such as lifestyle, dietary habits, and genetic predispositions, is needed to provide a better understanding of these gaps.
