Skip to main content
Download PDF
Download ePub

The spatial analysis of multimorbidity in Thai Cohort Study

Archives of Public Health volume 83, Article number: 120 (2025) Cite this article

Abstract

Background

This study used Thai Cohort Study (TCS) data to investigate the spatial and sociodemographic determinants of multimorbidity (two or more chronic conditions coexistence on one person) prevalence in Thailand in 2013.

Methods

Crude and age-adjusted prevalence were calculated for each province. Hotspot analysis was conducted to identify regions with statistically significant hotspots and cold spots, including areas without significant clustering. Then, ordinal logistic regression was used to identify sociodemographic background variables that predict hotpots.

Results

The highest age-adjusted provincial level prevalence of multimorbidity was in Sing Buri (18.26%). Sak Lek District in Phichit Province also had the highest age-adjusted district level prevalence of multimorbidity at 37.13%. The cold spots region in crude and age-adjusted prevalence of multimorbidity were clustered in Southern Thailand. Forty-eight districts were identified as hotspots in both crude and age-adjusted multimorbidity prevalence, 19 of which are in Bangkok (the capital). Population density (person/km2, odd ratio, provincial level: OR:1.00, 95% CI: 1.00–1.01; district level: OR: 1.01, 95% CI: 1.00–1.01), Aging index (provincial level: OR:1.03, 95% CI: 1.01–1.04; district level: OR: 1.01, 95% CI: 1.00–1.01), and average educational years (provincial level: OR:1.92, 95% CI: 1.07–3.48; district level: OR: 1.27, 95% CI: 1.02–2.26) were greater in hot spots areas.

Conclusion

This study shows that the prevalence of multimorbidity in Thailand is positively correlated with the degree of development of the region. Spatial cluster analysis provides new evidence for policymakers to design tailored interventions to target multimorbidity and allocate health resources to areas of unmet need.

Peer Review reports

Text box 1. Contributions to literature

• There is limited evidence on multimorbidity and its prevalence in developing countries is lacking.

• Spatial analysis and quantifying socio-demographic background factors confirm that the places with high prevalence of multimorbidity in Thailand are concentrated in developed areas such as the capital.

• Prevention and intervention treatment of multimorbidity specified by regional prevalence differences are necessary.

Introduction

Multimorbidity, defined as the coexistence of two or more chronic diseases in an individual, is becoming increasingly common with global population ageing [1]. It leads to reduced physical functioning and quality of life, increased healthcare costs and disease burden, and a higher risk of disability and mortality in adults [2]. Consequently, multimorbidity has emerged as a major challenge to global public health [3].

The prevalence of multimorbidity is not uniformly distributed across geographical regions. Spatial analyses play a critical role in understanding health conditions across different areas by providing insights into the social and environmental determinants of disease prevalence [4]. When applied to multimorbidity, spatial analysis helps identify geographic clusters with elevated prevalence and evaluates the contextual and environmental factors contributing to these patterns [5].

Similar to many low- and middle-income countries (LMICs), Thailand is undergoing a health transition, with the rising prevalence of multimorbidity driven by demographic shifts, lifestyle changes, and urbanization, posing significant challenges to the healthcare system [3]. For instance, multimorbidity may be more prevalent in economically developed regions [4, 6]. However, few studies have investigated its spatial distribution within Thailand [6]. Understanding this distribution is essential for identifying high-risk areas, assessing environmental risk factors, and developing targeted interventions for affected populations [6,7,8].

Spatial analysis of multimorbidity can inform evidence-based decision-making by pinpointing “hotspot” areas where interventions can be prioritized to optimize healthcare delivery and resource allocation [9, 10]. Furthermore, understanding the spatial distribution of multimorbidity supports the development of context-specific policies and strategies tailored to the unique challenges of different regions [11]. The aim of this study is twofold: (1) to examine the geographic variation of multimorbidity in Thailand, and (2) to quantify the potential impact of socio-demographic contextual factors on areas with a higher risk of multimorbidity. We hypothesize that the distribution of multimorbidity is not random but shows distinct geographic clustering, with higher prevalence in areas characterized by specific socio-demographic attributes.

Methods

The study population

The Thai Cohort Study (TCS) is a nationwide project aimed at investigating Thailand's ongoing health risk transition, which refers to the shift in disease burden from infectious diseases to non-communicable diseases (NCDs). Launched in 2005, the project has successfully enlisted over 87,000 individuals who initially enrolled at Sukhothai Thammathirat Open University (STOU) in Thailand [12, 13]. The participants in this study were drawn from across the country and spanned an age range of 15 to 87 years at baseline. These respondents were reasonably representative of both STOU's student population and the broader Thai population in terms of median age, income, regional distribution and ethnic diversity [14]. However, it's worth noting that this cohort displayed a lower mean age (around 41 years old) and boasted a higher level of education compared to the average adult population in Thailand [15]. Additionally, a larger proportion of STOU students were residing in urban areas [15].

Study design and data collection

Data were collected through three mailed questionnaires. The first was in 2005 (baseline) and the follow-up surveys were in 2009 (midpoint) and 2013 (endpoint). At each round of follow-up, approximately 70% of cohort members were retained in the study [14, 15]. The present study is a cross-sectional study based on the 2013 mailed questionnaire, which surveyed a total of 42,785 respondents to obtain district (second administrative level) and province (first administrative level) information from their home addresses [14, 15].

The definition of multimorbidity

In this research, multimorbidity refers to the concurrent presence of two or more chronic diseases in an individual. This widely adopted definition is commonly used in epidemiological studies [15]. During the 2013 follow-up survey, respondents were asked if a physician had diagnosed them with any of eleven specified diseases. A"yes"response indicated the presence of these conditions, while"no"denoted their absence [15]. These eleven diseases encompassed diabetes, high cholesterol, hypertension, ischemic (coronary) heart disease, stroke, kidney disease, liver cancer, lung cancer, stomach cancer, colon cancer, and breast cancer. Additionally, participants provided their height and weight, allowing for the calculation of the body mass index (BMI) [15]. BMI values were classified based on the specific cut-off for Asian populations, where individuals with a BMI over 25 kg/m2 were considered obese, following the World Health Organization (WHO) recommendations for this group [16]. Therefore, multimorbidity was determined when an individual reported"yes"for two or more of these twelve conditions (which included obesity) [15, 16].

Spatial data

The administrative boundaries, used as the spatial unit of analysis were downloaded from the Human Data Exchange-Thailand—Subnational Administrative Boundaries [17]. The province level is the first administrative level in Thailand, with 76 in total. Bangkok, as the capital, is the special administrative area and is not classified as a province (Bangkok has 50 districts). The second administrative level is the districts, of which there are 878 (the 50 districts of the capital being not included) [18].

To improve accuracy, we only analysed multimorbidity prevalence in Provinces or Districts where 6 or more TCS participants resided [18]. All Thai provinces (including the Capital Special Administrative Region) had more than five participants, but only 870 of the 928 districts (including the 50 districts in capital) had more than five participants. We listed the 58 districts with less than or equal to 5 participants as blank.

Age-adjusted multimorbidity prevalence

Due to data limitations and differences in age structure between the Thai Cohort Study (TCS) and the national population, we calculated both crude and age-standardized multimorbidity prevalence at the district and provincial levels [4, 18]. Age-standardized prevalence was derived using demographic data from the 2013 Thai capital region, provinces, and districts obtained from the National Statistics Office [19] and Official registration systems [20]. We first calculated the percentage of the population in each age group based on national data. Then, the crude prevalence of multimorbidity for each age group within the TCS sample across capital areas, provinces, and districts was standardized using the national age structure. Specifically, the crude prevalence for each age group in the TCS was multiplied by the corresponding population proportion from national data, and the results were summed to obtain the age-adjusted prevalence [4]. These adjusted and crude prevalence estimates were subsequently mapped using ArcGIS Pro 2.6 (ESRI, Redlands, CA).

Spatial analysis of multimorbidity

Spatial Autocorrelation (Global Moran's I)

The Global Moran's I statistic was used to assess whether there was a clustering of multimorbidity prevalence across districts and provinces in Thailand. The null hypothesis was that multimorbidity was randomly distributed across districts and provinces. The Global Moran's I value generally ranged from − 1 (perfect dispersion) to + 1 (perfect clustering). The Global Moran's I value would be used in conjunction with the G statistic to provide a better understanding of local spatial patterns [4, 18, 21].

High/Low Clustering (Getis-Ord General G)

Since Global Moran's I can assess spatial autocorrelation but cannot specifically determine high or low clustering, the Getis-Ord General G index is used to better identify the clustering of spatial data. In other words, Getis-Ord General G examines whether districts and provinces with similar multimorbidity prevalence are spatially clustered together. Similar to Global Moran's I, the Getis-Ord General G value ranges from − 1 to 0, indicating low clustering (areas with dissimilar prevalence), and from 0 to + 1, indicating high clustering (areas with similar prevalence). A value of 0 represents randomness, meaning no significant clustering is present [4, 18, 21].

Hotspot analysis (Getis-Ord Gi*)

The study utilized the Getis-Ord Gi* statistic in conducting Hotspot analysis to pinpoint statistically significant areas of multimorbidity prevalence within Thailand's districts and provinces. Within this context, a hotspot indicated a cluster of high prevalence (indicated by shades of red on map), while a cold spot denoted a cluster of low prevalence (indicated by shades of blue on map). The adjacent regions that are not statistically significant were displayed in white on the map [18, 22]. Spatial weights were assigned using contiguity edges corners technique. Contiguity defines neighbouring polygons as those touching each other, including polygons that share edges or corners, which are considered neighbours in the analysis. To address variations in neighbour counts among features, the study used row standardization to create proportional weights [18].

Statistical analysis

We used ordinal logistic regression to investigate the impact of socio-demographic contextual factors, including population density (persons/km2) [17, 23], aging index (an indicator comparing the ratio of elderly population (aged 60 and above) to child population (under 15 years old)) [19, 20], monthly average income per household (Baht) [19, 24,25,26], average annual personal income (Baht) [25,26,27,28], the poverty line (expenditure, Baht) [19, 24,25,26], average years of education [24, 28, 29], and housing density (n/km2) [24, 26, 28] on the adjusted prevalence of multimorbidity hotspot areas. Ordinal logistic regression was chosen because the dependent variables (the geographic classification of multimorbidity hotspots), is inherently ordinal, with “2” representing hot spots, “1” for non-significant areas, and “0” for cold spots. This method allowed us to examine the impact of these contextual indicators on hot spots, cold spots, and non-significant areas. Comparing to ordinal logistic regression, the logistic regression with binary outcome would collapse categories, losing important information, while multinomial logistic regression treats categories as unordered, ignoring their natural progression. As for spatial autoregressive models, they were not necessary for our study as our dependent variable is ordinal, not continuous. Our focus is on examining socio-demographic factors influencing multimorbidity hotspots, rather than spatial dependence between regions. And we have already used hotspot analysis to capture the spatial distribution. These contextual indicators for 2013, at the provincial and/or district level, were obtained from the website of the Thai government and relevant authorities. The analysis was performed using Stata statistical software (version 16.0, Stata Corp, College Station, Texas, USA).

Results

The characteristics of members in Thai Cohort Study (TCS)

Table 1 presents the characteristics of the 2013 Thai Cohort Study (TCS), which included 42,785 members, with more females (23,455) than males (19,330). The largest age group was under 39 years (51.3%). Fewer people were single (36.0%) than living with partners (64.0%). The East region had the lowest population (8.0%). About 79.7% had some university education. Regarding income, 9.6% earned less than 7,000 Baht per month, while 69.2% lived in households earning over 20,000 Baht monthly.

Table 1 Demographic and health characteristics of individuals in Thai Cohort Study (TCS) in 2013
Full size table

Spatial analysis of multimorbidity prevalence

The crude prevalence of multimorbidity by province (Fig. 1A) showed that 47 provinces have a crude prevalence of less than 10%, with the lowest being Ranong Province at 3.51%, and the highest being Sing Buri Province at 15.82%. As for the crude prevalence of multimorbidity in districts (Fig. 1C), 178 districts had zero crude prevalence rate, while Don Chedi District in Suphan Buri Province had the highest crude prevalence of multimorbidity at about 54.55%.

Fig. 1
figure 1

Geographic distribution of multimorbidity prevalence in Thai Cohort Study (TCS) in 2013. Figure 1 A Crude prevalence of multimorbidity by provinces in TCS; Fig. 1B Age-adjusted prevalence of multimorbidity by provinces in TCS; Fig. 1 C Crude prevalence of multimorbidity by districts in TCS and Fig. 1D Age-adjusted prevalence of multimorbidity by districts in TCS

Full size image

In terms of the age-adjusted prevalence of multimorbidity, Fig. 1B indicated that the highest prevalence was Sing Buri Province (18.26%), but the lowest age-adjusted prevalence of multimorbidity was Ranong Province (2.20%). Sak Lek district in Phichit Province had the highest age-adjusted prevalence of multimorbidity at 37.13% but there were also 178 districts with zero age-adjusted prevalence of multimorbidity (Fig. 1D).

The spatial analysis

In assessing the spatial autocorrelation of multimorbidity by province and district, the Global Moran's I test revealed statistically significant clustering patterns (p < 0.05) between provinces and districts in both crude (provinces: Moran’s I: 0.19, z-score: 2.69, p < 0.05; districts: Moran’s I: 0.09, z-score: 3.90, p < 0.001) and age-adjusted prevalence (provinces: Moran’s I: 0.11, z-score: 1.72, p < 0.1; districts: Moran’s I: 0.11, z-score: 5.22, p < 0.001) of multimorbidity.

Getis-Ord General G (High/Low Clustering) test showed the similar results with the Global Moran's I test. The pattern of crude and age-adjusted prevalence of multimorbidity in provinces (crude: Observed General G: 0.06, z-score: 2.45, p < 0.05; age-adjusted: Observed General G: 0.06, z-score: 2.05, p < 0.05) and districts (crude: Observed General G: 0.001, z-score: 4.33, p < 0.001; Observed General G: 0.001, z-score: 6.21, p < 0.001) would be high-clustering.

Hotspot analysis (Getis-Ord Gi*) was used to identify statistically significant areas with higher rates of multimorbidity, referred to as “hotspots”. Hotspot is defined as a geographic area with a significantly higher prevalence of multimorbidity compared to its surroundings. Statistically significant hotspots were identified in Nonthaburi, Ang Thong, Sing Buri, Chai Nat and Nakhon Pathom provinces for both crude (Fig. 2A) and age-adjusted (Fig. 2B) multimorbidity prevalence. Statistically significant cold spots were identified in Krabi, Phangnga, Surat Thani, Ranong, Pattani, Yala and Narathiwat provinces, all of which are located in Southern Thailand. These provinces were consistently identified as cold spot regions in both crude (Fig. 2C) and age-adjusted analyses (Fig. 2D). Forty-eight districts were identified as both hotspots for crude and age-adjusted multimorbidity prevalence, 19 of which are located in Bangkok. Additionally, 30 districts were identified as cold spots.

Fig. 2
figure 2

The Hotspot analysis (Getis-Ord Gi) of multimorbidity prevalence in Thai Cohort Study (TCS) in 2013. Figure 2 A The Hotspot analysis (Getis-Ord Gi*) of crude prevalence by provinces in TCS; Fig. 2B The Hotspot analysis (Getis-Ord Gi*) of age-adjusted prevalence by provinces in TCS; Fig. 2 C The Hotspot analysis (Getis-Ord Gi*) of crude prevalence by districts in TCS and Fig. 2D The Hotspot analysis (Getis-Ord Gi*) of age-adjusted prevalence by districts in TCS

Full size image

The analysis of ordinal logistic regression

The population density (odd ratio, provincial level: OR:1.00, 95% CI: 1.00–1.01; district level: OR: 1.01, 95% CI: 1.00–1.01), the ageing index (provincial level: OR:1.03, 95% CI: 1.01–1.04; district level: OR: 1.01, 95% CI: 1.00–1.01), and average educational years (provincial level: OR:1.92, 95% CI: 1.07–3.48; district level: OR: 1.27, 95% CI: 1.02–2.26) were greater in hotspot places (see Table 2). However, in Supplementary Table 1, in crude multimorbidity prevalence, only the aging index in hot spots was associated with higher multimorbidity prevalence at the provincial level (OR: 1.02, 95% CI: 1.01–1.05) and the district level (OR: 1.03, 95% CI: 1.02–1.07). In addition, monthly average income per household (Baht) (OR:1.01, 95% CI: 1.00–1.03) would be greater in provinces (hot spots) of high age-adjusted multimorbidity prevalence (see Table 2).

Table 2 Ordinal logistic regression of socio-demographic factors associated with age-adjusted multimorbidity prevalence in 2013
Full size table

Discussion

This study utilized data from the Thai Cohort Study (TCS) to explore the spatial variation of multimorbidity. The adjusted prevalence was found to be higher in Bangkok and metropolitan areas and lower in several southern provinces. Hotspots were mainly concentrated in Bangkok and its surrounding regions, whereas cold spots were predominantly located in the south.

We observed that multimorbidity prevalence was higher in areas with greater population density, higher household incomes, and longer average years of education. This finding suggests that more prosperous areas tend to have higher prevalence, which contrasts with some studies showing higher multimorbidity in socioeconomically disadvantaged areas [30,31,32]. Typically, low socioeconomic status (SES) is associated with poor living conditions, environmental pollution, and limited access to healthcare [31,32,33]. However, in some low- and middle-income countries, such as Kenya and South Africa [34], higher SES is linked to higher multimorbidity rates due to better access to healthcare, sedentary lifestyles, stress, and polypharmacy [33, 34].

Regions with more medical resources may also have higher rates of diagnosis and reporting, leading to a higher observed prevalence due to more comprehensive case recording [1, 35, 36]. In Thailand, disparities in wealth and healthcare access, particularly between Bangkok and other regions, may contribute to these patterns. Wealthier regions, like Bangkok, benefit from better healthcare infrastructure and thus show higher reported rates, whereas underreporting may occur in less-resourced southern regions [1, 37].

Our study identified population density, aging index, and average years of education as the three key contextual factors associated with multimorbidity hotspots, with education showing the strongest effect based on odds ratio (OR) results. These findings underscore the need for tailored prevention and management strategies, particularly in under-resourced regions [38].

Strengthens and Limitations

This study is the first to apply spatial analysis to multimorbidity using data from the Thai Cohort Study. This innovative approach provides fresh perspectives and contributes to the literature on multimorbidity prevalence. Spatial analysis offers a geographic lens to reveal differences and patterns across regions, enhancing understanding of location-based factors and informing regional healthcare interventions [4, 6].

Nonetheless, the study has some limitations. First, although the composition of TCS participants is broadly like the general Thai population, there are still differences that may limit generalizability [4]. Second, data completeness varied across regions, with some areas having missed or incomplete information on certain diseases, affecting the comprehensiveness of the analysis [12, 14, 15]. Third, potential confounding factors such as environmental or socio-economic variables, could not be fully controlled, potentially influencing the findings [4, 6]. Furthermore, due to data availability, socio-demographic information was analyzed at the provincial rather than district level, which may obscure finer-scale spatial patterns. The cross-sectional design also limits causal interpretation. Lastly, as multimorbidity diagnoses in the TCS were self-reported, the study may be subject to recall and reporting bias, which could impact data accuracy and the reliability of conclusions [12, 14, 15].

The future research

Although this study used data from 2013, the spatial patterns of multimorbidity and its associations with socio-demographic factors provide valuable insights into long-term trends. While specific health outcomes may have evolved, fundamental determinants such as population density, aging index, and education level, remain relevant [1, 4, 5]. Future research could employ hierarchical spatial models to examine cross-level interactions or variance components, offering a deeper understanding of multimorbidity distribution [6, 7]. Moreover, incorporating additional factors covering environmental exposures (e.g., air pollution), urbanization, lifestyle behaviours, and healthcare access (including the number, type, and quality of medical facilities across districts) could further enrich spatial analyses and inform targeted public health policies [4,5,6,7]. Additionally, the identification of geographic hotspots and cold spots of multimorbidity provides valuable evidence for policymakers to prioritize high-burden areas [1, 4, 15]. Such spatial insights can support more equitable distribution of healthcare resources and guide the development of region-specific prevention and management strategies [1, 4, 15].

Conclusion

The spatial analysis of multimorbidity prevalence in Thailand revealed that hotspots were concentrated in the central regions, including Bangkok, which are also economically developed areas. In contrast, cold spots were found in the relatively underdeveloped southern region. Multimorbidity was associated with higher population density, a greater aging index, and longer average years of education. Therefore, its prevalence tends to be higher in developed regions than in less developed ones. Future studies should explore additional factors in spatial analyses to gain a more comprehensive understanding of the geographic characteristics of multimorbidity, leading to improved prevention and intervention strategies.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

TCS:

Thai Cohort Study

OR:

Odd ratio

LMICs:

Low- and middle-income countries

NCDs:

Non-communicable diseases

STOU:

Sukhothai Thammathirat Open University

BMI:

Body Mass index

WHO:

World Health Organization

SES:

Socio-economic status

CI:

Confidential interval

References

  1. Skou ST, Mair FS, Fortin M, Guthrie B, Nunes BP, Miranda JJ, et al. Multimorbidity Nat Rev Dis Primers. 2022;8(1):48. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41572-022-00376-4.

    Article  PubMed  Google Scholar 

  2. PLOS Medicine Editors. Multimorbidity: Addressing the next global pandemic. PLoS Med. 2023;20(4):e1004229. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pmed.1004229.

  3. Basto-Abreu A, Barrientos-Gutierrez T, Wade AN, Oliveira de Melo D, Semeão de Souza AS, Nunes BP, et al. Multimorbidity matters in low and middle-income countries. J Multimorb Comorb. 2022;12:26335565221106074. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/26335565221106074.

  4. Weimann A, Dai D, Oni T. A cross-sectional and spatial analysis of the prevalence of multimorbidity and its association with socioeconomic disadvantage in South Africa: A comparison between 2008 and 2012. Soc Sci Med. 2016;163:144–56. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.socscimed.2016.06.055.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Rong P, Chen Y, Dang Y, Duan X, Yan M, Zhao Y, et al. Geographical specific association between lifestyles and multimorbidity among adults in China. PLoS ONE. 2023;18(6):e0286401. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0286401.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Laohasiriwong W, Puttanapong N, Singsalasang A. Prevalence of hypertension in Thailand: Hotspot clustering detected by spatial analysis. Geospat Health. 2018;13(1):608. https://doiorg.publicaciones.saludcastillayleon.es/10.4081/gh.2018.608.

    Article  PubMed  Google Scholar 

  7. Cromley EK, Wilson-Genderson M, Heid AR, Pruchno RA. Spatial Associations of Multiple Chronic Conditions Among Older Adults. J Appl Gerontol. 2018;37(11):1411–35. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/0733464816672044.

    Article  PubMed  Google Scholar 

  8. Bentué-Martínez C, Mimbrero MR, Zúñiga-Antón M. Spatial patterns in sociodemographic factors explain to a large extent the prevalence of hypertension and diabetes in Aragon (Spain). Front Med (Lausanne). 2023;10:1016157. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fmed.2023.1016157.

    Article  PubMed  Google Scholar 

  9. Gebregziabher M, Ward RC, Taber DJ, Walker RJ, Ozieh M, Dismuke CE, et al. Ethnic and geographic variations in multimorbidity: Evidence from three large cohorts. Soc Sci Med. 2018;211:198–206. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.socscimed.2018.06.020.

    Article  PubMed  Google Scholar 

  10. Palo SK, Nayak SR, Sahoo D, Nayak S, Mohapatra AK, Sahoo A, et al. Prevalence and pattern of multimorbidity among chronic kidney disease patients: a community study in chronic kidney disease hotspot area of Eastern India. Front Med (Lausanne). 2023;10:1131900. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fmed.2023.1131900.

    Article  PubMed  Google Scholar 

  11. Hernández B, Voll S, Lewis NA, McCrory C, White A, Stirland L, et al. Comparisons of disease cluster patterns, prevalence and health factors in the USA, Canada, England and Ireland. BMC Public Health. 2021;21(1):1674. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-021-11706-8.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Sleigh AC, Seubsman SA, Bain C; Thai Cohort Study Team. Cohort profile: The Thai Cohort of 87,134 Open University students. Int J Epidemiol. 2008;37(2):266–72. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ije/dym161.

  13. Seubsman SA, Kelly M, Sleigh A, Peungson J, Chokkanapitak J, Vilainerun D. Methods used for successful follow-up in a large scale national cohort study in Thailand. BMC Res Notes. 2011;4:166. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/1756-0500-4-166.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Feng X, Kelly M, Seubsman S, Sleigh A. Cardiovascular and cerebrovascular disease incidence among 42785 adults: The Thai Cohort Study, 2005–2013. Glob J Health Sci. 2020;12:81–101. https://doiorg.publicaciones.saludcastillayleon.es/10.5539/gjhs.v12n7p81.

    Article  Google Scholar 

  15. Feng X, Sarma H, Seubsman SA, Sleigh A, Kelly M. Spatial analysis of patterns of multimorbidity in the Thai Cohort Study using latent class analysis. J Epidemiol Glob Health. 2025;15(1):24. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s44197-025-00352-7.

    Article  PubMed  PubMed Central  Google Scholar 

  16. World Health Organization, Regional Office for the Western Pacific. The Asia-Pacific perspective: Redefining obesity and its treatment. Sydney, Australia: Health Communications Australia. 2000. https://iris.who.int/handle/10665/206936. Accessed March 5, 2025.

  17. Human Data Exchange. Thailand - Subnational Administrative Boundaries. Bangkok, Thailand: Human Data Exchange. 2023. https://data.humdata.org/dataset/cod-ab-tha. Accessed December 5, 2024.

  18. Soleimani M, Bagheri N. Spatial and temporal analysis of myocardial infarction incidence in Zanjan province. Iran BMC Public Health. 2021;21(1):1667. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-021-11695-8.

    Article  PubMed  Google Scholar 

  19. National Statistics Office. Statistics Tables. Bangkok, Thailand: National Statistics Office. 2022. http://statbbi.nso.go.th/staticreport/page/sector/en/index.aspx. Accessed December 5, 2024.

  20. Official statistics registration systems. Statistics. Bangkok, Thailand: Official statistics registration systems. 2022. https://stat.bora.dopa.go.th. Accessed December 5, 2024.

  21. Bagheri N, Batterham PJ, Salvador-Carulla L, et al. Development of the Australian neighborhood social fragmentation index and its association with spatial variation in depression across communities. Soc Psychiatry Psychiatr Epidemiol. 2019;54(10):1189–98. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00127-019-01712-y.

    Article  PubMed  Google Scholar 

  22. Bagheri N, Wangdi K, Cherbuin N, Anstey KJ. General Practice Clinical Data Help Identify Dementia Hotspots: A Novel Geospatial Analysis Approach. J Alzheimers Dis. 2018;61(1):125–34. https://doiorg.publicaciones.saludcastillayleon.es/10.3233/JAD-170079.

    Article  PubMed  Google Scholar 

  23. City Population. THAILAND: Administrative Division. Bangkok, Thailand: City Population. 2023. https://www.citypopulation.de/en/thailand/admin/#google_vignette. Accessed December 5, 2024.

  24. National Statistics Office. National Statistics Office. Bangkok, Thailand: National Statistics Office. 2023. https://www.nso.go.th/. Accessed December 5, 2024.

  25. National Statistics Office. Service. Bangkok, Thailand: National Statistics Office. 2023. http://service.nso.go.th/. Accessed December 5, 2024.

  26. Digital Government Development. Open Government Data of Thailand. Bangkok, Thailand: Digital Government Development. 2015. https://data.go.th/dataset/. Accessed December 5, 2024.

  27. National Statistics Office. Provinces. Bangkok, Thailand: National Statistics Office. 2023. https://province.nso.go.th/. Accessed December 5, 2024.

  28. FAQThai. FAQThai. Bangkok, Thailand: FAQThai. 2023. https://www.faqthai.org/. Accessed December 5, 2024.

  29. TerraBKK. Education for Thai people in 77 provinces. Bangkok, Thailand: TerraBKK. 2023. https://www.terrabkk.com. Accessed December 5, 2024.

  30. Firouraghi N, Bagheri N, Kiani F, Goshayeshi L, Ghayour-Mobarhan M, Kimiafar K, et al. A spatial database of colorectal cancer patients and potential nutritional risk factors in an urban area in the Middle East. BMC Res Notes. 2020;13(1):466. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13104-020-05310-z.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Head A, Fleming K, Kypridemos C, Pearson-Stuttard J, O’Flaherty M. Multimorbidity: the case for prevention. J Epidemiol Community Health. 2021;75(3):242–4. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/jech-2020-214301.

    Article  PubMed  Google Scholar 

  32. Odland ML, Payne C, Witham MD, Siedner MJ, Bärnighausen T, Bountogo M, et al. Epidemiology of multimorbidity in conditions of extreme poverty: a population-based study of older adults in rural Burkina Faso. BMJ Glob Health. 2020;5(3): e002096. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/bmjgh-2019-002096.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Pathirana TI, Jackson CA. Socioeconomic status and multimorbidity: a systematic review and meta-analysis. Aust N Z J Public Health. 2018;42(2):186–94. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/1753-6405.12762.

    Article  PubMed  Google Scholar 

  34. Mtintsilana A, Craig A, Mapanga W, Dlamini SN, Norris SA. Association between socio-economic status and non-communicable disease risk in young adults from Kenya, South Africa, and the United Kingdom. Sci Rep. 2023;13(1):728. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41598-023-28013-4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Matthias AT, Fernando GVMC, Somathilake BGGK, Prathapan S. Predictors and patterns of polypharmacy in chronic diseases in a middle-income country. Int J Physiol Pathophysiol Pharmacol. 2021;13(6):158–65.

    PubMed  PubMed Central  Google Scholar 

  36. Chowdhury SR, Chandra Das D, Sunna TC, Beyene J, Hossain A. Global and regional prevalence of multimorbidity in the adult population in community settings: a systematic review and meta-analysis. EClinicalMedicine. 2023;57:101860. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.eclinm.2023.101860.

    Article  PubMed  PubMed Central  Google Scholar 

  37. The ASEAN Post Team. Growing gap between richest and poorest Thais. Bangkok, Thailand: The ASEAN Post Team. 2023. https://theaseanpost.com/article/growing-gap-between-richest-and-poorest-thais. Accessed December 5, 2024.

  38. Zhou Y, Dai X, Ni Y, Zeng Q, Cheng Y, Carrillo-Larco RM, et al. Interventions and management on multimorbidity: An overview of systematic reviews. Ageing Res Rev. 2023;87:101901. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.arr.2023.101901.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We are very grateful to the staff of Sukhothai Thammathirat Open University (STOU) for their help in contacting students and to them who participated in this study.

Funding

Not applicable.

Author information

Authors and Affiliations

  1. National Centre of Epidemiology and Population Health, the Australian National University, Canberra, Australia

    Xiyu Feng, Haribondhu Sarma, Tsheten Tsheten, Adrian Sleigh & Matthew Kelly

  2. Health Research Institute, University of Canberra, Canberra, Australia

    Nasser Bagheri

  3. School of Human Ecology, Sukhothai Thammathirat Open University, Nonthaburi, Thailand

    Sam-ang Seubsman

  4. National Centre of Epidemiology and Population Health, Building 62, Mills Road, Acton 2601, Canberra, Australia

    Xiyu Feng, Haribondhu Sarma, Tsheten Tsheten, Adrian Sleigh & Matthew Kelly

Authors
  1. Xiyu Feng
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Haribondhu Sarma
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Nasser Bagheri
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Tsheten Tsheten
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Sam-ang Seubsman
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Adrian Sleigh
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Matthew Kelly
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

X.F. conceived the study, analyzed and interpreted the data and drafted the manuscript. X.F., N.B., H.S., T.T. and M.K. designed the study. S.S., A.S. and MK. provided the dataset. X.F., H.S., N.B., T.T. and M.K. contributed to subsequent drafts and final editing.

Corresponding author

Correspondence to Xiyu Feng.

Ethics declarations

Ethics approval and consent to participate

The studies involving humans were approved by the ethics committee of the Sukhothai Thammathirat Open University Research and Development Institute on October 22, 2005. Number: 0522/10. Date: October 22, 2005. For this sub-sample analysis, we also got approval from Human Research Ethics Committee of the Australian National University on January 25, 2022. Number: 2021/796. Date: January 25, 2022. The participants provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Feng, X., Sarma, H., Bagheri, N. et al. The spatial analysis of multimorbidity in Thai Cohort Study. Arch Public Health 83, 120 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13690-025-01605-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13690-025-01605-4

Keywords

Archives of Public Health

ISSN: 2049-3258