The association between climate variables and tuberculosis in Kolaka District, Southeast Sulawesi Province, Indonesia, 2013–2020: a Bayesian autoregressive model

Background Tuberculosis is one of the diseases that requires comprehensive treatment. This disease is highly contagious and can be transmitted through the air. Climate factors play a role in the increasing cases of tuberculosis. This study aimed to determine the correlation between climatic variables and TB in Kolaka District, Southeast Sulawesi Province, Indonesia, Methods This research was modeled using an autoregressive (AR) Bayesian model with three possible likelihoods; Gaussian, Poisson and Negative Binomial responses. Results Minimum temperature and average temperature, a coefficient of 4.234 suggests that for every 1 degree increase in minimum temperature, there is an estimated increase of approximately four cases, assuming other variables remain constant. Maximum temperature, a coefficient of 17.851 suggests that for every 1 degree increase in maximum temperature, there is an estimated increase of around 17-18 cases, assuming other variables remain constant. Humidity, a coefficient of -13.413 suggests that for every 1% increase in humidity, there is an estimated decrease of around 13 cases, assuming other variables remain constant. Rainfall, a coefficient of -0.327 suggests that for every 1 mm increase in rainfall, there is an estimated decrease of around 0.327 cases, assuming other variables remain constant. Light, a coefficient of -4.322 suggests that for every 1-hour increase in light duration, there is an estimated decrease of around four cases, assuming other variables remain constant. Conclusions Climate change has a significant impact on tuberculosis through temperature-related factors. These factors influence the prevalence, spread, and vulnerability to TB. Addressing these challenges requires a holistic approach involving adaptation planning. Strong public health systems and healthcare infrastructure can help mitigate the risks and impacts of climate change-related tuberculosis.


Introduction
Tuberculosis is an infectious disease caused by the bacterium Mycobacterium tuberculosis (MTB). 1 This disease primarily affects the respiratory system, particularly the lungs, but it can also attack other organs in the body. 2,3TB has become one of the deadliest diseases in the world, with millions of people infected and thousands of deaths occurring annually.Globally, there are 6.4 million cases of tuberculosis, which accounts for 64% of tuberculosis incidents.tuberculosis still remains among the top 10 causes of death worldwide, with an estimated 1.3 million tuberculosis related deaths globally. 4In Indonesia, TB still remains one of the diseases with the highest number of cases, requiring continued attention from the government.
According to the Global Tuberculosis Report 2017, it is estimated that there were 1.020.000tuberculosis cases in Indonesia, ranking second in the world for the highest number of TB cases after India.The number of new tuberculosis cases recorded in Indonesia in 2017 was 420.994. 5According to the Tuberculosis Data and Information Center of the Ministry of Health of the Republic of Indonesia in 2017, the indicators used for evaluating and monitoring the success of tuberculosis control efforts are the detection of new tuberculosis cases (case detection rate (CDR)) and the treatment success rate.These indicators serve as important measures to assess the progress and effectiveness of tuberculosis management in Indonesia.
The tuberculosis cases in Indonesia are generally high and rapidly growing among individuals aged 15 years and above, people living in poverty, those with low educational levels, and socioeconomically disadvantaged communities.West Java is the province that contributes the highest number of cases in Indonesia, with the majority occurring among individuals aged 15 years and above. 6Climate change has become an increasingly urgent global issue in the past few decades. 7In recent years, the impact of climate change on human health has been a major concern for scientists and health experts worldwide. 8One significant aspect of health that is significantly affected by climate change is the spread of infectious diseases, including TB. 9,10 Climate change and the spread of TB involve various complex factors. 11One key factor is the changes in temperature and rainfall patterns that can affect the environment where TB bacteria can survive and multiply. 124][15] Additionally, changes in rainfall patterns can affect the availability of clean water, which is crucial for ensuring adequate cleanliness and sanitation, both of which are important factors in TB prevention. 16,17 Brazil, it has been found that tuberculosis cases are more frequent under the following conditions: ultraviolet radiation (UVR) above 17 MJ/m 2 (67.8%; p ≤ 0.001); relative humidity between 31.0%and 69.0%(95.8%; p ≤ 0.001); rainfall less than 1 mm (71.7%; p ≤ 0.001); daily sun exposure for 12 hours (40.6%; p = 0.001); and temperatures between 20°C and 23°C (72.4%; p ≤ 0.001). 18It has been found that extremely high temperatures have a positive effect on the reported morbidity of pulmonary tuberculosis in Binzhou (RR = 0.924, 95% CI: 0.856-0.997)and Weihai (RR = 0.910, 95% CI: 0.843-0.982). 19In Hong Kong, it has also been found that lower average temperatures (<22.0°C) in the later months (>10 months) are associated with an increased risk of tuberculosis notifications.Specifically, the risk increases within the temperature range of 16.3 to 17.3°C at lag 13-15 months and reaches the highest risk at a temperature of 16.8°C at lag 14 months. 20According to Zhang et al. (2019) using a Geographically Weighted Regression (GWR) model, a positive correlation was found between TB incidence and annual average rainfall (AR), but a negative correlation was observed REVISED Amendments from Version 1 In the new version, we have explained various important aspects related to scientific progress.The methods section includes explanations of the Poisson distribution, the negative binomial distribution, and the model formation process.
Results and Abstract: The terms minimum temperature and average temperature are used interchangeably.Therefore, the sentence has been simplified to: "Minimum temperature and average temperature: a coefficient of 4.234 suggests that for every 1-degree increase in minimum temperature, there is an estimated increase of approximately four cases, assuming other variables remain constant."Discussion: The following sentence has been removed from the discussion section: "Additionally, changes in climate can impact water resources, food production, and availability of clean air, all of which are essential for maintaining good health.Extreme weather events, such as hurricanes, floods, or heatwaves, can cause injuries, displacement, and even loss of life.Moreover, these events can also lead to the contamination of water sources, the destruction of infrastructure, and the disruption of healthcare services." Any further responses from the reviewers can be found at the end of the article with other meteorological factors.Average relative humidity (ARH) also showed a negative correlation with TB incidence in all prefectures with a significance level of p < 0.05. 21midity is indeed a climate factor that can potentially affect the spread of tuberculosis. 22,23Several studies have shown that high humidity can support the survival of MTB bacteria in the environment. 24,25High humidity can also influence the dispersal of respiratory droplets containing MTB bacteria. 26,27Under low humidity conditions, respiratory droplets can dry more quickly, and the bacteria-containing particles become lighter, enabling them to remain suspended in the air for longer periods. 28,29Therefore, low humidity can also increase the risk of tuberculosis transmission.Indeed, climate change can also have an impact on the overall human health system.Fluctuations in extreme temperatures, extreme weather events such as floods or droughts, and changes in seasonal patterns can cause social and economic instability, which in turn can affect the spread of TB.These factors can lead to human migration, food insecurity, reduced nutritional status, and vulnerability to infectious diseases, including TB.

Location of study
Astronomically, Kolaka Regency is located along the equator section of the equator, extending from North to South between 3°36 0 -4°35 0 South Latitude (SL) and extending from West to East between 120°45 0 -121°5 2 0 East Longitude (EL).Based on its geographical position, the territorial boundaries of Kolaka Regency in the north is Kolaka Utara Regency, in the south is Bombana regency, in the east is Kolaka Timur Regency and in the west is Sulawesi Selatan Province in Teluk Bone.

Date sources
This study utilized climate data and tuberculosis disease data.The climate data were obtained from the Meteorology and Geophysics Agency of Southeast Sulawesi Province, including temperature, minimum temperature, maximum temperature, humidity, rainfall, and light data from 2013 to 2020.The data are accessible on the DATA ONLINE -PUSAT DATABASE -BMKG website.Meanwhile, the tuberculosis data were obtained through reports from the Health Department of Kolaka District from 2013 to 2020.The data on tuberculosis disease originates from the health profile of Kolaka district.This data can be accessed openly through the Kolaka District website.

Statistical analysis
The association between meteorological factors and the number of TB cases was modeled using an autoregressive (AR) Bayesian model with three possible likelihoods; Gaussian, Poisson and Negative Binomial responses.Although TB cases are in fact discrete as both the Poisson and NB distributions, the continuous Gaussian distribution is also suitable to use as the number of TB cases is large.In case the dispersion statistic is close to one, a Poisson model is better to apply than NB distribution.In this study, however, we use both WAIC and DIC values as the model selection criterion to evaluate their performance.Suppose y t denote the monthly/annual counts of TB in the District of Kolaka, t ¼ 1,2, …,8 representing the period of 2013-2020 and u t represents meteorological covariates; i.e., minimum temperature, maximum temperature, average temperature, humidity, rainfall and light.Then, the response y t follows one of three response likelihoods; For Gaussian distribution with mean α þ u t ; The term u t represents meteorological covariates; i.e., minimum temperature, maximum temperature, average temperature, humidity, rainfall and light.
Then, the temporal model is formulated as autoregressive order 1, which reads where α is an intercept, ρ is a temporal correlation term (with ρ j j < 1) and ε t is a Gaussian error term with zero mean and a fixed large precision τ u so that the error term is close to zero.
The performance of all models on the TB data in the district of Kolaka, Southeast Sulawesi from 2013 to 2020, based on the DIC (WAIC) values, is summarized in Table 1.
Table 1 shows that the performances of the autoregressive model with Poisson likelihood is much better, compared to those with Negative Binomial and Gaussian responses.This indicates that Poisson is suitable for analyzing tuberculosis data.Therefore, using Poisson autoregressive model, we determined the influence of meteorological factors on TB incidence in the district of Kolaka, Southeast Sulawesi from 2013 to 2020.

Ethical considerations
The study obtained ethical approval from the Research Ethics Committee of the Indonesian Public Health Association (IPHA) Southeast Sulawesi Province under approval number 134/KEPK-IAKMI/II/2023.

Results
Table 2 73 shows that the linear regression model conducted to test the correlation between climate factors and tuberculosis disease resulted in values above 0.05, indicating that there was no significant influence of each variable.Furthermore, a test using the Bayesian autoregressive model will be conducted to further investigate the relationship.Table 3 indicates that the minimum temperature, maximum temperature, and average temperature have positive values of one, which are quite high, indicating a positive association with TB cases.On the other hand, humidity, rainfall, and light have negative values, indicating a negative association with tuberculosis cases.Further analysis of the coefficient model revealed the following relationships: i) minimum temperature and average temperature, a coefficient of 4.234 suggests that for every 1 degree increase in minimum temperature, there is an estimated increase of approximately four cases, assuming other variables remain constant.ii) Maximum temperature, a coefficient of 17.851 suggests that for every 1 degree increase in maximum temperature, there is an estimated increase of around 17-18 cases, assuming other variables remain constant.iii) Humidity, a coefficient of -13.413 suggests that for every 1% increase in humidity, there is an estimated decrease of around 13 cases, assuming other variables remain constant.iv) Rainfall, a coefficient of -0.327 suggests that for every 1 mm increase in rainfall, there is an estimated decrease of around 0.327 cases, assuming other variables remain constant.v) Light, a coefficient of -4.322 suggests that for every 1-hour increase in light duration, there is an estimated decrease of around four cases, assuming other variables remain constant.

Discussion
In recent decades, climate change has become a major global issue that has garnered significant attention.1][32] Furthermore, human health is also significantly affected by climate change, with tuberculosis being one of the diseases impacted. 33,34Our findings indicate that climate change is likely to affect individual vulnerability to tuberculosis by increasing the prevalence of underlying risk factors, particularly in developing countries.
Indeed, climatic conditions are one of the factors that can influence the development of disease-causing microorganisms. 18,35Certain microorganisms, including bacteria, viruses, and fungi, can thrive or struggle depending on the environmental conditions they are exposed to.7][38] For example, warm and humid environments may promote the proliferation of certain bacteria and fungi, while extreme temperatures or drought conditions may inhibit their growth.Understanding the relationship between climate and the development of pathogenic microorganisms is crucial for assessing and addressing public health risks. 39Climate change can lead to changes in ecosystems, altering the patterns of interaction between humans and the environment.These changes can have significant impacts on human health.For example, shifts in temperature and precipitation patterns can affect the distribution of disease vectors such as mosquitoes, ticks, and other carriers of infectious diseases.[41] Climate change can also contribute to the exacerbation of respiratory and cardiovascular diseases due to increased air pollution, the release of allergens, and the intensification of heatwaves.Moreover, it can have indirect effects on mental health and well-being, particularly among communities affected by climate-related disasters or forced migration. 42,43opulation density can increase the risk of increased exposure for individuals suffering from pulmonary tuberculosis, which in turn facilitates the spread of the bacteria causing it. 44,45In addition to density factors, the occurrence of tuberculosis is also influenced by other risk factors such as high poverty levels, low coverage of healthy housing, and poor hygiene and health practices.Therefore, the high number of detected tuberculosis cases in a particular area may also be influenced by the characteristics of other variables. 6,46st bacteria and fungi are able to survive in air humidity reaching or exceeding 70%. 47In conditions where the monthly average air humidity in Kabupaten Serang ranges from 70% to 87%, bacteria and fungi can thrive in the outdoor environment.The presence of a negative relationship between air humidity and the number of new tuberculosis cases may suggest that indoor air humidity is not associated with the presence of TB-causing agents inside the home. 48This finding is supported by research conducted by Duffield and Young (1984), which demonstrated that pathogenic MTB bacteria can survive equally well in both damp and dry soil. 49lobal temperature increase can impact the ecology and geography of tuberculosis. 18,50The role of climate in tuberculosis spread cannot be seen as a single factor.Temperature can influence the transmission of tuberculosis.The MTB bacteria can survive in various temperatures.Both low and high temperatures can affect the survival of the bacteria outside the human body.In specific areas, low temperatures can create more favorable conditions for the bacteria to survive outside the human body and trigger an increased risk of transmission.Global climate change can affect the patterns of tuberculosis spread.Seasons and local climates can also influence the occurrence of tuberculosis.Several studies indicate a relationship between specific seasons and increased tuberculosis cases.Factors such as temperature, air humidity, and rainfall patterns can influence the transmission rate and severity of the disease.[53] The research findings from the present study did not find any correlation between humidity and tuberculosis in Kolaka Regency.The influence of humidity on tuberculosis involves multiple factors.It should be noted that TB is a contagious disease most commonly transmitted through the air (via infected airborne droplets), and other factors such as direct contact with untreated patients with tuberculosis or a weakened immune system also play a significant role in its transmission. 17,41,54gh relative humidity in indoor environments can affect the transmission of tuberculosis.When the relative humidity is above 80%, the risk of contracting tuberculosis significantly increases.High humidity in indoor spaces can impact the spread of tuberculosis by prolonging the survival time of infected airborne droplets in the air, thus allowing for a longer duration of inhalation and transmission.However, other studies have found a correlation between humidity and tuberculosis.Low humidity, especially during the dry season, is associated with an increased risk of tuberculosis infection.This occurs because low humidity can affect the respiratory tract and make individuals more vulnerable to tuberculosis infection.[57] In this research, no correlation was found between rainfall and tuberculosis.However, during the period of 2013-2020, the presence of tuberculosis in Kolaka Regency tended to increase.Rainfall and tuberculosis are two different factors that can be related to each other.Rainfall refers to the amount of precipitation that falls in a specific period in a given area.
High rainfall can create a humid environment.This humidity can affect the survival of MTB bacteria in the external environment.These bacteria are more resistant to moist conditions and can survive in airborne water droplets for a longer time.This means that in areas with high rainfall, these bacteria can remain active and easily spread through water droplets.
Rainfall can also determine the amount of time hosts spend indoors, thus influencing the transmission of MTB within households. 38deed, there are other factors that can contribute to the increase of tuberculosis.Areas with high rainfall tend to have a higher population density, and high population density can lead to poor environmental conditions. 16,58Poor environmental conditions, such as high density, lack of ventilation, and poor sanitation, can worsen the spread of tuberculosis.People living in areas with high rainfall may be forced to reside in unhealthy environments, such as overcrowded settlements in urban areas that lack adequate sanitation facilities, which can increase the risk of TB transmission.Poor economic conditions can result in unemployment, poverty, and diverted funds, reducing access to proper healthcare.[61][62] Furthermore, in Kolaka Regency, no relationship was found between sunlight exposure and the occurrence of tuberculosis.Indirect sunlight exposure can influence TB cases by affecting the host's condition.Areas with limited sunlight exposure have been reported to have a higher prevalence of vitamin D deficiency. 63Sunlight plays a crucial role in the production of vitamin D in the body.When the skin is exposed to sunlight, it synthesizes vitamin D, which is essential for maintaining healthy bones and overall health.Adequate sunlight exposure is necessary for the body to produce an optimal amount of vitamin D. 64 Absolutely, vitamin D deficiency can indeed affect the antimicrobial peptide system, which is responsible for regulating the immune response in the human body. 65,66Vitamin D plays a crucial role in modulating the immune system and enhancing the function of antimicrobial peptides, which are essential for fighting against microbial infections. 67,68Insufficient levels of vitamin D can compromise the immune system's ability to defend against pathogens, including MTB, the bacteria responsible for TB. 69,70Impaired host immunity due to vitamin D deficiency will support the development and reactivation of TB disease. 71n addition, sunlight exposure can also affect the life cycle of TB-causing agents.MTB can survive in the environment for more than 74 days if protected from light. 49MTB in sputum will die within a short period of time when exposed to sunlight, up to seven hours. 18Therefore, access to sunlight inside buildings is crucial to prevent the formation of conducive environments for MTB within the house.
Changes in rainfall patterns and temperature can impact the life cycle of MTB bacteria and the transmission vectors of this disease. 53High humidity and abundant rainfall can create ideal environmental conditions for the bacteria to survive and proliferate. 52Climate change can affect population migration, whether caused by natural disasters or socioeconomic changes. 51Migration can lead to the spread of TB to areas previously free from the disease.Migration can also disrupt access to adequate healthcare, worsen the health conditions of individuals infected with TB, and increase the risk of transmission. 37,72 control the spread of tuberculosis, especially in high-risk areas, it is important to follow recommended preventive measures such as early treatment for patients with tuberculosis, ensuring good ventilation in rooms, maintaining personal and environmental hygiene, and promoting awareness about the importance of respiratory hygiene and healthy living practices.Research has limitations, as factors like population density, ventilation quality, direct exposure to untreated tuberculosis patients, and individual immune status also play significant roles in the spread of the disease.Therefore, future studies should consider examining these factors in order to gain a comprehensive understanding.

Introduction:
The authors mentioned that according to the Global Tuberculosis Report 2017, it is estimated that there were 1.020.000tuberculosis cases in Indonesia, ranking second in the world for the highest number of TB cases after India, and that the number of new tuberculosis cases recorded in Indonesia in 2017 was 420.994.I understand that the paper explored TB transmission in 2013-2020, but I would suggest adding a citation of the recent report because the TB prevalence could be high or low.Otherwise, this section was well explained about how factors related to climate change contribute to TB transmission. 1.

Methods:
Methodology is one of the key parts of the paper, and it should be clearly explained so that other researchers can adopt or populate it in their studies.I would suggest the authors further explain the parameter μt to clarify the difference between the Poisson distribution and the negative binomial distribution, and how they are different from the Gaussian distribution. 1.
Based on the model, it's not clear how Table 1 was formed, and it would be worth it if it were clarified how those values were obtained and which data/value used to create the table.

2.
It would be good to see some more critical discussion of the limitations of the model, and what experimental data would be useful to improve the model.

Results and abstract
The author postulated that "minimum temperature, a coefficient of 4.234 suggests that for every 1 degree increase in minimum temperature, there is an estimated increase of approximately four cases, assuming other variables remain constant," and "average temperature, a coefficient of 4.234 suggests that for every 1 degree increase in average temperature, there is an estimated increase of approximately four cases, assuming other variables remain constant."These sentences sound like repetitions with the same value.I would suggest removing one of these sentences or connecting them (using minimum and average temperature) to one sentence to avoid repetition. 1.

Discussion:
The authors mentioned that "additionally, changes in climate can impact water resources, food production, and the availability of clean air, all of which are essential for maintaining good health.Extreme weather events, such as hurricanes, floods, or heatwaves, can cause injuries, displacement, and even loss of life.Moreover, these events can also lead to the contamination of water sources, the destruction of infrastructure, and the disruption of healthcare services."This paragraph has no direct link between climate change and TB.I would recommend removing this paragraph from the paper. 1.

Is the work clearly and accurately presented and does it cite the current literature? Partly
Is the study design appropriate and is the work technically sound?Yes

Are sufficient details of methods and analysis provided to allow replication by others? Partly
If applicable, is the statistical analysis and its interpretation appropriate?Partly Are all the source data underlying the results available to ensure full reproducibility?No

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Bioinformatics and infectious diseases mathematical modelling I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com

Table 1 .
The performance of autoregressive model with three likelihoods based on the values.

Table 2 .
Data analysis utilizing a linear regression model.

Table 3 .
Data analysis employing a Bayesian autoregressive model.

Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests:
Climate change has a significant impact on tuberculosis through temperature-related factors.These factors influence the prevalence, spread, and vulnerability to tuberculosis.Addressing these challenges requires a holistic approach involving adaptation planning.Strong public health systems and healthcare infrastructure can help mitigate the risks and impacts of climate change-related tuberculosis.Prevention efforts, early diagnosis, and effective treatment are crucial in controlling the spread of this disease.In the context of climate change, efforts to enhance capacity and resources are needed to address emerging challenges.No competing interests were disclosed.

have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
https://doi.org/10.5256/f1000research.167785.r294011©2024M. Issarow C.Chacha M. IssarowUniversiity of Cape Town, Cape Town, South Africa I can confirm that the issues raised have been addressed by the authors and that the paper can be accepted.

Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests:
No competing interests were disclosed.

have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Version 1
https://doi.org/10.5256/f1000research.152089.r247572