Mapping the distribution of tuberculosis cases identified through routine program and community based active screening in Central Ethiopia

doi:10.21203/rs.3.rs-4112076/v1

Download PDF

Research Article

Mapping the distribution of tuberculosis cases identified through routine program and community based active screening in Central Ethiopia

https://doi.org/10.21203/rs.3.rs-4112076/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Smear positive TB cases greatly contribute to community level transmission of the disease. Locating hotspots would make it easier to prioritize and target control interventions. This study is aimed at assessing the spatial distribution of smear positive index TB cases and their secondary cases and the predictors of clustering of smear positive TB cases.

Methods

The study was conducted in Silti Zone of Central Ethiopia from 2020–2022. Data of smear positive index TB cases were collected from the unit TB registers of healthcare facilities. Contacts of all index TB cases were screened in the community and tested to identify secondary TB cases. We performed spatial analysis including Moran’s I statistic, Getis-Ord Gi* statistic and geographically weighted regression (GWR) to assess the global distribution, local clustering and the predictors of clustering of smear positive TB cases respectively. Additionally, we used inverse distance weighting (IDW) interpolation to predict the distribution of smear postive TB cases and develop a continuous raster map for places with no data.

Results

Spatial autocorrelation results showed that the distribution of the smear positive TB cases showed signficant clustering (Moran’s I = 0.70029; p-value < 0.000). The Getis-Ord Gi* output indicated the presence of statistically significant hotspots as well as cold spots in the study area. Statistically significant hotspots were found in 11 Kebeles of Silti, Dalocha and Misrak Silti districts. Signficant coldspots were also found in five kebeles of the Silti and Misrak districts. GWR analysis revealed that no education, primary education, family size and thatched roof houses were significant predictors of the spatial clustering of the smear positive TB cases. We also found that majority of the secondary TB cases were found in the hotspot areas identified through the spatial analysis.

Conclusion

Our study showed that the distribution of smear positive TB cases in the study area was heterogenous and both statistically significant hotspots and coldspots were identified. Location based targeted interventions could improve TB control performances including reduction in the transmission of TB. Educational status, family size and housing type were some of the factors with significant influence on the spatial distribution of smear positive TB in the study area. Distribution of the secondary TB cases found through household contacts screening coincided with the identified hotspots indicating higher transmission of the disease in these places.

Spatial analysis

Interpolation

Hotspots

Coldspots

Clustering

Tuberculosis

Ethiopia

The global epidemiology of tuberculosis is uneven and mostly affects 30 countries labeled as high burden countries which are located in the Sub Saharan Africa and South East Asia(1). Even in these countries the distribution of TB still tends to vary considerably. Both the incidence and prevalence appear to concentrate in geographic areas with low socioeconomic status and dense population(2). Usually, areas within neighborhoods with higher concentration of TB cases indicate the existence of local transmission and is usually associated with unfavorable social determinants of health such as poverty, overcrowding or marginalization (3). Studies indicate that TB interventions targeting local hotspots could be effective in identifying and treating more cases that could interrupt the continued transmission (4, 5). However, for such interventions to be effective, localized hotspots or clusters should be identified based on routinely reported TB data and other assessments. Furthermore, such hotspots/clusters have to be stable enough for some time until interventions are carried out. Identifying localized areas meeting such criteria could help TB programs to prioritize interventions and make informed decisions to reduce TB incidence (6).

In most high burden countries, TB programs rely on patients passively presenting themselves to healthcare facilities for TB suggestive symptoms to identify cases. In a slightly modified approach in some settings, all patients attending both outpatient and inpatient services are screened for TB. But all of these methods have several drawbacks. Firstly, only patients having apparent symptoms are diagnosed and those having milder symptoms could be missed. Cases with mild symptoms could also hesitate to visit a healthcare facility. Secondly, patients having a limited access to care due to various factors are also not reached and hence not diagnosed (7, 8). Studies show that in Ethiopia, about a third of all TB cases are missed every year (9). Active case finding interventions are occasionally implemented but are usually challenging in high burden settings due to the enormous resources it requires. Besides, community wide implementation of active case finding may not be effective (5). Narrowing down interventions to prioritized high risk groups such as household contacts of TB patients is therefore necessary. Targeting geographic areas with a relatively higher burden of clustering of TB cases could also be another approach for a focused intervention. Such spatially targeted interventions, however, are usually not implemented due to factors including lack of evidence on its effectiveness and uncertainty on whether routine data can accurately pinpoint geographic hotspots or clusters of TB cases (5).

Existence of heterogeneity in the distribution of TB cases means geographic areas with higher clustering could be prioritized and targeted with interventions to identify more cases while places with dispersed distribution of cases could be addressed when resources allow(5, 10, 11). This approach could be cost effective especially in high burden settings where attempts to address all the various high-risk groups at the same time is usually not feasible and/or very challenging. It’s well established that smear positive TB cases are more infectious and contribute to the faster transmission of the disease in the community compared to smear negative and extrapulmonary TB cases(12–14). Hence, the advantages of identifying and targeting these particular groups of patients immediately after their diagnosis could be two folds. Firstly, the approach would narrow the interventions towards a few but effective transmitters of the disease saving time and other resources. Secondly, it could contribute to identifying and treating more cases there by interrupting the transmission chain in the community(10, 11). Hence, in this study we aimed to identify the spatial distribution of smear positive TB cases and the factors associated with their distribution in Silti Zone of Central Ethiopia.

Study setting and period

The current study was conducted in four districts and two town administrations of Silti zone of Central Ethiopia. The four districts were Hulbareg, Silti, Dalocha and Mistrak Silti where as the two town administrations were Worabe and Kibet. A district (Woreda) is the third administrative level in Ethiopia having a well-defined population, administrative structures and clearly demarcated geographic boundaries. The districts and town administrations are further divided into 102 kebeles (the fourth and lowest administrative unit in Ethiopia). The combined projected population of the study area for 2022 was 486,246 (15).

Healthcare facilities in the zone provide TB diagnosis and treatment services. Health posts refer presumptive TB cases (started since 2010) to hospitals and health centers where they are further assessed and tested for TB using either GeneXpert test or smear microscopy. Patients identified to have TB are referred back to the health posts or referring health centers to continue their treatment. DOTS services are provided by HEWs where patient’s adherence to treatment is supervised until the end of the treatment.

Patients on TB treatment are referred to health centers or hospitals a the second, fifth and sixth month of treatment to assess progress and smear conversion using AFB microscopy(16). Patients having severe weight loss are also provided with supplementary foods such as plumpy nuts and their weight is regularly monitored during scheduled visits to health centers. In 2022, the estimated all forms of TB prevalence in the study districts ranged from 60 per 100,000 population in Kibet town to 212 per 100,000 population in Silti district according to the zonal annual performance report (17). The current study was conducted from September, 2020 to December, 2022.

Study design

This study was a longitudinal community based study in Silti Zone, Central Ethiopia. The data of smear positive index TB cases was collected both retrospectively and prospectively from the unit TB registers from the selected healthcare facilities in the districts. Additionally, a cross sectional study was conducted at the community level which involved a house to house screening and testing of the household contacts of the index TB cases.

Definitions

An index TB case is a smear positive TB case that has been diagnosed through routine service and registered in the unit TB registers of the the healthcare facilities in the study area and are the first TB cases in the households during the study period. A smear positive TB case is a patient from whom at least one sputum specimen was positive through AFB smear microscopy. A secondary case is a TB case identified after the screening and testing of the household contacts of the smear positive index TB cases.

Data collection

Smear positive index TB cases were identified from the unit TB registers of healthcare facilities registered from Jan, 2019 to September, 2021 with the aim of screening the household contacts of TB patients. After identifying the smear positive index cases, the contact addresses of either the index case or treatment supporter was taken from the registers. Then, community level house to house visits were conducted to screen the household contacts. Individual data of index TB cases and their household contacts including sociodemographic, environmental and clinical variables were collected during the screening process. Additionally, their household level variables such as housing type and area, family size, number of windows and type of cooking methods used were collected. The geo-locations of the index TB cases cases were taken based on their household location during the data collection. For the purpose of protecting their privacy, coordinates were taken approximately 20 meters away from the households. After the community level active screening and testing of sputum samples, additional secondary TB cases were identified. GeneXpert MTB/RIF assay and TB cultures were performed to identify the secondary cases.

Data management and statistical analysis

Both spatial and non-spatial data were entered in to microsoft excel 2019 where data coding, arranging and cleaning were also performed. It was also used to convert the data into a CSV (comma-delimited) file for use in ArcGIS. Various epidemiologic methods were used to analyze the geographical distribution of TB cases and the associated factors. Stata version 17 (Stata Corporation, College Station, Texas 77845, USA) was used to calculate descriptive statistics such as median and interquartile ranges, frequencies and percentages regarding the characteristics of smear positive TB cases and their households.

Spatial analysis

We used ArcGIS version 10.5 to analyze the existence of geographical variations in the distribution of smear positive TB cases and to identify the factors associated with the variation in the distribution. Spatial autocorrelation was measured using the Global Moran’s I statistic in order to assess and understand the overall distributional pattern of smear positive index TB cases in the study area. This test statistic is used to analyze the global distribution of the data over the study area which produces values ranging between − 1 and + 1. Moran’s I index closer to -1 indicates significant dispersion of cases where as an index closer to + 1 indicates clustering of cases(18). We used incremental autocorrelation to calculate the maximum peak distance where the spatial dependence/correlation of the cases becomes prominent. The maximum peak distance was then used to conduct the hotspot analysis. Getis-Ord Gi* statistic was used explore and identify the local hot spots and cold spots of smear positive TB cases. After the hotspot analysis, we also used Inverse Distance Weighting (IDW) interpolation method using GIZ scores in order to predict the distribution of smear positive TB cases for locations with no data closer to the study area. The interpolation method uses a linear comibation of data at nearby measured places to predict the distribution of cases in locations that haven’t been measured(19).

Geographically Weighted Regression (GWR)

A set of various demographic, environmental and other variables were included in the spatial regression models in order to analyze the factors associated with the geographical distribution of the TB cases. We used the ordinary least square regression followed by geographically weighted regression model for this purpose. Through bivariate regression models, variables having a p-value of < 0.2 were selected to be included in the multivariate regression model. Model diagnostics such multicollinearity was checked on the final model to remove variables having a variance inflation factor (VIF) of six or more. We also used the residuals of the multivariate model to examine for the presence of spatial autocorrelation.

Study population and TB prevalence

The study area covered 102 kebeles in four districts and two town administrations in Silti Zone. The population in the study area ranged from 3,000–12,000 per kebele. Of the 102 kebeles, 37 kebeles had at least one smear positive TB case registered in healthcare facilities during the study period and the number of cases ranged from 2 to 58 per kebele. The lowest prevalence was 100 per 100,000 population while the highest prevalence was 1,390 per 100,000 population. The total number of smear positive index TB cases diagnosed and registered in healthcare facilities in the study area between 2020 and 2021 was 387. Screening of the household contacts of the smear positive index TB cases yielded 23 TB cases of all forms.

Individual and household characteristics of Smear Positive Index TB cases

The median age of the smear positive index TB cases was 32 (IQR; 23–45) and nearly 45% (n = 174) of them were female in gender. Most of the index TB cases were married (~ 62%) and Muslim (95%) in religion. (Table 1)

Table 1

Characteristics of Smear Positive Index TB cases in Silti Zone, Central Ethiopia, 2020–2021 (n = 387)
Variables		Frequency n (%)
Sex	Male	211 (54.8)
	Female	174 (45.2)
Age: median (IQR)		32 (23–45)
Marital status	Single	127 (33.1)
	Married	237 (61.7)
	Divorced	11 (2.9)
	Widowed	9 (2.3)
Religion	Christian	19 (4.9)
	Muslim	366 (95.1)
Educational status	None	187 (48.8)
	Primary	143 (37.2)
	Secondary	37 (9.6)
	Tertiary	17 (4.4)
BMI: median (IQR)		19.7 (17.6–22.4)
Duration between first symptom and treatment initiation in days: median (IQR)		30 (22–60)
Number of screened contacts: median (IQR)		3 (2–4)

In terms of their educational status close to 49% (n = 187) had no education at all and about 37% (n = 143) has only primary education. Their median body mass index was (BMI) 19.7 with an IQR of 17.6–22.4. The median duration between observation of the the first symptom and initiation of treatment was 30 days (IQR; 22–60). On average, three household contacts were screened per each smear positive index case. (Table 1)

In majority, (76%) of the households of the index TB cases, husbands were heads of the household and Wives were identified to be heads in about 18% of the households. About 60% of all the household heads had no education while 33% had primary education. The average family size in the househods was four but ranged between two to nine. Regarding the housing types, thatched hut houses with grass cover accounted for about 48% (n = 184). Most houses had at least one window but about 22% had no windows at all. Majority, (~ 93%) of the households used fire wood or charcoal as a source of energy for cooking. (Table 2)

Table 2

Household characteristics of Smear Positive Index TB cases in Silti Zone of Central Ethiopia, 2020–2021 (n = 387)
Variables		Frequency n (%)
Household head	Wife	71 (18.3)
	Husband	297 (76.4)
	Other	19 (4.9)
Education level of household head	illiterate	232 (59.9)
	Primary	128 (33.1)
	Secondary	12 (3.1)
	Tertiary	15 (3.9)
Family size: median (IQR)		4 (3–5)
House type	Thatched roof houses	184 (47.7)
House type	Houses with corrugated iron roof	202 (52.3)
Room number: median (IQR)		2 (1–3)
Number of windows: median (IQR)		1 (1–2)
House area in m²: median (IQR)		40 (30–64)
Cooking method	Electricity	15 (3.9)
	Fire/Charcoal	357 (92.7)
	Smoke free stove	13 (3.4)

Characteristics of secondary TB cases

Among the total identified 23 TB cases, 22 were smear positive while one case was smear negative but culture positive. Majority, 64% (n = 14) of the TB cases were male and their median age was 28 years (IQR = 18–45 years). About 60% of the cases had no education and 34% had only primary eduation. Close to 92% of the TB cases had no history of previous treatement and majority (96%) were non smokers. About 83% had reported chronic cough and 74% of thse had a productive cough. Only 13% of the cases reported blood stained sputum while about 30% had fever. While 35% of the cases reported night sweating, about 30% also reported weight loss. Nearly 17% reported chest pain and about 65% had a chronic fatigue. The median duration between observation of the first symptom and treatment initiation is 30 days (IQR = 23–30 days). The average daily duration of contact with the index TB cases was 12 hours (IQR = 8–18 hours).

Spatial distribution of Smear Positive Index TB cases in the study area

Results of spatial analysis indicates that the distribution of smear positive TB cases was clustered in the study area. The Global Moran’s I value was 0.70029 (p-value < 0.000) (Fig. 2 left). The maximum distance at which the clustering of the TB cases peaked was at 8 kilometers based on the incremental autocorrelation result (Fig. 2 right).

The highest concentration of cases were seen in Warabet Shama, Werabe town and Kibet town of study area where number of smear positive TB cases ranged from 32–58 per Kebele. Birhan Kitkita, Ashute Burako and Dalocha town also had a higher concentration of smear positive TB cases during the study period. Furthermore, the distribution map shows, higher clustering of the smear positive TB cases was seen in Kebeles situated at the Northern and North Western parts of the study area. On the other hand, most of the Eastern, Southern and South Western parts of the study area had a lower distribution of smear positive TB cases (Fig. 3).

Hotspots and coldsptos of smear positive TB

Hotspot analysis was performed using Gettis Ord GI* statistic to ideantify loal areas of hotspots and coldspots for smear positive TB cases in the study area. GIZ score and p-values were comupted to determine the statistical significance of the local clusters. The higher the GIZ score (both positively or negatively), the higher the significance level. Results with high GI* outputs indicate hotspots while low GI* output indicates coldspots.

As shown in Fig. 4 below, red circles indicate statistically significant hotspots with 99% confidence leve. Orange colored circles indicate places of statistically significant hotspot with 95% confidence level. On the other hand, deep green circles indicate coldspots with 99% confidence where as light green circles indicate coldspots with 95% confidence level. Yellow circles indicate areas where there were no significant clusters.

Thus, statistically significant hotspots were found in Yeteker, Woliya 6, Boze Sabola, Koto Baloso, Asano, Aredol Anshebele, Dildate Mezer, Albazerzemu Shidger, Ashute Burako, Sada Gora and Balo Keriso Kebeles of Silti, Dalocha and Misrak Silti districts. On the other hand, statistically significant coldspots were found in Dalocha town, Dalocha Telqasa, Burqa Delpa, Dube Godibamo and Dobo Bedono kebeles of the Silti and Misrak districts. Hulbareg district had a significant hotspot at Albazerzemu kebele but had no significant coldspot. Birhan Kitkita, Warabet Shama, Kibet twon, Sinana Gerara, Anshabeso, Arat Ber, Dacha Geisila and Agode Lubrera all had no statistically significant clusters. (Fig. 4)

Inverse distance weighting (IDW) interpolation based on the GIZ values predicted that most of the kebeles in Eastern, Western and Southern parts of the study area (shaded in Green and Yellow colors) mostly had a significantly lower clustering of smear positive TB cases. However, the Northern and Central parts (shared in Brown and White colors) had a higher predicted clustering of smear positive TB cases. (Fig. 5)

Predictors of spatial distribution of Smear Postive TB cases

The 387 smear positive TB cases were spatially aggregated into their respective 37 kebeles for the purpose of identifying hotspots and the associated factors. Ordinary Least Square (OLS) regression was first performed to assess the factors associated with the spatial distribution of smear positive TB cases in the study sites of Silti Zone. As it is shown in Table 3 below, the OLS model explained about 97% (Adjusted R2 = 0.973) of the spatial variation of smear positive TB cases in Silti zone of Central Ethiopia. Robust p-value was used to determine the statical significance of the variables included in the final model since the Koenker (BP) statistic was significant. The OLS diagnostics results also show that the Joint Wald statistic was statistically significant (p < 0.001) which indicates that the overall regression model was significant. There was no multicollinearity as evidenced by the small variance inflation factors (VIF < 7.5). Furthermore, the spatial autocorrelation results of the standardized regression residuals shows that they were not spatially correlated i.e. they were randomly distributed (Moran’s I = -0.041951, p < 0.850). (Fig. 6)

Table 3

Regression results and OLS diagnostics for smear positive TB cases in Silti Zone, Central Ethiopia (2020 − 200)
Variables	Coefficient		Standard Error	t-Statistic	Probability	Robust SE	Robust statistic	Robust probability	VIF
Intercept	-7.264		1.992	-3.65	< 0.001	2.069	-3.509	0.001	---
No education	1.241		0.093	13.42	< 0.001	0.080	15.472	< 0.001	2.23
Primary education	1.364		0.087	15.64	< 0.001	0.103	13.205	< 0.001	2.21
Family size	1.853		0.423	4.38	< 0.001	0.531	3.491	0.001	1.08
Hut houses with grass cover	-0.336		0.148	-2.47	0.019	0.115	-3.168	0.003	2.43
OLS Diagnostics					Akaike’s Information Criterion (AIC)				169.23
Multiple R squared		0.977029			Adjusted R squared				0.973324
Joint F statistic		263.71			Prob(> F), (5, 31) degrees of freedom				< 0.001
Joint Wald statistic		3460.6			Prob(> chi squared), (5) degress of freedom				< 0.001
Koenker (BP) statistic		9.236			Prob(> chi squared), (5) degress of freedom				0.097
Jarque-Bera statistic		9.619			Prob(> chi squared), (2) degress of freedom				0.081

In geographically weighted regressin analysis, having no education, having primary education, family size and number of thatched huts with grass cover were significantly associated with the distribution of smear positive TB in the study area. Both no education and primary education were positively associated with an increase in the number of smear positive TB cases. As the number of participants with no formal education increased, the number of smear positive TB cases increased in Misrak Silti kebeles such as Tuto Zagere, Agode Lubrera, Dobo Bedeno and Kuno Kurtefa but had a relatively weaker association in kebeles of Silti district. Similarly an increase in participants with primary eduation had a strong positive association with increased smear positive TB cases in kebeles of Hulbareg district such as Birhan Kitkita, Obisho Wacho and Kerate town. Family size also had a positive association with the number of smear positive TB cases per kebele. An increase in family size had a strong positive association with increased number of smear positive TB cases in kebeles of Silti district. By contrast this association was weaker in Kebeles of Misrak Silti district. On the other hand, living in huts with grass cover was negatively associated with the increased number of smear positivity. An increase in the number of houses with grass cover was associated with a decrease in smear postivity. This association was stronger in parts of Misrak silti district and was weaker in kebeles of Hulbareg and Silit districts.

Distribution of secondary TB cases

The smear positive index TB cases reported a total of 1,276 household contacts during community level house to house visits to screen and test symptomatic individuals. By testing 231 sputum smaples from the household contacts having symptoms of TB, an additional 23 TB cases of all forms were identified. As shown in Fig. 7 below, majority of the secondary TB cases were identified in Misrak Silti district. Fewer cases were identified in Hulbareg and Silti districts where as no secondary TB case was identified in Dalocah district. (Fig. 7)

In spite of the various efforts being made to control TB, it remains to be one of the greatest challenges of the health system in Ethiopia. Evidences on its spatial epidemiology could be helpful in its prevention and control. In this study, we found the distribution of smear positive TB cases to be varying in different parts of the study area and the prevalence ragned from 0.01–1.36%. Findings of the global Moran’s analysis confirms that the distribution of smear positive TB cases in the study area is clustered. Numerous studies report the heterogeniety of TB distribution and clustering of TB cases in certain geographic areas(3–5, 10, 20). Most of these studies describe the distribution of all forms of TB, but we focused on smear positive TB as these are fewer but efficient transmitters of the disease mostly in high burden and resource poor settings. It’s known that smear positive TB cases take the greatest share in the transmission of TB in the community(12), although smear negative cases can also transmit the disease to a lesser extent(21). Early identifiecation and prompt treatment of such cases would prove to be effective in interrupting the transmission and substantially reducing the incidence of new cases(14).

Hotspot analysis further showed areas of statistically significant hotspots and coldspots of smear positive TB cases. Kebeles in Misrak Silti and Silti districts had the commonest staticsticaly significant hotspots. And most of the coldspots were found in Dalocha district. A similar study conducted in Sidama zone in Ethiopia is in harmony with our finding where significant clustering of smear positive TB cases were found in 192 kebeles of eight districts(22). A study in Dabat district of Northen Ethiopia also describes the clustering of smear positive TB cases(23). Another study in one provice of China on smear positive TB also shows global and local spatial autocorrelation of such cases indicating the presence of significant hotspots as well as coldspots(24). This finding can have a double practical advantage in terms of narrowing down the scale as well as the areas of intervention. Programs could be cost and time effective by prioritizing smear positive TB cases through focusing on areas of hotspots. Previous studies consistently confirm the advantage of spatially targetted screening of contacts to improve the yield of TB(10, 11).

In this study, we used inverse distance weighting (IDW) interpolation method to predict the distribution of smear positive TB cases in places other than the measured points thereby developing a continuous rasterized map. According to the results, significant hotspots lie in the Central and Northern parts of the study area with the highest concentration of smear positive TB cases being in Yeteker, Woliya 6 and Boze Sabola kebeles. These places would beneft from a targeted screening of either the entire population or contacts of index cases. On the other hand, significant dispersion of smear positive TB cases was seen in the South Eastern parts of the study area. Such predictions to parts of an area with no measured data is helpful to make informed decisions including prioritization and resource allocation(25, 26). Studies indicate the beneficial application of IDW in identifying the geographical hetetrogeniety of TB and direction of tailored strategies in specific geographic areas(25–27). For instance, such data could be used for systematic surveillance of TB and this could inturn improve the coverage as well as the performance of TB control(23).

The GWR analysis revealed that no formal education, primary education, family size and tachtched houses with grass cover to have a statistically significant association with smear postivive TB cases in the study area. The possible explanation for the association between education and smear positive TB case could be that the less educated individuals are more likely to have inadequate awareness about TB(28, 29) and lower education could also be a barrier to prompt diagnosis and treatment(28). Furthermore, the more individuals are uneducated/less educated in a given area, they more likely face or live in poverity which is one of the known determinats of TB(30). Larger family size is associated with higher risk of developing TB(31) and is also directly linked with crowding. Crowded living conditions are also known to increase the prevalence of TB(32). In this study, hut houses with grass covers are negatively linked with smear positive TB unlike the other factors. These houses usually built in rural areas have a single door and room but the walls usually don’t reach the ceiling there by allowing adequate ventilation for the household.

We identified 23 secondary TB cases of all forms through screening and testing household contacts of the smear positive index TB cases. The screening of household contacts was not geographically targeted. However, distribution of the identified secondary cases closely followed the distribution of the smear positive index TB cases. More secondary cases were identified aroud areas found to have clustering of smear positive TB cases and a fewer or no cases were idenfied in areas with non significant clustering as well as areas of significant coldspots. Secondary cases identified in kebeles such as Ashute Burako and Woliya 6 correspond with some of the statistically significant hotspots of smear positive index TB cases we have found. Similarly, the absence of secondary TB cases in Dalocha district corresponds with the finding that the district was mostly a cold spot with a fewer/dispersed distribution of smear positive TB cases. This finding again confirms the importance of targeted intervention to improve case detection as well as reducting transmission of the disease in the community(10, 11).

The use of GIS has pinpointed the areas with statistically significant hotspots/clustering of smear positive TB cases which would otherwise be difficult through other means of statistical analysis. The other strength is the use of geographically weighted regression to identify predictors which directly links the coefficients of the predictors with specific places in the study area. The study also used both secondary data collected though routine service as well as primary data collected through community level house to house screening to assess relationships. On the other hand, the study had several limitations. Some important kebele level aggregate predictor variables were not included during data collection. Temporal dimension of the clustering of TB was not assessed as only two years data was used. This would have informed about the stability of the spatial clusters overe a period of time.

Our study has revealed that the distribution of smear positive TB cases is clustered in the study area. Statistically significant hotspots of smear positive TB cases were found in 11 Kebeles of Silti, Dalocha and Misrak Silti districts. Signficant coldspots were also found in five kebeles of the Silti and Misrak districts. Location based targeted interventions such as active screening in hotspot areas could yield more TB cases and would in turn reduce the transmission of TB. The GWR has shown that educational status, family size and housing type to be the factors with significant influence on the spatial distribution of smear positive TB in the study area. We have also found that the locations of the secondary TB cases found through household contacts screening coincides with the hotspots identified through the spatial analysis which indicates higher transmission of the disease in these places.

AIC

Akaike’s Information Criterion

EPHI

Ethiopian Public Health Institute

EPTB

Extrapulmonary Tuberculosis

GEE

Generalized Estimating Equations

GIS

Geographic Information System

HEW

Health Extension Worker

IQR

Inter Quartile Range

IDW

Inverse Distrance Weighting

Standard Error

aOLS

Ordinary Least Squares

RHB

Regional Health Bureau

Tuberculosis

WHO

World Health Organization

VIF

Variance Inflation Factor

Ethical approval and consent to participate

Ethical approval for this study was attained from Jimma University (Ref. no. IHRPGD/320/18) and the Ethiopian Public Health Institute (Ref. no. EPHI 6.13/592). Official support letters to conduct the study were also obtained from SNNPR RHB and Silti Zonal Health Office. All participants of the study were provided with clear and adequate information regarding TB disease and the current study. Additionally, written informed consents were obtained from all index cases and household contacts, including minors through their parents or guardians. We kept confidentially of the individual records strictly. Children aged five or younger were provided with TB preventive theraphy and confirmed TB patients were treated with standard TB treastment according to the national guidelines. Geographic coordinates of the households were taken from about 20 meters away from the gate to prevent identification.

Competing interests

The authors have no competing interests to declare.

Authors’ contributions

HM wrote the study protocol, supervised and participated in data collection, conducted data analysis and wrote the first draft of the manuscript. MG and GS performed the laboratory analysis of sputum specimens to identify secondary TB cases. GA and DY reviewed the study protocol and supervised the conduct of the study. GA, DY, MG and GS throughly reviewed the manuscript. All the authors have read and approved the final version of the manuscript.

Funding

This study was funded by Jimma University Insititute of Health as a routine part of supporting PhD research. However, the Universtiy did not have any role in the design and conduct of the research as well as the decision to publish.

Availability of data and materials

The raw dataset of this study is abvailable and could be obtained from the corresponding author upon reasonable request.

Acknowledgments

We are grateful to Jimma University, Institute of Health, SNNPR RHB and Silti Zonal Health Office for providing ethical approval and support letters to carry out this study. Our heartfelt thanks also goes to the Ethiopian Public Health Institute for providing ethical clearance, giving permission to use the laboratory testing facility with the necessary equipment and reagents needed to conduct this study. Finally, we would like to strongly acknowledge the participants of the study and the data collectors.

Consent for publication

Not applicable.

Global tuberculosis report 2022. Geneva: World Health organization. 2022. licence: cc bY-Nc-sa 3.0 iGo. World Health Organization. 2022.
Maciel ELN, Pan W, Dietze R, Peres RL, Vinhas SA, Ribeiro FK, et al. Spatial patterns of pulmonary tuberculosis incidence and their relationship to socio-economic status in Vitoria, Brazil. Int J Tuberculosis Lung Disease. 2010;14(11):1395–402.
Verver S, Warren RM, Munch Z, Richardson M, van der Spuy GD, Borgdorff MW, et al. Proportion of tuberculosis transmission that takes place in households in a high-incidence area. Lancet. 2004;363(9404):212–4.
Dowdy DW, Golub JE, Chaisson RE, Saraceni V. Heterogeneity in tuberculosis transmission and the role of geographic hotspots in propagating epidemics. Proc Natl Acad Sci U S A. 2012;109(24):9557–62.
Robsky KO, Robsky KO, Kitonsa PJ, Mukiibi J, Nakasolya O, Isooba D, et al. Spatial distribution of people diagnosed with tuberculosis through routine and active case finding: A community-based study in Kampala, Uganda. Infect Dis Poverty. 2020;9(1):1–10.
Theron G, Jenkins HE, Cobelens F, Abubakar I, Khan AJ, Cohen T, et al. Data for action: Collection and use of local data to end tuberculosis. Lancet. 2015;386(10010):2324–33.
Ho J, Fox GJ, Marais BJ. Passive case finding for tuberculosis is not enough. Int J Mycobacteriol. 2016;5(4):374–8.
Shamanewadi AN, Naik PR, Thekkur P, Madhukumar S, Nirgude AS, Pavithra MB, et al. Enablers and Challenges in the Implementation of Active Case Findings in a Selected District of Karnataka, South India: A Qualitative Study. Tuberc Res Treat. 2020;2020:1–10.
Mohammed H, Oljira L, Roba KT, Ngadaya E, Ajeme T, Haile T, et al. Burden of tuberculosis and challenges related to screening and diagnosis in Ethiopia. J Clin Tuberc Other Mycobact Dis. 2020;19:100158.
Cudahy PGT, Andrews JR, Bilinski A, Dowdy DW, Mathema B, Menzies NA et al. Spatially targeted screening to reduce tuberculosis transmission in high incidence settings: a systematic review and synthesis Patrick. 2020;19(3):1–16.
Nguyen L, Vo Q, Vu TN, Nguyen HT, Truong T, Khuu CM et al. Optimizing community screening for tuberculosis: Spatial analysis of localized case finding from door-to-door screening for TB in an urban district of Ho Chi Minh City. Viet Nam. 2018;1–15.
CDC. Guidelines for the investigation of contacts of persons with infectious tuberculosis. Recommendations from the National Tuberculosis Controllers Association and CDC. Vol. 54, MMWR. Recommendations and reports: Morbidity and mortality weekly report. Recommendations and reports / Centers for Disease Control; 2005.
Hopewell PC, Fair E, Miller C, World Health Organization. Communicable Diseases Cluster. Stop TB Department. Recommendations for investigating contacts of persons with infectious tuberculosis in low-and middle-income countries. Geneva: World Health Organization; 2012. p. 65.
WHO. WHO consolidated guidelines on tuberculosis Module 2: Screening Systemic screening for tuberculosis disease. World Health Organ. 2020. 1–68 p.
Agency CS. Ethiopian Central Statistical Agency. Ethiopian 2022 Project Population Census. 2022;(July).
MoH. Federal Democratic Republic of Ethiopia Minstry of Health,National Guidlines for TB,DR-TB and Leprosy in Ethiopia, sixth edition. 2017;12–31.
Silti Zone Health Department. Silti Zone 2014 E.C. annual TB performance report. 2022;1–9.
Chen Y. New Approaches for Calculating Moran’s Index of Spatial Autocorrelation. PLoS ONE. 2013;8(7).
Wu YH (Eva), Hung MC, Hung MC, editors. Rijeka: IntechOpen; 2016. p. Ch. 2.
Munch Z, Van Lill SWP, Booysen CN, Zietsman HL, Enarson DA. 3 ; Beyers N. uberculosis transmission patterns in a high-incidence area: a spatial analysis.
Tostmann A, Kik SV, Kalisvaart NA, Sebek MM, Verver S, Boeree MJ, et al. Tuberculosis transmission by patients with smear-negative pulmonary tuberculosis in a large cohort in the Netherlands. Clin Infect Dis. 2008;47(9):1135–42.
Dangisso MH, Datiko DG, Lindtjørn B. Spatio-temporal analysis of smear-positive tuberculosis in the Sidama Zone, Southern Ethiopia. PLoS ONE. 2015;10(6).
Tadesse T, Demissie M, Berhane Y, Kebede Y, Abebe M. The Clustering of Smear-Positive Tuberculosis in Dabat, Ethiopia: A Population Based Cross Sectional Study. PLoS ONE. 2013;8(5):1–6.
Jian-mei WANG, Nan QIN, Yue WANG, Lin ZHOU, Wei-bing WANG YZ. Spatial analysis on smear-positive pulmonary tuberculosis in Zhejiang province between year 2015 and 2017. 40, Chinese Journal of Antituberculosis. p. 1089–94.
Rasoul S, Shahram H, Jafar M, Mohsen S, Eslam MA. Spatial Epidemiology of tuberculosis in Ardabil Province: Based on Geographical Information System. Pakistan J Med Health Sci. 2019;13(3):965–9.
Rani RN, Rameshwari TKR, Kumar S. Detection of tuberculosis hotspots using spatial interpolation method in Mysuru District, Karnataka. J Appl Biol Biotechnol. 2021;9(2):124–30.
Lin H, Zhang R, Wu Z, Li M, Wu J, Shen X, et al. Assessing the spatial heterogeneity of tuberculosis in a population with internal migration in China: a retrospective population-based study. Front Public Health. 2023;11(May):1–12.
Abdu M, Balchut A, Girma E, Mebratu W. Patient Delay in Initiating Tuberculosis Treatment and Associated Factors in Oromia Special Zone, Amhara Region. Pulm Med. 2020;2020.
Mahmud S, Mohsin M, Irfan SH, Muyeed A, Islam A. Knowledge, attitude, practices, and determinants of them toward tuberculosis among social media users in Bangladesh: A cross-sectional study. PLoS ONE. 2022;17(10 October):1–19.
Marais BJ, Hesseling AC, Cotton MF. Poverty and tuberculosis: is it truly a simple inverse linear correlation? Eur Respir J. 2009;33(4):943. LP – 944.
Yohanes A, Abera S, Ali S. Smear positive pulmonary tuberculosis among suspected patients attending metehara sugar factory hospital; eastern Ethiopia. Afr Health Sci. 2012;12(3):325–30.
Baker M, Das D, Venugopal K, Howden-Chapman P. Tuberculosis associated with household crowding in a developed country. J Epidemiol Community Health (1978). 2008;62(8):715 LP – 721.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Mapping the distribution of tuberculosis cases identified through routine program and community based active screening in Central Ethiopia

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Background

Methods

Study setting and period

Study design

Definitions

Data collection

Data management and statistical analysis

Spatial analysis

Geographically Weighted Regression (GWR)

Results

Study population and TB prevalence

Individual and household characteristics of Smear Positive Index TB cases

Characteristics of secondary TB cases

Spatial distribution of Smear Positive Index TB cases in the study area

Hotspots and coldsptos of smear positive TB

Predictors of spatial distribution of Smear Postive TB cases

Distribution of secondary TB cases

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1