SORT IT KB Key Population Tuberculosis in key populations in Tajikistan – a snapshot in 2017

Introduction: WHO End TB Strategy aims at achieving targets of 90% mortality reduction and 80% reduction in tuberculosis (TB) incidence by 2030, recommending better addressing TB and multidrug-resistant TB (MDR-TB) issues in key populations. Aim: The study aimed at having a snapshot of the epidemiological characteristics of the key populations among the new TB patients, registered in Tajikistan during 2017. Methodology: A cross-sectional study was conducted, using official TB registration data for all new TB case notification in Tajikistan in 2017. Results: The key population included 1,029 (19.8%) patients among all 5,182 new TB cases registered in 2017. The following selected subpopulations were identified: migrant workers – 728 (70.7%), diabetics – 162 (15.7%), HIV-positive – 138 (13.4%), heavy drinkers – 74 (7.2%), drug users – 50 (4.8%), ex-prisoners – 50 (4.8%), and homeless – 9 (0.9%). Among the key population, 307 (29.8%) patients were smearpositive, 145 (14.1%) were drug-sensitive and 116 (11.3%) had MonoDR/MDR-TB. Time to treatment initiation for smear-positive cases was ≤ 5 days for 303 (98.7%) patients. Being a key population was inversely related to gender (female) (OR = 0.25, 95% CI (0.21, 0.29)) and population type (rural) (OR = 0.64, 95% CI (0.55, 0.74)). Conclusion: Among the key population the identified overlaps of selected sub-populations would enable more efficiently reaching the certain groups. TB case detection at PHC levels needs to be targeted for improved rates for key population detection. In the key population sub-group of migrant workers’ special migration destinations are recommended to be explored and find out possible associations with drug resistance.


Introduction
Tuberculosis (TB) is one of the most important public health problems in the Republic of Tajikistan, a lower-middle income country in Central Asia [1].
In 2017, the estimated TB incidence was reported to be 85 per 100,000 population and the prevalence 126.9 per 100,000 population [2,3]. Although the prevalence rate of TB has dropped to 1.8-fold in the last twelve years (from 238 cases in 2005 to 126.9 per 100,000 population in 2017) the proportion of multidrug-resistant TB (MDR-TB) among new TB cases increased from 12.5% in 2010 to 20% in 2017 [3]. The proportion of MDR-TB among previously treated cases was 53.6% in 2010 and decreased to 45% in 2017 according to the National Drug Resistance Survey of Tajikistan [4].
According to the Stop TB Partnership, the Global Plan to End TB recommends to reach at least 90 percent of all people with TB placing them on appropriate therapy, including first-and second-line treatment and preventative therapy; to reach at least 90 percent of the key populations, and to achieve at least 90 percent treatment success for all people diagnosed with TB through affordable treatment, services, adherence to complete and correct treatment and social support [5].
The identified sub-populations that are more prone to TB either due to more environmental, biological or behavioral risks or barriers in accessing public services they face. Many of these key populations overlap and 95S are exposed to multiple risks making delivering TB services to them even more challenging [6]. Therefore, depending on a specific country, the various factors impacting a national the TB response among its key populations might differ [6].
In order to attain the WHO End TB Strategy and to achieve the target of 90% mortality reduction and 80% reduction in TB incidence by 2030, TB and MDR-TB issues would need to be better addressed in key populations [7,8]. The 2015-2020 National Strategic TB Prevention Plan in Tajikistan recommends special attention among TB special key populations, such as the HIV, migrants, prisoners and other groups, for ensuring access to quality TB diagnosis based on the demands of the special key populations [9].
According to the 2017 WHO Report on TB in Tajikistan, the incidence of HIV/TB coinfection was estimated to be 3.1 per 100,000 population, while the proportion of patients with HIV-positive status among new and relapsed TB patients was reported to be about 4% [10,11].
The proportion of the migrants among newly registered TB cases increased from 13.5% of all cases in 2011 to 19.7% in 2015 [12]. In 2017, the rate of TB among the migrant workers was reported to be 15.3% [10].
In the reviewed literature form PubMed on key populations in Tajikistan, it was found that the TB transmission risk is considerably higher for healthcare workers who are in close contact with TB patents such as in dedicated TB care facilities [13]. Tuberculosis notification rates among healthcare workers ranged between two and ten-fold compared to those of the general population (6-year average: 562/100,000 vs 88/100,000, respectively) during 2009-2014 [14].
According to the literature, the prevalence of the TB among the prisoners in Tajikistan was about 4.5%, which is estimated to be about 50 times more compared to the general population rate in 2010 [15].
No published information was identified regarding the TB rates among other TB key populations in Tajikistan, such as drug users and diabetics, heavy drinkers and homeless.
Therefore, defining and targeting the TB key populations and understanding situations faced by them will further foster risks identification, the factors driving these risks, the challenges in providing services to key populations as well as to further improve TB prevention and control programs in Tajikistan in general.
The aim of the study is to provide a snapshot of the TB key populations, by analyzing the epidemiological characteristics of the new TB patients registered in Tajikistan, during the year 2017.
This study is the first attempt to identify and quantify the key populations in Tajikistan, among the new TB cases in 2017, including: HIV-positive, drug users, heavy drinkers, diabetics, migrant workers, exprisoners and homeless; demographic characteristics of the key populations are compared to non-key population.

Study Design
This is a cross-sectional study, based on the official TB registration data for all new TB case notification in Tajikistan in 2017.

General
The Republic of Tajikistan is a mountainous country (93% of mountains) in Central Asia covering an area of 142,600 km 2 . The population is approximately 9 million, of which 5.9 million (73.6%) live in rural areas [3]. Tajikistan borders Afghanistan, Uzbekistan, Kyrgyzstan and China.

National Tuberculosis Program
In 2002 the country set up a National Tuberculosis Program which follows WHO guidelines for TB [16,17], and implementation of Directly Observed Treatment Strategy (DOTS) Strategy was applied country-wide from 2002 to 2004. The National Tuberculosis Program is implemented by Ministry of Health and Social Protection of Republic of Tajikistan that owes Republic TB Center coordinating, Oblast (regional) TB centres and Dushanbe (capital city) TB centre. For prisoners, Ministry of Justice is also collaborating with Ministry of Health and Social Protection.
National TB control measures are implemented by a network of TB facilities and through the primary healthcare system. The country operates 76 TB ambulatory centres, 4 regional centres and 34 TB hospitals [9]. The total number of beds in TB hospitals in the Republic is more than 1,500. The prison system has one TB hospital with a specific ward for patients with MDR-TB. Funding for TB activities is mainly from international donor organizations (about 80%). The TB patients, including patients with drug-resistant forms, receive standardized treatment regimens in accordance with national guidelines, which are based on the WHO TB guidelines [9,16,17]. According to the National TB control strategy in Tajikistan, for all new 96S TB cases an official TB registration form-089 is completed by a TB specialists. The TB notification form includes demographic information, migration history, previous contacts with TB cases, history of imprisonment, drug and alcohol use, HIV positivity and main concomitant diseases, as well as results of laboratory tests (microscopy and Gene Expert (GX) of HAIN tests), facility that diagnosed TB case, circumstances of TB notification, presence of lung decay, etc.
The TB laboratory network in Tajikistan is organized as a three-level hierarchically organized diagnostic system, consisting of the NRL at the top level, three culture laboratories at medium level (the National Public Health Reference Laboratory in Dushanbe, the oblast reference laboratory of Kulyab, which serves the Khatlon oblast, and the reference laboratory of Digmoy, serving the Sugd oblast) and 72 microscopy laboratories at the base [18]. A diagnostic algorithm was developed to improve early detection of MDR-TB cases, using all available diagnostic methods. Besides smear microscopy, Xpert MTB/RIF is recommended for all suspected TB cases.

Study population
The research included the new TB cases registered in 2017 in Tajikistan. The exclusion criteria were the following: being imprisoned during the study period and "previously" treated TB patients, inadvertently reported as "new" cases on the national registration form.

Data collection
Data from the official registration form-089 for all new TB case notification were collected and entered into EpiInfo database (version 7.1.5) by trained staff and then reviewed and compared to the original registration forms for accuracy.

Analysis and statistics
Variables on social, demographic and clinical characteristics of TB patients, related to the study objectives were collected. A variable of "key population" was obtained by combining variables reflecting selected key sub-populations defined above: HIV-positive, drug users, heavy drinkers, diabetics, migrant workers, ex-prisoners and homeless. A variable of "heavy drinker" was based on TB specialist's judgement based on patient's provided information on their alcohol usage, extent and not as a result of Drug Dispensary expertise. The variables of "drug user", "migrant worker", "ex-prisoner" and "homeless" were captured based on the reported information in registration form-089.
Summary statistics were used to report the results. Descriptive analysis (mean ± standard deviation (SD) for continuous variables and frequencies for categorical variables) was conducted for the variables of interest and relevant socio-demographic variables. Differences in characteristics between the key populations were compared using χ 2 and Fisher's exact test for categorical variables and t-test for numerical variables. EasySTAT web based statistical application was used for data analysis (available at https://easystat.app/).

Ethics approval
The ethical approval for this study was obtained from the biomedical Ethics Committee of the Academy of Medical Science of the Ministry of Health and social protection of population of the Republic of Tajikistan (protocol#7/2019). This study was based only on the data of the registers collected by TB specialists. Based on this information, the informed consent was not deemed necessary. The personal identifiers were not databased, and the anonymization of the data was ensued.

Results
Data on a total of 5,182 new TB cases were captured out of 5,231 registered cases (99.1%). The key population included 1,029 (19.8%) patients among all new TB cases registered for the year 2017 in Tajikistan (Figure 1).
The characteristics of key population were compared to non-key population considering the following factors: age, gender, education level, marital status, employment status, population type, microscopy results, time to treatment initiation among smearpositive, drug resistance status, and TB detection facility. According to the results, females were by 75% (odds ratio (OR) = 0.25, 95% confidence interval (CI) (0.21, 0.29), p < 0.001) less likely to be a key population compared to males. Patients from rural areas were by 36% less likely (OR = 0.64, CI (0.55, 0.74), p < 0.001) to be a key population as compared to urban patients.

Discussion
This study aimed at identifying selected TB subpopulations as key population, including HIV-positive, drug users, heavy drinkers, diabetics, migrant workers, ex-prisoners and homeless, in order to more efficiently reach these groups to address targeted delivering of TB services, in accordance with WHO End TB Strategy [7,8]. The proportions of selected sub-populations were calculated, among which the data for drug users,

98S
homeless, heavy drinkers and diabetics previously unreported in the literature. Among the key population the high proportion (about 77%) of patients were unemployed, consistent with literature data [19]. Interestingly, approximately half of the key population (45.5%) were married: considering that about 70% of the key population were migrant workers, this could be explained by the need to migrate for work as a caregiver for family.
While in Tajikistan it is recommended to initiate a treatment after diagnosis not later than 5 days, interestingly, in only 4 (1.3%) cases among key population the time to treatment initiation for smearpositive patients was more than 5 days, which is indicative of adequate rates for treatment timely administration.
It was found out that PHC facilities detect about 30% of new cases among key population which implied TB case detection at PHC levels needs to be targeted in order to improve key population detection rates.
Since migrant workers represented about 70% of the key population, further study would be needed to define migration destinations (urban-rural, national and international) in the key population sub-group and find out possible associations with drug resistance.
The protective effect of being female for being a key population is explained by several patterns, like males being more vulnerable to TB due to genderspecific occupations, more likely to migrate for work, to smoke or use drugs in many societies and generally less likely to have their TB detected and reported than women [20]. For rural population the protective effect from being a key population might be explained by the fact of having more close communities in rural areas, where alcohol use, drug use, imprisonment, etc., are considered unfavorable behaviors.
Despite the sample size, there was a large amount of missing data, posing completeness and registration bias. This highlights the need for improved quality of data capture and more thorough reporting by TB specialists. Moreover, the collected information makes it unreasonable to suggest policy-related recommendations. However, characterization of the key population and calculation of selected subpopulations would enable reaching the certain groups more efficiently. Another limitation to consider would be the reporting and recall bias by the patients.
This study aimed at having a snapshot of the epidemiological characteristics of the key populations among the new TB patients in Tajikistan for the single year 2017, but does not describe any trends or time patterns in disease control protection of the population. Additionally, since outcome of the disease was not considered in the current study, further study on clinical outcomes and survival rates would be needed for a more comprehensive analysis of the situation in the country regarding targeting key populations.

Conclusion
In the current study the key population resulted to be about 20% of the whole new cases reported for the year 2017. Among the key population the overlaps of selected populations (HIV-positive, drug users, heavy drinkers, diabetics, migrant workers, ex-prisoners and homeless) were identified. This could enable further reaching the certain groups more efficiently for targeted delivery of TB services, in accordance with WHO End TB Strategy.
As observed by findings, in all the selected populations, the proportion of migrant workers predominated. This allows us recommending more proper targeting of this subgroup to more efficiently reach them. The obtained study results pose implications for practice for addressing the key population and its constituent sub-populations with high risk more efficiently. Additionally, case detection at PHC levels needs to be targeted for improved rates for key population detection. In the key population subgroup of migrant workers special migration destinations (urban-rural, national and international) are recommended to be researched further and find out possible associations with drug resistance.