Tuberculosis Notification Trends and Treatment Outcomes in Penitentiary and Civilian Health Care Sectors in the WHO European Region

Setting: Tuberculosis (TB) morbidity in penitentiary sectors is one of the major barriers to ending TB in the World Health Organization (WHO) European Region. Objectives and design: a comparative analysis of TB notification rates during 2014–2018 and of treatment outcomes in the civilian and penitentiary sectors in the WHO European Region, with an assessment of risks of developing TB among people experience incarceration. Results: in the WHO European Region, incident TB rates in inmates were 4–24 times higher than in the civilian population. In 12 eastern Europe and central Asia (EECA) countries, inmates compared to civilians had higher relative risks of developing TB (RR = 25) than in the rest of the region (RR = 11), with the highest rates reported in inmates in Azerbaijan, Kazakhstan, Kyrgyzstan, Republic of Moldova, Russian Federation, and Ukraine. The average annual change in TB notification rates between 2014 and 2018 was −7.0% in the civilian sector and −10.9% in the penitentiary sector. A total of 15 countries achieved treatment success rates of over 85% for new penitentiary sector TB patients, the target for the WHO European Region. In 10 countries, there were no significant differences in treatment outcomes between civilian and penitentiary sectors. Conclusion: 42 out of 53 (79%) WHO European Region countries reported TB data for the selected time periods. Most countries in the region achieved a substantial decline in TB burden in prisons, which indicates the effectiveness of recent interventions in correctional institutions. Nevertheless, people who experience incarceration remain an at-risk population for acquiring infection, developing active disease and unfavourable treatment outcomes. Therefore, TB prevention and care practices in inmates need to be improved.


Introduction
In 2019 in the world, 10 million people developed TB disease, and 1.4 million died from TB. Although WHO European Region carries only 3% of the global burden of tuberculosis (TB), it has one of the highest proportions of multidrug-resistant TB (MDR-TB). In 2019, an estimated 246,000 incident TB cases occurred in countries of the WHO European Region, equivalent to an average incidence of 26 cases per 100,000 population [1].

Setting
The WHO European Region consists of 53 Member States covering a vast geographical region from the Atlantic to the Pacific oceans and from the Mediterranean to the Baltic Sea [15]. Our study focused on comparative analyses of TB indicators in eastern Europe and central Asia (EECA) countries as well as of those from the rest of the region. The EECA region is made up of 12 of the 15 countries, which were formerly part of the Soviet Union, excluding the Baltics, and which are located in the east of the WHO European Region.

Study Population and Design
Data were collected for new and relapse TB cases and their outcomes from the civilian and penitentiary sectors reported in WHO European Region countries. Three selection criteria were applied: (1) countries that provided at least one report on new and relapse TB cases in both the civilian and penitentiary sectors between 2014 and 2018; (2) countries that provided at least two data points on new and relapse TB cases in both civilian and penitentiary sectors between 2014 and 2019 for enabling analysis of the trend; (3) countries that reported outcomes for TB cases on first-line drug (FLD) treatment schemes in both the civilian and penitentiary sectors for at least one cohort between 2012 and 2016.

Data Variables and Sources
Data were obtained from 3 sources: (1) The WHO global TB data collection system [14] has an extended set of indicators for TB in European Region prisons, and data were ex-tracted on: prison populations, the numbers of new and relapse TB cases in the civilian and penitentiary sectors for 2014 to 2018, and treatment outcomes for patients on FLD treatment schemes in the civilian and penitentiary sectors for 2012 to 2016; (2) total population estimates were extracted from World Population Prospects [16]; and (3) prison population estimates were taken from the World Prison Brief [17] for countries whose prison population data were missing from The WHO global TB data collection system.

Analysis and Statistics
For each country, we calculated annual notification rates per 100,000 population of new and relapse tuberculosis cases in civilian and penitentiary sectors separately. The Average Annual Percent Change (AAPC) was calculated by fitting a least-squares regression line to the natural logarithm of the rates, using the calendar year as a regressor variable.
As a measure of the effect of exposure to a prison setting on the risk of development of TB we computed the Relative Risk of TB in prison in reference to the civilian population and the corresponding confidence interval. Results were considered significant if the confidence interval did not include 1. The statistical analysis was performed using the online calculator VassarStats [18]. TB patients who were successfully treated or completed TB treatment were considered to have a favourable outcome; those who failed to complete treatment, were lost to follow-up or who died during the TB treatment were considered to have an unfavourable outcome, as per WHO standard definitions [19]. TB cases with no reported treatment outcomes were excluded from the analysis. We analysed the notification rate of incident TB cases and TB treatment outcomes (unfavourable versus favourable) for the civilian and penitentiary sectors.

Notification Rate of Incident TB Cases and Relative Risks of Developing TB Disease in the Penitentiary Sector Compared with the Civilian Sector
Out of the 53 countries of the WHO European Region, 42, including 10 from the EECA region, reported the number of new and relapse TB cases in the civilian and penitentiary sectors at least once in the five-year period between 2014 and 2018. During this time, 11 (21%) countries did not provide any reports on TB in prisons (Table 1 and Figure 1).   Czechia  458  489  497  474  −  16  19  14  25 22 Estonia  230  197  180  168  140  6  9  8  3 Georgia a  3099  3070  2926  2539  2272  101  82  57  58 43 Y   Netherlands  798  845  863  757  784  16  5  14  19  7  Y  North  Macedonia  280  278  260  206  214  4  4  3  10 3 Y Poland  6387  6065  5927  5365  5025  152  172  216  170  171  Y  Portugal  2198  2053  1833  1728  1812  53  61  39  32  44  Y  Republic of  Moldova a  3937  3484  3398  3259  2933  121  124  173  99 Tajikistan a  5677  5804  5866  5794  5605  130  90  99  101  121  Y  Turkey  12,966  12,413  12,035  11,696  11,421  142  137  151  125 Czechia  536  452  453  481  492  20  16  14  18 13       Table 2 shows the notification rate of incident TB cases (The notification rate of incident TB cases is the number of new and relapse tuberculosis cases reported per 100,000 population [1]) and percentage annual changes in notification rate of incident TB cases in the civilian and penitentiary sectors during 2014-2018 for the countries included in this study. In the penitentiary sectors of seven countries, all of which are in the EECA region, incident TB rates of more than 1000 per 100,000 population were reported: Azerbaijan, Kazakhstan, Kyrgyzstan, the Republic of Moldova, the Russian Federation, Tajikistan and Ukraine (Table 2 and Figure 2).
There was observed a decreasing trend in the notification of new TB cases and relapses both in the penitentiary and in the civil sector in 2014-2018 (Table 2 and Figure 3).
In the 42 countries analysed, the average annual change in incident TB rates during the study period was −7.0% in the civilian sector and −10.9% in the penitentiary sector. The decline in incident TB rates among inmates in the nine EECA countries included in this study should be noted (from −6.0% in Kyrgyzstan to −16.5% in Georgia) ( Table 2). TB cases registered in prison's inmates accounted for approximately 7% for all notified new and relapse TB patients in EECA countries, with the highest level in the Russian Federation (10%); in comparison, in the other countries in the region the proportion was 1.5%, with the highest level in Slovakia (6.3%) in 2014-2018.   Table 2 shows the notification rate of incident TB cases (The notification rate of incident TB cases is the number of new and relapse tuberculosis cases reported per 100,000 population [1]) and percentage annual changes in notification rate of incident TB cases in the civilian and penitentiary sectors during 2014-2018 for the countries included in this study. In the penitentiary sectors of seven countries, all of which are in the EECA region, incident TB rates of more than 1000 per 100,000 population were reported: Azerbaijan, Kazakhstan, Kyrgyzstan, the Republic of Moldova, the Russian Federation, Tajikistan and Ukraine (Table 2 and Figure 2).
There was observed a decreasing trend in the notification of new TB cases and relapses both in the penitentiary and in the civil sector in 2014-2018 (Table 2 and Figure 3).       In the 42 countries analysed, the average annual change in incident TB rates during the study period was −7.0% in the civilian sector and −10.9% in the penitentiary sector. The decline in incident TB rates among inmates in the nine EECA countries included in this study should be noted (from −6.0% in Kyrgyzstan to −16.5% in Georgia) ( Table 2). TB cases registered in prison's inmates accounted for approximately 7% for all notified new and relapse TB patients in EECA countries, with the highest level in the Russian Federation (10%); in comparison, in the other countries in the region the proportion was 1.5%, with the highest level in Slovakia (6.3%) in 2014-2018.   Prison's inmates in the Russian Federation and Slovakia had the highest risks of developing an active TB disease compared with their respective civilian populations (RR = 25, confidence interval CI: 25-26, and RR = 57 (CI: 35-92)), respectively, in the last reported year. (Table 3 and Figure 4).

Treatment Outcomes in TB Patients on First-Line Drug (FLD) Treatment Schemes
A total of 39 (74%) countries in the WHO European Region reported treatment outcomes for at least one cohort of TB patients, both civilians and inmates, who started on one of the FLD treatment schemes between 2012 and 2016 (Table 1 and Figure 1). Table 4 shows both the favourable and unfavourable treatment outcomes for civilians and inmates in these 39 countries.

Treatment Outcomes in TB Patients on First-Line Drug (FLD) Treatment Schemes
A total of 39 (74%) countries in the WHO European Region reported treatment outcomes for at least one cohort of TB patients, both civilians and inmates, who started on one of the FLD treatment schemes between 2012 and 2016 (Table 1 and Figure 1). Table 4 shows both the favourable and unfavourable treatment outcomes for civilians and inmates in these 39 countries.        Our study highlights a few countries where there were higher levels of unfavourable outcomes for inmates when compared with civilians, for example, Cyprus (100% vs. 0%), the Netherlands (29% vs. 9%) and Kazakhstan (21% vs. 9%). On the other hand, a higher proportion of unfavourable treatment outcomes among civilians than among inmates had been registered in the Czech Republic (28% vs. 8%), Andorra (25% vs. 0%), Estonia (20% vs. 0%), Armenia (19% vs. 0%), and Slovenia (16% vs. 0%).
A total of 12 of the 39 countries achieved TB treatment success rates of over 85% among inmates. In two EECA countries, Belarus and Tajikistan and in five other countries, Bulgaria, Latvia, Montenegro, Romania, Slovakia, the favourable outcomes were more than 85% in both sectors civilian and penitentiary.

Discussion
Recent systematic review by Cords et al. revealed a concerning scale of TB burden among people experiencing incarceration in different parts of the world and highlighted the high risk of contracting M tuberculosis infection and developing active disease, compared to the general population [12].
This is the first standardized study on TB morbidity and its treatment outcomes monitoring in the penitentiary sectors of such a scale in the WHO European Region. The main finding of our study is that from 2014 to 2018 the annual incident TB notification rates in prisons across the European Region decreased much faster than in the civilian population, which most likely reflects the decline of true burden in the prison populations. This finding highlights the positive impacts of the TB control interventions carried out by national governments, with the support of international agencies [20,21]. Another finding of our study is that, even though the annual decline of the TB burden in WHO European Region prisons was faster than in the civilian sector, the risk of developing TB disease in prisons is up to 57 times higher compared to the civilian sector. The increased risk of TB for inmates in EECA countries is a known feature of the region and has been previously described in several studies [11,[22][23][24]. High prevalence of active TB disease in correctional facilities is fuelled by intra-institutional transmission due to prolonged stays in overcrowded facilities with poor ventilation, along with risk factors, which amplify the risk of TB disease, such as HIV, malnutrition, diabetes, smoking, a history of alcohol and illicit drug consumption, and former TB disease [9,10,25,26]. Theoretically, prison settings offer great opportunities for TB control, and there are practical examples from the region's prisons in which significant improvements in their TB and rifampicin-resistant-TB burdens have been reported, and WHO-recommended screening, diagnostics, treatment, and linkage to civilian health care is ensured [27][28][29][30]. The high TB morbidity rates in the region's prisons, of up to 1623 per 100,000 population in 2018, underline the need for substantial improvements in TB control among inmates through wider application of the best practices in the field.
In the majority of EECA countries, treatment success rates for TB in inmates were lower compared with rates in civilian populations, which was not evident for the other countries in the region. This emphasizes the critical need for improvements in the TB services available to inmates. Although the specific reasons for unfavourable treatment outcomes in inmates were not analysed in this study, there is evidence that high drug-resistance rates, insufficient laboratory diagnosis capacities and weak integrations between civilian and prison healthcare services, including ensuring the continuity of TB treatment after release from prison, are major factors leading to poor treatment outcomes in prisons [6,31].
Decarceration and other countries' justice reforms that lead to it would reduce overcrowding, which is a major environmental factor for tuberculosis transmission, and would significantly reduce TB burden and its rising rates in prisons. Meanwhile, improving the TB situation and treatment outcomes for inmates can only be achieved with governmental commitment, inter-department cooperation for ensuring interventions equivalent to those in the civilian system and in close collaboration with it, and partnerships with civil society organizations. National tuberculosis programmes (NTPs) should develop operational plans and policies that optimize TB control in prisons and for inmates after their release and strengthen the capacities of prison health units for TB case management. Improving treatment outcomes for inmates will also prevent transmission of disease to other inmates, prison staff and community members. The End TB Strategy goals [32] will not be met without the implementation of effective measures in prisons where there are a large number of people who are vulnerable to TB and who engage in behaviours, which also put them at high risk for HIV infection.
One of the limitations of our current study is immediately apparent from the observation of the poor reporting of TB in prisons: in the 5-year study period, there was no available data for 11 WHO European Region countries. In addition, huge fluctuations in the reported annual incident TB rates in prisons reflect uncontrolled epidemics. It is important to note that poor TB reporting affects TB morbidity statistics and, consequently, TB estimates at national and international levels.
This study revealed that some high TB burden countries, such as Turkmenistan and Uzbekistan, have not reported any TB cases among inmates. Cooperation between the institutions responsible for health care in penitentiary systems and the ministries of health should further improve to allow proper TB recording and reporting in both the civilian and penitentiary systems of all countries in the WHO European Region.

Conclusions
This review provides an overview of active TB in prisons in the WHO European Region. The completeness of TB reporting for prisons by NTPs was 79% (42 out of 53 countries from the WHO European Region). Our analysis highlights the vulnerability of inmates to TB and emphasises the necessity of improving TB prevention and care policies and their practical application in prisons with respect to active TB detection, infection control, TB treatment and continuity of care. Most countries achieved a substantial decline of TB burden in prisons, which indicates the effectiveness of recent interventions in correctional institutions. These results provide the basis for an understanding that TB prevention and care in prisons should be elevated to be a health care priority and should facilitate intersectional collaboration between civilian health authorities and prison administrations to enable ending TB in the WHO European Region.

Evidence Available Prior to This Study
TB surveillance data from the WHO European Region were collected annually from countries via The WHO global TB data collection system [14]. The WHO Regional Office for Europe and the European Centre for Disease Prevention and Control (ECDC) have jointly coordinated the collection and analysis of TB surveillance data in the WHO European Region, aiming to ensure data timeline consistency and comparability, pan-European coverage and avoidance of data duplication [1].

Added Value of the Study
This article provides a full cascade analysis of the TB burden and treatment outcomes at the regional level, designated by countries and subregional groupings.

Implications of All Available Evidence
Further efforts should be made to reduce TB infection transmission, the development of active TB and acquisition of data on treatment outcomes either via WHO data collection or reports from individual countries. In particular, more attention needs to be placed on addressing the known risk factors associated with TB.