EUROPEAN JOURNAL OF PUBLIC HEALTH 2002; 12: ANDREI SLAVUCKIJ, VINCIANE SIZAIRE, LAURA LOBERA, FRANCINE MATTHYS, MICHAEL E.

Transcription

1 EUROPEAN JOURNAL OF PUBLIC HEALTH 2002; 12: Decentralization of the DOTS programme within a Russian penitentiary system How to ensure the continuity of tuberculosis treatment in pre-trial detention centres ANDREI SLAVUCKIJ, VINCIANE SIZAIRE, LAURA LOBERA, FRANCINE MATTHYS, MICHAEL E. KIMERLING * 94 Background: In Kemerovo region (Siberia), three pre-trial detention centres (SIZO; Ministry of Justice) serve as the gateway to the penitentiary system, comprised of 23 prisons and 30,000 detainees. The follow-up for tuberculosis (TB) patients released into civil society is unreliable. Due to varying detention times and frequent transfers to temporary detention centres (IVS; Ministry of Internal Affairs) for investigation and trial, and concerns about continuity of treatment, SIZOs were not included in the revised TB control programme initiated during Methods: To investigate the feasibility of DOTS (Directly Observed Therapy, Short-Course) expansion into SIZOs, general detainee release was studied by examining 10% of files from detainees admitted during 1998 (SIZOs 1,2,3). Then, 5% of general files from SIZO 1 were examined to determine SIZO IVS flow; 224 TB patient files from SIZO 3 were evaluated to determine the pattern of release/transfer. Results: TB patients in SIZO 3 have less chance of release before six months of detention than non-tb detainees (14/224, 6.3% versus 774/2276, 34%; p<0.001). Among detainees not released, 60% are not moved during the first six months of detention. For those who move, the mean stay in IVS was 9.5 (+/ 6) days. The incidence of active disease detected upon entry to SIZO 3 was 4,560/100,000, the subsequent rate during the same year of detention 880/100,000. Conclusion: Despite frequent detainee movements between institutions, DOTS should be introduced into the earliest stages of detention to prevent case mismanagement, and links to the civilian programme should be developed. Colony 33 is a referral colony for tuberculosis (TB) patients sent from 26 penitentiary institutions (23 general colonies and three SIZOs) in the Kemerovo Region of Siberia, Russia. It meets only a fraction of the needs required for effective tuberculosis control within an expansive penal system. Approximately 90% of the Colony 33 prison population comes from the region itself. Doctors Without Borders (Médecin Sans Frontières, MSF) began assisting penitentiary TB control efforts in 1996 through the introduction of a strict DOTS (Directly Observed Therapy, Short Course) programme according to international standards. 1 They found an existing system of control whereby case finding is based largely on fluorographic screening. All inmates, as policy, are screened upon entry then once every six months; those suspected of having active TB are collected in the * A. Slavuckij 1,2, V. Sizaire 3, L. Lobera 4, F. Matthys 2, M.E. Kimerling 1,5 1 Department of Epidemiology and International Health, School of Public Health, The University of Alabama at Birmingham, USA 2 Médecins Sans Frontières (MSF), Brussels, Belgium 3 Médecins Sans Frontières, Moscow 4 Médecins Sans Frontières, Mariinsk, Kemerovo (Siberia) 5 Department of General Internal Medicine, School of Medicine and Advisor MSF Kemerovo tuberculosis project Correspondence: Dr. Michael E. Kimerling, MD, MPH, UAB, RPHB 217, rd Avenue South Birmingham, Alabama , USA, tel , fax , kimerlin@uab.edu Keywords: DOTS, prisons, Russia, Siberia, tuberculosis different colonies then sent to Colony 33 for definitive diagnosis and treatment. Without an organizational structure to control the flow of suspects into Colony 33, initially the proportion of sputum smear positive (showing acid fast bacilli, AFB) patients was as low as 20%. The waiting list of newly screened suspects in the general colonies (according to Russian definitions) also increased rapidly. Many of these patients were treated with available but inadequate therapy, much of it from family members, thus contributing to development of resistance to first-line anti- TB drugs. Early programme data show the level of MDR- TB among those referred to Colony 33 at 23%. 2 Recognizing the care access barriers and treatment delays inherent in such a referral structure, MSF in collaboration with its counterparts undertook steps to decentralize the DOTS programme within the penitentiary system. The revised TB programme was expanded to a second TB referral centre, Colony 16, for prisoners under stricter terms of detention. Another general colony was developed as a dispensary for treated and/or cured cases (with a laboratory infrastructure to detect relapses by smear microscopy), creating a mechanism for earlier patient transfer from Colony 33, and thereby opening treatment beds for new cases. Tuberculosis therapy was also started in the only woman s colony, and a mobile

2 DOTS decentralization in Russian prisons laboratory screening team was organized to identify sputum AFB positive cases within the general colonies for rapid referral to the two TB colonies. Finally, efforts were made to educate regional authorities to gain support for the elimination of erratic, non-standardized TB treatment in all places of detention. The subsequent situation improved dramatically: the patient waiting list was reduced, delays to treatment initiation were reduced, the cure rate improved (cohort IX since the introduction of DOTS, April June 1998, cure rate based on smear conversion was 73%), the proportion of AFB positive cases accepted into the programme increased to nearly 60% of newly detected cases. However, the entrance gates into the penitentiary system, the three pre-detention centres or SIZOs, considered important breeding sites for TB, were not incorporated into the programme due to perceived difficulties in treatment continuation. During a detainee s stay in SIZO, he or she may be moved to IVS (jails) for undetermined periods of time. Depending upon the results of the investigation and trial, detainees may be convicted, sentenced and sent to the general colonies or released back to their communities (figure 1). The issue of treatment continuity for TB cases identified while in SIZO/IVS and upon release has never been studied within the context of establishing a DOTS programme. The major concern is whether DOTS within SIZO is possible without the augmentation of defaulter rates or worsening resistance rates to first-line drugs through subsequent erratic or unregulated treatment. We therefore evaluated detainee movements into and out of SIZO with special attention to the SIZO-IVS flow and the transfer/release of tuberculosis patients identified. COMPONENTS OF THE RUSSIAN PENAL SYSTEM: DETENTION, RELEASE AND CONVICTION IVS Isolator for Temporary Detention (jails) IVS is part of the structure of the Ministry of Internal Affairs (MVD). Detainees arrested at the crime location or suspected of committing a crime normally spend up to three days here. This period may be prolonged to ten days (exceptionally 30 days). From IVS, detainees may be released (due to a lack of evidence or the work of a skilled lawyer), or upon the sanction of the prosecutor, transferred to SIZO (see below). Detainees sent to SIZO may be returned to IVS for completion of the criminal investigation or for trial, normally lasting a few days to one week. One place where a detainee may be sent for a longer period of time is the psychiatric hospital for psychiatric or drug-abuse evaluation. Although there is a medical worker (felcher) in larger IVS, there is none in the smaller ones. SIZO Isolator for Investigation (pre-detention centre) SIZOs are part of the structure of the Ministry of Justice. These institutions are for detainees for whom the court has not yet rendered a verdict or who are waiting for completion of the criminal investigation. Detention in SIZO varies from several weeks to two and a half (or Before conviction: Detention Trial and Conviction IVS a SIZO b Not TB case, or not diagnosed in SIZO After conviction: Known TB case, or General diagnosed in SIZO Colonies TB Colony 16 TB Colony 33 a IVS Isolator for Temporary Detention (Ministry of Internal Affairs) b SIZO Pretrial Detention Center (Ministry of Justice) Figure 1 Detainee flow into, within and out of the Kemerovo penal system (adapted from Abramkin 3 ) more) years. The level of the public prosecutor s office determines the time period allowed for criminal investigation: two months, district prosecutor; three months, city or regional prosecutor; six months, prosecutors subject to the Russian Federation; two and a half years, general prosecutor from Moscow. 3 Legal practice is not closely adhered to as detainees may spend six months or more in SIZO with the local prosecutor s agreement. This often occurs due to the slow legal proceedings, the punitive tendency of the legal system, a lack of alternative to detention such as a bail system, inadequate means of communication and the bureaucratic methods of investigation. Taken together, the situation results in overcrowded SIZOs. Investigation outcomes are: trial and release without conviction (from IVS or SIZO); trial with conviction, but release as the time spent in SIZO is counted; trial and conviction with a longer term to be spent in a general colony. ORGANIZATION AND LINKS BETWEEN SIZO AND IVS IN KEMEROVO REGION There are three SIZOs in Kemerovo Oblast for detainees, covering different territories: SIZO 1 in the central part, SIZO 2 in the southern part and SIZO 3 in the northern part. Each SIZO covers several IVS. SIZO 1 covers six IVS, SIZO 2 covers five and SIZO 3 covers 14. Upon detention, persons are sent to IVS then to SIZO, according to the place where the crime occurred. If the person is suspected of committing a crime in one place, but arrested in another place, the person is sent to the nearest IVS before transfer to the IVS or SIZO where the crime was committed. Normally, the time spent away from SIZO depends on the distance to the respective IVS, severity of the committed crime, availability of transport and the time to organize trains with special coaches (for detainees) or special buses (for detainees). A convoy department within GUIN (Federal Penitentiary Authority) is responsible for transfers of detainees between SIZO and IVS. 95

3 EUROPEAN JOURNAL OF PUBLIC HEALTH VOL NO. 2 TUBERCULOSIS SCREENING AND MANAGEMENT IN SIZO Upon admission to SIZO, all detainees undergo fluorographic screening for tuberculosis. Because of overcrowding and lack of space, diagnosed TB patients are normally mixed among the other detainees. If drugs are available, the medical unit will start treatment. For patients with a known TB history, there may be a notice from the civil tuberculosis dispensary giving a diagnosis but without detailed information. In general, coordination between civil TB services and the penitentiary system is non-existent. When a TB patient is transferred from SIZO to either IVS, a psychiatric hospital or a TB colony (upon conviction), a referral medical note is included in the accompanying file. Families of detainees are encouraged to bring anti-tb drugs, and patients are not separated according to their sputum smear AFB status, either in SIZO or during transport to IVS. Recently, a regional prikaz (order) was issued to create separate detention cells for TB suspects and patients during their treatment in SIZO. Further, a position of TB specialist was created for each SIZO and a laboratory for smear examination should be organized. Finally, identified TB patients may no longer be transferred from SIZO to another prison during the first two months of detention, thus facilitating treatment without interruption. METHODS The roles of the various functional units within the Russian penitentiary system were studied. Specifically, the interactions between SIZO-IVS, SIZO-general colony and SIZO-release to civil society in Kemerovo Region were evaluated. The reference population studied was selected from detainees admitted to SIZOs during Ten per cent of all files from detainees admitted during 1998 were randomly selected (every tenth name from the detainee registries) for all three Kemerovo SIZOs, and the percentage of the general SIZO population that spent less than six months before release was determined (to correlate with the duration of standard TB therapy). For SIZO 3 (Mariinsk), the movements and releases of all tuberculosis patients identified amongst the detainees admitted during 1998 were also evaluated. For SIZO 1 (Kemerovo City), 5% (N=205) of detainee files were randomly selected (from among all detainees entering during 1998) to estimate the average time that detainees spent in each IVS (N=6) associated with this particular SIZO. All TB statistics are based on Russian TB case finding methods. 4,5 The Russian definition of active TB is: clinical symptoms and fluorographic evidence on chest x-ray (CXR) among new cases (never treated before), and persistent symptoms and CXR changes among old cases (previously registered as TB and treated). An inactive case is someone previously registered as TB and treated, and without symptoms, but still with residual CXR abnormalities. Quantitative data were analysed using EpiInfo (CDC, public domain); all odds ratios (OR) are reported with 95% confidence intervals (CI). RESULTS SIZO 3 (Mariinsk): diagnosis, movement and release of TB patients during 1998 During 1998, 2,500 detainees entered SIZO 3. In total, 224 TB patients were identified upon admission or during their stay, including active symptomatic cases and inactive cases without symptoms. Among these, 159 detainees (71%) were transferred out of SIZO 3: 141 to Colony 33 and four to Colony 16 for treatment; 14 (6.3 %) were released to home. Of the 136 detainees with active disease, 114 (84%) were diagnosed on admission, yielding a SIZO entry case detection rate of 4,560/100,000; 22 cases were subsequently diagnosed during the same year of incarceration, 1998 (880/100,000). The majority of TB cases (135/224=60%) were transferred to a TB colony after conviction; six patients (3%) were sent before conviction due to the severity of their disease state. Admissions to all SIZOs during 1998: judicial outcomes and detainee release A sample of total admissions to all three SIZOs during 1998 was studied (table 1); the proportion of detainees released before six months was determined according to judicial outcome (released or sentenced). Among the general population of SIZO 3, the percentage of detainees Table 1 Detainee admissions to all Kemerovo SIZOs in 1998 according to judicial outcome and time spent in SIZO Total admissions Detainees sampled <6 months released a <6 months sentenced b >6 months in SIZO c Unknown SIZO N N n % n % n % n % I 4, II 4, III 2, Total 11, a: Potential dropouts from the DOTS programme would likely come from the group of TB patients among detainees released before 6 months of detention. b: It is assumed that for TB patients sentenced before 6 months of detention in SIZO, once detected, patients will be transferred to Colony 33 (or Colony 16) and thus will not have their treatment course interrupted. c: Those TB patients who spend more than 6 months in SIZO and start their treatment on DOTS have more chance of completing this treatment while still in SIZO than those who are released before 6 months of detention (due to poor penal-civil sector programme linkage).

4 DOTS decentralization in Russian prisons released before six months was 32%, excluding TB patients. Among the TB patients, as noted above, the percentage released was considerably lower, 6.3% (14/224). TB patients therefore had less chance of release before six months of detention from SIZO 3 than did non-tb detainees (OR=0.13, 95% CI: ). For SIZO 1, the percentage of detainees released among the general population before six months was 28%, despite twice the number of admissions as in SIZO 3. In SIZO 2, the percentage of detainees released before six months of detention from the general population was extremely low (1.3%); the explanation for this difference is not known. Detainee flow between SIZO 1 and IVS in Kemerovo City Approximately 5% of all admissions to SIZO 1 during 1998 (205 of 4,596) were studied to understand the releases and movements of detainees between SIZO and IVS. There are six IVS to which detainees can be transported for investigation and a psychiatric hospital. Some 82% of the 205 detainees studied were moved to either of the two largest IVS. In total, 61 of 205 detainees were released (30%). Among these, 58 were released before six months of detention (28% of total sample); 29 were released before two months of detention. The average stay of those released between two and six months was 3.6 months (108 days +/ 29). Excluding those who were released (N=61) or died in SIZO (N=3), the number of movements during the first six months of detention was enumerated (N=141). The majority of detainees were not moved (N=84, 60%). Among those who did move during the first six months (N=57), 39 were moved during the first two months at least one time (39/141=27%); nine were moved more than once. The average stay in IVS for those moved only once during the first two months (N=30) was 12 days (+/ 12), including the psychiatric hospital where the average stay is longer. Excluding persons going to the psychiatric hospital, the average IVS stay was 9.5 days (+/ 6). DISCUSSION Tuberculosis is a critical health issue in the Kemerovo pre-detention centres (SIZOs) with an overall active TB case detection rate of 5,440/100,000 during a single year. Surprisingly, the vast majority of cases was found upon entry screening and indicates that tuberculosis has reached epidemic levels at the earliest stage of incarceration. Further, this finding reflects a parallel epidemic in the civilian sector, at least among those groups at-risk for incarceration, and is similar to findings in other settings. 6 9 The major problem in SIZO does not concern those TB patients who will be released before six months of detention. As was found, this group comprised less than 10% of all TB patients in SIZO 3 during The major issue is the high rate of disease found at entry and during the year of detention. One may hypothesize that those detainees most likely to enter with tuberculosis are repeat offenders incarcerated for a second, third or additional time and who therefore may have had TB exposure during a previous internment. It has been shown elsewhere that both the number of jail admissions and total jail time were associated with development of tuberculosis in prison, although the HIV context certainly played a major role in the New York City system. 10 Due to recidivism, detainees may also be less likely to obtain early release or resolution of the charges against them, again increasing their exposure risk and likelihood of developing active disease. Unfortunately, it is not known how many of these cases had a previous TB diagnosis or was previously incarcerated, so this hypothesis could not be tested. For detainees diagnosed with active tuberculosis and sentenced to a prison term, referral to either of two treatment colonies is available. To adequately address disease control in prisons, however, one must incorporate the entire penitentiary system. A potential barrier thus arises from the complicated interrelation between SIZO and IVS. In Kemerovo, 25 IVS are associated with three SIZOs. For a DOTS programme within SIZOs to be effective, it must be initiated in each one. As Colony 33 receives patients from the three SIZOs after months of erratic treatment, starting DOTS in only one SIZO would not bring about sufficient change to stop the additional creation of drug-resistant strains in the others. Organizing continuity of treatment for patients who move between multiple locations is challenging. In Tomsk Oblast, where DOTS was introduced to the civil sector in 1995 by the British relief organization, Medical Emergency Relief International (MERLIN), a telegram is sent to the local civilian TB clinic and a nurse goes to IVS to continue treatment. In Kemerovo Region, where DOTS has not yet been introduced into the civil sector, a similar strategy would require linking 25 IVS to numerous civil TB clinics. In Kemerovo, a detainee may be moved from SIZO to IVS for a period of one to twelve days. For longer periods, a detainee can stay only in the main psychiatric hospital. Since the latter is a medical institution, the personnel could be taught DOTS therapy along with standard monitoring procedures. Further, medical workers are available in all SIZOs and in the largest IVS, covering the vast majority of detainees. Thus, DOTS decentralization to the SIZO level is possible with careful planning. A local policy concerning the interdiction of movement for TB cases in the first two months of therapy was issued in 1999; investigators should now come to each SIZO to perform their case investigations. This arrangement will further serve to address the major concern of TB transmission to non-tb detainees within the SIZO-IVS network and also to ensure proper therapy. CONCLUSION Pre-trial detention centres are highly unstable settings for disease control, involving a complicated system of detainee flow. Due to inherent delays associated with the legal process, individual detainee length of stay is 97

5 EUROPEAN JOURNAL OF PUBLIC HEALTH VOL NO. 2 uncertain. In our study sample, TB patients have less chance of release before six months of detention than the general, non-tb detainees, which was the established pattern before the recently issued policy. Based on our findings, approximately 10% of diagnosed TB patients are expected to be released back to their communities before a standard six-month DOTS treatment course could be completed. As 82% of detainees studied in SIZO 1 were moved to either of the two largest IVS, where medical workers are present, continuation of DOTS therapy is feasible. Further, as the average length of stay in IVS is less than two weeks, unforeseen interruption of treatment should not lead to the development of resistance pending the resumption of adequate therapy (allowing that no medical staff are available in IVS for administering TB treatment). Therefore, the introduction of DOTS into SIZO should prove feasible, thereby limiting further creation of drug resistance, including MDR-TB, and addressing the prison-seated epidemic at the earliest stage of detention. However, this can be considered as only one step within a development framework for a broader regional control programme that links the civil and penal sectors, ultimately providing for uninterrupted care. Another conclusion of our investigation is that entry screening in SIZOs is essential. Although widely recommended and presumably done, the best screening method and required frequency remain undetermined. Such a high entry rate of active tuberculosis, as found in our analysis, again highlights the need for a comprehensive civil-prison strategy, including SIZO and IVS. While tuberculosis in prisons certainly impacts disease in the community, it is increasingly apparent that an uncontrolled civilian epidemic also threatens the ability to control the prison-seated one. Sources of support: USAID. Support for this work was also provided by the Global Tuberculosis Initiative/USAID project of the Gorgas Memorial Institute and in part by the Edmund S. Muskie and Freedom Support Act Graduate Fellowship Program (USIA). The authors would like to thank all the staff of Colony 33 and the Kemerovo SIZOs for their cooperation in making this study possible. REFERENCES 1 World Health Organization. Treatment of tuberculosis: guidelines for national programmes. Geneva: WHO, Kimerling ME, Kluge H, Vezhnina N, et al. Inadequacy of the current WHO re-treatment regimen in a central Siberian prison: treatment failure and MDR-TB. Int J Tuberculosis Lung Dis 1999;3(5): Abramkin V. Man and prison. Vol. 1. Moscow: Ford Foundation, Russian Ministry of Public Health: Order Number 33, Drobniewski F, Tayler E, Ignatenko N, et al. Tuberculosis in Siberia: 2. Diagnosis, chemotherapy and treatment. Tubercle Lung Dis 1996;77(4): Martin V, Gonzalez JA, Cayla J, et al. Case-finding of pulmonary tuberculosis on admission to a penitentiary center. Tubercle Lung Disease 1994;74: Puisis M, Feinglass J, Lidow E, et al. Radiographic screening for tuberculosis. Public Health Rep 1996;3: Layton M, Henning K, Alexander T, et al. Universal radiographic screening for tuberculosis among inmates upon admission to jail. Am J Public Health 1997;87(8): Bock N, Reeves M, La Marre M, et al. Tuberculosis case detection in a state prison system. Public Health Rep 1998;113: Bellin E, Fletcher D, Safyer S. Association of tuberculosis infection with increased time in or admission to the New York City jail system. JAMA 1993;269(17): Received 16 January 2001, accepted 15 June