Healthcare delivery for HIV-positive people with tuberculosis in Europe

Abstract

Background

In a 2013 survey, we reported distinct discrepancies in delivery of tuberculosis (TB) and HIV services in eastern Europe (EE) vs. western Europe (WE).

Objectives

To verify the differences in TB and HIV services in EE vs. WE.

Methods

Twenty-three sites completed a survey in 2018 (EE, 14; WE, nine; 88% response rate). Results were compared across as well as within the two regions. When possible, results were compared with the 2013 survey.

Results

Delivery of healthcare was significantly less integrated in EE: provision of TB and HIV services at one site (36% in EE vs. 89% in WE; P = 0.034), and continued TB follow-up in one location (42% vs. 100%; P = 0.007). Although access to TB diagnostics, standard TB and HIV drugs was generally good, fewer sites in EE reported unlimited access to rifabutin/multi-drug-resistant TB (MDR-TB) drugs, HIV integrase inhibitors and opioid substitution therapy (OST). Compared with 2013, routine usage of GeneXpert was more common in EE in 2018 (54% vs. 92%; P = 0.073), as was access to moxifloxacin (46% vs. 91%; P = 0.033), linezolid (31% vs. 64%; P = 0.217), and bedaquiline (0% vs. 25%; P = 0.217). Integration of TB and HIV services (46% vs. 39%; P = 1.000) and provision of OST to patients with opioid dependency (54% vs. 46%; P = 0.695) remained unchanged.

Conclusion

Delivery of TB and HIV healthcare, including integration of TB and HIV care and access to MDR-TB drugs, still differs between WE and EE, as well as between individual EE sites.

Introduction

In eastern Europe (EE) the HIV epidemic continues to increase, and a leading cause of death is tuberculosis (TB) [1–3]. Incidences of multi-drug-resistant TB (MDR-TB) are high in many countries in EE, with Russia alone accounting for 10% of the world’s MDR-TB cases [2]. We have previously reported on considerably higher mortality rates among HIV/TB-coinfected patients in EE compared with western Europe (WE) and this was related to modifiable factors such as lack of drug susceptibility testing (DST) and suboptimal initial TB treatment [3–5]. As a supplement to these results derived from data on individual patients, we investigated provision of HIV and TB services at a clinical level in a survey from 2013. The main findings were: disintegrated healthcare systems in EE, limited availability of TB drugs for treatment of MDR-TB in EE, low availability of DST for these drugs in EE, and limited availability of opioid substitution therapy (OST) for injecting drug users (IDUs) in EE [6]. In accordance with these findings, the World Health Organization (WHO) highlights the urgent need for increased access to new TB medicines, higher coverage of DST and a healthcare system that makes it easier to access and continue treatment to improve conditions for people with drug-resistant TB, i.e. integrated care [2,7]. Integrated care generally aims to bring together the delivery of diagnosis, treatment and rehabilitation, all in order to improve access to healthcare services and treatment outcome. This covers a broad spectrum, from all services being provided by the same care provider to access to all care services within the same health unit.

Since 2013, use of genotypic resistance testing has been widely recommended and TB drugs such as moxifloxacin, linezolid, clofazimine, bedaquiline and delamanid have come into broader use [8], and we therefore explored whether this has been implemented into clinic-specific guidelines and standard procedures in individual clinics in Europe.

We aimed to compare HIV and TB services provided in clinics in EE and WE and to compare our findings with results from the previous study in 2013.

Methods

This study was a cross-sectional survey conducted in a similar manner to the 2013 study using a slightly modified questionnaire (https://chip.dk/Studies/TBHIV/Documents) [6].

Of the 41 European HIV and TB sites participating in the previously established international ‘TB:HIV study collaboration’ [9] 26 were invited by email in September 2018 to complete a questionnaire. The 26 sites included all collaborating EE sites (except two of five Polish sites) and 10 WE sites acting as a reference point. These particular WE sites were selected to ensure a representative distribution across WE. All participating sites were larger hospital departments specializing in infectious diseases. Follow-up e-mails and phone calls were used to ensure a high response rate. Data collection within the study was terminated in November 2018.

Data were collected through a self-administered questionnaire consisting of 44 questions divided into five categories related to the availability and delivery of HIV and TB healthcare, as well as to clinical management strategies for coinfected patients (background information, integration of HIV and TB services, utilization of HIV healthcare, utilization of TB healthcare and follow-up of TB-HIV coinfected patients).

The responders to the questionnaire were senior consultants in charge of treatment of coinfected patients at the individual site. If the HIV and TB services were separated at two sites, only one questionnaire was completed, and responses coordinated in between.

Descriptive categorical data were obtained from the completed questionnaires. All statistical analyses were divided into 2 categories; (1) Comparison between WE and EE sites (2) Comparison with the 2013 survey.

In order to analyze the overall quality of TB-HIV management within the individual sites further, a ‘TB-HIV care score’ was constructed based on ‘WHO’s rapid communication 2018’ recommendations [10]. The index includes four aspects of care: (1) delivery of healthcare (max. nine points), (2) TB diagnostics (max. four points), (3) TB DST availability (max. three points), and (4) TB drug availability (max. six points), and a combined score for each site was calculated (max. 22 points).

The delivery of healthcare component was based on 13 questions from the questionnaire that covered integration of TB and HIV care, access to OST, payment for care, diagnostic procedures, access to directly observed therapy (DOT) if needed, and procedures in place to avoid loss to follow-up (question 1, 2, 7, 8, 12, 14, 20, 26, 27, 28, 40, 41 and 43).

The TB diagnostics component was based on four questions from the questionnaire that covered standard diagnostic procedures for TB in daily clinical work (question 20, 26, 27, and 28) For further details, please see the footnote to Fig. 2.

Fig 2.

Health Care Index. Delivery of health care; composed of 13 components from the questionnaire (Question 1, 2, 7, 8, 12, 14, 20, 26, 27, 28, 40, 41 and 43). 1 point granted if the following criteria were met: Question 1: HIV and TB services located within one hospital. Question 2: HIV and TB usually treated by the same doctor. Question 7: OST available for all in need. Question 8: All patients diagnosed with TB offered HIV testing. Question 12: No fee for HIV services Question 14: Initiation of ART treatment as soon as possible after TB diagnosis. Question 40: Usage of any form of DOT. Question 41: Patient follow-up at the same hospital for the entire period of TB treatment. Question 43: Procedures in place to prevent loss to follow-up. TB diagnostics; composed of 4 questions from the questionnaire (question20, 26, 27 and 28). 1 point granted if the following criteria were met: Question 20: HIV patients regularly screened for active TB disease. Question 26: NAAT, culture, microscopy followed by NAAT/culture or NAAT followed by culture is standard diagnostic procedure for TB. Question 27: Access to rapid TB diagnostic test. Question 28: DST routinely performed for all positive cultures. DST; composed of 3 components i.e. capability to test for resistance to i) first-line drugs, ii) cycloserine OR terizidone, and iii) at least one injectable AND at least one fluroquinolone. Drug availability; composed of 6 components i.e. unlimited availability to i) first-line drugs, ii) linezolid, iii) clofazimine, iv) bedaquiline, v) cycloserine OR terizidone, vi) and at least one injectable AND at least one fluroquinolone. Missing values; Site no. 2, 8 and 10 had 1 missing value, and site no. 19 had 3 missing values. One Eastern European site excluded from this analysis due to 9 missing values.

The DST component included three parts, i.e. capability to test for resistance to: (1) first-line drugs, (2) cycloserine or terizidone, and (3) at least one injectable and at least one fluoroquinolone.

The drug availability component was composed of six bits, i.e. unlimited availability to (1) first-line drugs, (2) linezolid, (3) clofazimine, (4) bedaquiline, (5) cycloserine OR terizidone, and (6) at least one injectable AND at least one fluoroquinolone.

Two-sided Fisher’s exact test and independent-samples t-test were used to explore associations and calculate P-values. All analyses were performed using SPSS v.24.0. A power calculation showed that if 99% of the WE sites had unlimited access to a particular drug, then less than 49% of the EE sites needed unlimited access in order to demonstrate a significant difference (statistical significance criterion at 0.05).

Results

Of the 26 sites invited to participate, 23 completed the survey (88% overall response rate), and participating sites in EE (n = 14; 87% response rate) were from Belarus (3), Georgia (1), Latvia (1), Lithuania (1), Poland (3), Romania (1) and the Russian Federation (4). Those in WE (n = 9; 90% response rate) were from Belgium (1), Denmark (1), France (1), Italy (1), Spain (2), Switzerland (1), and the UK (2). Three sites in Estonia, Crimea and Italy did not reply.

In total, the EE sites had an estimated 26.816 HIV-positive patients under regular follow-up with 1229 new and recurrent TB cases among HIV-positive patients within the last 12 months. Corresponding numbers in WE were 29.548 HIV-positive patients and 94 new TB cases.

Comparison of delivery of care in Eastern European vs. Western European sites

Results on organization and integration of healthcare services are shown in Fig. 1. In EE, HIV and TB services were less commonly provided at the same site and by the same doctor compared with WE. Most EE sites reported that TB treatment was initiated at special TB hospitals (79%), whereas in WE, TB treatment was generally initiated in the HIV or infectious diseases clinic (88%; P = 0.001). In WE, initial work-up, treatment and follow-up care were provided at the same facility, whereas follow-up care in EE was commonly provided at a separate location (Fig. 1). DOT was used at most sites in EE and WE, although the approach varied so that more sites in EE used DOT for all patients for the entire duration of treatment, whereas in WE, DOT was more commonly used in selected patient groups (P = 0.006).

Fig 1.

Organization and integration of healthcare services in eastern and western Europe. TB, tuberculosis; OST, opioid substitution therapy; DOT, directly observed therapy.

All sites reported using guidelines for HIV/TB-coinfected patients; national guidelines were more commonly used in EE (100% vs. 67%; P = 0.047). In addition, use of international (WHO and/or European AIDS Clinical Society) guidelines was reported by 57% of sites in EE vs. 67% in WE (P = 0.495). A similar and high proportion of sites in both regions reported that all TB patients were offered an HIV test (P = 1.000). In the majority of HIV sites in EE, PLWH was screened for TB once a year by chest X-ray and a clinical algorithm, whereas testing for TB in sites in WE was generally performed on clinical indication (P = 0.081) (Fig. 1). Use of interferon gamma release assay was less common in EE than in WE (57% vs. 100%; P = 0.048), where use of the tuberculin skin test was somewhat more common in EE than in WE (57% vs. 33%; P = 0.400).

Opioid substitution therapy for all coinfected patients with opioid dependency was only available at some sites in EE, but was available at all sites in WE (P = 0.037). When available, there was no significant regional difference in how the OST was provided. In both regions, most sites would refer to an OST department (63% vs. 67%; P = 1.000). There was no significant difference between the two regions regarding standard procedures in place to support adherence or to avoid loss to follow-up (data not shown). However, sites in WE would, to some degree, more frequently contact other facilities to locate patients lost to follow-up (29% vs. 56%; P = 0.383).

Comparison of availability of diagnostics and treatments in EE vs. WE sites

Microscopy and culture of sputum were reported to be standard diagnostic procedures for all sites, and GeneXpert was generally accessible in both regions (86% vs. 100%; P = 0.502). Both regions were reported to routinely perform DST at TB diagnosis (86% vs. 100%; P = 0.502); conventional DST was less commonly used in EE (57% vs. 100%; P = 0.048), whereas use of GeneXpert seemed more common in EE (86% vs. 67%; P = 0.343).

There was no significant difference between the two regions’ capability to perform DST for individual TB drugs, except for pyrazinamide (54% vs. 100%; P = 0.046) (Table 1). For clofazimine (15% vs. 44%), linezolid (23% vs. 56%) and bedaquiline (7% vs. 22%) substantial non-significant differences between regions were reported.

Table 1.

Availability of TB drug susceptibility testing in daily clinical work at Eastern European and Western European sites

Drug	Eastern Europe (n = 14, % of sites)	Western Europe (n = 9, % of sites)	P-value
Isoniazid	92.3	100	1.000
Ethambutol	84.6	100	0.494
Pyrazinamide	53.8	100	0.046
Rifampicin	92.3	100	1.000
Rifabutin	30.8	44.4	0.662
Streptomycin	57.1	66.7	0.495
Injectables (any)	100	100	NA
Fluoroquinolone (any)	84.6	100	0.494
Ethionamide or prothionamide	76.9	77.8	1.000
Cycloserine	61.5	55.6	1.000
Terizidone	14.3	22.2	1.000
Clofazimine	15.4	44.4	0.178
P-aminosalicylic acid	64.3	33.3	0.214
Linezolid	23.1	55.6	0.187
Bedaquiline	7.1	22.2	0.538
Delamanid	7.1	0	1.000

Any injectable; amikacin, kanamycin or capreomycin. Any fluoroquinolone; ciprofloxacin, levofloxacin, moxifloxacin or ofloxacin.

Two months of isoniazid, ethambutol, pyrazinamide and rifampicin followed by 4 months of rifampicin and isoniazid were generally reported as the standard TB treatment regimen for drug-susceptible TB (92% vs. 100%; P = 1.000), whereas use of a specified standard regimen for MDR-TB was more common in EE (64% vs. 13%; P = 0.059). Routine evaluation of culture conversion at the end of TB treatment was significantly more common in EE (93% vs. 22%; P = 0.001).

All EE and WE sites were reported to have unlimited access to isoniazid, ethambutol, pyrazinamide and rifampicin, but availability of rifabutin was low in EE (Table 2). For drugs such as bedaquiline, linezolid, cycloserine, clofazimine and carbapenems, there seemed to be profound, albeit generally non-significant, differences in availability between the two regions (Table 2).

Table 2.

TB drug availability at Eastern European and Western European sites

Drug	Eastern Europe (n = 14, % of sites)	Western Europe (n = 9, % of sites)	P-value
Rifampicin, isoniazid, pyrazinamide and ethambutol	100	100	NA
Rifabutin	23.1	88.9	0.008
Streptomycin	61.5	83.3	0.605
Amikacin	100	88.9	0.409
Kanamycin	69.2	50.0	1.000
Viomycin	0	20.0	0.294
Capreomycin	58.3	50.0	1.000
Ciprofloxacin	75.0	85.7	1.000
Levofloxacin	92.3	75.0	0.531
Moxifloxacin	91.7	88.9	1.000
Ofloxacin	75.0	66.7	1.000
Cycloserine	66.7	87.5	0.603
Terizidone	25.0	20.0	1.000
Clofazimine	27.3	87.5	0.020
Ethionamide	50.0	57.1	1.000
Prothionamide	83.3	57.1	0.305
Linezolid	58.3	88.9	0.178
P-aminosalicylic acid	66.7	71.4	1.000
Amoxicillin	84.6	87.5	1.000
Meropenem	33.3	77.8	0.080
Imipenem	53.8	85.7	0.329
Delamanid	16.7	22.2	1.000
Bedaquiline	23.1	44.4	0.376

Unlimited availability to a particular drug.

^*p<0.05.

Most sites reported starting antiretroviral therapy (ART) as soon as possible after TB diagnosis, irrespective of CD4 cell count (64% vs. 89%; P = 0.513). Sites in EE were more inclined to start a regimen with a nonnucleoside reverse transcriptase inhibitor (almost exclusively efavirenz), whereas integrase strand transfer inhibitors were more popular in WE (P = 0.012). Tenofovir disoproxil fumarate + emtricitabine/lamivudine was the most common nucleo(t)side reverse transcriptase inhibitor backbone in both regions (data not shown).

Sites in EE reported to routinely use co-trimoxazole for all TB/HIV patients more often than sites in WE (46% vs. 11%; P = 0.165).

The TB-HIV care score for the individual sites ranged from 10 to 21 with means ( ± SE) of 19.1 ± 0.7 points in WE and 15.0 ± 0.7 in EE, (P = 0.001) (Fig. 2).

In WE, eight out of nine (88.9%) had a TB-HIV care score of ≥ 17 compared with three out of 13 in EE (23.1%) with some variation in the segments of the score for the individual sites. For delivery of healthcare/integration of care, all sites in WE had at least eight points. Only two EE sites had eight points, and the remaining 11 sites had between three and six points.

One EE site had a score of six due to nine missing values and was excluded from the analysis. Within EE, the score varied from 10 to 19 with considerable heterogeneity within the region; three had a score ≥ 17, six had a score of 14–16, and the remaining four had a score of 10–13.

Comparison of data from the 2013 and 2018 surveys

Twenty-one sites participated in both surveys and were included in these analyses; 13 sites from EE [Georgia, Latvia, Lithuania, Romania (one each), and Belarus, Poland and the Russian Federation (three each)], and eight sites from WE (Belgium, Denmark, France, Italy, Spain, Switzerland (one each) and the United Kingdom (two)].

In EE, there was no improvement in integration of TB and HIV services, and patients were even less likely to be treated for TB and HIV by the same doctor (Fig. 3). Use of OST did not increase between 2013 and 2018. When looking at the component of delivery/integration of care of the TB-HIV care score, only three EE sites had an increase in this sub-score from 2013 to 2018 (data not shown).

Fig 3.

Organization and integration of healthcare services in eastern Europe from 2013 to 2018. TB, tuberculosis; OST, opioid substitution therapy; DOT, directly observed therapy.

Access to GeneXpert improved substantially from 2013 to 2018 (54% vs. 92%; P = 0.073). The standard TB treatment remained unchanged, although there was a tendency in EE to add second-line TB drugs to the initial regimen (0% vs. 18%; P = 0.082). A culture conversion test was used at the end of treatment in all but one site in 2018, compared with nine out of 13 in 2013 (P = 0.322). Development in access to TB drugs is illustrated in Fig. 4. Unlimited access to TB drugs improved significantly for moxifloxacin, but also seemed to improve for rifabutin, levofloxacin, ethionamide, prothionamide, linezolid and bedaquiline, although not significantly so (Fig. 4).

Fig 4.

Development of drug availability in eastern Europe in 2013 and 2018. PAS, Para-aminosalicylsyre.

The only notable change in WE sites was improved access to bedaquiline (0% vs. 50%; P = 0.050).

As for HIV treatment strategies, there was a tendency towards earlier initiation of ART in 2018, defined as ‘as soon as possible and within the first 8 weeks after TB diagnosis, regardless of CD4 cell count’ (46% vs. 62%; P = 0.134).

Discussion

This clinical survey suggests that while considerable progress in areas such as HIV testing of TB patients and access to first-line TB treatment has been made in EE, HIV and TB care remains fragmented and access to MDR-TB treatment highly variable. Delivery of healthcare was considerably more fragmented in EE than in WE, with TB treatment, follow-up, ART, and OST being provided at different sites and by different healthcare professionals. When combining survey variables in a TB-HIV care score, WE sites in general had a higher score than EE sites, and the score varied substantially within EE. Some sites in EE were close to having the same TB-HIV care score as WE sites, whereas other sites were considerably behind. It was not possible to identify a threshold for good clinical practice, but it is worth noting that eight out of nine sites in WE had a score of 17 or more, which may suggest this is an acceptable level for a site. Our results supplement a survey from 2016/2017 by the Wolfheze Working Group [11] that focuses on the programmatic aspects, i.e. policies and guidelines on management of coinfected patients, whereas our study focuses on data on the actual daily practical management, i.e. on the organizational aspects. Both studies found that all countries have guidelines for management of coinfected patients, whereas our survey also documents that usage of such national guidelines was more common in EE than in WE. Furthermore, both studies report on fragmented healthcare delivery in EE. In recent decades, numerous definitions of ‘integrated care’ have been put forward [12]; however, a detailed discussion of these is beyond the scope of this study. In short, general approaches to ‘integrated care’ are: (1) merging of departments; (2) establishing guidelines on specific health issues; and (3) collaboration across services, e.g. between healthcare professionals and social workers [12,13]. A review of 17 systemic reviews on integrated care concluded that healthcare integration potentially improves delivery of healthcare and treatment outcome [14]. Whereas the benefits of integrated services are well described, potential TB transmission among HIV-positive people in integrated TB/HIV clinics is a concern, and relevant measures such as use of masks, attention to hygiene and isolation should be taken to reduce this risk. Still, the most recent WHO guidelines on management of TB/HIV patients highlight the delivery of integrated services as an important tool to reduce the burden of disease [15]. This recommendation is supported by a study from 2007, in which co-location of TB/HIV services was associated with improved medical adherence and clinical outcomes [16]. The lack of improvement in TB-HIV healthcare delivery at most EE sites contrasts these recommendations and calls for action now, requiring profound changes in healthcare infrastructure.

It has been suggested that DOT – which is significantly more commonly used in EE than in WE – can, to some degree, compensate for a fragmented healthcare system regarding medical adherence. A review from 2015 found that DOT improved adherence compared with self-administered treatment. However, this effect was lost as contact to healthcare facilities became more frequent; and considering the cost implications, DOT did not seem to provide a solution to poor adherence to TB treatment [17].

Our results document the fact that access to OST in EE – which is an important component of care for HIV-positive people who inject drugs – has not improved from 2013 to 2018. Additionally, IDU is more common in EE than in WE among coinfected patients. In the TB:HIV Cohort Study, 68% of the participating patients at EE sites were reported to have a history of IDU compared witih 20% at WE sites [5]. The continuous restricted use of OST in EE is problematic for the health of individual patients and for the continuous HIV transmission among IDUs in EE, as numerous studies have shown that OST decreases the spread of HIV and increases adherence to ART [18–22]. OST is still formally illegal in the Russian Federation [23], but even in other countries, usage remains limited, signifying that this is not only an issue of legality. A recent systematic review acknowledges integration of healthcare, client-centred philosophies and attention to stigma as aspects that affect usage of OST [24].

The general TB diagnostic and treatment approach did not differ between sites in EE and WE, and our results suggest that GeneXpert has become widely available across EE. Although this reflects excellent progress, the high prevalence of MDR-TB (and likely high levels of XDR-TB) in EE calls for rapid genotypic DST beyond rifamycins (e.g. line-probe assay for fluoroquinolones and second-line-injectable drugs). Additional measures – such as coordination of MDR-TB care through virtual review in clinics of excellence – may allow rapid test information to be optimally used to design MDR-TB regimens and provide access to such regimens. The observed common practice in EE of adding one or more second-line drugs to the initial treatment – perhaps based on GeneXpert results or clinical suspicion of resistant TB – is of potential concern but cannot be analysed further in the present study.

There were no significant regional differences in access to DST for most TB drugs, perhaps due to the limited number of sites included. Seemingly fewer sites in EE had access to DST for rifabutin, linezolid, clofazimine and bedaquiline.

A similar pattern was observed for unlimited access to individual TB drugs, with lower levels of unlimited availability of rifabutin and clofazimine in EE, contrasting findings in the 2013 survey where EE sites had significantly less access to most drugs with activity against MDR-TB [6]. Unlimited access to DST and TB drugs – including the new WHO-recommended key components of MDR-TB treatment regimens such as levofloxacin/moxifloxacin, bedaquiline, linezolid, clofazimine and cycloserine/terizidone [25] – seems crucial if patient care and treatment outcome in EE are to be improved.

Compared with the 2013 survey, EE sites were more inclined to initiate ART as soon as possible, reflecting that results from large randomized trials have changed clinical practice [26–28]. In EE, efavirenz is still the preferred third component in half of the EE sites, although dolutegravir and raltegravir appear to be attractive first choices for TB/HIV-coinfected people due to favourable toxicity profiles and few drug–drug interactions.

There are some limitations to this study. First, it is based on self-reported data and is therefore sensitive to information biases, including ‘obsequiousness bias’, where responders systematically adjust their responses to fit the expected desire of the investigator. To address this, we visited eight EE sites and interviewed staff that completed the survey. In general, there were only few and minor discrepancies, which seemed to be due to differences in interpretation of a few questions in the survey. It is therefore likely that the survey data generally represent the local practices in the participating sites, although slight overestimation of the surveyed sites’ capability cannot be excluded.

Second, the collaborating EE sites are in general major clinics or referral hospitals and are therefore not necessarily representative of the entire TB/HIV healthcare system, but rather reflect best standard in their respective countries. It is worth noting that the response rate to the survey was high, and results are likely to reflect the situation at all sites in the TB:HIV Study.

Finally, the statistical power of the survey is limited due to few participating sites, as illustrated by relatively large, but insignificant, differences in, for example, access to individual TB drugs. The statistical analyses were not corrected for multiple testing and there is a risk of type 1 error, so the significant results should be interpreted conservatively.

To conclude, this clinic-based survey has demonstrated persistent and significant differences in healthcare management for TB-HIV patients across Europe, but also significant improvements in EE. However, the survey also showed considerable heterogeneity within EE, and in some EE sites care is still far from the level in WE. Whereas GeneXpert and TB drugs with activity against MDR-TB are now more commonly used in EE, there is an urgent need for improvements in integration of TB and HIV services and patient support in order to improve diagnostics, healthcare and treatment, and ultimately treatment outcome for future TB-HIV patients in EE.

Appendix 1

TB:HIV study group

Eastern Europe

Belarus:

Belarusian State Medical University, Department of Infectious Disease: I. Karpov (PI), A. Vassilenko (site coordinator); Republican Research and Practical Clinic for Pulmonology (Minsk): A. Skrahina (PI), D. Klimuk, A. Skrahin, O. Kondratenko and A. Zalutskaya; Gomel State Medical University (Gomel): V. Bondarenko (PI), V. Mitsura, E. Kozorez, O. Tumash. Gomel Region Clinic for Hygiene: O. Suetnov (PI) and D. Paduto.

Estonia:

East Viru Central Hospital (Kohtla-Jarve): V. Iljina (PI) and T. Kummik.

Georgia:

Infectious Diseases, AIDS and Clinical Immunology Research Clinic (Tiblisi): N. Bolokadze (PI), K. Mshvidobadze and N. Lanchava; National Clinic for Tuberculosis and Lung Diseases of Georgia (Tibilisi): L. Goginashvili, L. Mikiashvili and N. Bablishvili.

Latvia:

Infectology Clinic of Latvia (Riga): B. Rozentale (PI), I. Zeltina and I. Janushkevich.

Lithuania:

Clinic for Communicable Diseases and AIDS (Vilnius): I. Caplinskiene (PI), S. Caplinskas, Z. Kancauskiene.

Poland:

Wojewodski Szpital Zakanzy/Medical University of Warsaw (Warszawa): R. Podlasin (PI), A. Wiercinska-Drapalo (PI), M. Thompson and J. Kozlowska; Wojewodski Szpital Specjalistyczny/Medical University Teaching Hospital (Bialystok): A. Grezesczuk (PI); Jozef Strus Multidisciplinary City Hospital (Poznan): M. Bura (PI); Wroclaw University School of Medicine (Wroclaw): B. Knysz (PI) and M. Inglot; Jagiellonian University Medical College (Krakow): A. Garlicki (PI) and J. Loster.

Romania:

Dr Victor Babes Hospital (Bucharest): D. Duiculescu (PI) and S. Tetradov.

Russia:

Botkin Hospital of Infectious Diseases (St. Petersburg): A. Rakhmanova (PI), O. Panteleeva, A. Yakovlev, A. Kozlov, A. Tyukalova and Y. Vlasova; City TB Hospital No. 2 (St Petersburg): A. Panteleev; Clinic for Prevention and Control of AIDS (Veliky, Novgorod): T. Trofimov (PI); Medical University Povoljskiy Federal Region.

Ukraine:

Crimean Republican AIDS Clinic (Simferopol): G. Kyselyova (PI).

Central Northern Europe

Denmark:

Rigshospitalet (Cph): N. Obel (PI) and J. Gerstoft; Hvidovre University Hospital: G. Kronborg.

Belgium:

CHU Saint-Pierre (Brussels): M.C. Payen (PI), K. Kabeya and C. Necsoi.

France:

Aquitaine Cohort: Cohorte administration: F. Dabis (PI) and A. Tsaranazy (public health resident). Participating Clinics and Physicians: Bayonne Hospital; Bordeaux University Hospital: C. Cazanave.

Switzerland:

Swiss HIV Cohort Study (SHCS): Cohorte administration: H. Furrer (PI), M. Sagette (site manager) and M. Rickenbach (head data clinic). Participating Clinics and Physicians: University Hospital Basel: L. Elzi and M. Battegay; University Hospital Bern: H. Furrer; Hopital Cantonal Universitaire, Geneve: D. Sculier and A. Calmy; Clinic Hospitalaire Universitaire Vaudois, Lausanne: M. Cavassini; Hospital of Lugano: A. Bruno and E. Bernasconi; Cantonal Hospital St. Gallen: M. Hoffmann and P. Vernazza; University Hospital Zurich: J. Fehr and Prof. R. Weber.

United Kingdom:

Mortimer Market Clinic (London): R. Miller (PI) and N. Vora; St. Mary’s Hospital: G. Cooke (PI) and S. Mullaney; North Manchester General Hospital: E. Wilkins (PI) and V. George; Sheffield Teaching Hospitals: P. Collini (PI) and D. Dockrell; King’s College Hospital (London): F. Post (PI), L. Campbell, R. Brum, E. Mabonga and P. Saigal. Queen Elizabeth Hospital: S. Kegg (PI); North Middelsex University Hospital: J. Ainsworth (PI) and A. Waters. Leicester Royal Infirmary: J. Dhar (PI) and K. Ellis.

Southern Europe

Italy:

IRCCS – Ospedale L. Spallanzani (Rome): E. Girardi (PI), A. Rianda, V. Galati, C. Pinnetti and C. Tommasi; A.O. San Gerardo (Monza): G. Lapadula (PI); IRCCS AOU San Martino – IST di Genoa (Genova): A. Di Biagio (PI) and A. Parisini; Clinic of Infectious Diseases, University of Bari (Bari): S. Carbonara (PI), G. Angarano and M. Purgatorio; University of Brescia Spedali Civili: A. Matteelli (PI) and A. Apostoli.

Spain:

Barcelona Cohort funded by the Spanish HIV/AIDS Research Network: Hospital Clinic of Barcelona: J.M. Miro (PI), C. Manzardo, C. Ligero and J. Gonzalez and Jose A. Martinez-Martinez; Hospital del Mar: F. Sanchez, H. Knobel, M. Salvadó and J.L. Lopez-Colomes; Mutua de Terrassa: X. Martínez-Lacasa and E. Cuchí; Hospital Universitari Vall d’Hebrón: V. Falcó, A. Curran, M.T. Tortola, I. Ocaña and R. Vidal; Hospital Universitari de la Santa Creu i Sant Pau: M.A. Sambeat, V. Pomar and P. Coll; Hospital Universitari de Bellvitge: D. Pozamczer, M. Saumoy and F. Alcaide; Agenda de Salud Pública de Barcelona: J. Caylà, A. Moreno, J.P. Millet, A. Orcau, F. Fina, A. Romero, L.L. Roldan. Hospital Universitaria Donostia (San Sebastian): J.A. Iribarren (PI) and M. Ibarguren; Hospital Ramon y Cajal (Madrid): S. Moreno (PI) and A. González; Hospital Universitaria ‘Gregorio Marañon’ (Madrid): P. Miralles (PI) and T. Aldámiz-Echevarría.

Latin America

Argentina:

The CICAL Cohort: Cohorte administration: M. Losso (PI), J. Toibaro and L. Gambardella. Participating Clinics and Physicians: Hospital J. M. Ramos Mejía (Buenos Aires): J. Toibaro and L. Moreno Macias; Hospital Paroissien (BA): E. Warley (PI) and S. Tavella; Hospital Piñero (BA): O. Garcia Messina and O. Gear; Hospital Nacional Profesor Alejandro Posadas: H. Laplume; Hospital Rawson (Cordoba): C. Marson (PI); Hospital San Juan de Dios (La Plata): J. Contarelia and M. Michaan; Hospital General de Agudos Donación F. Santojani: P. Scapellato; Hospital Francisco Javier Muñiz (BA): B. Bartoletti and D. Palmero; Hospital Jujuy: C. Elias.

Chile:

Fundación Arriaran (Santiago): C. Cortes.

México:

INNcMZS (México DF): B. Crabtree (PI); Hospital General Regional de Leon- CAPACITS: J.L. Mosqueda Gomez; Hospital Civil de Guadalajara: J. A. Villanueva (PI); L.A. Gonzalez Hernandez and F. Badial.