TUBERCULOSIS IN AUSTRIA

TUBERCULOSIS IN AUSTRIA TB Data Analysis Report 1997-2006

TUBERCULOSIS IN AUSTRIA REPORT ON DATA ANALYSIS 1997-2006 Editors Hung-Wei Kuo, MSc 1,2,3 Dr. med Daniela Schmid, MSc 1 Co-Authors Sabine Pfeiffer 4 Mag. Dr. Alexander Indra 4 Univ.Prof.Dr. Franz Allerberger 1 (1) Kompetenzzentrum Infektionsepidemiologie, Austrian Agency for Health and Food Safety (AGES), Vienna, Austria (2) European Program for Intervention Epidemiology Training (EPIET) (3) Centers for Disease Control, R.O.C. (Taiwan) (4) Nationale Referenzzentrale für Tuberkulose, Austrian Agency for Health and Food Safety (AGES), Vienna, Austria 2009

INDEX Glossary 3 Summary 6 1. Introduction 7 2. Procedure of checking the crude data set and making the data set eligible for the analyses 13 2.1. Process for ensuring that each record presents a new case of TB 15 2.2. Creating further variables for the analyses 20 2.3. Handling missing data 21 2.4. Handling data discrepancies 25 3. Objectives 27 4. Materials and Method 30 4.1. Data sources 30 4.2. Study designs 31 4.3. Statistical analysis 31 5. Results 33 5.1. Primary Objective 33 Burden of TB in Austria, 1997-2006 5.2. Secondary Objective 92 Influence of immigration to Austria on the epidemiology of TB in Austria, 1997-2006 6. Interpretation 115 6.1. Interpretation of the findings on the research question I: Burden of TB disease in Austria 115 6.2. Interpretation of the findings on the research question II: Influence of immigration to Austria on the epidemiology of TB in Austria 119 7. Conclusions and Recommendations 140 8. Annex of Tables 143 9. References 153 Glossary 3

GLOSSARY Foreign born person Austrian inhabitant Austrian= Austrian citizen Non-Austrian Since data on country of birth for TB cases in nonnative residents of Austria are not available in the Austrian TB-surveillance database, the term TB case in foreign born persons is not used. Data about citizenship or data about country of origin (i.e. emigration country when there is no citizenship status) are available, and have been used in the study. Anyone residing in Austria. The term is synonymous with Austrian population or residents of Austria. A person with Austrian citizenship; the term citizenship is synonymous with nationality. A person residing in Austria who is a citizen of another country and does not hold Austrian citizenship. The National Statistical Institute of the Republic Austria Statistic Austria 1 provides data on the specific citizenship of non-austrian residents of Austria coming from the former Yugoslavia and Turkey. There are no available data about the specific citizenship or emigration country of non-austrian residents who are neither citizens of the former Yugoslavia nor of Turkey. These residents of Austria are grouped as non- Austrians of other citizenship in the Austrian population statistics. Statistic Austria also provided data on the immigrants (in German Zuwanderer) by country of origin per year for the years 1997-2006 The Austrian Ministry of Internal Affairs provides data on the specific citizenship (synonymous with nationality in the text) for the asylum seekers in Austria per year for 1997-2006 1 Bundesanstalt Statistik Österreich (STAT): National Statistical Institute of the Republic Austria; Statistik Austria ging durch das österreichische Bundesstatistikgesetz 2000 aus dem Österreichischen Statistischen Zentralamt hervor. Summary 4

TB case in a non- Austrian TB case in an Austrian TB case in short-stay non-austrians TB case in long-stay non-austrians Immigrant to Austria (i.e. entrants, migrant) Refugees Asylum seeker The term is synonymous with non-austrian TB case The term is synonymous with Austrian TB case < 5 years stay in Austria when becoming a case of TB > 5 years in Austria when becoming a case of TB In this document this refers to any person entering Austria with the purpose of establishing residence in Austria. A person who is outside his or her country of nationality or habitual residence; has a well-founded fear of persecution because of his or her race, religion, nationality, membership of a particular social group or political opinion; and is unable or unwilling to avail himself or herself of the protection of that country, or to return there, for fear of persecution according to 1951 refugees convention Article 1 by the UNHCR (Media Relations and Public Information Service P.O. Box 2500; 1211 Geneva 2). Refugees are fleeing persecution or war rather than moving for financial or personal reasons. People become refugees, either on an individual basis or as part of a mass exodus, because of political, religious, military and other problems in their home country. Austria is a member state party to the 1951 Convention and the 1967 Protocol relating to the Status of Refugees. The Convention does not provide automatic protection for all time. Many refugees have integrated permanently into their country of asylum, but some cease to be refugees when the basis for their original asylum claim ceases to exist. The national TB surveillance database provides data on the residence status of TB cases in non-austrians using the following terms: foreigner, migrant labour, refugee (see below) and asylum seeker (see below). The terms migrant labour and foreigner are no longer officially used by the Austrian Ministry of Internal Affairs, which deals with immigration. Summary 5

An asylum seeker is someone who has fled his or her own country and applies to the government of another country for protection as a refugee. The term asylum seekers refers to all people who apply for refugee protection, whether or not they are officially determined to be refugees. Under the 1951 Convention on Refugees, an asylum applicant must be able to demonstrate a well-founded fear of persecution in his or her country of origin for reasons of political opinion, religion, ethnicity, race and nationality, or membership of a particular social group. The applicant must also be able to demonstrate that they are unable to obtain any protection or assistance from their own national authorities. The essential parts of this Convention are reiterated in, and updated by, The Refugee or Person in Need of International Protection (Qualification) Regulations 2006. Refugee and Asylum seeker Approved asylum seeker=asylum holder High income country High TB incidence country 50 / 100,000 TB incidence countries Former Soviet Union and now European countries by WHO Drug susceptibility testing (DST) For the purpose of the analyses, refugees and asylum seekers among the non-austrian TB-cases were merged into one group and named refugees. A person who is recognized as having a right of asylum. A high-income country is defined by the World Bank as a country with a Gross National Income per capita of $11,456 or more. In this report, we defined a high TB incidence country as a country with 20 cases per 100,000 persons per year. In comparison, a low TB incidence country is defined as a country with <20 cases per 100,000 per year. Within the group of high incidence TB countries described in this report, we defined a sub-group of very high TB incidence countries, as having 50 cases per 100,000 persons per year. Georgia, Armenia, Moldavia, Lithuania, Latvia, Estonia, Kazakhstan, Uzbekistan, Kyrgyzstan, Ukraine, Azerbaijan. Antimicrobial susceptibility testing. Summary 6

SUMMARY Since many immigrants come from countries with a TB-incidence greater than that of the high-income reception countries, and since the TB incidence in the native population of the reception countries has decreased over the past decade, most reception countries, and in particular, the European countries, have observed within the past decade a substantial increase in the proportion of foreign born population with TB. The WHO Regional office for Europe stated in 2007 that foreign born populations settled in Europe have an increased probability of developing TB due to a higher risk of progression of latent TB infection contracted in their country of origin. According to our findings, from 1997-2006 a positive correlation was found between immigration from high TB incidence countries to Austria and the epidemiology of TB among non-austrians. This is most likely explained by the import of cases of latent TB among immigrants from high TB incidence countries. In the years 1999, 2000, 2004 and 2005 the epidemiology of TB among non-austrians was largely determined by the high proportion of refugees among immigrants from high TB incidence countries, who developed TB shortly after entry into Austria. The impact of immigrants, who developed TB disease within 5 years of their arrival in Austria, on the incidence of TB among long-stay non-austrian residents (length of residence in Austria 5 years), was minor. No correlation was found between immigration and the epidemiology of TB in Austrians between 1997 and 2006. Immigrants from high TB incidence countries represent a source of the re-emergence of TB in a low prevalence country such as Austria. Almost all high-income countries perform chest radiography screening for the detection of active TB among applicants for permanent residence. In Austria, refugee camp residents undergo TB screening by chest x-ray shortly after arrival. Measures for the detection of cases of latent TB are not performed. Chest radiography screenings for active TB among immigrants at entry has minimal impact on the TB burden and is not cost-effective. The most beneficial screening measure for TB control is screening upon entry for latent TB infection, followed by a preventive chemotherapy. We recommend targeted screening for latent TB in high TB risk groups among immigrants to Austria. Summary 7

1. Introduction Main Features of the Disease Tuberculosis (TB) is a disease primarily caused by M. tuberculosis, a mycobacterium, for which humans are the main reservoir. The disease is an important cause of disability and death in many countries of the world, especially in developing countries. Highincome countries showed downward trends in mortality and morbidity of TB for many years. The lifetime risk of progression from tuberculosis infection to TB disease in the general population is approximately 10%; half of the progressors will develop active disease within the first 2 years following infection [1]. An anti-tuberculosis-treatment in cases of latent tuberculosis infection (LTBI), i.e. preventive chemotherapy, considerably reduces the lifetime risk of TB disease [1]. TB disease may affect any organ or tissue. Extrapulmonary TB occurs less commonly than pulmonary TB (30% vs. 70%). Children and persons with immunodeficiency, such as AIDS, have a higher risk of extrapulmonary TB [1]. Recurrent TB disease can be due to endogenous reactivation of a latent focus or due to reinfection after treatment completion. The 5-year mortality in untreated cases of pulmonary tuberculosis with sputum smear positive for acid-fast bacilli is up to 65% [1]. The main mode of transmission of tuberculosis is the inhalation of tubercle bacilli in airborne droplet nuclei (aerosols), 1 to 5 microns (µm) in diameter, produced by people with respiratory tract tuberculosis (pulmonary TB, bronchial TB, laryngeal TB) during expiratory efforts (coughing, sneezing, singing) [1]. The period of infectiousness is as long as viable tubercle bacilli are excreted through respiratory secretions. The risk of infection with tubercle bacilli is directly related to the degree of infectiousness of the infected person given by the amount of bacteria excretion through respiratory secretions, to the intensity and duration of the infectious contact, the degree of susceptibility of the exposed person and to the virulence of the bacterial strain. The first 24 months after infection constitute the most risky period for developing active tuberculosis. The risk of TB disease is highest in children under 4 years, lowest in later childhood, high among young adults, very old persons and in immunosuppressed persons such as AIDS-patients [1]. Introduction 8

Tuberculosis in Austria In Austria, tuberculosis is a disease mandatorily notifiable by physicians to the district public health authorities [2]. There are 9 provincial public health directorates in Austria one for each province and a total of 99 district public health authorities. The Austrian Tuberculosis Act was established in 1968 and updated several times in the past; the last addendum took place in 2002. The aim of the Tuberculosis Act is to bring about the regulation of the measures for the prevention and control of tuberculosis in Austria. According to the Tuberculosis Act [2], case definitions for TB surveillance are as follows: A notifiable case (i) is a laboratory confirmed case of tuberculosis (a definite case, see below); a probable case of tuberculosis, as indicated by clinical or radiological signs or symptoms (other than definite case, see below); and (ii), a case of death due to confirmed or probable tuberculosis. The definitions of definite case and of other than definite case as used in this evaluation project, are based on the those given by EuroTB. A definite TB case is a patient with culture-confirmed disease due to M. Tuberculosis. The other than definite TB case is a patient meeting with both of the following conditions: (i) a clinician s judgement that the patient s clinical and/or radiological signs and/or symptoms are compatible with tuberculosis, and, (ii) a clinician s decision to treat the patient with a full course of anti-tuberculosis treatment [3]. The Austrian Tuberculosis Act [2] clearly states what measures have to be taken by law: Persons fulfilling one of the case definitions (as given above) have to be notified to the district public health authority. Contact investigations have to be performed by the district public health authority after the occurrence of an index case. TB case management in the outpatient setting has to be coordinated by the responsible district public health authority and conducted by a pulmonologist or qualified physician. The TB Act requires that examinations of TB cases, patients with suspected TB, and of contact persons of TB cases are performed according to current knowledge. Targeted screening in order to identify cases of active TB have to be performed in high risk populations, as determined by the provincial governor. Introduction 9

There are no national guidelines providing detailed recommendations for the procedure of case findings (contact investigation, TB screening), on the use of newer methods of diagnosis (e.g. gamma-interferon release test), the procedure of case management in inpatient and outpatient settings, or the programmatic management of drug-resistant tuberculosis. A simplified diagram of the Austrian TB-surveillance system: Microbiological diagnosis Self-presentation Hospital-based physician Outpatient setting Pulmonologist Test Result BSL-II /-III mycobacteriological Laboratories Test Result Case of active TB Detected by Contact investigation Screening Case Report Local level: District PH authority National TB Reference Laboratory Department for TB Epidemiology Data Consolidation: Laboratory data & notification data Monthly transfer of TB case reports Annual data set transfer Case Report Case Report Case Report National level: TB Coordination Center Ministry of Health Provincial level: Provincial PH authority Aggregated data annually sent to European level: EuroTB Figure 1: The flow of information from its sources, to the components of data collection, data transfer, data compilation, data analyses and dissemination. Detection of a TB case. A TB case can be passively detected through selfpresentation of the patient or during medical examinations for diagnostic purposes other than TB. A TB case can be actively detected through contact investigations or screening. Primary reporter. The detected case is reported to the relevant district public health authority via the initial TB case report by the TB case identifying physician. Introduction 10

District public health authority. Until 2009 data transfer occurred as follows: An extended TB case report was composed at the district level, and included additional case data (citizenship, residence status, laboratory test performance, comorbidities, clinical data, case finding). Upon completion, the extended TB case report was sent to the National TB Coordinator at the Ministry of Health. National Coordination Center and National Reference Laboratory. At the National TB-Coordination Centre the extended TB-case reports were collected and monthly sent to the National TB-Reference Laboratory. At the National TB-Reference Laboratory the notification data from the extended and final TB-case report were merged with the laboratory data provided by the biosafety level 3-mycobacteriology laboratories into the TB surveillance data base. National TB Coordination Centre and EuroTB. The 1-year data set extracted from the TB surveillance data base was sent to the National TB Coordination Centre at the end of each year, where the data were aggregated according to the requirements of EuroTB, and then sent to EuroTB. According to these data the incidence of TB disease in Austria has steadily declined since 1997: from 1,400 cases in 1997 to 894 cases in 2006 (17.55/100,000 to 10.82/100,000), which accounts for an incidence reduction of 38%. The proportion of multi-drug resistant tuberculosis (MDR-TB) in all definite cases of TB has increased from 1997 until 2004 by 6.1 fold from 0.48% to 2.95% and declined to 1.88% in 2006 [4-9]. The Austrian TB surveillance database Before 2000, the tuberculosis notification data obtained from the public health districts were entered into a database using EPIINFO Version 6 software at the National Tuberculosis Control Coordinating Centre at the Ministry of Health. In addition, laboratory data (including results of drug susceptibility testing) were entered into a database at the National TB reference laboratory, also using EPIINFO Version 6 software. In the year 2000 the National TB reference laboratory was assigned to manage the entire TB surveillance data, including notification data and laboratory data merged in one database, with the aid of the EPIINFO Version 6 software. In the year Introduction 11

2006 a new electronic database - Oracle 10g using oracle 10g web toolkit - was established for the TB surveillance data. The entire data set for the years 1995-2006 was transferred from the EPIINFO database into this new database. The Department of Infectious Disease Epidemiology in AGES was mandated by the Ministry of Health, Family and Youth to perform a 10-year TB surveillance data analyses. Introduction 12

The following case definitions given by EuroTB [11] are applied for the Austrian TB surveillance system: Classification of TB cases Definite case A patient with culture-confirmed TB disease due to Mycobacterium tuberculosis complex. Other than definite case Pulmonary case A patient with clinical and/or radiological signs compatible with TB, and a clinician s decision to treat with full treatment. A patient with TB of the lung parenchyma, tracheobronchial tree, or larynx. Extra-pulmonary case Primary drug 2 resistance Acquired drug 2 resistance A patient with TB of organs other than the lung parenchyma, tracheobronchial tree, or larynx. Resistance to one or more anti-tb drugs at the start of treatment in a TB case not previously treated Resistance to one or more anti-tb drugs at the start of treatment in a TB case previously treated for at least 1 month Multi-drug resistance Resistance to at least isoniazid and rifampicin. (MDR) Categories of treatment outcome at 12 months Cured Treatment completed and: - samples taken at the end of treatment and on at least one previous occasion are culture negative, or - sputum microscopy for acid-fast bacilli (AFB) is negative both at the end of treatment and on at least one previous occasion. Completed Treatment completed, but does not meet criteria for cured or failed. Died Death before commencement of treatment or during treatment, irrespective of cause. Failed Defaulted Transfer Still on treatment Unknown Culture or sputum microscopy remains or becomes positive again at 5 months or later during treatment. Treatment interrupted for 2 or more consecutive months. Patient referral to another clinical unit for treatment and final outcome unknown. Patient still being treated after 12 months and does not meet criteria for any other outcome. Information on outcome not available, for cases not known to have been transferred to another clinical unit 2 Anti-tuberculosis drug resistance in the world. The WHO/IUATLD Global project on anti-tuberculosis drug surveillance1994 1997. WHO/TB/97.229. Geneva: WHO, 1997. Procedure for data control 13

2. Procedure of checking the crude data set and making the data set eligible for the analyses Procedure of making this data set eligible for analyses The Austrian 10-year tuberculosis surveillance data were provided by the Austrian National Reference Laboratory for Tuberculosis (TB-NRL). An IT expert conducted the data transfer from the electronic TB-data base into an excel file document including a data set of 11,537 records. The crude data set was checked for the variables required for each case record by Euro TB (see variables below) [10]. Each case record was reviewed for repeated entry, for missing data and for data discrepancies on these variables. Variables required by Euro TB According to the EuroTB surveillance guideline (2006), variables to be reported for each TB case include: country of report year of report sex age country of origin (of birth or citizenship) localization of disease (pulmonary, extrapulmonary TB) clinical criteria for defining TB case 3 previous TB diagnosis year of previous TB diagnosis previous treatment for TB outcome of previous treatment for TB culture result sputum smear result 3 Clinical criteria according to Definition for EUROTB Joint Tuberculosis Data Collection 2007: Signs, symptoms and/or radiological findings consistent with active tuberculosis in any site Procedure for data control 14

other test results (pathology, nucleic acid) antibiotic susceptibility (first line), M. tuberculosis species outcome at 12 month The following section includes: 2.1. Process of ensuring that each case record represents a new case of TB 2.2. Creating further variables for the analyses 2.3. Handling with missing data 2.4. Handling with data discrepancies Procedure for data control 15

2.1. Process of ensuring that each case record represents a new case of TB Microsoft Excel program was used for data checking process [Table 1, Figure 2]. Records with identical ID codes Among the 11,537 records, case records with non-unique ID codes of case records were observed. The existing ID codes were declared as unreliable for identifying case records. For case records with identical ID codes, new ID codes were assigned. Records with identical surname AND prename combination Step 1 Identifying records with identical surname and identical prename Process We combined the surname and prename to a keyword (e.g. Mustermann Franz => MustermannFranz) Findings Identical surname and prename combinations were identified in 609 records, included: 254 records occurred 2 times (ntotal=508) with identical surname and prename, 23 records occurred 3 times (ntotal=69) with identical surname and prename, 4 records occurred 4 times (ntotal=16) with identical surname and prename, 2 records occurred 5 times (ntotal=10) with identical surname and prename, 1 record occurred 6 times (ntotal=6) with identical surname and prename. Step 2 Reviewing the date of birth of these 609 records Process We compared the date of birth for the records with identical surname and prename combinations for all the 609 records in order to identify records of the same persons. Procedure for data control 16

Findings In 212 records out of these 609, the dates of birth were different. Therefore, each of these 212 records was considered as a record of a case. In the remaining 397 records with identical name/surname combinations, there were 359 in which the date of birth was identical (resulting from 172 persons, including persons with > 2 case records). For the remaining 38 records with identical name/surname combinations, slightly different dates of birth were found. For these records, the data on dates of birth were cross checked with the original notification forms (see below). Step 3 Identifying the true values of the date of birth in the records with identical surname and prename but slightly different dates of birth Process We checked the hard copies of the original notification forms of the 38 records Findings and correction Based on the hand written entries on the hard copies of the original notification forms 6 case records out of the 38 turned out to origin from 6 different TB cases (the slightly different dates of births were in accordance with the original documents). For the remaining 32 records the incorrectly entered data on the dates of birth were corrected according to the true values in the original documents. Finally, 391 records with identical surname/ prename / date of birth combinations remained (giving 188 persons being recorded more than once in the data set). Procedure for data control 17

Step 4 Further processing with multiple records from the same person Process Consider: the variable date of laboratory report means the date on which the report of the drug-susceptibility testing were provided to the TB database manager; the lab report was usually provided per quarter. We checked the remaining 391 records with identical surname/prename/date of birth combinations (i.e. MustermannFranz01011999) giving 188 persons being recorded more than once in the data set) on the date of diagnosis, date of laboratory report and the results of drug susceptibility testing. The following corrections of case records were conducted: multiple records from the same person with identical dates of diagnosis AND identical dates of laboratory report were deleted except one in case of multiple records from the same person with dates of diagnosis differing < 12 months the record including the earliest date of diagnosis was kept in the data set in case multiple records from the same person with identical dates of diagnosis but different dates of laboratory report the record including the most recent date of laboratory report was kept in the data set in case of multiple records from the same person with dates of diagnosis differing 12 months each of the records from that person was kept as record of a new case in the data set Result of the crude data check 11,496 records remained in the data set. Each of the 11,496 records represents a new case of TB. Procedure for data control 18

Table 1; Process of checking multiple records and the result Records in crude data set N total = 11,537 n (%) Records fulfilling the criteria identical surname AND prename 609 (5.3) Records fulfilling this criteria occurring 2 times 508 (4.4) Records fulfilling this criteria occurring 3 times 69 (0.6) Records fulfilling this criteria occurring 4 times 16 (0.1) Records fulfilling this criteria occurring 5 times 10 (0.09) Records fulfilling this criteria occurring 6 times 6 (0.05) Records fulfilling the criteria identical surname AND prename BUT different birth date Records fulfilling the criteria identical surname AND prename AND identical date of birth Records fulfilling the criteria identical surname AND prename AND identical date of birth AND date of diagnosis differing 12 months (def.: new case) 218 Process of checking multiple records This kind of records were kept as records of new cases in the data set 391 (3.4) Multiple records from same persons 306 (2.7) This kind of records were kept as records of new cases in the data set Records fulfilling the criteria identical surname AND prename AND identical date of birth AND date of diagnosis differing < 12 months (def.: a not to be counted TB-episode ) 40 (0.4) Out of this kind of multiple records from the same person the one record with the earliest date of diagnosis was kept in the data set Records fulfilling the criteria identical surname AND prename AND identical date of birth AND identical date of diagnosis Records fulfilling the criteria identical surname AND prename AND identical date of birth AND identical date of diagnosis AND identical date of laboratory report Records fulfilling the criteria identical surname AND prename AND identical date of birth AND identical date of diagnosis AND different date of laboratory report Records remained in the data set (each of these record represents a new case) 45 (0.4) 37 (0.3) Out of this kind of multiple records from the same person one record was kept in the data set 8 (0.07) Out of this kind of multiple records from the same person the one record with the most recent date of laboratory report was kept in the data set 11,496 Procedure for data control 19

Figure 2; Process of data check concerning multiple records of the same person Procedure for data control 20

2.2 Creating further variables for the analyses Define definite case and other-than-definite case (variable name: definite case ) We created the binary variable definite case (yes/no). When there was either at least one specimen (e.g. sputum, urine, liquor) culture positive for MTC given or result of drug-susceptibility testing was given, the value yes was entered for the variable definite case in that case record, otherwise the value no was entered (is equal to an other-than-definite TB case). Findings There were 7,482 records of fulfilling the criteria above and 4,014 records otherwise. Define the MDR-TB of the records of a definite case (variable name: MDR- TB ) We created a binary variable MDR-TB (yes/no). In case of susceptibility result resistant for isoniazid and rifampicin we entered yes, otherwise we entered no. Define the poly resistant TB (variable name: poly resistant TB ) We created the binary variable poly resistant TB. In case of susceptibility result resistant to isoniazid and at least to one of the other anti-tb drugs other than rifampicin, or resistant to rifampicin and at least to one of the other anti-tb drugs other than isoniazid, we entered yes, otherwise we entered no. Procedure for data control 21

2.3 Handling missing data (in terms of no data entry=blanks) Microsoft excel program was used to deal with the missing data [Table 3]. Step 1 Checking missing data (in terms of no data entry) for variable date of diagnosis Process We used the filter function to select the records with missing data on variable date of diagnosis. Findings and solution There were 219 records with missing data on the variable date of diagnosis. Apart from the date of diagnosis the variable date of laboratory report is available. The date of the laboratory report means the date of completion of the drug-susceptibility testing which is provided per quarter. We entered date of the laboratory report if available as substitute for the diagnosis date. Step 2 Checking missing data (in terms of no data entry) for variable date of laboratory report Process We used the filter function to select the records with missing data for variable date of laboratory report Findings All records with data on drug-susceptibility testing were also provided with the date of the laboratory report. Hence, there were no missing data. Procedure for data control 22

Step 3 Checking missing data (in terms of no data entry) for variable drugsusceptibility testing when date of the laboratory report is available Process We used the filter function to select the records with date of laboratory report given but missing data on drug susceptibility testing. Findings There were 13 records fulfilling these criteria. For the variable drug susceptibility testing the value no result obtained was entered Step 4 Checking missing data (in terms of no data entry) for variable treatment outcome Process We used the filter function to select the records with missing data on variable treatment outcome. Table 2. Records with missing data on treatment outcome by notification year, 1997-2006 Notification No of records with missing year treatment outcome data 1997 115 1998 74 1999 53 2000 17 2001 34 2002 26 2003 20 2004 49 2005 49 2006 134 total 571 (4.97%) Procedure for data control 23

Findings and solution There were 571 records out of a total of 11,496 case records with missing data on treatment outcome. Table 2 is displaying the number of records with missing data on treatment outcome by year. For these records no original hard copies are any more available. We entered missing data for treatment outcome in all these records. Step 5 Check missing data (in terms of no data entry) for variable citizenship Process We used the filter function to select the records with missing data (=blank) or with the entry unknown for the variable citizenship. Findings All records had an entry for variable citizenship. In 217 records citizenship was given as unknown. Step 6 Check missing data (in terms of no data entry) for variable previous date of TB diagnosis Process We used the filter function to select the records with missing data on variable previous date of TB diagnosis. Findings and solution There were 139 records with missing data for the variable previous date of TB diagnosis. We entered missing data for this variable. Procedure for data control 24

2.4. Handling data discrepancies Microsoft excel program was used to check the data discrepancies [Table 3]. Step 1 Checking the discrepancies between data on culture result and availability of data on drug-susceptibility testing (i.e laboratory data ) Process We used pivot table to create a 2x2-table with the variables Definite case and result on drug-susceptibility testing in order to identify whether for the case record fulfilling the criteria of a case record of a definite case in terms of culture positive specimen given data on the drug-susceptibility testing is available. Findings and solution There were 473 records with at least one specimen culture positive given but with no data on drug-susceptibility testing available. In these case records the value yes for the variable definite case was changed to the value no. Finally there were 7009 case records fulfilling the criteria of a case record of a definite case : availability of either the data on drug-susceptibility testing or of the date of the laboratory report; remaining 4487 case records defined as case records of other-than definite cases. Procedure for data control 25

Step 2 Checking the discrepancies between data on treatment outcome and data on culture result or microscopy result Process We used the filter function to select case records with treatment outcome cured ; and among these records we checked whether the criteria for cured were fulfilled; def.: for outcome cured: a case with treatment completion AND culture becoming negative on samples taken at the end of treatment OR sputum microscopy becoming negative for AFB at the end of treatment. Findings and solution In 78 case records there were no entry positive result for culture or microscopy for any specimen (e.g. sputum, urine, stool, liquor). The criteria for treatment outcome cured were not fulfilled. The entry was changed from cured to completed in all these 78 case records. Procedure for data control 26

Records Table 3; Process of handling records with missing data on key variables and records with data discrepancies N total =11,537 (%) N new case =11,496 (%) Missing Data 1 735 (6.4) 725 (6.3) Data on date of diagnosis 219 (1.9) 215 (1.9) Handling with missing data Entered date of completion of drug-susceptibility testing or date of report on drug-susceptibility test result (quarterly provided) Data on outcome of treatment 571 (4.8) 561 (4.9) Entered missing data (to be clarified for the years 2004, 2005, 2006) Data on citizenship 0 0 217 records with entry unknown for citizenship accepted Data on history of TB 216 (1.9) 214 (1.9) Entered missing data Data on date of previous TB diagnosis 139 (1.2) 139 (1.2) Entered missing data Data discrepancy Entry of value positive culture but missing data on drugsusceptibility Entry of data on drug-susceptibility but missing data on the specimen(s) culture positive for MTC 473 (4.1) 473 (4.1) Entry cured for variable treatment outcome in other-thandefinite cases in which no data on specimen positive for AFB 78 (0.7) 78 (0.7) by microscopy at the beginning of treatment were available 2 Missing Data in case records of definite TB-cases Considered entry of culture positive as an entry error; value yes for the variable definite case was changed into no 1,104 (9.6) 1,098 (9.6) Entering culture positive specimen unknown N definite case = 7,482 N definite case = 7,009 Value cured was changed to completed 2 Case Records with data on the date of the laboratory report BUT Entered no result obtained 13 (0.17) 13 (0.18) missing data on drug-susceptibility Records with data on drug-susceptibility BUT missing data on 0 0 date of the laboratory report Missing Data in case records of lethal TB case N deaths =1,298 N deaths =1,297 Data on cause of death (entry option: TB or other than TB) for Entered other than TB 15 (1.2) 14 (1.1) cases of TB with lethal outcome 1 records with missing data at least on one of the key variable 2 def.: classified as cured : treatment completion And culture becoming negative on samples taken at the end of treatment and on at least one previous occasion OR- sputum microscopy becoming negative for AFB at the end of treatment and on at least one previous occasion; treatment completion AND not meeting the criteria to be classified as Cured or Failed Procedure for data control 27

3. Objectives Primary Objective To describe the burden of tuberculosis in Austria over the past 10 years, 1997-2006 by the following measures TB in Austria, 1997-2006 Crude annual incidence of TB, 1997-2006 Annual incidence of TB by case classification, 1997-2006 Tuberculosis by province, 1997-2006 Annual sex-specific incidence of TB in Austria, 1997-2006 Annual age-group specific incidence of TB in Austria, 1997-2006 Mean annual age-group- and sex-specific incidence rate of TB, 1997-2006 Annual incidence of TB in Austrians and non-austrians, 1997-2006 Mean annual incidence rate of TB in Austrians and non-austrians, 1997-2006 TB in non-austrians by country of origin, 1997-2006 Annual incidence of TB in Austrians and non-austrians by sex, 1997-2006 Mean annual age- and sex-specific incidence rate of TB in Austrians and non- Austrians, 1997-2006 Age-adjusted annual incidence of TB in non-austrians and Austrians by province, 1997-2006 Age-group specific annual incidence of TB in non-austrians and Austrians, 1997-2006 MDR-TB in Austria, 1997-2006 Distribution of resistance pattern to anti-tb drugs in cases of definite TB in Austria, 1997-2006 Crude annual incidence of MDR-TB in Austria, 1997-2006 Annual sex-specific incidence of MDR-TB in Austria, 1997-2006 Mean annual age-group and sex-specific incidence rate of MDR-TB, 1997-2006 MDR-TB in Austrians and non-austrians, 1997-2006 Mean annual incidence rate of MDR-TB in Austrians and non-austrians, 1997-2006 Annual sex-specific incidence of MDR-TB in non-austrians, 1997-2006 Objectives 28

Annual number of MDR-TB cases in non-austrians by country of origin, 1997-2006 Annual number of MDR-TB cases in Austrians and non-austrians by province of occurrence, 1997-2006 TB-Outcome, 1997-2006 TB-cases by treatment outcome Annual proportion of cured TB cases, 1997-2006 Annual proportion of TB cases with treatment completion, 1997-2006 Annual proportion of TB cases lost to follow up, 1997-2006 Annual number of MDR-TB cases by treatment outcome Mortality of TB in Austria, 1997-2006 Secondary Objective To elucidate the influence of immigration to Austria on the epidemiology of TB in Austria between 1997 and 2006 by the following analyses Relationship between TB in non-austrians and TB in Austrians in Austria Statistics of Immigration to Austria, 1997-2006 Immigration to Austria and TB among non-austrians Non-Austrians who developed TB by year(s) elapsed since immigration to Austria, 1997-2006 Non-Austrians who developed TB by year(s) elapsed since immigration, and residence status in Austria, 1997-2006 5-year-Incidence of TB in the annual immigrant-cohorts, 1997-2006 Annual proportion of TB in non-austrians and annual number of immigrants from high TB incidence countries, 1997-2006 Correlation between immigration from high TB incidence countries and TB among non-austrians Correlation between immigration from 50/100,000 TB incidence countries and TB among non-austrians, 1997-2006 Correlation between immigration and TB in non-austrians by the level of TB incidence in the countries of origin, 1997-2006 Correlation between immigration from high TB incidence countries and MDR-TB among non-austrians Objectives 29

Correlation between immigration from high TB incidence countries and MDR-TB among non-austrians by countries/regions of origin Immigration to Austria and TB in long-stay non-austrians Immigration to Austria and TB in Austrians Objectives 30

4. Material and Methods 4.1. Data sources Three sets of data were used. Data Set 1: Austrian 10-year tuberculosis surveillance data (individual case data) were provided by the Austrian National Reference Laboratory for Tuberculosis (TB-NRL) which was running the TB database until 2009. An IT expert conducted the data transfer from the electronic TB database into an excel file that included 11,537 records. Data were checked for plausibility as described above. The final data set for analyses included 11,496 records. Data Set 2: Austrian population statistics for 1997 to 2006 (aggregated data) provided by the National Statistical Institute of the Republic Austria named Statistic Austria contained data on sex, age, province of residence and data on citizenship in terms of Austrian citizens, citizens from the former Yugoslavia, citizens from Turkey, and citizens of other unspecified citizenships (see glossary for definition). Statistic Austria also provided data on the immigrants by country of origin and per year for the years 1993-2006 (aggregated data). Data Set 3: Statistics on asylum seeker by nationality per year for 1997 to 2006 were provided by the Austrian Ministry of Internal Affairs (aggregated data). Materials and Methods 31

The year of onset of TB was chosen as the time variable for analyses as data on the month of onset are not reliable. In Austria the district public health authority or an assigned pulmonologist have to complete an extended TB case report, which includes a date of diagnosis including day, month and year. How the date of diagnosis is determined is not clear to us. As a uniform definition for TB disease onset is unavailable, it could refer to any of the following: the date of detection of acid-fast-bacilli in any organ secretion, the date of detection of MTCnucleic acid in any organ secretion, the date of detection of MTC in any organ secretion by culture, the date of receiving the TB-case report at the district public health authority and finally, the date of clinical onset. When the date of diagnosis was not available, it was substituted with the date of completion of the drugsusceptibility testing (n=15), or with the date of the provision of the laboratory report on the drug-susceptibility testing (n=200). This means that there were 215 with substitutes for the true date of diagnosis. In cases of records having substituted the date of diagnosis with the date of report on drug-susceptibility test results, the date of diagnosis was March 31, June 30, September 30 or December 31, as reports on drug-susceptibility test results are provided quarterly. 4.2. Study designs Incidence studies and a correlational study were performed. The correlational study explored the statistical relationship between tuberculosis in Austria and immigration to Austria. The Austrian TB surveillance database and the Austrian immigration database (source of database given above) were cross-referenced. 4.3. Statistical analysis The disease burden of TB (including MDR-TB) was described by the annual incidence of TB and of MDR-TB for the years 1997-2006, and by the trends of TB (including MDR-TB) across the same 10-year period. Materials and Methods 32

The annual TB-incidence and MDR-TB-incidence were stratified by sex, agegroups, citizenship (Austrians, non-austrians), TB case classification (i.e. definite case, other than definite case) and province of residence. The chi-square test was used for testing the difference in the annual TB-incidence and MDR-TB incidence between females and males, between age-groups (by using age-group 10-14 as a reference group), between Austrians and non-austrians, and for testing the incidence percentage change between selected surveillance years. ANOVA test was performed in order to analyse the difference in age of the TB cases of the years 1997-2006. Student s t-test was used in order to test the difference in age between females and males for each year, 1997-2006. For plotting the province-specific annual TB incidences, age-adjustment was performed by direct standardisation using WHO Standard Population [12]. A simple linear regression model or, when appropriate, a log-linear regression model were used for testing the changes in TB (MDR-TB) incidences (sexspecific, age-specific, case-classification-specific, province-specific, nationalityspecific), and the TB outcome across the 10-year period (1997-2006). The average annual percentage change (AAPC) was calculated by fitting a regression line to the natural logarithm of the rates using calendar year as a regressor variable (i.e. y=mx+b, where y=ln(rate) and x=calendar year; EAPC=100 (e m - 1)) [13]. The trend in the annual average age across the 10-year period was tested by using a simple linear regression model. The differences in the slope or average annual percentage change (AAPC) between sexes, citizenships (i.e. Austrians and non-austrians), TB case classes (i.e. definite case, other than definite case) and provinces, were tested by t-test (WHO-TB-workshop: http://www.who.int/tb/surveillanceworkshop/). Linear relationships between immigration to Austria during the years 1997 to 2006 and TB (MDR-TB) in Austria were examined by calculating the Pearson correlation coefficient. SAS version 9.1 or STATA version 10 was used for data analysing. Materials and Methods 33

5. Results 5.1. Primary Objective: The burden of TB in Austria, 1997-2006 TB in Austria, 1997-2006 Crude annual incidence of TB, 1997-2006 From 1997 to 2006, a total of 11,496 tuberculosis (TB) cases were reported in Austria, including 7,009 definite cases and 4,487 other than definite cases. Out of 11282 cases on which information on TB site were available 9269 were of pulmonary TB and 2013 of extrapulmonary TB. In 6,736 cases the mycobacterial organism was identified: in 4,018 cases, Mycobacteria tuberculosis complex was given; in 2,689 cases, M. Tuberculosis was detected; and in 29 cases, M. bovis was detected. Between 1997 and 2006 the crude annual incidence decreased significantly by 0.77 cases per 100,000 population per year (slope= -0.77, 95%CI: -0.96 to -0.58, R 2 =0.92, p<0.001) [Figure 3, see annex Table 4]. The average annual percentage change (AAPC) in TB incidence was -5.2% (95% CI: -6.3% to -4.0%). From 1997 to 2003 the incidence decreased even more linearly (slope= -1.04, 95%CI: -1.21 to -0.87, R 2 =0.98, p<0.001). From 2003 to 2004 the incidence increased slightly from 12.22 to 13.26/100,000 population (incidence percentage change (IPC): 8.5%, p=0.06); after 2004 the incidence decreased to 10.9/100,000 population in 2006 (IPC: 17.8%, p<0.001). The averaged annual proportion of pulmonary TB in all TB cases from 1997 to 2006 was 82.2% (80.3-84%). Results 34

Annual incidence of TB by case classification, 1997-2006 The incidence of definite TB and of other than definite TB decreased significantly between 1997 and 2006 (definite TB: slope= -0.45, 95%CI: - 0.57 to -0.34, R 2 =0.91, p<0.0001; other than definite TB: slope= -0.31, 95%CI: - 0.47 to - 0.16, R 2 =0.73, p<0.01) [Figure 4]. The AAPC between the incidence of definite TB and of other than definite TB showed no difference [definite TB: -5.1%, 95% CI: -6.4% to -3.8% vs. other than definite TB: -5.2%, 95% CI: -7.4% to -2.9%). There was a steep decline in the incidence of other than definite TB from 8.07/100,000 population in 1997 and to 5.6/100,000 population in 1999 (IPC: - 30.6%; p<0.001) compared to no significant change in the incidence of definite TB within this time period (1997: 10.51; 1998: 10.25, 1999: 10.54). The incidence of definite TB decreased from 10.53 in 1999 to 8.30/100,000 population in 2001 (IPC: 21.3%; p<0.001). Between 2001 and 2002 the incidence of definite TB experienced again a slight increase followed by a decrease till 2003 (8.3 to 8.53 to 7.45/ 100,000 population), whereas the incidence of other than definite TB decreased steadily from 2001 to 2003 (5.21 to 4.86 to 4.76/100,000 population). The slight increase in the incidences of definite TB and other than definite TB from 2003 to 2004 was not significant (IPC in definite TB: 6.3%, p=0.31; IPC in other than definite TB: 12.6%, p=0.09). A significant decrease in both forms of TB was observed between 2004 and 2006 (ICP in definite TB: 18.7%, p<0.01; IPC in other than definite TB: 16.4%, p=0.01). Analysing the changes in the ratio of definite TB cases to other than definite TB cases over the 10-year period of analysis, an increase in the ratio was observed from 1.3 in 1997 to 1.9 in 1999, and, after a decrease to 1.6 in the years 2000 and 2001, again an increase to 1.8 in 2002. Between 2002 and 2006 the ratio decreased to 1.4 [Figure 4, see annex Table 4]. Results 35

Figure 3: Crude annual incidence of TB in Austria, 1997-2006 20 18.58 Incidence (cases/100,000 population) 15 10 5 17.10 16.14 15.30 13.51 13.39 12.22 13.26 y = -0.77x + 18.49 R 2 = 0.92 12.27 10.90 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 4: Annual incidence of TB by case classification, 1997-2006 12 Annual incidence of definite TB Annual incidence of other-than-definite 10 Incidence (cases/100,000 population) 8 6 4 2 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 36

TB by province, 1997-2006 There are 9 provinces in Austria: Vienna, Upper Austria, Lower Austria, Salzburg, Carinthia, Styria, Tyrol, Vorarlberg and Burgenland [Table 5]. In this 10-year study, Vienna was the dominant province in terms of place of residence of TB cases, accounting on average for 30.7 % of all cases. Following Vienna was the province of Upper Austria, with an average of 16.7 % of all cases. The third most prevalent province of cases residence was Lower Austria, and the fourth most prevalent province was Styria. Figure 5 displays the age-adjusted mean annual province-specific incidence rate of TB (direct standardization by using WHO standard population from 2000-2025 [13]) from 1997 to 2006 by regional mapping. The province Vienna was most seriously affected with a mean annual incidence rate of 20.7/100,000 population, followed by the province Upper Austria with a mean annual incidence rate of 11.9/100,000 population, and the province Salzburg with a mean annual incidence of 11.4/100,000 population. Between 1997 and 2006, the age-adjusted annual TB incidence decreased significantly in all 9 provinces of Austria except Styria. The highest annual average percentage change (AAPC) was observed in the province of Carinthia with an AAPC of -10.1% (95%CI: -14% to -6%), followed by the province of Burgenland (AAPC: - 8.1%; 95%CI: -14.7% to -1%), the province of Vorarlberg (AAPC: -8%; 95%CI: -10.6% to -5.3%), and the provinces of Upper Austria, Salzburg and Lower Austria. The lowest AAPC was observed in the province of Vienna with -1.8% (95%CI: -3.6% to -0.02%) [Figure 6a-6d; Table 5]. Results 37

Figure 5: Age-adjusted mean annual incidence rate of TB by province, 1997-2006 Results 38

Table 5: Age-adjusted annual incidence of TB by province, and the province-specific annual average percentage change (AAPC) of TB, 95% CI, 1997-2006 Age-adjusted annual incidence of TB Year Vienna Upper Lower Austria Austria Salzburg Carinthia Styria Tyrol Vorarlberg Burgenland 1997 22.83 18.94 10.83 13.31 17.81 14.41 8.98 15.88 14.39 1998 21.74 15.03 11.42 13.03 14.11 11.79 11.43 14.82 11.56 1999 20.93 15.43 10.46 11.15 12.99 11.99 11.11 11.36 7.97 2000 20.59 11.59 10.88 16.03 13.43 10.55 11.57 10.63 10.03 2001 20.82 10.07 8.81 13.58 12.10 8.32 9.90 9.77 6.76 2002 20.88 10.94 8.38 8.84 8.86 11.14 7.42 9.15 6.98 2003 19.89 9.12 7.32 10.82 8.02 9.56 8.09 10.58 8.91 2004 22.14 9.03 9.53 9.28 5.99 12.98 8.03 9.87 10.31 2005 20.81 10.58 8.92 9.51 6.44 9.11 7.55 7.04 8.16 2006 16.57 8.72 8.19 8.05 8.84 8.34 7.01 6.71 3.60 AAPC (95%CI) -1.8% (-3.6%, -0.02%) -7.4% (-10.4%, -4.4%) -3.5% (-6 %, -1%) -5.5% (-9.1%, -1.7%) -10.1% (-14%, -6%) -3.8% (-7.5%, 0.1%) -4.8% (-7.8%, -1.7%) -8% (-10.6%, -5.3%) -8.1% (-14.7%, -1%) Results 39

Figure 6a: Age-adjusted province-specific annual incidence of TB for the provinces Vorarlberg and Tyrol (Western Austria), 1997-2006 25 Age adjusted annual incidence of TB in the province Tyrol Age adjusted annual incidence of TB in the province Vorarlberg 20 Incidence (cases/100,000 population) 15 10 5 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 6b: Age-adjusted province-specific annual incidence of TB for the provinces of Salzburg, Upper Austria, Lower Austria (Northern Austria), 1997-2006 25 Age adjusted annual incidence of TB in the province Salzburg Age adjusted annual incidence of TB in the province Upper Austria Age adjusted annual incidence of TB in the province Lower Austria 20 Incidence (cases/100,000 population) 15 10 5 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 40

Figure 6c: Age-adjusted province-specific annual incidence of TB for the provinces Styria, Carinthia (Southern Austria), 1997-2006 25 Age adjusted annual incidence of TB in the province Steyr Age adjusted annual incidence of TB in the province Carynthia 20 Incidence (cases/100,000 population) 15 10 5 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 6d: Age-adjusted province-specific annual incidence of TB for the provinces of Vienna and Burgenland (Eastern Austria), 1997-2006 25 Age adjusted annual incidence of TB in the province Vienna Age adjusted annual incidence of TB in the province Burgenland 20 Incidence (cases/100,000 population) 15 10 5 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 41

Annual sex-specific incidence of TB in Austria, 1997-2006 From 1997 to 2006 the TB incidence in males decreased by 4.7% on average per year (95% CI of the AAPC: -6.1% to -3.4%), compared with 6.1% on average per year in females (95% CI of the AAPC: -7.1% to -5.0%) [Figure 7, see annex Table 6]. There was no difference in the AAPC between males and females. A slight increase in the incidence in both sexes occurred from 2003 to 2004, whereas the increase was steeper in males (IPC from 2003 to 2004 in males 10% vs. in females 5.9%). Figure 7: Annual sex-specific incidence of TB in Austria, 1997-2006 30 25 24,04 Annual incidence of TB in males Annual incidence of TB in females Incidence (cases/100,000 population) 20 15 10 13,49 21,75 12,76 21,34 11,27 19,34 11,51 17,52 9,76 17,61 9,43 15,96 8,70 17,55 9,21 16,66 8,13 14,48 7,51 5 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 42

Figure 8: Ratio of TB incidence in males to TB incidence in females in Austria, 1997-2006 2,5 2 Incidence Ratio 1,5 1 0,5 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Year The incidence of TB was significantly higher in males compared to females in all years analysed [see annex Table 6]. Figure 8 displays the annual incidence ratio in males compared to females from 1997 to 2006. After a peak of male to female incidence ratio in the year 1999, there was an increasing trend in the incidence ratio from 2000, with a peak in 2005 (slope=0.06, 95%CI: 0.03-0.09, R 2 =0.87, p<0.01). Results 43

Annual age-group specific incidence of TB in Austria, 1997-2006 The study adopted age-groups as given by EuroTB: age group 1: 0-4 yrs, age group 2: 5-14 yrs, age group 3: 15-24 yrs, age group 4: 25-34 yrs, age group 5: 35-44 yrs, age group 6: 45-54 yrs, age group 7: 55-64 yrs and age group 8: 65 yrs. The lowest incidence of TB was observed in the age group 5-14 yrs in all the years analysed. From 1997 to 2006 there was no significant change of TB incidence in the 5-14 age-group (AAPC: 3.1%; 95%CI: -3.4% to 10.2%). The highest TB incidence was observed in the age-group 65 yrs in all 10 years analysed except 2003 Age group 0-4: after a decrease of TB incidence per 100,000 population from 8.72 in 2000 to 2.99 in 2002 (IPC: 70%, p<0.01), the incidence increased by 1.08/100,000 population per year (slope=1.08, 95%CI: 0.87 to 1.29, R 2 =0.99, P<0.001) from 2002 to 2006. Age group 15-24: From 1997 to 2005 this age-specific incidence increased significantly with an AAPC of 3.63% (95%CI: 0.12%-7.26%) and decreased from 14.2 to 8.1 (IPC: 42.9 %, p<0.01) in 2006. Age group 25-34: From 1997 to 2002 this age-specific incidence decreased significantly with an AAPC of 6.7% (95%CI: -9.5% to -3.8%) and increased by 47.5% in 2004 (p<0.01). After that, the incidence in that age-group decreased to 11.41/100,000 population in 2006 (IPC: 40%, p<0.01). Age group 35-44: From 1997 to 2006 the incidence in this age-group decreased significantly with an AAPC of 7.4% (95%CI: -8.9% to -5.8%). Age group 45-54: From 1997 to 2006 the incidence in this age-group decreased significantly with an AAPC of 6.6% (95%CI: -8.4% to -4.7%). Age group 55-64: From 1997 to 2006 the incidence in this age-group decreased significantly with an AAPC of 6.4% (95%CI: -8.2% to -4.5%). Age group 65yrs: From 1997 to 2006 the incidence in this age-group, with an AAPC of 8.5%, experienced the largest decrease compared to the age-groups 25-64 years (95%CI: -10% to -7%) from 1997 to 2006 (simple linear regression model: slope=-2, 95%CI: -2.5 to -1.6, R 2 =0.94, P<0.001) [Figure 9a-9c, see annex Table 7]. Results 44

Figure 9a: Age-group specific annual incidence of TB in Austria, age-groups 0-4, 5-14, 15-24, 1997-2006 Incidence (cases/100,000 population) 35 30 25 20 15 10 5 0 Annual incidence of TB in age group 0-4 years Annual incidence of TB in age group 5-14 years Annual incidence of TB in age group 15-24 years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 9b: Age-group specific annual incidence of TB in Austria, age-groups 25-34, 35-44, 45-54, 1997-2006 Incidence (cases/100,000 population) 35 30 25 20 15 10 5 0 Annual incidence of TB in age group 25-34 years Annual incidence of TB in age group 35-44 years Annual incidence of TB in age group 45-54 years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 9c: Age-group specific annual incidence of TB in Austria, age-groups 55-64, 65+, 1997-2006 Incidence (cases/100,000 population) 35 30 25 20 15 10 5 Annual incidence of TB in age group 55-64 years Annual incidence of TB in age group older or equal to 65 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 45

There was no significant difference in the annual average age in female TB cases across the 10 years 1997-2006, whereas the annual average age in male TB cases decreased significantly across this 10-year period. The increase in the average age in all cases of TB (48.9 yrs to 49.5 yrs) from 1997 to 1998 was not significant (t-test: p=0.2). From 1998 to 2005 there was a significant decreasing trend in the annual average age of TB cases (slope= -0.46, 95%CI: -0.66 to -0.25, R 2 =0.77, p< 0.001). The annual average age decreased between 1998 and 2000 (49.5 vs. 47.9yrs; p=0.051). From 2000 to 2002 the average annual age in the TB cases did not change (slope= 0.25, 95%CI: -1.58 to 2.08, R 2 =0.75, p=0.33). After then the annual average age decreased steeply to 45.3 yrs in 2005 (slope= -1.03, 95%CI: -2.32 to 0.27, R 2 = 0.85, p=0.08) and increased significantly to 46.5 yrs in 2006 (t-test; p=0.11) [Figure 10]. Figure 10: Annual average age in all cases of TB; annual average age by sex, 1997-2006 51 50 Average age of all male TB cases Average age of all female TB cases Average age of all TB cases 49 48 Average age 47 46 45 44 43 42 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year There was no difference in the average age between males and females in the years 1997 to 2003 and 2005 to 2006. In 2004 the average age in males was higher compared to females at borderline significance [see annex Table 8]. Results 46

Mean annual age-group and sex-specific incidence rate of TB, 1997-2006 The mean annual age-specific incidence rates indicate that within the Austrian population there was an increase of TB incidence with age, beginning from the 5-14 age group (Chi-square test for a trend, p<0.0001). The highest mean annual incidence rate was in the age group older than 64 years, and the lowest mean annual incidence rate was in the age group 5-14 years. The risk of TB in the age group 0-4 years was 2.32 times higher compared to the age group 5-14 years, and in the age group older than 64 years 7.9 times higher compared to the age group 5-14 years [Figure 11; see annex Table 9]. In the age group 0-4 years and 5-14 years there was no difference in the TB incidence rate between females and males. The mean annual age- and sexspecific TB incidence rates increased significantly with age, and, in the age groups 5-14 and above, the mean annual incidence rate was higher in males than in females. The highest relative risk of TB (RR) in males compared to females was in the age group 45-54 years [RR: 3.08, 95%CI: 2.78-3.43, p<0.001; Figure 11]. In the age group 55-64, males were at a 2.68 higher risk, and, in the age group of 65 and above, males were at a 1.92 higher risk of contracting TB compared to females [Figure 11; see annex Table 9]. Stratifying TB incidence into the two age groups <45 years and 45 years, the risk of TB in males aged <45 years males was 1.55 time higher (95%CI: 1.46-1.63; p<0.001) compared to females. In comparison, the risk of TB in males aged 45 years was 2.26 times higher (95%CI: 2.14-2.38; p<0.001) compared to females. Results 47

Figure 11: Mean annual age- and sex-specific TB incidence rate, 1997-2006 35 30 Mean annual age-specific TB incidence rate in males Mean annual age-specific TB incidence rate in total Mean annual age-specific TB incidence rate in females 27,6 32,3 Incidence rate ( cases/ 100,000 pys at risk) 25 20 15 10 5 0 23,7 22,7 19,9 17,3 16,9 18,2 16,1 12,7 14,9 14,4 10,9 11,5 7,4 9,8 8,9 8,9 8,9 6,7 3,1 6,0 2,9 2,6 0-4 5-14 15-24 25-34 35-44 45-54 55-64 64+ Age groups Results 48

Political Background regarding immigration activities relevant to Austria between 1990 and 2006 According to the 2001 census, out of Austria s approximate eight million inhabitants, more than 730,000 (9.1%) were foreign-born, out of which 62.8 % came from the successor states of the former Yugoslavia and from Turkey. Over the past decade 1997 to 2006 immigration to Austria increased by 5,129 immigrants annually (p=0.002). Beginning in 1991, political upheavals in the Balkans, such as the breakup of Yugoslavia, displaced about 2,700,000 people by mid-1992, of which over 700,000 of them sought asylum in Europe. In 1999, about one million Albanians escaped from alleged Serbian persecution. Over 200,000 Serbs and other non-albanian minorities fled or were expelled from Kosovo after the Kosovo War in 1999. The consequences of the war in Chechnya have inevitably spilled across borders with implications for the refugee protection regime in Europe. According to statistics from the United Nations High Commissioner for Refugees (UNHCR), around 120,000 Russian citizens sought asylum in the industrialised countries from 2000-2004. In both 2003 and 2004 asylum seekers from the Russian Federation were the largest group of people claiming asylum in the countries of Europe. Although statistical information on asylum seekers from Chechnya is not separately recorded from that of asylum seekers from other parts of the Russian Federation, UNHCR estimates that the vast majority of asylum seekers from the Russian Federation during those years were Chechens [14]. Results 49

Annual incidence of TB in Austrians and non-austrians, 1997-2006 The incidence of TB was significantly higher in non-austrians compared to Austrians in all years analysed (1997-2006) [Figure 12; see annex Table 10]. Between 1997 and 2002 the incidence ratio of TB in non-austrians compared to Austrians were almost steady with a median of 3.5 (min 3.09, max: 3.99), whereas from 2002 to 2005 the incidence ratio increased from 3.99 to 6.96 (1.7 times), giving in an increase of the TB incidence ratio by almost 1 per year (slope=0.98; 95%CI: 0.6-1.36; R 2 =0.96; p<0.01) [Figure 13]; between 2005 and 2006 the incidence ratio decreased to 5.41. Figure 12: Annual incidence of TB among Austrians and non-austrians, and crude annual incidence of TB, 1997-2006 70 Annual incidence of TB among Austrian citizens Annual incidence of TB among Non-Austrian citizens Crude annual incidence of TB 60 57,59 Incidence (cases/100,000 population) 50 40 30 20 10 54,00 18,58 15,25 44,72 17,10 14,48 47,27 16,14 13,15 48,16 15,30 12,12 38,01 13,51 11,07 41,84 13,39 10,48 45,29 12,22 8,78 13,26 8,54 53,89 12,27 7,74 40,98 10,90 7,58 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Incidence of TB in non-austrians: After a decrease of the TB incidence in the non-austrians from 54 to 44.72 cases/100,000 population (p<0.05) from 1997 to 1998, the annual TB incidence increased to 48.16/100,000 population in 2000, and declined again to 38.01 cases/100,000 population in 2001 (p<0.01). From 2001 to 2004 the annual incidence increased steeply (highest incidence Results 50

percentage change [IPC] of 51.5 %, p<0.01), and decreased again from 2004 to 2006 (IPC: - 28.9%, p<0.01). Incidence of TB in Austrians: The annual incidence of TB in Austrians decreased linearly between 1997-2006, by 9.1 cases/1,000,000 population (slope=-0.91; 95%CI: -1.02 to -0.8; R 2 =0.98; p<0.001). The average annual percentage change in TB incidence was 8.1% (95% CI: 7.3% - 8.8%) and the IPC between 1997 and 2006 50.3% [Figure 12; see annex Table 10]. Figure 13: Ratio of the incidence of TB in non-austrians to the incidence of TB in Austrians per year, 1997-2006 8 7 6 Incidence Ratio 5 4 3 2 1 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Year Results 51

Mean annual incidence rate of TB in Austrians and non-austrians, 1997-2006 The mean annual incidence rate of TB in non-austrians was significantly 4.4 times higher than the mean annual incidence rate in Austrians (non-austrians: 47.7/100,000 person years vs. Austrians: 10.9/100,000 person years, p<0.001). When standardising for age, the age-standardised 10-year mean annual incidence rate was 5.4 times higher in non-austrians compared to Austrians. The mean annual incidence rate (IR) of TB in Austrians of the period 1997-2001 was significantly higher compared to the 2002-2006 period (13.2/100,000 person years versus 8.6/100,000 person years; p<0.001). In contrast, there was no difference in the mean annual IR between these two 5-year periods in non- Austrians (1997-2001: 46.5/100,000 pyr versus 2002-2006: 48.2/100,000 pyr; p=0.31). The age-standardised mean annual TB IRs in Austrians was 10.6/100,000 pyr for the 1997-2006 period and 6.8/100,000 pyr for the 2002-2006 period. The mean annual IRs in non-austrians for these two time-periods was 46.8/100,000 pyr and 46.2/100,000 pyr after standardisation for age. Results 52

TB in non-austrians by country of origin, 1997-2006 Table 11 shows the TB cases in non-austrians by the country of origin of the non-austrians over the 10-year period from 1997 to 2006. From 1997 to 2006, TB cases in the non-austrians having immigrated from the former Yugoslavia accounted for the largest proportion of cases in non-austrians. Between 1997 and 2003, the origin of the second largest proportion of non- Austrian cases was Turkey. The proportion of cases in non-austrians who originated in Africa increased from 1997 to 2005. Cases in non-austrians from the Russian Federation increased from 2001 to 2004, followed by a slightly decrease. The proportion of cases in non-austrians who immigrated from Romania increased from 1997 to 2001. In 2006, cases in non-austrians from former Yugoslavia accounted for the largest proportion of cases in non-austrians (34.3%), followed by cases in non-austrians of Asian origin (14.6%), the Russian Federation (12.8%), and Turkey (12.8%) [Table 12]. Table 11: The citizenship of TB-cases among non-austrians in Austria 1997 to 2006 (Number of cases indicated in brackets) Former Yugoslavia (1,319 ), Turkey (500), Russian Federation (221), Romania (148), Asia: India (167), the Philippines (66), China (50), Thailand (29), Mongolia (28), Bangladesh (21), Vietnam (16), Indonesia (9), Nepal (6), Cambodia (3), South Korea (3), Sri Lanka (3) Malaysia (1), North Korea (1), Taiwan (1), unknown (3). Other European high incidence countries: Georgia (78), Albania (36), Poland (32), Moldavia (29), Bulgaria (17), Armenia (16), Ukraine (15), Hungary (13), Portugal (7), Azerbaijan (4), Greece (4), Spain (4), Uzbekistan (3), Kasachstan (2), Lithuania (1), Slovenia (1). European low incidence countries: Germany (21), Czech Republic (4), Italy (4), Netherlands (3), Sweden (3), Israel (2), Belgium (1), France (1), Great Britain (1), Norway (1), Switzerland (1), unknown (17). Africa: Nigeria (84), Gambia (18), Congo (15), Sierra Leone (15), Ethiopia (13), Angola (12), Algeria (11), Ghana (11), Liberia (11), Cameroon (10), Kenya (9), Uganda (4), Guinea (4), Eritrea (3), Mauretania (3), Senegal (3), Côte d'ivoire (2), DR Congo (2), South Africa (1), Tanzania (1), Zambia (1), Zimbabwe (1), unknown African countries (7). Eastern Mediterranean Region: Pakistan (68), Afghanistan (60), Somalia (34), Egypt (15), Iran (15), Morocco (10), Tunisia (9), Iraq (8), Sudan (6), Libya (3), Syria (3), Saudi Arabia (1). Latin-America: Dominican Republic (11), Brazil (5), Peru (3), Bolivia (1), Colombia (1), Ecuador (1), Nicaragua (1), Paraguay (1). Unknown origin (57) Results 53

Year Table 12: Annual distribution of cases of TB in non-austrians by country/ region of origin; the 4 European countries of origin with the dominant number of cases are separately displayed, other countries of origin are summarised according to the WHO regions, 1997-2006 Turkey Former Yugoslavia Romania Cases of TB in non-austrians n (%) other high TB Russian incidence Federation European EMRO 4 Africa Asia countries Latin America 1997 68 (21.6) 180 (57.1) 9 (2.9) 0 (0) 0 (0) 13 (4.1) 9 (2.9) 32 (10.2) 4 (1.3) 315 1998 63 (23.5) 138 (51.5) 8 (3) 0 (0) 3 (1.1) 16 (6) 9 (3.4) 30 (11.2) 1 (0.4) 268 1999 51 (17.1) 164 (54.8) 13 (4.3) 1 (0.3) 2 (0.7) 19 (6.4) 23 (7.7) 24 (8) 2 (0.7) 299 2000 51 (16.2) 154 (49) 19 (6.1) 4 (1.3) 3 (1) 22 (7) 14 (4.5) 46 (14.6) 1 (0.3) 314 2001 49 (19.1) 95 (37) 19 (7.4) 1 (0.4) 5 (1.9) 28 (10.9) 18 (7) 41 (16) 1 (0.4) 257 2002 47 (16.2) 120 (41.2) 12 (4.1) 7 (2.4) 15 (5.2) 29 (10) 25 (8.6) 33 (11.3) 3 (1) 291 2003 43 (13.1) 127 (38.8) 15 (4.6) 29 (8.9) 23 (7) 23 (7) 29 (8.9) 36 (11) 2 (0.6) 327 2004 39 (9) 116 (26.9) 20 (4.6) 81 (18.8) 43 (10) 24 (5.6) 47 (10.9) 59 (13.7) 2 (0.5) 431 2005 48 (11.5) 115 (27.4) 16 (3.8) 57 (13.6) 37 (8.8) 31 (7.4) 49 (11.7) 59 (14.1) 7 (1.7) 419 2006 41 (12.8) 110 (34.3) 17 (5.3) 41 (12.8) 17 (5.3) 28 (8.7) 19 (5.9) 47 (14.6) 1 (0.3) 321 Total N 4 EMRO: WHO Regional Office for the Eastern Mediterranean. Results 54

Year Table 13: Cases of TB in non-austrians from countries other than former Yugoslavia or Turkey Number of TB cases in non-austrians from countries other than former Yugoslavia and Turkey 1997 120 1998 103 1999 113 2000 130 2001 129 2002 139 2003 173 2004 295 2005 275 2006 187 5 most frequent countries of origin Unknown (23) other European countries (12) Romania (9) India (9) Philippines (7), China (7) Unknown (12) India (10) Romania (8) Albania(7) Philippines (7) Romania (13) India (10) Unknown (9) Albania(7), Sierra Leone(5), Poland (5), Philippines (5), Pakistan (5) India (22) Romania (19) Philippines (12) Pakistan (8) Afghanistan (7) India (21) Romania (19) Afghanistan (12) Pakistan (8) China (7) India (15) Romania (12) Nigeria (10) Afghanistan (9) Georgia (9) Russian Federation (29/173; 16.8%) India (18) Romania (15) Georgia (14) Nigeria (11) Russian Federation (81/295; 27.5%) Georgia (30) Romania (20) Nigeria (20) India (20) Russian Federation (57/275; 20.7%) India (25) Nigeria (20) Romania (16) Georgia (14) Russian Federation (41/187; 21.9%) Romania (17) India (17) Mongolia (10) Somalia (9) Results 55

Table 13 shows the five most frequent countries of origin of non-austrians other than Turkey and former Yugoslavia in which TB occurred per year, between 1997-2006. Between 2003 and 2006 the majority of cases occurred in non- Austrians having immigrated from the Russian Federation, accounting in 2003 for 16.8% of the cases coming from countries other than from former Yugoslavia and Turkey, 27.5% in 2004, 20.7% in 2005, and 21.9% in 2006. Data on the non-austrians living in Austria were segregated into three groups: group I: non-austrians of Turkish citizenship, group II: non-austrians who were citizens of former Yugoslavia, group III: non-austrians who were citizens of countries other than Turkey or former Yugoslavia. Between 1997 and 2006 the annual incidence of TB decreased significantly among non-austrians who originated in Turkey or former Yugoslavia (Turkey: slope=-1.42, 95%CI: -2.49 to -0.34, R 2 =0.54, p<0.05; former Yugoslavia: slope=-1.99, 95%CI: -3.53 to -0.45, R 2 =0.53, p<0.05). The annual incidence of TB in non-austrians who were nationals of countries other than Turkey or former Yugoslavia increased slightly (slope=0.22, 95%CI: -2.17 to 2.61, R 2 =0.02, p=0.81) between 1997 and 2002, followed by a steep increase from 47.4/100,000 population in 2002 to 89.2/100,000 population in 2004 (by 88%, p<0.001) [Figure 14; see annex Table 14]. From 2004 to 2006, the incidence of TB in this group of non-austrians decreased again (by 46.5% (p<0.001). Results 56

Figure 14: Annual incidence of TB in non-austrians who are citizens of former Yugoslavia, Turkey, and other countries (the country of origin non-austrians who are not from Turkey or former Yugoslavia is unknown). Annual incidence of TB in Non-Austrian citizens by country of origin, 1997-2006 Incidence (cases/100,000 population) 100 90 80 70 60 50 40 30 20 10 Turkey Former Yugoslavia Others 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 57

Annual incidence of TB in Austrians and non-austrians by sex, 1997-2006 Figure 15: Annual sex-specific TB incidence in Austrians and non-austrians, 1997-2006 Incidence (per 100,000population) 90 80 70 60 50 40 30 20 10 Incidence of TB in male Austrians Incidence of TB in female Austrians Incidence of TB in male Non-Austrians Incidence of TB in female Non-Austrians 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year The annual TB incidence in Austrians decreased linearly in both sexes (male: slope=-1.2, 95%CI: -1.4 to -1.0, R 2 =0.96, p<0.001; female: slope=-0.64, 95%CI: -0.69 to -0.58, R 2 =0.99, p<0.001). There was no difference in the AAPC between males and females (male: -8.2%, 95% CI: -9.2% to -6.4%; female: - 8.0%, 95%CI: -8.7& to -7.3%). TB incidence in male non-austrians decreased from 1997 to 1998 (IPC: 24.8%, p<0.01), increased between 1998 and 1999 by 24.5% (p<0.05), and then decreased again between 1999 and 2001. From 2001, TB incidence in male non- Austrians increased again and reached its peak with 77.3/100,000 in the year 2004, and then by 2006 it had declined again to 50/100,000 (IPC: -35.4%, p<0.001). The TB incidence in female non-austrians decreased from 1997 to 1999 (IPC: -21.6%, p=0.052), between 1999 and 2000 it increased by 27.6% (p=0.048), and then decreased from 2000 to 2001 by 41.7% (p<0.001). From 2001, TB incidence in female non-austrians increased again and reached its peak with 36.2/100,000 in the year 2004. It then declined between 2004 and 2006 to 31.4 /100,000 (IPC: -13.4%, p=0.25) [Figure 15]. Results 58

Mean annual age- and sex-specific TB incidence rates in Austrians and non-austrians, 1997-2006 Figure 16: Mean annual age- and sex-specific TB incidence rates in Austrians Figure 16: 10-year mean annual age- and sex-specific TB incidence rates in Austrians Incidence rate (cases/100,000 pys at risk) 35 30 25 20 15 10 5 0 4,88 4,09 3,33 0-4 Mean annual age-specific TB incidence rate in male Austrian citizens Mean annual age-specific TB incidence rate in all Austrian citizens Mean annual age-specific TB incidence rate in female Austrian citizens 1,52 1,34 5-14 1,17 4,41 15-24 3,96 4,18 8,39 7,45 6,50 14,90 25-34 35-44 Age groups 11,07 23,38 7,24 7,34 7,51 45-54 15,20 21,33 14,15 55-64 31,62 21,78 15,80 65+ The mean annual age-specific incidence rates show an increase in the TB incidence with age in Austrians aged older than 4 years (Chi-square test for trend, p<0.0001). The highest incidence rate of TB in Austrians was observed in the age group of 65 years, with 21.78/100,000 pyr in both sexes [Figure 16; Table 14]. The lowest incidence rate was in the age group 5-14 years (1.34/100,000 pyr), and taking this group as a reference group, the risk of TB in the age group 0-4 years was 3.05 times higher, and in the age group 65 years, 16.25 times higher [Figure 16; Table 14]. In the age groups 0-4, 5-14, and 15-24 there was no difference in the TB incidence rate between the female and male Austrians. The mean annual ageand sex-specific TB incidence rates indicated a significant increase of TB risk with age in both sex, and from 25 years of age and onwards TB incidence rates Results 59

increased far more significantly in males compared to females [RR in age group 25-34: 1.29; RR in age group 35-44: 2.06; RR in age group 45-54: 3.18; Chi- Square test for a trend, p<0.001; Figure 16]. In the age group 55-64 males were at a 2.84 higher risk, and, in the age group 65 years, at a 2.00 higher risk of TB compared to females. Figure 17: Mean annual age- and sex-specific TB incidence rates in non-austrians, 1997-2006 100 90 90.30 Mean annual age-specific TB incidence rate in female Non-Austrians Mean annual age-specific TB incidence rate in male Non-Austrians Mean annual age-specific TB incidence rate in total Non-Austrians 80 77.23 Incidence rate (cases/100,000 pyr) 70 60 50 40 30 20 10 25.84 25.17 24.55 17.75 16.82 15.95 69.26 47.86 59.71 43.00 58.41 48.68 36.04 60.60 62.04 52.26 46.16 39.49 26.70 63.78 59.58 54.67 0 0-4 5-14 15-24 25-34 35-44 45-54 55-64 65+ Age groups The highest mean incidence rate of TB in non-austrians was in the 15-24 year age group (69.26/100,000 person years). In this age group males were affected far more than females (90.3/100,000 pyr vs. 47.87/100,000 person years); female non-austrians were mostly affected in the age group 65 years, (63.78/100,000 pyr) [Figure 17, Table 14]. The lowest mean annual incidence rate in both sexes was in the age group 5-14 years (16.82/100,000 pyr). The risk of TB in the age group 0-4 was 1.5 times higher, and in the age group 65 years 3.54 times higher compared to the reference age group 5-14 years [Figure 17; Table 14]. In the age groups 0-4, 5-14 and 65 there was no difference in the mean annual TB incidence rate between the female and male sex. The mean annual Results 60

age- and sex-specific TB incidence rates were significantly higher in males compared to females from the age group 15-24 years to the age group 55-64 (RR in age group 15-24: 1.89, RR in age group 25-34: 1.8, RR in age group 35-44: 1.62, RR in age group 45-54: 2.27 and RR in age group 55-64: 1.57; Chi- Square test for a trend, p=0.002; Table 14). In all age groups the mean annual incidence rate was significantly higher in non- Austrians compared to Austrians. The largest difference in the mean annual incidence rate between non-austrians and Austrians was in the age-group 15-24 years, with a rate ratio of 16.57, followed by the age group 5-14 years, with a incidence rate ratio of 12.55. The average age in non-austrians was 53.7 years (range: 0-90 yeas ), and 69.9 years for Austrians (range: 0-99 years). The distribution of the age-group specific mean annual incidence rate by sex demonstrated that the greatest difference between female non-austrians and female Austrians was in the age group 5-14 years (incidence rate ratio: 11.69). Male non-austrians differed mostly from male Austrians in the age group 15-24 years (incidence rate ratio: 22.78) [Table 15]. Results 61

Table 14: Mean annual age-group-specific TB incidence rates [IR] in both sexes, and mean annual age- and sex-specific TB IRs in Austrians and non-austrians; incidence rate ratios [RR] of age groups to the age group 5-14 as the reference age group), and agegroup specific RRs of males to females in Austrians and non-austrians, 1997-2006 Age groups 0-4 5-14 15-24 25-34 35-44 45-54 55-64 64+ Age-group IR Both sexes 4.09 RR age-group (x) : age-group ref (95%CI) 3.05 (2.39-3.89) Austrian Age-group IR Female Age-group IR Male 4.88 3.33 1.34-1.52 1.17 4.18 7.45 11.07 15.20 14.15 21.78 3.12 (2.53-3.85) 5.56 (4.57-6.77) 8.26 (6.82-10.00) 11.34 (9.37-13.72) 10.56 (8.72-12.8) 16.25 (13.47-19.6) 4.41 3.96 6.50 8.39 7.24 14.90 7.34 23.38 7.51 21.33 15.80 31.62 RR m : f (95%CI) 0.68 (0.49-0.95) 0.77 (0.53-1.12) 0.9 (0.73-1.11) 1.29 (1.12-1.49) 2.06 (1.84-2.31) 3.18 (2.82-3.59) 2.84 (2.51-3.22) 2.00 (1.85-2.16) p Age-group IR Both sexes 0.02 25.17 RR age-group (x) : age-group ref (95%CI) 1.50 (1.18-1.89) Non-Austrian Age-group IR Female Age-group IR Male 25.84 24.55 0.17 16.82-17.75 15.95 0.31 69.26 <0.001 59.71 <0.001 48.68 <0.001 46.16 <0.001 52.26 <0.001 59.58 4.12 (3.46-4.89) 3.55 (3.00-4.20) 2.89 (2.43-3.44) 2.74 (2.29-3.29) 3.11 (2.54-3.80) 3.54 (2.87-4.37) 47.86 90.30 43.00 77.23 36.04 58.41 26.70 60.60 39.49 62.04 63.78 54.67 RR m : f (95%CI) 0.95 (0.67-1.34) 0.90 (0.66-1.23) 1.89 (1.61-2.21) 1.80 (1.58-2.05) 1.62 (1.38-1.91) 2.27 (1.84-2.80) 1.57 (1.20-2.06) 0.86 (0.64-1.14) p 0.77 0.5 <0.001 <0.001 <0.001 <0.001 <0.001 0.29 Results 62

Table 15: Mean annual age-group specific incidence rate (IR) in both sexes in Austrians (A) and non-austrians (non-a), mean annual age-group and sex-specific TB IR in A and non-a, and rate ratio (RR) in non-a and A, 1997-2006 Age groups Age-group IR Both sexes A Age-group IR Both sexes non-a 0-4 4.09 25.17 5-14 1.34 16.82 15-24 4.18 69.26 25-34 7.45 59.71 35-44 11.07 48.68 45-54 15.20 46.16 55-64 14.15 52.26 64+ 21.78 59.58 Avg. age (range) 69.9 (0-99) 53.7 (0-90) RR non-a : A (95%CI) 6.16 (4.82-7.87) 12.55 (9.80-16.11) 16.57 (14.56-18.87) 8.02 (7.28-8.81) 4.40 (4.00-4.83) 3.04 (2.73-3.74) 3.69 (3.20-4.25) 2.74 (2.35-3.17) p Age-group IR Female A Age-group IR Female non-a <0.01 4.88 25.84 <0.01 1.52 17.75 <0.01 4.41 47.86 <0.01 6.50 43.00 <0.01 7.24 36.04 <0.01 7.34 26.70 <0.01 7.51 39.49 <0.01 15.80 63.78 RR non-a : A (95% CI) 5.30 (3.76-7.43) 11.69 (8.31-16.52) 10.86 (8.93-13.24) 6.62 (5.68-7.70) 4.98 (4.21-5.88) 3.64 (2.92-4.50) 5.26 (4.05-6.74) 4.04 (3.29-4.91) p Age-group IR Male A Age-group IR Male non-a <0.01 3.33 24.55 <0.01 1.17 15.95 <0.01 3.96 90.30 <0.01 8.39 77.23 <0.01 14.90 58.41 <0.01 23.38 60.60 <0.01 21.33 62.04 <0.01 31.62 54.67 RR non-a : A (95%CI) 7.37 (5.11-10.64) 13.62 (9.44-19.80) 22.78 (19.08-27.30) 9.21 (8.13-10.43) 3.92 (3.49-4.40) 2.59 (2.29-2.93) 2.91 (2.44-3.45) 1.73 (1.36-2.17) p <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Results 63

Age-adjusted annual incidence of TB in Austrians and non- Austrians by province, 1997-2006 From 1997 to 2006, the age-adjusted annual TB incidence in Austrians decreased significantly in all 9 provinces of Austria. The highest AAPC was observed in the province of Carinthia with an AAPC of -15.4% (95%CI: -18.8% to -11.8%), followed by the province of Upper Austria (AAPC: - 11.4%; 95%CI: -13.7% to - 9.1%), Vorarlberg (AAPC: -10.4%; 95%CI: -13.6% to -7.1%), and then followed in descending order by the provinces of Burgenland, Lower Austria, Salzburg, Styria and Vienna. The lowest AAPC was observed in the province Tyrol with -5% (95%CI: -8.3% to -1.6%) [Table 16]. The age-adjusted annual TB incidence in non-austrians decreased significantly only in the province Vorarlberg (AAPC: - 6.9%; 95%CI: -12% to -1.5%) [Table 17]. Results 64

Table 16: Age-adjusted annual incidence of TB in Austrians by province, 1997-2006, and the annual average percentage change (AAPC) by province Age-adjusted annual incidence of TB in Austrians Year Upper Lower Vienna Austria Austria Salzburg Carinthia Styria Tyrol Vorarlberg Burgenland 1997 16,26 14,50 8,24 9,14 15,61 12,76 7,80 9,74 11,95 1998 15,84 11,93 9,51 10,57 13,33 9,64 10,60 11,24 8,79 1999 14,77 12,01 8,49 8,54 10,43 9,27 8,89 8,38 6,47 2000 13,27 8,40 9,25 9,08 10,04 8,23 10,26 8,18 7,16 2001 14,21 7,92 7,08 11,48 9,80 6,90 9,15 7,51 6,65 2002 13,37 8,42 6,35 6,03 6,09 8,16 7,13 7,11 4,64 2003 12,16 5,61 4,66 4,66 6,40 7,43 7,57 6,52 6,10 2004 9,84 5,83 5,22 6,23 3,50 8,57 5,97 4,97 6,61 2005 10,43 5,63 4,09 6,22 4,08 5,72 6,82 3,48 5,06 2006 9,52 4,77 4,95 6,30 4,10 5,82 6,04 4,67 3,13 AAPC (95%CI) -5.9% (-7.4% to -4.5%) -11.4% (-13.7% to -9.1%) -8.7% (-12% to -5.4%) -6.6% (-11.6% to -1.3%) -15.4% (-18.8% to -11.8%) -6.6% (-9.7% to -3.4%) -5% (-8.3% to-1.6%) -10.4% (-13.6% to -7.1%) -9.6% (-14.2% to -4.8%) Table 17: Age-adjusted annual incidence of TB in non-austrians by province, 1997-2006 and the annual average percentage change (AAPC) by province. Age-adjusted annual incidence of TB in non-austrians Year Upper Vienna Austria Lower Austria Salzburg Carinthia Styria Tyrol Vorarlberg Burgenland 1997 57,71 78,41 45,38 41,35 58,73 56,57 22,44 56,68 52,30 1998 55,89 63,35 35,28 35,75 23,54 54,51 20,23 41,92 55,64 1999 52,46 53,42 33,76 30,14 53,48 70,93 33,13 39,69 29,30 2000 53,93 49,88 30,86 63,24 61,90 70,70 21,28 30,93 57,49 2001 52,82 37,40 34,33 25,81 49,04 35,24 16,97 27,21 5,48 2002 50,27 47,02 32,50 26,83 50,68 60,87 10,24 25,41 51,78 2003 54,07 48,63 38,14 47,02 31,53 46,84 11,04 33,52 55,90 2004 71,67 39,91 57,30 26,59 39,85 75,25 24,31 41,91 70,24 2005 59,18 66,33 64,70 31,45 36,50 63,10 14,19 26,38 51,21 2006 41,87 52,23 47,45 20,70 66,48 49,12 16,37 20,85 14,50 AAPC (95%CI) -0.9% (-4.1% to 3.1%) -2.8% (-8.1% to 2.9%) 4.9% (-0.9% to 11%) -5.1% (-12% to 2.6%) 0.4% (-8.1% to 9.8%) -0.4% (-6.4% to 6%) -5.5% (-13.3% to 3%) -6.9% (-12% to -1.5%) -3.3% (-22% to 19.9%) Results 65

Age-group specific annual incidence of TB in non-austrians and Austrians, 1997-2006 Between the years 1997 and 2006 non-austrians in the age-group 5-14 years were the least affected, except in the years 2001 to 2003, when the TB incidence in the 0-4 age-group fell slightly below the incidence of the 5-14 age-group. Age group 0-4: After an increase in the TB incidence from 25.1/100,000 population in 1997 to 37.9 in 2000 (IPC: 51%, p=0.2), the incidence decreased from 37.9 to 11.8 (IPC: 68.9%, p<0.01) between 2000 and 2001. After 2001 the incidence increased annually by 5.9/100,000 population until 2006 (slope=5.9, 95%CI: 2.5 to 9.2, R 2 =0.86, P<0.001). Age group 15-24: After a decrease from 62.1 in 1997 to 47.1 in 1999 (IPC: 24%, p=0.15), from 1999 to 2005 the incidence increased by 7.7/100,000 population per year (slope=7.7, 95%CI: 3.6 11.8, R 2 =0.83, P<0.001). Age group 25-34: From 1997 to 2001 age-group specific incidence decreased from 60.7 to 47.5 (IPC: 21.7%, p=0.1), after which, the incidence increased to 86.5 in 2004 (IPC: 82%, p<0.01), and then the incidence decreased to 42.6 in 2006 (IPC: 48.4%, p<0.01). Age group 35-44: The incidence decreased slightly from 53.1 in 1997 to 42.8 in 1998 (IPC: 19.4%, p=0.24), after which, the incidence increased to 54.5 in 2000 (IPC: 27.3%, p=0.18), and then decreased to 36.7 in 2001 (IPC: 32.7%, p=0.03). From 2001 to 2004 the incidence increased linearly by 5.3/100,000 population per year (slope=5.3, 95%CI: 1.2 9.5, R 2 =0.94, P=0.03) and then decreased slightly until 2006. Age group 45-54: The incidence decreased slightly from 64.8 in 1997 to 48.8 in 1998 (IPC: 24.7%, p=0.17), after which, the incidence increased to 62.1 in 1999 (IPC: 27.3%, p=0.24), and then decreased to 35.6 in 2001 (IPC: 42.7%, p=0.01). From 2001 to 2003 the incidence had an IPC of 43.5% (P=0.16), and then decreased slightly until 2006. Results 66

Age group 55-64: From 1997 to 2006 the incidence in this age-group decreased significantly with an AAPC of 11.8% (95%CI: -16.1% to -7.2%). The incidence increased slightly from 35.0 in 2004 to 46.6 in 2005 (IPC: 33.1%, p=0.3). Age group 65 years: From 1997 to 2006 the incidence in this age-group had an AAPC of -6.1% (95%CI: -11.4% to -0.5%). After a considerable increase in the incidence between 1997 and 1998, there was a decrease up to 2002. Between 2002 and 2004 the incidence increase from 41.4 to 59.0 was significant (IPC: 42.5%, p<0.01); from then the Tb incidence decreased slightly in this age group [Figure 18a-18c, see annex Table 18]. Among Austrians aged 15 years the incidence of TB decreased significantly during the study period with a minimum AAPC of 6.2% in age-group 55-64 years, and a maximum AAPC of 11.6% in the age-group 35-39 years [see annex Table 18]. Results 67

Figure 18a: Age-group specific annual incidence of TB in non-austrians, age-groups 0-4, 5-14, 15-24, 1997-2006 Incidence (cases /100,000 population) 120 90 60 30 0 1997 1998 1999 Annual incidence of TB among non-austrians in age group 0-4 years Annual incidence of TB among non-austrians in age group 5-14 years Annual incidence of TB among non-austrians in age group 15-24 years 2000 2001 2002 Year 2003 2004 2005 2006 Figure 18b; Age-group specific annual incidence of TB in Non-Austrians, age-groups 25-34, 35-44, 45-54; 1997-2006 Incidence (cases /100,000 population) 120 90 60 30 0 Annual incidence of TB among non-austrians in age group 25-34 years Annual incidence of TB among non-austrians in age group 35-44 years Annual incidence of TB among non-austrians in age group 45-54 years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 18c; Age-group specific annual incidence of TB in Non-Austrians, age-groups 55-64, 65 plus; 1997-2006 120 Incidence (cases /100,000 population) 90 60 30 0 Annual incidence of TB among non-austrians in age group 55-64 years Annual incidence of TB among non-austrians in age group older or equal to 65 years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 68

MDR TB in Austria, 1997-2006 Distribution of resistance pattern to anti-tb drugs in cases of definite TB in Austria, 1997-2006 Between 1997 and 2006, the Mycobacteria tuberculosis complex (MTC) was isolated in 7,009 TB cases. Of these culture confirmed cases, 6,485 (92.52%) were infected with MTC that was sensible to any of the first line anti TB drugs (Isoniazid, Rifampicin, Pyrazinamid, Ethambutol and Streptomycin), and 341 cases (4.87%) were infected with MTC resistant to only one of the first line anti- TB drugs (i.e. defined as monoresistance). Of these 168 cases (49.3%) were infected with MTC resistant to Isoniazid. In 99 cases (1.41%), the isolate of MTC showed polyresistance to anti-tb drugs (defined as resistant to more than one first line drug other than both Isoniazid and Rifampicin), and in 84 cases (1.2%) the MTC isolate was resistant to at least both Isoniazid and Rifampicin (defined as multi-drug resistance). Figure 19 displays the annual distribution of cases of resistance to anti-tb drugs by type of resistance in Austria from 1997 to 2006. Figure 19: Annual distribution of type of resistance to anti-tb drugs in Austria, 1997-2006 8% 7% Monoresistance Polyresistance MDR-TB 6% Proportion (%) 5% 4% 3% 2% 1% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 69

Figure 20: Proportion of MDR-TB cases among TB cases with antimicrobial susceptibility testing in Austria, 1997-2006 10% Proportion of MDR-TB among all TB cases with antimicrobial susceptibility testing in Austria 8% Proportion (%) 6% 4% 2% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 20 displays the annual proportion of multi-drug resistant tuberculosis (MDR-TB) cases among the TB cases with antimicrobial susceptibility testing. In 2004 the proportion of MDR-TB cases was 5 times higher compared to the average proportion of MDR cases from the previous 6 years, 1997 to 2002 (average p= 0.6%). Results 70

Crude annual incidence of MDR-TB in Austria, 1997-2006 From 1997 to 2006, a total of 84 multi-drug resistant tuberculosis (MDR-TB) cases were reported in Austria, including 59 males (70.2%), and 72 non- Austrians (85.7%). The crude annual incidence of MDR-TB increased slightly from 1997 to 1998 (0.05 to 0.08/100,000), remained stable in 1999, and declined to 0.04/100,000 in 2002. From 2002 the MDR-TB incidence increased steeply from 0.04/100,000 to a peak of 0.23/100,000 in 2004 (p<0.01). Between 2004 and 2006 the MDR-TB incidence decreased by 50% (0.23 to 0.12/100,000) [Figure 21; see annex Table 19]. Figure 21: Annual crude incidence of MDR-TB in Austria, 1997-2006 0,25 0,23 Incidence (cases/100,000 population) 0,20 0,15 0,10 0,05 0,05 0,08 0,08 0,06 0,06 0,04 0,15 0,17 0,12 0,00 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 71

Annual sex-specific incidence of MDR-TB in Austria, 1997-2006 There was no difference in the risk of MDR-TB between males and females in the years 1997-2003 and 2006. In 2004 the risk of MDR-TB was 2.97 times, and in 2005, 13.97 times higher in males compared to females [Figure 21, see annex Table 19]. Figure 21: Annual sex-specific incidence of MDR-TB in Austria, 1997-2006 0,40 Male Incidence (cases/100,000 population) 0,35 0,30 0,25 0,20 0,15 0,10 0,05 Female 0,00 1997 1998 1999 2000 2001 Year 2002 2003 2004 2005 2006 Mean annual age-group and sex-specific incidence rates of MDR- TB in Austria, 1997-2006 The highest mean annual age-specific incidence rate of MDR-TB was in the age group 25-34 years with 0.25/100,000 pyr, and the lowest mean annual incidence rate was 0.01/100,000 pyr in the age group 5-14 years. The risk of MDR-TB in the age group 15-24 years was 21.65 times higher compared to the reference age group age group 5-14 year, in the age group 25-34 years 23.25 times higher and in the age group 35-44 years 9.91 times higher [Figure 22, see annex Table 20]. Results 72

The average age of MDR-TB cases across the 10-year period was 31.6 years (min: 0 year, max: 72 years). Figure 22: Mean annual age- and sex-specific MDR-TB incidence rates, 1997-2006 0,35 0,31 0,33 Mean annual age-specific MDR-TB incidence rate in males Mean annual age-specific MDR-TB incidence rate in total Mean annual age-specific MDR-TB incidence rate in females 0,30 Incidence rate ( cases/ 100,000 pyr) 0,25 0,20 0,15 0,10 0,05 0,00 0,10 0,07 0,05 0-4 0,02 0,01 0, 00 5-14 0,15 15-24 0,23 0,25 0,17 25-34 0,16 0,10 0,05 35-44 0,08 0,06 0,04 45-54 0,09 0,04 0,00 55-64 0,08 0,03 0,00 64+ Age groups Results 73

MDR-TB in Austrians and in non-austrians, 1997-2006 Figure 23 demonstrates the crude proportion of MDR-TB cases among all TB cases with antimicrobial susceptibility testing (blue line) and the proportion of MDR-TB cases stratified into Austrians and non-austrians (bars). In the last decade the proportion of MDR-TB cases among culture confirmed TB cases - regardless of the country of origin - ranged between a minimum of 0.4% (in 2002) and a maximum of 3% (in 2004), with a median of 0.75%. Between 1997 and 2002 the crude proportion of MDR-TB cases remained stable at a low level (range: 0.4%-0.8%), followed by a steep increase towards 2004 which was exclusively due to a significant rise of MDR-TB cases in non-austrians [trend line in Figure 23]. Figure 23: Proportion of MDR TB cases among culture positive TB cases per year; proportion of MDR-TB cases among culture positive TB cases in Austrians, proportion of MDR TB cases among culture positiv TB cases in non-austrians 5% 4% % of MDR-TB case in all TB cases in Austrians % of MDR-TB case in all TB cases in non-austrians % of MDR-TB case in all TB cases Proportion (%) 3% 2% 1% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 74

Multidrug resistant (MDR)-TB in Austrians: The annual incidence of MDR-TB in Austrians ranged between a minimum of 0 and a maximum of 0.03 per 100,000 Austrians with a 10-year average of 0.02/100,000 Austrians. No change over time was observed [Figure 24, see annex Table 21]. Figure 24: Annual incidence of MDR-TB in non-austrians and Austrians 3,0 2,5 Annual incidence of MDR-TB in Austrians Annual incidence of MDR-TB in non-austrians Incidence (cases/100,000 population) 2,0 1,5 1,0 0,5 0,0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year MDR-TB in non-austrians: There was slight increase in the annual MDR-TB incidence in non-austrians from 0.29/100,000 population in 1997 to 0.72/100,000 population in 1999. The proportion of MDR-TB cases in all cases tested among non-austrians increased from 0.5% in 1997 to 1.5% in 1999 (p> 0.05). The MDR-TB incidence decreased significantly from 0.72/100,000 population in 1999 to 0.40/100,000 population in 2002 by an AAPC of -18.5% (95%CI: -31.5% to -3.1%). From 2002 to 2004 the incidence increased 6-fold (from 0.40 to 2.45/100,000; p<0.001), and then decreased between 2004 and 2006.The IPC was 55% (from 2.45 to 1.1/100,000; p=0.04) [Figure 24]. The year 2004 had the highest proportion of MDR-TB cases of all cases tested among non-austrians (proportion= 4.3%) within the 10-year period of analysis, whereas no cases of MDR-TB in Austrians were reported that year [Figure 23, see annex Table 21]. Results 75

Mean annual incidence rate of MDR-TB in Austrians and non-austrians, 1997-2006 The mean annual incidence rate of MDR-TB in non-austrians was significantly higher than the mean annual incidence rate in Austrians (non-austrians: 0.98/100,000 pyr, Austrians: 0.02/100,000 pyr; p<0.001). There was no difference in the mean annual incidence rate of MDR-TB between the two 5-year periods in Austrians (1997-2001: 0.02/100,000 pyr versus and 2002-2006: 0.01/100,000 pyr). The mean annual incidence rate of MDR-TB in non-austrians differed significantly between these two 5-year periods: 0.52/100,000 pyr in the 5-year period 1997-2001 versus 1.39/100,000 pyr in the 5-year period 2002-2006 (p<0.001). Results 76

Annual sex-specific incidence of MDR-TB in non-austrians, 1997-2006 Figure 25: Annual sex-specific incidence of MDR-TB in non-austrians, 1997-2006 Annual MDR-TB incidence in female non-austrians 4,0 Annual MDR-TB incidence in male non-austrians 3,5 3,0 Incidence (cases/100,000 population) 2,5 2,0 1,5 1,0 0,5 0,0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year There was no difference between males and females in the annual incidence of MDR-TB in non-austrians, except in the year 2005, when male non-austrians had a relative MDR-TB risk of 11.13 compared to female non-austrians [Figure 25, see annex Table 22]. Between 1997 and 2002 there was no significant change in the annual incidence of MDR-TB among male non-austrians (AAPC: 1.67%; 95%CI: -24.2% to 36.4%; P> 0.05). Between 2002 and 2004 the 1-year incidence of MDR-TB in male non- Austrians increased annually by 1.4 cases/100,000 population (slope: 1.4; 95%CI: 0.4-2.3; R 2 =0.99; p=0.03) to 3.51/100,000 population, followed by a 66% decrease between 2004 and 2006. Results 77

Year Annual number of MDR-TB cases in non-austrians by country of origin, 1997-2006 Table 23: Annual number of cases of MDR-TB in non-austrians by country of origin, 1997-2006 MDR-TB cases Former Yugoslavia Albania Turkey Romania Russian Federation Other European countries 1 Africa 2 S-/E-Asia, West- Pacific 3 1997 0 0 1 (50%) 0 0 0 1 (50%) 0 2 1998 1 (25%) 0 1 (25%) 0 0 1 (25%) 0 1 (25%) 4 1999 1 (20%) 0 1 (20%) 1 (20%) 0 1 (20%) 0 1 (20%) 5 2000 0 0 1 (25%) 1 (25%) 0 0 1 (25%) 1 (25%) 4 2001 0 1 (33%) 0 0 1 (33%) 0 0 1 (33%) 3 2002 1 (33%) 0 0 0 0 1 (33%) 0 1 (33%) 3 2003 0 0 0 1 (10%) 5 (50%) 2 (20%) 0 2 (20%) 10 2004 0 0 0 0 9 (47%) 7 (37%) 0 3 (16%) 19 2005 1 (8%) 0 1 (8%) 2 (15%) 5 (38%) 4 (31%) 0 0 13 2006 0 0 0 0 4 (44%) 3 (33%) 0 2 (22%) 9 total 4 (6%) 1 (1%) 5 (7%) 5 (7%) 24 (33%) 19 (26%) 2 (3%) 12 (17%) 72 Total Country of origin of non-austrians with MDR-TB: 1 Armenia, Azerbaijan, Georgia, Moldavia, Ukraine (high TB incidence Eastern European countries). 2 Nigeria, unknown African country. 3 China, India, Mongolia and Vietnam. Between 1997 and 2006, 24 out of 72 (33%) MDR-TB cases in non-austrians occurred in those who immigrated from the Russian Federation, 19 (26%) MDR TB patients originated (26%) in other Eastern European TB-endemic countries such as Armenia, Azerbaijan, Georgia, Moldavia and the Ukraine, 12 cases (17%) occurred in non-austrians who originated in countries in South-East Asia and the WHO Western Pacific Region such as China, India, Mongolia and Vietnam [Table 23]. From 1997 to 2000 the countries of origin of non-austrian MDR-TB cases were Turkey, former Yugoslavia, Albania, Nigeria, Asia, Romania, Armenia, Azerbaijan, Georgia, Moldavia, Ukraine. Results 78

In 2001 MDR cases in non-austrians having immigrated from the Russian Federation accounted for 33% (1 out of 3 cases), in 2003 for 50% (5/ 10 cases), in 2004 for 47% (9/ 19 cases), in 2005 for 38.5% (5/ 13 cases), and in 2006 for 44% (4 out of 9 cases); MDR-TB cases in non-austrians originating in Romania accounted for 33.3% in 2002 (1 out of 3 cases), 30% in 2003 (3 out of 10 cases), 36.8% in 2004 (7 out of 19 cases), 46.2% in 2005 (6 out of 13 cases), and 33.3% in 2006 (3 out of 9 cases) and MDR cases in non-austrians originating in other high TB incidence European countries as Armenia, Azerbaijan, Georgia, Moldavia, Ukraine accounted for 33% (1/3 cases) in 2002, 20% (2/10 cases) in 2003, 37% (7/19 cases) in 2004, 31% (4/13 cases) in 2005 and 33% (3/9 cases) in 2006; non-austrian MDR TB cases having immigrated from the Asian countries China, India, Mongolia and Vietnam accounted in 2001 and 2002 each for 33% of all non-austrian MDR-TB cases (1/3 cases each), for 20% (2/10) in 2003, for 16% (3/19) in 2004, and for 22% (2/9) in 2006. Annual number of MDR-TB cases in Austrians and non-austrians by province of occurrence, 1997-2006 From 1997 to 2006, 34 out of 84 (40.5%) MDR-TB cases were reported from Vienna, 15 cases (17.9%) were reported from the province of Upper Austria, 12 cases (14.3%) from the province Lower Austria, 2 cases from Burgenland, 2 cases from Carinthia, 3 cases from Salzburg, 7 cases from Styria, 7 cases from Tyrol, and 2 cases from Vorarlberg [Table 24]. MDR-TB cases in non-austrians that occurred between 1997 and 2006 accounted for 30 out 34 cases in Vienna, 13 out of 15 cases in Upper Austria, 10 out of 12 cases in Lower Austria, 3 out of 3 cases in Salzburg, 1 out of 2 cases in Carinthia, 6 out of 7 cases in Styria, 6 out 7 cases in Tyrol, 1 out of 2 cases in Vorarlberg, and for 2 out of 2 cases in Burgenland. The provincial distribution of MDR-TB cases in non-austrians dominated by refugees is due to the location of the two refugee camps; one located in Upper Austria and the other one located in Vienna. Results 79

Table 24: Annual number of MDR-TB cases in Austrians (A) and non-austrians (nona) by province of occurrence, 1997-2006 MDR-TB cases Year Vienna Upper Austria Lower Austria Salzburg Carinthia Styria Tyrol Vorarlberg Burgenland A nona A nona A nona A nona A nona A nona A nona A nona A nona 1997 1 1 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 1998 0 2 1 1 0 1 0 0 0 0 0 0 1 0 0 0 0 0 1999 0 2 0 0 1 1 0 1 0 0 0 0 0 1 0 0 0 0 2000 1 1 0 1 0 0 0 0 0 0 0 1 0 1 0 0 0 0 2001 1 1 0 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 0 1 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2003 0 3 0 3 0 1 0 1 1 0 0 0 0 1 1 0 0 1 2004 0 7 0 2 0 3 0 0 0 1 0 3 0 2 0 1 0 0 2005 0 7 1 1 0 2 0 1 0 0 0 1 0 0 0 0 0 1 2006 1 5 0 2 0 1 0 0 0 0 0 0 0 1 0 0 0 0 Results 80

TB-outcome, 1997-2006 TB cases by treatment outcome In 7,450 out of 10,904 cases of TB (68.3%) from 1997 to 2006, for which treatment outcome data were available, 1,125 TB cases (10.3%) were listed as cured [Table 25]. Figure 26 shows the annual distribution of TB cases by treatment outcome, from 1997 to 2006. Table 25: Annual number of TB cases by treatment outcome (%), 1997-2006 Cases n/n (%) Year Died Cured Treatment completion Final report pending Lost to follow-up Missing data Total 1997 173 (11,69) 85 (5,7) 1022 (69,1) 0 (0) 83 (5,6) 117 (7,9) 1480 1998 176 (12,91) 181 (13,3) 853 (62,6) 0 (0) 76 (5,6) 77 (5,6) 1363 1999 177 (13,74) 173 (13,4) 786 (61,0) 0 (0) 97 (7,5) 55 (4,3) 1288 2000 155 (12,66) 126 (10,3) 840 (68,6) 0 (0) 84 (6,9) 19 (1,6) 1224 2001 122 (11,25) 117 (10,8) 723 (66,7) 0 (0) 85 (7,8) 37 (3,4) 1084 2002 123 (11,39) 123 (11,5) 706 (65,4) 0 (0) 91 (8,4) 37 (3,4) 1080 2003 104 (10,51) 96 (9,7) 672 (67,9) 0 (0) 91 (9,2) 27 (2,7) 990 2004 88 (8,16) 101 (9,4) 712 (66,0) 34 (3,2) 97 (9,0) 47 (4,4) 1079 2005 83 (8,24) 75 (7,4) 653 (64,8) 50 (5,0) 99 (9,8) 47 (4,7) 1007 2006 96 (10,65) 48 (5,3) 483 (53,6) 88 (9,8) 57 (6,3) 129 (14,3) 901 Total 1297 (11,28) 1125 (9,8) 7450 (64,8) 172 (1,5) 860 (7,5) 592 (5,1) 11496 Figure 26: Annual distribution of TB cases by treatment outcome (%) in Austria, 1997-2006 100% 90% 80% 70% Proportion (%) 60% 50% 40% 30% Missing data on treatment outcom Treatment withdrawal Only interim report Died Cured Treatment completed 20% 10% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 81

Note: The following analyses on treatment outcome were restricted to cases for which data on treatment outcome were available Annual proportion of cured TB cases, 1997-2006 The proportion of cured TB cases increased steeply from 1997 to 1998 (from 5.7% (85/1,480) to 13.3% (181/1,363)), decreased to 10.3% (126/1,224) in 2000, and increased again to 11.4% (123/1,080) in 2002. After then, the proportion of cured TB cases decreased annually by 1.4% until 2006 (slope= - 0.014, 95%CI: -0.02 to -0.004, R 2 =0.96, p<0.01 [Figure 27]. There was no difference in the proportion of cured TB cases between Austrians and non- Austrians (10-year cumulative proportion 9.6% vs 10.1%) [Figure 28, see annex Table 26]. There was also no difference in the proportion of cured TB cases between males and females (cumulative 10.2% versus 9.1%) [Figure 29]. Figure 27: Annual proportion of cured TB cases among TB cases in Austria, 1997-2006 (N tota l= 7,450, excluding cases with missing data on treatment outcome) 16% Proportion of TB cases with missing data on treatment outcome Proportion of cured TB cases 14% 13.3% 13.4% 12% 11.4% Proportion 10% 8% 6% 5.7% 10.3% 10.8% 9.7% 9.4% 7.4% 5.3% 4% 2% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 82

Figure 28: Annual proportion of cured TB cases among Austrian cases and non-austrian cases, 1997-2006 (N total Austrians = 7,612; N total non-austrians = 3,323, excluding cases with missing data on treatment outcome) 20% Proportion of cured TB cases among all TB cases w ith Austrian citizenship Proportion of cured TB cases among all TB cases w ithout Austrian citizenship 15% Proportion 10% 5% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 29: Annual proportion of cured TB cases among male cases and female cases, 1997-2006 (N total male = 6,888; N total female = 4,047, excluding cases with missing data on treatment outcome) 16% 14% Proportion of cured TB cases among all female TB cases Proportion of cured TB cases among all male TB cases 12% Proportion 10% 8% 6% 4% 2% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 83

Annual proportion of cases with treatment completion, 1997-2006 Note: Since at the time of analysis only partial data on treatment completion for TB cases reported in 2006 were available, data from that year has not been included in this analysis. Between 1997 and 2006 a slight decrease in the annual proportion of TB cases with treatment completion (from 69.1% to 64.8%) was observed (slope= 0, 95%CI: -0.01 to 0.01, R 2 =0.004, p=0.87) [Figure 30; see annex Table 26]. Figure 30: Annual proportion of TB cases with treatment completion among TB cases in Austria, 1997-2005 100% 90% 80% Proportion 70% 60% 50% 40% 69.1% 62.6% 61.0% 68.6% 66.7% 65.4% 67.9% 66.0% 64.8% 30% 20% 10% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 Year Results 84

From 1997 to 2005 the annual proportion of TB cases with treatment completion in non-austrian cases was stable. In comparison, the annual proportion of TB cases with treatment completion in Austrian cases increased by 1% (slope= 0.01, 95%CI: 0.001 to 0.02, R 2 =0.48, p=0.037) [Figure 31]. Figure 31: Annual proportion of TB cases with treatment completion among Austrian cases and non-austrian cases, 1997-2005 100% 90% Annual proportion of TB cases with treatment completion in Austrians Annual proportion of TB cases with treatment completion in non-austrians 80% 70% Proportion 60% 50% 40% 30% 20% 10% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 Year Results 85

Annual proportion of TB cases lost to follow-up, 1997-2006 Throughout the study period, the annual proportion of TB cases lost to follow-up was substantially higher among non-austrian cases compared to Austrian cases (cumulative 10-year proportion: 17.3% vs 3.2%, p< 0.001) [Figure 32, see annex Table 26]. Figure 32: Annual proportion of TB cases lost to follow-up among Austrian cases and non-austrian cases, 1997-2006 30% 25% Proportion of lost to follow -up TB cases among all TB cases w ith Austrian citizenship Proportion of lost to follow -up TB cases among all TB cases w ithout Austrian citizenship 20% Proportion 15% 10% 5% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 86

Annual number of MDR-TB cases by treatment outcome, 1997-2006 Table 27 shows the annual number of MDR-TB cases in Austrians and non- Austrians by treatment outcome, from 1997 to 2006. Table 27: Annual number of MDR-TB cases in Austrians (A) and non-austrians (nona) by treatment outcome, 1997-2006 MDR-TB cases Year Died Cured Treatment completion Lost to follow-up A NonA A NonA A NonA A NonA 1997 1 0 1 0 0 1 0 1 1998 0 1 0 0 2 1 0 2 1999 1 1 0 1 0 1 0 2 2000 0 0 0 2 1 2 0 0 2001 1 0 0 0 1 3 0 0 2002 0 1 0 1 0 1 0 0 2003 2 0 0 0 0 3 0 7 2004 0 0 0 5 0 6 0 4 2005 0 0 0 2 0 4 0 4 2006 0 0 0 0 1 1 0 1 Results 87

Mortality of TB in Austria, 1997-2006 From 1997-2000 The crude mortality from TB in Austria ranged between 0.67 deaths per 100,000 population in 1997, to 0.82 deaths per 100,000 population in 2000. The annual average percentage change was not significant (mortality percentage change 22.4%, p=0.24). From 2000 to 2006 mortality from TB decreased annually by 7 deaths per 10,000,000 population (slope= -0.07, 95%CI: -0.13 to -0.01, R 2 =0.64, p<0.05) [Figure 33]. The average time period between date of diagnosis and date of death from TB was 3.5 months (range 0-81 months). The age-standardized mortality did not change significantly between 1997 and 2000 (from 0.43 deaths per 100,000 in 1997, to 0.53 deaths per 100,000 population in 2000, mortality percentage change: 23.4%; p=0.31). From 2000 to 2006 the mortality decreased by 5 deaths per 10,000,000 persons per year (slope= -0.05, 95%CI: -0.08 to -0.01, R 2 =0.67, p<0.05) [Figure 34]. Figure 33: Crude annual mortality from TB in Austria, 1997-2006 1,0 Mortality (deaths/100,000 population) 0,8 0,6 0,4 0,2 0.67 0.65 0.75 0.82 0.42 0.55 0.53 0.36 0.28 0.35 0,0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Results 88

Figure 34: Age-standardised annual mortality from TB in Austria, 1997-2006 0,6 Mortality (deaths/100,000 population) 0,5 0,4 0,3 0,2 0,1 0.43 0.44 0.50 0.53 0.29 0.37 0.37 0.23 0.18 0.22 0,0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 35: Age standardised annual mortality from TB in Austrians compared to non- Austrians, 1997-2006 1,20 Age-adjusted mortality of TB among Austrians Age-adjusted mortality of TB among non-austrians 1,00 Mortality (deaths/100,000 population) 0,80 0,60 0,40 0,20 0,00 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year The age-standardized annual mortality from TB in non-austrians was significantly higher compared to Austrians in the years 1997, 1999, 2001, 2002, 2003, 2005 and 2006 [Figure 35, Table 28]. Table 28 shows the difference in age- Results 89

standardized TB mortality between Austrians and non-austrians, expressed as a ratio. The average time period between date of diagnosis and date of death due to TB did not differ significantly (Austrians: 3.5 months vs. non-austrians: 4.1 months, p=0.38). Table 28: Age-standardized annual mortality from TB in Austrians and non-austrians, 1997-2006 Year Mortality (per 100.000 population) Austrians Mortality (per 100.000 population) non-austrians Mortality ratio non-austrians: Austrians (95%CI) p 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 AAPC (95%CI) 0.40 0.82 0.43 0.50 0.47 1.12 0.53 0.40 0.25 0.88 0.33 0.71 0.34 0.63 0.21 0.32 0.17 0.45 0.15 0.86-11.7% -3.9% (-16.8% to -6.3%) (-12.8% to 7.3%) 2.05 (1.4-2.99) <0.01 1.16 (0.77-1.75) 0.47 2.38 (1.7-3.35) <0.01 0,75 (0.5-1.14) 0.18 3.52 (2.26-5.49) <0.01 2.15 (1.42-3.25) <0.01 1.85 (1.22-2.81) <0.01 1.52 (0.88-2.64) 0.13 2.65 (1.52-4.62) <0.01 5.73 (3.31-9.92) <0.01 Between 1997 and 2006 the age-adjusted annual mortality from TB in Austrians decreased by 0.04 deaths per 100,000 population per year (slope= -0.04, 95%CI: -0.06 to -0.02, R 2 =0.69, p<0.01). The age-adjusted annual mortality in non-austrians from 1997 to 2006 was at its highest in 1999, with a total of 1.12 deaths per 100,000 non-austrians. A second and a third peak occurred in 2001 and 2006 with 0.88 deaths and 0.86 deaths per 100,000 [Figure 35, Table 28]. The average annual percentage change (AAPC) in the age-adjusted annual mortality among non-austrians across the 10-year period was not shown to be significant (AAPC: -3.9%, 95%CI: -12.8% to 7.3%) [Figure 36, 37]. Results 90

Figure 36: Log-transformed age-adjusted annual mortality from TB in Austrians compared to non-austrians, 1997-2006 10,00 Age-adjusted mortality of TB among Austrians Age-adjusted mortality of TB among non-austrians Mortality (deaths/100,000 population) 1,00 0,10 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 37: Age-adjusted annual mortality from TB in non-austrians, 1997-2006 and 10- year mean annual age-adjusted mortality rate 1,20 Age-adjusted mortality of TB among non-austrians 10-year mean annual mortality rate 1,20 Mortality (deaths/100,000 population) 1,00 0,80 0,60 0,40 0,20 1,00 0,80 0,60 0,40 0,20 Mortality rate (deaths/100,000 person-year at risk) 0,00 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0,00 Results 91

The age-specific mean annual TB mortality rates were significantly higher in non-austrians as compared to Austrians in the age-groups 15-24 years, 25-34 years, 35-44, years and 55-64 years. In all age groups 15 years, TB mortality increased significantly with age in Austrians and non-austrians [Figure 38]. Figure 38: Age-specific mean annual TB mortality rate in Austrians and non-austrians, 1997-2006 12 10 mean annual TB mortality rate in Austrians mean annual TB mortality rate in non-austrians Mortality rate (death/100,000 pyr) 8 6 4 2 0 0-4 5-14 15-24 25-34 35-44 45-54 55-64 65+ Age groups Results 92

5.2. Secondary objective: Influence of immigration to Austria on the epidemiology of TB in Austria, 1997-2006 Relationship between TB in non-austrians and TB in Austrians Does TB spread between these two population groups? Between 1997 and 2006 the epidemiology of TB among non-austrians regardless of their length of residence in Austria did not correlate with the epidemiology of TB in Austrians: No correlation was observed between age-adjusted TB incidences in Austrians and in non-austrians (r=0.01; p=0.98) [Figure 39 scatter diagram]; No correlation was found between the proportion of MDR-TB cases among definite TB cases in Austrians and the proportion of MDR-TB cases among definite TB cases in non-austrians (r=-0.45; p=0.2) [Figure 40, no scatter diagram shown]; The provincial distribution of TB cases in non-austrians showed no correlation with the provincial distribution of TB cases in Austrians [Figure 41]. Figure 39: Scatter diagram of the age-adjusted annual TB incidences in non-austrians and in Austrians, 1997-2006 18 16 Age-adjusted TB incidence in Austrians 14 12 10 8 6 4 2 0 0 10 20 30 40 50 60 Age-adjusted TB incidence of in non-austrians 70 Results 93

Figure 40: Proportion of MDR-TB among the tested TB cases in non-austrians and Austrians Proportion of MDR-TB among all TB cases with antimicrobial susceptibility testing in Austrians 10% Proportion of MDR-TB among all TB cases with antimicrobial susceptibility testing in non-austrians 8% proportion (%) 6% 4% 2% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Figure 41: Scatter diagram of the age-standardised mean annual province-specific TB incidence rates in Austrians and non-austrians Age-adjusted TB rate in Austrians (cases/100,000pyr) 18 16 14 12 10 8 6 4 2 Tyrol Vorarlberg Salzburg Lower Austria Burgenland Carinthia Upper Austria Styria Vienna 0 0 10 20 30 40 50 60 70 Age-adjustedr TB rate in non-austrians (cases/100,000pyr) Results 94

Year Statistics on immigration to Austria, 1997-2006 A total of 814,760 immigrants to Austria were registered between 1997 to 2006. The number of immigrants increased from 56,895 persons in 1997 to 108,947 persons in 2004. After 2004 immigration decreased to a total of 85,384 immigrants in 2006. The annual proportions of immigrants from Turkey and former Yugoslavia decreased between 1997 to 2006 (Turkey: 11% to 5.7%; former Yugoslavia: 26.5% to 17.3%). In contrast, the annual proportion of immigrants from Romania increased from 2.6% in 1998 to 5.6% in 2006, and the proportion of immigrants from the Russian Federation increased from 0.9% in 1998 to 6% in 2004, followed by a decrease to 2.9% in 2006 [Table 29]. Table 29: Annual Distribution of immigrants (%) to Austria by country/region of origin, 1997-2006 Number of immigrants former Yugoslavia Turkey Romania Russian Federation Other European countries 1 Africa 2 Annual proportion South- /East-Asia, West- Pacific 3 4 EMRO Latin-/ South- America 5 Rest 6 1997 56895 26.5% 11.0% 2.8% 1.1% 1.0% 2.1% 4.3% 5.3% 1.8% 44.1% 1998 59229 28.7% 10.0% 2.6% 0.9% 1.1% 2.1% 4.8% 5.4% 1.8% 42.4% 1999 72379 32.2% 10.1% 2.6% 1.0% 1.4% 2.3% 4.4% 7.2% 1.6% 37.4% 2000 65954 25.2% 10.8% 2.9% 1.4% 2.0% 2.7% 5.8% 8.5% 2.0% 38.8% 2001 74786 27.2% 10.4% 3.2% 1.2% 2.6% 2.4% 4.6% 6.3% 1.8% 40.3% 2002 92567 20.7% 11.6% 4.7% 1.8% 3.2% 2.7% 6.3% 4.5% 1.6% 43.0% 2003 97164 20.7% 10.5% 5.5% 3.7% 3.7% 2.9% 6.3% 4.4% 1.6% 40.7% 2004 108947 20.2% 7.6% 5.0% 6.0% 4.2% 3.7% 5.5% 3.9% 1.6% 42.3% 2005 101455 20.8% 7.7% 5.2% 3.9% 4.1% 2.9% 5.8% 4.1% 1.8% 43.9% 2006 85384 17.3% 5.7% 5.6% 2.9% 3.7% 2.3% 5.1% 5.5% 1.7% 50.4% total 814760 Legend: Immigrants arrived in Austria from the following countries: 1 Other very high TB-incidence countries (incidence 50/100.000): Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Tajikistan, Turkmenistan, Ukraine and Uzbekistan. 2 Africa: Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, Chad, Comoros, Cote d'ivoire, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Republic of Congo, DR Congo, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, South Africa, Swaziland, Tanzania, Togo, Uganda, Zambia, Central African Republic, and Zimbabwe. 3 South-/East-Asia, West-Pacific: Bangladesh, Bhutan, Brunei, Cambodia, China, Cook Islands, Fiji, India, Indonesia, Japan, Kiribati, North Korea, South Korea, Laos, Malaysia, Maldives, Marshall Islands, Mongolia, Myanmar, Nauru, Nepal, Niue, Palau, Papua New Guinea, Philippines, Singapore, Solomon Islands, Sri Lanka, Taiwan, Thailand, Tonga, Tuvalu, Vanuatu and Vietnam (Australia and New Zealand excluded). 4 EMRO: Afghanistan, Bahrain, Djibouti, Egypt, Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Pakistan, Qatar, Saudi Arabia, Somalia, Sudan, Syria, Tunisia, United Arab Emirates, and Yemen 5 Latin-/ South-America: Antigua and Barbuda, Argentina, Bahamas, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador, Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Micronesia, Nicaragua, Palestine, Panama, Paraguay, Peru, Samoa, St. Lucia, St. Kitts and Nevis, St. Vincent and the Grenadines, Suriname, Trinidad and Tobago, Uruguay and Venezuela 6 Rest: North America, New Zealand, Australia, other European countries with incidence <50/100.000 (i.e. Albania, Andorra, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Liechtenstein, Luxembourg, Malta, Monaco, Netherlands, Norway, Poland, Portugal, San Marino, Slovakia, Spain, Sweden, Switzerland, United Kingdom and Vatican City). Results 95

Figure 42: Proportion (%) of immigrants coming from countries/regions with TB incidence of < 20/100.000 population, countries/regions with TB incidence of 20-49/100.000 population, and countries/regions with TB incidence of 50/100.000 Table 30: Annual distribution of immigrants (%) from countries/regions with TB incidence 20/100,000 population by country/ region of origin, 1997-2006 and the annual proportion of immigrants coming from countries/ regions of origin with TB incidence> 20/100.000 Year Other high Other high Countries/ Other high Other high endemic endemic regions of Former Russian TB endemic 2 endemic countries in Romania Africa countries in origin with TB Yugoslavia Federation European countries Latin-/ countries 1 S-/E-Asia, W- Pacific-Region 3 in EMRO 4 incidence> South- America 5 20/100.000 Annual proportion 1997 35.2% 3.7% 1.5% 27.9% 2.8% 6.4% 6.9% 1.0% 61.2% 1998 38.4% 3.5% 1.2% 25.3% 2.9% 7.3% 7.0% 1.1% 61.1% 1999 41.6% 3.3% 1.2% 21.6% 3.0% 6.3% 9.1% 1.0% 65.0% 2000 33.5% 3.8% 1.8% 22.3% 3.6% 8.6% 11.0% 1.1% 63.1% 2001 37.2% 4.3% 1.6% 23.1% 3.2% 7.2% 8.3% 1.0% 62.0% 2002 31.4% 6.9% 2.6% 20.0% 4.0% 10.4% 6.5% 1.0% 62.8% 2003 29.8% 7.8% 5.3% 21.4% 4.2% 9.8% 6.1% 1.0% 65.7% 2004 28.5% 6.9% 8.2% 26.5% 5.1% 8.3% 5.2% 0.9% 65.7% 2005 29.1% 7.1% 5.3% 28.6% 3.9% 8.9% 5.4% 1.1% 63.8% 2006 26.0% 8.2% 4.2% 32.6% 3.3% 8.4% 7.8% 1.1% 60.1% Legend: Immigrants arrived from the following countries: 1 Other high TB endemic European countries: Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Tajikistan, Turkmenistan, Ukraine and Uzbekistan. 2 Africa: Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, Chad, Comoros, Cote d'ivoire, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Republic of Congo, DR Congo, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, South Africa, Swaziland, Tanzania, Togo, Uganda, Zambia, Central Africa. Republic and Zimbabwe. 3 South-/East-Asia, West-Pacific: Bangladesh, Bhutan, Brunei, Cambodia, China, Cook Islands, Fiji, India, Indonesia, Kiribati, North Korea, South Korea, Laos, Malaysia, Maldives, Mongolia, Myanmar, Nauru, Nepal, Palau, Papua New Guinea, Philippines, Solomon Islands, Sri Lanka, Taiwan, Thailand, Tuvalu, Vanuatu and Vietnam. 4 EMRO: Afghanistan, Djibouti, Iraq, Morocco, Pakistan, Qatar, Somalia, Sudan and Yemen. Results 5 Latin-/ South-America: Bolivia, Brazil, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, 96 Honduras, Micronesia, Föd.Staaten, Nicaragua, Panama, Paraguay, Peru and Suriname.

A small peak in the proportion of immigrants coming from countries with TB incidence of 20/100,000 [65% of all immigrants in 1999; Figure 42] was observed in 1999. This peak was dominated primarily by immigrants from former Yugoslavia, accounting for 41.6% of all immigrants from countries with TB incidence 20/100,000 of that year, and secondly, by immigrants from other high TB incidence European countries, accounting for 21.6% of this entrant population [Table 30]. In 2001, immigration from countries with TB incidence 20/100,000 [62% of all immigrants in 2001; Figure 42] was still dominated by persons arriving from former Yugoslavia (accounting for 37.2%), and other high TB incidence European countries (accounting for 23.1%). In 2004, 65% of all immigrants arrived from countries/regions with TB incidence 20/100,000; out of these, the 5 dominating regions/ countries of origin were: former Yugoslavia (28.5%), other high TB incidence European countries (26.5%), South Asia and East Asia (8.3%), the Russian Federation (8.2%), and Romania (6.9%) [Figure 38, Figure 42, and Table 29]. The proportion of immigrants coming from Romania, the Eastern European countries Georgia, Armenia, Moldavia, Lithuania, Latvia, Estonia, Kazakhstan, Uzbekistan Kyrgyzstan, Ukraine, Azerbaijan, and from the Russian Federation, increased from 1999, while the proportion of immigrants coming from former Yugoslavia decreased from 2001 [Table 30]. Results 97

Immigration to Austria and TB among non-austrians Non-Austrians who developed TB by year(s) elapsed since immigration to Austria, 1997-2006 Figure 43: Distribution of TB diagnosis among non-austrians by years elapsed since entering Austria, 1997-2006 [Immigrants with TB disease upon arrival are not included in this analysis]. 70 60 50 % of cases 40 30 20 10 0 1 2 3 4 5 6 7 8 9 10 11-20 > 20 Years elapsed since entering Austria Between 1997 and 2006, 94.31% of the TB cases in non-austrians developed TB within 5 years after immigration to Austria and 60.32% within the first year after immigration to Austria [Figure 43]. Results 98

Figure 44: Distribution of non-austrians who fell sick from TB in Austria by year(s) elapsed between immigration to Austria and occurrence of the disease, 1997-2006 (%) (time category I: 0 year =at entry, time category II: within 1 year, time category III: within 2 years, time category IV: within 3-5 years and time category V: later than 5 years following arrival in Austria) 70% 60% 59.0% 50% Proportion (%) 40% 30% 20% 13.5% 19.7% 10% 0% 2.2% Diagnosed as a TB case upon entry into Austria 5.5% 1 year 2 years 3-5 years >5 years Years elapsed since entering Austria Between 1997 and 2006, 2.2% of TB cases among non-austrians immigrated already as case of TB to Austria; 92.2% of TB cases among non-austrians contracted the disease within 5 years after arrival, 72.5% within 2 years, and 59% within the first year after entering Austria [Figure 44]. Results 99

Non-Austrians who contracted TB by year(s) elapsed since immigration, and residence status in Austria, 1997-2006 Figure 45: Distribution of non-austrians who contracted TB by duration (in year(s)) between immigration to Austria and TB occurrence, stratified by residence status, 1997-2006 Taking the residence status of immigrants to Austria, three distinct categories were created. These are, category 1: asylum seekers and refugees, category 2: migrant labourers, and category 3: foreigners and others. Between 1997 and 2006, 99.3% of asylum seekers and refugees who became a case of TB in Austria fell ill within 5 years after entering Austria: 77.8% of those fell ill within the first year after immigration; 10.88%, in the second year after immigration; 6.03%, in the third year after immigration; 3.68%, in the fourth year after immigration; and 0.88%, in the fifth year after immigration. 87.8% of TB cases among migrant labourers contracted TB within 5 years after arriving in Austria: in contrast, 26.2% of those, within the first year after immigration (21.5% within the second year, 16.15% within the third year 16.15% within the fourth year; and 7.7% within the fifth year after immigration. In the category of foreigners and others, 89.4% of individuals who sickened TB in Austria, did so within 5 years following immigration into Austria; and 45%, in the first year [Figure 45]. Results 100

5-year incidence of TB in the annual immigrant cohorts, 1997-2001 For computing the 5-year TB incidence among immigrants, the annual number of immigrants to Austria between 1997 and 2001 was used as the denominator (i.e. annual immigrant cohorts for 1997-2001); immigrants were stratified according to the level of TB incidence in the country of origin (i.e. <20 cases/100,000 population, 20 cases/100,000 population. Figure 46: 5-year incidence of TB in annual immigrant cohorts of 1997-2001; 4-year risk in the immigrant-cohort of 2002; 3-year risk in the immigrant-cohort of 2003; 2-year risk in the immigrant-cohort of 2004; 1-year risk in the immigrant-cohort of 2005 and the < 1-year risk in the immigrant cohort in 2006 stratified into immigrants coming from countries with TB incidence of < 20/100.000 population and from countries with TB incidence 20/100.000 Between 1997 and 2001, a peak in the 5-year TB incidence occurred in the 1999- immigrant-cohort from countries with TB incidence of 20/100,000 (220.4 cases /100,000 immigrants in 1999). These immigrants came predominantly from former Yugoslavia (62.9%), Southeast Asia and the WHO Western Pacific Region (10.3%, dominated by India and China), Romania (6.7%), EMRO (6%, dominated by Afghanistan and Iraq), from other Eastern European countries (3.54%, dominated by the Ukraine), Africa (5.96%), Russian Federation (2.5%), and from Latin and South America (2.1%). Results 101

For computing the 4-year TB incidence among immigrants, the number of immigrants to Austria in 2002, for the 3-year incidence the number of immigrants to Austria in 2003, for the 2-year incidence the number of immigrants in 2004, for the 1-year incidence the number of immigrants in 2005 and for the < 1-year incidence the number of immigrants in 2006 were used as the denominator; immigrants were stratified according to the level of TB incidence in the country of origin (i.e. <20 cases/100,000 population, 20 cases/100,000 population. Those immigrating in 2004 had a 2-year risk of becoming a case of TB (280.7 cases per 100,000 immigrants) higher than the 4-year risk in the immigrantcohort of 2002, the 3-year risk in the immigrant-cohort of 2003, the 1-year risk in the immigrant-cohort of 2005 and the < 1-year risk in the immigrant cohort in 2006. Even though these values are hardly to compare, the 2004-incidence peak is according to our previous findings. The 2004 immigrant-cohort was dominated by immigrants from former Yugoslavia (36.4%), Russian Federation (14.27%), Southeast Asia and the Western Pacific Region (12.4%), Romania (12.08%), Eastern European countries (10.03%), Africa (8.81%); the Latin-/ South-America (2.12%) contributed marginally. EMRO (3.88%), and Between 1997-2001, the 5-year incidence of TB in the cohort of immigrants coming from countries with TB incidence of 20/100,000 population was significantly higher compared to the cohort of immigrants coming from countries with TB incidence of <20/100,000 population for the years analysed (p < 0.01) [Figure 46]. The 5-yrear risk of becoming a case of TB in immigrants from >20/100,000 population TB incidence countries was 17 times higher compared to immigrants from <20/100,000 population 1997-2001 (p< 0.001). TB incidence countries in the period Results 102

Year of entry Table 31: Number of cases that occurred in non-austrians (non-a) immigrating from countries with TB incidence of 20/100,000 population within 5 years after entry into Austria, by year of entry and by country/region of origin; the annual distribution of immigrants (%) from 20/100,000 population TB-incidence countries/regions by country/ region,1997 until 2006; 5-year incidence of TB in the annual entrant cohorts, 1997-2006 Former Yugoslavia Cases in non-a Immigrants Romania Cases in non-a Immigrants Russian Federation Cases in non-a Immigrants Other high TB endemic European countries 1 Cases in non-a Immigrants Cases in non-a 2 Africa Immigrants Other high endemic countries in S- /E-Asia, W- Pacific-Region 3 Cases in non-a Immigrants Other high endemic countries in EMRO 4 Cases in non-a Immigrants Other high endemic countries in Latin-/ South- America 5 Cases in non-a Immigrants 5-year incidence (n/n) x 100,000 n % n % n % n % n % n % n % n % 1997 16 35.2% 3 3.7% 0 1.5% 0 27.9% 7 2.8% 15 6.4% 7 6.9% 1 1.0% 213.9 1998 16 38.4% 2 3.5% 0 1.2% 4 25.3% 11 2.9% 18 7.3% 7 7.0% 1 1.1% 232.0 1999 48 41.6% 8 3.3% 1 1.2% 1 21.6% 21 3.0% 10 6.3% 9 9.1% 2 1.0% 297.7 2000 15 33.5% 5 3.8% 1 1.8% 3 22.3% 15 3.6% 24 8.6% 16 11.0% 1 1.1% 277.7 2001 19 37.2% 5 4.3% 2 1.6% 8 23.1% 14 3.2% 23 7.2% 17 8.3% 1 1.0% 262.0 2002 18 31.4% 11 6.9% 7 2.6% 16 20.0% 21 4.0% 28 10.4% 15 6.5% 3 1.0% 307.2 2003 24 29.8% 8 7.8% 36 5.3% 18 21.4% 24 4.2% 24 9.8% 15 6.1% 2 1.0% 343.4 2004 16 28.5% 6 6.9% 62 8.2% 32 26.5% 32 5.1% 37 8.3% 12 5.2% 3 0.9% 392.1 2005 9 29.1% 9 7.1% 44 5.3% 27 28.6% 12 3.9% 25 8.9% 13 5.4% 1 1.1% 307.9 2006 5 26.0% 5 8.2% 8 4.2% 10 32.6% 9 3.3% 14 8.4% 6 7.8% 0 1.1% 172.1 Table 31 displays the number of cases that occurred in non-austrians immigrating from countries with TB incidence of 20/100,000 population within 5 years after entry into Austria, by year of entry and by country/region of origin The table also shows the annual distribution of immigrants (%) from countries with a TB incidence 20/100,000 population by country/ region of origin. The far right hand column of the table shows that the 5-year risk of TB in the annual entrant cohorts peaked first in 1999 and secondly in 2004. Immigration from former Yugoslavia peaked in 1999, and immigration from the Russian Federation peaked in 2004. Results 103

Annual proportion of TB in non-austrians, and annual number of immigrants from high TB incidence countries, 1997-2006 The annual proportion of non-austrian TB cases among the TB cases in Austria between 1997 and 2006 correlated positively with the annual number of immigrants from high TB incidence countries (TB incidence: 20/100,000) between 1997 and 2006, with a correlation coefficient of r= 0.87 (p<0.01) [Figure 47]. 75% of the variation in the annual proportion of TB cases among non-austrians between 1997 and 2006 can be explained by the linear relationship with the annual immigration from high TB incidence countries to Austria in this period. Figure 47: Annual proportion of non- Austrian TB cases and annual number of immigrants from high TB incidence countries (TB incidence 20/100,000) Proportion (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% Annual proportion of non-austrian TB cases Number of entrants from high TB incidence countries 79443 68806 56249 62615 55715 42814 49919 44327 41.4% 24.9% 27.6% 34.7% 28.8% 25.5% 25.2% 22.5% 73866 58132 42.9% 37.2% 90000 80000 70000 60000 50000 40000 30000 20000 Number of immigants 10% 10000 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 From 1997 to 2006 the majority of TB cases among non-austrians originated from the following countries and regions: Former Yugoslavia, with a total of 1,319 cases (37.9%); Turkey, with 500 cases (14.4%); and Asian countries, with 407 cases (11.7%). TB cases among non-austrians having immigrated from other high TB incidence European countries (i.e. Romania, Georgia, Armenia, Results 104

Poland, Bulgaria etc.) accounted for 8.5% (n=296); cases among non-austrians from African countries, for 6.9% (n=242); cases among non-austrians from Middle Eastern countries, for 6.7% (233); cases among non-austrians having immigrated from the Russian Federation, for 6.3% (n=221); from Latin and South American countries, 0.7% (n=24); and from other low TB incidence countries of the world, 6.9% (n=241). Figure 48 shows the proportional distribution of TB cases among non-austrians by origin (country, or region) and year of case occurrence, displaying only countries/ regions with TB-incidence 20 /100,000. Figure 48: Annual proportional distribution of TB cases among non-austrians by origin and year, displaying only cases in non-austrians coming from high TB incidence countries/regions ( 20 /100,000 TB incidence) Throughout the study period, TB cases among non-austrians from former Yugoslavia accounted always for the highest proportion of TB cases among non- Austrians. The proportion of cases in non-austrians originally from Turkey decreased slightly over the 10-year period. Results 105

From 1999, cases among non-austrians from the high TB incidence European countries such as Armenia, Azerbaijan, Bulgaria, Georgia, Hungary, Kazakhstan, Lithuania, Poland, Portugal, Romania, Spain, Ukraine and Uzbekistan increased steadily. From 2002, the proportion of cases in non-austrians originating from the Russian Federation increased tremendously. Results 106

Correlation between immigration from high TB incidence countries and TB among non-austrians Figure 49: Annual number of all immigrants and annual number of immigrants coming from countries with TB incidence of 20/100,000 (in purple bars, striped purple bars), annual number of TB cases in non-austrians (trend line), 1997-2006 1999: Refugee wave from former Yugoslavia 2003-2004: Refugee wave from Chechnya Figure 49 displays the trend in the number of TB cases among non-austrians between 1997-2006 (blue trend line), and in the number of immigrants from countries with TB incidence of 20/100,000 (striped bar chart). Between 1998 and 2000 the peak for the annual number of cases in non-austrians correlated with the peak in immigration from high TB incidence countries that occurred (primarily determined by immigration from former Yugoslavia) one year later. Between 2001 and 2006 the annual number of TB cases in non-austrians correlated with the annual number of immigrants from the high TB incidence countries (r 2001-2006 = 0.93; p<0.01). 86% of the variation in the annual number of TB cases in non-austrians between 2001 and 2006 can be explained by the linear relationship with immigration activities from high TB incidence countries during the same period. Results 107

Correlation between immigration from 50/100,000 TB incidence countries and TB among non-austrians, 1997-2006 Figure 50: Annual number of all immigrants, annual number of immigrants coming from countries with TB incidence of 50/100,000 (in purple bars, striped purple bars), and annual number of TB cases in non-austrians (trend line), 1997-2006 1999: Refugee wave from former Yugoslavia 2003-2004: Refugee wave from Chechnya Figure 50 displays the trend in the number of TB cases among non-austrians between 1997-2006, and the number of immigrants from countries with TB incidence of 50/100,000 (very high TB incidence countries) within the same period. Immigration from very high TB incidence countries steadily increased from 1997 to 2004. No correlation was found between the annual number of TB cases in non-austrians between 1997-2000, and the annual immigration figures from very high TB incidence countries during the same period. In contrast, the annual number of TB cases in non-austrians did correlate with the annual number of immigrants from very high TB incidence countries (r 2001-2006 = 0.91; p=0.01) between 2001 and 2006. 83% of the variation in the annual number of TB cases in non-austrians between 2001 and 2006 can be explained by the linear relationship with the immigration figures from very high TB incidence countries to Austria during this period. Results 108

Correlation between immigration and TB in non-austrians by the level of TB incidence in the countries of origin, 1997-2006 Analyses of the relationship between immigration to Austria and the epidemiology of TB in non-austrians segregated into countries with TB-incidence of 20/100,000 population (high TB incidence country), and with TB-incidence < 20/100,000 population (low TB incidence country) produced the following results: A positive correlation was found between immigration from high TB incidence countries, and the number of TB cases among non-austrians having immigrated from high TB incidence countries (r=0.75; p=0.01) [Figure 51a, Figure 51b in the annex]: 56.3% of the variation in the annual number of TB cases in non- Austrians between 1998 and 2006, can be explained by the linear relationship with the immigration from high TB incidence countries to Austria. No correlation could be found between immigration from low TB incidence countries and the number of TB cases in non Austrians having immigrated from low TB incidence countries (r=-0.31; p=0.39) [Figure 51c, Figure 51d in the annex]: Immigration activities from low TB incidence countries increased between 1997 and 2006, but the number of TB cases in non-austrians having immigrated from low TB incidence countries remained steadily low during the study period. Results 109

Figure 51a: Annual number of immigrants from high TB incidence countries and the annual number of TB cases among immigrants from these countries, 1997-2006 90000 80000 Number of immigrants coming from high TB incidence countries Number of TB cases in non-austrians coming from high TB incidence countries 500 450 70000 400 60000 350 Number of immigrants 50000 40000 30000 300 250 200 150 Number of cases 20000 100 10000 50 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 Figure 51c: Annual number of immigrants from low TB incidence countries and the annual number of TB cases among immigrants from these countries, 1997-2006 90000 80000 Number of immigrants coming from low TB incidence countries Number of TB cases in non-austrians coming from low TB incidence countries 500 70000 400 60000 Number of immigrants 50000 40000 30000 300 200 Number of cases 20000 100 10000 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 Results 110

Correlation between immigration from high TB incidence countries, and MDR-TB among non-austrians The incidence of MDR-TB cases in non-austrians peaked twice during the study period, first in 1999, and second in 2004. with a 3.5 fold higher peak (0.72 versus 2.54/100,000). The number of immigrants coming from high TB incidence countries (TB incidence of 20/ 100,000) peaked also in 1999 and 2004. According to the correlation analysis of MDR-TB in non-austrians, and immigration from high TB incidence countries, 74.5% of the variation between 1997 and 2006 in the annual number of MDR-TB cases among non-austrians can be explained by the linear relationship with the immigration from high TB incidence countries to Austria (r=0.86; p= 0.03) [Figure 52]. Figure 52: Annual number of MDR-TB cases in non-austrians and annual number of immigrants from high TB incidence countries, 1997-2006 90000 Number of immigrants coming from high TB incidence countries Number of MDR-TB cases in non-austrians 20 80000 70000 15 Number of immigrants 60000 50000 40000 30000 20000 10 5 Number of cases 10000 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 Results 111

Correlation between immigration from high TB incidence countries and MDR-TB among non-austrians, by countries/ regions of origin Within the 6-year period of 1997 to 2002, the average annual number of MDR- TB cases was 4.8 (including on average 3.5 non-austrian MDR-TB cases). The average annual number of MDR-TB cases within the successive 4-year period (2003 to 2006) was 14.8, with a ratio of non-austrians to Austrians of 13.8. The steep increase of MDR-TB cases in non-austrians between 2003 and 2004 was due to MDR-TB cases in non-austrians originating from the Russian Federation and from high TB incidence Eastern European countries. MDR-TB cases among non-austrians originating from the Russian Federation (n=9) peaked in 2004, as did immigration from the Russian Federation in that year. The same pattern was observed in relation to the high TB incidence Eastern European countries: In 2004 there was a peak of 7 MDR-TB cases from that region, as well as a peak in immigration to Austria from the same region (n=19,371) [Figure 53]. Figure 53: Annual number of MDR-TB cases in non-austrians coming from countries with TB incidence of 20/100,000 population, by countries/regions of origin; and annual number of immigrants from high TB incidence countries, 1997-2006 10 Africa & Asia Other high TB incidence European countries Turkey Former Yugoslavia Russian Federation Number of immigrants coming from high TB incidence countries 90000 9 8 80000 70000 Number of cases 7 6 5 4 3 60000 50000 40000 30000 Number of immigrants 2 20000 1 10000 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 Results 112

Analyses of the relationship between immigration and the number of MDR-TB in non-austrians, between 1997 and 2006, when segregated into countries with a TB incidence of 20-49/100,000 population, and a TB incidence of 50/100,000 population yielded the following results: A positive correlation was found between immigration from countries with a TB incidence of 50/100,000 population, and the number of MDR-TB cases in non- Austrians having immigrated from these countries (r=0.89; p<0.01). 79.2% of the variation in the annual number of MDR-TB cases in non-austrians between 1997 and 2006 can be explained by the linear relationship with immigration from very high TB incidence countries to Austria. No correlation was found between immigration from countries with a TB incidence of 20-49/100,000 population, and the number of MDR-TB cases in non- Austrians having immigrated from these countries (r=0.03; p=0.93) [Figure 54]. Figure 54: Annual number of immigrants, and annual number of MDR-TB cases, from countries with a TB incidence of 20-49/100,000 population, and countries with a TB incidence of 50/100,000 population, 1997-2006 60000 Number of immigrants coming from very high TB incidence countries ( 50/100,000 population) Number of immigrants coming countries with a TB incidence of 20-49/100,000 population Number of MDR-TB cases coming from very high TB incidence countries ( 50/100,000 population) Number of MDR-TB cases coming from countries with a TB incidence of 20-49/100,000 population 20 50000 18 16 Number of immigrants 40000 30000 20000 14 12 10 8 6 Number of cases 10000 4 2 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 Results 113

Immigration to Austria and TB in long-stay non-austrians Tb cases in foreign-born Austrian residents (non-austrians) were defined as longstay non-austrian cases who have lived in Austria for at least 5 years before having contracted TB, otherwise defined as short-stay non-austrian cases (newly-immigrated). Figure 55 demonstrates the 10-year trend for TB-cases in non-austrians who fell ill from TB more than 5 years after their arrival in Austria. In parallel to the decreasing trend of TB incidence in Austrians between 1997 and 2006, the number of TB-cases among long stay non-austrians also decreased (slope= -10.3; 95%CI: -14.6 to -5.9; R 2 = 0.79; p=0.0006) except in 2000 and 2005 following high immigration activities to Austria in 1999 and 2004. The association between the percentage change in the number of TB cases among the long stay non-austrians and the annual number of immigration activities from high TB-incidence countries was significant. Figure 55; Annual number of TB in non-austrians who developed TB more than 5 years after arriving in Austria, and the annual number of immigrants coming from high TB incidence countries, 1997-2006 250 Number of TB cases in non-austrians developed TB more than 5 years after entry to Austria Number of immigrants from high TB-incidence countries 90000 80000 200 70000 Number of cases 150 100 60000 50000 40000 30000 Number of immigrants 50 20000 10000 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year 0 Results 114

Immigration to Austria and TB in Austrians There was no positiv correlation observed between immigration to Austria and the epidemiology of TB among Austrians: Between 1997 and 2006 there was an increasing trend in the number of immigrants to Austria (by 5,129 immigrants annually; p=0.002), whereas TB-incidence, MDR-TB-incidence and TB-mortality among the Austrians decreased steadily. Interpretation 115

6. Interpretation We describe the results of the analyses of Austrian TB surveillance data from the years 1997 to 2006. An analysis of 5-year TB surveillance data, 1995 to 1999, was performed in 2003 by Strauss et al. [15]. The objectives for the analyses of the 10-year surveillance data were first, to describe the burden of tuberculosis (TB) in Austria from 1997 to 2006, and, second, to elucidate the influence of immigration to Austria on the epidemiology of TB in Austria. The 10-year trends in TB incidence, MDR-TB incidence, and TB mortality were analysed in terms of the demographic factors age, sex, and province of residence; and residence status, for which there were two categories: Austrian citizens (Austrians), and immigrants living in Austria who are not citizens (non-austrians). Based on the findings of the second research question, recommendations for TB control in immigrants are provided. 6.1. Interpretation of the findings on research question I: Burden of TB disease in Austria. Tuberculosis. From 1997 until 2006 the crude annual incidence decreased by 7.7 cases per 1,000,000 Austrian residents per year. The annual decrease in TB incidence among Austrians (i.e. Austrian citizens) was of 9.1 cases per 1,000,000 persons per year. A significant decrease was recorded in all 9 provinces. In comparison to other high-income European countries such as Belgium (annual reduction of 1.7 cases/ 1,000,000), Denmark (annual reduction of 2.6 cases/1,000,000), Finland (annual reduction of 6.7 cases/1,000,000), France (annual reduction of 4.6 cases/1,000,000), the reduction in annual TB incidence between the years 1997 and 2006 was relatively high in Austria, and comparable to Germany, with an annual reduction of 8.2 cases/1,000,000 [16]. TB incidence increased with age in Austrian inhabitants aged 4 years and older. Austrian inhabitants aged > 64 years were with a 10-year mean annual incidence rate of 21.8/100.000 pyr at highest affected. This pattern of TB incidence and age distribution has also been observed in the populations of other industrialised Interpretation 116

countries, such as Italy, Germany, Switzerland and North-America [16, 17]. The increase in the risk of TB with age was more apparent in male compared to female Austrians. These findings were also observed in other low TB incidence European countries [18]. MDR-TB. From 1997 to 2006, out of the 11,496 TB cases, the Mycobacteria tuberculosis complex (MTC) was isolated and tested for antimicrobial susceptibility in 7,009 TB cases. Out of these, 84 cases of multi-drug resistant tuberculosis (MDR-TB) (1,2%) were reported. During the study period, the proportion of MDR-TB cases among TB cases in which antimicrobial susceptibility testing was performed ranged between a minimum of 0.4% (in 2002), and a maximum of 3% (in 2004), with an proportion of 0.75% on average. Between 1997 and 2002 the crude proportion of MDR-TB cases remained at a low and stable level (average: 0.6%, range: 0.4%-0.8%), followed by a steep increase towards 2004, which was exclusively due to the rise of MDR-TB cases in non- Austrians [see Figure 39 in result part]. During the study period, only 12 cases of MDR-TB occurred in Austrians, resulting in an annual MDR-TB incidence of between 0 and 0.03 cases per 100,000 Austrians. Annual changes were not significant. Between 2002 to 2006, an increasing trend in the proportion of MDR-TB among TB cases with antimicrobial susceptibility testing was also observed in other highincome European countries, including France, Germany and Spain, [19]. In the same observation period, a decreasing trend in the proportion of MDR-TB was observed in Italy, the Netherlands, Norway, Sweden, and Switzerland [19]. Between 2002-2006, the proportion of MDR-TB cases in Denmark remained stable, at a level of 0% to 1.5%; in Finland, at a level of 0% to 1%; and in the United Kingdom, at a level of 0.8% to 1.3%. Further research is required in order to establish the reasons for the differences in MDR-TB trends between high income European countries. Interpretation 117

Treatment outcome. From 1997 to 2006, 1,125 (10.3%) out of a total of 11,496 TB cases in Austria were declared as cured. The proportion of cured TB cases decreased in both Austrians and non-austrians during the period 1998-2006. The proportion of treatment completion was stable in non-austrian TB cases, whereas the annual proportion of TB cases in Austrians with treatment completion increased slightly. The decreasing trend in the proportion of cases cured, in parallel to the slightly increasing trend in the proportion of treatment completion cases, might be due to the fact that the number of cases that underwent the test for sputum conversion at the time of treatment completion decreased or, where recorded as treatment completion before the tests for sputum conversion were available. Throughout the entire study period, the annual proportion of TB cases lost to follow-up among non-austrians was substantially higher compared to TB cases lost to follow-up among Austrians. This is due to the greater difficulty of case management particularly in newly arrived non-austrians. However, there were no obvious trends in the proportion of lost to follow-up cases, in Austrians and in non-austrians. TB mortality. Between 1997 and 2000 the age standardised crude TB mortality remained constant, around an average of 0.48 deaths per 100,000 population, but decreased significantly by 5 deaths per 10,000,000 population per year from 2000 to 2006. Similar decreasing trends in TB mortality were observed in other high income Western European countries such as in France, Germany, Spain and Switzerland. In Eastern Europe and the Russian Federation the situation was reversed: between 2000 and 2006, an increasing trend in TB mortality was observed in Moldova, Ukraine, and Russia [19]. Age-standardised TB mortality in Austrians showed a significant decreasing trend by 4 deaths per 10,000,000 persons per year. Since no significant decrease in the incidence of MDR-TB (considered as risk factor of death from TB) occurred in the Austrians between 1997-2006, this decreasing trend might be primarily explained by improved case management. Interpretation 118

Age-adjusted annual mortality among non-austrians remained steady during the 10-year study period. The fact that the TB mortality in non-austrians did not reflect the increase in MDR-TB incidence among non-austrians between 2002 and 2006 (see below), could also be explained by the improvement in case management of non-austrian TB cases during the last 5 years of the study period. During the study period, without considering the difference in the distribution of age and MDR-TB, there was no difference in the average time elapsed between the date of TB diagnosis and death from TB between Austrians and non- Austrians More research on this subject is needed. Conclusion on the Burden of TB in Austria As observed in other high income countries, the TB-incidence in Austrian residents (including Austrians and non-austrians) decreased significantly between 1997 and 2006. The annual decrease in TB incidence among Austrians (i.e. Austrian citizens) was of 9.1 cases per 1,000,000 persons per year during this study period. The annual MDR-TB incidence in Austrians were at a low and stable level during the stud period (range: 0-0.03 cases per 100,000 Austrians). Annual changes were not significant. Age-standardised TB mortality in Austrians showed a significant decreasing trend by 4 deaths per 10,000,000 persons per year. Interpretation 119

6.2. Interpretation of the findings on research question II: Influence of immigration to Austria on the epidemiology of TB in Austria 6.2.a) TB in non-austrians and TB in Austrians Does TB spread between these two population groups? Between 1997 and 2006 the epidemiology of TB (including MDR-TB) among non- Austrians, regardless of their length of residence in Austria, did not correlate with the epidemiology of TB in Austrians (r=0.01; p=0.98) [see Figure 39, Figure 40 in result part]. The age-standardised 10-year mean annual incidence rate was 5.4 times higher in non-austrians compared to Austrians (47.1/100,000 versus 8.7/100,000). These rates are consistent with the findings from previous analyses of the TB burden in immigrants in Milan, Italy [17]. The agestandardised mean annual incidence rate in non-austrians remained unaltered and high during the two 5-year periods 1997-2001 and 2002-2006 (46.8/100,000 pyr at risk versus 46.2/100,000 pyr). In contrast, the age-standardised 5-year mean annual incidence rate of TB in Austrians decreased significantly from 10.6/100,000 pyr in the first 5-year period (1997-2001), to 6.8/100,000 pyr in the second 5-year period (2002-2006). The provincial distribution of TB cases among Austrians and non- Austrians showed no correlation which would have indicated TB spread from non-austrians to Austrians [see Figure 41 in result part]. The age-standardised mean annual incidence rate in Austrians was highest in the province of Vienna (13/100,000), followed by the province of Upper Austria (8.44/ 100,000), and by Carinthia (8.35/100,000). In comparison, the age-standardised mean annual incidence rate in non-austrians was highest in the province of Styria (59.7/100,000), followed by the province of Vienna (57.2/ 100,000), and the province of Upper Austria (53.8/100,000). The variation of TB incidence in Interpretation 120

Austrians by province could not be explained by the provincial distribution of TB in non-austrians [see Figure 41 in result part]. Consistent with our findings Matthew et al. also observed no correlation between state-specific TB rates among U.S.- born persons and state-specific TB rates among foreign born persons (see below the scatter diagram from their publication) [20]. The highest age-specific mean annual TB incidence rate among non- Austrians was observed in the age group 15-24 years (69.3/100,000pyr, 16.6 times higher than among Austrians of the same age group), followed by the age group 25-34 years. Similarly, the age distribution for TB incidence among immigrants from developing countries to Milan, Italy [17], showed that the highest incidence rate was in the age group 25-34 years. By comparison, among Austrians it was the > 64 age group that was most affected. These findings on the differences in age distribution of TB cases among Austrians and non- Austrians indicate that there is minimal social interaction between Austrians and non-austrians of the same age group. The age-adjusted annual TB mortality was significantly higher in non- Austrians compared to Austrians in 7 of the 10 years analysed (1997, 1999, Interpretation 121

2001, 2002, 2003, 2005 and 2006). While throughout the study period the ageadjusted annual TB mortality of non-austrians did not change significantly, between 1997 and 2006 TB mortality in Austrians decreased significantly by an annual average percentage change of 11.7%. Conclusion I on the Influence of immigration to Austria on TB in Austria TB in non-austrians and TB in Austrians Between 1997 and 2006 no linear relationship (correlation) was found between the burden of TB in non-austrians and that in Austrians. Our findings are similar to observations from other studies from Denmark, US-America and Canada, where the same research question was posed: Based on molecular epidemiological analyses, the proportion of cases of TB among native-born individuals that were attributed to the transmission of mycobacteria from foreignborn persons, was estimated as low as 0.9% in Denmark, 2% in U.S. and 11% in Canada [21-24]. Interpretation 122

6.2.b) Immigration activities to Austria between 1997 and 2006 Background on immigration procedures in Austria. According to Austrian immigration law, those who plan to migrate to Austria must apply in advance for an Austrian residence permit in their home country. There are several types of residence permits, reflecting different statuses and reasons for immigration. These include: (1) residence permit ( Aufenthaltsbewilligung ), (2) settlement permit ( Niederlassungsbewilligung ), (3) dependant ( Familienangehöriger ), (4) permanent residence ( Daueraufenthalt ), and (5) permanent residence family membership ( Daueraufenthalt für Familienmitglieder ). Individuals who migrate to Austria in order to apply for asylum due to persecution in their home country (in accordance with the United Nations 1951 refugee convention) initially receive the status of refugees, and, after formally applying for asylum, become asylum seekers, and receive a residence permit while the process of verification of the justification for asylum is conducted by the authorities. Refugees are hosted initially in two refugee camps situated in Upper Austria and Lower Austria. Within the first three days after arrival, camp residents undergo screening for TB disease, including a chest x-ray. Individuals with screening results that are suggestive of tuberculosis are transferred to a relevant hospital for further diagnostic procedures and case management. After several days in the refugee camps, the refugees are usually transferred to asylum seeker hostels located across all the nine provinces. Between 1997 and 2006 a total of 814,760 immigrants to Austria were registered by the Austrian Ministry of Internal Affairs. Immigrants originating from European countries with TB incidence of 20/100,000 persons, accounted for 58% of all immigrants to Austria, and those coming from European countries with TB incidence of < 20/100,000 persons, for 24.9% of all immigrants. 14.7% of immigrants to Austria during the 10-year study originated from non-european countries with TB incidence of 20/100,000 persons, and 2.4% from non- European countries with TB incidence of < 20/100,000 persons. Immigrants from Interpretation 123

former Yugoslavia entered Austria in two immigration waves (1999 and 2004), and accounted for the majority of the immigrant cohort for the years 1997-2006 (n=191,769; 23.5 %). After Germany (n=100,868; 12.4%), Turkey was ranked as the third most frequent country of origin among immigrants (n=76,194; 9.4%). From 1997 until 2004, immigration to Austria increased almost twofold (56,895 in 1997 and 108,947 migrants in 2004), reflecting population mobility across Europe due to political crises. The number of immigrants from high TB incidence countries ( 20 cases/ 100,000 persons) to Austria peaked in the years 1999 and 2004. Figure 56 shows in absolute numbers the 10-year trend of immigration activity to Austria. Figure 56: Annual number of all immigrants and annual number of immigrants coming from countries with TB incidence 20/100,000 (in purple bars, striped purple bars) This immigration pattern is explained by the following historical events: Beginning in 1991, political upheavals shook the former Socialist Federal Republic of Yugoslavia. In 1999, about one million Albanians fled from alleged Serbian persecution, and over 200,000 Serbs and other non-albanian minorities left Kosovo after the Kosovo War. In that year, immigrants to Austria from former Interpretation 124

Yugoslavia reached a record of 23,391, resulting in the first immigration peak from that region. The second peak of immigrants (22,618) from former Yugoslavia was in 2004, when violent unrest in Kosovo broke out. Immigrants from the Russian Federation also contributed to the 2004 peak. As a consequence of the war in Chechnya, around 120,000 Russian citizens sought asylum in the European industrialized countries between 2000 and 2004. The number of immigrants from the Russian Federation to Austria peaked in 2004 and 2005; and the proportion of immigrants from Russia increased from 1.1% in 1997, to 6% in 2004. According to estimate of the UNHCR (United Nations High Commissioner for Refugees), the vast majority of asylum seekers from the Russian Federation were Chechens. The Russian Federation has been a high TB incidence country for the past decade, ranging between 65/100,000 population in 1995 and 108/100,000 population in 2005 [16]. In addition, the opening of boarders by Eastern European countries induced an enormous East to West migration wave. The proportion of immigrants from high TB incidence European countries, such as Romania, Georgia, Armenia, Moldavia, Lithuania, Latvia, Estonia, Kazakhstan, Uzbekistan Kyrgyzstan, Ukraine, Azerbaijan, and from the Russian Federation, increased from 1999 to 2004. After the year 2004 the number of immigrants decreased again, and in 2006 there were 85,384 immigrants to Austria. 6.2.c) Immigration to Austria and TB among non-austrians Tuberculosis. Between 1997 and 2006, 3,483 (30.3%) of a total of 11,496 recorded TB cases in Austria were among non-austrians. The majority of TB cases among non-austrians originated from former Yugoslavia, with a total of 1,319 cases (37.9%), followed by 500 cases (14.4%) from Turkey, and 407 cases (11.7%) from Asian countries. TB cases in non-austrians coming from high TB incidence European countries other than former Yugoslavia (Eastern European countries including Romania, Georgia, Armenia, Moldova, Lithuania, Latvia, Estonia, Kazakhstan, Uzbekistan, Kyrgyzstan, Ukraine, Azerbaijan) accounted for 8.5% (n=296), cases from African countries accounted for 6.9% Interpretation 125

(n=242), cases from Middle Eastern countries for 6.7% (233), cases from the Russian Federation for 6.3% (n=221), cases from Latin and South American countries for 0.7% (n=24), and cases from other low TB incidence countries around the world accounted for 6.9% (n=241) of the total number of cases among non-austrians. Analysis of the distribution of TB cases among non-austrians by country or region of origin demonstrated the increasing proportion of cases in non-austrians coming from the Russian Federation between 2002 and 2004, and an increasing proportion of cases in non-austrians coming from other high TB incidence European countries between 1999 and 2004. In contrast, the proportion of cases in non-austrians coming from former Yugoslavia and Turkey decreased during the 10-year study period [see Figure 48]. The annual proportion of TB cases in non-austrians among all TB cases reported in Austria between 1997 and 2006 correlated positively with the annual number of immigrants from high TB incidence countries (1-year TB incidence: 20/100,000) between 1997 and 2006, with a correlation coefficient of r= 0.87 [see Figure 47]. This means that 76% of the variation in the annual proportion of TB cases among non-austrians between 1997 and 2006 can be explained by the linear relationship of the proportion of TB cases in non-austrians with the number of immigrants to Austria. When considering a time lag of 1 year between annual immigration activity to Austria and its possible impact on the tuberculosis burden among non-austrians, the variation in the proportion of non-austrian TB cases correlated almost perfectly with the number of immigrants from the previous year. 90% of the changes between 1997 and 2006 in the annual proportion of TB cases in non-austrians can be explained by the linear relationship with the immigration activity from high TB incidence countries from the previous year [see Figure 47]. Analysing the relationship between the annual number of TB cases in non- Austrians and the annual immigration activities to Austria from high TB incidence Interpretation 126

countries ( 20/100.000 persons) between 1997 and 2006 the following observations were made: Between 1998 and 2001, the 2000 peak in the annual number of cases in non- Austrians followed a peak in immigration from high TB incidence countries with a one year delay. Between 2001 and 2006 the annual number of TB cases among non-austrians highly correlated with the annual number of immigrants from high TB incidence countries. The number of TB cases in non-austrians peaked in 2004, the same year in which immigration activities from high TB incidence countries peaked. 86% of the variation in the annual number of TB cases in non- Austrians could be explained by the linear relationship with immigration activities between 2001 and 2006 from high TB incidence countries to Austria. The 2004 peak in the number of cases in non-austrians was determined by cases in non- Austrians coming from former Yugoslavia, Eastern European countries, the Russian Federation and Asia. These findings fit perfectly to the 2004 peak in immigration activities which was determined by immigrants from former Yugoslavia, Eastern European countries, the Russian Federation, and Asia [Figure 57; Figure 58, not given in the result part]. An analysis of the linear relationship between immigration and the number of TB cases in non-austrians from 1997 to 2006, grouped into countries of origin with TB incidence of 20/100,000, and TB incidence of < 20/100,000 yielded the following results: A positive correlation was found between immigration from high TB incidence countries and the annual number of TB cases in non-austrians coming from high TB incidence countries between 1998 and 2006 (r=0.75; p=0.01). 56.3 % of the variation in the annual number of TB cases in non-austrians between 1998 and 2006 can be explained by the linear relationship with the immigration from high TB incidence countries to Austria. No correlation was found between immigration from low TB incidence countries and the number of TB cases in non-austrians coming from low TB incidence countries (r=-0.31; p=0.39). While immigration activities from low TB incidence Interpretation 127

countries increased steadily, the number of TB cases in non-austrians originating from low TB incidence countries remained steady between 1997 and 2006 [Figure 51]. We also analysed the relation between the epidemiology of TB in non-austrians and immigration to Austria by country of origin. TB incidence in non-austrians originating from former Yugoslavia peaked in 1999 and 2000 (51.7 and 47.8/100,000) [Figure 14, Figure 57, number of cases in non-austrians from former Yugoslavia, see purple arrows], and the number of immigrants from former Yugoslavia also peaked in 1999, with a total of 23,391 [Figure 57, number of immigrants from Former Yugoslavia, see purple arrow]. After 2000 the TB-incidence in Non-Austrians from former Yugoslavia decreased significantly until the year 2006. In 2004, TB incidence in non-austrians originating from countries other than former Yugoslavia and Turkey (data on the origin of non-austrians from countries/region other than former Yugoslavia and Turkey are not available) was 89.2/100,000 population, the highest incidence observed in all non-austrian groups throughout the 10-year study [see brown arrow in Figure 14). Cases in non-austrians originating from the Russian Federation contributed 27.5% of the 2004 incidence peak (n=81), cases from Eastern European countries, 21.4% (n=63), cases from Asian countries, 20% (n=59), cases from African countries, 15.9% (n=47), and cases from Middle Eastern countries, 8.1% (n=24). There was a peak in the number of cases in non- Austrians from the Russian Federation in 2004 (n=81, including the highest proportion of all refugees entering Austria between 1997 and 2006), and immigration activities from the Russian Federation also peaked that year (n=6,500) [see grey arrow in Figures 57 and 58, information not given in the result part]. The annual number of TB cases in persons originating from Africa, South and East Asia, Western Pacific, and the Middle East peaked in 2004 (n=132), and 2005 (n=146) [Figure 57, see green arrows]. The number of immigrants coming from these regions also peaked in 2004, with a total of 15,961 immigrants [Figure 58, green arrow]. The annual number of TB cases in Interpretation 128

persons originating from high TB incidence Eastern European countries peaked in 2004 (n=63) and 2005 (n=53) [Figure 57, dark green arrows], again mirroring the peaks in immigration from this region in 2004 (n=19,371) and 2005 (n=19,253) [Figure 58, dark green arrows]. Figure 57: Annual distribution of TB cases in non-austrians from countries with TB incidence of 20/100,000, by countries/regions of origin (in numbers) Figure 58: Annual distribution of immigrants to Austria from countries with TB incidence of 20/100,000, by countries/regions of origin, 1997-2006 (in numbers) Interpretation 129

In summary, the comparative analyses of the origin of immigrants to Austria from 1997 to 2006, and of TB cases in non-austrians by country of origin, 1997-2006, supported our previous findings on the relationship between TB in non- Austrians, and immigration activities from high TB incidence countries to Austria. MDR-Tuberculosis. From 1997 to 2006, 72 (85.7%) out of the 84 MDR-TB cases occurred in non-austrians. All of the 72 MDR-TB cases in non-austrians came from high TB incidence countries, 24 (33%) were from the Russian Federation, 19 (26%) from other Eastern European high TB endemic countries such as Armenia, Azerbaijan, Georgia, Moldova and Ukraine, and 12 (17%) from countries in South and South-East Asia, and the Western Pacific region, such as China, India, Mongolia and Vietnam. Our findings were consistent with the WHO report of 2007 which stated that indigenous MDR-TB occurrence is rare in Western and Central Europe as compared to the Russian Federation and Eastern European countries, that were previously part of the former Soviet Union [25]. The trend in the annual proportion of MDR-TB cases among all tested TB cases in non-austrians between 1997-2006, showed that the highest proportion of MDR-TB cases was in 1999 and 2004. The proportion of MDR-TB cases in all cases in Austrians remained unchanged throughout the study period. The incidence of MDR-TB cases in non-austrians peaked between 1997-2006, at first in 1999, and again in 2004 (0.72 and 2.54/100,000 population respectively). The number of immigrants from high TB incidence countries (TB incidence 20/ 100,000 population) also peaked in 1999 and 2004. Correlation analysis of the annual number of MDR-TB cases in non-austrians with the annual number of immigrants from high TB incidence countries, showed that 74.5% of the variation between 1997 and 2006 in the annual number of MDR-TB cases among non-austrians could be explained by the linear relationship with annual immigration from high TB incidence countries to Austria [Figure 52]. Interpretation 130

During the 6-year period between 1997 and 2002, the average annual number of MDR-TB cases was 4.8, primarily dominated by cases in non-austrians. The non- Austrians to Austrians ratio of MDR-TB was 2.2. The average annual number of MDR-TB cases during the successive 4-year period, 2003 to 2006 was 14.8, with a non-austrians to Austrians ratio of 13.8. The steep increase of MDR-TB cases in non-austrians between 2003 and 2004 was determined by MDR-TB cases in non-austrians originating from the Russian Federation, and from the high TB incidence Eastern European countries. As immigration from the Russian Federation peaked in 2004, so did the MDR-TB cases in non-austrians originating from that country (n=9). The same observations were made in relation to high TB incidence Eastern European countries, with a peak in MDR-TB cases (n=7) in 2004, along with a peak in immigration from that region (n=19,371) in the same year [Figure 53]. Interpretation 131

Risk of becoming a case of TB after arriving in Austria. According to our 10-year data set on TB cases among immigrants who entered Austria between 1997 and 2006 and became cases of TB, 92.2% of these immigrants developed the disease within 5 years after arrival; 72.5% developed TB within 2 years; and 59% within 1 year after arrival in Austria [Figure 59]. Figure 59: Distribution of non-austrian TB cases by time elapsed from arrival until date of diagnosis of TB (categories: date of diagnosis at entry, within 1 year, within 2 years, 3-5 years, more than 5 years after arrival in Austria) 70% 60% 59.0% 50% Proportion (%) 40% 30% 20% 13.5% 19.7% 10% 0% 2.2% Diagnosed as a TB case upon entry into Austria 5.5% 1 year 2 years 3-5 years >5 years Years elapsed since entering Austria Coker et al. (http://www.tbalert.org/resources/paper_pub/tb_in_britain/coker.pdf) described similar observations in immigrants to the UK: 50% of those born abroad who developed TB in the UK, did so within 5 years after entry (see above figure: Interpretation 132

Tuberculosis case reports born abroad by time since entry into the UK, England and Wales, 1998-2001 n=9,725). Dilip et al. [25] discovered similar findings among immigrants to New Zealand, born in high TB incidence countries. The highest number of TB cases among these immigrants occurred within the first year after arrival, followed by decreasing numbers of TB cases in subsequent years (see figure below; graph left). In contrast, very few people from low TB incidence countries were diagnosed with TB within the first year after arrival in New Zealand. Most of those who developed the disease did so 20 years or more after arrival (see graphs below) [26]. A TB study among entrants to the U.S. consistently demonstrated that the length of residence in the U.S. is inversely related to the risk of tuberculosis among foreign-born persons [27]. Immigrants from high TB-incidence countries Immigrants from low TB-incidence countries Only 30 (2.2%) entrants out of the 1,338 TB cases among non-austrians who immigrated between 1997 and 2006 and in which month/year of immigration was known, presented with active tuberculosis at the time of entry into Austria. A study of the screening procedure among U.S.-bound immigrants coming from Vietnam demonstrated an 8.4% proportion of entrants with active TB present upon entry into the U.S. [28]. This reflects the high TB burden in that Asian country. About 90% of untreated infected individuals will never develop active TB. Of the remaining 10% who do develop the disease, about 50% will usually develop TB Interpretation 133

during the first 2 years following infection [1]. Almost 73% of the TB cases among non-austrians who entered into Austria (98.5% (955/970) of these cases were coming from high TB incidence countries) between 1997 and 2006, developed the disease within 2 years after their arrival and 92.2% of these immigrants (98.4% (1,214/1,234) of these cases were coming from high TB incidence countries) did so within 5 years after arrival. This short period of latency in the majority of immigrants indicates that in the majority of TB cases among non-austrians who entered Austria between 1997-2006 from high TB-incidence countries, the infection was acquired in the countries of origin (i.e. imported cases) [Figure 59]. Spread of TB from native Austrians to newly-immigrated non-austrians is due to the low incidence of TB in Austrians and the minimal social interaction between these two population groups very unlikely. These findings are consistent with observations from a study on TB among Somali immigrants in Demark. 62.4% of TB cases among Somali immigrants occurred during their first 2 years of residence in Denmark [21]. Tuberculin skin testing of 300 Somalis in Denmark indicated that 80%-90% of all adults (16-49 years of age), and 25% of all children were infected with M. tuberculosis at the time of arrival. Similar observations were also made in the U.S. as the great increase in the number of immigrants between 1989 and 1992 correlated positively with the increase in TB incidence among the foreign-born population [20]. The findings that immigrants to Austria from countries with TB incidence of 20/100,000 (primarily former Yugoslavia, Turkey, the Russian Federation and other Eastern European countries) were 17 times more likely to become a TB case compared to immigrants from countries with a TB incidence of < 20/100,000 (mostly from Western European countries, e.g. Germany), reflect the TB burden in the countries of origin. This also underlines the conclusion that the majority of non-austrians in Austria with TB acquired the infection already in their home countries. The WHO / Regional office for Europe concluded in a 2007 Interpretation 134

publication that the incidence of TB among foreign-born populations living in Europe was up to 50 times greater than the incidence among native populations. (http://www.euro.who.int/tuberculosis/publications/20071204_10). Between 1997 and 2006 there were 970 TB cases (27.8%) among non-austrian refugees. The annual proportion of refugees among all TB cases in non-austrians increased from 11% in 1997, to the first peak of 22% in 1999, followed by a slight decrease in 2000, and a steep increase to the second peak of 45.9% in 2004 [Figure 60, not given in the result part]. Figure 60: Annual proportion of refugees among all non-austrian TB cases, and the annual number of immigrants coming from high TB incidence countries The annual proportion of refugees among TB cases that occurred in non- Austrians between 1997 and 2006 correlated positively with the annual number of immigrants from high TB incidence countries between 1997-2006. 94.1% of the variation in the annual proportion of refugees among non-austrian TB cases between 1997-2006, could be explained by the linear relationship with the annual immigration figures from high TB incidence countries to Austria. Interpretation 135