Understanding Tuberculosis Transmission in the United Kingdom: Findings From 6 Years of Mycobacterial Interspersed Repetitive Unit–Variable Number Tandem Repeats Strain Typing, 2010–2015

Understanding Tuberculosis Transmission in the United Kingdom: Findings From 6 Years of... Abstract Genotyping provides the opportunity to better understand tuberculosis (TB) transmission. We utilized strain typing data to assess trends in the proportion of clustering and identify the characteristics of individuals and clusters associated with recent United Kingdom (UK) transmission. In this retrospective cohort analysis, we included all culture-confirmed strain-typed TB notifications from the UK between 2010 and 2015 to estimate the proportion of patients that clustered over time. We explored the characteristics of patients in a cluster using multivariable logistic regression. Overall, 58.5% of TB patients were concentrated in 2,701 clusters. The proportion of patients in a cluster decreased between 2010 (58.7%) and 2015 (55.3%) (P = 0.001). Being a clustered patient was associated with being male and UK-born, having pulmonary disease, having a previous TB diagnosis, and having a history of drug misuse or imprisonment. Our results suggest that TB transmission in the UK decreased between 2010 and 2015, during which time TB incidence also decreased. Targeted cluster investigation and extended contact tracing should be aimed at persons at risk of being in a transmission chain, including UK-born individuals with social risk factors in clusters with a high proportion of patients having pulmonary disease. disease transmission, genotyping, surveillance, tuberculosis Advances in genotyping methods and their application to the molecular epidemiology of infectious diseases, including tuberculosis (TB), have enabled utilization of outputs by public health experts. TB is a complex disease, requiring multifaceted control methods. One aim of control strategies is to decrease transmission, and many high-resource countries have now incorporated genotyping methods to achieve this (1–8). In response to increasing TB incidence in the United Kingdom (UK) during the early 2000s (5), improvements in TB control were recommended (9), including the incorporation of genotyping into surveillance. In 2010, prospective routine typing of TB isolates using 24-locus mycobacterial interspersed repetitive unit-variable number tandem repeats (MIRU-VNTR) was established. This identified clustered patients with culture-positive TB isolates with indistinguishable MIRU-VNTR profiles, which may reflect individuals being part of a transmission chain. Culture confirmation in the UK has been stable since this time at around 60%, while coverage of genotyping has fluctuated over time (between 70% and 88%) and by geographical region (between 67% and 88%) (5). Cluster results were distributed weekly to local health protection teams through an electronic system which linked epidemiologic and molecular data to facilitate cluster review and, where necessary, investigation (10). This process led to detection of transmission settings in which public health action, such as enhanced screening or awareness raising, could be conducted. This identified additional individuals with active disease or latent infection, ensuring early diagnosis of these cases and preventing further onward transmission. However, because of the low discriminatory power of MIRU-VNTR, clustered patients could also reflect common endemic strains circulating in the UK or abroad, necessitating additional epidemiologic information to assess the likelihood of recent transmission. In the UK, where a large proportion of TB cases are probably due to infection acquired abroad, identifying the characteristics of persons who may have transmitted TB to others within the UK, as well as those recently infected in the UK, could identify settings for public health action and enable better targeted policies to interrupt transmission. Additionally, genotyping data are informative for assessing the epidemiology of transmission at a population level. Our objectives in this study were 1) to assess trends in TB clustering in the UK, using 6 years of available MIRU-VNTR data to identify the proportion of patients in a cluster as a proxy measure for transmission, and 2) to identify populations that may have been involved in recent UK transmission, based on the characteristics of clusters and the individuals in clusters. Identification of such populations will facilitate better targeted cluster investigation and public health action. METHODS Study population From 2010 onward, at least 1 isolate from all culture-confirmed Mycobacterium tuberculosis complex cases was prospectively typed using 24-locus MIRU-VNTR. We included all patients with notifications in the UK between January 1, 2010, and December 31, 2015, for whom at least 23 loci were typed. Data collection Notifications of TB cases made to the Enhanced TB Surveillance System in England, Wales, and Northern Ireland and the Enhanced Surveillance of Mycobacterial Infections (ESMI) system in Scotland between 2010 and 2015 were matched (11) to culture-positive results received from various branches of the National Mycobacterium Reference Laboratory. Strain typing and drug susceptibility results were available for culture-positive isolates. The following information was collected with notification of each patient’s case: demographic characteristics (age, sex, place of residence, ethnicity, country of birth), clinical characteristics (site of disease, previous diagnosis), and social risk factors (current or history of imprisonment, drug use, alcohol misuse, or homelessness). Definitions A molecular cluster (hereafter referred to as a cluster) was defined as a group of TB cases containing at least 2 patients with isolates with indistinguishable MIRU-VNTR profiles (at least 1 with complete 24-locus MIRU-VNTR and others with at least 23 loci typed) (10). A clustered patient was defined as a TB patient in a cluster, and a nonclustered patient was defined as a TB patient with a strain type distinguishable from that of all other patients. Two time periods were used to define patients as being in a cluster: the entire period of 2010–2015, to identify annual proportions of patients who were clustered with any other patient(s) during that time period, and 2-year intervals, where at least 2 cases from a cluster occurred within those 2 years. For the annual analysis, clusters were defined as new in the year in which the second case occurred. In the 2-year interval analysis, clusters were defined as new when both the first case and the second case occurred within the 2-year period. Clusters were defined as accruing when 2 or more cases (regardless of order; first, second, or subsequent) occurred within the period of interest (annual or 2-year). Within this paper, where year is presented, this is the year of patient notification. The sizes of clusters were defined as small (<5 patients), medium (5–20 patients), or large (>20 patients). Lineage was derived from MIRU-VNTR strain type (12) as Euro-American, Central Asian strain (CAS), East African–Indian (EAI), Beijing, Mycobacterium africanum, or Mycobacterium bovis. Large lineage diversity was defined as <50% of all patients having the same lineage, moderate lineage diversity as ≥50% but <80% of all patients having the same lineage, and little lineage diversity as ≥80% of patients having the same lineage. Data analysis We calculated the annual proportions of patients in a cluster with other patient(s) between 2010 and 2015 to assess the trend over time. To account for the fact that patients whose TB occurred in earlier years had a longer time to form a cluster with future patients, 2-year intervals were used. The maximum proportion of patients involved in transmission was estimated using the “n – 1” method: (number of clustered patients minus number of clusters)/number of patients with a strain type of 23 loci (13). The significance of trends was assessed using the χ2 test for trend. The characteristics of clustered patients were compared with those of nonclustered patients. Univariable logistic regression was performed to identify characteristics associated with being in a cluster, using odds ratios. Any characteristics associated with the outcome at P ≤ 0.2 were included in multivariable analysis. The significance of the main effects was assessed using likelihood ratio testing, with P ≤ 0.05 considered significant. Interactions between biologically and statistically plausible variables in the model were tested using likelihood ratios. Medium and large TB clusters are of greater public health significance than small ones, as more than 1 transmission event will probably have occurred. We described the characteristics of medium and large clusters using summary statistics and proportions, for all clusters in total and separately for clusters containing only UK-born patients (which are most likely to represent UK transmission), non-UK-born patients, and all other patients (both UK- and non-UK-born patients). We characterized the proportions of all clustered and nonclustered patients from each lineage by ethnic group for UK-born patients and by the 8 most frequent non–European Union countries of birth and the 2 most frequent European Union countries of birth for non-UK-born patients. RESULTS Between 2010 and 2015, 82.3% (23,646/28,741) of culture-confirmed TB patients had an isolate with a MIRU-VNTR profile typed to at least 23 loci (Figure 1). Overall, 58.5% (13,844/23,646) of these patients were concentrated in 2,701 clusters. There were 12,380 different strains identified. The proportion of patients in a cluster varied by geographical region of the UK (see Web Table 1, available at https://academic.oup.com/aje) and were closely aligned with regional TB incidence; the lowest proportion was identified in Wales (24.6%) and the highest in London, England (51.5%). Over half (55.1%; 1,487/2,701) of clusters were found only within 1 UK region; a much smaller proportion (12.5%; n = 337) of clusters crossed regional boundaries; and the remaining clusters (32.5%; n = 877) were only national clusters with no regional focus (i.e., there were not 2 or more patients within a single region). Figure 1. View largeDownload slide Numbers of tuberculosis (TB) patients, culture-confirmed patients, strain-typed patients, and TB clusters, United Kingdom (UK), 2010–2015. Figure 1. View largeDownload slide Numbers of tuberculosis (TB) patients, culture-confirmed patients, strain-typed patients, and TB clusters, United Kingdom (UK), 2010–2015. Trends in proportion of patients in a cluster The annual proportions of patients in clusters remained stable between 2010 and 2013 (range, 58.7%–60.7%) and decreased in 2014 and 2015 (56.9% and 55.3%, respectively) (χ2 test: P-trend = 0.001). The number of new clusters formed annually ranged from 410 in 2010 to 586 in 2012, with the lowest being 285 in 2015 (Table 1). Overall, the maximum recent transmission estimate (n − 1) over the 6-year study period was 47%. The annual estimate increased between 2010 and 2013 (range, 27.4%–31.2%) before decreasing to its lowest level (24.9%) in 2014. Table 1. Annual and 2-Year Proportions of Clustering of Tuberculosis Cases and Number of Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Abbreviations: TB, tuberculosis; UK, United Kingdom. a For single years, this is the number of clusters (new or existing) that cases occurring in that year are in, regardless of whether there are any other patients in the same cluster that year. For 2-year intervals, this is the number of clusters in which 2 or more cases occurred within the 2-year period. b For single years, this is the number of clusters in which the second case forming the cluster occurred in that year. For 2-year intervals, this is the number of clusters in which the first and second cases in the cluster occurred within that 2-year period, forming the cluster. Table 1. Annual and 2-Year Proportions of Clustering of Tuberculosis Cases and Number of Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Abbreviations: TB, tuberculosis; UK, United Kingdom. a For single years, this is the number of clusters (new or existing) that cases occurring in that year are in, regardless of whether there are any other patients in the same cluster that year. For 2-year intervals, this is the number of clusters in which 2 or more cases occurred within the 2-year period. b For single years, this is the number of clusters in which the second case forming the cluster occurred in that year. For 2-year intervals, this is the number of clusters in which the first and second cases in the cluster occurred within that 2-year period, forming the cluster. Analysis of 2-year periods showed that the proportion of patients in a cluster was highest in 2011–2012 (50.5%) and lowest in 2014–2015 (43.7%) (χ2 test: P-trend < 0.001). The number of accruing clusters formed within the 2-year periods followed the same pattern (1,146 clusters in 2011–2012 vs. 777 clusters in 2014–2015), and the numbers of new clusters formed decreased year-on-year between 2010–2011 (976 clusters) and 2014–2015 (179 clusters). The proportion of clustering was consistently higher in UK-born patients than in non-UK-born patients, but the trend over time was similar for all patients. No change in the size of clusters was observed over time (small clusters of <5 patients: 80.7% in 2010–2011 vs. 81.2% in 2014–2015 (χ2 test: P-trend = 0.162)). Additionally, there was no change in the age profile (age 15–44 years: 67.2% in 2010–2011 vs. 62.4% in 2014–2015 (χ2 test: P-trend = 0.162)) or place of birth (UK-born: 32.1% in 2010–2011 vs. 34.9% in 2014–2015 (χ2 test: P-trend = 0.162)) for clustered patients. There was a decrease in the proportion of non-UK-born clustered patients who had recently (<2 years previously) entered the UK (24.9% in 2010–2011 vs. 15.8% in 2014–2015 (χ2 test: P-trend = 0.072)), which mirrored the overall change in time between UK entry and TB for all patients observed from 2010 to 2015 (data not shown). Characteristics of patients in clusters The proportion of patients in a cluster varied by place of birth: 69.8% (4,202/6,022) of UK-born patients were in a cluster, as compared with 54.7% (9,241/16,894) of non-UK-born patients. Additionally, the proportion of patients in a cluster varied by age: 74.6% (308/413) of those aged ≤14 years, as compared with 47.9% (1,571/3,281) of those aged ≥65 years. Within each age group, a higher proportion of UK-born patients were in a cluster—most notably, 82.0% (191/233) of UK-born patients aged ≤14 years as compared with 63.6% (103/162) of non-UK-born patients aged ≤14 years. Characteristics associated with being in a cluster Characteristics associated with being in a cluster were identified in a comparison of 13,844 clustered patients with 9,802 nonclustered patients (Table 2). In univariable analysis, clustered patients were more likely to be under age 20 years, to be male, to be UK-born (or, if non-UK-born, to have lived in the UK for 6 years or more), to have a previous TB diagnosis, to have a social risk factor, and to have a strain with CAS or Beijing lineage. Table 2. Characteristics Associated With Being in a Tuberculosis Cluster (Univariable and Multivariable Logistic Regression), United Kingdom, 2010–2015a Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Abbreviations: CI, confidence interval; OR, odds ratio; SS+, sputum-smear–positive; UK, United Kingdom. a Totals for some characteristics do not equal the total number of patients because of missing information. b Years since entry into the UK was not included in the multivariable analysis, since it was only relevant to non-UK-born patients. c Isoniazid-resistant TB was not included in multivariable analysis because it was found not to be associated with the outcome in univariable analysis (see Methods section). d Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. Table 2. Characteristics Associated With Being in a Tuberculosis Cluster (Univariable and Multivariable Logistic Regression), United Kingdom, 2010–2015a Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Abbreviations: CI, confidence interval; OR, odds ratio; SS+, sputum-smear–positive; UK, United Kingdom. a Totals for some characteristics do not equal the total number of patients because of missing information. b Years since entry into the UK was not included in the multivariable analysis, since it was only relevant to non-UK-born patients. c Isoniazid-resistant TB was not included in multivariable analysis because it was found not to be associated with the outcome in univariable analysis (see Methods section). d Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. The multivariable model adjusted for age, sex, ethnic group, site of disease, social risk factors, previous TB diagnosis, area of residence, having multiple-drug-resistant TB, and strain lineage. Men were more likely to be in a cluster than women (adjusted odds ratio (aOR) = 1.1, 95% confidence interval (CI): 1.0, 1.2). Compared with those aged 20–29 years, patients aged 60–69 years (aOR = 0.8, 95% CI: 0.7, 1.0) and patients aged 70 years or more (aOR = 0.5, 95% CI: 0.5, 0.6) were less likely to be in a cluster. UK-born patients were more likely to be in a cluster, irrespective of ethnic group. Patients with pulmonary disease (smear-positive: aOR = 1.4 (95% CI: 1.3, 1.5); smear- negative/unknown: aOR = 1.3 (95% CI: 1.2, 1.4)) were more likely to be in a cluster than those with extrapulmonary disease only. Additionally, patients with a previous TB diagnosis (aOR = 1.3, 95% CI: 1.1, 1.5), a history of drug misuse (aOR = 1.7, 95% CI: 1.4, 2.1), or a history of imprisonment (aOR = 1.4, 95% CI: 1.1, 1.7) were more likely to be in a cluster. Compared with patients with a strain of Euro-American lineage, patients with a strain of Beijing (aOR = 2.3, 95% CI: 2.0, 2.8) or CAS (aOR = 1.5, 95% CI: 1.4, 1.6) lineage were more likely to be in a cluster, while patients with an EAI lineage were less likely to be (aOR = 0.7, 95% CI: 0.6, 0.7). We identified an interaction between some ethnic groups and strain lineages. UK-born Pakistani patients with an EAI lineage had lower odds of being in a cluster (odds ratio (OR) = 0.3, 95% CI: 0.1, 0.9), as did non-UK-born Bangladeshi patients with an EAI lineage (OR = 0.4, 95% CI: 0.2, 0.6). UK-born white patients with a CAS lineage had higher odds of being in a cluster (OR = 1.8, 95% CI: 1.2, 2.7), as did non-UK-born black African patients with a CAS lineage (OR = 1.4, 95% CI: 1.1, 1.9). Finally, non-UK-born white patients with a Beijing lineage had increased odds of being in a cluster (OR = 2.8, 95% CI: 1.2, 6.7). The interaction effect estimates corresponded to a further difference in odds, in addition to the main effects presented in Table 2 for ethnicity and lineage. Characteristics of clusters The 13,844 clustered patients were concentrated in 2,701 different clusters. The median number of patients in a cluster was 3 (range, 2–242). The majority of clusters (74.3%; 2,008/2,701) were small (<5 patients) in size; 22.6% (610 clusters) were medium (5–20 patients), and 3.1% (83 clusters) were large (>20 patients). Over half of clusters (51.3%; 1,386/2,701) contained only non-UK-born patients, and 8.3% (224 clusters) contained only UK-born patients (Figure 1, Table 3). For clusters comprised of only UK-born patients and only non-UK-born patients, the median number of patients in a cluster was 2 (range, 2–12 for UK-born patient clusters and 2–24 for non-UK-born patient clusters). Table 3. Characteristics of Tuberculosis Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Abbreviations: EPTB, extrapulmonary tuberculosis; PTB, pulmonary tuberculosis; SS+, sputum-smear–positive; UK, United Kingdom. a Clusters were defined as small (<5 patients), medium (5–20 patients), or large (>20 patients). b Values are presented as median (range). c Values are presented as median (interquartile range). d Current or history of imprisonment, drug use, alcohol misuse, or homelessness. e Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. Table 3. Characteristics of Tuberculosis Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Abbreviations: EPTB, extrapulmonary tuberculosis; PTB, pulmonary tuberculosis; SS+, sputum-smear–positive; UK, United Kingdom. a Clusters were defined as small (<5 patients), medium (5–20 patients), or large (>20 patients). b Values are presented as median (range). c Values are presented as median (interquartile range). d Current or history of imprisonment, drug use, alcohol misuse, or homelessness. e Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. In medium and large clusters, the median age in clusters containing only UK-born patients (210 clusters) was higher (45 years; interquartile range, 39–57) than the median age of patients in all clusters (35 years; interquartile range, 30–43) (Table 3). Sixty-eight percent (67.7%; 21 clusters) of such clusters (with only UK-born patients) contained only pulmonary patients, and in 25.8% (8 clusters) more than half of the cluster members had a social risk factor. Conversely, in 4 clusters containing only non-UK-born patients, all had extrapulmonary disease. Lineage diversity Where the lineage of the bacteria infecting patients could be determined (87.8% of patients), the most frequent lineages were Euro-American (45.3%; 9,412 patients) and CAS (30.6%; 6,362 patients) strains, with fewer EAI (15.8%; 3,279 patients) and Beijing (6.5%; 1,356 patients) strains and very small numbers of M. africanum (1.0%; 208 patients) or M. bovis (0.7%; 149 patients) strains. The lineage distribution differed between clustered and nonclustered patients (χ2 test: P < 0.001), most notably for EAI (10.8% vs. 23.1%) and Beijing (8.4% vs. 3.9%) strains. Euro-American lineage was the most frequent lineage in patients from sub-Saharan African countries (Nigeria, 81.7%; Zimbabwe, 83.0%) and Eastern European countries (Romania, 98.9%; Poland, 97.4%). The CAS lineage was the most frequent in patients from Pakistan (73.2%) and India (51.4%), while the EAI lineage was most frequent in patients from the Philippines (93.6%) and Bangladesh (62.8%). The majority of UK-born patients had cases with Euro-American lineage (67.8%); however, lineage varied by ethnic group: Pakistani and Indian patients had a high proportion of CAS strains (52.6% and 42.0%, respectively), similar to patients born in those countries (Pakistan, 65.1%; India, 45.8%). The proportion of patients with TB strains from each lineage differed between clustered and nonclustered patients (Web Figure 1), particularly among countries of birth that showed large (Somalia and Nepal) or moderate (India, Pakistan, Bangladesh, and the UK) lineage diversity. This was particularly evident in patients from Bangladesh with a Euro-American lineage (31.2% clustered vs. 13.2% nonclustered) and an EAI lineage (39.7% clustered vs. 75.3% nonclustered), patients from Somalia with a CAS lineage (36.1% clustered vs. 20.2% nonclustered), and patients from Nepal with an EAI lineage (3.5% clustered vs. 24.9% nonclustered). For all countries of birth, other than the Philippines and Nigeria, a higher proportion of clustered patients than nonclustered patients had Beijing and CAS lineages, while fewer clustered patients had an EAI lineage. DISCUSSION In this paper, we report results from a population-based study carried out in the UK using MIRU-VNTR to describe trends and features of possible TB transmission based on the proportion of patients in a cluster over time, as well as the characteristics associated with being in a cluster. We identified decreases in the proportions of patients in clusters and in numbers of new clusters between 2010 and 2015. Annual proportions of patients in a cluster (55.3%–60.7%) and the maximum transmission estimate (47%) identified were higher than findings from other low-incidence countries and previous UK estimates provided for shorter time periods or at a local level (2, 3, 14). The low discriminatory power of MIRU-VNTR is known to overestimate transmission (15), since a proportion of patients who are part of a cluster will have TB due to reactivation or importation of common endemic strains. Although measuring TB transmission at a population level is difficult, combining several pieces of evidence can help us understand trends in population-level transmission over time. The rate of TB in UK-born children, which is used as an indicator of recent transmission, decreased between 2010 and 2015 (5). Additionally, the number of TB notifications and the rate of TB decreased over the same period, with this decrease occurring among both the UK-born and non-UK-born populations (5). The reduction in the proportion of patients in a cluster, combined with these additional epidemiologic trends, is suggestive of a decrease in TB transmission in the UK during the study period. Improvements in TB control methods which promote case-finding and contact tracing may have affected transmission, contributing to the decrease in the proportion of patients identified as being in a cluster, or may have increased the number of sporadic (not clustered) cases identified. In addition to the timely screening of contacts and additional settings guided by strain typing (16, 17), interventions include “find and treat” (18) and cohort review, which includes presentation of and accountability for the number of contacts identified and screened (19). It is also possible that part of the decrease in the proportions of patients in clusters is due not to changes in transmission but to a reduction in the number of non-UK-born TB patients in the UK, particularly recent migrants (5), because of preentry screening and latent TB infection screening (20–22). With fewer non-UK-born patients, the reduction in clustering among these patients may have led to fewer introductions of common strains into the UK. Our study identified the following characteristics associated with being a clustered patient: being UK-born, being non-UK-born of black ethnicity, current or past drug misuse, and current or past imprisonment. Note that these are the characteristics of patients in clusters; therefore, they include persons who may have transmitted TB as well as those to whom TB may have been transmitted. People who misuse drugs may be less likely to seek health care or may be more likely to have a negative health-care experience, contributing to increased transmission within this population (23–25). Additionally, they may be less willing to name contacts due to the illegal nature of their activities or may have complex social networks, resulting in missed opportunities for contact tracing. Imprisonment is a risk factor associated with TB in both the UK, with sporadic transmission previously reported (26, 27), and globally (28). Imprisonment reported here included current or past imprisonment in the UK or abroad; therefore, the identification of an association between clustering and imprisonment resulted from historical circulation of strains in prisons in addition to recent transmission. UK-born black Caribbean and black African and non-UK-born black Caribbean patients were most likely to be in a cluster, while the Bangladeshi patients were more likely to have a unique strain. The reasons for these observed differences between ethnic groups are unknown but could include socioeconomic and/or behavioral differences, including engagement with health services. Nevertheless, our findings suggest that transmission dynamics within these ethnic groups, as well as among those with social risk factors, should be considered during the prioritization of cluster investigation and the targeting of resources for public health action. This may include the use of enhanced data collection to identify behaviors and settings that may be associated with increased transmission risk (16, 29). The patient and cluster characteristics we identified suggest that recent transmission is less likely to have occurred among non-UK-born patients, particularly those in clusters containing only non-UK-born individuals. Many of these clusters probably represent reactivation of common endemic strain types circulating abroad. Similar findings have been identified elsewhere (15). However, transmission does still occur among non-UK-born individuals in the UK; this maybe particularly true in urban areas (16) and in specific migrant populations (30). The lineage of M. tuberculosis complex strains isolated from patients differed by country of birth and largely corresponded to the strains that were predominant in their countries of origin (31, 32). This provides additional evidence that a large proportion of TB in the UK is imported, due to either arrival of patients with active disease or subsequent reactivation of latent infection. Among UK-born patients, lineage differed by ethnic group and often closely resembled that in the countries of origin of previous generations of migrants (i.e., parents or grandparents). This could be due to acquisition of infection after receiving visitors from, or traveling to, the family’s country of origin. Additionally, for both UK-born and non-UK-born populations, household or community interaction with persons from the country of origin may occur (16, 33, 34). CAS and Beijing lineage strains were more frequent among clustered patients, while the EAI lineage was more frequent in those with a unique strain. These findings suggest that lineage may play a role in transmission dynamics. However, in analysis of household transmission in the UK, where patients are molecularly and epidemiologically linked, associations between lineage and clustering were not identified (33). It is also possible that our results were due to differences in the ability of MIRU-VNTR typing to distinguish between strains of different lineages. Whole-genome sequencing analysis identifies single nucleotide polymorphism differences between isolates, a more discriminatory measure of relatedness than that yielded by MIRU-VNTR (35). In 2017, the UK replaced MIRU-VNTR with whole-genome sequencing to determine relatedness between TB isolates. This will bring many benefits, including refined estimates of the level of transmission and more accurate identification of the characteristics of persons transmitting TB. However, only data collected over several years enables investigation of trends and risk groups. Therefore, the findings from this study using MIRU-VNTR data, particularly trends in clustering over time, are helpful in evaluating recent TB control efforts and in targeting further TB control efforts, including enhanced contact tracing, which can later be refined on the basis of whole-genome sequencing results. Systematic collection of epidemiologic links between patients will also facilitate the identification of transmission settings in the UK (16, 33, 34) and could additionally refine the estimation of transmission rates within settings (33). Developing and applying more sophisticated methods for estimating transmission beyond those utilized here, thereby refining the estimates provided by our current analysis to account for reactivation, super-spreaders, and the plausibility of transmission, including temporal and spatial analysis, would be beneficial. Such methods have been used in the United States (4, 8, 36, 37) and locally in the UK (38), resulting in refined transmission estimates. Our study benefited from having a large national data set spanning a period of 6 years in which prospective routine strain typing was performed. It will take several years of continuous use of whole-genome sequencing before such trends in transmission can be measured again. In addition, high-quality data on demographic factors, clinical characteristics, and social risk factors were collected for the TB patients. However, our study also had several limitations, as we have acknowledged throughout this Discussion where relevant. Additionally, between 2010 and 2015, only 61% of TB cases in the UK were culture-confirmed, although this proportion was higher (72%) among pulmonary patients (5). Among culture-confirmed cases, a small proportion were not typed (2.4%) or typed to 23 loci (15.3%). In conclusion, our findings suggest that TB transmission in the UK decreased between 2010 and 2015, during which time TB incidence in the UK also decreased. Our results emphasize the need for TB control efforts for reducing transmission to be targeted toward UK-born populations, particularly persons belonging to black Caribbean and black African ethnic groups, and those with social risk factors. While we were not able to determine why transmission may be more frequent in these groups, additional research should be conducted to investigate this in order to further tailor public health interventions. ACKNOWLEDGMENTS Author affiliations: Tuberculosis Unit, National Infection Service, Public Health England, London, United Kingdom (Jennifer A. Davidson, H. Lucy Thomas, Colin N. J. Campbell, Maeve K. Lalor); Field Service, National Infection Service, Public Health England, London, United Kingdom (Helen Maguire, Neil Macdonald); Institute for Global Health, University College London, London, United Kingdom (Helen Maguire, Colin N. J. Campbell, Maeve K. Lalor); National Mycobacterium Reference Service South, National Infection Service, Public Health England, London, United Kingdom (Timothy Brown); and Field Service, National Infection Service, Public Health England, Newcastle, United Kingdom (Andy Burkitt). No funding was obtained for this work. We thank Ross Harris of the Statistics Unit, National Infection Service, Public Health England, for statistical review and guidance provided at multiple points during this analysis. Conflict of interest: none declared. Abbreviations aOR adjusted odds ratio CAS Central Asian strain CI confidence interval EAI East African–Indian MIRU-VNTR mycobacterial interspersed repetitive unit–variable number tandem repeats OR odds ratio TB tuberculosis UK United Kingdom REFERENCES 1 Al-Hajoj S , Varghese B , Al-Habobe F , et al. . Current trends of Mycobacterium tuberculosis molecular epidemiology in Saudi Arabia and associated demographical factors . Infect Genet Evol . 2013 ; 16 : 362 – 368 . Google Scholar Crossref Search ADS PubMed 2 Lambregts-van Weezenbeek CS , Sebek MM , van Gerven PJ , et al. . Tuberculosis contact investigation and DNA fingerprint surveillance in the Netherlands: 6 years’ experience with nation-wide cluster feedback and cluster monitoring . Int J Tuberc Lung Dis . 2003 ; 7 ( 12 suppl 3 ): S463 – S470 . Google Scholar PubMed 3 Kamper-Jørgensen Z , Andersen AB , Kok-Jensen A , et al. . Clustered tuberculosis in a low-burden country: nationwide genotyping through 15 years . J Clin Microbiol . 2012 ; 50 ( 8 ): 2660 – 2667 . Google Scholar Crossref Search ADS PubMed 4 Moonan PK , Ghosh S , Oeltmann JE , et al. . Using genotyping and geospatial scanning to estimate recent Mycobacterium tuberculosis transmission, United States . Emerg Infect Dis . 2012 ; 18 ( 3 ): 458 – 465 . Google Scholar Crossref Search ADS PubMed 5 Public Health England . Tuberculosis in England: 2016 Report (Presenting Data to End of 2015) . London, United Kingdom : Public Health England ; 2016 . https://www.gov.uk/government/publications/tuberculosis-in-england-annual-report. Accessed October 26, 2016. 6 Sandgren A , Schepisi MS , Sotgiu G , et al. . Tuberculosis transmission between foreign- and native-born populations in the EU/EEA: a systematic review . Eur Respir J . 2014 ; 43 ( 4 ): 1159 – 1171 . Google Scholar Crossref Search ADS PubMed 7 Yang C , Shen X , Peng Y , et al. . Transmission of Mycobacterium tuberculosis in China: a population-based molecular epidemiologic study . Clin Infect Dis . 2015 ; 61 ( 2 ): 219 – 227 . Google Scholar Crossref Search ADS PubMed 8 Yuen CM , Kammerer JS , Marks K , et al. . Recent transmission of tuberculosis—United States, 2011–2014 . PLoS One . 2016 ; 11 ( 4 ): e0153728 . Google Scholar Crossref Search ADS PubMed 9 United Kingdom Department of Health . Stopping Tuberculosis in England: An Action Plan from the Chief Medical Officer . London, United Kingdom : Department of Health ; 2004 . http://webarchive.nationalarchives.gov.uk/+/http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4090417. Accessed January 23, 2017. 10 Public Health England . TB Strain Typing and Cluster Investigation Handbook. Third Edition, February 2014 . London, United Kingdom : Public Health England ; 2014 . https://www.gov.uk/government/publications/tb-strain-typing-and-cluster-investigation-handbook. Accessed October 28, 2015. 11 Aldridge RW , Shaji K , Hayward AC , et al. . Accuracy of probabilistic linkage using the Enhanced Matching System for public health and epidemiological studies . PLoS One . 2015 ; 10 ( 8 ): e0136179 . Google Scholar Crossref Search ADS PubMed 12 Brown T , Evans J , Sails A . HPA Mycobacterium tuberculosis Strain Typing: A Guide to Data Production and Distribution. London, United Kingdom : Health Protection Agency ; 2012 . https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/433731/HPA_VNTR_280812_final.pdf. Accessed July 3, 2016. 13 Glynn JR , Vynnycky E , Fine PE . Influence of sampling on estimates of clustering and recent transmission of Mycobacterium tuberculosis derived from DNA fingerprinting techniques . Am J Epidemiol . 1999 ; 149 ( 4 ): 366 – 371 . Google Scholar Crossref Search ADS PubMed 14 Hamblion EL , Le Menach A , Anderson LF , et al. . Recent TB transmission, clustering and predictors of large clusters in London, 2010–2012: results from first 3 years of universal MIRU-VNTR strain typing . Thorax . 2016 ; 71 ( 8 ): 749 – 756 . Google Scholar Crossref Search ADS PubMed 15 Stucki D , Ballif M , Egger M , et al. . Standard genotyping overestimates transmission of Mycobacterium tuberculosis among immigrants in a low-incidence country . J Clin Microbiol . 2016 ; 54 ( 7 ): 1862 – 1870 . Google Scholar Crossref Search ADS PubMed 16 Black AT , Hamblion EL , Buttivant H , et al. . Tracking and responding to an outbreak of tuberculosis using MIRU-VNTR genotyping and whole genome sequencing as epidemiological tools [published online ahead of print July 5, 2017]. J Public Health (Oxf) . ( doi:10.1093/pubmed/fdx075 ). 17 Smith CM , Trienekens SC , Anderson C , et al. . Twenty years and counting: epidemiology of an outbreak of isoniazid-resistant tuberculosis in England and Wales, 1995 to 2014 . Euro Surveill . 2017 ; 22 ( 8 ): 30467 . Google Scholar Crossref Search ADS PubMed 18 Jit M , Stagg HR , Aldridge RW , et al. . Dedicated outreach service for hard to reach patients with tuberculosis in London: observational study and economic evaluation . BMJ . 2011 ; 343 : d5376 . Google Scholar Crossref Search ADS PubMed 19 Anderson C , White J , Abubakar I , et al. . Raising standards in UK TB control: introducing cohort review . Thorax . 2014 ; 69 ( 2 ): 187 – 189 . Google Scholar Crossref Search ADS PubMed 20 Panchal RK , Browne I , Monk P , et al. . The effectiveness of primary care based risk stratification for targeted latent tuberculosis infection screening in recent immigrants to the UK: a retrospective cohort study . Thorax . 2014 ; 69 ( 4 ): 354 – 362 . Google Scholar Crossref Search ADS PubMed 21 Usdin M , Dedicoat M , Gajraj R , et al. . Latent tuberculous screening of recent migrants attending language classes: a cohort study and cost analysis . Int J Tuberc Lung Dis . 2017 ; 21 ( 2 ): 175 – 180 . Google Scholar Crossref Search ADS PubMed 22 Zenner D , Loutet MG , Harris R , et al. . Evaluating 17 years of latent tuberculosis infection screening in north-west England: a retrospective cohort study of reactivation . Eur Respir J . 2017 ; 50 ( 1 ): 1602505 . Google Scholar Crossref Search ADS PubMed 23 Lloyd C . The stigmatization of problem drug users: a narrative literature review . Drugs Educ Prev Policy . 2013 ; 20 ( 2 ): 85 – 95 . Google Scholar Crossref Search ADS 24 Neale J , Tompkins C , Sheard L . Barriers to accessing generic health and social care services: a qualitative study of injecting drug users . Health Soc Care Community . 2008 ; 16 ( 2 ): 147 – 154 . Google Scholar Crossref Search ADS PubMed 25 Van Boekel LC , Brouwers EP , van Weeghel J , et al. . Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: systematic review . Drug Alcohol Depend . 2013 ; 131 ( 1-2 ): 23 – 35 . Google Scholar Crossref Search ADS PubMed 26 Story A , Murad S , Roberts W , et al. . Tuberculosis in London: the importance of homelessness, problem drug use and prison . Thorax . 2007 ; 62 ( 8 ): 667 – 671 . Google Scholar Crossref Search ADS PubMed 27 Anderson C , Story A , Brown T , et al. . Tuberculosis in UK prisoners: a challenge for control . J Epidemiol Community Health . 2010 ; 64 ( 4 ): 373 – 376 . Google Scholar Crossref Search ADS PubMed 28 Baussano I , Williams BG , Nunn P , et al. . Tuberculosis incidence in prisons: a systematic review . PLoS Med . 2010 ; 7 ( 12 ): e1000381 . Google Scholar Crossref Search ADS PubMed 29 Gardy JL , Johnston JC , Ho Sui SJ , et al. . Whole-genome sequencing and social-network analysis of a tuberculosis outbreak . N Engl J Med . 2011 ; 364 ( 8 ): 730 – 739 . Google Scholar Crossref Search ADS PubMed 30 Munang ML , Browne C , Khanom S , et al. . Tuberculosis microepidemics among dispersed migrants, Birmingham, UK, 2004–2013 . Emerg Infect Dis . 2015 ; 21 ( 3 ): 524 – 527 . Google Scholar Crossref Search ADS PubMed 31 Gagneux S , Small PM . Global phylogeography of Mycobacterium tuberculosis and implications for tuberculosis product development . Lancet Infect Dis . 2007 ; 7 ( 5 ): 328 – 337 . Google Scholar Crossref Search ADS PubMed 32 Reed MB , Pichler VK , McIntosh F , et al. . Major Mycobacterium tuberculosis lineages associate with patient country of origin . J Clin Microbiol . 2009 ; 47 ( 4 ): 1119 – 1128 . Google Scholar Crossref Search ADS PubMed 33 Lalor MK , Anderson LF , Hamblion EL , et al. . Recent household transmission of tuberculosis in England, 2010–2012: retrospective national cohort study combining epidemiological and molecular strain typing data . BMC Med . 2017 ; 15 : 105 . Google Scholar Crossref Search ADS PubMed 34 Munang ML , Browne C , Evans JT , et al. . Programmatic utility of tuberculosis cluster investigation using a social network approach in Birmingham, United Kingdom . Int J Tuberc Lung Dis . 2016 ; 20 ( 10 ): 1300 – 1305 . Google Scholar Crossref Search ADS PubMed 35 Borgdorff MW , van Soolingen D . The re-emergence of tuberculosis: what have we learnt from molecular epidemiology? Clin Microbiol Infect . 2013 ; 19 ( 10 ): 889 – 901 . Google Scholar Crossref Search ADS PubMed 36 Althomsons SP , Kammerer JS , Shang N , et al. . Using routinely reported tuberculosis genotyping and surveillance data to predict tuberculosis outbreaks . PLoS One . 2012 ; 7 ( 11 ): e48754 . Google Scholar Crossref Search ADS PubMed 37 Althomsons SP , Hill AN , Harrist AV , et al. . Statistical method to detect tuberculosis outbreaks among endemic clusters in a low-incidence setting . Emerg Infect Dis . 2018 ; 24 ( 3 ): 573 – 575 . Google Scholar Crossref Search ADS PubMed 38 Saavedra-Campos M , Welfare W , Cleary P , et al. . Identifying areas and risk groups with localised Mycobacterium tuberculosis transmission in northern England from 2010 to 2012: spatiotemporal analysis incorporating highly discriminatory genotyping data . Thorax . 2016 ; 71 ( 8 ): 742 – 748 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Epidemiology Oxford University Press

Understanding Tuberculosis Transmission in the United Kingdom: Findings From 6 Years of Mycobacterial Interspersed Repetitive Unit–Variable Number Tandem Repeats Strain Typing, 2010–2015

Loading next page...
 
/lp/ou_press/understanding-tb-transmission-in-the-uk-findings-from-six-years-of-qIEbRwNiU3
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0002-9262
eISSN
1476-6256
D.O.I.
10.1093/aje/kwy119
Publisher site
See Article on Publisher Site

Abstract

Abstract Genotyping provides the opportunity to better understand tuberculosis (TB) transmission. We utilized strain typing data to assess trends in the proportion of clustering and identify the characteristics of individuals and clusters associated with recent United Kingdom (UK) transmission. In this retrospective cohort analysis, we included all culture-confirmed strain-typed TB notifications from the UK between 2010 and 2015 to estimate the proportion of patients that clustered over time. We explored the characteristics of patients in a cluster using multivariable logistic regression. Overall, 58.5% of TB patients were concentrated in 2,701 clusters. The proportion of patients in a cluster decreased between 2010 (58.7%) and 2015 (55.3%) (P = 0.001). Being a clustered patient was associated with being male and UK-born, having pulmonary disease, having a previous TB diagnosis, and having a history of drug misuse or imprisonment. Our results suggest that TB transmission in the UK decreased between 2010 and 2015, during which time TB incidence also decreased. Targeted cluster investigation and extended contact tracing should be aimed at persons at risk of being in a transmission chain, including UK-born individuals with social risk factors in clusters with a high proportion of patients having pulmonary disease. disease transmission, genotyping, surveillance, tuberculosis Advances in genotyping methods and their application to the molecular epidemiology of infectious diseases, including tuberculosis (TB), have enabled utilization of outputs by public health experts. TB is a complex disease, requiring multifaceted control methods. One aim of control strategies is to decrease transmission, and many high-resource countries have now incorporated genotyping methods to achieve this (1–8). In response to increasing TB incidence in the United Kingdom (UK) during the early 2000s (5), improvements in TB control were recommended (9), including the incorporation of genotyping into surveillance. In 2010, prospective routine typing of TB isolates using 24-locus mycobacterial interspersed repetitive unit-variable number tandem repeats (MIRU-VNTR) was established. This identified clustered patients with culture-positive TB isolates with indistinguishable MIRU-VNTR profiles, which may reflect individuals being part of a transmission chain. Culture confirmation in the UK has been stable since this time at around 60%, while coverage of genotyping has fluctuated over time (between 70% and 88%) and by geographical region (between 67% and 88%) (5). Cluster results were distributed weekly to local health protection teams through an electronic system which linked epidemiologic and molecular data to facilitate cluster review and, where necessary, investigation (10). This process led to detection of transmission settings in which public health action, such as enhanced screening or awareness raising, could be conducted. This identified additional individuals with active disease or latent infection, ensuring early diagnosis of these cases and preventing further onward transmission. However, because of the low discriminatory power of MIRU-VNTR, clustered patients could also reflect common endemic strains circulating in the UK or abroad, necessitating additional epidemiologic information to assess the likelihood of recent transmission. In the UK, where a large proportion of TB cases are probably due to infection acquired abroad, identifying the characteristics of persons who may have transmitted TB to others within the UK, as well as those recently infected in the UK, could identify settings for public health action and enable better targeted policies to interrupt transmission. Additionally, genotyping data are informative for assessing the epidemiology of transmission at a population level. Our objectives in this study were 1) to assess trends in TB clustering in the UK, using 6 years of available MIRU-VNTR data to identify the proportion of patients in a cluster as a proxy measure for transmission, and 2) to identify populations that may have been involved in recent UK transmission, based on the characteristics of clusters and the individuals in clusters. Identification of such populations will facilitate better targeted cluster investigation and public health action. METHODS Study population From 2010 onward, at least 1 isolate from all culture-confirmed Mycobacterium tuberculosis complex cases was prospectively typed using 24-locus MIRU-VNTR. We included all patients with notifications in the UK between January 1, 2010, and December 31, 2015, for whom at least 23 loci were typed. Data collection Notifications of TB cases made to the Enhanced TB Surveillance System in England, Wales, and Northern Ireland and the Enhanced Surveillance of Mycobacterial Infections (ESMI) system in Scotland between 2010 and 2015 were matched (11) to culture-positive results received from various branches of the National Mycobacterium Reference Laboratory. Strain typing and drug susceptibility results were available for culture-positive isolates. The following information was collected with notification of each patient’s case: demographic characteristics (age, sex, place of residence, ethnicity, country of birth), clinical characteristics (site of disease, previous diagnosis), and social risk factors (current or history of imprisonment, drug use, alcohol misuse, or homelessness). Definitions A molecular cluster (hereafter referred to as a cluster) was defined as a group of TB cases containing at least 2 patients with isolates with indistinguishable MIRU-VNTR profiles (at least 1 with complete 24-locus MIRU-VNTR and others with at least 23 loci typed) (10). A clustered patient was defined as a TB patient in a cluster, and a nonclustered patient was defined as a TB patient with a strain type distinguishable from that of all other patients. Two time periods were used to define patients as being in a cluster: the entire period of 2010–2015, to identify annual proportions of patients who were clustered with any other patient(s) during that time period, and 2-year intervals, where at least 2 cases from a cluster occurred within those 2 years. For the annual analysis, clusters were defined as new in the year in which the second case occurred. In the 2-year interval analysis, clusters were defined as new when both the first case and the second case occurred within the 2-year period. Clusters were defined as accruing when 2 or more cases (regardless of order; first, second, or subsequent) occurred within the period of interest (annual or 2-year). Within this paper, where year is presented, this is the year of patient notification. The sizes of clusters were defined as small (<5 patients), medium (5–20 patients), or large (>20 patients). Lineage was derived from MIRU-VNTR strain type (12) as Euro-American, Central Asian strain (CAS), East African–Indian (EAI), Beijing, Mycobacterium africanum, or Mycobacterium bovis. Large lineage diversity was defined as <50% of all patients having the same lineage, moderate lineage diversity as ≥50% but <80% of all patients having the same lineage, and little lineage diversity as ≥80% of patients having the same lineage. Data analysis We calculated the annual proportions of patients in a cluster with other patient(s) between 2010 and 2015 to assess the trend over time. To account for the fact that patients whose TB occurred in earlier years had a longer time to form a cluster with future patients, 2-year intervals were used. The maximum proportion of patients involved in transmission was estimated using the “n – 1” method: (number of clustered patients minus number of clusters)/number of patients with a strain type of 23 loci (13). The significance of trends was assessed using the χ2 test for trend. The characteristics of clustered patients were compared with those of nonclustered patients. Univariable logistic regression was performed to identify characteristics associated with being in a cluster, using odds ratios. Any characteristics associated with the outcome at P ≤ 0.2 were included in multivariable analysis. The significance of the main effects was assessed using likelihood ratio testing, with P ≤ 0.05 considered significant. Interactions between biologically and statistically plausible variables in the model were tested using likelihood ratios. Medium and large TB clusters are of greater public health significance than small ones, as more than 1 transmission event will probably have occurred. We described the characteristics of medium and large clusters using summary statistics and proportions, for all clusters in total and separately for clusters containing only UK-born patients (which are most likely to represent UK transmission), non-UK-born patients, and all other patients (both UK- and non-UK-born patients). We characterized the proportions of all clustered and nonclustered patients from each lineage by ethnic group for UK-born patients and by the 8 most frequent non–European Union countries of birth and the 2 most frequent European Union countries of birth for non-UK-born patients. RESULTS Between 2010 and 2015, 82.3% (23,646/28,741) of culture-confirmed TB patients had an isolate with a MIRU-VNTR profile typed to at least 23 loci (Figure 1). Overall, 58.5% (13,844/23,646) of these patients were concentrated in 2,701 clusters. There were 12,380 different strains identified. The proportion of patients in a cluster varied by geographical region of the UK (see Web Table 1, available at https://academic.oup.com/aje) and were closely aligned with regional TB incidence; the lowest proportion was identified in Wales (24.6%) and the highest in London, England (51.5%). Over half (55.1%; 1,487/2,701) of clusters were found only within 1 UK region; a much smaller proportion (12.5%; n = 337) of clusters crossed regional boundaries; and the remaining clusters (32.5%; n = 877) were only national clusters with no regional focus (i.e., there were not 2 or more patients within a single region). Figure 1. View largeDownload slide Numbers of tuberculosis (TB) patients, culture-confirmed patients, strain-typed patients, and TB clusters, United Kingdom (UK), 2010–2015. Figure 1. View largeDownload slide Numbers of tuberculosis (TB) patients, culture-confirmed patients, strain-typed patients, and TB clusters, United Kingdom (UK), 2010–2015. Trends in proportion of patients in a cluster The annual proportions of patients in clusters remained stable between 2010 and 2013 (range, 58.7%–60.7%) and decreased in 2014 and 2015 (56.9% and 55.3%, respectively) (χ2 test: P-trend = 0.001). The number of new clusters formed annually ranged from 410 in 2010 to 586 in 2012, with the lowest being 285 in 2015 (Table 1). Overall, the maximum recent transmission estimate (n − 1) over the 6-year study period was 47%. The annual estimate increased between 2010 and 2013 (range, 27.4%–31.2%) before decreasing to its lowest level (24.9%) in 2014. Table 1. Annual and 2-Year Proportions of Clustering of Tuberculosis Cases and Number of Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Abbreviations: TB, tuberculosis; UK, United Kingdom. a For single years, this is the number of clusters (new or existing) that cases occurring in that year are in, regardless of whether there are any other patients in the same cluster that year. For 2-year intervals, this is the number of clusters in which 2 or more cases occurred within the 2-year period. b For single years, this is the number of clusters in which the second case forming the cluster occurred in that year. For 2-year intervals, this is the number of clusters in which the first and second cases in the cluster occurred within that 2-year period, forming the cluster. Table 1. Annual and 2-Year Proportions of Clustering of Tuberculosis Cases and Number of Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Year or Period All TB Patients/Clusters UK-Born TB Patients/Clusters Only Non-UK-Born TB Patients/Clusters Only Maximum Transmission Estimate (n − 1), % Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb Clustered Patients No. of Accruing Clustersa No. of Newly Identified Clustersb No. % No. % No. % Single years  2010 27.4 2,051 58.7 1,095 410 569 68.6 25 25 1,373 55.3 122 122  2011 29.6 2,691 58.7 1,334 566 804 71.5 27 15 1,804 54.7 160 95  2012 31.2 2,849 60.7 1,384 586 857 73.4 28 15 1,908 56.6 187 89  2013 28.6 2,351 60.0 1,228 424 720 70.7 19 8 1,586 56.1 115 42  2014 24.9 2,067 56.9 1,164 430 654 67.1 15 8 1,373 53.0 113 41  2015 26.4 1,835 55.3 958 285 598 65.8 16 7 1,197 51.5 80 24 2-year intervals  2010–2011 36.0 3,883 48.1 976 976 1,193 61.1 68 68 2,527 43.7 369 369  2011–2012 38.2 4,689 50.5 1,146 660 1,462 63.8 77 53 3,086 46.3 459 326  2012–2013 37.5 4,298 49.9 1,068 398 1,371 62.7 70 38 2,817 45.5 406 207  2013–2014 34.2 3,469 45.9 887 262 1,132 56.8 52 24 2,267 41.8 331 151  2014–2015 32.5 3,038 43.7 777 179 1,038 55.1 46 21 1,934 39.3 278 102 Abbreviations: TB, tuberculosis; UK, United Kingdom. a For single years, this is the number of clusters (new or existing) that cases occurring in that year are in, regardless of whether there are any other patients in the same cluster that year. For 2-year intervals, this is the number of clusters in which 2 or more cases occurred within the 2-year period. b For single years, this is the number of clusters in which the second case forming the cluster occurred in that year. For 2-year intervals, this is the number of clusters in which the first and second cases in the cluster occurred within that 2-year period, forming the cluster. Analysis of 2-year periods showed that the proportion of patients in a cluster was highest in 2011–2012 (50.5%) and lowest in 2014–2015 (43.7%) (χ2 test: P-trend < 0.001). The number of accruing clusters formed within the 2-year periods followed the same pattern (1,146 clusters in 2011–2012 vs. 777 clusters in 2014–2015), and the numbers of new clusters formed decreased year-on-year between 2010–2011 (976 clusters) and 2014–2015 (179 clusters). The proportion of clustering was consistently higher in UK-born patients than in non-UK-born patients, but the trend over time was similar for all patients. No change in the size of clusters was observed over time (small clusters of <5 patients: 80.7% in 2010–2011 vs. 81.2% in 2014–2015 (χ2 test: P-trend = 0.162)). Additionally, there was no change in the age profile (age 15–44 years: 67.2% in 2010–2011 vs. 62.4% in 2014–2015 (χ2 test: P-trend = 0.162)) or place of birth (UK-born: 32.1% in 2010–2011 vs. 34.9% in 2014–2015 (χ2 test: P-trend = 0.162)) for clustered patients. There was a decrease in the proportion of non-UK-born clustered patients who had recently (<2 years previously) entered the UK (24.9% in 2010–2011 vs. 15.8% in 2014–2015 (χ2 test: P-trend = 0.072)), which mirrored the overall change in time between UK entry and TB for all patients observed from 2010 to 2015 (data not shown). Characteristics of patients in clusters The proportion of patients in a cluster varied by place of birth: 69.8% (4,202/6,022) of UK-born patients were in a cluster, as compared with 54.7% (9,241/16,894) of non-UK-born patients. Additionally, the proportion of patients in a cluster varied by age: 74.6% (308/413) of those aged ≤14 years, as compared with 47.9% (1,571/3,281) of those aged ≥65 years. Within each age group, a higher proportion of UK-born patients were in a cluster—most notably, 82.0% (191/233) of UK-born patients aged ≤14 years as compared with 63.6% (103/162) of non-UK-born patients aged ≤14 years. Characteristics associated with being in a cluster Characteristics associated with being in a cluster were identified in a comparison of 13,844 clustered patients with 9,802 nonclustered patients (Table 2). In univariable analysis, clustered patients were more likely to be under age 20 years, to be male, to be UK-born (or, if non-UK-born, to have lived in the UK for 6 years or more), to have a previous TB diagnosis, to have a social risk factor, and to have a strain with CAS or Beijing lineage. Table 2. Characteristics Associated With Being in a Tuberculosis Cluster (Univariable and Multivariable Logistic Regression), United Kingdom, 2010–2015a Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Abbreviations: CI, confidence interval; OR, odds ratio; SS+, sputum-smear–positive; UK, United Kingdom. a Totals for some characteristics do not equal the total number of patients because of missing information. b Years since entry into the UK was not included in the multivariable analysis, since it was only relevant to non-UK-born patients. c Isoniazid-resistant TB was not included in multivariable analysis because it was found not to be associated with the outcome in univariable analysis (see Methods section). d Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. Table 2. Characteristics Associated With Being in a Tuberculosis Cluster (Univariable and Multivariable Logistic Regression), United Kingdom, 2010–2015a Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Characteristic Clustered Patients (n = 13,844) Nonclustered Patients (n = 9,802) Univariable Regression Multivariable Regression No. % No. % OR 95% CI P Value OR 95% CI P Value Sex 13,833 9,790 <0.001 0.001  Female 5,347 38.7 4,036 41.2 1.0 Referent 1.0 Referent  Male 8,486 61.3 5,754 58.8 1.1 1.0, 1.2 1.1 1.0, 1.2 Age group, years 13,844 9,802 <0.001 <0.001  ≤4 78 0.6 16 0.2 3.3 1.9, 5.7 1.6 0.9, 3.0  5–9 46 0.3 13 0.1 2.4 1.3, 4.5 2.0 0.9, 4.5  10–14 184 1.3 76 0.8 1.7 1.3, 2.2 1.0 0.8, 1.4  15–19 807 5.8 353 3.6 1.6 1.4, 1.8 1.2 1.0, 1.4  20–29 3,759 27.2 2,574 26.3 1.0 Referent 1.0 Referent  30–39 3,210 23.2 2,340 23.9 0.9 0.9, 1.0 1.0 0.9, 1.1  40–49 2,207 15.9 1,423 14.5 1.1 1.0, 1.2 1.0 0.9, 1.1  50–59 1,489 10.8 921 9.4 1.1 1.0, 1.2 1.0 0.9, 1.2  60–69 914 6.6 728 7.4 0.9 0.8, 1.0 0.8 0.7, 1.0  ≥70 1,150 8.3 1,358 13.9 0.6 0.5, 0.6 0.5 0.5, 0.6 Place of birth and ethnic group 13,278 9,337 <0.001 <0.001  UK   White 2,646 20.0 1,350 14.4 2.0 1.8, 2.1 2.0 1.8, 2.3   Black Caribbean 255 1.9 42 0.5 6.0 4.3, 8.4 4.6 3.2, 6.7   Black African 232 1.8 53 0.6 4.4 3.2, 5.9 3.2 2.3, 4.5   Indian 362 2.7 131 1.4 2.8 2.2, 3.4 2.3 1.8, 2.8   Pakistani 381 2.9 123 1.3 3.1 2.5, 3.8 2.3 1.9, 3.0   Bangladeshi 56 0.4 43 0.5 1.3 0.9, 1.9 1.2 0.8, 1.9   Other 229 1.7 58 0.6 3.9 2.9, 5.3 3.0 2.2, 4.2  Non-UK   White 715 5.4 545 5.8 1.3 1.2, 1.5 1.2 1.0, 1.4   Black Caribbean 154 1.2 57 0.6 2.7 2.0, 3.7 3.0 2.1, 4.3   Black African 2,327 17.5 1,333 14.3 1.7 1.6, 1.9 1.9 1.5, 1.9   Indian 2,567 19.3 2,555 27.4 1.0 Referent 1.0 Referent   Pakistani 1,493 11.2 1,181 12.6 1.3 1.1, 1.4 1.1 1.0, 1.2   Bangladeshi 230 1.7 407 4.4 0.6 0.5, 0.7 0.7 0.6, 0.9   Other 1,631 12.3 1,459 15.6 1.1 1.0, 1.2 1.1 1.0, 1.2 Years since entry into the UK 8,343 6,876 <0.001  ≤1 1,616 19.4 1,491 21.7 1.0 Referent —b —  2–5 2,291 27.5 2,099 30.5 1.0 0.9, 1.1 — —  6–10 1,754 21.0 1,362 19.8 1.2 1.1, 1.3 — —  ≥11 2,682 32.1 1,924 28.0 1.3 1.2, 1.4 — — Site of disease 13,838 9,797 <0.001 <0.001  Pulmonary SS+ TB 4,143 29.9 2,234 22.8 1.7 1.6, 1.9 1.4 1.3, 1.5  Pulmonary non-SS+ TB 5,299 38.3 3,458 35.3 1.4 1.3, 1.5 1.3 1.2, 1.4  Extrapulmonary TB only 4,396 31.8 4,105 41.9 1.0 Referent 1.0 Referent Social risk factors  Drug misuse 12,151 8,484 <0.001 <0.001   Yes 722 5.9 169 2.0 3.1 2.6, 3.7 1.7 1.4, 2.1   No 11,429 94.1 8,315 98.0 1.0 Referent 1.0 Referent  Alcohol misuse 12,117 8,437 <0.001 0.440   Yes 775 6.4 317 3.8 1.8 1.5, 2.0 1.1 0.9, 1.3   No 11,342 93.6 8,120 96.2 1.0 Referent 1.0 Referent  Imprisonment 11,902 8,251 <0.001 0.005   Yes 635 5.3 183 2.2 2.5 2.1, 2.9 1.4 1.1, 1.7   No 11,267 94.7 8,068 97.8 1.0 Referent 1.0 Referent  Homelessness 12,299 8,531 <0.001 0.569   Yes 599 4.9 270 3.2 1.6 1.4, 1.8 0.9 0.8, 1.2   No 11,700 95.1 8,261 96.8 1.0 Referent 1.0 Referent Previous TB 13,151 9,237 <0.001 <0.001  Yes 796 6.1 460 5.0 1.2 1.1, 1.4 1.3 1.1, 1.5  No 12,355 94.0 8,777 95.0 1.0 Referent 1.0 Referent Place of residence 13,087 9,132 0.058 0.131  Urban area 12,671 96.8 8,799 96.4 1.2 1.0, 1.3 1.2 1.0, 1.4  Rural area 416 3.2 333 3.7 1.0 Referent 1.0 Referent Isoniazid-resistant TB 13,516 9,612 0.947  Yes 765 5.7 546 5.7 1.0 0.9, 1.1 —c —  No 12,751 94.3 9,066 94.3 1.0 Referent — — Multiple-drug-resistant TB 13,733 9,737 0.050 0.691  Yes 224 1.6 128 1.3 1.2 1.0, 1.5 1.1 0.8, 1.4  No 13,509 98.4 9,609 98.7 1.0 Referent 1.0 Referent Strain lineage 13,844 9,802 <0.001 <0.001  Euro-American 5,723 41.3 3,689 37.6 1.0 Referent 1.0 Referent  Central Asian strain 4,067 29.4 2,295 23.4 1.1 1.1, 1.2 1.5 1.4, 1.6  East African–Indian 1,328 9.6 1,951 19.9 0.4 0.4, 0.5 0.7 0.6, 0.7  Beijing 1,030 7.4 326 3.3 2.0 1.8, 2.3 2.3 2.0, 2.8  Otherd 1,696 12.3 1,541 15.7 0.7 0.7, 0.8 0.8 0.7, 0.9 Abbreviations: CI, confidence interval; OR, odds ratio; SS+, sputum-smear–positive; UK, United Kingdom. a Totals for some characteristics do not equal the total number of patients because of missing information. b Years since entry into the UK was not included in the multivariable analysis, since it was only relevant to non-UK-born patients. c Isoniazid-resistant TB was not included in multivariable analysis because it was found not to be associated with the outcome in univariable analysis (see Methods section). d Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. The multivariable model adjusted for age, sex, ethnic group, site of disease, social risk factors, previous TB diagnosis, area of residence, having multiple-drug-resistant TB, and strain lineage. Men were more likely to be in a cluster than women (adjusted odds ratio (aOR) = 1.1, 95% confidence interval (CI): 1.0, 1.2). Compared with those aged 20–29 years, patients aged 60–69 years (aOR = 0.8, 95% CI: 0.7, 1.0) and patients aged 70 years or more (aOR = 0.5, 95% CI: 0.5, 0.6) were less likely to be in a cluster. UK-born patients were more likely to be in a cluster, irrespective of ethnic group. Patients with pulmonary disease (smear-positive: aOR = 1.4 (95% CI: 1.3, 1.5); smear- negative/unknown: aOR = 1.3 (95% CI: 1.2, 1.4)) were more likely to be in a cluster than those with extrapulmonary disease only. Additionally, patients with a previous TB diagnosis (aOR = 1.3, 95% CI: 1.1, 1.5), a history of drug misuse (aOR = 1.7, 95% CI: 1.4, 2.1), or a history of imprisonment (aOR = 1.4, 95% CI: 1.1, 1.7) were more likely to be in a cluster. Compared with patients with a strain of Euro-American lineage, patients with a strain of Beijing (aOR = 2.3, 95% CI: 2.0, 2.8) or CAS (aOR = 1.5, 95% CI: 1.4, 1.6) lineage were more likely to be in a cluster, while patients with an EAI lineage were less likely to be (aOR = 0.7, 95% CI: 0.6, 0.7). We identified an interaction between some ethnic groups and strain lineages. UK-born Pakistani patients with an EAI lineage had lower odds of being in a cluster (odds ratio (OR) = 0.3, 95% CI: 0.1, 0.9), as did non-UK-born Bangladeshi patients with an EAI lineage (OR = 0.4, 95% CI: 0.2, 0.6). UK-born white patients with a CAS lineage had higher odds of being in a cluster (OR = 1.8, 95% CI: 1.2, 2.7), as did non-UK-born black African patients with a CAS lineage (OR = 1.4, 95% CI: 1.1, 1.9). Finally, non-UK-born white patients with a Beijing lineage had increased odds of being in a cluster (OR = 2.8, 95% CI: 1.2, 6.7). The interaction effect estimates corresponded to a further difference in odds, in addition to the main effects presented in Table 2 for ethnicity and lineage. Characteristics of clusters The 13,844 clustered patients were concentrated in 2,701 different clusters. The median number of patients in a cluster was 3 (range, 2–242). The majority of clusters (74.3%; 2,008/2,701) were small (<5 patients) in size; 22.6% (610 clusters) were medium (5–20 patients), and 3.1% (83 clusters) were large (>20 patients). Over half of clusters (51.3%; 1,386/2,701) contained only non-UK-born patients, and 8.3% (224 clusters) contained only UK-born patients (Figure 1, Table 3). For clusters comprised of only UK-born patients and only non-UK-born patients, the median number of patients in a cluster was 2 (range, 2–12 for UK-born patient clusters and 2–24 for non-UK-born patient clusters). Table 3. Characteristics of Tuberculosis Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Abbreviations: EPTB, extrapulmonary tuberculosis; PTB, pulmonary tuberculosis; SS+, sputum-smear–positive; UK, United Kingdom. a Clusters were defined as small (<5 patients), medium (5–20 patients), or large (>20 patients). b Values are presented as median (range). c Values are presented as median (interquartile range). d Current or history of imprisonment, drug use, alcohol misuse, or homelessness. e Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. Table 3. Characteristics of Tuberculosis Clusters by Patient’s Place of Birth, United Kingdom, 2010–2015 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Characteristic Cluster Type Total Only UK-Born Patients Only Non-UK-Born Patients Both UK- and Non-UK-Born Patients No. % No. % No. % No. % All clusters  No. of patients clustered 13,844 679 4.9 4,212 30.4 8,953 64.7  No. of clusters 2,701 224 8.3 1,386 51.3 1,091 40.4  Cluster sizea, no. of patientsb 3 (2–242) 2 (2–12) 2 (2–24) 4 (2–242) Medium and large clusters  No. of patients clustered 8,746 210 2.4 1,244 14.2 7,292 83.4  No. of clusters 693 31 4.5 174 25.1 488 70.4  Cluster size, no. of patientsb 7 (5–242) 6 (5–12) 6 (5–24) 8 (5–242)  ≥50% of cluster was male 535 77.2 26 83.9 121 69.5 388 79.5  Age of patients, yearsc 35 (30–43) 45 (39–57) 34 (29–38) 35 (30–44)  % of cluster members with a social risk factord   ≥25 136 19.6 12 38.7 13 7.5 111 22.7   ≥50 42 6.1 8 25.8 5 2.9 29 5.9   ≥75 4 0.6 2 6.5 0 0.0 2 0.4  Site of disease   PTB (with/without EPTB) 82 11.8 21 67.7 11 6.3 50 10.2   EPTB 4 0.6 0 0.0 4 2.3 0 0.0   ≥50% of patients with PTB were SS+ 322 46.5 16 51.6 78 44.8 228 46.7  Strain lineage   Euro-American 313 45.2 19 61.3 50 28.7 244 50.0   Central Asian strain 177 25.5 1 3.2 62 35.6 114 23.4   East African–Indian 66 9.5 0 0.0 37 21.3 29 5.9   Beijing 46 6.6 3 9.7 10 5.7 33 6.8   Othere 91 13.1 8 25.8 15 8.6 68 13.9 Abbreviations: EPTB, extrapulmonary tuberculosis; PTB, pulmonary tuberculosis; SS+, sputum-smear–positive; UK, United Kingdom. a Clusters were defined as small (<5 patients), medium (5–20 patients), or large (>20 patients). b Values are presented as median (range). c Values are presented as median (interquartile range). d Current or history of imprisonment, drug use, alcohol misuse, or homelessness. e Undetermined, Mycobacterium bovis, or Mycobacterium africanum lineage. In medium and large clusters, the median age in clusters containing only UK-born patients (210 clusters) was higher (45 years; interquartile range, 39–57) than the median age of patients in all clusters (35 years; interquartile range, 30–43) (Table 3). Sixty-eight percent (67.7%; 21 clusters) of such clusters (with only UK-born patients) contained only pulmonary patients, and in 25.8% (8 clusters) more than half of the cluster members had a social risk factor. Conversely, in 4 clusters containing only non-UK-born patients, all had extrapulmonary disease. Lineage diversity Where the lineage of the bacteria infecting patients could be determined (87.8% of patients), the most frequent lineages were Euro-American (45.3%; 9,412 patients) and CAS (30.6%; 6,362 patients) strains, with fewer EAI (15.8%; 3,279 patients) and Beijing (6.5%; 1,356 patients) strains and very small numbers of M. africanum (1.0%; 208 patients) or M. bovis (0.7%; 149 patients) strains. The lineage distribution differed between clustered and nonclustered patients (χ2 test: P < 0.001), most notably for EAI (10.8% vs. 23.1%) and Beijing (8.4% vs. 3.9%) strains. Euro-American lineage was the most frequent lineage in patients from sub-Saharan African countries (Nigeria, 81.7%; Zimbabwe, 83.0%) and Eastern European countries (Romania, 98.9%; Poland, 97.4%). The CAS lineage was the most frequent in patients from Pakistan (73.2%) and India (51.4%), while the EAI lineage was most frequent in patients from the Philippines (93.6%) and Bangladesh (62.8%). The majority of UK-born patients had cases with Euro-American lineage (67.8%); however, lineage varied by ethnic group: Pakistani and Indian patients had a high proportion of CAS strains (52.6% and 42.0%, respectively), similar to patients born in those countries (Pakistan, 65.1%; India, 45.8%). The proportion of patients with TB strains from each lineage differed between clustered and nonclustered patients (Web Figure 1), particularly among countries of birth that showed large (Somalia and Nepal) or moderate (India, Pakistan, Bangladesh, and the UK) lineage diversity. This was particularly evident in patients from Bangladesh with a Euro-American lineage (31.2% clustered vs. 13.2% nonclustered) and an EAI lineage (39.7% clustered vs. 75.3% nonclustered), patients from Somalia with a CAS lineage (36.1% clustered vs. 20.2% nonclustered), and patients from Nepal with an EAI lineage (3.5% clustered vs. 24.9% nonclustered). For all countries of birth, other than the Philippines and Nigeria, a higher proportion of clustered patients than nonclustered patients had Beijing and CAS lineages, while fewer clustered patients had an EAI lineage. DISCUSSION In this paper, we report results from a population-based study carried out in the UK using MIRU-VNTR to describe trends and features of possible TB transmission based on the proportion of patients in a cluster over time, as well as the characteristics associated with being in a cluster. We identified decreases in the proportions of patients in clusters and in numbers of new clusters between 2010 and 2015. Annual proportions of patients in a cluster (55.3%–60.7%) and the maximum transmission estimate (47%) identified were higher than findings from other low-incidence countries and previous UK estimates provided for shorter time periods or at a local level (2, 3, 14). The low discriminatory power of MIRU-VNTR is known to overestimate transmission (15), since a proportion of patients who are part of a cluster will have TB due to reactivation or importation of common endemic strains. Although measuring TB transmission at a population level is difficult, combining several pieces of evidence can help us understand trends in population-level transmission over time. The rate of TB in UK-born children, which is used as an indicator of recent transmission, decreased between 2010 and 2015 (5). Additionally, the number of TB notifications and the rate of TB decreased over the same period, with this decrease occurring among both the UK-born and non-UK-born populations (5). The reduction in the proportion of patients in a cluster, combined with these additional epidemiologic trends, is suggestive of a decrease in TB transmission in the UK during the study period. Improvements in TB control methods which promote case-finding and contact tracing may have affected transmission, contributing to the decrease in the proportion of patients identified as being in a cluster, or may have increased the number of sporadic (not clustered) cases identified. In addition to the timely screening of contacts and additional settings guided by strain typing (16, 17), interventions include “find and treat” (18) and cohort review, which includes presentation of and accountability for the number of contacts identified and screened (19). It is also possible that part of the decrease in the proportions of patients in clusters is due not to changes in transmission but to a reduction in the number of non-UK-born TB patients in the UK, particularly recent migrants (5), because of preentry screening and latent TB infection screening (20–22). With fewer non-UK-born patients, the reduction in clustering among these patients may have led to fewer introductions of common strains into the UK. Our study identified the following characteristics associated with being a clustered patient: being UK-born, being non-UK-born of black ethnicity, current or past drug misuse, and current or past imprisonment. Note that these are the characteristics of patients in clusters; therefore, they include persons who may have transmitted TB as well as those to whom TB may have been transmitted. People who misuse drugs may be less likely to seek health care or may be more likely to have a negative health-care experience, contributing to increased transmission within this population (23–25). Additionally, they may be less willing to name contacts due to the illegal nature of their activities or may have complex social networks, resulting in missed opportunities for contact tracing. Imprisonment is a risk factor associated with TB in both the UK, with sporadic transmission previously reported (26, 27), and globally (28). Imprisonment reported here included current or past imprisonment in the UK or abroad; therefore, the identification of an association between clustering and imprisonment resulted from historical circulation of strains in prisons in addition to recent transmission. UK-born black Caribbean and black African and non-UK-born black Caribbean patients were most likely to be in a cluster, while the Bangladeshi patients were more likely to have a unique strain. The reasons for these observed differences between ethnic groups are unknown but could include socioeconomic and/or behavioral differences, including engagement with health services. Nevertheless, our findings suggest that transmission dynamics within these ethnic groups, as well as among those with social risk factors, should be considered during the prioritization of cluster investigation and the targeting of resources for public health action. This may include the use of enhanced data collection to identify behaviors and settings that may be associated with increased transmission risk (16, 29). The patient and cluster characteristics we identified suggest that recent transmission is less likely to have occurred among non-UK-born patients, particularly those in clusters containing only non-UK-born individuals. Many of these clusters probably represent reactivation of common endemic strain types circulating abroad. Similar findings have been identified elsewhere (15). However, transmission does still occur among non-UK-born individuals in the UK; this maybe particularly true in urban areas (16) and in specific migrant populations (30). The lineage of M. tuberculosis complex strains isolated from patients differed by country of birth and largely corresponded to the strains that were predominant in their countries of origin (31, 32). This provides additional evidence that a large proportion of TB in the UK is imported, due to either arrival of patients with active disease or subsequent reactivation of latent infection. Among UK-born patients, lineage differed by ethnic group and often closely resembled that in the countries of origin of previous generations of migrants (i.e., parents or grandparents). This could be due to acquisition of infection after receiving visitors from, or traveling to, the family’s country of origin. Additionally, for both UK-born and non-UK-born populations, household or community interaction with persons from the country of origin may occur (16, 33, 34). CAS and Beijing lineage strains were more frequent among clustered patients, while the EAI lineage was more frequent in those with a unique strain. These findings suggest that lineage may play a role in transmission dynamics. However, in analysis of household transmission in the UK, where patients are molecularly and epidemiologically linked, associations between lineage and clustering were not identified (33). It is also possible that our results were due to differences in the ability of MIRU-VNTR typing to distinguish between strains of different lineages. Whole-genome sequencing analysis identifies single nucleotide polymorphism differences between isolates, a more discriminatory measure of relatedness than that yielded by MIRU-VNTR (35). In 2017, the UK replaced MIRU-VNTR with whole-genome sequencing to determine relatedness between TB isolates. This will bring many benefits, including refined estimates of the level of transmission and more accurate identification of the characteristics of persons transmitting TB. However, only data collected over several years enables investigation of trends and risk groups. Therefore, the findings from this study using MIRU-VNTR data, particularly trends in clustering over time, are helpful in evaluating recent TB control efforts and in targeting further TB control efforts, including enhanced contact tracing, which can later be refined on the basis of whole-genome sequencing results. Systematic collection of epidemiologic links between patients will also facilitate the identification of transmission settings in the UK (16, 33, 34) and could additionally refine the estimation of transmission rates within settings (33). Developing and applying more sophisticated methods for estimating transmission beyond those utilized here, thereby refining the estimates provided by our current analysis to account for reactivation, super-spreaders, and the plausibility of transmission, including temporal and spatial analysis, would be beneficial. Such methods have been used in the United States (4, 8, 36, 37) and locally in the UK (38), resulting in refined transmission estimates. Our study benefited from having a large national data set spanning a period of 6 years in which prospective routine strain typing was performed. It will take several years of continuous use of whole-genome sequencing before such trends in transmission can be measured again. In addition, high-quality data on demographic factors, clinical characteristics, and social risk factors were collected for the TB patients. However, our study also had several limitations, as we have acknowledged throughout this Discussion where relevant. Additionally, between 2010 and 2015, only 61% of TB cases in the UK were culture-confirmed, although this proportion was higher (72%) among pulmonary patients (5). Among culture-confirmed cases, a small proportion were not typed (2.4%) or typed to 23 loci (15.3%). In conclusion, our findings suggest that TB transmission in the UK decreased between 2010 and 2015, during which time TB incidence in the UK also decreased. Our results emphasize the need for TB control efforts for reducing transmission to be targeted toward UK-born populations, particularly persons belonging to black Caribbean and black African ethnic groups, and those with social risk factors. While we were not able to determine why transmission may be more frequent in these groups, additional research should be conducted to investigate this in order to further tailor public health interventions. ACKNOWLEDGMENTS Author affiliations: Tuberculosis Unit, National Infection Service, Public Health England, London, United Kingdom (Jennifer A. Davidson, H. Lucy Thomas, Colin N. J. Campbell, Maeve K. Lalor); Field Service, National Infection Service, Public Health England, London, United Kingdom (Helen Maguire, Neil Macdonald); Institute for Global Health, University College London, London, United Kingdom (Helen Maguire, Colin N. J. Campbell, Maeve K. Lalor); National Mycobacterium Reference Service South, National Infection Service, Public Health England, London, United Kingdom (Timothy Brown); and Field Service, National Infection Service, Public Health England, Newcastle, United Kingdom (Andy Burkitt). No funding was obtained for this work. We thank Ross Harris of the Statistics Unit, National Infection Service, Public Health England, for statistical review and guidance provided at multiple points during this analysis. Conflict of interest: none declared. Abbreviations aOR adjusted odds ratio CAS Central Asian strain CI confidence interval EAI East African–Indian MIRU-VNTR mycobacterial interspersed repetitive unit–variable number tandem repeats OR odds ratio TB tuberculosis UK United Kingdom REFERENCES 1 Al-Hajoj S , Varghese B , Al-Habobe F , et al. . Current trends of Mycobacterium tuberculosis molecular epidemiology in Saudi Arabia and associated demographical factors . Infect Genet Evol . 2013 ; 16 : 362 – 368 . Google Scholar Crossref Search ADS PubMed 2 Lambregts-van Weezenbeek CS , Sebek MM , van Gerven PJ , et al. . Tuberculosis contact investigation and DNA fingerprint surveillance in the Netherlands: 6 years’ experience with nation-wide cluster feedback and cluster monitoring . Int J Tuberc Lung Dis . 2003 ; 7 ( 12 suppl 3 ): S463 – S470 . Google Scholar PubMed 3 Kamper-Jørgensen Z , Andersen AB , Kok-Jensen A , et al. . Clustered tuberculosis in a low-burden country: nationwide genotyping through 15 years . J Clin Microbiol . 2012 ; 50 ( 8 ): 2660 – 2667 . Google Scholar Crossref Search ADS PubMed 4 Moonan PK , Ghosh S , Oeltmann JE , et al. . Using genotyping and geospatial scanning to estimate recent Mycobacterium tuberculosis transmission, United States . Emerg Infect Dis . 2012 ; 18 ( 3 ): 458 – 465 . Google Scholar Crossref Search ADS PubMed 5 Public Health England . Tuberculosis in England: 2016 Report (Presenting Data to End of 2015) . London, United Kingdom : Public Health England ; 2016 . https://www.gov.uk/government/publications/tuberculosis-in-england-annual-report. Accessed October 26, 2016. 6 Sandgren A , Schepisi MS , Sotgiu G , et al. . Tuberculosis transmission between foreign- and native-born populations in the EU/EEA: a systematic review . Eur Respir J . 2014 ; 43 ( 4 ): 1159 – 1171 . Google Scholar Crossref Search ADS PubMed 7 Yang C , Shen X , Peng Y , et al. . Transmission of Mycobacterium tuberculosis in China: a population-based molecular epidemiologic study . Clin Infect Dis . 2015 ; 61 ( 2 ): 219 – 227 . Google Scholar Crossref Search ADS PubMed 8 Yuen CM , Kammerer JS , Marks K , et al. . Recent transmission of tuberculosis—United States, 2011–2014 . PLoS One . 2016 ; 11 ( 4 ): e0153728 . Google Scholar Crossref Search ADS PubMed 9 United Kingdom Department of Health . Stopping Tuberculosis in England: An Action Plan from the Chief Medical Officer . London, United Kingdom : Department of Health ; 2004 . http://webarchive.nationalarchives.gov.uk/+/http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4090417. Accessed January 23, 2017. 10 Public Health England . TB Strain Typing and Cluster Investigation Handbook. Third Edition, February 2014 . London, United Kingdom : Public Health England ; 2014 . https://www.gov.uk/government/publications/tb-strain-typing-and-cluster-investigation-handbook. Accessed October 28, 2015. 11 Aldridge RW , Shaji K , Hayward AC , et al. . Accuracy of probabilistic linkage using the Enhanced Matching System for public health and epidemiological studies . PLoS One . 2015 ; 10 ( 8 ): e0136179 . Google Scholar Crossref Search ADS PubMed 12 Brown T , Evans J , Sails A . HPA Mycobacterium tuberculosis Strain Typing: A Guide to Data Production and Distribution. London, United Kingdom : Health Protection Agency ; 2012 . https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/433731/HPA_VNTR_280812_final.pdf. Accessed July 3, 2016. 13 Glynn JR , Vynnycky E , Fine PE . Influence of sampling on estimates of clustering and recent transmission of Mycobacterium tuberculosis derived from DNA fingerprinting techniques . Am J Epidemiol . 1999 ; 149 ( 4 ): 366 – 371 . Google Scholar Crossref Search ADS PubMed 14 Hamblion EL , Le Menach A , Anderson LF , et al. . Recent TB transmission, clustering and predictors of large clusters in London, 2010–2012: results from first 3 years of universal MIRU-VNTR strain typing . Thorax . 2016 ; 71 ( 8 ): 749 – 756 . Google Scholar Crossref Search ADS PubMed 15 Stucki D , Ballif M , Egger M , et al. . Standard genotyping overestimates transmission of Mycobacterium tuberculosis among immigrants in a low-incidence country . J Clin Microbiol . 2016 ; 54 ( 7 ): 1862 – 1870 . Google Scholar Crossref Search ADS PubMed 16 Black AT , Hamblion EL , Buttivant H , et al. . Tracking and responding to an outbreak of tuberculosis using MIRU-VNTR genotyping and whole genome sequencing as epidemiological tools [published online ahead of print July 5, 2017]. J Public Health (Oxf) . ( doi:10.1093/pubmed/fdx075 ). 17 Smith CM , Trienekens SC , Anderson C , et al. . Twenty years and counting: epidemiology of an outbreak of isoniazid-resistant tuberculosis in England and Wales, 1995 to 2014 . Euro Surveill . 2017 ; 22 ( 8 ): 30467 . Google Scholar Crossref Search ADS PubMed 18 Jit M , Stagg HR , Aldridge RW , et al. . Dedicated outreach service for hard to reach patients with tuberculosis in London: observational study and economic evaluation . BMJ . 2011 ; 343 : d5376 . Google Scholar Crossref Search ADS PubMed 19 Anderson C , White J , Abubakar I , et al. . Raising standards in UK TB control: introducing cohort review . Thorax . 2014 ; 69 ( 2 ): 187 – 189 . Google Scholar Crossref Search ADS PubMed 20 Panchal RK , Browne I , Monk P , et al. . The effectiveness of primary care based risk stratification for targeted latent tuberculosis infection screening in recent immigrants to the UK: a retrospective cohort study . Thorax . 2014 ; 69 ( 4 ): 354 – 362 . Google Scholar Crossref Search ADS PubMed 21 Usdin M , Dedicoat M , Gajraj R , et al. . Latent tuberculous screening of recent migrants attending language classes: a cohort study and cost analysis . Int J Tuberc Lung Dis . 2017 ; 21 ( 2 ): 175 – 180 . Google Scholar Crossref Search ADS PubMed 22 Zenner D , Loutet MG , Harris R , et al. . Evaluating 17 years of latent tuberculosis infection screening in north-west England: a retrospective cohort study of reactivation . Eur Respir J . 2017 ; 50 ( 1 ): 1602505 . Google Scholar Crossref Search ADS PubMed 23 Lloyd C . The stigmatization of problem drug users: a narrative literature review . Drugs Educ Prev Policy . 2013 ; 20 ( 2 ): 85 – 95 . Google Scholar Crossref Search ADS 24 Neale J , Tompkins C , Sheard L . Barriers to accessing generic health and social care services: a qualitative study of injecting drug users . Health Soc Care Community . 2008 ; 16 ( 2 ): 147 – 154 . Google Scholar Crossref Search ADS PubMed 25 Van Boekel LC , Brouwers EP , van Weeghel J , et al. . Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: systematic review . Drug Alcohol Depend . 2013 ; 131 ( 1-2 ): 23 – 35 . Google Scholar Crossref Search ADS PubMed 26 Story A , Murad S , Roberts W , et al. . Tuberculosis in London: the importance of homelessness, problem drug use and prison . Thorax . 2007 ; 62 ( 8 ): 667 – 671 . Google Scholar Crossref Search ADS PubMed 27 Anderson C , Story A , Brown T , et al. . Tuberculosis in UK prisoners: a challenge for control . J Epidemiol Community Health . 2010 ; 64 ( 4 ): 373 – 376 . Google Scholar Crossref Search ADS PubMed 28 Baussano I , Williams BG , Nunn P , et al. . Tuberculosis incidence in prisons: a systematic review . PLoS Med . 2010 ; 7 ( 12 ): e1000381 . Google Scholar Crossref Search ADS PubMed 29 Gardy JL , Johnston JC , Ho Sui SJ , et al. . Whole-genome sequencing and social-network analysis of a tuberculosis outbreak . N Engl J Med . 2011 ; 364 ( 8 ): 730 – 739 . Google Scholar Crossref Search ADS PubMed 30 Munang ML , Browne C , Khanom S , et al. . Tuberculosis microepidemics among dispersed migrants, Birmingham, UK, 2004–2013 . Emerg Infect Dis . 2015 ; 21 ( 3 ): 524 – 527 . Google Scholar Crossref Search ADS PubMed 31 Gagneux S , Small PM . Global phylogeography of Mycobacterium tuberculosis and implications for tuberculosis product development . Lancet Infect Dis . 2007 ; 7 ( 5 ): 328 – 337 . Google Scholar Crossref Search ADS PubMed 32 Reed MB , Pichler VK , McIntosh F , et al. . Major Mycobacterium tuberculosis lineages associate with patient country of origin . J Clin Microbiol . 2009 ; 47 ( 4 ): 1119 – 1128 . Google Scholar Crossref Search ADS PubMed 33 Lalor MK , Anderson LF , Hamblion EL , et al. . Recent household transmission of tuberculosis in England, 2010–2012: retrospective national cohort study combining epidemiological and molecular strain typing data . BMC Med . 2017 ; 15 : 105 . Google Scholar Crossref Search ADS PubMed 34 Munang ML , Browne C , Evans JT , et al. . Programmatic utility of tuberculosis cluster investigation using a social network approach in Birmingham, United Kingdom . Int J Tuberc Lung Dis . 2016 ; 20 ( 10 ): 1300 – 1305 . Google Scholar Crossref Search ADS PubMed 35 Borgdorff MW , van Soolingen D . The re-emergence of tuberculosis: what have we learnt from molecular epidemiology? Clin Microbiol Infect . 2013 ; 19 ( 10 ): 889 – 901 . Google Scholar Crossref Search ADS PubMed 36 Althomsons SP , Kammerer JS , Shang N , et al. . Using routinely reported tuberculosis genotyping and surveillance data to predict tuberculosis outbreaks . PLoS One . 2012 ; 7 ( 11 ): e48754 . Google Scholar Crossref Search ADS PubMed 37 Althomsons SP , Hill AN , Harrist AV , et al. . Statistical method to detect tuberculosis outbreaks among endemic clusters in a low-incidence setting . Emerg Infect Dis . 2018 ; 24 ( 3 ): 573 – 575 . Google Scholar Crossref Search ADS PubMed 38 Saavedra-Campos M , Welfare W , Cleary P , et al. . Identifying areas and risk groups with localised Mycobacterium tuberculosis transmission in northern England from 2010 to 2012: spatiotemporal analysis incorporating highly discriminatory genotyping data . Thorax . 2016 ; 71 ( 8 ): 742 – 748 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

American Journal of EpidemiologyOxford University Press

Published: Oct 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off