- A total of 310 cases of TB were reported to the Enhanced Surveillance of Mycobacterial Infections (ESMI) scheme in 2016, representing an annual incidence of 5.7 cases per 100,000 population. This was a decrease of 3% in the number of cases and 5% in the annual incidence on 2015 (320 cases, 6.0 per 100,000 population).
- This represents a sixth consecutive decrease in numbers and incidence of TB cases reported since 2010 and is the lowest number of cases and incidence reported since enhanced surveillance began in 2000.
- The number and proportion of cases known to have been born outside the UK increased by 4% in 2016 (from 57.3%) and accounted for almost two thirds of all TB cases notified in Scotland (61.2%), the highest reported since surveillance began.
- The incidence of TB among those born outside the UK increased while the incidence among those born in the UK decreased further in 2016. As a consequence, the incidence of TB was approximately 20 times higher among those born outside the UK than those born in the UK (45.3 cases per 100,000 populations compared with 2.3 cases per 100,000 population respectively).
- As seen in previous previous years, most cases of TB were observed in those aged 25-34 years (86 cases, 27.7%, 11.8 cases per 100,000 population) and fewest cases in those aged 5-14 years (five cases, 1.6%, 0.9 cases per 100,000 population).
- In 2016, resistance to at least one first-line drug at the start of treatment was reported for 12 cases (5.8%), eight (3.9%) isolates being resistant to isoniazid, four (1.9%) were resistant to pyrazinamide, one (0.5%) was resistant to rifampicin and one (0.5%) was resistant to ethambutol. One case (0.5%) was resistant to both isoniazid and rifampicin and therefore defined as multidrug-resistant tuberculosis (MDR-TB).
- Successful treatment outcomes of TB cases first notified in 2015 were down slightly when compared with 2014 (79% compared with 80%) and failed to meet the ECDC target of 85%. Twenty-one tuberculosis cases (6.8% case fatality ratio) were known to have died, which is similar to 2014 (21 cases; 6.2%).
National surveillance of tuberculosis provides information on the numbers, distribution and characteristics of cases, drug resistance patterns and treatment outcomes. It supports the early identification and treatment of cases and enables the identification of high-risk populations. Tuberculosis surveillance was introduced in Scotland in 2000 through the Enhanced Surveillance of Mycobacterial Infections (ESMI) scheme. This report describes tuberculosis in Scotland for the year 2016 including treatment outcomes of cases first reported in 2015 (12-month outcomes) and 2014 (24-month outcomes).
Overall numbers, rates and geographical distribution
In 2016, the ESMI scheme received 310 provisional notifications of tuberculosis, an incidence of 5.7 (95% CI 5.1-6.4) cases per 100,000 population. This was a decrease of 3.1% in the number of cases and a decrease of 5.0% in the incidence, compared with 2015i (Table 1). This represented the sixth consecutive decrease in annual numbers of tuberculosis cases and incidence since 2010 and the lowest reported case numbers and incidence since enhanced surveillance began (Figure 1). This is similar to the 1% reduction in numbers of cases and 2% reduction in incidence reported for the whole of the UK in 2016.1
Historically, the largest numbers of cases have been notified from Greater Glasgow and Clyde, and Lothian NHS Boards (Table 2). In 2016, the majority of cases were from Greater Glasgow and Clyde (118 cases; 38.1%; 10.2 per 100,000 population), Lothian (68 cases; 21.9%; 7.7 per 100,000 population) and Grampian (29 cases; 9.4%; 4.9 per 100,000 population) NHS Boards (Figure 2).
Compared with 2015, there has been a decrease in the number of cases of tuberculosis reported in NHS Greater Glasgow and Clyde, NHS Tayside and NHS Western Isles, while case numbers in NHS Borders, NHS Dumfries and Galloway, NHS Fife, NHS Highland and NHS Lothian increased (Table 2).
i The number of cases reported for 2015 has increased from the 315 cases initially reported in last year’s report, to 320 cases following reporting delays.
Gender and age
In 2016, more than half of tuberculosis cases occurred in males (179 cases, 57.7%, 6.8 cases per 100,000 population) (Table 3). Most tuberculosis cases occurred in those aged 25-34 years (86 cases; 27.7%; 11.8 cases per 100,000) and fewest in those aged 5-14 years (five cases, 1.6%, 0.9 cases per 100,000 population) as shown in Figure 3. Males were more common in every age group except those aged under five.
The rate of tuberculosis among children aged under five years (which is an accepted indicator of recent transmission) increased from no reported cases in 2015 to 3.1 cases per 100,000 population in 2016. In addition, the child-to-adult notification rateii (which is also an accepted indicator of ongoing transmission) was 0.25. This represents an increase in the ratio when compared with 2015 (0.13), suggesting that transmission of tuberculosis may be ongoing in Scotland.
ii The child-to-adult notification rate is the ratio between the case notification rate in children under 15 and that in adults. A declining trend in the ratio suggests a decrease in ongoing transmission.
Place of birth
Place of birth was known for 93.9% (291/310) of cases in 2016. Of these, 61.2% (178/291) were born outside the UK. This is an increase in the percentage of non-UK born cases when compared with 2015 (57.3%) and the highest percentage recorded since surveillance began (Figure 4).
In 2016, the rate of tuberculosis among those born outside the UK increased slightly to 45.3 cases per 100,000 population from 45.0 cases per 100,000 per population in 2015 (Figure 5)iii. The rate of tuberculosis among those born in the UK decreased to 2.3 cases per 100,000 population in 2016 from 2.7 cases per 100,000 population in 2015. The rate of tuberculosis among non-UK born individuals was approximately 20 times higher than the rate in the UK born population (45.3 versus 2.3 cases per 100,000 population). This is the highest rate difference reported since these population data became available.
The mean age of presentation in 2016 was 44 years and the median was 38 years. However, the age at presentation of tuberculosis cases varies by place of birth as shown in Figure 6. The mean age of those born in the UK (58 years) was significantly higher (p<0.01) than the mean age of those born outwith the UK (38 years).
Place of birth was known for all cases born outside the UK. As in previous years, India (47 cases; 26.4%) and Pakistan (46 cases; 25.8%) were the most commonly recorded countries of birth, followed by Romania (10 cases; 5.6%) and Poland (eight cases; 4.5%) in 2016 (Table 4).
Information on the time from their entry into the UK to tuberculosis diagnosis was available for 87.1% (155/178) of non-UK born cases. Of these, 83.2% (129/155) had entered the UK two or more years before diagnosis; 59.4% had entered five or more years earlier and 36.1% had entered ten or more years before diagnosis (Figure 7). The mean time between entry and diagnosis was ten years, the median time being six years.
iii These rates are displayed from 2004 onwards due to the lack of available denominator population data in Scotland from 2000-2003. NRS recently reviewed historic population data on country of birth and as a consequence these data have been reviewed meaning they may differ from previously published data.2
Ethnicity was recorded for 299 cases (96.5%). The majority were white-Caucasian (116; 38.8%), Indian (53; 17.7%), Pakistani (52; 17.4%), and Other/Mixed (33; 11.0%) (Table 5). The majority of tuberculosis cases reported from all ethnic origins except white-Caucasian were born outside the UK (range 89.5-100%) as shown in Table 5 and Figure 8.
Site of disease
Of the 310 cases reported in 2016, 180 (58.1%) were classified as pulmonary tuberculosis (26 with and 154 without extra-pulmonary disease) and 130 (41.9%) as non-pulmonary tuberculosis (Table 6). Historically, pulmonary disease is more common than non-pulmonary disease in Scotland. However, non-pulmonary disease was more likely in those born outside the UK (OR=4.0, 95% CI 2.4-6.8, p<0.01), those of non-white Caucasian ethnicity (OR=4.2, 95% CI 2.5-7.1, p<0.01) and those aged 25-44 years (OR=1.8, 95% CI 1.2-2.9, p<0.01).
The majority of non-pulmonary cases were in extra-thoracic lymph nodes (62; 47.7%), pleural sites (17; 13.1%) and intra-thoracic lymph nodes (15; 11.5%) (Table 8). Cervical (40 cases), axillary (six cases) and abdominal (three cases) lymph nodes were the most commonly reported extra-thoracic sites.
Information on patients’ previous diagnoses was available for 96.8% (300/310) of cases. Of these, 20 (6.7%) had a previous diagnosis of tuberculosis more than one year before their current date of diagnosis. The mean length of time between their two tuberculosis diagnoses was 12 years and the median time was five years (range 2-63 years).
Whether or not a case was a hospital inpatient was known for all cases in 2016, 152 cases (49.0%) having been admitted to hospital for treatment.
Information on BCG vaccination was available for around two thirds (204/310; 65.8%) of cases in 2016, of whom 150 (73.5%) had previously received BCG vaccination. Of the cases aged under 15, 27.3% (3/11) were known to have received BCG vaccination. However, currently no information is collected around the eligibility of children for BCG vaccination (i.e. the country of birth of their parents and grandparents).
Duration of symptoms
Of the 310 reports of tuberculosis received in 2016, 267 (86.1%) were known to have been symptomatic at the time of notification. The duration of symptoms was known for 253 (94.8%) of these cases. The mean length of time between symptom onset and diagnosis was 21 weeks and the median length of time was 13 weeks (range 1-522 weeks). As in previous years, there was a significant difference in duration of symptoms between those presenting with pulmonary and non-pulmonary disease (16 weeks and 28 weeks, respectively, p<0.05). Approximately 15% (39/267) of tuberculosis cases had been symptomatic for longer than six months at notification (Figure 9) which was slightly higher than in 2015 (30/253; 11.9%).
Identification of cases of tuberculosis
The method of identification of tuberculosis was known for all cases reported in 2016. Of these, 267 cases (86.1%) presented with an illness subsequently diagnosed as tuberculosis, 25 cases (8.1%) were identified through contact tracing, 12 cases (3.8%) were incidental findings, four cases (1.3%) were identified post mortem, one case (0.3%) was diagnosed in another country, and one case (0.3%) was identified through new entrant screening.
At the time of notification, treatment had been commenced for almost all cases (298/310; 96.1%). Treatment regimens were known for all cases that had started treatment, the majority of whom (272; 91.3%) were recorded as being given at least the standard four-dose first-line drug regimen of isoniazid, rifampicin, pyrazinamide and ethambutol. This is similar to that reported in previous years and maintains the sustained increase in the proportion of cases on at least the standard four-dose first-line therapy since 2004 when 45% of cases commenced on the standard four-drug regimen.
Information on risk factors were known for the majority of cases in 2016 (302/310; 97.4%). Risk factors were identified for 86 cases (28.5%), of whom nine cases had more than one known risk factor. Risk factors identified include immunosuppression (25 cases; 8.3%), alcohol misuse (22 cases; 7.3%), being a refugee (21 cases; 7.0%), working in healthcareiv (18 cases; 6.0%), homelessness (four cases; 1.3%), residency in a residential or corrective institution (four cases; 1.3%) and drug misuse (two cases; 0.7%) (Figure 10).
Risk factors were identified more often in TB cases born in the UK (35.3% in UK born versus 25.4% in non-UK born). Alcohol misuse, immunosuppression and residency in a corrective or residential institution were more commonly identified among UK born TB cases, while working in healthcare and having refugee status were more commonly identified among non-UK born TB cases.
iv One third of TB cases who work in a healthcare setting were born in countries with a high incidence of tuberculosis or had travelled to an area with a high incidence of tuberculosis for two months or more in the two years preceding their TB diagnosis.
Information on Scottish Index of Multiple Deprivation (SIMD) quintile was available for 92.6% (287/310) of TB cases reported in 2016, 23 cases (7.4%) having either missing or partial postcode information recorded. Almost one third of TB cases (90/287; 31.4%) were in the most deprived category (SIMD1) compared with 12% (35/287) in the least deprived category (SIMD5) which is similar to previous years where 35.5% were in the most deprived category and 12.1% were in the least deprived category (Figure 11).
Directly observed therapy
Information on whether a patient was due to have their treatment administered using directly observed therapy (DOT) was known for almost all cases (307/310; 99.0%) in 2016. Of these, 38 patients (12.4%) were administered treatment using DOT, which is higher than the 24 cases (7.5%) enrolled on DOT in 2015. Patients were observed either three (27 cases), seven (10 cases) or six (one case) times per week. Patients enrolled on DOT were more likely to have been born in the UK (22/113; 19.5% versus 16/178; 9.0%, p<0.05) and more likely to have one or more identified risk factor for TB (23/86; 26.7% versus 15/224; 6.7%, p<0.05) with alcohol and immunosuppression being the most common risk factors for DOT.
Suspected source of infection
Information on source of infection was recorded for 297 cases (95.8%) reported in 2016: 223 cases (75.1%) were classified as having no clear source of infection, 35 (11.8%) were recorded as other sources, 31 (10.4%) as household, four (1.3%) as workplace, two (0.7%) as healthcare premises (one was a retired HCW and one was previously a HCW in a high incidence country for TB), one (0.3%) as hostel/homeless, and one prison (0.3%). The sources recorded as other source of infection were infection in country of origin (23; 65.7%), contact of a known tuberculosis case (four; 11.4%), reactivation of latent tuberculosis infection (four; 11.4%), travel to a high-risk country (two; 5.7%) and volunteering with high-risk groups (two; 5.7%).
In 2016, 1077 case contacts were known to have been traced in Scotland with a mean of three contacts for each TB case (range 0-97, median two contacts). The majority of contacts identified were from contact with pulmonary TB cases (878/1077; 81.5%) and pulmonary TB cases had a mean of five contacts per case, significantly more than a mean of two contacts per non-pulmonary TB case (p<0.05). This compares with 1274 contacts traced in 2015 (range 0-238, mean four contacts, median two contacts) of whom 62.6% were from contact with non-pulmonary TB cases.
The result of a smear test was known for 64.4% (116/180) of pulmonary notifications. Of these, 87 (75.0%) were smear-positive and 29 (25.0%) were smear-negative. In 2016, 90.3% of all cases (280/310) had specimens sent for culture: 88.9% (160/180) of pulmonary and 92.3% (120/130) of non-pulmonary notifications. Of these, 74.3% (208/280) were confirmed by culture, 80.6% (129/160) of pulmonary and 65.8% (79/120) of non-pulmonary notifications. This is similar to the proportion of pulmonary and non-pulmonary cases confirmed by culture when compared with 2015 (130/162; 80.2% and 80/125; 64.0%, respectively).
Of the 208 culture confirmed cases in 2016, 206 (98.6%) were due to infection with M. tuberculosis, one (0.5%) each with M. bovis and M. africanum.
Isoniazid, rifampicin, ethambutol and pyrazinamide are considered first-line drugs for the treatment of tuberculosis in the UK. Of the 209 culture confirmed cases of tuberculosis reported in 2016, 207 (99.0%) had drug susceptibility tests for both isoniazid and rifampicin. Resistance to at least one first-line drug at the start of treatment was reported for 12 cases (5.8%), eight (3.9%) isolates being resistant to isoniazid, four (1.9%) were resistant to pyrazinamide, one (0.5%) was resistant to rifampicin and one (0.5%) was resistant to ethambutol (Table 9). However, one of the pyrazinamide-resistant cases was identified as M. bovis and was therefore not included in the analysis below.
One case (0.5%) was resistant to both isoniazid and rifampicin and therefore defined as multidrug-resistant tuberculosis (MDR-TB). This case case was male, aged 35-44 years, non-UK born, presented with pulmonary disease, was symptomatic for three months or more prior to their diagnosis and was not known to have a previous diagnosis of tuberculosis. No tuberculosis cases reported in 2016 had extensively drug-resistant tuberculosis (XDR-TB).
All first patient M. tuberculosis complex (MTBC) isolates since 2010 have been genotyped using 24VNTR PCR, aiding cluster investigation and management. If a new patient has MTBC isolates from different anatomical sites then both are included. In 2016, isolates from all culture-confirmed TB patients were genotyped with 96.2% having results for at least 23 loci available; 56.4% being in 88 molecular clusters with other Scottish isolates and 43.6% having a unique strain type.
Treatment outcomes information
Information on treatment outcome is reported at 12 months after initial tuberculosis notification and 24 months for those recorded as still on treatment at 12 months. For this reason, the treatment outcomes described here are for cases notified in 2015 and cases reported to still be on treatment in 2014. Outcomes for cases notified in 2016 will be reported in the 2017 annual report.
TB outcomes in the drug-sensitive cohort at 12 months
In 2015, 320 cases of tuberculosis were reported to ESMI of which two cases were rifampicin-resistant and were therefore excluded from outcomes analysis. Outcomes data were available for 310 (97.5%) cases, of whom 245 (79.0%) had successfully completed treatment at 12 months, which is slightly lower than that reported in 2014 when 80.3% of cases successfully completed their treatment (Figure 12). There were no significant differences in treatment completion by sex, place of birth, site of disease or having an identified risk factor for TB. However, those cases aged 65 years or older were less likely to successfully complete their treatment than those aged under 65 years (68.5% vs 82.3%, p=0.02).
The most common reason for not completing treatment was death (21 cases; 6.8%), having treatment stopped (11 cases; 3.5%), being lost to follow up (seven cases; 2.3%), or still being on treatment (nine cases; 2.6%). In addition, 20 cases (6.5%) were not evaluated (Table 11). The cases reported as still being on treatment at 12 months were either on their initially planned course of treatment (five cases), or had had their treatment changed (one case).
In the 21 cases (6.8%) reported to have died, tuberculosis was believed to have contributed to death in eight cases (38.1%), to have been incidental to death in seven (33.3%), to have caused death in two (9.5%) and to have had an unknown relationship to death in four cases (19.0%) (Figure 13).
Of the cases known to have died, more than two thirds (15/21; 71.4%) were male and three quarters were aged 65 years or older (16/21; 76.2%). The majority of deaths had pulmonary TB (14/21; 66.7%), were known to have been born in the UK (13/21; 61.9%) or were of white-Caucasian ethnicity (17/21; 81.0%).
TB outcomes in the drug-sensitive cohort at 24 months
In 2014, there were 15 cases of TB recorded as still being on treatment at 12 months. Treatment outcomes information at 24 months was available for 11 (73.3%) of these cases, all (100%) of whom had successfully completed their treatment at 24 months.
Surveillance data for Scotland show that the number of cases of tuberculosis remained relatively stable from 2000-2005 with between 350 and 400 cases per year. From 2005 to 2010 the number of cases steadily increased to 503, which was the highest number reported since ESMI began in 2000. Since 2010, however, there has been a steady reduction in case numbers with 310 cases reported in 2016, a reduction of 38% and the lowest number of cases reported since surveillance began. The incidence in Scotland has decreased significantly from 9.6 cases per 100,000 population in 2010 to 5.7 per 100,000 population in 2016 (a 41% reduction; p<0.001). As such, Scotland’s incidence remains well within the WHO target of under 10 cases per 100,000 and remains lower than that of the UK as a whole (9.4 cases per 100,000) and those of many other European countries (mean 11.7 cases per 100,000 population).1,4
There have been no changes to the surveillance system, data collection methods, laboratory testing or national screening policy and so the decrease in tuberculosis cases is not believed to be due to a lack of case ascertainment. However, accepted indicators of recent tuberculosis transmission suggest that transmission may be occurring in Scotland; the rate of tuberculosis among children under five years of age increased from no cases in 2015 to 3.1 in 2016 and the child under 15 years to adult notification ratio also increased, from 0.13 in 2015 to 0.25 in 2016.
In 2016, the rate of tuberculosis among those born outside the UK was approximately 20 times higher than among those born in the UK (45.3 cases per 100,000 population versus 2.3 cases per 100,000 population), the highest rate ratio reported since surveillance began. Cases born outside the UK were younger and presented more often with non-pulmonary disease than those born in the UK. The proportion of tuberculosis cases born outside the UK further increased in 2016, and now accounts for almost two thirds of cases (61.2%), the highest proportion reported since surveillance began. While this is lower than that observed for the whole of the UK (73% in 2016), it is higher than that reported in many other European countries (mean 29.8%). More than half of the non-UK born cases reported to ESMI originated from South Asian countries with high incidences of tuberculosis. The majority of non-UK born cases occurred in people who had lived in the UK for two years or more suggesting that the current screening system (pulmonary TB screening among migrants who have been resident in a high incidence country prior to applying for a visa to enter the UK) would probably have failed to detect them. Screening for latent tuberculosis infection on entry to the UK could however have detected those cases whose subsequent disease was as a result of re-activation of latent infection.
In Scotland the greatest burden is borne by Greater Glasgow and Clyde NHS Board (with more than half of all reported cases and an incidence of 10.2 cases per 100,000). However, in any one year individual NHS boards can have marked variations in the number and complexity of cases they must deal with. These local variations in incidence over short time periods continue to prove a challenge for tuberculosis control programmes for NHS boards in Scotland.
The reporting of risk factors identifies at-risk groups within the Scottish population. Just over one quarter of tuberculosis cases reported in 2016 had a known risk factor. Immunosuppression, alcohol misuse, refugee status and working in healthcare continue to be the most commonly reported risk factors in Scotland and pose significant challenges for tuberculosis treatment and control. For cases who work in a healthcare setting this is a challenge for occupational health services and employee screening as many originate from or travel to countries with a high incidence of tuberculosis. The TB Action Plan for Scotland5 recommends that local services are aware of these high risk groups and design approaches to better reach them, including awareness raising among GPs, and homelessness and alcohol services.
Efforts to control tuberculosis include measures to reduce the time interval between the development of a patient’s first symptoms and the start of their treatment. Patients with untreated pulmonary tuberculosis pose the highest risk of transmission to contacts. In 2016, 13.5% of pulmonary cases were symptomatic for more than six months before commencing treatment, which is higher than that reported in 2015 (7.4%). Historically, non-pulmonary cases are symptomatic for longer before commencing treatment than pulmonary cases. However, non-pulmonary tuberculosis poses a much lower risk of transmission than pulmonary tuberculosis.
Personal, social and cultural issues including deprivation are all factors which may increase a person’s risk of being exposed to and infected with tuberculosis. In order to explore the effect of deprivation on TB, this annual report for the first time presents information on TB case numbers by Scottish Index of Multiple Deprivation quintile (SIMD 2016). In 2016, almost one third of TB cases were in the most deprived category while 12% were in the least deprived, suggesting that TB is strongly associated with deprivation in Scotland. This trend is reflected historically, with just over one third of cases resident in the most deprived category and 12% in the least deprived category between 2000 and 2015.
The TB Action Plan for Scotland recommends that there should be an increase in the percentage of cases seen by the TB service within two weeks of first presentation to the NHS. Where dates were available in 2016, 20 cases (6.8%) had an interval of greater than two weeks from diagnosis to being commenced on therapy, which is higher than that reported in 2015 (12 cases, 3.8%).
In order to achieve optimal detection of infectious cases and detect drug resistance, a high proportion of cases should be bacteriologically confirmed. In 2016, the percentage of pulmonary tuberculosis cases confirmed by culture was 80.6%, which is similar to 2015 (80.2%) and meets the ECDC target6 which recommends that at least 80% of pulmonary cases be confirmed by culture.
Drug-resistant tuberculosis is more difficult and expensive to treat and because it takes more time to treat, the patient can remain infectious for longer. Following a peak in the number of isolates resistant to any first-line drugs in 2011, the number has reduced by more than 50%, mainly due to a decrease in the number and proportion of isolates resistant to isoniazid. None of the drug-resistant cases were known to have a previous TB diagnosis.
Direct molecular detection of M. tuberculosis complex on first film-positive sputa and resistance to rifampicin and isoniazid is available free of charge. Generally this can be done within one working day of specimen receipt where there is urgent need. Direct detection of other specimens is charged. These initiatives should help improve the timeliness of treatment and public health interventions in patients with tuberculosis and drug-resistant tuberculosis. In December 2016, PHE Midlands & North incorporated whole genome sequencing (WGS) into their routine service, discontinuing some ‘traditional’ diagnostic tests. Validation of WGS to determine mycobacteria species identification, some drug resistance prediction and typing in a single test is currently underway at SMRL. This will in time replace 24VNTR typing to better inform public health teams and guide TB cluster investigations.
The Department of Health’s toolkit7 specifies that 95% of patients with tuberculosis should have the outcome of their treatment recorded and the ECDC target for successful treatment completion at 12 months is 85%.5 For patients first diagnosed in 2015, treatment outcomes at 12 months were recorded for 97.5% of cases, of whom 79.0% successfully completed treatment at 12 months. This was a decrease in the percentage of cases successfully completing treatment when compared with 2014 (80.3%) and does not meet the ECDC target of 85%. As in previous years, the most common reason for not successfully completing treatment within one year was death. However, 6.5% (20 cases) were not evaluated and this is much higher than that reported for the whole of the UK in 2016 (1.8%).1 In total, 21 deaths were reported in 2015 representing a case fatality ratio of 6.8%. This is in line with the number of deaths and case fatality ratio recorded in 2014 (21 deaths; 6.2%).
- Public Health England. Reports of cases of tuberculosis to enhanced tuberculosis surveillance systems: United Kingdom, 2000 to 2016. London: PHE; 2016. Available from: https://www.gov.uk/government/statistics/reports-of-cases-of-tb-to-uk-enhanced-tuberculosis-surveillance-systems. (accessed 26 October 2017).
- National Records of Scotland. Population by country of birth and nationality. Available from: https://www.nrscotland.gov.uk/statistics-and-data/statistics/statistics-by-theme/population/population-estimates/estimates-of-special-populations/population-by-country-of-birth-and-nationality. (accessed 26 October 2017).
- World Health Organisation. Definitions and reporting framework for tuberculosis - 2013 revision. Geneva: WHO; 2013. Available from: http://apps.who.int/iris/handle/10665/79199. (accessed 26 October 2017).
- European Centre for Disease Prevention and Control/WHO Regional Office for Europe. Tuberculosis surveillance and monitoring in Europe, 2017. Stockholm: ECDC; 2016. Available from: https://ecdc.europa.eu/en/publications-data/tuberculosis-surveillance-and-monitoring-europe-2017. (accessed 26 October 2017).
- Scottish Government. A TB Action Plan for Scotland. Edinburgh: Scottish Government; 2011. Available from: http://www.gov.scot/Publications/2011/03/18095603/0. (accessed 26 October 2017).
- European Centre for Disease Prevention and Control. Progressing towards TB elimination. Stockholm: ECDC; 2010. Available from: https://ecdc.europa.eu/en/publications-data/progressing-towards-tb-elimination. (accessed 26 October 2017).
- Department of Health. Tuberculosis prevention and treatment: a toolkit for planning, commissioning and delivering high-quality services in England. London: Department of Health; 2007. Available from: http://webarchive.nationalarchives.gov.uk/20130123192158/http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_075621. (accessed 26 October 2017).
The authors would like to thank all of the clinical, microbiology, nursing and public health staff who have contributed to the operation of the ESMI scheme. Health Protection Scotland would also like to acknowledge the contribution made to ESMI by the Scottish Mycobacteria Reference Laboratory.