Factors Influencing the Diagnosis and Treatment of Latent Tuberculosis among Contacts in Congregated Settings in Korea

Ahreum Kim; Minhyeok Choi

doi:10.3346/jkms.2019.34.e138

Journal List > J Korean Med Sci > v.34(Suppl 1) > 1122118

Go to TopGo to Top Go to BottomGo to Bottom

TOOLS

Kim and Choi: Factors Influencing the Diagnosis and Treatment of Latent Tuberculosis among Contacts in Congregated Settings in Korea

Original Article

Preventive & Social Medicine

Journal of Korean Medical Science 2019; 34(18): e138.

Published online: 7 May 2019

DOI: https://doi.org/10.3346/jkms.2019.34.e138

Factors Influencing the Diagnosis and Treatment of Latent Tuberculosis among Contacts in Congregated Settings in Korea

Ahreum Kim

, Minhyeok Choi

Regional Center for Respiratory Diseases, Pusan National University Hospital, Busan, Korea.

Address for Correspondence: Minhyeok Choi, MD, PhD. Regional Center for Respiratory Diseases, Pusan National University Hospital, 179 Gudeok-ro, Seo-gu, Busan 49241, Republic of Korea. come2mh@gmail.com

Received 27 March 2019 Accepted 30 April 2019

The Korean Academy of Medical Sciences

(open-access):

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

This study aimed to compare the indicators (the rates of diagnosis, need for treatment, treatment initiation, and treatment completion) of management of latent tuberculosis infection (LTBI) in contacts and to identify the impact of active tuberculosis (TB) index case characteristics on the exposed population in congregated settings, such as schools, workplaces, and medical institutes.

Methods

The data of 8,648 clusters in the TB epidemiological investigation database between 2013 and 2016 were extracted and analyzed to evaluate the indicators and perform multilevel logistic regression (MLR) analyses to identify the factors affecting each indicator.

Results

The rates of total LTBI diagnosis, need for treatment, treatment initiation, and treatment completion were 15.2%, 10.2%, 69.4%, and 76.6%, respectively. After adjusting for other factors on MLR, the probability of diagnosis and need for treatment of latent TB in contacts was higher in most types of facilities than in schools. Conversely, treatment completion rates in these facilities were lower. Notably, the correctional institutions showed the highest odds ratio (OR) relative to school for LTBI diagnosis (OR, 6.37) and need for treatment (OR, 4.49) and the lowest OR for treatment completion (OR, 0.10).

Conclusion

This study provided evidence for the implementation of latent TB control policies in congregated settings.

Graphical Abstract

Keywords: Tuberculosis, Latent Tuberculosis Infection, Congregated Settings, Tuberculosis Epidemiological Investigation

INTRODUCTION

Tuberculosis (TB) is an infectious disease, most commonly of the respiratory system, caused by Mycobacterium tuberculosis. The symptoms on presentation include chronic cough, fever, and production of sputum, among others. According to the global tuberculosis report, 2017, approximately 1.6 million new cases of TB were reported worldwide in 2016, with approximately 1.67 million deaths.1

Asymptomatic individuals who do not transmit the infection despite being infected with M. tuberculosis are considered to have latent tuberculosis infection (LTBI).1 It is estimated that 1 out of 3 adults are infected with latent TB in Korea.2 About 10% of people infected with latent TB progress to active TB, so it is important to control the TB epidemic.3 4 In 2016, the Korea Centers for Disease Control & Prevention (KCDC) prepared an “Implementation plan” for a TB-safe country. The focus of the policy has been directed to the need for the reduction of TB cases through active detection and management of LTBI patients.2

Many Koreans are exposed to congregated settings (settings in which a large numbers of people spend a considerable amount of time together in relatively enclosed spaces) such as in schools and military institutions. Since the risk of a TB epidemic is high in congregated settings, TB interventions are particularly relevant in their context. Since 2013, the KCDC has performed epidemiological investigations in contacts of contagious TB cases with an exposure period of more than one day in congregated settings.2 In particular, the KCDC has performed epidemiological investigations and TB control projects in these settings using the National Tuberculosis Control Guidelines,2 the Tuberculosis Epidemiological Investigation Guidelines,5 the Tuberculosis Control Guidelines for Schools,5 the Tuberculosis Control Guidelines for Correctional Institutions,5 and the Tuberculosis Control Guidelines for the Military and Police.5 The guidelines, including the implementation criteria for epidemiological investigations, were revised along with the progress made on the TB control project in congregated settings since 2013. Studies on TB control in congregated settings in Korea and in other countries have comprised various populations including children, youths, and school-goers.6 7 8 These studies include those conducted among healthcare workers, patients in medical institutions,9 10 11 inmates in social welfare facilities,12 13 military and police units,14 15 16 correctional institutions,17 18 19 20 and workers in certain workplaces.21 There are also studies on prevalence, diagnosis, and completion of treatment for latent TB, which were mainly conducted to report on the effects of treatment initiation and completion among LTBI-infected people.22 23 24 25 In addition, some studies were performed on TB contacts.26 27 28 These preceding studies have partially investigated the characteristics or influencing factors in patients with LTBI, based on the type of facility. However, no studies have determined the indicators of the management of latent TB (diagnosis, judgment regarding need for treatment, and the start and completion of treatment) among contacts and LTBI patients, and the influence of the characteristics of the index case and the type of facility where the infection occurred.

The present study aimed to identify the factors influencing the occurrence of latent TB in the exposed population in congregated settings, based on the type of facility and the year. Indicators of different congregated settings were also calculated and compared, accounting for the type of facility and index case characteristics.

METHODS

Data sources

The present study utilized raw data from the TB epidemiological investigation databases in congregated settings, which were collected by the KCDC between January 1, 2013 and December 31, 2016. TB epidemiological investigations are performed according to the National Tuberculosis Control Guidelines when more than 2 active TB patients are identified within 6 months or more than 1 case of infectious TB occurs.2 For the management of TB in Korea, the databases are the only source of data for the onset of TB in the first and additional patients in congregated settings, their contacts, outcomes of diagnosis, and the treatment of latent TB. The TB databases include information collected from congregated settings and include data on epidemic clusters, TB patients, their contacts (i.e., those who had been in contact with a patient for more than 1 day during the period of communicability), and information on patients with LTBI. Based on the four most recent years (2013–2016) of epidemiological investigations, data for 8,648 TB clusters (occurring in congregated settings) were available for the present study. The number of clusters and contacts analyzed by process are shown in Fig. 1.

Fig. 1

Flow chart of clusters and contacts by process.

LTBI = latent tuberculosis infection.

Definitions of indicators and variables in the TB process

In this study, the general characteristic variables included were: gender of the first index case diagnosed with active TB in the cluster and age (0–12, 13–18, 19–29, 30–39, 40–49, 50–59, 60–69, and 70 years or older). TB-related variables including sputum test results (negative and nontuberculous mycobacterium [NTM], unknown, culture positive, and smear positive), chest radiography findings (normal, unknown, cavity-negative TB, cavity-positive TB), presence or absence of symptoms, and duration of symptoms (no symptom, 1–3 weeks, 4–5 weeks, 6–7 weeks, and 8 weeks and over).

To examine the effect of facility type, congregated settings were categorized as: school, workplace, medical institution, social welfare facility, military and police units, correctional institution, and other facilities. To find the effect of the year when an epidemiological investigation was performed, the year cluster was divided into two groups (2013–2014, 2015–2016). Since the database did not include information on the sub-type of schools, categorization according to school type was not possible.

For examining the variables related to the diagnosis and treatment of latent TB contacts, latent TB diagnosis, need for treatment, treatment initiation, and treatment completion were treated as outcome variables; rates for each indicator were then calculated. The diagnosis rate of latent TB was defined as the number of persons diagnosed with LTBI (n = 46,535 in 8,648 clusters) among the contacts who completed the test (n = 488,735). The rate of the need for treatment, was the percentage of people diagnosed with latent TB and deemed necessary to treat (individuals aged 35 years or younger, susceptible to isoniazid and rifampin, not immunosuppressed, no history of TB infection or treatment, and non-diabetic) (n = 32,938).2 The rate of treatment initiation was defined as the percentage of people who needed and were started on treatment (n = 25,002); the rate of completion of treatment was defined as the percentage of people who completed the treatment schedule (n = 19,562).

L T B I d i a g n o s i s r a t e (%) = \frac{N o . o f p e r s o n s w h o w e r e d i a g n o s e d w i t h L T B I}{N o . o f p e r s o n s w h o c o m p l e t e d t e s t f o r L T B I} \times 100

N e e d f o r t r e a t m e n t r a t e (%) = \frac{N o . o f p e r s o n s w h o w e r e j u d g e d t o n e e d t r e a t m e n t f o r L T B I}{N o . o f p e r s o n s w h o w e r e d i a g n o s e d w i t h L T B I} \times 100

R a t e o f i n i t i a t i o n o f t r e a t m e n t (%) = \frac{N o . o f p e r s o n s w h o w e r e s t a r t e d o n t r e a t m e n t f o r L T B I}{N o . o f p e r s o n s w h o w e r e j u d g e d t o n e e d t r e a t m e n t f o r L T B I} \times 100

R a t e o f c o m p l e t i o n o f t r e a t m e n t (%) = \frac{N o . o f p e r s o n s w h o c o m p l e t e d t r e a t m e n t f o r L T B I}{N o . o f p e r s o n s w h o s t a r t e d t r e a t m e n t f o r L T B I} \times 100

Analytic methods

The general characteristics of the first index case in the TB outbreak groups were described using univariate analysis according to the types of facilities. Mean values and 95% confidence intervals (CIs) were also evaluated to determine the rates of diagnosis, need for treatment, treatment initiation, and treatment completion of LTBI in each TB outbreak group. In addition, t-tests and variance analysis were performed to identify differences between clusters depending on the characteristics of the first index case.

To identify factors affecting the diagnosis and treatment of latent TB, multilevel logistic regression analysis was performed. Multilevel analysis is known as an analytic method that considers upper level contextual effects, such as the ‘area belonged to at the local level’ or ‘time for example, year cluster’.29 In the multilevel logistic regression model used for this study, dependent variables included diagnosis and need for treatment of latent TB, as well as the initiation and completion of latent TB treatment in TB outbreak groups. Independent variables corresponding to level 1 (individual level) were biological characteristics (gender and age), TB characteristics (sputum test results, chest radiography results, and presence or absence of symptoms and its duration) of the index case, as well as the facility type. Independent variables for level 2 (year level) were year clusters (2013–2014 and 2015–2016). In addition, the median odds ratio (MOR) was calculated to examine variations by year. MOR is calculated as follows:

M O R = e x p (\sqrt{2 {\times σ}^{2} \times ϕ^{- 1} (0.75)})

Here, ö refers to the cumulative distribution function in the normal distribution with a mean of 0 and a variance of 1, where ö⁻¹(0.75) is the 75th percentile.30 31 MOR refers to an indicator that quantifies the variation between levels, by comparison between two individuals between different levels (year cluster in the present study); this is the median value of the odds ratios (ORs) in one group with a higher tendency and another with a lower tendency. MOR is a value that evaluates the degree of variation in outcome variables at the higher-level cluster (year in the present study) compared to the individual level (when considering variables at the individual level), which is equivalent to the fixed-effects ORs.30 31 32 The MOR value is always 1 or more. An MOR value of 1 implies no variation between levels; the MOR value increases when there are considerable variations between levels. The present study used the PROC GLIMMIX command in the SAS 9.4 software package (SAS Institute Inc., Cary, NC, USA) for multilevel logistic regression analysis. The statistical significance level in this study was reported at P < 0.05.

Ethics statement

The present study was performed after approval from the Institutional Review Board (IRB) of the Pusan National University Hospital (IRB No. 1708-030-058). Informed consent was waived by the IRB.

RESULTS

General characteristics

Data pertaining to the characteristics for each epidemiological cluster is presented in Table 1. Among the 8,648 TB clusters investigated from 2013 to 2016, schools (2,925 clusters, 33.8%), accounted for the highest proportion of outbreaks according to facility type, followed by medical institutions (2,100 clusters, 18.8%), and social welfare facilities (1,624 clusters, 13.2%). In terms of the gender and age of the first index case, the total cluster had a higher proportion of men (4,870 clusters, 56.3%) than women; and the age group of 19–34 years accounted for the highest proportion (2,483 clusters, 28.7%). Regarding the sputum test of the index case, the highest rate of smear positivity was found to be 59.2% (5,188 clusters), while culture positivity was 31.7% (2,740 clusters). On chest radiography, the proportion of cases with cavity-negative TB was the highest at 61.5% (5,320 clusters), followed by cavity-positive cases, at 26.7% (2,306 clusters). The highest duration of symptoms for the index case was 1–3 weeks (3,371 clusters, 39.0%), followed by those with no symptoms (2,551 clusters, 29.5%), and those with symptoms for 4–5 weeks (1,275 clusters, 14.7%). For epidemiological investigation by year, the rate for the year 2015–2016 was 69.0% (2,678 clusters), which was 2.23-fold higher than that of the year 2013–2014.

Table 1

Number and percentages of the index case of cluster

Variables		No. (%)
Total		8,648 (100.0)
Types of facilities
	School	2,925 (33.8)
	Workplace	2,100 (24.3)
	Medical institute	1,624 (18.8)
	Social welfare facility	1,139 (13.2)
	Military	503 (5.8)
	Correctional facility	121 (1.4)
	Others	236 (2.7)
Gender
	Men	4,870 (56.3)
	Women	3,778 (43.7)
Age, yr
	0–12	47 (0.5)
	13–18	1,266 (14.6)
	19–29	2,483 (28.7)
	30–39	1,025 (11.9)
	40–49	989 (11.4)
	50–59	849 (9.8)
	60–69	397 (4.6)
	≥ 70	1,592 (18.4)
Sputum test
	Negative & NTM	645 (7.5)
	Unknown	145 (1.7)
	Culture (+)	2,740 (31.7)
	Smear (+)	5,118 (59.2)
Chest radiography
	Normal	241 (2.8)
	Unknown	781 (9.0)
	TB but cavity (−)	5,320 (61.5)
	TB and cavity (+)	2,306 (26.7)
Duration of symptom, wk
	None	2,551 (29.5)
	1–3	3,371 (39.0)
	4–5	1,275 (14.7)
	6–7	225 (2.6)
	≥ 8	1,226 (14.2)
Year
	2013–2014	2,678 (31.0)
	2015–2016	5,970 (69.0)

NTM = nontuberculosis mycobacterium, TB = tuberculosis.

Table 2 shows the rates of need for treatment, treatment initiation, and treatment completion by type of facility and year group. The total LTBI diagnosis rate for contacts who completed the latent TB test was 15.2% (95% CI, 14.8%–15.6%). Highest LTBI diagnosis rates by type of facility, were noted in correctional institutions (32.5%; 95% CI, 27.8%–37.2%), and by year, in the 2015–2016 group; the diagnosis rate increased with the advancement of years. The rate of need for treatment of contacts who were diagnosed with latent TB was 10.2% (95% CI, 9.9%–10.6%). Correctional institutions showed the highest rate of need for treatment of contacts at 18.9%, followed by the workplace, at 16.8%. By year, the need for treatment of contacts was 12.0% in 2015–2016; this value was approximately 2-fold higher than that in 2013–2014.

Table 2

Rate of indicator by process for LTBI diagnosis and treatment by cluster

Variables		Rate of diagnosis LTBI		Rate of need for treatment		Rate of initiation of treatment		Rate of completion of treatment
Variables		Mean (95% CI)	P value	Mean (95% CI)	P value	Mean (95% CI)	P value	Mean (95% CI)	P value
Total		15.2 (14.8–15.6)		10.2 (9.9–10.6)		69.4 (68.3–70.5)		76.6 (75.5–77.7)
Type of facilities			< 0.001		< 0.001		< 0.001		< 0.001
	School	8.3 (8.0–8.7)		6.3 (6.0–6.6)		72.9 (71.4–74.4)		81.1 (79.7–82.5)
	Workplace	25.6 (24.5–26.8)		16.8 (15.8–17.8)		60.2 (57.7–62.6)		69.6 (67.0–72.2)
	Medical institute	13.7 (12.7–14.6)		8.9 (8.1–9.7)		64.7 (61.6–67.9)		73.5 (70.3–76.7)
	Social welfare facility	16.4 (15.2–17.5)		9.8 (8.8–10.7)		76.0 (72.6–79.4)		75.0 (71.5–78.5)
	Military	11.5 (10.3–12.7)		8.4 (7.4–9.4)		77.3 (73.5–81.0)		78.6 (74.5–82.6)
	Correctional facility	32.5 (27.8–37.2)		18.9 (14.6–23.1)		69.6 (60.3–78.8)		69.9 (59.9–80.0)
	Others	14.4 (11.8–17.1)		11.3 (8.8–13.7)		73.2 (66.3–80.0)		71.4 (63.7–79.1)
Year			< 0.001		< 0.001		< 0.001		< 0.001
	2013–2014	11.3 (10.8–11.8)		6.5 (6.1–6.8)		76.2 (74.6–77.8)		83.1 (81.7–84.6)
	2015–2016	17.1 (16.5–17.7)		12.0 (11.5–12.5)		65.8 (64.4–67.2)		72.6 (71.2–74.1)

LTBI = latent tuberculosis infection, CI = confidence interval, NTM = nontuberculosis mycobacterium, TB = tuberculosis.

The treatment initiation rate of contacts diagnosed with latent TB needing treatment was 69.4% (95% CI, 68.3%–70.5%). By year, the treatment initiation rate of contacts significantly and progressively decreased in recent years, with a decline from 76.3% in 2013 to 64.2% in 2016 (P < 0.001).

The treatment completion rate of contacts who started treatment was 76.6% (95% CI, 75.5%–77.7%). Among facility type, schools showed the highest treatment completion rate at 81.1%, compared with the workplace, at 69.6%. The treatment completion rate of contacts progressively decreased in the recent years.

Influential factors in the process of latent TB: multilevel regression analysis

As shown in Table 3, multilevel logistic regression analysis was performed using the diagnosis of latent TB in contacts, need for treatment, treatment initiation, and treatment completion as dependent variables, and the characteristics of the index case and cluster as independent variables; years were considered as a random effect. All indicators showed significant differences depending on year (MOR, 1.133–1.189). Compared to school, the other facilities that the index case belonged to, such as the workplace, showed a significantly higher rate of latent TB diagnosis in contacts (P < 0.001). Correctional institutions had a high risk for latent TB diagnosis (OR, 6.37; 95% CI, 5.92–6.86). By year cluster, LTBI became significantly lower with the progression of years (OR, 0.75; 95% CI, 0.61–0.91). Among contacts, men index cases had a 1.16-fold higher LTBI diagnosis rate than women index cases. Compared to the 70 years or older age group, all other age groups showed a significantly higher LTBI diagnosis rate in contacts; however, there was no apparent pattern (P < 0.001).

Table 3

Impact of factors for diagnosis of contacts, need for treatment, initiation and completion of treatment for latent TB using multilevel logistic regression

Factors (index case)		Diagnosis of LTBI (n = 488,735)	Need for treatment (n = 46,535)	Initiation of treatment (n = 32,938)	Completion of treatment (n = 25,002)
Factors (index case)		aOR (95% CI)	aOR (95% CI)	aOR (95% CI)	aOR (95% CI)
Gender
	Women	1.00	1.00	1.00	1.00
	Men	1.16^a (1.13–1.19)	1.11^a (1.08–1.14)	1.19^a (1.12–1.27)	1.00 (0.92–1.08)
Age, yr
	0–12	1.90^a (1.64–2.21)	2.09^a (1.77–2.47)	1.19 (0.81–1.73)	0.42^a (0.28–0.64)
	13–18	1.41^a (1.33–1.49)	1.57^a (1.47–1.67)	1.48^a (1.27–1.72)	0.87 (0.73–1.04)
	19–29	1.30^a (1.24–1.37)	1.56^a (1.47–1.66)	0.94 (0.82–1.08)	0.55^a (0.47–0.65)
	30–39	1.71^a (1.62–1.80)	1.80^a (1.68–1.92)	1.01 (0.87–1.16)	0.58^a (0.49–0.69)
	40–49	2.04^a (1.94–2.15)	2.07^a (1.95–2.21)	1.32^a (1.14–1.52)	0.64^a (0.54–0.76)
	50–59	1.73^a (1.64–1.82)	1.76^a (1.66–1.88)	1.15 (1.00–1.33)	0.95 (0.80–1.13)
	60–69	1.32^a (1.24–1.41)	1.38^a (1.27–1.49)	0.94 (0.79–1.13)	0.98 (0.78–1.22)
	≥ 70	1.00	1.00	1.00	1.00
Sputum test
	Negative & NTM	1.00	1.00	1.00	1.00
	Unknown	0.99 (0.91–1.07)	0.91 (0.82–1.00)	0.91 (0.71–1.17)	0.97 (0.73–1.28)
	Culture positive	1.07^b (1.03–1.10)	1.08^b (1.04–1.12)	0.75^a (0.68–0.83)	0.95 (0.85–1.06)
	Smear positive	1.13^a (1.10–1.17)	1.15^a (1.11–1.20)	0.84^b (0.76–0.93)	1.11 (1.00–1.24)
Chest radiography
	Normal	1.00	1.00	1.00	1.00
	Unknown	1.11^b (1.03–1.20)	1.11^c (1.02–1.21)	0.70^b (0.56–0.87)	0.76^c (0.58–0.98)
	TB but cavity	1.15^a (1.08–1.23)	1.13^b (1.05–1.22)	0.73^b (0.60–0.89)	0.72^b (0.57–0.90)
	TB and no cavity	1.28^a (1.20–1.37)	1.25^a (1.16–1.35)	0.67^a (0.55–0.83)	0.59^a (0.47–0.75)
Duration of symptom, wk
	None	1.00	1.00	1.00	1.00
	1–3	1.00 (0.98–1.03)	1.03^c (1.00–1.06)	1.14^a (1.06–1.22)	0.75^a (0.69–0.81)
	4–5	1.09^a (1.05–1.12)	1.07^b (1.03–1.11)	1.08 (0.99–1.18)	0.83^b (0.74–0.93)
	6–7	1.18^a (1.10–1.28)	1.20^a (1.10–1.31)	1.10 (0.91–1.32)	1.21 (0.94–1.56)
	≥ 8	1.28^a (1.24–1.32)	1.34^a (1.29–1.40)	1.18^b (1.08–1.28)	0.73^a (0.66–0.81)
Type of facilities
	School	1.00	1.00	1.00	1.00
	Workplace	3.55^a (3.42–3.70)	2.42^a (2.31–2.53)	0.72^a (0.65–0.79)	0.71^a (0.62–0.80)
	Medical institute	3.05^a (2.93–3.18)	2.13^a (2.03–2.23)	1.17^b (1.05–1.30)	0.61^a (0.54–0.69)
	Social welfare facility	3.29^a (3.13–3.45)	2.28^a (2.15–2.42)	2.21^a (1.91–2.55)	0.76^b (0.65–0.89)
	Military	1.55^a (1.48–1.63)	1.53^a (1.45–1.61)	1.46^a (1.29–1.66)	0.98 (0.85–1.12)
	Correctional facility	6.37^a (5.92–6.86)	4.49^a (4.13–4.89)	1.74^a (1.42–2.15)	0.10^a (0.09–0.13)
	Others	2.07^a (1.91–2.24)	1.75^a (1.60–1.92)	1.48^a (1.20–1.82)	0.94 (0.75–1.20)
Year
	2013–2014	1.00	1.00	1.00	1.00
	2015–2016	0.75^b (0.61–0.91)	0.82^c (0.67–0.99)	0.65^b (0.49–0.87)	0.84 (0.62–1.13)
AIC		61,411.81	53,875.28	21,131.55	13,905.68
MOR		1.137	1.133	1.185	1.189

aOR = adjusted odds ratio, CI = confidence interval, LTBI = latent tuberculosis infection, NTM = nontuberculosis mycobacterium, MOR = median odds ratio.

^aP < 0.001; ^bP < 0.01; ^cP < 0.05.

When the need for treatment of latent TB was used as a response variable, most variables showed significance similar to that of latent TB diagnosis. When compared by type of facility that index cases belonged to, other facilities showed significantly high ORs for rates of need for treatment of latent TB compared to schools (P < 0.001). In particular, the risk for the need for treatment was the highest in correctional facilities (OR, 4.49; 95% CI, 4.13–4.89; P < 0.001). When OR was examined by year, the rate of need for treatment of contacts with latent TB was significantly lower in the most recent year (OR, 0.82; 95% CI, 0.67–0.99; P < 0.05). Men index cases had a 1.11-fold higher rate of need for treatment of contacts with latent TB compared to women cases; this was significantly higher in all ages of index cases (P < 0.001).

When treatment initiation of latent TB was used as a response variable, most variables were statistically significant. Other facilities except the workplace had significantly higher ORs of treatment initiation rate in contacts (P < 0.01), while the OR of the workplace was significantly lower, at 0.72 (95% CI, 0.654–0.793). By year cluster, the treatment initiation rate in latent TB became significantly lower in the most recent year (OR, 0.65; 95% CI, 0.49–0.87; P < 0.01). Compared to women cases, men index cases had a 1.19-fold higher rate of treatment initiation among contacts with latent TB. By age group, the 13–18 years and 40–49 years groups showed significantly higher rates than the group aged 70 years and above (P < 0.001); the rates in the other age clusters were not significant.

Multilevel logistic regression analysis using treatment completion in latent TB as the response variable, found that all variables except the gender of the index case and sputum test results were significant. When examined by facility type of index cases, all facilities except other facilities demonstrated significantly lower treatment completion rates among contacts (P < 0.01). Unlike other indicators, when OR was examined by year, it showed no significant results.

DISCUSSION

The present study is the first to calculate the indicators related to latent TB and to investigate the influencing factors among congregated settings in Korea. Depending on the type of facility, each indicator was found to have significant differences along the process, from diagnosis of latent TB to treatment completion; differences were also noted depending on year cluster.

According to the results of this study, the most extensive epidemiological investigation was conducted at schools, compared to other facilities. This may be explained by the fact that a longer time is spent in communal living in schools, with longer contact times. In addition, schools have a better reporting system and accessibility to medical systems than other facilities. Epidemiological investigations for TB among congregated facilities in Korea were started in schools from 2013. Remarkably, the number of epidemiological investigations rapidly increased with age, beginning from when the first index case was aged > 13 years; this was probably due to the fact that this is a rapid period of growth in youths, and they spend longer times in high school.6 As reported by Fox et al.26 in 2013, contact investigation is known to be important for children, because unlike older age clusters, in congregated settings with many children, the risk of developing active TB is higher, and the response to the latent TB test is often unreliable; more controls are therefore required.

Compared to schools, other facilities had significantly higher rates of diagnosis of LTBI and need for treatment and the rate was still high even after adjusting for other factors. In this study, correctional institutions showed the highest diagnosis rate and OR. In comparison to international studies, the LTBI rate of contacts in correctional institutions was higher in Korea than that reported in the United States (17%)17; however, it was lower than that of Spain (45.5%)18 and Brazil (49%–73%).20 It may be speculated that the LTBI risk in correctional institutions was higher than of other facilities owing to the overpopulated setting and small rooms, in addition to medical accessibility issues. Correctional institutions in Korea accommodate 1,098.8 inmates on average in each institution, corresponding to the highest level in the Organisation for Economic Co-operation and Development (OECD) member countries. Furthermore, the accommodation rate is 121.8%, corresponding to the second largest in the OECD member countries.33 Correctional institutions hold 4 inmates per 8.48 m² of room space and 6 inmates per 12.75 m² of room space; this corresponds to approximately 2 m² per person. The setting in which inmates stay, living together in small rooms with short periods of outdoor activities, seemed to enhance TB transmission. In 2015, 87 doctors working in correctional institutions were in charge of 54,581 prisoners, and one doctor treated about 200 prisoners every day.34 Under such circumstances, there could be delayed identification of prisoners infected with active TB, thus the possibility of transmission to contacts increases.

In this study, medical institutions had higher rates of latent TB diagnosis, need for treatment, and treatment initiation than schools. However, the treatment completion rate was lower. This finding may be explained by the fact that medical institutions perform more tests and monitor treatment more strictly. The hospital-ward characteristics in Korea possibly had some impact. In 2017, there were a total of 154,753 wards, of which 99,523 (64.3%) were general wards accommodating at least four inpatients, indicating that a high proportion are dormitory-type wards, and 85.9% of the total 534,157 beds were in non-isolated wards.35 The setting with the most bed supplies were the non-isolated wards; this was identified as the major cause of nosocomial infections during the Middle East respiratory syndrome (MERS) outbreak in 2015.36 Although treatment of latent TB was initiated, considering the short hospitalization period of within 1 month, a relatively low treatment completion rate of latent TB indicated a poor post-discharge control. To date, there is no clear regulation for post-discharge control of latent TB patients. A system has been established to share the information about TB-infected soldiers after discharge with the corresponding community health center. In contrast, for patients with LTBI, there is no system that registers these patients after discharge in the corresponding community health center or the TB control information system. Thus, there is need to establish a system that provides continuous treatment for patients with LTBI after discharge, and the same is applicable to social welfare facilities, as well as the medical and other institutions that have admission and discharge processes.

Although the rates of initiation and completion of treatment for latent TB have reduced, patients with LTBI aged ≥ 35 years have been included as treatment targets since 2015. After 2015, in view of the lower ORs for treatment initiation and completion, the general workplace has been particularly considered for active TB control interventions in congregated settings. In a similar study in Japan, workers with latent TB had a lower treatment completion rate than patients in other occupation groups such as school students, healthcare professionals, and homemakers.22 Workplaces had lower treatment and completion rates of latent TB; this could be attributed to the large numbers of individuals aged ≥ 35 years who were recently included as treatment subjects. It could also be explained by the fact that unlike schools or correctional institutions, the guidelines for the management of infected cases in the workplace congregated settings were unclear leading greater difficulties in the management of infected individuals. Overall, these results show that unlike the case of active TB patients a system of management from initiation to completion of treatment has not been fully established for LTBI, despite the expansion in numbers of treatment subjects with latent TB. An additional system, which can monitor and manage the treatment process in patients with LTBI aged 35 years or older is also necessary.

Despite the significant findings of this study, there are a few limitations. First, the database contained only the information of Korean individuals; non-Korean individuals were not included in the analysis. The number of new TB patients among foreigners has increased in recent years, while that of Koreans has tended to decrease, showing an opposite trend.37 Follow-up studies should be performed among foreigners with TB, whose numbers seemed to be on an increase. Secondly, the characteristics (gender and age, among others) of contacts of TB patients were not considered. However, the present study primarily intended to investigate the characteristics of clusters including type of facility and effect of timing after considering the characteristics of the index case; therefore, characteristics of contacts were not a major issue. Since the age of the contacts was a characteristic of the cluster they belonged to, the range was relatively similar and its influence was considered minor. Future studies should be performed with individual units considering the characteristics of contacts. Thirdly, the characteristics of the sub-categories in the congregated settings were not considered in this study. Schools maybe classified into kindergarten, elementary school, middle school, high school, and college; other facilities also have sub-types. In previous studies, the highest number of contact cases in Korean schools was found in high schools. Kindergartens and daycare centers also showed high LTBI rates.6 As for correctional institutions in the United States, prisons where they serve their terms, had a higher treatment completion rate than detention centers that held inmates temporarily.17 Since the details supplied by the facilities were incomplete, the data in the present study did not completely reflect all sub-types of facilities. However, this study had a different goal and those variables were not considered. Further studies are necessary to evaluate individual facilities according to sub-types. Lastly, the jobs among the first index cases in congregated settings were not classified. School populations could be sub-categorized as students and teachers, and medical institutions could be sub-divided into patients and healthcare professionals (doctor and nurse). These characteristics are likely to vary. For instance, the index cases in a study had different types of occupations depending on school type6; in another study, healthcare professionals had a higher risk of LTBI due to contact with patients.13 However, the analysis in the present study was based on various types of facilities. Therefore, they were not sub-divided to ensure consistency in the analysis. In future studies, analysis should be performed by facility type as well as by occupation in individual facilities. Despite these limitations, this study identified the impact of characteristics of the first index case, type of facility, and timing in each cluster on latent TB; the findings of this study are therefore significant.

These results have considerable implications. First, the overall incidence level in the school population needs to be reduced. Since this population has more contact with the family than that of other collective facilities, it is also necessary to manage the family contacts.

Second, a strategy is needed to increase the TB-treatment completion rate among contacts in other facilities. In Korea, the management of TB in collective facilities has been focused on finding additional TB patients and identifying LTBI cases through epidemiological investigations in case of epidemics.37 38 Conversely, efforts to monitor treatment in cases with LTBI were relatively less enthusiastic.37 38

In Korea, patients with active TB are managed by the TB public-private mix model (PPM) recommended by the World Health Organization.1 38 The PPM is regarded as an effective policy for TB control in Korea and other countries.1 37 38 However, till date, people with LTBI have not been included in the target population of the PPM. For improving levels of completion of LTBI treatment, the active management of LTBI is essential. This may be achieved by sharing information about patients to medical institutions participating in the PPM; also, public health centers need to monitor and manage the completion of treatment via telephone and text communication.

The directly observed treatment (DOT) service needs to be extended to include patients with LTBI. In Korea, the DOT service has been implemented since 2013 for patients with TB.38 However, till date, patients with LTBI are not included in the target population for DOT. Since DOT is known to be effective in improving the compliance of TB patients,1 it should be implemented to increase the treatment completion rate in these patients.

Third, a cohort of contact and latent TB cases should be established to monitor the status of management and the occurrence of active TB in the long-term. It is known that individuals with LTBI have a 10% of risk of latent TB for the rest of their lives.38 In contact control programs, epidemiological investigations can take up to 9 months.2 However, a period of 9 months is insufficient for obtaining information about the occurrence of active TB in contacts.

Fourth, tests should be performed for latent TB before including individuals in congregated settings and also during the period of stay. Fifth, accessibility to medical institutions should be improved to facilitate timely access to medical services in case of suspected TB symptoms. Lastly, it is necessary to establish a system that transfers information about latent TB patients to the corresponding community health centers after discharge from the congregated settings; this will serve to raise treatment completion rates in patients with latent TB.

The present study investigated the states and influential factors along the process, extending from the diagnosis of latent TB to the completion of treatment in Korea. The findings showed that compared to schools, other facilities carried high risks of LTBI. Therefore, efforts should be made to enhance the early detection of the first patients with TB. In addition, efforts should be made to establish policies to determine the rates of treatment completion among latent TB cases in correctional institutions, medical institutions, and social welfare facilities, and to improve the treatment initiation rate in workplaces, particularly among those aged 35 years or older. There is a growing understanding of the need to manage LTBI, but the evidence for this is lacking. Further research is needed to facilitate the implementation of evidence-based policies for the control of LTBI.

Notes

^Funding This research was supported by a fund (No. 2017-71) by research of Korea Centers for Disease Control & Prevention (KCDC).

^Disclosure The authors have no potential conflicts of interest to disclose.

^{Author Contributions}

Conceptualization: Choi MH.
Data curation: Kim AR.
Formal analysis: Kim AR.
Investigation: Choi MH, Kim AR.
Methodology: Choi MH.
Supervision: Choi MH.
Validation: Choi MH.
Writing - review & editing: Kim AR, Choi MH.

References

1. World Health Organization. Global Tuberculosis Report 2017. Geneva: World Health Organization;2017.

2. Korea Centers for Disease Control & Prevention (KCDC). 2017 National Tuberculosis Control Guideline. Cheongju: KCDC;2017.

3. Kim HW, Kim JS. Treatment of latent tuberculosis infection and its clinical efficacy. Tuberc Respir Dis (Seoul). 2018; 81(1):6–12.

4. Choi JC. Diagnosis and treatment of latent tuberculosis infection. J Korean Med Assoc. 2014; 57(1):12–18.

5. Korea Centers for Disease Control & Prevention (KCDC). Guidelines for Management of Tuberculosis Epidemiological Investigation. Cheongju: KCDC;2012.

6. Eun BW, Kwon YH, Lee YK. Contact investigation for tuberculosis among children and adolescents in Korea, 2013–2015. Public Health Wkly Rep. 2016; 9(39):776–778.

7. Lee PH, Chang FY. Latent infection treatment to prevent TB transmission in school settings. J Formos Med Assoc. 2012; 111(10):525–526.

8. Jung DH, Jo KW, Shim TS. Prevalence of latent tuberculosis infection among medical students in South Korea. Tuberc Respir Dis (Seoul). 2012; 73(4):219–223.

9. Pan SW, Kou YR, Hu TM, Wu YC, Lee YC, Feng JY, et al. Assessment of latent tuberculosis infection in psychiatric inpatients: a survey after tuberculosis outbreaks. J Microbiol Immunol Infect. 2016; 49(4):575–583.

10. He GX, van den Hof S, van der Werf MJ, Wang GJ, Ma SW, Zhao DY, et al. Infection control and the burden of tuberculosis infection and disease in health care workers in china: a cross-sectional study. BMC Infect Dis. 2010; 10(1):313.

11. Jo KW, Hong Y, Park JS, Bae IG, Eom JS, Lee SR, et al. Prevalence of latent tuberculosis infection among health care workers in South Korea: a multicenter study. Tuberc Respir Dis (Seoul). 2013; 75(1):18–24.

12. Park YS, Cheong K. A case study of pulmonary tuberculosis outbreak in a facility for the physically or mentally disabled. In : Proceeding of the 50th Fall Conference of the Korean Society for Preventive Medicine; 1988 Oct 1; place unknown. Seoul: Korean Society for Preventive Medicine;1988. p. 131–132.

13. Hwang J, Cheong J, Choi BS. Prevalence of and risk factors for latent tuberculosis infection among employees at a workers' compensation hospital. J Korean Soc Occup Environ Hyg. 2017; 27(3):238–244.

14. Choi CM, Kang CI, Kim DH, Kim CH, Kim HJ, Lee CH, et al. The role of TST in the diagnosis of latent tuberculosis infection among military personnel in South Korea. Int J Tuberc Lung Dis. 2006; 10(12):1342–1346.

15. Moon CG, Park SJ, Cho MG, Kim YJ, Kim SY, Kim YK, et al. Clinical characteristics of tuberculosis in combat and auxiliary police in Korea. Infect Chemother. 2006; 38(6):383–388.

16. Ji SH, Kim HJ, Choi CM. Management of tuberculosis outbreak in a small military unit following the Korean national guideline. Tuberc Respir Dis (Seoul). 2007; 62(1):5–10.

17. Lobato MN, Leary LS, Simone PM. Treatment for latent TB in correctional facilities: a challenge for TB elimination. Am J Prev Med. 2003; 24(3):249–253.

18. García-Guerrero J, Marco Mouriño A, Sáiz de la Hoya Zamácola P, Vera-Remartínez EJ. Grupo de estudio PREVALHEP de prisiones. Multi-centre study of the prevalence of latent tuberculosis infection amongst inmates in Spanish prisons. Rev Esp Sanid Penit. 2010; 12(3):79–85.

19. Dara M, Acosta CD, Melchers NV, Al-Darraji HA, Chorgoliani D, Reyes H, et al. Tuberculosis control in prisons: current situation and research gaps. Int J Infect Dis. 2015; 32:111–117.

20. Carbone AS, Paião DS, Sgarbi RV, Lemos EF, Cazanti RF, Ota MM, et al. Active and latent tuberculosis in Brazilian correctional facilities: a cross-sectional study. BMC Infect Dis. 2015; 15(1):24.

21. Wang HY, Wang KC, Tsai SH, Lai PF, Lin WC, Lee HC, et al. A tuberculosis outbreak at a workplace in Central Taiwan, 2016. Taiwan Epidemiol Bull. 2018; 34(4):59–64.

22. Kawatsu L, Uchimura K, Ohkado A. Trend and treatment status of latent tuberculosis infection patients in Japan - Analysis of Japan TB surveillance data. PLoS One. 2017; 12(11):e0186588.

23. Goswami ND, Gadkowski LB, Piedrahita C, Bissette D, Ahearn MA, Blain ML, et al. Predictors of latent tuberculosis treatment initiation and completion at a U.S. public health clinic: a prospective cohort study. BMC Public Health. 2012; 12(1):468.

24. Stagg HR, Zenner D, Harris RJ, Muñoz L, Lipman MC, Abubakar I. Treatment of latent tuberculosis infection: a network meta-analysis. Ann Intern Med. 2014; 161(6):419–428.

25. Kan B, Kalin M, Bruchfeld J. Completing treatment for latent tuberculosis: patient background matters. Int J Tuberc Lung Dis. 2013; 17(5):597–602.

26. Fox GJ, Barry SE, Britton WJ, Marks GB. Contact investigation for tuberculosis: a systematic review and meta-analysis. Eur Respir J. 2013; 41(1):140–156.

27. Jaganath D, Zalwango S, Okware B, Nsereko M, Kisingo H, Malone L, et al. Contact investigation for active tuberculosis among child contacts in Uganda. Clin Infect Dis. 2013; 57(12):1685–1692.

28. Park JS. Prevalence and risk factors for Mycobacterium tuberculosis infection among contacts of pulmonary tuberculosis patients. Tuberc Respir Dis (Seoul). 2012; 72(2):140–148.

29. Kawachi I, Berkman LF. Neighborhoods and Health. New York, NY: Oxford University Press;2003.

30. Larsen K, Merlo J. Appropriate assessment of neighborhood effects on individual health: integrating random and fixed effects in multilevel logistic regression. Am J Epidemiol. 2005; 161(1):81–88.

31. Sanagou M, Wolfe R, Forbes A, Reid CM. Hospital-level associations with 30-day patient mortality after cardiac surgery: a tutorial on the application and interpretation of marginal and multilevel logistic regression. BMC Med Res Methodol. 2012; 12(1):28.

32. Matozinhos FP, Gomes CS, Andrade AC, Mendes LL, Pessoa MC, Friche AA, et al. Neighbourhood environments and obesity among adults: a multilevel analysis of an urban Brazilian context. Prev Med Rep. 2015; 2:337–341.

33. Ministry of Justice. Korea Correctional Service. Updated 2017. Accessed March 15, 2019. http://www.corrections.go.kr/corrections/1193/subview.do.

34. Park IH. Prison doctors treat 216 people a day. Herald Corporation [Internet]. 2016. 10. 10. Updated 2016. Accessed March 15, 2019. http://news.heraldcorp.com/view.php?ud=20161010000897.

35. Health Insurance Review & Assessment Service. 2017 Health Insurance Statistical Yearbook. Seoul: National Health Insurance Service;2018.

36. World Health Organization. WHO statement on the ninth meeting of the IHR Emergency Committee regarding MERS-COV. Updated 2015. Accessed May 2, 2019. https://www.who.int/mediacentre/news/statements/2015/ihr-ec-mers/en.

37. Cho KS. Tuberculosis control in the Republic of Korea. Epidemiol Health. 2018; 40:e2018036.

38. Cho KS. Tuberculosis control in the Republic of Korea. Health Soc Welf Rev. 2017; 37(4):179–212.

TOOLS

ORCID iDs

Ahreum Kim
https://orcid.org/0000-0002-2930-1232

Minhyeok Choi
https://orcid.org/0000-0002-9611-7232

Similar articles