National TB prevalance survey - DPRK - searo.who.int · aimed at understanding the real burden of the TB disease within the country and finding ways to improve TB detection and control

National prevalence surveyDemocratic People’s Republic of Korea

2017

TB

National TB Prevalence Survey

Democratic People’s Republic of Korea

Technical Brief

iiiNational TB Prevalence Survey – Democratic People’s Republic of Korea

Contents

Acknowledgement ................................................................iv

Acronyms ...............................................................................v

Executive summary .............................................................. vii

1. Introduction .................................................................1

1.1 Trend of TB prevalence in DPR Korea .................... 1

1.2 National TB Prevalence Survey ............................... 2

2. Survey results ...............................................................8

2.1 Survey participation by sex and age ...................... 8

2.2 Survey participation by Urban, Rural and Regions . 9

2.3 Occupation distribution of survey participants ....... 9

2.4 Smoking status among the survey participants ...... 9

2.5 Contact status with known TB cases ................... 10

2.6 Screening ............................................................ 10

2.7 Laboratory examination ...................................... 11

2.8 TB cases .............................................................. 13

3. Success, challenges and lessons learnt from survey design to finalization of survey result .....................18

iv National TB Prevalence Survey – Democratic People’s Republic of Korea

Acknowledgement

The achieved success is due to NTP’s contribution of USD 481 963 to the survey and the comparative use of less external funds (USD 896 026 from Global Fund) than other Asian countries for a successful implementation of National TB Prevalence Survey.

Technical assistance for countries which do not have any past survey experience is essential during the survey preparation, field operation and data analysis. In country, the UNICEF country office was responsible for the procurement and WHO country office provided the technical support for protocol development, field operation, final data analysis and report preparation. Through the WHO support, technical consultant who was involved in the Indonesian National TB Prevalence Survey visited DPR Korea in June 2016, reviewed survey implementation and provided technical support for data analysis. In the end of July 2016, two lab experts from Hong Kong SNRL reviewed the laboratory progress and validated survey lab results.

In 2017, WHO HQ experts reviewed the whole survey database two times and provided the technical support for a joint survey data review and final analysis.

vNational TB Prevalence Survey – Democratic People’s Republic of Korea

Acronyms

AFB-S Acid-fast bacilli smear

BSC Biosafety cabinet

C&DST Culture and drug susceptibility testing

CDMU Central data management unit

CI Confidence interval

CNR Case notification rates

CP Continuation phase

CPC Cetylpyridinium chloride

CTPI Central TB Preventive Institute

CXR Chest X-ray

DM Diabetes mellitus

DOT Directly Observed Treatment

DOTS Directly observed treatment—short course

DPR Korea Democratic People’s Republic of Korea

DRTB Drug-resistant tuberculosis

DR-TB Drug resistant TB

EA Enumeration area

EQA External quality assurance

FLD First Line Drugs

FM Fluorescent microscopy

GFATM The Global Fund to fight AIDS, Tuberculosis and Malaria

HIV Human immunodeficiency virus

IP Intensive phase

IPW Inverse probability weighting

KAP Knowledge, attitude and practice

L-J Lowenstein Jensen

M & E Monitoring and evaluation

vi National TB Prevalence Survey – Democratic People’s Republic of Korea

MDGs Millennium Development Goals

MDR-TB Multidrug-resistant tuberculosis

MI Multiple imputation

MoPH Ministry of Public Health

MOTT Mycobacterium other than tuberculosis

MTB Mycobacterium tuberculosis

NA Not available

NPS National TB Prevalence Survey

NRL National TB Reference Laboratory

NTM Non Tuberculosis Mycobacterium

NTP National Tuberculosis Control Programme

OR Odds ratio

PIN Personal identification number

PMU Project Management Unit

PPS Probability proportional to size

PR Principal Recipient

PTB Pulmonary tuberculosis

PTPI Provincial TB Preventive Institute

RSE Relative standard error

SD Standard deviation

SE Standard error

SLD Second-line drugs

SNRL Supra national reference laboratory

SOP Standard operating procedure

TA Technical assistance

TAG Technical advisory group

TB Tuberculosis

UNICEF United Nations Children’s Fund

WHO World Health Organization

ZN Ziehl-Neelsen

viiNational TB Prevalence Survey – Democratic People’s Republic of Korea

Executive summary

The National Tuberculosis (TB) Prevalence Survey conducted in DPR Korea (2015–2016) aimed at understanding the real burden of the TB disease within the country and finding ways to improve TB detection and control activities. This is the first nationwide TB prevalence survey in DPR Korea.

The National TB Prevalence Survey is cross-sectional survey utilizing multi-stage sampling methods. Multi-stage sampling methods include the stratified sampling method and the random cluster sampling method.

The capital of DPR Korea is Pyongyang city and the country is geopolitically split into nine provinces and three municipalities, which are then divided into 210 cities and counties.

Counties are subdivided into almost 4000 smaller administrative areas such as Ri, Ub, Gu and Dong. Survey clusters for the National TB Prevalence Survey were selected throughout the entire country with the exception of Jagang Province. There was no major sex ratio difference found with the survey census data when compared with the national data published by Central Bureau of statistics in 2015,

There were a total of 100 survey clusters with a total 71 877 eligible survey population out of which 60 683 were the eligible survey participants after excluding children population under 15 years of age.

The prevalence survey identified high levels of TB overall, highlighting that TB remains a public health concern in DPR Korea. Based on the survey results, the estimated national TB prevalence rate (in all forms and among all ages) is 641 as per 100 000 populations. This estimation is 1.31 times higher than 2012 national TB prevalence rate estimation (490 cases per 100 000 population) when the National TB Prevalence Survey design was conducted and 1.25 times higher than WHO estimation (511 cases per 100 000 population).

In the National TB Prevalence survey, a total of 356 bacteriologically confirmed pulmonary TB cases were detected, out of which 340 cases were confirmed as survey TB cases based on the follow up results and central expert committee discussion.

Among the 356 bacteriologically confirmed pulmonary TB cases, 247 persons (69.4%) were males and 109 persons (30.6%) were females. The sex ratio of smear

viii National TB Prevalence Survey – Democratic People’s Republic of Korea

positive pulmonary TB cases was 2.4:1 and that of bacteriologically confirmed pulmonary TB cases were 2.3:1. This result shows that survey result is almost similar to past NTP notification data.

Among 203 smear positive cases, 16 cases with culture negative results were excluded from the survey TB cases based on the central expert committee’s decision. Therefore there was a total of 340 survey TB cases, out of which 187 were smear positive TB cases and 153 were smear negative TB cases.

Among the smear positive TB cases, the culture positive was 84.5% and the culture negative was 15.5%. Among the smear negative TB cases, the culture positive was 39.2% and scanty culture was 60.8%.

There were a total of 760 cases that had missing lab results (those who did not produce sputum samples or cultures are contaminated). Such cases have been taken into account and multiple imputations applied for calculating the final TB prevalence.

Among the survey TB cases, 187 cases (55.0%) were detected both by interview and chest X-ray. 146 cases (42.9%) were screened only by chest X-ray. 7 cases (2.1%) were suspected as a TB case by interview alone. Among these 7 cases, 5 persons could not undergo chest X-ray.

Among the 187 smear positive cases, 62.6% were identified both by interview and chest X-ray, while 34.2% were identified by chest X-ray alone and 3.2% by interview alone. This result shows that there is high probability of getting TB in TB suspects identified both by interview and CXR.

Distribution of survey TB cases by sex was 69.7% of males and 30.3% of females. In males survey TB cases were found 2.3 times higher than in females. Survey TB cases increased with age until 55 years.

Based on the survey TB cases and the number of survey participants, the survey team calculated the prevalence of bacteriologically confirmed pulmonary TB among 15 years and above population as 567 per 100 000 population (510–631). As per WHO recommendation, the team used model 3 with the survey data to estimate TB prevalence. First the survey team conducted multiple imputations for missing cases and applied inverse probability weighting to the eligible non participants and no sputum collection candidates for TB prevalence estimation. The result was 587 per 100 000 population.

ixNational TB Prevalence Survey – Democratic People’s Republic of Korea

Prevalence of males is 2.9 times higher than that of females and in 45–54 age group it was highest as 877 per 100 000 population. TB prevalence in rural area is 1.14 times higher than in urban areas.

After applying the sex proportion weighting to the eligible population in the country, the result was 597 per 100 000 populations.

Proportion of males among enumerated population was 44.8% and among national population it was 46.54 %( excluding army), thus weighting for males have been made.

Given that the proportion of under 15 years of age population is 20.7% and ratio of pediatric TB prevalence to adult TB prevalence is 26.6% and proportion of extra-pulmonary cases is 21% by using last 5 years of NTP notification data, TB prevalence of all age, all forms is estimated as 641 per 100 000 population.

1.27% of smokers were identified as survey TB cases, which was 3 times higher than the proportion of survey TB cases among non-smokers (0.43%).

This result shows that smoking is one of key risk factors for TB and explains why the TB prevalence in males is much higher than the one in females in the country.

Among the survey participants who had contact with TB case in family, 3.76% were confirmed as survey TB cases, which was 7.7 times higher than those who did not have TB case in family.

2.91% of survey participants who had contact with TB case were diagnosed as survey TB case with a rate of 6.3 times higher than those who did not have any contact with TB case in past two years.

x National TB Prevalence Survey – Democratic People’s Republic of Korea

1National TB Prevalence Survey – Democratic People’s Republic of Korea

1 Introduction

Throughout the world, TB is a disease with a high incidence rate and high rate of mortality among infectious diseases. Due to the vicious strategy of isolation and squeezing by the USA and its satellite countries, and natural disasters occurring in rapid succession, TB emerged once again as a high-burden disease in the country.

The first step for treatment and prevention of TB is to identify the negative effect of TB on the public health sector with an emphasis on identifying the TB epidemiological situation in the country.

Knowing the precise situation of TB epidemiology in the country is of high importance for the evaluation on the current implementation status of TB control strategy and for further development of the succeeding TB control strategy.

Previous evaluation on the case notification rate, treatment success rate and regional TB epidemiology had been conducted for some years based on the number of estimated TB cases from Annual Risk of Tuberculosis Infection (ARTI), though some discrepancies were observed in the evaluation and some questions rose about the reliability of ARTI estimations of TB cases.

Therefore, despite the high survey cost, many countries planned and conducted National TB Prevalence Surveys to more accurately estimate the number of TB cases. As a result the countries made a strong improvement in case detection, registration and management, achieving beyond what was expected.

The MoPH of DPR Korea decided to conduct a National TB Prevalence Survey using a sampling method representative of the whole country in order to establish the precise situation of TB epidemiology in the country and to organize TB prevention and treatment in a scientific way.

This was the first time in the country to conduct national TB Prevalence Survey. Therefore, DPR Korea evaluated the experience of other country surveys, adopted WHO-recommended survey methodology and developed all survey forms, SOPs and processes of data analysis specific to country context. Based on the preparation, we conducted the National TB Prevalence Survey successfully.

1.1 Trend of TB prevalence in DPR Korea

To estimate the real burden of disease caused by TB in the entire country, DPR Korea National TB Control Programme (NTP) conducted the Annual Risk of Tuberculosis

2 National TB Prevalence Survey – Democratic People’s Republic of Korea

Infection (ARTI) survey under WHO technical support in 2007. The ARTI of DPR Korea was thus estimated at 3.1% (2.8~3.3%). Based on this result the new smear positive TB cases were estimated at 155±34 per 100 000 population.

According to the WHO Global TB Report published in 2011, the estimate for all forms of TB in DPR Korea was 84 000 (72 000~97 000) and TB incidence was estimated at 345 (100~698) per 100 000 population in 2010. Death due to TB was estimated at 5700 (4100~9400) giving TB a mortality rate of 23 (17~39) per 100 000 population.

In 2011, NTP in DPR Korea reported that notifications for all forms of TB cases were 386 per 100 000 population and TB notification for new smear positive TB were 134 cases per 100 000 population.

1.2 National TB Prevalence Survey

1.2.1 Survey objectives

Primary objective

The National TB Prevalence Survey in DPR Korea (2015–2016) aimed to have a new understanding of the real burden of disease caused by TB and to find solutions to improve TB control activities.

Secondary objectives

§ To measure bacteriologically confirmed pulmonary TB among population aged 15 years and above in DPR Korea.

§ To re-estimate TB burden (TB prevalence and mortality) by comparing survey results with previous TB prevalence estimation and notification data and to set the baseline for assessing trends in TB disease burden.

§ To assess the proportion of people with TB symptoms.

§ To evaluate the relationship between social, demographic, health factors and TB.

§ To measure the CXR epidemiology suggestive of TB among the adult population.

1.2.2 Sampling method

3 tier/stage stratified multistage cluster samplings were used. First, the country was stratified into urban and rural areas. Second, urban and rural areas were divided

–


into east coast region, west coast region and in-land region separately. Thirdly, all administrative units were enumerated and clusters were selected according to probability proportional to size (PPS) sampling.

The eligibility criteria for the survey were as follows:

§ Individuals aged ≧15 years. The prevalence of smear positive pulmonary TB among those aged <15 was very low. In DPR Korea, more than 90% of smear-positive pulmonary TB cases developed from the adults aged ≧15 years, plus it was difficult to collect sputum samples from children.

§ I nd i v i dua l s r e s i d ed in flats in a peripheral administrative unit for more than 2 weeks (A l l r e s ident s were registered in their living administrative unit 2 weeks before the survey census)

§ From households in the community.

The exclusion criteria for the survey were as follows:

§ Travelers staying at hotels and hospitalized individuals in survey areas: The chance of collecting samples from these individuals was very rare. Moreover, travelers could not be representative of resident people in the survey population.

§ Individuals serving in the army and security forces: These individuals were the same sex and cannot represent the survey population.

§ Prisoners.

1.2.3 Screening

A screening method recommended by WHO was used. Participants suspected of having TB either by chest X-ray or by the interview were examined using sputum smear microscopy and culture.

Reception of survey participants


The eligibility for sputum collection in terms of chest X-ray was defined as all those with an abnormal chest radiograph in the lung field or mediastinum other than a single small calcification nodule with a size less than 10 mm or pleural adhesion at costo-phrenic angle(s). Even if there were abnormal findings incompatible with TB such as bronco-pneumonia or bronchiectasis, sputum collection was requested for the purpose of screening. The team leader interpreted all films taken that day the same night for quality assurance.

The final decisions from the survey TB cases were made by a central expert committee referencing the screening data.

1.2.4 Survey implementation

The survey implementation and refinement underwent an Assessment visit, Pre-survey visit, Pilot survey before the actual Field operation

Phase 1 of the field operation was conducted from Oct to Dec 2015 and Phase 2 from Feb to May 2016. During Phase 1, two field teams conducted the field operation in 16 clusters within Pyongyang city and during phase 2, six field teams undertook the operation within the remaining 84 clusters nationwide.

Smear microscopy & Culture

Interview, Chest X-ray

No Smear microscopy & Culture

Normal (Interview, Chest X-ray)

Abnormal (Interview, Chest X-ray)

Figure 1: Survey Screening

Pilot Survey at Kaesong Dong, Morangbong District


1.2.5 Quality assurance measures

The protocol was designed in line with WHO guideline and reviewed by WHO Task Force Members on several occasions. Regular and mid-term monitoring was conducted throughout field implementation.

1.2.6 Survey field operation

Symptom Screening

Doctors in the interview unit undertook careful interviews with survey participants and recorded their answers to the Questionnaire.

X-ray screening

The X-ray reader interpreted the film carefully to discern whether it had any abnormal shadow and recorded the findings of the film in the section of “Field reading”. Primary field results were classified into three types: “Normal,” “Abnormal,” and “Not taken.” Participants with “abnormal” chest X-ray results were identified to produce sputum samples, as did elderly, pregnant and disabled participants who did not take chest X-ray but had symptoms suggestive of TB.

The team leader interpreted all films taken that day the same night for quality assurance. If there was any discrepancy, a sputum collection was requested for suspected participants before the team left.

Sputum collection and field report

The ear ly morning sputum and spot sputum from eligible survey participants through the network of household doctors were collected and packaged for transportation to NTRL.

Central survey operation

Under the leadership of the steering committee for the National TB Prevalence Survey, the central survey operation

Sample collection at field


composed of laboratory work, final interpretation and organization of the technical advisory groups, and data management including categorization, input and analysis of data, and quality assurance.

Laboratory

All laboratory tests of Mycobacterium tuberculosis were performed at National tuberculosis Reference Laboratory with biosafety equipment and experienced laboratory staff in the Central Tuberculosis Preventive Institute. All laboratory materials infected by M. tuberculosis were incinerated or autoclaved for reuse.

Sputum smear microscopy

All sputum specimens transported by the field survey teams were put through a decontamination processed by using N-acetyl-L-cysteine sodium hydroxide (NALC-NaOH). Smear samples were prepared according to the ID number and an AFB microscopy was performed after fluorescent staining.

The results of sputum smear microscopy are recorded as follows:

Culture Examination

After processing sputum samples and preparing smears, deposits were inoculated with 3–4 drops (approximately 0.1–0.15ml) on slopes of egg-based medium labelled with the ID number. Contaminated cultures were removed and rapidly growing mycobacteria were identified using SD Bioline testing.

Contaminated cultures were recorded when detected.

Identification Test

The identification of MTB was performed by SD Bioline TB Ag MPT64 rapid test with cultures.

(Note: SD TB Ag MPT64 rapid test have a 105 CFU/ml detection limit, so TB confirmation can be available with only 3 colonies from an early stage of culture (around 3–4 weeks)).

Culture at NRL, CTPI


Re-checking of slides

When a specimen turned out to be smear-negative but culture-positive, the slide derived from the specimen was rechecked with FM in order to avoid the false-negative smear result and verify the original smear result.

Final interpretation of X-ray films

All the chest X-ray films taken in the field were sent for central reading to CTPI after each field operation.

Quality assurance of laboratory tests

The protocol was designed in line with WHO guideline and reviewed by WHO Task Force Members on several occasions.

Regular monitoring and mid-term monitoring were performed to avoid inter-survey team bias, including inter-lab (among lab facilities).

For internal quality assurance of smear microscopy, 5~10% of positive slides were routinely rechecked. To prevent false positive smear results, rechecking all smear positive slides was performed by the referee of National Tuberculosis Reference Laboratory (NRL). To prevent false negative smear results, all smear negative slides with culture positive were rechecked by the referee of NRL.


2 Survey results

2.1 Survey participation by sex and age

Among the eligible population, survey participants were 60 683 and actual survey participant rate was 84.4%, almost reaching the target of 85%.

By sex, male participants were 27 034 (83.9%) and female participants were 33 649 (84.9%). Participation rate in females was 1% higher than the one in female. Out of the 33 649 female participants 123 (0.35%) females were pregnant.

Individuals enumerated in Census 90 466

Eligible study population71 877(79.45%)

Participants 60 683(84.4%) -Interview only 1 034 (1.7%)

-Interview and CXR 59 649 (98.3

Eligible for Sputum Examination 4 802 (7.9%)

Submitted at least one specimen 4 586 (95.5%)

Submitted two specimens 4 462 (92.5%)

Lab Results available 4 586 (95.5% of eligible)

(2 Smear 2 Culture results: 4 037, 84.1%)

Culture result MTB + 158 MOTT+ 9 Contaminated 1 Negative 44

Ineligible individuals: 18 589(20.55%) Excluding Children& Adult due to residential

criteria

Symptom + CXR + : 1 028(21.4%) Symptom + CXR - : 1 916(39.9%) Symptom - CXR - : 1 858(38.7%) Missed cases: 216 (4.5%)

Smear + Culture + 158 Smear + Culture -: 29 Smear + Culture + (NTM):9 Smear + Culture ND/contained: 1 Smear - Culture +:153 Smear not done, Culture +: (NTM):1 Positive Bacteriological MTB result: 340

Culture contamination rate: Spot: 220/4 504=4.9% Morning: 208/4 544=4.6%

S+ C+ 158 S+ C- 29 MOTT 9 Excluded cases 16

S- C+ 153 MOTT 1 Excluded cases 3964

Any Smear Positive 212 Morning S+ only: 39 Spot S+ only: 14 Both S+: 159

Any Smear Negative: 4 374 Morning S- only: 56 Spot S- only: 121 Morning S- & Spot S-: 4 250

Culture result MTB + 153 MOTT+ 1 Contaminated 1

Negative 44

Central Panel


Figure 2: Sex and age of eligible population VS survey participation

65+

55-64

45-54

35-44

25-34

15-24

Male Female

8000 80006000 60004000 40002000Number

2000

Age g

roup

Eligible Participants

2.2 Survey participation by Urban, Rural and Regions

Survey participation rate in urban and rural areas were 83% and 84.95% respectively, which was higher in rural area than in urban area. Participation rate in construction units was 98.3%.

2.3 Occupation distribution of survey participants

The occupational distribution among participants were 30.2% of workers, 21.8% of farmers, 10.4% of officers, 8.1% of students, 0.03% of sports players and 29.6% of the dependent and retired

2.4 Smoking status among the survey participants

18.3% of survey participants (11 106 persons) were identified as smokers, out of which 11 098 was male smokers and 8 was female smokers. Smoking rate among male participants was 41.1%, while female smoking rate was 0.02% close to 0%. 0.11% of participants (69 persons) stopped smoking. Male smoking rate increased with age.


2.5 Contact status with known TB cases

100% of survey participants answered to question of any TB cases in their family and 1485 participants (2.8%) responded having TB case in their family. There was no significant difference between male and female responses.

Table1.1: Proportion of survey participants who have TB patients in family within the past two years (%)

SexNo TB case in family TB case in family

N % N %

Male 26332 97.4 702 2.6

Female 32677 97.1 972 2.9

Total 59009 97.2 1674 2.8

99.8% of participants answered the question relating to any contact history with Pul-TB cases within the past two years. Among those responders, 3089 persons (5.1%) responded they had a history of contact with Pul-TB cases. It was noted that there were more contacts in males (5.9%) than in females (4.4%). There were non-responders of 82 females and 45 males.

2.6 Screening

2.6.1 Interview

85.1% of survey participants did not have any symptoms and only 14.9% of survey participants had at least one symptom.

Among the participants who had the positive symptoms, the males was 18.7% and the females was 11.9%.

Among the positive symptoms, 10.1% was cough, 7.5% was breathless, 6.7% was sputum, 2.3% was ineffectualness, 2.2% was fever, 2% was night sweating with other symptoms being less than 2%. Males had 1.3 or 2.3 times more symptoms than females.

2.6.2 TB presumptive cases by interview and CXR

In survey protocol, those who have more than 15 days of cough/hemoptysis/blood in sputum/abnormal chest X-ray were defined as TB presumptive cases.


Through the interview and Chest X-ray, 7.9% of the total survey participants (4802 persons) were identified as sputum collection candidates. Among them, 3.2% were identified by interview only, 3.1% by CXR only and 1.7% by both of interview and CXR.

Differences in the sex variable were identified, where 10.8% of male survey participants and 5.6% of female survey participants were screened positive. Male presumptive cases were 2 times higher than the female presumptive cases.

The TB presumptive cases increased with age except for the over 55’s with lower trend. No significant difference was identified by the stratums. There were more TB presumptive cases in rural areas than in urban areas. By provinces, proportion of TB presumptive cases among the survey participants ranged from 4.8% to 11%.

2.7 Laboratory examination

2.7.1 Smear result

The spot sputum was collected from 4544 survey participants (94.6% of TB presumptive cases) and morning sputum was collected from 4504 survey participants (93.8% of TB presumptive cases). Among spot sputum providers, 164 persons had positive smear results with 34.1% of scanty, which was 3.6% of survey participant who produced spot sputum.

Among morning sputum providers 189 persons had positive smear results with 36.5% of scanty, which equaled to 4.2% of morning sputum providers.

14 persons had only positive spot smear results and 39 persons had only positive morning smear result. 150 persons had positive smear result both by spot and morning sputum. Thus, a total number of smear positive cases was 203 persons, out of which 73.9% were positive both by spot and morning sputum and only 26.1% positive either by morning or spot sputum

216 persons did not submit either spot or morning sputum (4.5% of TB presumptive cases).

2.7.2 Culture result

Culture test was done for all collected sputum samples. 5.7% of spot sputum were culture- positive and 6.5% of morning sputum were culture-positive. 0.2% of samples were NTM and contamination rate was 4.6% and 4.95% respectively.


16 persons had culture positive results by spot sputum only, 50 persons culture-positive by morning sputum only and 245 persons were culture-positive both by morning and spot sputum. The total number of culture positive cases was 311 persons.

Among the culture-positive results, 78.8% was positive both by morning and spot sputum and 21.2% was positive either by morning or spot specimen.

Among culture-positive results of spot sputum, 34.5% was scanty and among the culture-positive results of morning sputum 41.4% was scanty.

2.7.3 Comparison of Smear and Culture result

Among 164 spot smear positive cases, 36 persons (21%) had culture negative results and one sample had been contaminated. Other 127 persons were spot smear positive & culture positive. This meant that spot smear and culture positive results were 42.64% of the spot smear positive results.

Among 4371 spot smear negative cases, 134 persons (3.1%) were culture positive cases.

When combining spot smear and culture results, a total of 298 persons were either spot smear-positive or culture positive.

Out of 208 contaminations, 207 were smear negative and only 1 was smear-positive.

Among 189 morning smear positive cases, 37 persons (21%) had culture negative results and 147 persons were morning smear positive& culture positive. This meant that morning smear and culture positive results were 44.3% of the morning smear positive results.

Among 4306 morning smear negative cases, 148 persons (3.4%) were culture positive cases.

When combining morning smear and culture results, a total of 322 persons were either morning smear positive or culture positive.

Out of 220 contaminations, 5 were smear-positive.


2.8 TB cases

2.8.1 Definite and Probable TB cases

Out of 203 smear positive cases, 16 persons were identified as culture negative cases in the follow-up, thus 16 participants were excluded from survey TB cases by Central expert committee.

The central expert committee confirmed 187 smear positive TB cases and 153 smear negative TB cases, therefore a total of bacteriologically confirmed TB cases were 340.

Among the smear positive TB cases, 84.5% were culture positive and 15.5% were culture negative TB cases.

Among the smear negative TB cases, 60.8% were culture positive and 39.2% were culture scanty TB cases.

There were 760 persons who were eligible for smear examination, but didn’t produce sputum samples or contaminated during lab examination. Therefore missing value imputation was done for 760 cases for estimation of TB prevalence.

2.8.2 Survey TB case by screening result

Among the survey TB cases, 55% were confirmed by both interview and CXR. 42.9% of survey TB cases were screened positive only by CXR. 2.1% of survey TB cases were confirmed only by interview.

Among the smear positive TB cases, 62.6% were confirmed by both interview and CXR. 34.2% of survey TB cases were screened positive only by CXR. 3.2% of survey TB cases were confirmed only by interview.

This shows that CXR is very important tool for detection of TB cases.


Table 2: Survey TB cases by interview and CXR results

ClarificationSmear positive

Bacteriologically positive

N % N %

Symptom positive, CXR Normal 1 0.5 2 0.6

Symptom positive, CXR abnormal 117 62.6 187 55.0

Symptom negative, CXR Abnormal 64 34.2 146 42.9

Symptom positive, CXR not done 5 2.7 5 1.5

Total 187 100.0 340 100.0

2.8.3 Distribution of survey TB case by characteristics

Distribution of survey TB cases by sex was 69.7% of males and 30.3% of females. Survey TB cases of males was two times higher than that of females. After 55 years of age band, the number of TB cases decreased. There were more TB cases in urban than in rural areas.

Table 3: TB case distribution by sex, age, urban and rural

Clarification Number of TB case

Total 340

Male 237

Female 103

Age

15–24 15

25–34 58

35–44 95

45–54 103

55–64 41

≥65 28

Urban 194

Rural 143

Construction unit 3


2.8.4 Final CXR reading result and survey TB case

Field CXR reading results showed that 4.8% had abnormal results.

Comparing the field and final CXR results, 72.3% of abnormal CXR results at field was assessed as normal at final reading and 13.3% of active pulmonary TB, 2.8% of suspected TB, 7.8% of scars after TB lesions were related to TB by final CXR reading.

When analyzing survey TB cases by final CXR result, 75.6% had active pulmonary TB lesion, 14.4% was suspected having TB and 5.9% had scars after TB at final CXR reading. Only 0.6% of those having normal CXR result at final reading was identified as survey TB cases.

2.8.5 TB prevalence

Based on the survey TB cases and the number of survey participants, the survey team calculated the prevalence of bacteriologically confirmed pulmonary TB among 15 years and above population as 567 per 100 000 population (510–631). As per WHO recommendation, the team used method 3 with the survey data to gain an estimation of TB prevalence. First the survey team conducted multiple imputations for missing cases and applied inverse probability weighting to the eligible non participants and no sputum collection candidates for TB prevalence estimation. The result was 587 per 100 000 population. Prevalence of males is 2.9 times higher than that of females and in 45–54 age group it was highest as 877 per 100 000 population. TB prevalence in rural area is 1.14 times higher than in urban areas.

After applying the sex proportion weighting to the eligible population in the country, the result was 597 per 100 000 populations.

𝑃𝑎: National Pulmonary TB prevalence rate

Pa = 597.3 (528–666) per 100 000 population

(Annual report of DPR Korea Central Bureau of Statistic: for the weighting of male and female variables, we used the national data of CBS (male population is 46.54% except military). Survey census data was 44.8% for male population).

𝑃= (1−𝑐+𝑐𝑟)/(1−𝑒) 𝑃𝑎

𝑃: all age, all forms of national TB Prevalence rate, 641 per 100 000 populations


𝑐: proportion of less than 15 years of age population; 20.7%

𝑟: ratio of pediatric TB prevalence to adult TB prevalence; 26.6%

𝑒: Proportion of extra-pulmonary cases; 21%

The team estimated the national TB prevalence rate at 641 per 100 000 populations.

Table 4: Prevalence of bacteriologically confirmed pul-TB cases per age, sex, urban and rural area among 15 years and above population

Clarification

Cluster level analysis, Model-1

Individual level analysis, Model-2

Individual level analysis, Model-3

Prevalence (100 000

population) (95%CI)

Prevalence (100 000

population) (95% CI)

Prevalence (100 000

population) (95% CI)

Total 567(510–631) 582(555–609) 587(520–655)

Male 892(778–1005) 908(857–958) 917(783–1052)

Female 309(249–369) 320(294–348) 319(256–382)

Age

15–24 146(72–220) 161(127–196) 155(70–240)

25–34 525(390–660) 534(474–595) 579(410–748)

35–44 732(585–879) 753(687–819) 764(611–916)

45–54 825(665–984) 839(768–910) 877(705–1049)

55–64 591(410–771) 611(530–693) 595(410–781)

≥65 450(283–617) 448(374–522) 444 (264–624)

Urban 571(490–651) 585(549–621) 577(489–665)

Rural 627(524–730) 643(597–689) 659 (555–764)

Construction units

96(0–205) 108(57–159) 102 (0–219)


2.8.6 Treatment status of survey TB cases

Among the 340 bacteriologically confirmed pulmonary TB cases, 106 cases (31.2%) have been diagnosed as TB and enrolled on treatment. 82.1% of them were currently having TB related symptoms. Out of 87 cases with symptoms enrolled on treatment before the survey, 59.8% was smear-positive and 40.2% was smear-negative.

Health centers which provided medical service to TB or pleurisy patients in the past were city/district/county hospitals, taking a 66.1% proportion, followed by clinics (12.1%).

12.2% of TB patients were received treatment in every TB facility and level and 1% had administered self-care.

2.8.7 TB risk factors

1.27% of Among the TB patients under this TB prevalence survey, smokers were identified as survey TB cases, which1.27% compared with that of non-smokers (0.43%). The TB prevalence in smokers was 3 times higher than the proportion of survey TB cases among in non-smokers (0.43%).

This result shows that smoking is one of key main risk factors for TB and explains why one of reasons for the high TB prevalence in males is much higher than . 41% of men in the one in females in the country prevalence survey reported smoking.

Among the survey participants who had contact with TB case in family, 3.76% were confirmed as survey of persons had TB cases, which was patients in their family within the past two years, a rate 7.7 times higher than those who did not have a TB case inpatient within their family.

2.91% of survey participants who had infected persons were in contact with active TB case were diagnosed as survey TB case with cases, a rate of that was 6.3 times higher than those who did not have any contact with TB case in past two years.

From this point of view active case finding in families having TB case should be strengthened along with chemo-prophylaxis to prevent new cases emerging and contacts with TB cases.


Discussion of central expert committee

3 Success, challenges and lessons learnt from survey design to finalization of survey result

The achieved successes of the National TB Survey is due to the National TB Control Programmes contribution of USD 481, 963 to the survey and the comparative use of less of external funds (USD 896 026 from Global Fund) than other Asian countries for a successful implementation of successfully completed National TB Prevalence Survey. A high participation rate in the survey was achieved. Though the survey used paper based data collection, conventional X-ray machines and L-J solid culture for TB laboratory examination, it was conducted very scientifically with the correct final data analysis for the national TB prevalence and incidence estimation given under the close cooperation with WHO HQ.

A major challenge for the National TB Prevalence Survey was the delay in the procurement of mobile conventional X-ray machines and the interrupted cash flow in the country. Thus the main survey was postponed by one year from the planned year of 2014start date. The final report preparation has taken a long time due to of lack of human resources.

The lessons learnt from national TB Prevalence Survey are as follows:

The planning phase is very important for the successful implementation of the survey, especially when developing the procurement plan for the capacity building of the human resources and getting the necessary laboratory and X-ray equipment and capacity building for such a survey design and data analysis.

Technical assistance for countries which do not have any past survey experience is essential necessary during the survey preparation, field and operation and data analysis period. In country, the UNICEF country office


was responsible for processed the procurement need for the survey and WHO country office provided the technical support for protocol development, field operation, final data analysis and report preparation. Through the WHO support, technical consultant who was involved in supported the Indonesian National TB Prevalence Survey in 2016 visited DPR Korea in June 2016, reviewed survey implementation our country and provided technical support for data analysis. In the end of July 2016, two lab experts survey implementation. Data management and technical assistance from Hong Kong SNRL reviewed the laboratory progress and validated survey lab laboratory test results.

In 2017, WHO HQ experts reviewed the whole survey database two times and provided the technical support for a joint survey data review and final analysis.

Through open discussion and transparent collaboration with all stakeholders for the 2015–2016 National TB Prevalence Survey, we have adopted the new TB prevalence estimates rate estimation and as a result, new baseline data has been prepared for the National Strategic Plan for TB control 2018–2021 and the next Global Fund TB grant.

National prevalenceDemocratic People’s Republic of Korea

2017

TB

National TB prevalance survey - DPRK - searo.who.int · aimed at understanding the real burden of the TB disease within the country and finding ways to improve TB detection and control

Documents