1. INTRODUCTION

Indonesia is one of the countries with the highest tuberculosis (TB) burden in the world. TB prevalence was estimated at 660 per 100,000 population, which means that there would be more than 1,600,000 people with TB in Indonesia [1]. According to TB global report 2017, the TB incidence rate in Indonesia was 395 per 100,000 population and the mortality rate was 40 per 100,000 per year. The target of TB program in Indonesia is to achieve TB incidence rate of <10 per 100,000 population and mortality rate of <2 per 100,000 population by 2030 [2]. The main strategy of TB program is to prevent its transmission by optimizing case finding and treating people with TB. However, to date, the case finding rate is very low. The national Case Notification Rate (CNR) in 2016 was 128 per 100,000 population, a slight increase from that in 2015 (125 per 100,000 population). The Case Detection Rate (CDR) was very low at 32% in 2015 and 33% in 2016. In Badung District, the TB case finding was lower compared with that of national level. The Badung CNR was 60 per 100,000 population in 2015 and 65 per 100,000 population in 2016. Meanwhile, the CDR was extremely low at 15% in 2015 and 19% in 2016, which were still far lower compared with the national CDR target of 70% [3,4].

The household contacts of TB patients (index case) are at the highest risk for TB infection. Around a quarter (25%) of household contacts have been infected at the time the index case was diagnosed [5]. The TB active case finding program among at-risk population has already been administered based on The Regulation of The Minister of Health Indonesia Number 67 in 2016 as a strategy to improve case finding [6,7]. Active case finding among TB contacts is a systematic, efficient, and effective strategy to look for new TB cases [8], including a home visit in contact investigation. A home visit aims to actively identify TB contacts and to interview them for TB symptom screening [7]. However, screening TB using symptoms, particularly chronic cough to identify presumptive TB cases, may cause missed opportunity. As per the Indonesia TB prevalence survey in 2014, 42.5% TB cases were found without showing common symptoms (i.e., cough for 2 weeks or longer) [1]. Alternately, chest X-ray examination and other sensitive methods for TB screening are needed to reduce missed opportunities [9]. The chest X-ray examination may be able to find any abnormalities in lung parenchymal that is suggestive of TB as an early detection of presumptive TB cases, especially before symptoms occur.

The Ministry of Health of the Republic of Indonesia has initiated family approach for Healthy Indonesia Program where TB is part of its priority. The main objectives of the family approach in TB program are to increase the coverage of TB services among family members and to involve family members in TB case finding. The family of TB patients should provide support for TB treatments and participate in active TB case finding. All TB household contacts should participate in the TB screening and evaluation at public health centers (PHCs) [10].

The Indonesian National TB Program has been implementing contact investigation, however, its implementation was lacking evaluation. There were no indicators available for TB recording and reporting system to evaluate the performance of contact investigation [7]. This study aims to assess the implementation of TB early detection in contact investigation, its result, and contribution in improving TB case finding.

2. MATERIALS AND METHODS

3. RESULTS

We found a total of 124 index cases. The study team had successfully home visited all of them. On the basis of interviews of index cases, we identified 498 contacts, thus the ratio contacts to cases was 4:1 Figure 1. Among the 498 contacts, 122 (24.5%) were children (under 15 years), 254 (51.0%) were female, and 328 (65.9%) were the nuclear family of index cases. All contacts were interviewed regarding the TB symptoms. There were 114 (22.9%) of contacts who had one or more TB symptoms with the most frequent symptoms being cough (15.9%), breathing difficulty (6.8%), and fever (6.4%) Table 1.

Figure 1

Flow diagram describing the implementation and result of contact investigation in Badung District, Bali Province, Indonesia. The dotted boxes indicate potential missed opportunities for tuberculosis case finding.

On the basis of the contacts’ follow-up results we found 100 (20.1%) out of 498 who attended at least one TB examination session at a health facility (PHC or hospital) Figure 1. Twenty-two of them were children and 78 were adults. Anamnesis and physical examination were completed for all contacts that visited PHC. The most common symptoms found were cough, fever, and breathing difficulty. Twenty-one adults were requested to provide sputum smear for examination and three (14.3%) of them were found positive. There were 13 children undergoing TST, five (38.5%) of them were found positive. Seventeen children and 70 adults were undergoing chest X-ray examination in Badung local hospital, 14 (82.4%) children and 55 (78.6%) adults were found have lung parenchymal abnormalities Table 2. Out of 100 contacts that attended at least one examination session, there were 41 contacts that completed participation until diagnosis was confirmed. The participation was higher among children compared with adults (46.2% vs 14.3%), among nuclear family compared with extended family (9.0% vs 1.9%), among contacts with education level ≤9th grade compared with those with higher education (13.1% vs 6.5%), and among contacts who had at least one or more TB symptoms identified during home visit compared with those without symptoms (14.0% vs 6.5%) Table 3.

Those TB evaluations had successfully found 10 additional new TB cases consisted of six childhood TB, three bacteriologically confirmed TB, and one clinically confirmed TB. There was a child who had TST result negative but diagnosed as TB by the physician at PHC because the chest X-ray result showed bilateral bronchopneumonia and perihilar lymph, and the score was 6. The positivity rate was higher among children compared with that in adults (46.2% vs 14.3%), among males compared with females (30.0% vs 19.1%), among contacts with education level ≤9th grade compared with higher education (47.1% vs 8.3%), among contacts who had at least one or more TB symptoms identified during home visit compared with those without symptoms (43.8% vs 12.0%) Figure 1 and Table 3. The overall contribution of early detection in contact investigation to improve TB case finding was 8.1% or an increase from 124 to 134 cases Table 4.

4. DISCUSSION

This study found that the ratio of contacts to index cases was 4:1, similar to other studies that found the ratio of 3–5:1 [12–14]. This result indicated that most of the identified contacts were a nuclear family to index cases. The contact identification should be optimized because most of the people in Bali including Badung District are living with extended family. The ratio of contacts to index cases was relatively low compared with the target stated on The Regulation of The Minister of Health Indonesia No. 67 in 2016, where contact investigation should identify 10–15 contacts from one index case [6]. Based on review toward the TB register 01 book, technically the number of rows to put contact tracing results was limited. Only five rows were provided and it led the officer to provide only a maximum five identified contacts, even though there were more contacts that could be found. The trouble also came from index cases. They had experienced fear of infection, loneliness, and isolation from society, which delayed them to seek healthcare [15]. The stigma experienced by index cases prevented them from declaring their close and household contacts during the investigation [16]. The TB officers at PHCs should be trained to improve their capacity in delivering TB education, building trust, and interviewing people with TB to trace more contacts.

The proportion of TB contacts who had cough in this study was similar to another study (15%); however, the proportions for other symptoms were found to be fewer [17]. The proportion of contacts with at least one TB symptom in this study was 22.9%, a lot fewer compared with that in another study (45%) [17]. This result shows that the awareness about the TB symptoms among contacts, except for cough, was still low. The TB officer when conducting home visit should perform with a good skill on how to ask, explain, and describe (giving examples) TB symptoms clearly. The questions should also be clear toward duration and other characteristics of the symptoms. For example, fever in TB is low grade or <38.5°C [5,18]. Further education about TB symptoms will help contacts to be aware, to remember, and to reveal the symptoms that occurred.

The implementation of early detection in contact investigation had two points of missed opportunity, causing a low participation (8.2%). First was the participation of contacts attending PHC and second was the participation of contacts to complete TB evaluation until diagnosis is confirmed Figure 1. An earlier study regarding barriers to adherence with TB contact investigation found that knowledge, attitudes, and practices were associated with a contact’s participation [19]. Another study also found that the barriers among index cases and contacts in contact investigation were their lack of TB knowledge, perception to related TB stigma, and fear of TB diagnosis [16]. Both studies showed that the greatest challenge of the TB early detection participation was health behavior [20]. This finding implicates the importance of health education for contacts, which can be delivered while conducting a home visit. The factors associated with health-seeking behavior on TB include knowledge, perception regarding TB, TB-related stigma, fear of TB diagnosis, and their previous experience visiting health facilities. Therefore, the health education material and method should comprehensively address the cognitive and psychological aspects of contacts [21,22].

The implementation of early detection in contact investigation yielded 10 additional new TB cases among 41 contacts that completed participation. It means one additional TB case was found for every four contacts evaluated. Other study found that for one index case, the number required to screen contacts was at least 11 household members/contacts [23]. The prevalence of TB among contacts in this study (24.4%) was higher compared with that from a previous study in the Philippines (12.8%) [24]. Higher positivity among contacts with at least one TB symptom indicates that they should be made as a priority for screening. In addition, a result of 12% TB positive among contacts without symptoms was similar to positivity rate among presumptive TB in the general population (5–15%) [7]. This indicates that TB early detection using chest X-ray was feasible to be implemented. Moreover, this result shows that the early detection among contacts was an efficient and effective strategy for TB case finding. The strategy should be continued and initiated in other districts and provinces in Indonesia [17,25,26].

It has potential to improve the contribution of early detection in TB case finding (8.1%), especially if the contacts’ participation level can be optimized. A previous study found that the contribution of systematic contact investigation with a home visit to improve case finding was 63%. The high contribution in this study was due to the high level of contacts’ participation in TB evaluation (66%) [23]. Furthermore, we also estimated the expected contribution by calculating the expected contacts’ participation multiplied by the positivity rate based on symptoms and age groups. We expected that if the complete participation among child and adult contacts with TB symptoms can achieve 100% and adults without TB symptoms is 50%, then the expected additional new TB cases found will be 91 people in quarter of a year or 73.4% expected contribution to improve case finding Table 4.

The contribution of early detection in TB case finding was high among child contacts. In the third quarter of 2017, the program found two index cases of children, meanwhile the early detection program among child contacts found six children to be TB positive. It means that the strategy has improved childhood TB case finding about three times in the same period. The finding of a total of eight children with TB was close to the 10% target of total TB patients found. Other studies also reported a high level of TB case finding among children during contact investigation [27,28]. The prevalence of childhood TB among child contact tracing was also high [29]. This finding has an important significance in increasing the TB program’s performance. As the childhood TB case finding is mostly low in Indonesia, therefore this strategy can be offered as a solution.

The indicators of contact investigation calculated from this study should be continued and integrated into the TB reporting system. It is important to maintain a good standard in implementing contact investigation, establish systematic evaluation, and determine the next strategy to improve the TB program’s performance. Those standard indicators will also lead the TB officers at PHCs and at district level to be more concerned about the performance of contact investigation. Another study proved that the use of standard indicators had improved the program’s performance [23].

This study has some limitations. The participation of contacts in screening and evaluation was very low. The contacts with complete participation might indicate a self-selection bias. They participated because of their condition and a close relationship to the index case. Contacts with any TB symptoms and from a nuclear family of index cases were more likely to complete participation and thus might have caused an overestimation of positivity rate in this study. The officers’ capacity and facilities in each PHC to support contact investigation, TB screening, and evaluation varied among PHCs. This might have ct have caused missed opportunity to find new TB cases.

5. CONCLUSION

The early detection in contact investigation is feasible to be implemented perpetually and can be adopted also by other districts/cities and provinces. This strategy showed to be effective and efficient in finding new TB cases, especially among children, and thus contributed to the increase in TB case finding. The main challenge was the poor participation of TB contacts to undergo TB screening and evaluation. A comprehensive education covering both cognitive and psychological aspects is needed to encourage contacts in participating in the TB screening until their diagnosis is confirmed.

CONFLICTS OF INTEREST

The authors declare that they have no potential conflicts of interest with respect to this research, authorship and/or publication of this manuscript.

AUTHOR CONTRIBUTIONS

I.W.G.A.E.P. designed the study, developed methodology, and data collection tools, conducted data analysis, and drafted the manuscript. N.M.D.K. contributed to the design of the study and the data collection tools, conducted the data collection, imputation, and drafted the manuscript. N.P.E.P.D. provided background information about the TB program in Badung District, coordinated the implementation of early detection in contact investigation, and data collection. I.K.S. contributed to the development of methodology and data collection, provided TB-related epidemiological information, and assisted in data collection. I.M.K.D. and I.K.H.M. contributed to data collection and imputation. P.R. critically reviewed the study protocol and supervised the research implementation. B.A. critically reviewed the study protocol, supervised the research implementation, and assisted in drafting the manuscript. A.P. critically reviewed all aspects of the study and assisted in drafting the manuscript. H.B.N. assisted in drafting the manuscript, and C.U.W. critically reviewed all aspects of the study and assisted in drafting the manuscript. All of the authors have approved the final manuscript.

FUNDING

This research was funded by a grant from Global Fund.

ACKNOWLEDGMENTS

The authors would like to thank Global Fund for facilitating a workshop on study protocol finalization and funding the research implementation. Furthermore, the authors are grateful to the National TB Program (NTP) Ministry of Health of the Republic of Indonesia for their support and facilitation. The authors would like to thank TB Experts Commission for their constructive inputs on study protocol and report finalization. The authors sincerely thank Badung District Health Office for their support and enthusiasm to adopt the concept of early detection in the implementation of the contact investigation. The authors would like to thank all respondents who kindly consented to participate in this study and to all surveyors who contributed in data collection.