Factors contribute to Tuberculin Skin Test (TST) Positive among Urban Islander, Malaysian Borneo

Introduction: Latency and the identification of risk groups have become a challenge and a priority to achieve the control and eradication of the TB disease. The objective of this study is to determine the prevalence and risk factors of tuberculin skin test (TST) positive in Pulau Gaya, Sabah, Malaysian Borneo. Methods: A cross-sectional study was conducted among 197 from Pulau Gaya residents by using non-probability convenient sampling method from March to May 2017. Questionnaire on socio-demographic and risk factors of TB was used to assess the risk factors for TST positive. The data was collected by trained health personnel. TST was administered and measured after 72 hours by a trained lab technician. The standardised cut-offs of ≥10 mm are defined as reactive positive TST. TST positive cases were referred to Luyang Health Clinic for further investigations and management. SPSS version 24 was used to analyse the associated factors. Results: The response rate was 100% (197 and the Prevalence of TST positive was 46.2%, which is considered high compare to other areas in Malaysia. The significant risk factors in bivariate analysis for TST positive were old age (p<0.005 ), increase BMI (p<0.033), marriage (p<0.001), employment (p<0.05), closed contact with TB patient (p<0.005), and Bacillus Calmette-Guerin (BCG) scars (p<0.005). In multivariate analysis, all the significant variables were entered, resulting in variables of employment (aOR 2.327 (95% CI:1.175-4.609)) and no BCG scar (aOR 2.473 (95% CI:1.175-4.446)). Conclusion: The risk factors for positive TST were unemployment and no BCG scar. These high-risk group individuals need to be screened and monitored to detect the latent TB or early symptoms of active PTB. A health facility needs to be built in Pulau Gaya to improve TB vaccination activity and management programme.

Latency and the identification of risk groups have become a challenge and a priority to achieve the control and eradication of the TB disease (3,4). Available evidences suggested that age and comorbidity (such as HIV, steroid user, or chronic kidney disease), are predictors for both LTBI and TB disease (5). However, studies are controversial on whether social factors such as smoking and alcohol abuse should be considered (6).
Data from Kota Kinabalu Health Office, the incident rate of TB in Pulau Gaya is two times higher than the Sabah incident rate in 2015. The incident rate for TB in Pulau Gaya for 2015 is near 234 per 100,000 population, and the incident rate for Sabah is 125.98 per 100,000 population (7). Thus TB is one of the vital health problems among the Pulau Gaya population. In 2016, non-Malaysians contributed 54 per cent of the active TB cases in Pulau Gaya. Previous studies have revealed a high prevalence of LTBI in high-risk groups such as health care workers, prisoners and gold miners (8). In the community, there also vary in the prevalence of LTBI (9). However, it is challenging to find a study on the prevalence of LTBI in the small island community, like Pulau Gaya. The proximity of the island to the city and the geographical factor is a factor of rapid movement in and out of immigrant in Pulau Gaya, especially a country with a high burden of TB like Philippine, which is adjacent to the Sabah. The population of Malaysians, as well as immigrant, contributes to the high incident rate in Gaya (10). Due to the mixture of population, the cultural and the socioeconomic in Pulau Gaya are slightly different from the nearest area.
An estimated one-third of the world population currently presents LTBI, and only between 5 and 10% will develop TB disease during their lifetime (11). Most of these patients develop active TB within the first five years unless they are diagnosed and treated with anti-TB medications (12). Identification LTBI will increase the rate of case detection of active TB. This situation may influence in developing new intervention policies to eliminate the spread of TB (13). Therefore, the present study was done to determine the prevalence and risk factors for positive TST of among residents of Pulau Gaya, Sabah.

METHODS
A community-based cross-sectional study was conducted at Pulau Gaya, Kota Kinabalu, Sabah, Malaysia from March to May 2017. A total of 197 subjects were recruited. Non-probability convenient sampling method was utilised to select the study participants. The participants who are i) aged 15 years and over, ii) staying in Pulau Gaya and iii) No previous history of TB disease iv) agreeable to give consent to participate in the study were selected. For the subject under 18 years old, the consent was taken from their parents/guardians. Subjects with cognitive or any others speech impairment and did not meet the inclusion criteria were excluded. The ethical clearance was obtained from University Malaysia Sabah. The permission of conducting this research was obtained by the Institutional Review Board University Malaysia Sabah of Faculty of Medicine and Health Sciences before data collection in Pulau Gaya community.
After getting the informed and written consent, a validated questionnaire was administered with guided in this study due to some of the participant have a lower educational level. The questionnaire used for the study consists of two parts, including socio-demographic data and individual risk factors towards TB. The information required from the participants included a socio-demographic profile, house conditions, smoking habit, alcohol consumption, TB contact, diabetes mellitus, and end-stage renal failure. BCG scar at the left arm and body mass index (BMI) was examined during the interview. Using the intradermal technique, TST was performed by a trained health personal by administering 0.1 ml of purified protein derivative. The result was obtained after 72 hours by calling the individual participant, and positive TST was measured based on an induration of ≥10 mm.
Data were analysed using the IBM statistical package for social science (SPSS) Statistics ver. 24.0 (IBM Co., Armonk, NY, USA). The categorical variables were used in frequencies and percentage while the continuous variable was using mean and median in the descriptive analysis of data. Simple logistic regression was performed for bivariate analysis for categorical data and Mann Whitney U test for continuous not parametric data. Multiple logistic regression analysis was used for multivariate analysis. Statistical significance was determined at p<0.05.

RESULTS
A total of 196 participants from Pulau Gaya were enrolled in the study. Out of these, 91 participants were positive with TST. Hence, the prevalence obtained was 46.2% with a response rate of 100%. Among them, 74 (37.6%) participants were male, and 123 (62.4%) were female. The descriptive data on sociodemographic and individual risk factors have been shown in Table 1. The sociodemographic characteristic of subjects in our study showed that the age of the participants ranged 15-88 years with a mean age of 30.25 and a standard deviation of 16.08 years. In terms of occupation, among 197 participants being interviewed, 45 (22.8%) subjects were working, followed by not working 152 subjects (77.2%), Of the total 197 respondents, only 75 respondents (38.1%) had close contact with who have ever been diagnosed with TB and 122 respondents (61.9%) no known history of close contact with TB patient. Out of a total of 196 interviewees, 120 (60.9%) participants are having BCG scar followed by 79(39.1%) with no BCG scar. Among sociodemographic variables, only marital status and employment were significantly associated with TST positive in bivariable analysis. Among individual risk factors, the presence of BCG scar and TB contact showed a statistical difference in bivariate analysis with a p-value of <0.05. The rest of the variables were not statistically significant. It is shown in Table 2. There is no significant difference in age and BMI category between people with TST positive and negative. However, for a continuous variable using the Mann Whitney U test, older age and increasing BMI were statistically significant (Table  3).

DISCUSSION
Research of TB study in Pulau Gaya is still in infancy. TB is a global disease worldwide, even in Sabah, Malaysia. It afflicts the poor and uneducated communities living in congested areas based on a previous study (14). Diagnosis of TB requires certain investigations, namely chest radiography and sputum acidfast bacilli. Also, TST acts as an adjunct test before going to a definitive investigation modality as described earlier. In this study, the prevalence rate for TST positive is 46.2% which is near half of the subjects. The figure almost the same with the overall rate of LTBI among healthcare worker in Malaysia, which is 46% (15) . However, for the community base study, the figure is considered as high compared to other countries (16).
This study did not find a significant difference between people with TST positive and with TST negative in gender but not in the age variable. In this study, older age has more risk of TST positive. In previous studies, TB and LTBI were predominant in a male and elderly population (17)(18)(19). Our result also was contradicted with the previously published study showing no significant difference in education, overcrowding and family income between participants with TST positive and TST negative. Other risk factors for TB are smoker, alcohol consumer, IVDU, ex-prisoner, history of closed contact with TB patient, Undernutrition, and some chronic disease (17). In this study, all variables mentioned above were not significant except for the nutritional status (BMI) and history of TB contact. However, for the BMI, the results contradicted with the previous study, which is high or increase BMI were significantly associated with TST positive in the bivariate analysis but not in multivariate analysis.
For the history of TB contact, this study result in the bivariate analysis also was contradicted from previous literature, which is there was no history of closed contact have a higher risk of getting TST positive compared with having a history of close contact with active PTB patient (15,16), However, in multivariate analysis, the results were not significantly associated. In a subanalysis of the TB contact variable, the majority of the subjects that do not have close contact with PTB patient were associated with the absence of BCG scar, which is significantly associated with TST positive.
The other significant findings in the bivariate analysis were marital status, no BCG scar and employment. For the marital status, marriage subjects showed more risk than single/divorced group. The subjects without BCG scar higher risk of getting TST positive. The results were consistent with previous literature (16). The employed residence also had a higher risk of developed TST positive as compared to unemployed participants in bivariate analysis.
In multivariate analysis among the positive risk factors in this study, two risk factors significantly associated with positive TST, which is employment and absent of BCG scar. The employed participant has risk compared to the unemployed due to they have more contact with the people and exposed to overcrowding environment. Health promotion and health education will improve their knowledge and attitude toward the preventive measure of PTB disease.
In this study, the presence of BCG scar was a protective factor from TST positive. However, only two-third of the participant in Pulau Gaya had not been vaccination of the BCG. This show previous BCG vaccination coverage in Pulau Gaya was low compare to Malaysia policy. Vaccination of BCG is one of the beneficial interventions for the TB control programme. A meta-analysis of BCG trial data also suggested that the protective efficacy of BCG may persist ten years or more after infant vaccination (20). A study showed that BCG vaccination during infancy and adolescence could induce the immunological memory to mycobacterial antigens that are persistent up to 14 years (21). Additionally, a study looking at long term BCG efficacy among American Indians and Alaska natives showed that protection could last up to 60 years (22) and in Brazilians vaccinated as neonates for up to 20 years (23).
The limitations of this study include that majority of previous studies compared TB and non-TB cases as well as latent TB infection and non-latent TB infection. However, in this study, we only managed to make a comparison between positive and negative TST. These happen due to the logistic restriction of the participant, limitation of time and budget constraint. Which is the majority of subjects were unable to go nearest health facility to do further investigation. So, we were unable to proceed with definitive investigations which are sputum acid-fast bacilli and chest radiography for most of the subjects due to Further follow up need to be done to the subject to conform the LTBI and TB status of the TST positive subject.
Another limitation was that we only manage to do a non-probability convenient sampling method to select the study participants. Random sampling was unable to be done due to the logistic challenging to the researcher and the subjects, Most of the Pulau Gaya residence and the subjects were working in the mainland of Kota Kinabalu. This study needs the Mantoux test reading in the 2 nd follow up. So, we need the subject that contactable and available for the second to follow up that was difficult to choose randomly.
Further research will have found new significant value and can be used in the control of TB cases in Pulau Gaya. It includes as following (1) further research regarding health-seeking behaviour to understand the problems in controlling TB, (2) further proper investigated of TST positive subject for TB confirmation and (3) large sample size with an easy flow for subject doing the further investigation (chest radiography). Treatment of LTBI is an essential component of control and elimination of TB. Knowing the risk factors for LTBI and improving the screening technique will help in the management of LTBI and active TB.

CONCLUSION
Unemployment and the absence of BCG scar had proven to be the main risk factors of TST positive cases in this study. Health education regarding transmission, risk factors and symptoms of TB need to be emphasis to the highrisk population, specially employed people and no BCG scar groups. Vaccination of BCG is one of the vital interventions for control the spread of TB. Coverage of BCG vaccination needs to be improved in Gaya Island to control the spread of TB. There also need Health facilities to build in the Pulau Gaya to improve TB detection, and vaccination activities need to increase in Pulau Gaya.