403 JOURNAL OF SCIENCE, Hue University, N 0 61, 2010 PREVALENCE OF HIV INFECTION AND RISK FACTORS OF TUBERCULIN INFECTION AMONG HOUSEHOLD CONTACTS IN AN HIV EPIDEMIC AREA: CHIANG RAI PROVINCE, THAILAND Pornnapa Suggaravetsiri Khon Kaen University, Thailand Juthatip Putthasorn Chiang Rai Prachanukraow Hospital, Chiang Rai Province, Thailand SUMMARY During the course of the HIV epidemic in Chiang Rai Province in Northern Thailand, incidence of tuberculosis also sharply increased. Up to the year 2002, there were no specific interventions for the prevention and control of TB infection among the household contacts of an infected person. The study aimed to identify prevalence of HIV infection and risk factors of tuberculosis infection among household contacts. Methodology: A cross-sectional analytic study was conducted among household contacts of TB cases between January 2001 and January 2002. We collected data using structured interviews, physical examinations, tuberculin skin test (TST), mumps skin test and pre-test and post-test counseling for HIV testing. Results: The prevalence of positive tuberculin skin test among 1,211 household contacts was 57.55% and the HIV positive rate among household contacts was 7.74% (72 out of 930 contacts who did agree to HIV testing). Based on the variables entered into the model, gender, age, HIV status, caregivers of TB patients and living in the same room as the infected patient were significantly related to TB infection. However, the presence of a BCG scar was not significantly associated with TB infection. Conclusion: TB screening every 6 months should be provided to household members and children who had contacts with HIV positive individuals and the introduction of anti TB prophylaxis should be provided to HIV infected persons. Keywords: Tuberculosis, Households, Risk factors, HIV, Thailand 1. Introduction Tuberculosis (TB) and Human immune deficiency virus (HIV) are currently the major leading causes of death worldwide among infectious diseases. It is estimated that TB kills more than two million people each year. This problem is further aggravated by the alarming spread of HIV and the emergence of drug resistance. According to WHO estimates there were 8.8 million new case of TB in 2003 (140/100,000) of which 3.9 million (62/100,000) were positives smear and 674,000 (11/100,000) were infected with 404 HIV. Annually 5% to 15% of HIV- positive individuals develop TB and at least one- third of the world population, that is about 1.9 billion people are infected with TB. Within Chiang Rai province in Northern Thailand, 1990-1998 data revealed that the proportions of TB cases attributable to HIV was 72% among males- and 66% among female patients. The TB notifications increased from 40/100,000 population in 1990 to 144/ 100,000 population in 2002, which amounted to a three- fold increase in TB. This was mostly attributable to HIV. Obviously the problem of TB/HIV is an issue of critical importance, and challenging for investigation and control. Several studies were conducted to explored risk factors among TB patients, household members, and health care workers. However, most of them investigated the risk factors relating to the development of TB rather than its source of infection. This study aimed to identify the risk factors of TB infection among household contacts in a HIV epidemic area, Chiang Rai province, Thailand. 2. Methodology A cross- sectional study was carried out between January 2001 and January 2002. Three hospitals located in the epidemic area of Chiang Rai Province, Chiangrai Prachanukroa, Phan and MaeJan, were the study sites. Tuberculosis index cases (the first suspected case in each household) were diagnosed by chest radiography (CXR) and tested for acid-fast bacilli (AFB) in sputum. TB index cases and the household of each TB index case were visited by a researcher assistant. Household contacts were defined as members of an extended family of the index case who lived in the same house as the TB index case. Specific information about TB infection from the household contacts was collected by their response to a list of typical clinical symptoms of TB. Physical examination, mumps skin test and a tuberculin skin test (TST) were performed at the hospital or at home by well trained nurses. Indurative diameters of mumps skin test and the TST were measured within 48 to 72 hours. An indurative diameter of ≥ 10 mm of TST was considered to be a positive test (TB infection). Double data entry by using the software of Epi-info Version 6.02 (Atlanta, USA) was done to validate the data. The data file then was transferred to STATA 7.0 (Texas, USA) for statistical analysis to calculate the proportion of infection among household contacts and examine the effects of different factors related to the TB infection. Multiple Logistic regressions were carried out by the backward elimination method. The study was approved by the Ethics Committee of the Public Health Ministry of Thailand. 3. Results From the 526 TB index cases, 221 (42.01%) were TB/HIV- positive and 305 (57.98%) were TB/HIV- negative. Among 1,300 household contacts were identified, 1,248 household contacts consented to participant in the study. Of these 1,240 (99.36% of 1,248) agreed for TST and 930 (74.52% of 1,248) consented for HIV testing. Twenty-nine persons were excluded because of an allergy to TST. The remaining 1,211 405 study participants were included in the analyses. The sex ratio between males and females was almost one to one. There were 690 (56.98% of 1,211) with tuberculosis infection. Of those 366 (30.22%) were children younger than 14 years with TB infection. There were 930 HIV testing results among household contacts, of which 72 (7.74% of 930) were HIV positive and 685 (56.56 %) had a BCG scar. Using a univariate analysis, the risk of being TST- positive was found to be higher among household contacts of HIV negative- than HIV positive index cases. The risk of household contacts of developing a positive TST response was associated with the TB index cases and elaborates demographic variables. A low educational status increased the risk of TB infection. Household members being involved in agricultural and trade activities had higher risks of acquiring TB compared to government officers. To be a student was a protective factor of having the infection. Married and divorced family members were more at risk of infection than those who were never married. Among the family members, spouses had a higher risk of having the infection than other family members of the index cases (Table 1). Table 1. Rate and odds ratio of tuberculin skin test (TST) positive and characteristics of household contacts Characteristic Number and Prevalence (%) of TST-positive Crude odds ratio 95% CI p- value Sex male 327/528 (61.93) 1.00 Female 363/683 (53.15) 0.51 0.40-0.65 <0.01 Age (years) <10 75/239 (31.38) 1.00 10-19 111/189 (58.73) 3.11 2.09-4.63 <0.01 20-29 107/146 (73.29) 6.00 3.80-9.48 <0.01 30-39 119/171 (69.59) 5.00 3.27-7.66 <0.01 40-49 103/148 (69.59) 5.01 3.21-7.80 <0.01 50-59 82/124 (66.13) 4.27 2.69-6.77 <0.01 ≥ 60 93/194 (47.94) 2.01 1.36-2.98 <0.01 Race / Ethnicity Thai 639/1126 (56.75) 1.00 Hill tribe 51/ 85 (60.00) 1.14 0.73-1.79 0.56 406 Education More than high school 126/196 (64.29) 1.00 No education 209/431 (48.49) 0.52 0.37-0.74 <0.01 Elementary 355/584 (60.79) 0.86 0.62-1.21 0.38 Occupational Government official 138/253 (54.55) 1.00 Agriculture & Trade 219/318 (68.87) 1.84 1.31-2.60 <0.01 Labour 175/251 (69.27) 1.92 1.33-2.77 <0.01 Student 158/389 (40.62) 0.57 0.41-0.78 <0.01 Marital status Single 244/508 (48.03) 1.00 Married 358/558 (64.16) 1.94 1.52-2.48 <0.01 Divorced 88/145 (60.69) 1.67 1.15-2.43 <0.01 Relationship to Index case Relative 179/366 (48.91) 1.00 Spouse 203/282 (71.99) 2.68 1.93-3.74 <0.01 Children 121/238 (50.84) 1.08 0.78-1.50 0.64 Parent 187/325 (57.54) 1.42 1.05-1.91 0.02 Alcohol drinking and smoking were associated with a risk of having infection than non- drinkers and non-smoker. The presence of a BCG scar was a protective factor. Variables that had a close contact with the index case, such as giving care to them, sleeping in the same bedroom and sleeping in the same bed as the index case, were associated with an increased risk of being infected with TB (Table 2). Table 2. Univariate analysis of risk factors associated with tuberculosis infectious among household contacts (1) Variable TST positive TST negative OR 95 % CI p-value Alcohol drinking No 577 478 1.00 407 Yes 113 43 2.18 1.50-3.16 <0.01 Cigarette smoking No 449 380 1.00 Yes 241 141 1.45 1.13-1.85 <0.01 BCG scar present Positive 411 274 1.00 Negative 279 247 0.75 0.60-0.95 <0.01 HIV status Positive 22 20 1.00 Negative 539 316 1.55 0.83-2.87 0.17 Care giver to the TB patient No 376 406 1.00 Yes 314 115 2.95 2.83-3.81 <0.01 Sleeping in the same bedroom as the TB patient No 379 362 1.00 Yes 311 159 1.87 1.47-2.37 <0.01 Sleeping in the same bed as the TB patient No 448 395 1.00 Yes 242 126 1.69 1.31-2.18 <0.01 Open the window Everyday 479 371 1.00 Seldom 75 62 0.94 0.65-1.35 0.73 Never 136 88 1.20 0.89-1.62 0.24 Note: (1) Total may not be equal to 1,211 due to missing data. OR = 1.00 is the reference group for each category A multivariate model was fitted to assess the potential confounding factors and other factors, age and sex, BCG scar present, HIV status of contact, care giver of the TB patient, and sleeping in the same bedroom as the TB patient. A potential confounder was only the age group; the multivariate analysis was done to test the magnitude of the 408 single factor to develop TB infection. If this was true then the R 2 value would be additional valuable information as it would show the magnitude of how much a given variable determined the variation of the dependent variable or the infection with TB among the household contacts. The factors significantly related to being TST positive were being male, older age, and being in close contact to the TB index cases. The presence of a BCG scar was not a risk factor for TB infections among household contacts (Table 3). 4. Discussion This study found that more than 50% of the household contacts of people with TB and HIV were infected with TB. This was lower than studies conducted in Sub- Saharan Africa but higher than studies conducted in other regions. The differences between this study and those undertaken in other areas could be due to the use of different methods for identification of cases and family relationships. It was not known whether poverty could lead to poor nutrition and low immunity of people when they led to varying degree of risk of having the TB infection. One unexpected result was that those household contacts to an index case who were HIV negative had a higher risk of having TB compared to household contacts with an index case who was HIV positive. The reason for this probably was that household contacts knew the HIV positive status of the index case and had prevention in close contact with the HIV infected person. Household contacts with HIV negative index cases might not take that precaution. The level of TST indurations of a person with previous BCG immunization ranged from 5.0 mm to 11.3 mm even without natural infection. Another possible explanation to the contradictory finding was that a high degree of illness in HIV-infected patients could result in weaken and perhaps less effective dissemination of M. tuberculosis the patients’ environment. In addition, HIV-infected persons probably had a compromised ability to react to tuberculin skin testing, because HIV infection is associated with an elevated risk of skin test allergy. Table 3. Risk factors associated with tuberculosis infection among household contacts of TB index cases (Multivariate analysis) Variable Adjust OR (2) 95 % CI p-value Sex Male 1.00 Female 0.45 0.33-0.61 <0.01 Age (years) <10 1.00 10-19 3.70 1.89-7.25 <0.01 409 20-29 6.45 3.21-12.96 <0.01 30-39 5.34 2.70-10.57 <0.01 40-49 5.55 2.79-11.05 <0.01 50-59 6.26 2.99-12.87 <0.01 ≥ 60 2.56 1.29-5.08 <0.01 BCG scar present Positive 1.00 Negative 0.74 0.53-1.04 0.08 HIV status Positive 1.00 Negative 3.15 1.57-6.33 <0.01 Care giver to the TB patient No 1.00 Yes 2.04 1.48-2.81 <0.01 Sleeping in the same bedroom as the TB patient No 1.00 Yes 1.99 1.44-2.74 <0.01 Note: (2) Variables entered in the Multiple Logistic Regression Analysis were sex, age, BCG scar present, HIV status, care giver to the TB patient, and Sleeping in the same bedroom as the TB patient similar to other studies, the risk of having positive TST was related being male and older age group. Males might be less careful and less aware of the danger in acquiring TB. The decreasing immune stimulation probably was related to older age and a longer term exposure to TB might contribute an increasing risk among older age groups. Furthermore, it was found that hill tribe people had a higher risk to be infected with M. tuberculosis then other Thai citizen. However the difference between both groups was not significance. Type I error could occur as these were only 85 hill tribe persons included in this study. Other factors such as malnutrition, low income, poor access to health care, crowding, and poor hygiene were not explored in this study and required further investigation. Another factor was immune status, which was found to be lower among hill tribe people as discussed by Wanda Walton and Patricia Simone. The presence of a BCG scar was not a risk factor for TB infection. This was similar to what was found in West Africa and might be explained by some protective mechanism of being exposed to the causative agent of TB. These also require 410 further investigation. 5. Conclusion The presence of a positive TST result was consistently higher over all ages in individuals with a BCG scar. BCG vaccination was reported to induce cross-reactivity with tuberculin PPD (Purified protein derivative), but the degree of tuberculin sensitivity is highly variable, depending on the vaccine strain used, the dosage, the method of administration, and other factors known to influence the reaction to TST. Currently, BCG is given immediately to all babies born in Thailand. The coverage is more than 95% nationwide. BCG scars consistently had no relationship with TB infection status. Therefore, the presence of a BCG scar should not be a factor in the decision-making to treat latent TB infection among children and HIV-infected persons, as BCG did not appear to prevent the activation of a latent TB infection induced by HIV-related immunodeficiency. Household contacts that had been in close contact with the index case included family members, coworker or friends, and customers as found in other studies. In this investigation occupational and marital status was an independent risk factor to TB infection and also family relation, such as being husband or wife, had an impact on the TB infections of the contacts. Table 4. A Multiple logistic regression analysis of risk factors associated with tuberculosis infection among household contacts of TB index cases Variable Adjust OR (2) 95 % CI p-value Sex Male 1.00 Female 0.45 0.33-0.61 <0.01 Age (years) <10 1.00 10-19 3.70 1.89-7.25 <0.01 20-29 6.45 3.21-12.96 <0.01 30-39 5.34 2.70-10.57 <0.01 40-49 5.55 2.79-11.05 <0.01 50-59 6.26 2.99-12.87 <0.01 ≥ 60 2.56 1.29-5.08 <0.01 BCG scar present 411 Positive 1.00 Negative 0.74 0.53-1.04 0.08 HIV status Positive 1.00 Negative 3.15 1.57-6.33 <0.01 Care giver to the TB patient No 1.00 Yes 2.04 1.48-2.81 <0.01 Sleeping in the same bedroom as the TB patient No 1.00 Yes 1.99 1.44-2.74 <0.01 Note: (3) Variables entered in the Multiple Logistic Regression Analysis were sex, age, BCG scar present, HIV status, Care giver to the TB patient, and Sleeping in the same bedroom as the TB patient REFERENCES 1. Narain JP, Lo YR. Epidemiology of HIV-TB in Asia. Indian J Med Res. 2004; 120(4): 277-89. 2. World Health Organization, UNAIDS. Preventive therapy against TB in people living with HIV.1999. 3. Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC. Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. JAMA. 1999; 282(7):677-86. 4. WHO. Global Tuberculosis Control: Surveillance, planning, financing. 2003. 5. WHO. Global Tuberculosis Control: Surveillance, planning, financing. 2005. 6. Annual Report of Chiang Rai Health Office. TB Control Programme. 2004. 7. Tansuphasawadikul S, Amornkul PN, Tanchanpong C. Clinical presentation of hospitalized adult patients with HIV infection and AIDS in Bangkok, Thailand. J AIDS. 1999; 21(4): 326-32. 8. Lui K.J. Statistical Estimation of Epidemiological Risk. Chichester: John Wiley&Sons Ltd; 2004. 412 9. Nunn P, Gicheha C, Hayes R. Cross-sectional survey of HIV infection among patients with tuberculosis in Nairobi, Kenya. Tubercle and lung dis. 1992; 73:45-51. 10. Braun MM, Badi N, Ryder RW, et al. A retrospective cohort study of the risk of tuberculosis among women of childbearing age with HIV infection in Zaire. Am Rev Respi .Dis 1991; 143:501-4. . University, Thailand Juthatip Putthasorn Chiang Rai Prachanukraow Hospital, Chiang Rai Province, Thailand SUMMARY During the course of the HIV epidemic in Chiang Rai Province in Northern Thailand,. University, N 0 61, 2010 PREVALENCE OF HIV INFECTION AND RISK FACTORS OF TUBERCULIN INFECTION AMONG HOUSEHOLD CONTACTS IN AN HIV EPIDEMIC AREA: CHIANG RAI PROVINCE, THAILAND Pornnapa Suggaravetsiri . contacts with HIV positive individuals and the introduction of anti TB prophylaxis should be provided to HIV infected persons. Keywords: Tuberculosis, Households, Risk factors, HIV, Thailand