|Year : 2013 | Volume
| Issue : 1 | Page : 38-43
Daily contact with a patient and poor housing affordability as determinants of pulmonary tuberculosis in urban Pakistan
Batool A Haider1, Saeed Akhtar2, Juanita Hatcher3
1 Department of Community Health Sciences, Faculty of Health Sciences, Aga Khan University, Stadium Road, Karachi 74800, Pakistan; Department of Epidemiology, Harvard School of Public Health, Harvard University, 677 Huntington Avenue, Boston, MA 02115, USA
2 Department of Community Health Sciences, Faculty of Health Sciences, Aga Khan University, Stadium Road, Karachi 74800, Pakistan; Department of Community Medicine and Behavioural Sciences, Faculty of Medicine, Kuwait University, Jabriya 90805, Kuwait
3 Department of Community Health Sciences, Faculty of Health Sciences, Aga Khan University, Stadium Road, Karachi 74800, Pakistan
|Date of Web Publication||27-Feb-2017|
Department of Community Medicine and Behavioural Sciences, Faculty of Medicine, Kuwait University, Jabriya 90805
Source of Support: None, Conflict of Interest: None
Objective: This study aimed to evaluate the factors associated with pulmonary tuberculosis (TB) among individuals aged 15 years or more in urban Karachi, Pakistan.
Design and setting: A case–control design was implemented in three major tertiary-care hospitals to select cases (n = 342) with active pulmonary TB (i.e. two sputum smears positive for Mycobacterium tuberculosis with clinical and radiographic evidence of current pulmonary TB and diagnosed between August 2002 and October 2003. Selected controls (n = 342) were surgery patients from the same hospitals at time of recruitment of the cases, without clinical and radiographic evidence of pulmonary TB.
Results: Multivariable logistic regression model showed that daily contact with a pulmonary TB patient (adjusted odds ratio [ORadj])= 5.07; 95% CI: 3.31, 7.78), and poor housing affordability (i.e. rented vs. owned) (ORadj= 1.59; 95% CI: 1.13, 2.26) were significantly associated with pulmonary TB status. The overall adjusted summary population attributable risk (%) for both the risk factors together was 38.7.
Conclusion: Reaching out to underprivileged TB patients for delivery of DOTS and focused education of patients and their contacts about M. tuberculosis transmission mode may substantially minimize pulmonary TB risk in this and similar settings.
Keywords: Case–control study, Tuberculosis, Risk factors, Population attributable risk, Pakistan
|How to cite this article:|
Haider BA, Akhtar S, Hatcher J. Daily contact with a patient and poor housing affordability as determinants of pulmonary tuberculosis in urban Pakistan. Int J Mycobacteriol 2013;2:38-43
|How to cite this URL:|
Haider BA, Akhtar S, Hatcher J. Daily contact with a patient and poor housing affordability as determinants of pulmonary tuberculosis in urban Pakistan. Int J Mycobacteriol [serial online] 2013 [cited 2019 Aug 19];2:38-43. Available from: http://www.ijmyco.org/text.asp?2013/2/1/38/201045
| Background|| |
Globally, tuberculosis (TB) remains the second leading cause of death from an infectious disease with 8.8 million (range, 8.5–9.2 million) incident TB cases, 1.1 million (range, 0.9–1.2 million) deaths from TB among HIV-negative people and an additional 0.35 million (range, 0.32–0.39 million) deaths from human immunodeficiency virus (HIV) - associated TB in 2010 ,,. TB affects mostly adults in the economically productive age groups; around two-thirds of cases are estimated to occur among people aged 15–59 years and also more common among men than women .
In 2010, based on an annual incidence (231 cases per 100,000 population) and prevalence (364 cases per 100,000 population) of all types of TB, Pakistan ranked 6th among the 22 countries with the highest burden of TB in the world. Furthermore, Pakistan contributes about 44% of TB burden in the Eastern Mediterranean Region . Despite the availability of highly efficacious treatment for decades and the adoption of DOTS by the NTP (Pakistan's National Tuberculosis Control Program) in 1995, these latest statistics showed an alarming increase over previously reported burden of TB in Pakistan .
Risk factors for TB identified in other parts of the world include intensive immigration from high prevalence to low prevalence countries, contact with a TB patient, poor socio-demographic factors, tobacco smoking, alcohol abuse, HIV infection and institutionalization ,,,,,,,. The role of these and other locally prevalent risk factors threatening TB control in Pakistan are yet to be fully established. Therefore, this case–control study aimed to evaluate the factors associated with pulmonary TB among individuals aged 15 years or more in urban Pakistan.
| Methods|| |
Study design and setting
This case–control study was conducted at three tertiary care hospitals in Karachi – the largest city and economic hub of Pakistan with a metropolitan area of 3530km2 and 17 million ethnically diverse populations. The publicly funded health care system lacks adequate financial and human resources, facilities and equipment, appropriate organizational structure with poor quality of services . Pakistan's private sector makes a major contribution to providing health care, particularly in major cities. For this study, Civil Hospital (CH), Jinnah Postgraduate Medical Center (JPMC) and Liaquat National Hospital (LNH) were selected. Of these hospitals, CH and JPMC are public whereas LNH is a semi-private hospital. These hospitals were selected because they receive patients of almost all socioeconomic strata of Karachi.
Definitions and selection of cases and controls
A case of pulmonary TB was defined as an individual aged 15 years or more with active pulmonary TB, i.e. two positive sputum smears for Mycobacterium tuberculosis, clinical and radiographic evidence of current pulmonary TB and had been diagnosed for the first time between August 1, 2002 and October 31, 2003. A control was defined as an individual aged 15 years or more, attending the surgical departments of the above-mentioned hospitals at the time of recruitment of the cases, without clinical and radiographic evidence of pulmonary TB, as recorded in the medical file of the hospital. Cases were selected from the TB clinics of the study hospitals. Since the source population of the cases cannot be easily identified in a hospital-based study, therefore, controls were selected from the surgical departments of the same hospital in order to have a similar referral pattern as the cases. This provided some reassurance that all subjects come from the same source population ,,,. Controls admitted with complaints or diseases of the respiratory system or with a past history of TB were excluded from the study. Cases and controls were also excluded if they were unable to give an interview due to ill health or unstable state of mind.
Questionnaire and data collection
A structured and pre-tested questionnaire was used for data collection. The questionnaire comprised six components, including questions on socio-demographic characteristics, exposure to smoke from a biomass stove, contact with a pulmonary TB patient, history of institutionalization, tobacco smoking status and history of chronic diseases. Two trained interviewers collected the data from all the enrolled cases and controls.
A written informed consent was sought from all patients/guardians after explaining the purpose of the study and assuring the confidentiality of collected information. The study protocol was approved by the institutional ethics review committee, and permission for data collection was sought from the heads of TB clinics and the surgery departments of the study hospitals.
For this study, a sample size of at least 185 cases and 185 controls was estimated based on the 33% prevalence of smoking among potential control subjects , an alpha error of 0.05 , study power of 80% and an odds ratio (OR) ≥ 2 to relate most of the risk factors with pulmonary TB status.
Data were double entered using EPI-INFO version 6.04 (Centers for Disease Control and Prevention, Atlanta, GA, USA) and analyzed with SPSS version 14 (SPSS Inc., Chicago, IL, USA). Descriptive statistics were computed to describe the distribution of various characteristics of cases and controls. Using simple logistic regression, unadjusted OR and their 95% confidence intervals (CI) were calculated to assess the association of each variable with pulmonary TB. All independent variables associated (P⩽0.25) with the pulmonary TB status in univariable analyses were considered for inclusion in the multivariable logistic regression model . These included age, gender, ethnicity, educational level, marital status, housing affordability, overcrowding, past hospitalization, exposure to a relative with TB and daily contact with a TB patient. Adjusted ORs (ORadj) and their 95% CI were used for interpretation of the final model. Additionally, the parameters' estimates from the final model were used to compute the population attributable risk (PAR) percent of TB for each risk factor and summary PAR (%) for all the risk factors together .
| Results|| |
A total of 342 cases and 342 controls were recruited in the study. Response rate was 99.8%. The mean (SD) age (years) of cases and controls was 33.3 (16.7) and 40.7 (15.7), respectively. The proportions of male cases and controls were 52% and 48%, respectively. Furthermore, 55% of the cases and 73% of the controls were currently married. The proportions of cases and controls with no formal school education were 25% and 30% respectively. Additionally, 77% of the cases and 70% of the controls were living in overcrowded housing ([Table 1]). Univariable logistic regression analyses showed that age, marital status, house ownership, ethnicity, overcrowding, occupation, a family member who smoked, past hospitalization, concurrent chronic disease, exposure to a relative with TB, daily contact and sharing of unwashed utensils with a TB patient were significantly (p ⩽ 0.25) associated with the TB status ([Table 2]).
|Table 1: Distribution and univariate analysis of demographic factors associated with pulmonary tuberculosis in a case–control study, Karachi, Pakistan, August 2003–October 2003.|
Click here to view
|Table 2: Distribution and univariable analyses of risk factors associated with pulmonary tuberculosis in a case–control study, Karachi, Pakistan, August 2003–October 2003.|
Click here to view
The final multivariable logistic regression model included daily contact with at least one pulmonary TB patient, poor housing affordability (rented vs. owned), age and past hospitalization ([Table 3]). Pulmonary TB cases compared with the controls were more likely to have had daily contact with at least one pulmonary TB patient (ORadj= 5.07; 95% CI: 3.31, 7.78) or have had poor housing affordability (ORadj= 1.59; 95% CI: 1.13, 2.26) after adjusting for the effects of age and past hospitalization. The adjusted PARs (%) for daily contact with a TB patient and poor housing affordability was 29.7 and 12.7 respectively. An overall adjusted summary PAR (%) for both the factors together was 38.7 ([Table 3]).
|Table 3: Multivariable logistic regression model of determinants for pulmonary tuberculosis: a case–control study, Karachi, Pakistan, August 2003–October 2003.|
Click here to view
| Discussion|| |
Daily contact with at least one pulmonary TB patient and housing affordability (rented vs. owned house as an indicator of socioeconomic status) were identified as independent risk factors for pulmonary TB among individuals aged 15 years or older. The finding of daily contact with at least one pulmonary TB patient at home or in the work place as an independent risk factor in this evaluation is consistent with those of earlier studies ,,,. This finding has an important implication for prevention of M. tuberculosis transmission. It reiterates the call for screening of those having frequent contact with TB patient(s) and warrants curative or prophylactic therapy as required ,,. Pakistan's national TB control program is being implemented mainly through the public sector infrastructure in a setting, wherein 80–96% of TB cases are first seen by the private health care providers ,. Notably, TB control efforts in settings like this are convoluted by unawareness and poor adherence to the World Health Organisation's (WHO) and national TB treatment guidelines by private practitioners and poor treatment compliance by the patients ,,. Therefore, focused education of TB patients and their family members during their initial contact with health workers in the private/ public sector about how M. tuberculosis is spread and the significance of treatment compliance may help interrupt the transmission chain of the causative agent in settings like this.
Corroborating the findings of earlier studies ,,,, lack of home ownership as a proxy variable for low socioeconomic status was found to be an independent risk factor for pulmonary TB. Exactly how lack of home ownership as a measure of low socioeconomic status leads to increased TB risk is still unclear. Perhaps it results in poor nutrition, and/or psychological stress rendering the immune system vulnerable to M. tuberculosis infection and development of overt disease.
The presented data had some limitations. First, controls were recruited from surgical in-patients assuming that they represented the same source population as the cases, and had surgical patients developed pulmonary TB, they would have attended the same hospital for treatment. Second, pulmonary TB incident and prevalent cases diagnosed in the last year were included in the study. Controls were recruited on the same day as the cases. Therefore, time windows for exposures assessment for cases and controls might not have coincided. Third, controls were not tested with the same rigor as cases to establish their pulmonary TB status. Since, it was neither possible nor desirable to subject potential controls to the same diagnostic protocol to determine their TB free status. Therefore, some misclassification of controls cannot be ruled out, though the possibility of such occurrence was very low. Fourth, case–control design has an inherent problem of recall bias. Recall can also be differential with cases being more likely to remember exposures to putative risk factors as compared with controls. However, cases and controls were patients in this instance, therefore, it is likely that exposure misclassification if occurred, indeed would have been non-differential in nature.
| Conclusion|| |
In, urban Pakistan daily contact with at least one pulmonary TB patient and an SES indicator (home ownership) are directly and independently associated with pulmonary TB status. Reaching out to underprivileged TB patients for delivery of DOTS and focused education of patients and their contacts may lessen pulmonary TB risk in this and similar settings.
| Conflict of interest|| |
| Acknowledgments|| |
We are grateful to the study participants and their family members for their time and cooperation. We also thank the study data collectors for their assistance in efficient data collection and recognize the cooperation extended to us by the head of the departments and management of the study hospitals for their permission to conduct data collection in their respective facilities.
| References|| |
C. Dye, S. Scheele, P. Dolin, V. Pathania, M.C. Raviglione, Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project, JAMA 282 (7) (1999) 677–686.
M.P. Small, Tuberculosis research: balancing the portfolio, JAMA 276 (1996) 1512–1513.
WHO, Global Tuberculosis Control, World Health Organization, Geneva, 2011.
WHO, Global tuberculosis control; epidemiology, strategy, financing, in: WHO Report 2009, Geneva, Switzerland, 2009.
J. Alcaide, M.N. Altet, P. Plans, I. Parron, L. Folguera, E. Salto, et al, Cigarette smoking as a risk factor for tuberculosis in young adults: a case–control study, Tuber. Lung Dis. 77 (2) (1996) 112–116.
S.E. Buskin, J.L. Gale, N.S.Weiss, C.M. Nolan, Tuberculosis risk factors in adults in King County, Washington, 1988 through 1990, Am. J. Public Health 84 (11) (1994) 1750–1756.
A.C. Crampin, J.R. Glynn, S. Floyd, S.S. Malema, V.K. Mwinuka, B.M. Ngwira, et al, Tuberculosis and gender: exploring the patterns in a case control study in Malawi, Int. J. Tuberc. Lung Dis. 8 (2) (2004) 194–203.
J.R. Glynn, D.K. Warndorff, S.S. Malema, V. Mwinuka, J.M. Ponnighaus, A.C. Crampin, et al, Tuberculosis: associations with HIV and socioeconomic status in rural Malawi, Trans. R. Soc. Trop. Med. Hyg. 94 (5) (2000) 500–503.
P. Gustafson, V.F. Gomes, C.S. Vieira, P. Rabna, R. Seng, P. Johansson, et al, Tuberculosis in Bissau: incidence and risk factors in an urban community in sub-Saharan Africa, Int. J. Epidemiol. 33 (1) (2004) 163–172.
R. Kempainen, K. Nelson, D.N. Williams, L. Hedemark, Mycobacterium tuberculosis
disease in Somali immigrants in Minnesota, Chest 119 (1) (2001) 176–180.
M. Tekkel, M. Rahu, H.M. Loit, A. Baburin, Risk factors for pulmonary tuberculosis in Estonia, Int. J. Tuberc. Lung Dis. 6 (10) (2002) 887–894.
K. Tocque, M.A. Bellis, N.J. Beeching, Q. Syed, T. Remmington, P.D. Davies, A case–control study of lifestyle risk factors associated with tuberculosis in Liverpool, North-West England, Eur. Respir. J. 18 (6) (2001) 959–964.
A. Islam, Health-related millennium development goals: policy challenges for Pakistan, J. Pak. Med. Assoc. 54 (4) (2004) 175–181.
O. Miettinen, Theoretical Epidemiology: Principles of Occurrence Research in Medicine, John Wiley & Sons, Inc., New York, 1985.
K.J. Rothman, S. Greenland, Case–control studies, in: Modern Epidemiology, second ed., Lippinxott-Raven, Philadelphia, 1998.
S. Wacholder, J.K. McLaughlin, D.T. Silverman, J.S. Mandel, Selection of controls in case–control studies. I. Principles, Am. J. Epidemiol. 135 (9) (1992) 1019–1028.
S. Wacholder, D.T. Silverman, J.K. McLaughlin, J.S. Mandel, Selection of controls in case–control studies. II. Types of controls, Am. J. Epidemiol. 135 (9) (1992) 1029–1041.
R. Maher, S. Devji, Prevalence of smoking among Karachi population, J. Pak. Med. Assoc. 52 (6) (2002) 250–253.
J.J. Schlesselman, Case–control Studies: Design, Conduct, Analysis, Oxford University Press, Oxford, UK, 1982.
D.W. Hosmer, S. Lemeshow, Applied Logistic Regression, John Wiley & Sons Inc., New York, 1989.
P. Bruzzi, S.B. Green, D.P. Byar, L.A. Brinton, C. Schairer, Estimating the population attributable risk for multiple risk factors using case–control data, Am. J. Epidemiol. 122 (5) (1985) 904–914.
N.J. Claessens, F.F. Gausi, S. Meijnen, M.M. Weismuller, F.M. Salaniponi, A.D. Harries, High frequency of tuberculosis in households of index TB patients, Int. J. Tuberc. Lung Dis. 6 (3) (2002) 266–269.
CDC, Essential components of a tuberculosis prevention and control program. Recommendations of the Advisory Council for the Elimination of Tuberculosis, MMWR Recomm. Rep. 44 (RR-11) (1995) 1–16.
Ministry of Health, Pakistan, National Tuberculosis Control Program, Islamabad, Pakistan, 2009.
H. Sadiq, A.D. Muynck, Health care seeking behavior of pulmonary tuberculosis patients visiting TB Center Rawalpindi, J. Pak. Med. Assoc. 51 (1) (2001) 10-16.
K. Arif, S.A. Ali, S. Amanullah, I. Siddiqui, J.A. Khan, P. Nayani, Physician compliance with national tuberculosis treatment guidelines: a university hospital study, Int. J. Tuberc. Lung Dis. 2 (3) (1998) 225-230.
K.A. Karamat, S. Rafi, S.A. Abbasi, Drug resistance in Mycobacterium tuberculosis: a four years experience, J. Pak. Med. Assoc. 49 (11) (1999) 262-265.
R.G. Liefoogghe, Treatment adherence of tuberculosis patients in Bethania hospital, Sialkot, J. Coll. Physicians Surg. Pak. 7 (1997) 140-144.
[Table 1], [Table 2], [Table 3]