Factors associated with multi-drug resistant tuberculosis in Kenya

Abstract

Multi-drug resistant tuberculosis (MDR-TB) and weak health systems threaten global tuberculosis control. Kenya is ranked 13th among the 22 high TB burden countries worldwide,and currently has an estimated 2,300 MDR-TB patients. A case-controlstudy to determine factors associated with MDR-TB in Kenya was conducted toinform policy in designing public health interventions that are best suited to thecountry’s needs.This was an unmatched case control study conducted in 41 health facilities in 20 districts across the eight provinces in Kenya from September 2009 to January 2010.Cases were confirmed MDR-TB (resistance to at least rifampicin and isoniazid)patients while controls were sputum- smear positive TB patients with clinicalresponse and negative sputum smear at the fifth month of treatment with first-lineanti-tuberculosis drugs. Study approval was sought and obtained from relevantinstitutions. Using the health facility TB register as the sampling frame, MDR-TBpatients and two randomly selected unmatched controls per case were enrolled. Apretested structured interviewer administered questionnaire was used for patientinterviews and to abstract information from records. Data on socio-demographic,behavioural, and clinical exposure history were obtained. Data were entered andanalyzed using Epi-info and Stata versions 3.5 and 9.0 software respectively.A total of 81cases {mean age: 32 years (SD: 10), 62% males} and 162 controls{mean age: 35 years (SD: 13), 59% males} there was no statistically significantdifference with respect to baseline socio-demographic characteristics. Six (7.4%) ofthe MDR-TB cases having no previous history of TB, reported living in the samehouse with a known MDR-TB patient. Cases were more likely to have history ofprevious exposure to first line anti-Tuberculosis drugs (OR= 85, 95% CI=29.7-243.3; P<0.0001) and be non Kenyan (OR=5.5, 95% CI=1.4-21.8; P=0.007). Casepatientswith positive HIV status (OR=0.34, 95% CI= 0.1-0.9;P=0.025) and thosewho had received TB treatment under the Directly Observed Therapy program(DOT) (OR=0.23, 95% CI= 0.1-0.6; P=0.002) were less likely to have MDR-TB.The study established that MDR-TB was associated with previous TB treatment, andbeing non Kenyan while use of DOT was protective. MDR-TB could be transmittedto otherwise healthy individuals. The protective association with HIV positive serostatusmay reflect selective survival of HIV negative MDR-TB and thus need to beinvestigated.We recommend strengthening of MDR-TB surveillance among previously treated TBcases and refugees and active MDR-TB case finding among HIV infected TBpatients. More rapid MDR-TB diagnostic tests should be used among the HIVinfected patients. Access to TB care services by all population groups includingimmigrants, implementation of DOT, MDR-TB contact tracing and screening andinfection prevention should be strengthened in Kenya.