Table 4– Characteristics of the cohort study design, outcome and exposure

First author [ref.]	Country	Recruitment year	Setting	Inclusion	DST method	Genotyping method	Treatment adherence	Treatment regimen	Sample size
Sonnenberg [28]	South Africa	1995	Hospital serving four gold mines	Culture-positive TB patients without MDR and cured of TB	Bactec system or conventional proportional count method and Lowenstein–Jensen medium	IS6110 DNA fingerprinting	DOT	2HRZE/4HR	326^#
Quy [16]	Vietnam	1996–1998	TB control programme	New smear- positive TB	Proportion method on Lowenstein–Jensen medium	IS6110 DNA fingerprinting	DOT	2HRZS/6HE	2901
Cox [27]	Uzbekistan	2001–2002	DOTS programme	Smear-positive pulmonary TB patients	Lowenstein–Jensen media using proportion method or modified proportion method in Bactec 460TB	IS6610 DNA fingerprinting and spoligotyping	DOT	2HRZE (with or without S)/4HR	209
Cox [27]	Uzbekistan	2001–2002	DOTS programme	Smear-positive pulmonary TB patients		IS6610 DNA fingerprinting and spoligotyping	DOT	2HRZES/1HRZE/5HRE	173
Matthys [15]	Russian Federation	1997–1998	TB treatment programme in a penitentiary hospital	Newly admitted patients diagnosed with TB through sputum smear and culture	Proportion method on Lowenstein–Jensen medium	IS6110 DNA fingerprinting	Daily strict supervision	2HRZES/1HREZ/5HRE	233

DST: drug susceptibility testing; TB: tuberculosis; MDR: multidrug resistant; DOTS: directly observed treatment, short course; H: isoniazid; R: rifampicin; Z: pyrazinamide; E: ethambutol; S: streptomycin. ^#: initial pre-treatment DST result missing for three TB patients.