Introduction Tuberculosis (TB) is a major cause of death worldwide. Control of Tuberculosis is a serious challenge to global health. A new and potentially devastating threat to TB control is the emergence of multi-drug resistant TB (MDR-TB). South Africa was ranked fourth among the countries with the highest number of confirmed MDR-TB cases. Aim The aim was to investigate the annual MDR-TB prevalence and associated risk factors for MDR-TB from 2001 to 2007 at the Prince Cyril Zulu Communicable Disease Centre. To investigate previous TB treatment duration, previous TB treatment outcome, and duration of previous TB treatment interruption in a subgroup of patients who were previously treated for TB. To determine the average length of time from diagnosis of TB to diagnosis of MDR-TB and commencement of MDR-TB treatment. Methods An observational analytic nested case-control study design was used. All patients who were diagnosed with pulmonary TB and who had a sputum culture performed between 2001 and 2007 were included in the study. The cases were all MDR-TB cases diagnosed on sputum culture between 2001 and 2007. The controls were drug susceptible TB cases which had a sputum culture done at diagnosis, and were diagnosed in the same month as the MDR-TB case Results There were 10 205 sputum cultures performed from 2001 to 2007. MDR-TB was found in 445 patients. An increase in the prevalence of MDR-TB occurred in 2007, due to a significant increase in prevalence among new TB cases. The MDR-TB prevalence was 11.7% among new TB cases and 4.7% among previously treated TB cases in 2007. There was no significant association between demographic characteristics and MDR-TB. Previous TB treatment failure and a duration of previous TB treatment of greater than 32 weeks was found to be significantly associated with MDR-TB. The median time from TB diagnosis to MDR-TB diagnosis was 98 day and from MDR-TB diagnosis to MDR-TB treatment 10 days. Discussion Delays in the diagnosis of MDR-TB, long waiting times before MDR-TB treatment commencement and lack of isolation have contributed to the spread of primary MDR-TB and was most likely responsible for the increase in prevalence of MDR-TB among new TB cases. Recommendations It was suggested that a sputum specimen should be obtained for culture and sensitivity from all new TB patients in areas which have an MDR-TB prevalence of greater than 3% among new TB patients. Ensure patient education on basic infection control measures. Improve MDR-TB diagnosis and reduce waiting times for MDR-TB treatment. / Thesis (M.Med.Sc.)-University of KwaZulu-Natal, Durban, 2011.