Investigation of the comparative cost-effectiveness of different strategies for the management of multidrug-resistant tuberculosis

Rockcliffe , Nicole (2003) Investigation of the comparative cost-effectiveness of different strategies for the management of multidrug-resistant tuberculosis. Masters thesis, Rhodes University.




The tuberculosis epidemic is escalating in South Africa as well as globally. This escalation is exacerbated by the increasing prevalence of multidrug-resistant tuberculosis (MDRTB), which is defined by the World Health Organisation (WHO) as resistance of Mycobacteria to at least isoniazid and rifampicin. Multi-drug resistant tuberculosis is estimated to occur in 1-2% of newly diagnosed tuberculosis (TB) patients and in 4-8% of previously treated patients. MDRTB is both difficult and expensive to treat, costing up to 126 times that of drug-sensitive TB. Resource constrained countries such as South Africa often lack both the money and the infrastructure to treat this disease. The aim of this project was to determine whether the performance of a systematic review with subsequent economic modelling could influence the decision making process for policy makers. Data was gathered and an economic evaluation of MDRTB treatment was performed from the perspective of the South African Department of Health. Three treatment alternatives were identified: a protocol regimen of second line anti-tuberculosis agents, as recommended in the South African guidelines for MDRTB, an appropriate regimen designed for each patient according to the results of culture and drug susceptibility tests, and non-drug management. A decision-analysis model using DATA 3.0 by Treeage® was developed to estimate the costs of each alternative. Outcomes were measured in terms of cost alone as well as the ‘number of cases cured’ and the number of ‘years of life saved’ for patients dying, being cured or failing treatment. Drug, hospital and laboratory costs incurred using each alternative were included in the analysis. A sensitivity analysis was performed on all variables in order to identify threshold values that would change the outcome of the evaluation. Results of the decision analysis indicate that the individualised regimen was both the cheaper and more cost-effective regimen of the two active treatment options, and was estimated to cost R50 661 per case cured and R2 070 per year of life saved. The protocol regimen was estimated to cost R73 609 per case cured and R2 741 per year of life saved. The outcome of the decision analysis was sensitive to changes in some of the variables used to model the disease, particularly the daily cost of drugs, the length of time spent in hospital and the length of treatment received by those patients dying or failing treatment. This modelling exercise highlighted significant deficiencies in the quality of evidence on MDRTB management available to policy makers. Pragmatic choices based on operational and other logistic concerns may need to be reviewed when further information becomes available. A case can be made for the establishment of a national database of costing and efficacy information to guide future policy revisions of the South African MDRTB treatment programme, which is resource intensive and of only moderate efficacy. However, due to the widely disparate range of studies on which this evaluation was based, the outcome of the study may not be credible. In this case, the use of a systematic review with subsequent economic modelling could not validly influence policy-makers to change the decision that they made on the basis of drug availability.

Item Type:Thesis (Masters)
Uncontrolled Keywords:Tuberculosis, Multidrug resistance
Subjects:R Medicine > RS Pharmacy and materia medica
Divisions:Faculty > Faculty of Pharmacy
ID Code:2286
Deposited By: Mrs Carol Perold
Deposited On:01 Dec 2011 06:51
Last Modified:06 Jan 2012 16:22
19 full-text download(s) since 01 Dec 2011 06:51
19 full-text download(s) in the past 12 months
More statistics...

Repository Staff Only: item control page