Value Health. 2018 04;pii: S1098-3015(17)33409-5. [Epub ahead of print]21(4): 482-490
OBJECTIVES: To summarize the costs of tuberculosis (TB) diagnosis and treatment in human immunodeficiency virus (HIV)-infected patients and to assess the methodological quality of these studies.
METHODS: We included cost, cost-effectiveness, and cost-utility studies that reported primary costing data, conducted worldwide and published between 1990 and August 2016. We retrieved articles in PubMed, Embase, EconLit, CINAHL plus, and LILACS databases. The quality assessment was performed using two guidelines-the Consolidated Health Economic Evaluation Reporting Standards and the Tool to Estimate Patient's Costs. TB diagnosis was reported as cost per positive result or per suspect case. TB treatment was reported as cost of TB drugs, TB/HIV hospitalization, and treatment. We analyzed the data per level of TB/HIV endemicity and perspective of analysis.
RESULTS: We included 34 articles, with 24 addressing TB/HIV treatment and 10 addressing TB diagnosis. Most of the studies were carried out in high TB/HIV burden countries (82%). The cost of TB diagnosis per suspect case varied from $0.5 for sputum smear microscopy to $175 for intensified case finding. The cost of TB/HIV hospitalization was higher in low/medium TB/HIV burden countries than in high TB/HIV burden countries ($75,406 vs. $2,474). TB/HIV co-infection presented higher costs than TB from the provider perspective ($814 vs. $604 vs. $454). Items such as "choice of discount rate," "patient interview procedures," and "methods used for valuing indirect costs" did not achieve a good score in the quality assessment.
CONCLUSIONS: Our findings point to the need of generation of more standardized methods for cost data collection to generate more robust estimates and thus, support decision-making process.
Keywords: TB/HIV co-infection; costs; diagnosis; treatment