Overview of the Burden of Disease Review Manager
Burden of Disease studies aim to provide a comprehensive assessment of mortality, ill-health and disability experienced in the population. The Disability Adjusted Life Year (DALY) is a summary measure that combines the health loss from premature mortality (Years of Life Lost [YLLs]) as well as the loss from disability associated with non-fatal outcomes (Years of Life Lived with Disability [YLDs]) (Murray et al., 1996; Mathers et al., 2001).
Certain epidemiological parameters are required for estimating YLDs and attributable burden of modifiable risk factors; these are prevalence, incidence, case-fatality rate, relative risk, odds ratio, hazard ratio, mean, incidence ratio, severity, duration and remission.
To ensure the use of quality estimates, a systematic review, including a risk-of-bias assessment of studies producing these estimates may be useful (Mathers et al., 2001). While critical appraisal of randomised control trials is well established by the Cochrane Collaboration (Higgins et al., 2011), similar tools across other study designs are not standardised.
The Burden of Disease Research Unit (BODRU) of the South African Medical Research Council has developed a standardised risk-of-bias tool that can be used for cross-sectional (including population-based surveys) studies, case-control studies, cohort studies and surveillance. A risk-of-bias score (0-20) can be obtained based on a set of defined criteria to assess the external and internal validity of a study. This tool can be applied to assess the risk-of-bias of observational studies and guide the decision on whether to use or disregard the epidemiological parameters from that particular study. The risk-of-bias-score can also guide statisticians on the size of the weight (based on the quality) that can be applied to estimates from various studies pooled in a meta-analysis. A web-based system, the Burden of Disease Review Manager (BODRevMan), has been developed to manage the systematic review process, including the risk-of-bias assessment and summary of data extracted from articles/studies.