"Equity-informative health technology assessment - A commentary on Ngalesoni, Ruhago, Mori, Robberstad & Norheim".

Leading health commentators have called for health equity to be made a measureable endpoint of health technology assessment (HTA) (Daniels et al., 2016; Horton, 2013). Nowhere is this more important than in low income countries on the path to universal health coverage, where reducing inequalities in health care, health outcomes and protection from the financial risks of ill-health are central policy concerns (Norheim et al., 2014; World Health Organization, 2015). A practical toolkit of methods now exists for providing decision makers with useful quantitative information about the equity implications of health technologies (Asaria et al., 2016; Cookson et al., 2016; Johri and Norheim, 2012; Verguet et al., 2016). As this study by Ngalesoni et al. on cardiovascular disease prevention in Tanzania illustrates, these methods can now be applied in low income country settings to generate useful policy insights and are becoming increasingly methodologically sophisticated. Studies of this kind will hopefully help pave the way for a new wave of applied research and methodological refinements in this area. If researchers and research funding bodies grasp this mettle, in the not too distant future the HTA evidence presented to decision makers will routinely address the health equity concerns that are so essential to achieving fairness on the path to universal health coverage. Who gains and who loses from a cost-increasing health programme depends on differences between people across a number of factors, including health risks, uptake of services, adherence to treatment and capacity to benefit. All of these factors may vary systematically by equity-relevant variables such as socioeconomic status, ethnicity, gender and geographical location. The overall impact on health equity also depends on how the programme is funded and who bears the largest opportunity costs of diverting scarce resources from other uses e which again may vary by social group. Careful quantitative analysis of these factors is becoming increasingly feasible as datasets and methods continue to improve, allowing