- Democratic governance is linked with declines in cardiovascular disease mortality and road deaths, as well as increases in government health spending
- Life expectancy improved faster in countries that transitioned to democracy between 1970 and 2015 compared to those that did not--increasing by an average of 3% after 10 years
- Democracy appears to play a bigger part in health outcomes than a country's GDP--accounting for about 25% of the reductions in deaths from cardiovascular disease and transport injuries over time
- Increasing funding for development agency-led programmes promoting open and accountable democratic institutions and processes may help improve health and increase investment in high-quality, accessible healthcare
Democratic rule, enforced by regular free and fair elections, appears to make an important contribution to adult health by increasing government spending on health and potentially reducing deaths from several non-communicable diseases (NCDs) and transport injuries. Conversely, autocracies that escape this general scrutiny, and do not have the same external pressures or support from global health donors to tackle NCDs and injuries, may have less incentive to finance their prevention and treatment, and seem to underperform as a result.
The findings are from the first comprehensive assessment of the impact of democracy on adult health and cause-specific mortality using detailed political, economic, and population health information for 170 countries over the past 46 years (1970-2016), published in The Lancet.
"The results of this study suggest that elections and the health of the people are increasingly inseparable," says Thomas Bollyky from the Council on Foreign Relations, USA, who led the research. "Without the same pressure or validation from voters or foreign aid agencies, autocratic leaders have less incentive than their democratic counterparts to finance the more expensive prevention and treatment of heart diseases, cancers, and other chronic illnesses. Despite being responsible for an estimated 58% of the death and disability in low- and middle-income countries, just 2% of development assistance for health was devoted to non-communicable diseases in 2016." 
The findings suggest that average adult life expectancy (after controlling for HIV/AIDS ) improved faster in countries that transitioned to democracy between 1970 and 2015 compared to those that did not transition--increasing by an average of 3% after 10 years (figure 1). Moreover, as levels of democracy increased, governments spent more on health, irrespective of a country's economic wellbeing (gross domestic product [GDP] per capita).
The causes of mortality that appear to be most affected by democratic experience--cardiovascular diseases, tuberculosis, transport injuries, and several other non-communicable diseases--are responsible for over a quarter of all the death and disability in individuals aged 70 years and younger in low- and middle-income countries.
"Our estimates represent a potentially major change in how we think about tackling global health challenges", says co-author Dr Joseph Dieleman from the Institute for Health Metrics and Evaluation, USA. "In a time of stagnant aid budgets, and as the burden of disease rapidly shifts to non-communicable diseases, international health agencies and donors may increasingly need to consider the implications of regime type in order to maximise health gains." 
"Efforts to improve the health of adults might benefit from funding programmes that help countries to strengthen their democratic processes and build more accountable institutions. So would directing more of the scarce development assistance for health to causes where democratic performance has the most effect on health, such as cardiovascular diseases." 
The past decade has seen falling levels of democracy around the world, with an estimated 2.5 billion people--a third of the world's population--living in countries where democratic qualities (e.g., freedom of expression, the right to vote, and freedom of association) are in decline .
The extent to which a country's democratic experience impacts population health has been poorly understood. So far, studies have focused on broad measures of child and infant health (e.g., mortality rates and life expectancy) that have reported conflicting findings, and have been unable to conclude whether the democratic process itself is affecting health, or if other factors such as country income, or the quality of government institutions, might be responsible.
To explore this further, the authors analysed political, economic, and population health data from the Global Burden of Disease study, the University of Gothenburg's Varieties of Democracy project, and Financing Global Health database . They modelled the effect of democracy on cause-specific mortality, HIV-free life expectancy at age 15 years, and health spending in 170 countries.
Given that democratic and autocratic countries alike received substantial amounts of international aid for HIV/AIDS, the authors controlled for HIV in the analysis of life expectancy at age 15 to better isolate and assess the effect of regime type on adult health between 1970 and 2015. The researchers also investigated the pathways by which democratic rule might improve health .
Results indicate that democratic experience accounted for more of the reduction in mortality within a country than GDP for cardiovascular diseases (22% vs 12%), transport injuries (18% vs 7%), cancers (10% vs 6%), cirrhosis (6% vs 2%), and other non-communicable diseases such as congenital heart disease and congenital birth defects (13% vs 9%) between 1995 and 2015 (figure 2). In contrast, democracy explained little of the declines in mortality for some of the leading communicable causes of death such as HIV (3%) and malaria (4%), which are more heavily targeted by international aid.
Democracy also did not appear to have substantial effects on mortality from all NCDs. Democratic experience appears to be important for causes dependent on good quality care, but accounts for less than 1% of the changes in mortality from diabetes, which is largely driven by non-utilisation of health-care services. The low number of deaths from mental health and musculoskeletal disorders globally make the study's results on these causes hard to analyse.
The researchers also estimated that the average country's increase in democratic experience (via direct and indirect effects such as increased government spending and economic growth) reduced deaths from cardiovascular disease and other NCDs by around 9%, and tuberculosis by roughly 8% between 1995 and 2015 (figure 4).
The authors acknowledge that many global health practitioners may be concerned that the more political global health assistance becomes, the more it could undermine productive relationships with local governments. They say: "This reticence about democracy promotion is understandable, but it ignores the inevitably political nature of many current global health objectives...Ignoring the role of civil society, a free media, and open and accountable government in resolving these debates undermines efforts to build institutional capacity and the popular support needed for sustained population health improvements. Pretending otherwise is akin to believing that the solution to a nation's crumbling roads and infrastructure is just a technical schematic and cheaper materials."
The authors note that the link between democracy and population health is difficult to measure because of the association of democracy with other factors, such as country income or total health expenditure. The authors used multiple statistical measures to reduce the risk of confounders, but they remain impossible to rule out. They also point out that the factors in the analysis together explained less than three-quarters of the total reductions in mortality for some causes, so there may be other factors that might play important roles, which they were unable to identify. Lastly, they note that whilst GBD 2016 provides the only comprehensive data on cause-specific mortality in all countries, it relies on modelling estimates that may be affected by a lack of data in low-income countries, particularly for causes like cardiovascular diseases and other NCDs.
Commenting on the implications of the findings in a linked Comment, Dr Helen Epstein from Bard College, USA, writes: "Global health advocacy groups need to do more than clamour for more funding and occasionally bemoan corruption. They need to call on Washington (USA), Brussels (Belgium), and London (UK), to impose sanctions on dictators, including those who cooperate with western military aims."
Peer-reviewed / Observational and modelling study / People
NOTES TO EDITORS
This study was funded by Bloomberg Philanthropies and the Bill & Melinda Gates Foundation. It was conducted by researchers from the Council on Foreign Relations, Washington, DC, USA; Stanford University School of Medicine, Stanford, CA, USA; Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA, USA; Bilkent University, Ankara, Turkey.
The labels have been added to this press release as part of a project run by the Academy of Medical Sciences seeking to improve the communication of evidence. For more information, please see: http://www.
 Quotes direct from authors and cannot be found in text of Article.
 HIV-free life expectancy estimates life expectancy in each country if the HIV pandemic had not happened, excluding war and natural disasters.
 The Varieties of Democracy projects provides numerous indicators of regime characteristics including electoral fraud, multiparty elections, freedom of expression, and media freedom in more than 200 countries from 1789 to 2017. The Institute for Health Metrics and Evaluation Financing Global Health database includes national government health spending estimates for 188 countries from 1990 to 2015.
 The authors estimated the direct and indirect effect of democracy on health including changes in mortality as a function of changes in democratic experience, GDP per capita, average years of education, urbanicity, skilled birth attendance, and government health expenditure.
IF YOU WISH TO PROVIDE A LINK FOR YOUR READERS, PLEASE USE THE FOLLOWING, WHICH WILL GO LIVE AT THE TIME THE EMBARGO LIFTS: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(19)30235-1/fulltext