Cigarette taxation and neonatal and infant mortality: A longitudinal analysis of 159 countries.
Márta K RadóAnthony A LavertyThomas V HoneKiara ChangMohammed JawadChristopher J MillettJasper V BeenFilippos T FilippidisPublished in: PLOS global public health (2022)
Previous studies on the associations between cigarette taxes and infant survival have all been in high-income countries and did not examine the relative benefits of different taxation levels and structures. We evaluated longitudinal associations of cigarette taxes with neonatal and infant mortality globally. We applied country-level panel regressions using 2008-2018 annual mortality and biennial WHO tobacco taxation data. Complete data was available for 159 countries. Outcomes were neonatal and infant mortality. We conducted analyses by type of taxes (i.e. specific cigarette taxes, ad valorem taxes, and other taxes, import duties and VAT) and the income group classification of countries. Covariates included scores for other WHO recommended tobacco control policies, socioeconomic, health-care, and air quality measures. Secondary analyses investigated the associations between cigarette tax and cigarette consumption. We found that a 10 percentage-point increase in total cigarette tax as a percentage of the retail price was associated with a 2.6% (95% Confidence Interval [CI]: 1.9% to 3.2%) decrease in neonatal mortality and a 1.9% (95% CI: 1.3% to 2.6%) decrease in infant mortality globally. Estimates were similar for both excise and ad valorem taxes. We estimated that 231,220 (95% CI: 152,658 to 307,655) infant deaths could have been averted in 2018 if all countries had total cigarette tax at least 75%. 99.2% of these averted deaths would have been in low- and middle-income countries (LMICs). The secondary analysis supported causal interpretation of results by finding that a 10 percentage-point increase in taxes was associated with a reduction of 94.6 (95% CI: 32.7 to 156.5) in annual cigarette consumption per capita. Although causal inference is precarious due to the quasi-experimental design, we used a robust analytical approach and focused on within-country changes. Limitations include an inability to include data on roll-your-own tobacco, other forms of tobacco use, and reliance on taxation data only for the cigarette brands most sold in each country. In line with limited existing evidence conducted in HICs, we found that raising taxes on tobacco was associated with a reduction in neonatal and infant mortality globally. Implementing recommended levels of taxation in LMICs should be a priority since this is where the lowest levels of taxation and the largest potential infant mortality benefits exist.