From sex differences to sex inequalities in life expectancy: A cross-country observational benchmarking analysis.

BACKGROUND: The answer to whether females or males have better health, and which sex is the more disadvantaged, has depended in part on the metric and how the inequality is measured. This study introduces a new method for analyzing and interpreting sex inequalities in health outcomes-defined as the avoidable sex differences in health outcomes-that is systematic and potentially more objective. For this paper, we focus on life expectancy at different ages. METHODS AND FINDINGS: We introduce the adjusted sex ratio as a measure of sex inequalities and determining sex disadvantage. First, we calculated the sex ratio of life expectancy at ages 0, 5, 15, 35, 50, and 70. To understand what is achievable under favorable conditions, we identified countries in the 5th percentile of the highest life expectancy for each sex and used these values as benchmarks, and calculated the sex ratio of these best-performing countries ("frontier"). We calculated the country- and age-specific adjusted sex ratio by dividing country sex ratios by frontier sex ratios. This assumes that theoretically, under the current risk and healthcare environments, females all over the world have the potential to live up to the life expectancy of the females in the frontier countries, and separately, all males to their male-specific frontier. An adjusted ratio of greater than one indicates male disadvantage, while below one indicates female disadvantage. To avoid overinterpreting small differences, we defined a narrow range around equality (ratio of 1) within which we do not label either sex as disadvantaged. Before adjustment, males in all countries (except two) and at all ages had lower life expectancy than females. After adjustment, between 13% (at age 0) and 33% (at age 70) of the 237 countries shift from male to female disadvantage in life expectancy. More than half of the countries remain male-disadvantaged, indicating that males are generally disadvantaged in terms of life expectancy in most countries, even after our adjustments. India and approximately half of the countries in the Middle East and North Africa, North Atlantic, sub-Saharan Africa, and Western Pacific and Southeast Asia show female disadvantage. The number of countries with female disadvantage rises with age, especially in sub-Saharan Africa and Western Pacific and Southeastern Asia. Central and Eastern Europe show substantial male disadvantage across nearly all ages, even with adjustment. Our frontier selection and buffer range are empirical choices, and other definitions could be equally valid. Although our sex-specific benchmarks use the best-performing countries for each sex, they are not meant to represent purely biological differences, as observed sex gaps in life expectancy may also reflect unmeasured genetic variation, environmental exposures, and their interactions with sex. CONCLUSION: This study provides a novel, potentially more objective method for assessing sex inequalities in health outcomes, and presents the trends across countries, age, and time.