Analyze WHO Health Outcomes by Sex

Health comparisons can hide important differences when male, female, and combined observations are collapsed into one country-year record. The WHO Global Health Observatory Data Scraper exposes the Dim1 sex code so each value remains attached to its published population.

Request the dimensions explicitly

When presenting gaps, include the original male and female estimates, not only the calculated difference. Analysts can then inspect whether a change came from one population, both populations, or a revised bound. Keep terminology aligned with the source dimension rather than inferring identity from a statistical code.

{"indicatorCodes":["WHOSIS_000001"],"countryCodes":["USA","IND","GBR"],"sexCodes":["SEX_MLE","SEX_FMLE"],"yearFrom":2010,"yearTo":2026,"maxResultsPerIndicator":250}

Join male and female records only when indicator, country, year, and all other relevant dimensions match. Calculate a difference or ratio downstream and retain both original values plus uncertainty bounds. Do not reconstruct a combined-sex estimate by averaging the two rows; weighting and methodology can differ from that shortcut.

Charts should state whether they show an absolute gap, relative gap, or separate trends. Small changes can be less meaningful than the confidence intervals suggest, and cross-country interpretation still depends on coverage and methodology. Store WHO's formatted value alongside the numeric estimate because it communicates published precision.

Some indicators do not provide every sex dimension or year. Treat those gaps as unavailable and make coverage visible. Keep source URLs, region, and update dates in exported tables. A careful dimensional model prevents a common analytical failure: producing an elegant health-equity chart from records that did not actually describe comparable populations.