Charles Henry de Soysa, the Ceylonese philanthropist whose 1879 maternity home in Colombo seeded the midwife-training system behind Sri Lanka's maternal health outcomes.

Map maternal mortality by country and the pattern looks almost like a map of income. Data: Our World in Data (CC BY). Sub-Saharan Africa and a handful of conflict-scarred states glow dark red: South Sudan loses more than 1,200 mothers for every 100,000 births, Chad and Nigeria over 1,000. Rich countries in Europe and East Asia sit at the pale end, some below two. The global average in 2023 was about 197. It is one of the clearest income gradients in all of public health — except for a small island off the southern tip of India that refuses to sit where its bank balance says it should.

Sri Lanka's maternal mortality ratio has been sitting in the high teens to twenties per 100,000 births in recent years — in the range of wealthy Western European countries, not its South Asian neighbors. Afghanistan, by comparison, loses roughly 500 mothers per 100,000 births; Pakistan around 155. Sri Lanka only crossed into the World Bank's upper-middle-income bracket in the last couple of years; for most of the twentieth century, and the period when this outcome was built, it was a poor to lower-middle-income country by any measure. Maternal death is normally one of the tightest correlates of national wealth anywhere in the data. Sri Lanka has spent nearly a century quietly breaking the correlation.

The story starts earlier than independence. In 1879, a Ceylonese businessman named Charles Henry de Soysa funded a "lying-in home" in Colombo, and in 1881 it opened the island's first school for training midwives as a distinct, credentialed profession — not traditional birth attendants learning informally from elders, but government-recognized practitioners taught obstetric basics. The colonial administration built on this instinct rather than importing a different model: organized public-health units began appearing from 1926, and by 1952 they covered the whole country. The core decision, repeated at every stage, was to put a trained midwife within reach of essentially every village, rather than concentrate resources in a few well-equipped city hospitals that most rural mothers would never reach in time.

The numbers this produced are almost hard to believe on paper. One widely cited estimate puts the national maternal mortality ratio at roughly 2,100 per 100,000 live births in 1881; by the mid-1990s it had fallen to about 24. Sri Lanka's ratio had dropped through the 1950s to something like 500 to 600 per 100,000 — comparable to today's worst-off African states — and then continued falling for another forty years while the country fought a civil war, endured coups and currency crises, and never got rich. The health system did not wait for prosperity to arrive first.

What makes the model durable is that it never depended on a single hospital or a single funding stream. Sri Lanka layered a public-health midwife assigned to a defined population of households, a small maternity home for routine deliveries within a short journey of most villages, and a referral hospital for genuine emergencies — three tiers, each doing only the job suited to it, all free at the point of use under the health system built out after independence in 1948. A World Bank study that compared Sri Lanka with Malaysia, another rare high-performer relative to its income, found the same pattern in both: political commitment sustained across changes of government, decades of investment in mid-level health workers rather than only doctors, and decentralization that put skilled care close to where women actually gave birth. Neither country waited to become rich; both decided maternal survival was a baseline, not a reward for development.

The system's toughest test came recently. Sri Lanka's 2022 economic collapse — a currency crisis so severe the country defaulted on its foreign debt and ran short of fuel, medicine and foreign exchange — threatened to unravel decades of gains overnight. Ratios ticked up in the immediate crisis years but did not collapse back toward South Asian norms, a sign that the decentralized, low-cost structure of public-health midwives and local maternity homes is harder to break than a system built around a few flagship city hospitals would have been.

The map of maternal mortality, read carelessly, looks like a map of GDP with a health overlay. Read carefully, it is a map of where governments decided, generations ago, that reaching the last village mattered more than building the most impressive hospital in the capital. Sri Lanka's patch of pale color, floating incongruously among much richer countries, is what that decision looks like nearly a hundred and fifty years later.

Sources: Our World in Data, "Maternal Mortality"; WHO, UNICEF, UNFPA, World Bank Group and UNDESA, Trends in Maternal Mortality 2000 to 2023; Indrani Pathmanathan et al., Investing in Maternal Health: Learning from Malaysia and Sri Lanka (World Bank, 2003); R. Haththotuwa et al., "Models of Care That Have Reduced Maternal Mortality and Morbidity in Sri Lanka," International Journal of Gynecology & Obstetrics (2012).

You're reading Lines & Legends, the Forummapping newsletter — the lines that move the world, and the legends behind them. Follow @forummapping on X for daily history maps, and forward this email to a friend who loves history.