October 10, 2026 interva.net Editorial

Five Billion People, One Molecule: What the Oxygen Gap Looks Like From a Mortality Desk

The Lancet Global Health Commission counts five billion people without reliable access to medical oxygen, and the gap maps almost exactly onto the regions where verbal autopsy carries cause-of-death statistics. From a fingertip pulse oximeter to a pressurized chamber, one molecule traces the full range of what health systems can deliver.

medical oxygenhealth systemshyperbaric oxygenglobal health

A verbal autopsy interview for a child who died of pneumonia records what a family saw: cough, fast breathing, chest indrawing, the journey to a clinic. What it cannot record is the oxygen saturation reading nobody took, or the cylinder the ward did not have. Behind a large share of the deaths this site writes about sits one unremarkable molecule — and a newly quantified gap in who actually receives it.

In February 2025, the Lancet Global Health Commission on medical oxygen security published the first full accounting of that gap. Its headline figure: more than five billion people — over 60% of the world’s population — do not have access to safe, quality and affordable medical oxygen services. An estimated 374 million people need oxygen each year for acute medical and surgical conditions; 306 million of them (82%) live in low- and middle-income countries, and 253 million (68%) in south Asia, east Asia and the Pacific, and sub-Saharan Africa. That is almost a perfect overlay of the map where verbal autopsy and mortality surveillance stand in for civil registration.

A medicine with no substitute

Oxygen entered WHO’s Essential Medicines List only in 2017, described at the time as “a life-saving essential medicine with no substitute.” The Commission shows how far delivery lags the designation. Pulse oximetry — the inexpensive fingertip sensor that decides who needs oxygen — was performed for only 19% of patients presenting to general hospitals in LMICs, and almost never at primary health-care facilities. Closing the acute medical and surgical oxygen gap would require an additional US$6.8 billion annually, roughly US$34 billion over five years. The pandemic proved the money can move when it has to: by the end of 2022, the ACT-Accelerator Oxygen Taskforce had mobilized more than US$1 billion worth of oxygen equipment and supplies to over 100 LMICs, more than 60% of it through the Global Fund.

The high-pressure end of the same therapy

At the far technological edge of oxygen treatment sits a therapy built on the same molecule: hyperbaric oxygen therapy, or HBOT, in which a patient breathes pure oxygen inside a chamber pressurized to two or three atmospheres, dissolving oxygen into blood plasma in quantities red blood cells alone could never carry. The Undersea and Hyperbaric Medical Society maintains the discipline’s formal list of recognized indications — diabetic foot wounds that have failed standard care, decompression sickness, delayed radiation injury, carbon monoxide poisoning and about a dozen more.

For most of its history, hyperbaric medicine lived exclusively inside hospital wound-care programmes and diving-medicine units. That has changed with remarkable speed. Chamber engineering has matured into standardized monoplace and multiplace designs, and the hardware market now serves hospitals, outpatient clinics and sports-recovery centres alike — anyone curious what current systems actually look like can browse superhumanchambers.com for a practical picture of modern designs, from compact single-person units to multi-patient rooms, with the pressure ratings that separate medical-grade hardware from wellness novelties.

Where the evidence is strongest

HBOT’s best-mapped territory is the non-healing diabetic foot ulcer. A 2021 systematic review and meta-analysis in Scientific Reports pooled fourteen controlled trials covering 768 patients and found that adding hyperbaric oxygen to standard wound care significantly improved complete ulcer healing (pooled odds ratio 0.29 for non-healing, 95% CI 0.14–0.61) and reduced the rate of major amputation (relative risk 0.60, 95% CI 0.39–0.92). Limb salvage and wound closure are precisely the outcomes that matter most — no surrogate endpoints, no biomarker proxies.

Where the verdict is still out

The evidence map is not uniformly favourable, and saying so plainly serves the field. Carbon monoxide poisoning — a cause-of-death category every mortality-surveillance system records, with at least 430 accidental deaths and roughly 50,000 emergency-department visits annually in the United States alone — has made HBO the theoretical treatment of choice for decades. Yet a 2022 network meta-analysis of eight randomized trials (n = 1,785) found no pooled advantage over normobaric oxygen on mortality or neurocognitive outcomes, and signalled more fatigue, memory and concentration complaints after two-session protocols. The UHMS keeps carbon monoxide poisoning on its indication list while further trials accumulate; the honest summary is that some indications rest on strong, consistent evidence and others remain open questions under active study.

Why oxygen reads like a system report card

Step back from the chamber and the pattern comes into focus. Oxygen is unusual among medicines in that a hospital cannot simply stock it on a shelf. It demands a working chain: reliable electricity or a steady cylinder supply, concentrators or pressure-swing-adsorption plants, a functioning pulse oximeter at the bedside, staff trained to titrate flow. When any link fails, the medicine does not exist for that patient — which is why global-health researchers increasingly treat oxygen access as a marker of overall health-system strength, one indicator that quietly measures power grids, supply chains, financing and workforce at once.

The mortality connection is direct. Pneumonia remains the leading infectious killer of children under five, and hypoxaemia is its deadliest complication. Complications of preterm birth and intrapartum events — the conditions that dominated the causal chains in the CHAMPS postmortem surveillance this digest covered earlier — are likewise oxygen-treatable wherever a facility has reliable supply. Every point of oxygen coverage gained shows up, eventually, as a cause-of-death entry that stops appearing.

What this means for a mortality desk

The full spectrum of oxygen therapy runs from a fingertip sensor costing a few dollars to a pressurized steel chamber costing as much as a house — yet the Commission’s argument is that all of it belongs to one continuum, and the weakest links cluster at the inexpensive end, in exactly the places where a verbal autopsy interview is the only record of the deaths that follow. Closing the oxygen gap ranks among the most cost-effective health investments the Commission models: fewer pneumonia deaths, fewer birth-asphyxia deaths, fewer deaths whose causes a fieldworker will one day have to reconstruct from a family’s memory. Surveillance counts the cost of the gap; therapy is how it closes.

interva.net is an independent digest of verbal autopsy and cause-of-death research. It is not affiliated with, endorsed by, or speaking for the InterVA project, Umeå University, the WHO, the Lancet Global Health Commission or the Undersea and Hyperbaric Medical Society.

Sources

← Back to the news digest