Rethinking the Role of Oxygen Saturation in Assessing Childhood Illness

In the late 1970s, the introduction of pulse oximetry revolutionized the way clinicians could assess blood oxygen levels. By simply clipping a small sensor on a patient’s finger or toe, these devices could quickly and painlessly measure oxygen saturation in the blood. Over the following decades, pulse oximeters became indispensable tools in hospital emergency rooms, operating theaters, and intensive care units – allowing doctors to rapidly identify and treat dangerously low oxygen levels, known as hypoxemia.

However, the role of pulse oximetry has remained less clear when it comes to assessing and managing childhood illness in primary care and outpatient settings. While World Health Organization (WHO) guidelines recognize hypoxemia as a critical warning sign for severe illness in children, they provide limited guidance on how to interpret and act on different levels of oxygen saturation outside of the hospital. This has left clinicians and health programs uncertain about when and how to incorporate this vital sign into routine care.

In a new Viewpoint article published in The Lancet Global Health, an international team of child health experts argue that it’s time to rethink the use of pulse oximetry and oxygen saturation thresholds in the management of childhood illness, particularly in resource-limited primary care settings. Their analysis of data from diverse hospital and outpatient studies across low- and middle-income countries suggests that even moderate decreases in oxygen levels carry significantly elevated risks of death – highlighting the importance of pulse oximetry as a risk assessment tool, not just for guiding oxygen therapy.

“Pulse oximetry is often seen as a tool to simply identify who needs oxygen, but our data shows it has much broader value in assessing severity of illness and guiding referrals and follow-up, even in primary care settings,” explains lead author Dr. Hamish Graham from the Murdoch Children’s Research Institute in Australia. “We need to move beyond a binary ‘normal or abnormal’ view of oxygen saturation and instead recognize it as a powerful vital sign that can help frontline health workers make more informed triage and management decisions.”

Oxygen Saturation and Risk of Death

The authors’ reanalysis of data from nearly 50,000 children across multiple studies found a clear and consistent relationship between oxygen saturation levels and risk of death. Compared to children with normal oxygen levels (98-100%), the odds of death increased gradually as saturation fell:

– 94-95%: Doubled odds of death
– 92-93%: Tripled odds of death
– 90-91%: Quadrupled odds of death
– 88-89%: 6-fold increase in odds of death

This pattern held true across both inpatient and outpatient settings, with similarly elevated risks observed in the hospital and primary care populations.

“What’s really striking is just how high the risk is, even for children with ‘moderate’ hypoxemia in the 90-93% range,” says co-author Dr. Carina King from the Karolinska Institute in Sweden. “These are not insignificant decreases in oxygen – our data shows they come with a 2-3 times greater chance of dying compared to children with normal levels.”

The authors argue these findings have important implications for how clinicians and health programs approach oxygen saturation measurement and interpretation, particularly in primary care settings where pulse oximetry is still not widely used.

“Hypoxemia has typically been viewed through a binary lens – either you have it (below 90%) or you don’t. But this overlooks the reality that oxygen saturation exists on a spectrum, and even small decreases can signal serious underlying illness,” explains co-author Dr. Eric McCollum from the Johns Hopkins Bloomberg School of Public Health.

Pulse Oximetry in Primary Care

While pulse oximeters are now considered an essential medical device by the WHO and are widely used in hospitals, their role in primary care and community settings has remained unclear. Current WHO guidelines for the Integrated Management of Childhood Illness (IMCI) only briefly mention pulse oximetry, advising referral for any child with an oxygen saturation below 90%.

However, the new analysis suggests this threshold may be too low, potentially missing many children at high risk. The authors argue that moderate hypoxemia in the 90-93% range should also prompt careful clinical reassessment, consideration of referral, and close follow-up – even if the child does not meet other severe illness criteria.

“Health workers in primary care facilities are often working with limited resources and facing high patient loads. Adding pulse oximetry to their assessment toolkit could be a game-changer, but we need to give them clear guidance on how to interpret the results and what actions to take,” says Dr. Graham.

One key barrier to wider adoption of pulse oximetry in primary care has been concerns about feasibility – the time required to obtain an accurate reading, the need for appropriately-sized probes, and integration into busy clinical workflows. But the authors argue these challenges are not insurmountable, pointing to emerging examples of successful implementation.

“In Malawi, we found that with some targeted training and support, frontline health workers were able to routinely incorporate pulse oximetry into their assessment of sick children, using the results to guide referrals and follow-up,” says co-author Dr. Carina King. “The key is not just providing the devices, but ensuring health workers understand how to use them effectively and have the systems in place to act on the information.”

Beyond guiding oxygen therapy, the authors highlight pulse oximetry’s potential value in risk stratification and triage. Studies have shown that pulse oximetry can identify children with severe illness who may be missed by clinical signs alone. In Bangladesh, for example, the authors found that WHO’s IMCI guidelines implemented without oximetry missed 88% of children with severe hypoxemia, including all the children who subsequently died.

“Pulse oximetry isn’t a silver bullet, but it can be a really valuable addition to the toolkit, helping frontline workers to quickly identify the sickest children who need the most urgent attention,” says Dr. McCollum.

Rethinking Oxygen Saturation Thresholds

As countries and programs consider expanding pulse oximetry in primary care, a critical question is what oxygen saturation thresholds should guide clinical decision-making. The traditional cutoff of 90% has been primarily focused on identifying who needs oxygen therapy. But the new analysis suggests this may be too low for broader triage and risk assessment purposes.

“The 90% threshold makes sense for directing oxygen treatment, but our data indicates we should be concerned about children with saturations even as high as 90-93%,” explains Dr. Graham. “At these moderate levels of hypoxemia, the risks start to climb significantly, and health workers need to respond accordingly.”

Rather than a single cutoff, the authors propose a more nuanced, risk-based approach. Severe hypoxemia (<90%) should prompt urgent referral to a higher-level facility. Moderate hypoxemia (90-93%) should trigger careful clinical re-evaluation, consideration of referral, and close follow-up. And even children with borderline low saturations (94-95%) may warrant additional monitoring.

Importantly, the authors caution that oxygen saturation should never be viewed in isolation, but rather as one vital sign to be interpreted alongside other clinical assessments. A child with an SpO2 of 92% but no other signs of severe illness may require a different response than one with the same oxygen level who also appears lethargic or is struggling to breathe.

“Pulse oximetry gives us crucial information, but it has to be combined with a holistic clinical evaluation. Health workers need to be empowered to use their judgment, weighing the oxygen reading alongside the full picture of the child’s condition and their local context,” says Dr. King.

Looking Ahead

As countries work to expand access to pulse oximetry and strengthen oxygen services, the authors emphasize the importance of taking a system-wide approach. Introducing new technologies like pulse oximeters is not enough – health facilities also need reliable supplies of oxygen, functioning referral pathways, and high-quality inpatient care to actually save lives.

“We’ve seen too many examples of pulse oximeters being provided without the necessary supporting systems. It’s like giving a Ferrari to someone who only has dirt roads to drive on,” says Dr. McCollum. “Strengthening the entire continuum of care for sick children is essential if we want pulse oximetry to have its full impact.”

The authors also highlight the need for more research to guide implementation, particularly on effective training approaches, workflow integration, and strategies for maintaining equipment in resource-limited settings. Innovative technologies like smartphone-based oximeters may also expand access, but will require careful evaluation.

Ultimately, the team hopes their findings will spur a rethinking of oxygen saturation thresholds and the role of pulse oximetry, not just in hospitals, but in primary care facilities and throughout health systems.

“Pulse oximetry shouldn’t be seen as a niche tool for the sickest kids – it has tremendous potential to improve care for children at all levels,” concludes Dr. Graham. “By recognizing oxygen saturation as a vital sign and using it to guide both clinical management and health system strengthening, we can save many more young lives.”