Bridging the diagnostics gap: AI-enabled CGM is the next leap in diabetes care

Bridging the diagnostics gap: AI-enabled CGM is the next leap in diabetes care

Could new diagnostic technologies, enabled by AI, empower people living with diabetes to better manage their condition and live happier, healthier lives?

Diabetes is one of the world’s most complex and costly chronic diseases¹. Left unmanaged, it can lead to a long list of secondary complications that includes cardiovascular disease, nerve damage, kidney disease, blindness, lower limb amputation and mental health issues.

The number of people living with diabetes worldwide is increasing dramatically, placing even more pressure on healthcare systems that are already grappling with ageing and growing populations, stretched budgets and inequitable patient access.

In the Western Pacific region, defined by the International Diabetes Foundation as including the ASEAN countries, China, Mongolia and Oceania, 131.9 million adults were living with diabetes in 2011. By 2024, that number had increased to 215.4 million — and by 2050, is expected to reach 253.8 million².

For every one of those people, managing fluctuations in blood sugar day and night is a demanding, exhausting task. Using data from conventional continuous glucose monitoring (CGM) devices, they have to interpret, calculate and act immediately to avoid hypoglycemia — or risk loss of consciousness and a potential ER visit or ICU admission³.

This solution is far from ideal. Conventional CGMs only sound a blood sugar threshold alarm when glucose is already out of range, leading to alarm fatigue and reduced adherence, especially when alarm frequency is high and accuracy is suboptimal⁴.

The trend arrows displayed in conventional CGMs, meanwhile, only indicate direction and speed of retrospective glucose change. They don’t provide predictions for future glucose trends, don’t reflect sudden highs or lows and don’t give people living with diabetes an understanding of what caused the change⁵.

As a result, people living with diabetes are always one step behind their blood glucose, relying on guesswork to make the right intervention.

The global diagnostics access gap

The limitations of conventional CGM reflect a broader challenge across the global healthcare landscape: lack of access to diagnostic testing. Only 53% of the world’s population has access to diagnostics, falling to just 19% of the population in low- and middle-income countries (LMICs) in a primary care setting⁶.

Diagnostics are the cornerstone of resilient healthcare systems, essential for preventing and diagnosing disease as well as determining the right therapy and checking to see if it’s working.

In Asia Pacific, approximately 70% of clinical decisions rely on diagnostic test results, but only 2-3% of healthcare budgets are spent on diagnostic services⁷.

And living with a chronic condition like diabetes requires constant diagnostics and intervention — up to 180 times per day⁸. Effective diabetes management is about having the right information at the right time to take action.

AI-enabled CGM addresses these challenges

But more proactive, earlier intervention to improve diabetes outcomes is possible — with effective diagnostic tools. And new technologies are bridging the gap. Decentralised testing is bringing life-changing, personalised insights closer to home for people living with many different chronic conditions.

For people living with diabetes, next-gen, AI-enabled CGM technologies make blood glucose management easier, less stressful and more proactive. AI-enabled CGM isn’t about having more data; it’s about better interpretation, earlier warnings and intelligent support. About feeling confident enough to make the right decisions. About feeling safe, having the freedom to plan the day ahead and getting a good night’s sleep.

Trained on large datasets of blood glucose information, insulin doses, lifestyle inputs and circadian data, machine learning algorithms can predict glucose patterns before hypoglycaemic events occur, helping to prevent blood sugar spikes and lows rather than simply observing them happening^{9, 10}.

This addresses unmet diagnostic needs in three ways:

• Providing predictive alerts, rather than reactive alarms — flagging impending risk based on machine learning and giving the person more time to act¹¹
• Replacing retrospective trend arrows with two-hour prediction curves, refreshed every 5 minutes. The curve provides a forward-looking, continuously updated projection.
• Predicting overnight blood sugar lows by scanning for risk at night, to address undetected nocturnal hypoglycemia and alert the user before they go to sleep.

In a real-world CGM study of adults with type 2 diabetes over 5 days, CGMs identified mild or severe hypoglycaemia in 53 of 108 people (49%). Importantly, participants only recognised 25% of these incidents happening, indicating a high rate of hypoglycemia unawareness and proving the need for predictive alerts¹².

A call to action for collaborative change

The access gap to diagnostics is a global challenge. But every healthcare system is unique, requiring tailored and innovative approaches to both centralised and decentralised diagnostic solutions. So meaningful change is only possible if healthcare leaders, industry, NGOs and governments pool expertise and collaborate to drive greater diagnostics capacity.

As the WHO puts it, ‘increasing access to diagnostics could reduce annual premature deaths’, but providing ‘equitable access to safe, effective and quality assured diagnostics requires a comprehensive health-systems approach that addresses all stages of the value chain’.¹³

AI-enabled CGM can also ease the clinical and economic burden on healthcare systems under heavy pressure by preventing hospital admissions from hypoglycemic events and other complications. These technologies will soon become the gold standard for increasing self-efficacy in diabetes management, empowering hundreds of millions of people living with the condition with diagnostic data¹⁴.

It’s a clear imperative for action: we must all work together to put the power of diagnostics into individuals’ own hands, helping people living with diabetes to manage their condition proactively, rather than reactively. We must give them the insight they need to act sooner.

Only then can we rise to one the world’s most urgent public health challenges, and begin to turn the tide on diabetes.

You can help to drive change for the many millions of people living with diabetes in our region, too: please like, share or repost this article to broaden our reach for maximum impact. Thank you.