Every day, patients move through admission with a clinically significant condition that never makes it onto their problem list. It isn’t rare, it isn’t subtle in its consequences, and it is almost entirely actionable — yet it remains one of the most consistently overlooked findings in acute care: elevated blood glucose at the point of admission.

For hospital and health-system leaders, this is not a narrow clinical footnote. Unrecognized hyperglycemia sits at the intersection of nearly every metric that matters to the C-suite — mortality, length of stay, readmissions, surgical complications, and cost per case. And because it so often goes unidentified and undocumented, it operates as a silent driver of poor outcomes and lost revenue that most quality dashboards never capture. Getting admission-stage glycemic identification right is one of the highest-leverage, lowest-cost improvements a hospital can make.

For organizations without dedicated inpatient diabetes resources, this gap is usually operational rather than informational: hospitals know the standard of care — what they lack is the staffing model, workflow design, and data visibility to execute it reliably at every admission. That distinction, as we’ll see, is exactly where the opportunity lives.

How Common Is Hyperglycemia Among Hospitalized Patients?
The scale is larger than most administrators assume. In the foundational 2002 study published in The Journal of Clinical Endocrinology & Metabolism, Umpierrez and colleagues reviewed more than 2,000 consecutive admissions and found that 38% of patients had hyperglycemia — 26% with known diabetes and 12% with no prior diabetes history at all. Later observational studies and guideline reviews have broadly replicated that 38–40% range across general inpatient populations.

The numbers climb higher in intensive care, where roughly 46% of ICU patients experience hyperglycemia at some point in their stay. And among adults 65 and older — the population that fills most Medicare-funded beds — more than 29% have diabetes, a prevalence that only compounds the identification challenge.

Stress Hyperglycemia: The Hidden Population
Not everyone with high glucose in the hospital has diabetes. Stress hyperglycemia — transient glucose elevation triggered by acute illness in patients without a prior diagnosis — affects an estimated 11–12% of all hospitalized patients, and a substantially higher share of the critically ill. This is the population that matters most and gets watched least: these patients carry no diabetes label to trigger a protocol, no home regimen to reconcile, and often no follow-up plan when they leave.

Who Goes Undiagnosed — the Documentation Gap
The identification problem has two layers. The first is undiagnosed diabetes. In one large analysis of ICU admissions at a major academic center, roughly 9% of all ICU patients had unknown diabetes — an elevated A1c with no documented history — mirroring national CDC estimates that about a quarter of people with diabetes are undiagnosed.

The second layer is even more striking: even when hyperglycemia is detected, it frequently is never recorded. A 2021 study in Scientific Reports examined discharge summaries for hospitalized patients with clear laboratory evidence of stress hyperglycemia and found that the condition was almost never carried into the discharge problem list, rarely mentioned anywhere in the narrative, and essentially never communicated to the outpatient providers who would manage these patients next. In practical terms, a documented, measurable glucose abnormality vanished from the record before the patient reached follow-up. That is not a clinical-competence issue — it is a workflow and accountability issue, and it is fixable.

Why Unrecognized Hyperglycemia Can Be More Dangerous Than Known Diabetes
Here is the counterintuitive finding that should reframe how leaders think about this: the patient with new, unrecognized hyperglycemia is often at higher risk than the patient with acknowledged diabetes.

In the Umpierrez data, in-hospital mortality was 16% among patients with new hyperglycemia, compared with 3% in patients with known diabetes and 1.7% in normoglycemic patients — and the elevated risk in the new-hyperglycemia group persisted after adjustment for age, comorbidities, and illness severity. The likely explanation is intuitive: a patient with a diabetes diagnosis triggers protocols, monitoring, and specialist attention. A patient with unlabeled hyperglycemia triggers none of that, so the condition goes untreated precisely where it is most dangerous.

More recent evidence reinforces that hyperglycemia is not merely a bystander marker of how sick a patient already is. A 2024 prospective cohort study in The Journal of Clinical Endocrinology & Metabolism found that in-hospital hyperglycemia was an independent contributor — modest in magnitude but statistically robust after adjustment for clinical covariables — to healthcare-associated infection, acute kidney injury, and in-hospital stroke. The signal matters: even after accounting for how sick patients are, elevated glucose adds incremental harm.

That pattern is sharpest in surgical patients. Perioperative hyperglycemia is one of the most consistently cited risk factors for surgical-site infection across specialties, with landmark surgical studies documenting several-fold higher serious-infection rates in patients whose perioperative glucose ran high. For a hospital measured on SSI rates and perioperative outcomes, admission-stage identification is where the work pays off first.

The Financial Case for Systematic Glycemic Screening
The clinical stakes translate directly into financial ones.

Length of stay. In the Umpierrez cohort, patients with new hyperglycemia had a mean length of stay of 9.7 days, versus 4.5 days for normoglycemic patients — roughly a five-day gap. The average expense of an inpatient day in the U.S. runs around $3,025 (2022). Average day-cost isn’t the same as marginal savings, but excess length of stay creates capacity, staffing, and reimbursement pressure that compounds quickly across a hospital’s admission volume.

Readmissions and CMS penalties. Best estimates place 30-day readmission rates for hospitalized patients with diabetes at 16–20.4%, and adults with diabetes account for a disproportionate share of all unplanned readmissions. This feeds directly into the CMS Hospital Readmissions Reduction Program (HRRP), which penalizes hospitals up to 3% of inpatient Medicare revenue. HRRP tracks specific high-volume conditions — heart failure, pneumonia, COPD, and others — that frequently co-occur with diabetes. Because diabetes and stress hyperglycemia so often accompany those conditions, poor glycemic control can raise readmission risk the program never attributes to glucose, especially when hyperglycemia is never identified, managed, or handed off effectively. Notably, a classic randomized study of a multidisciplinary inpatient diabetes management service cut the readmission rate to roughly half that of usual care — early evidence that structured glycemic programs can pay for themselves.

Aggregate burden. The total cost of diabetes in the U.S. reached $412.9 billion in 2022, and inpatient hospital care is the single largest direct-cost category. For any given health system, the inpatient slice of that number is not abstract — it is bed-days, complications, and readmissions that begin, or could be prevented, at admission.

What the Guidelines Say About Admission Identification
The standard of care is already clear; the gap is in execution.

The American Diabetes Association’s Standards of Care in Diabetes (2026), Section 16, calls for an A1c test on admission for any patient with diabetes or hyperglycemia (a random glucose above 140 mg/dL) when no A1c result is available from the prior three months. The clinical logic is simple and powerful: an admission A1c distinguishes true stress hyperglycemia (normal A1c) from previously unrecognized or poorly controlled diabetes (elevated A1c) — two conditions with very different management and very different follow-up needs. The Endocrine Society’s 2022 guideline for hyperglycemia in non-critical care settings similarly emphasizes systematic glucose monitoring on admission and clear escalation pathways.

Yet real-world implementation lags badly. Published audits have repeatedly found admission A1c testing performed in only a small fraction of eligible hyperglycemic patients. The recommendation exists; the operational muscle to execute it consistently — at scale, across every unit and every admission — usually does not.

Building a Systematic Approach to Glycemic Identification
Closing this gap is less about clinical knowledge than about designing the workflow so the right thing happens automatically. Three questions define whether a hospital identifies hyperglycemia reliably: Who screens? Who flags? Who acts?

In most hospitals, the honest answer is “no one owns it.” Screening depends on whichever clinician happens to notice a glucose value; flagging depends on manual vigilance; and action waits for a consult that is often placed only after a complication has already emerged. A systematic approach replaces that ad hoc chain with defined ownership: every admission with a glucose above threshold receives an A1c, a documented assessment, and — when warranted — a management plan and a follow-up pathway, without relying on any individual to remember.

This is where embedded specialists and real-time data change the equation. Glucometrics — the structured measurement of glucose control across patient-days, whole stays, and hypoglycemic events — turns thousands of raw glucose readings into a population-level view of who is hyperglycemic, how severely, and where the gaps are. That intelligence layer is what makes identification proactive instead of reactive, and it gives quality and C-suite leaders the benchmarking data they need for regulatory reporting and program evaluation.

From Identification to Action: A Framework for Health Systems
The evidence points to a clear conclusion: the value isn’t in detecting hyperglycemia once, but in building a repeatable system that catches it at admission, acts on it during the stay, and carries it into post-discharge care. That end-to-end continuity is precisely the gap most hospitals struggle to staff and sustain.

An embedded model — diabetes-focused advanced practice providers (APPs) participating in inpatient workflows, with registered dietitians (RDs) extending management into ambulatory follow-up — is structurally designed to close it. Rather than waiting for a complication-triggered consult, embedded APPs can flag hyperglycemia at admission, order the A1c, initiate protocol-based management from day one, and ensure the condition is documented and handed off rather than lost at discharge. This is the model JM2 Health operates: an end-to-end glycemic management program that treats admission-stage identification as the front door to better outcomes, not an afterthought.

For rural and mid-size hospitals that lack the volume to justify full-time endocrinology or diabetes APPs, the same standard can be delivered through virtual APP and RD coverage — remote specialists reviewing glucose data and entering recommendations directly into the EHR, a model shown to reduce both hyperglycemia and hypoglycemia. And JM2 Health’s glucometrics reporting supplies the data infrastructure beneath all of it, giving leaders the population-level visibility to identify hyperglycemia, quantify severity, benchmark performance, and demonstrate improvement over time.

Admission-stage glycemic identification is a rare thing in hospital operations: a single point of leverage that improves safety and quality and reduces cost at the same time. The data have been clear for two decades. The opportunity is to finally operationalize it.

Ready to identify glycemic risk earlier? JM2 Health helps hospitals and health systems build end-to-end glycemic management programs that connect admission screening, inpatient workflows, glucometrics reporting, and post-discharge follow-up. Schedule a consultation to see how JM2 Health can strengthen your organization’s diabetes and hyperglycemia management strategy.

Frequently Asked Questions
What percentage of hospital patients have undiagnosed diabetes or hyperglycemia?
Roughly 38% of hospitalized patients have hyperglycemia, and about 12% have no prior diabetes diagnosis at all. In ICU populations, close to one in ten patients has previously unrecognized diabetes.

Why does hyperglycemia in the hospital increase mortality risk?
Patients with new, unrecognized hyperglycemia had markedly higher in-hospital mortality than patients with known diabetes in landmark data — largely because unlabeled hyperglycemia doesn’t trigger the monitoring and protocols that a diabetes diagnosis does, so it goes untreated where it is most dangerous. Recent research also shows hyperglycemia independently contributes to infection, kidney injury, and stroke.

Should all hospitalized patients be screened for diabetes on admission?
Current ADA guidance recommends an A1c test on admission for any patient with diabetes or a random glucose above 140 mg/dL if no A1c is available from the prior three months. This distinguishes true stress hyperglycemia from previously unrecognized diabetes.

How does inpatient hyperglycemia affect length of stay and readmissions?
Hyperglycemia is associated with substantially longer stays — several additional days in landmark data — and 30-day readmission rates for patients with diabetes run 16–20.4%, feeding directly into CMS readmission penalties of up to 3% of Medicare revenue.

What is stress hyperglycemia and why does it go undocumented?
Stress hyperglycemia is transient high blood glucose caused by acute illness in patients without diabetes. It frequently goes undocumented because these patients carry no diabetes diagnosis to trigger a protocol, and studies show the condition is rarely recorded in discharge summaries or communicated to outpatient providers.

This article is intended as thought leadership for hospital and health-system decision-makers and is not clinical guidance. Medically reviewed by [Reviewer name, credentials], [Month Year].

References

Umpierrez GE, et al. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002;87(3):978–982.
Barmanray RD, et al. Hyperglycemia in hospital: an independent marker of infection, acute kidney injury, and stroke for hospital inpatients (DINGO). J Clin Endocrinol Metab. 2024;109(11):e2048–e2056.
American Diabetes Association. 16. Diabetes Care in the Hospital: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(Suppl. 1):S339–S355.
Endocrine Society. Management of Hyperglycemia in Hospitalized Adult Patients in Non-Critical Care Settings (2022 Clinical Practice Guideline).
Moyer AL, et al. Lack of recognition and documentation of stress hyperglycemia is a disruptor of optimal continuity of care. Sci Rep. 2021.
American Diabetes Association. Economic Costs of Diabetes in the U.S. in 2022. Diabetes Care. 2023.
CMS. Hospital Readmissions Reduction Program (HRRP).
Ostling S, et al. Pre- and post-discharge risk factors for hospital readmission among patients with diabetes.