Stress hyperglycemia (SH) is associated with poor clinical outcomes in hospitalized patients. Delays in recognizing SH impact timing of treatment, however, the best way to identify those at risk is unclear. We assessed the value of a validated diabetes risk test, the Cambridge Risk Score (CRS), to predict SH in patients admitted to the hospital compared to using undiagnosed diabetes and prediabetes based on HbA1c. Adults admitted without diabetes or use of diabetes medication to medical and surgical services of an academic hospital over 4 years was analyzed. Patients with at least one measurement of blood glucose ≥ 140mg/dL during the stay were identified as having SH. CRS was calculated using age, sex, BMI, family history of diabetes, smoking status, and use of antihypertensives or steroids. Prediabetes was defined based on HbA1c 5.7-6.4 and undiagnosed diabetes was defined as HbA1c >=6.5 without ICD code of diabetes in the medical record. Multivariate regression was used to assess the association between CRS, prediabetes and undiagnosed diabetes, and risk of SH controlling for relevant covariates. 5,535 adults met inclusion criteria, 25.9% developed SH. CRS as a continuous variable was significantly associated with risk of SH after adjustment (OR 2.25, 1.84 - 2.74). Prediabetes was significantly associated with risk of SH after adjustment (OR 1.31, 1.13-1.51). However, undiagnosed diabetes was not significantly associated with SH, but many individuals were missing A1c measures which could have influenced results.

In conclusion results of this study suggest CRS risk stratification and early assessment for prediabetes may allow early identification of individuals at risk for SH.


C.E. Mendez: Advisory Panel; Self; Monarch Medical Technologies. R.J. Walker: None. K. Lu: None. A.Z. Dawson: None. L.E. Egede: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at