Precision Prescribing of SGLT2 Inhibitors for Heart Failure Prevention in Type 2 Diabetes
For people living with type 2 diabetes, preventing future heart failure is an important clinical goal. Sodium-glucose cotransporter 2 inhibitors, or SGLT2 inhibitors, are a class of glucose-lowering medications that have transformed diabetes care because they not only lower blood sugar, but also reduce the risk of heart failure and slow kidney disease progression. They are now widely recommended for patients with type 2 diabetes who already have atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease.
However, most people with type 2 diabetes do not have those conditions at the time treatment decisions are made. For this much larger group, current guidelines do not clearly identify who is most likely to gain the greatest heart failure prevention benefit from starting an SGLT2 inhibitor. This study aimed to solve that problem by developing and validating a prediction model that estimates an individual patient’s absolute heart failure benefit from SGLT2 inhibitor therapy.
Why a Personalized Model Was Needed
Clinical trials have shown that SGLT2 inhibitors reduce the relative risk of heart failure, but relative risk is only part of the story. A medication may have the same proportional effect across many patients, yet the actual number of events prevented can differ greatly depending on a person’s baseline risk. Someone with a higher chance of developing heart failure over the next few years stands to gain more in absolute terms than someone at very low risk.
That is the key idea behind precision prescribing. Rather than using a one-size-fits-all approach, clinicians can estimate who is most likely to benefit from treatment and prioritize therapy accordingly. This is especially useful when balancing treatment burden, cost, side effects, and patient preferences.
How the SABRE Model Works
The investigators developed the SGLT2i Absolute Benefit Response, or SABRE, model. It combines two important pieces of information. First, it estimates a patient’s baseline 5-year risk of developing heart failure using the validated QDiabetes-HF model. Second, it applies the heart failure relative treatment effect seen with SGLT2 inhibitors from a meta-analysis of randomized trials, which found a hazard ratio of 0.63 for hospitalization for heart failure. In practical terms, that means SGLT2 inhibitors were associated with a 37% relative reduction in heart failure hospitalization in the trial evidence used for the model.
By combining absolute baseline risk with the expected treatment effect, SABRE estimates the likely absolute heart failure benefit for an individual patient over 5 years. This approach allows clinicians to move from general population evidence to patient-specific decision support.
Study Design and Validation
The model was tested using United Kingdom primary care data linked to hospital and death records from 2013 to 2020. The study included 57,368 new users of SGLT2 inhibitors and 111,673 new users of comparator drugs, specifically dipeptidyl peptidase-4 inhibitors or sulfonylureas. These comparator treatments are common options in type 2 diabetes management and provided a real-world reference group.
In this observational validation, SGLT2 inhibitor use was associated with a 30% lower risk of new-onset heart failure compared with the comparator drugs, with a hazard ratio of 0.70 and a 95% confidence interval of 0.63 to 0.78. This real-world estimate was consistent with the benefit seen in clinical trials, supporting the reliability of the treatment effect used in the SABRE model.
Importantly, the relative benefit did not differ according to a patient’s baseline absolute heart failure risk. In other words, SGLT2 inhibitors appeared to reduce risk by a similar proportion across risk groups. What changed was the absolute number of heart failure events prevented, which was much larger in people who started with higher baseline risk.
Key Findings
The SABRE model predicted 5-year absolute heart failure benefit ranging from less than 0.1% to 14.1%, with a median benefit of 1.0% and an interquartile range of 0.6% to 1.8%. This wide range shows how differently patients may respond in absolute terms, even when the relative treatment effect is similar.
The model calibrated well against observed outcomes, meaning its predictions closely matched what actually happened in the validation dataset. Good calibration is essential for a clinical prediction tool because it suggests the model is trustworthy when used in practice.
When compared with current guideline-based prescribing, SABRE provided more targeted prevention of heart failure among people with type 2 diabetes who did not already have atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease. This is especially relevant because many patients in routine diabetes care fall into this “no established cardiorenal disease” category, yet may still have meaningful hidden risk.
What the Results Mean for Clinical Practice
This study supports a more individualized way to decide who should receive an SGLT2 inhibitor for primary prevention of heart failure. Rather than prescribing broadly to all patients with type 2 diabetes, clinicians may be able to focus treatment on those with the greatest expected absolute benefit. That could improve the efficiency of care, reduce overtreatment in very low-risk individuals, and help align medication choices with patient values.
For example, a patient with type 2 diabetes, older age, hypertension, obesity, or other risk factors may have a substantially higher 5-year risk of heart failure than a younger person without those features. Even if both patients receive the same relative reduction from an SGLT2 inhibitor, the older or higher-risk patient may avoid more heart failure events over time. SABRE is designed to quantify that difference.
In practice, this kind of model could be integrated into electronic health records or decision-support tools. A clinician could enter routine clinical data and receive an estimate of expected heart failure benefit, helping guide shared decision-making. Such tools are most useful when they are easy to use, transparent, and supported by robust validation.
Clinical Context: Benefits and Considerations of SGLT2 Inhibitors
SGLT2 inhibitors, including empagliflozin, dapagliflozin, canagliflozin, and ertugliflozin, are well known for lowering blood glucose through urinary glucose excretion. Beyond glycemic effects, they have favorable cardiovascular and renal actions. They reduce heart failure hospitalization, may lower blood pressure modestly, and can support weight loss. In people with chronic kidney disease or established cardiovascular disease, their benefits are particularly well established.
As with any medication, there are also potential downsides to consider. These may include genital fungal infections, volume depletion in some patients, and rare but important adverse effects such as diabetic ketoacidosis. Therefore, predicting who is most likely to benefit is not only clinically useful, but may also help ensure that treatment is chosen thoughtfully and appropriately.
Strengths and Limitations
A major strength of this study is its combination of trial evidence and real-world validation. The model was not developed from theory alone; it was built using a validated heart failure risk model and a treatment effect supported by randomized trial data, then checked against a large primary care population.
Another strength is its focus on a group often under-addressed by current guidelines: patients with type 2 diabetes who do not yet have established atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease. This population represents a large proportion of people seen in everyday diabetes care.
There are also limitations to keep in mind. Prediction models depend on the quality of the underlying data, and risk estimates may vary across health systems or populations with different demographics, comorbidities, or prescribing patterns. As with any model, SABRE should complement, not replace, clinical judgment. Patient preferences, medication access, side effect risk, and other treatment goals still matter.
Bottom Line
The SABRE model offers a practical way to personalize SGLT2 inhibitor prescribing for people with type 2 diabetes who do not already have major cardiorenal disease. By estimating both baseline heart failure risk and the expected treatment benefit, it identifies patients most likely to gain meaningful absolute protection against heart failure over 5 years.
In short, this study advances the field of precision diabetes care. It shows how combining validated risk prediction with robust treatment effect data can support more targeted prevention, helping clinicians choose SGLT2 inhibitors for the right patient at the right time.
Citation: Young KG, McGovern AP, Hopkins R, Jansz TT, Cardoso PM, Holman RR, Pearson ER, Hattersley AT, Jones AG, Docherty K, Sattar N, Shields BM, Dennis JM, MASTERMIND Consortium. Precision Prescribing of SGLT2 Inhibitors in Individuals With Type 2 Diabetes for Primary Prevention of Heart Failure: Model Development and Validation Study. Diabetes Care. 2026;49(6):1115-1123. PMID: 42160591.
