GLP-1 drugs: how background therapy shifts

When a patient with type 2 diabetes starts a GLP-1 receptor agonist or a dual GIP/GLP-1 receptor agonist, what happens to the other glucose-lowering drugs on the prescription? A 2026 observational study by Dziewierz and colleagues, published in Diabetes Research and Clinical Practice, tracked that question across 3,660 real-world patients in Poland^[1]. The short answer is: a higher background medication count was not inevitable. In fact, more than a third of patients saw no change at all, and about one in six had the number of background agents reduced.

What the study measured

The research team pulled electronic health records for adults with type 2 diabetes who initiated a GLP-1 receptor agonist or a dual GIP/GLP-1 receptor agonist between 2022 and 2025. The primary question was whether the recorded count of background glucose-lowering substances increased, decreased, or stayed the same after the new agent was added. Drugs in the analysis included metformin, SGLT-2 inhibitors, DPP-4 inhibitors, sulfonylureas, and basal insulin, among others^[1].

The 3,660 patients had a median age of 60 years; 53.9% were female. Background medication count at baseline was a median of one substance.

The three-way split

After initiation of the incretin-class agent, the researchers observed three distinct patterns^[1]: 44.0% of patients had an increase in recorded background therapy, 38.8% showed no change, and 17.2% had a reduction. The median count rose from one to two substances, but because fewer than half of patients drove that shift, the median alone understates how varied the prescribing response was.

Across the specific agents studied, the pattern differed by drug. Dulaglutide initiators showed the highest rate of background therapy increase at 52.9%. Tirzepatide initiators showed the lowest at 20.4%^[1]. The authors caution that these figures describe prescription-record patterns only; they do not establish concomitant exposure, treatment intensity, or a causal relationship.

Why tirzepatide stands out in the data

The 20.4% increase rate for tirzepatide is meaningfully lower than for the GLP-1-only agents in the same dataset. One plausible interpretation is that tirzepatide's dual GIP and GLP-1 receptor activity produces enough glycaemic improvement, on its own, that clinicians feel comfortable holding or reducing concurrent agents rather than intensifying the full regimen. In the SURPASS-2 trial, tirzepatide at all three doses produced greater HbA1c reductions than semaglutide 1 mg at 40 weeks^[2]. Larger glycaemic drops in trials may translate to less perceived need to pile on additional agents in practice.

That said, the observational design means confounding is possible. Tirzepatide was the most recently approved agent in the cohort, and newer prescriptions are often for patients who were already on multiple agents, leaving less room to add more. Prescriber experience with a newer drug may also be different from experience with agents in wider circulation for longer.

What deprescribing means here

The 17.2% reduction rate matters clinically. For patients who had been on sulfonylureas or insulin before starting an incretin-class drug, a clinician removing one of those agents is reducing hypoglycaemia risk and pill burden at the same time. This fits a pattern visible in other real-world datasets on GLP-1 class agents: effective glycaemic control through the new agent sometimes creates space to simplify the regimen^[3].

The study does not distinguish between deprescribing that was planned (a clinician replacing a sulfonylurea with tirzepatide as a deliberate therapeutic swap) and deprescribing that was reactive (a patient stopping a medication because of side effects or cost, with the prescriber recording the change). Both scenarios appear in real-world prescribing data, and the Dziewierz dataset cannot separate them.

Limitations worth knowing

The study describes recorded prescriptions, not confirmed dispensing or patient adherence. A prescription reduced or removed from the electronic record may reflect a prescriber's intention, a pharmacy substitution, or a patient's own decision. The follow-up window and exact timing of the medication counts relative to the incretin initiation date are not specified in the abstract. And because the cohort is Polish, patterns may differ in health systems with different prescribing cultures, formulary structures, or access to newer agents.

The authors are explicit that the findings do not establish causation and should be read as hypothesis-generating for prospective work. Longer follow-up, with adherence confirmation and HbA1c trajectories linked to the prescribing changes, would be needed to draw firmer conclusions.

What does not yet have an answer

The study cannot say whether the pattern of background therapy change affected glycaemic outcomes, rates of hypoglycaemia, or long-term cardiovascular endpoints. It also does not address what happens at one or two years: whether the 38.8% who saw no change will eventually have agents added, or whether the 17.2% who had reductions will need reinstatement. Those are the questions that a prospective follow-up study would be well placed to answer.

For patients and clinicians

If you are starting or about to start a GLP-1 or GIP/GLP-1 receptor agonist for type 2 diabetes, background therapy adjustment is a clinical decision specific to your HbA1c trajectory, hypoglycaemia risk, and existing regimen. A prescriber using real-world patterns like those in this study might use them to set expectations for follow-up reviews. They do not translate directly into a recommendation for any individual patient. For context on how tirzepatide compares to GLP-1-only agents on glycaemic outcomes, see the tirzepatide overview and the piece on GLP-1 cardiovascular outcomes in type 2 diabetes.

Tirzepatide and the GLP-1 receptor agonists discussed in this study are prescription-only medicines in all countries where they are approved. Consult a qualified healthcare professional before making any changes to a diabetes treatment regimen. For country-specific prescribing rules, see the tirzepatide regulation section.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.