The demographic profile of those starting the drug was consistent across both periods. Mean age was 59 ± 12 years in the early cohort and 60 ± 12 years in the later one, roughly 54% of individuals in each group were male, and more than 90% were covered by public insurance. Comorbidities were frequent, with hypertension affecting more than 70% of individuals in both cohorts and dyslipidaemia more than 40%.

Kidney disease was recorded in 24.7% of early initiators compared with 14.6% of later initiators, a difference the authors suggest may reflect the tendency of specialists to act as early adopters of newly launched therapies, treating patients with more advanced or poorly controlled disease first. That specialist-led early adoption was also visible in the prescribing data. In the early cohort, records from 223 physicians contributed, of whom 145 (65.0%) were general practitioners and 71 (31.8%) were diabetologists; diabetologists accounted for nearly two-thirds of index prescriptions in that period.

In the later cohort, data came from 512 physicians, with general practitioners representing 80.5% and diabetologists 16.6%. Clinically, most individuals arrived at tirzepatide with suboptimal glycaemic control and excess weight. HbA1c measurements taken within six weeks before initiation were available for 43.3% of the early cohort and 35.7% of the later cohort; among those with data, the majority exceeded 7.5%, with mean values of 7.9% and 7.8% respectively.

Obesity, defined as a body mass index above 30 kg/m², was present in 67% of the early cohort and 55% of the later cohort. Where liver and kidney parameters were recorded, median values fell within normal ranges, including an estimated median glomerular filtration rate of approximately 83 mL/min/1.73 m². The authors note that the study population's age, sex distribution and obesity prevalence were consistent with the broader German population of people with type 2 diabetes.

Prior treatment history emerged as one of the study's most striking findings. Among individuals with a complete six-month baseline record, more than 80% had received prior antidiabetic treatment, and the majority had been treated with insulin and/or a glucagon-like peptide 1 receptor agonist before switching to tirzepatide: 72.1% in the early cohort and 59.6% in the later cohort. Only 17.8% of the early cohort and 23.5% of the later cohort initiated tirzepatide as first-line therapy or as second-line treatment after metformin monotherapy.

In addition, 43.3% and 50.2% of the two cohorts had previously received antihypertensive medication, while roughly a third had received lipid-lowering therapy. The authors interpret the substantial proportion switched from GLP-1 receptor agonists as suggesting ongoing difficulties achieving glycaemic targets, consistent with the elevated baseline HbA1c values observed. To examine how dosing evolved, the researchers analysed a subgroup of 458 individuals from the early cohort who had at least six months of follow-up and uninterrupted tirzepatide prescriptions, defined as no more than six weeks between the calculated end of one prescription and the next.

The overwhelming majority, 71.2%, began at the 2.5 mg starting dose, in line with the product label's titration instructions, although higher starting doses of 5.0 mg and 7.5 mg were also documented. The authors suggest these higher initial doses may reflect physicians' choices when switching patients already exposed to GLP-1 receptor agonists, or possible misclassification if initial prescriptions issued by non-panel physicians were not captured in the database. Over the six-month observation window, dose escalation followed the labelled titration schedule but remained conservative.

The 5.0 mg dose became the most prevalent early on, taken by 68.1% of the subgroup by month two and 35.9% by month six. Use of higher doses of 7.5 mg, 10.0 mg, 12.5 mg and 15.0 mg increased gradually over time, and by month six 18.6% of individuals were receiving 10.0 mg. Nevertheless, more than half of the subgroup remained on doses below 10.0 mg throughout the first six months, with 5.0 mg the single most common maintenance dose.

The most frequent escalation step, from 2.5 mg to 5.0 mg, occurred in 70.5% of individuals, and escalation from 5.0 mg to 7.5 mg was the second most common. De-escalation was rare, with the most frequent pattern, a reduction from 7.5 mg to 5.0 mg, seen in only 9.8% of the subgroup. Concomitant antidiabetic therapy remained common alongside tirzepatide.

Among the 458 individuals with uninterrupted six-month use, 449 had a sufficient look-back period for prior medication assessment; 88.2% of these had received prior antidiabetic treatment, and 32.6% had been on insulin with or without a GLP-1 receptor agonist before starting tirzepatide. During follow-up, 77.7% of the subgroup received concomitant antidiabetic medication, with 45.9% taking only oral antidiabetic drugs and 31.9% continuing insulin-based therapy. These figures indicate that in routine German practice, tirzepatide was typically added to existing regimens rather than replacing them outright, particularly among patients previously treated with insulin.

The study's context is the growing global burden of type 2 diabetes, which accounted for an estimated 589 million adults aged 20 to 79 years in 2024, a figure projected to reach 853 million by 2050, with type 2 diabetes representing more than 90% of cases. In Germany, prevalence is estimated at approximately 7 to 12%. Despite guideline recommendations for multifactorial management targeting both glucose and weight, prior studies indicate that only 33 to 51% of people achieve glycaemic goals, typically an HbA1c of 7% or below, and fewer than half achieve weight loss of at least 5% of body weight.

Tirzepatide, approved by the European Medicines Agency in 2022 for insufficiently controlled type 2 diabetes and in 2023 for obesity or overweight with a weight-related comorbidity, demonstrated significant HbA1c and body weight reductions versus placebo, semaglutide 1 mg, and titrated insulin degludec or glargine in the SURPASS registrational trials, with meta-analyses and secondary analyses suggesting additional benefits for renal function, liver-related laboratory parameters and cardiovascular safety. The authors emphasise that the real-world cohorts overlapped demographically with the SURPASS trial populations in age, sex and comorbidities, supporting the relevance of the trial findings to German clinical practice. They also caution that the analysis has important limitations: the database captured only panel physicians, so prescriptions from non-panel physicians were missed; laboratory values were frequently missing and were not imputed, with results reported descriptively on observed data only; prescription records reflect prescribing rather than dispensing or adherence; and the six-month window may have missed later dose changes.

No clinical outcome data were available to inform dose-selection decisions, and no sex- or gender-based analysis was performed. As a purely descriptive study without inferential testing, the findings characterise early adoption and utilisation patterns rather than treatment effectiveness, and the authors call for further research into tirzepatide prescribing patterns in Germany.

Subject of Research: Real-world utilisation patterns of tirzepatide among adults with type 2 diabetes in German office-based practice

Article Title: Characteristics and Treatment Utilisation Patterns of Individuals Using Tirzepatide for Type 2 Diabetes in Germany

Article References: Füchtenbusch, M., Kraus, M., Heitmann, E., Wilhelm, S., Schwarz, L., Bendel, J., & Hennies, N. (2026). Characteristics and Treatment Utilisation Patterns of Individuals Using Tirzepatide for Type 2 Diabetes in Germany. Advances in Therapy. https://doi.org/10.1007/s12325-026-03767-1

Image Credits: AI Generated

DOI: 10.1007/s12325-026-03767-1

Keywords: tirzepatide, type 2 diabetes, GLP-1 receptor agonist, real-world evidence, Germany, dosing patterns, electronic medical records, observational study, obesity, HbA1c, insulin, pharmacoepidemiology