Abstract
Background Limited recent data exist on prescribing patterns and patient characteristics for glucagon-like peptide 1 receptor agonists (GLP-1 RAs), an important drug class used as anti-diabetic medication (ADM) for patients with type 2 diabetes mellitus (T2D) and/or anti-obesity medication (AOM) in patients with overweight or obesity.
For brevity, we use the term GLP-1 RA to refer to both GLP-1 RA and dual GLP-1 RA/GIP medications.
Objective To describe recent trends in prescribing and dispensing of GLP-1-based medications in the US
Methods Using a subset of real-world electronic health record (EHR) data from Truveta, a growing collective of health systems that provide more than 18% of all daily clinical care in the US, we identified people who were prescribed a GLP-1-based medication between January 01, 2018 and December 31, 2023. We describe prescribing volumes and patient characteristics over time, by medication, and by FDA- labeled use. Among the subset of patients for whom post-prescription dispensing data is available, we describe the proportion and characteristics of patients who were and were not dispensed a GLP-1 RA following their prescription.
Results 1,063,200 patients were prescribed a GLP-1 RA between January 2018 and December 2023, with 4,213,888 total prescriptions during this period. Among first-time prescriptions for which use could be established, ADMs accounted for 86.6% and AOMs accounted for 13.4%. The most common first-time medication was semaglutide (n = 559,880). Among those with available data, 71.5% of first-time ADM prescriptions overall and 74.9% in September 2023 had a fill within 60 days. By comparison, 45.6% of first-time AOM prescriptions overall and 32.1% in September 2023 had a fill within 60 days.
Trends in Prescribing
Overall prescribing rates (GLP-1 RA prescriptions per total prescriptions) in December 2023 remained approximately the same as rates in November 2023 and October 2023.
The number of patients prescribed an ADM in December 2023 remained approximately the same as the number in both November 2023 and October 2023.
Among ADMs, semaglutide was the most prescribed medication. In late 2023, tirzepatide surpassed dulaglutide as the second most commonly prescribed ADM.
The number of patients prescribed an AOM in December 2023 was 1.2 times November 2023 and 1.1 times October 2023.
Semaglutide was the most commonly prescribed AOM, followed by AOM liraglutide. Prescribing of AOM tirzepatide was observed in the first month of availability in pharmacies, December 2023.
Trends in First-Time Prescribing
First-time prescribing rates (first-time GLP-1 RA prescriptions per total prescriptions) in December 2023 were slightly lower than rates in November 2023 and October 2023.
First-time prescribing of ADMs was lower in the most recent month, with first- time prescribing rates in December 2023 being 0.8 times that of November 2023.
Similar to overall prescribing, semaglutide was the most commonly first- prescribed GLP-1 RA overall and in the most recent period.
First-time prescribing of AOMs was similar in the most recent month, with first- time prescribing rates in December 2023 being 1.0 times that of November 2023.
While overall prescribing of semaglutide increased throughout 2023, first-time prescribing of semaglutide slowed after the first quarter of 2023 from 0.9% of prescriptions in May 2023 to 0.5% in December 2023. Similar declines were observed for ADM, AOM, and unknown semaglutide. In contrast, new prescribing of tirzepatide increased throughout 2023.
Trends in Dispensing
During the full time period, 74.2% of GLP-1 RA prescriptions had a dispense within 60 days of their prescription.
60-day initiation rates were higher for ADMs compared to AOMs. 71.5% of patients first prescribed an ADM overall and 74.9% in September 2023 filled a GLP-1 RA within 60 days, compared to 45.6% of patients first prescribed an AOM overall and 32.1% in the September 2023.
60-day AOM initiation rates for the 65+ population (23.6% in September 2023) are lower than AOM initiation rates for the population aged 18-64 (32.9% in September 2023).
About this report
This report provides recent trends in the prescribing and dispensing of glucagon-like peptide-1 receptor agonist (GLP-1 RAs)-based medications, including dual GLP-1 RA/gastric inhibitory polypeptide (GIP) over time, by medication, and by FDA-labeled use. We used near real-time, aggregated electronic health record (EHR) data from a collective of US healthcare systems to describe prescribing and dispensing of GLP-1 RA medications from January 2018 to December 2023.
Introduction
Obesity and type 2 diabetes mellitus (T2D) are interconnected epidemics with a high prevalence and individual and combined negative impacts on health (CDC, 2020, Hales, 2020). Glucagon-like peptide-1 receptor agonist (GLP-1 RAs) -based medications, including dual GLP-1 RA/gastric inhibitory polypeptide (GIP), have been shown to improve glycemic control in patients with T2D and reduce weight in patients with and without T2D (Davies, 2015, Davies, 2021, Garvey, 2023, Jastreboff, 2022, Wilding, 2023, Yeh, 2023). GLP-1 RA have also demonstrated cardiovascular benefit in patients with and without T2D (Lincoff, 2023, Ussher, 2023).
Interest in these medications has recently accelerated, largely for their weight-loss effects; however, limited empiric data exist on prescribing, dispensing, and patient characteristics. Access to and use of GLP-1 RA may be impacted by high cost, limited insurance coverage for patients without T2D and medication shortages.
Methods
Data
We evaluated prescribing and dispensing of GLP-1 RAs between January 2018 and December 2023 using a subset of Truveta Data. Truveta provides access to continuously updated, linked, and de-identified electronic health record (EHR) from a collective of US health care systems that provide 18% of all daily clinical care in the US, including structured information on demographics, encounters, diagnoses, vital signs (e.g., weight, BMI, blood pressure), medication requests (prescriptions), medication administration, laboratory and diagnostic tests and results (e.g., HbA1c tests and values), and procedures. Updated EHR data are provided daily to Truveta by member health care systems. In addition to EHR data for care delivered within Truveta member health care systems, medication dispensing and social drivers of health (SDOH) information are made available through linked third-party data. Medication dispense (via e-prescribing data) includes fills for prescriptions written both within and outside Truveta member health care systems, providing greater observability into patients’ medication history. Medication dispense histories are updated at the time of the encounter, and include fill dates, NDC or RxNorm codes, quantity dispensed, and days of medication supplied. SDOH data include individual-level factors, including income and education.
Population
We identified GLP-1 RA prescriptions (medication requests) and dispenses among adults using RxNorm and NDC codes for dulaglutide, exenatide, liraglutide, lixisenatide, semaglutide, and tirzepatide. Exenatide and lixisenatide were grouped as ‘other’ given small relative sizes. Brands were identified when possible using RxNorm and NDC codes. Brands labeled for treatment of T2D were classified as ADMs and brands labeled for treatment of overweight and obesity were classified as AOMs. When brand could not be established, the labeled use was classified as unknown.
Prescribing Analysis
Characteristics of adult patients ever prescribed a GLP-1 RA were summarized, including demographics, social drivers of health (SDOH), and presence of T2D and/or overweight or obesity. Characteristics are provided by medication and by year of first prescription.
Overall and first-time prescribing rates of GLP-1 RA were summarized over time by medication and labeled use. The prescribing rate was calculated monthly as the number of patients with a GLP-1 RA prescription divided by the number of patients with any prescription in the month. Absolute differences and percentage changes in the rate of GLP-1 RA prescribing over the most recent 3-month period were summarized overall and by medication and labeled use.
Dispensing Analysis
We describe the proportion of GLP-1 RA prescriptions (overall and first-time) with a dispense occurring within 60 days. Dispensing analyses were restricted to prescriptions for which complete 60-day dispensing outcomes could be observed. Dispensing outcomes were considered observable if (1) a dispense was recorded in the 60 days following the prescription, and/or (2) an encounter took place after the 60- day period ended (at which time any dispenses would be observed). To allow sufficient follow-up, we report dispensing with a 90-day delay. Given shortages and expected substitutions at the pharmacy, we considered the dispense of any GLP-1 RA, which could differ from the prescribed GLP-1 RA.
We plot the proportion of prescriptions (overall and first-time) dispensed within 60 days over time, by medication and use. Differences in dispensing were expected between ADMs and AOMs given ADMs are typically covered by insurance, whereas insurance coverage for AOMs is limited. As such, we provide dispensing rates stratified by FDA labeled use. Further, we stratify dispensing by age 65+ vs. 18-64 as a proxy for Medicare coverage. Medicare never provides coverage for AOMs, whereas commercial insurers may.
Results
Overall population
The demographics of patients by first-prescribed medication are as follows:
Prescribing over time
Overall
We describe the monthly rate of prescriptions by medication and FDA labeled use over time:
First-time
We describe the monthly rate of first-time prescriptions by medication and FDA labeled use over time.
Characteristics of patients by year of first GLP-1 RA use are as follows:
60-day dispensing over time
We describe the proportion of patients with any GLP-1 RA prescriptions (first-time and repeat) in a given month who filled any GLP-1 RA in the 60 days following their prescription.
60-day initiation over time
We describe the proportion of patients first-prescribed a GLP-1 RA who filled any GLP- 1 RA in the subsequent 60 days over time.
Patient characteristics are given in Table 3:
By Age
As a proxy for Medicare insurance, we segment initiation rate by populations 65+ versus 18-64.
Patient characteristics for patients aged 65+ are given in Table 4:
Patient characteristics for patients aged 18-65 are given in Table 5:
Discussion and Conclusion
This analysis summarized GLP-1 RA prescribing trends and patient characteristics from January 2018 to December 2023. During this period, 1,063,200 patients were prescribed a total of 4,213,888 total prescriptions. ADMs accounted for 90.4% of prescriptions, and AOMs accounted for 9.6%, among prescriptions where use was known. A larger proportion of ADMs, compared to AOMs, were dispensed within 60 days: 71.5% of first-time ADM prescriptions and 45.6% of first-time AOM prescriptions had a fill within 60 days.
Several limitations of our analysis should be noted. All data are preliminary and may change as additional data are obtained. These findings are consistent with data accessed January 15, 2024. Data presented in this analysis represent raw counts and/or rates, and post-stratification methods have not been conducted. In addition, cohorts with small counts may be suppressed during the de-identification process leading to the appearance of zero patients for a given time period. Given relatively large numbers of patients prescribed and dispensed GLP-1 RA, we do not expect substantial suppression in the most recent periods. The unknowns in this report either indicate the value was not included in the individual’s electronic health record or that it was excluded from the data to protect an individual’s identity as a part of Truveta’s commitment to privacy (Truveta, 2024). Finally, this report is a description of prescribing trends and patient characteristics only. Results should not be interpreted as causal.
Data Availability
The data used in this study are available to all Truveta subscribers and may be accessed at studio.truveta.com.
Acknowledgements
The authors thank Dr. Ty Gluckman for his contributions and feedback on describing GLP-1 RA prescribing trends.
