Japan tirzepatide dispensing and dose mix

Japan dispensed 7,829,974.7 kit units of tirzepatide in outpatient care during fiscal year 2024, and 86.2% of that volume was handed over somewhere other than a medical institution^[1]. The figure comes from a Cureus paper published on 30 July 2026 by Takuya Omura and Takahiro Kamihara of the National Center for Geriatrics and Gerontology in Obu, working from Japan's national open claims data^[1]. Monthly dispensing rose from roughly 392,000 kit units to 1.03 million between June 2024 and March 2025, and the mix of pen strengths moved upward over the same ten months^[1].

What the claims data actually counted

The source is the National Database of Health Insurance Claims and Specific Health Checkups of Japan, usually shortened to NDB. The Ministry of Health, Labour and Welfare publishes aggregate tables from it as NDB Open Data, downloadable without an application^[2]. Published reviews of the database put its coverage above 98% of all Japanese health insurance claims, which makes it the closest thing the country has to a full national picture of what gets prescribed^[4]. Omura and Kamihara used the injectable-drug quantity tables for fiscal year 2024 and tracked all six Mounjaro strengths across dispensing settings. Their results held in sensitivity analyses that included public-expense claims^[1].

The counting unit matters more than it looks. NDB Open Data reports dispensed quantity, not people^[4]. A patient on a weekly injection contributes several kit units a month, and someone who started and stopped inside the study window is indistinguishable in these tables from someone who never missed a week. Nothing here links a pen to a patient.

Almost every pen is handed over at a pharmacy

Of the total, 6,748,743.5 kit units, or 86.2%, were dispensed outside medical institutions^[1]. In practice that points to community pharmacies rather than clinic or hospital dispensaries. The operational read is that community pharmacists, not clinic staff, are carrying the refrigerated storage and most of the conversation about pen handling and missed doses. Anyone trying to size tirzepatide use in Japan from hospital pharmacy data alone would miss the large majority of it.

The strength mix climbed

Mounjaro is sold in six strengths: 2.5, 5, 7.5, 10, 12.5 and 15 mg per weekly injection^[3]. Across the study window, the share of dispensed quantity accounted for by 7.5 mg and above went from 13% to 24%, while the 2.5 mg share fell from 40.6% to 27.1%^[1]. The 5 mg presentation did most of the heavy lifting in absolute terms, accounting for more than half of the total growth in dispensed quantity^[1].

There is a mechanical explanation worth sitting with before reaching for a behavioural one. The US prescribing information describes 2.5 mg as a starting dosage that is not intended for glycaemic control, with stepwise increases every four weeks as needed^[3]. A market where a lot of people started recently is a market with a heavy 2.5 mg share. Ten months later, if those same people are still on treatment, the mix shows more 5 mg and more 7.5 mg without anybody changing their mind about anything. What the claims tables cannot separate is how much of the shift is a cohort ageing and how much is prescribers titrating differently than they did at launch.

What the study does not show

No patient counts. No HbA1c, no weight change, no adverse events, no discontinuation rates. The authors are explicit that they documented dispensing-setting dynamics and product-mix shifts rather than individual patient outcomes or clinical effectiveness^[1]. The methods literature on NDB Open Data flags the same ceiling: the tables report prescription volume without the number of patients receiving the drug, and the aggregation blocks most of the questions a clinician would want answered^[4]. The same blind spot covers every other injectable incretin sitting in those tables, semaglutide included.

Indication is invisible too. Tirzepatide was approved in Japan in April 2023 for type 2 diabetes, and the Japan Diabetes Society consensus algorithm places it in the obesity and insulin-resistance arm while declining to name it as a candidate in non-obese patients, citing a shortage of data below a BMI of 23^[5]. Quantity tables cannot separate on-label diabetes prescribing from anything else, so these numbers should not be read as a measurement of weight-management use in Japan.

Where this lands

This is a drug-utilisation paper and it earns its keep as one. It tells you where the product physically moves and which pens are moving, which is useful for supply planning and pharmacy capacity, and much less useful for anything clinical. For the clinical side of Japanese real-world use, our write-up of the multicentre retrospective study in routine Japanese practice covers HbA1c and weight outcomes in 324 patients. The background on the molecule itself, including the trial programme and regulatory status, sits on our peptide page, and the closest comparator is semaglutide.

One closing caveat. Dispensing volume is a market signal, not a clinical recommendation. Nothing in these tables says which tirzepatide strength suits any individual, and that decision belongs with a clinician who knows the patient. This article is educational and is not medical advice.