The NHIS established the National Health Information Database (NHID) (http://nhiss.nhis.or.kr) in 2012. The NHID comprises five databases: the entitlement database, the national health screening database, the healthcare utilization database based on NHI claims, the long-term care insurance database, and the healthcare provider database [21]. The NHI claims data accumulate information on the basis of medical records, covering all services that medical institutions and pharmacies provide to the insured.

We extracted 10-year data, covering the period of 2013–2022, including substance name, dosage, dosage form, unit price, quantity prescribed, drug costs, type of medical institution (clinic, hospital, general hospital, university hospital), ambulatory or inpatient service, International Classification of Diseases (ICD-10, tenth revision) code, visit date, and patient information, as recorded in the NHI claims prescribing database.

As presented in Table 1, 15 substances, including 11 single substances and 4 fixed combinations, were available in Korea at the time of interest. Omeprazole was approved in 1990, but potassium-competitive acid blockers (P-CABs) were launched more recently, starting with revaprazan in 2006, followed by tegorpazan in 2019, and fexuprazan in 2022. All relevant data for the 10 years under study were extracted from the NHID. Compared with single forms, fixed combinations showed no differences in reducing heartburn and acid regurgitation or in safety profiles among patients with non-erosive reflux disease [22].

This study evaluated PPI prescribing patterns over 10 years, spanning 2013–2022, and analyzed the annual trends in drug spending, utilization volume, unit prices of each substance, and proportion of each PPI in the ambulatory setting. Changes in utilization trends and spending were also observed over time.

Drug costs are incurred when physicians issue an ordonnance. Therefore, we collected annual PPI drug costs from medical claims data. These costs represent the amounts paid by the NHIS, the single payer, to healthcare providers after review and approval by the HIRA. However, the data may not capture 100% of the actual prescriptions, since physicians tend to prescribe medicines as extra billings. This portion is assumed to be minimal. According to the NHIS report, the extra billing rate was found to be 2.8% in 2021 [23].

PPI utilization was analyzed on the basis of the WHO’s defined daily doses (DDD) (https://atcddd.fhi.no) for omeprazole (20 mg), esomeprazole (30 mg), pantoprazole (40 mg), lansoprazole (30 mg), dexlansoprazole (30 mg), and rabeprazole (20 mg). The daily maintenance doses for S-pantoprazole, ilaprazole, fexuprazan, revaprazan, and tegoprazan, as well as the four fixed combination doses, were calculated on the basis of the general principles for DDD assignment by the WHO (https://atcddd.fhi.no/ddd/definition_and_general_considera/).

The actual purchase prices for each pharmaceutical, which are lower than the MRPs, were used for analysis. The unit price was calculated by dividing the cost of each substance by the total volume of each substance in the DDD.

We selected a first-choice PPI on the basis of cost-effectiveness, prioritizing the option with the lowest cost under the assumption of identical clinical effectiveness of PPIs. To estimate potential cost savings, we utilized the Dirichlet distribution, a statistical model that generates vectors of parameters that sum to 1. Thus, the proportion of each PPI was randomly generated between 0 and 1, and the sum of these proportions equaled 1 [24]. Using these proportions of each PPI, potential cost savings were calculated by subtracting the estimated total cost from the current expenditure. Thus, the distribution with the largest proportion of first-choice PPI resulted in the largest potential savings. For example, given a variation in the share of the lowest cost PPI from its current share to the maximum of 100%, the remaining share could be distributed to the remaining thirteen PPIs using the Dirichlet method, and then we could calculate the potential savings. The best scenario, which maximizes the potential cost savings, is to allocate 100% of the market share to the first-choice of PPI on the basis of its cost-effectiveness.

Drug spending, unit price, and cost savings were reported in South Korean won (KRW) (1 USD Purchasing Power Parity [PPP] = 810.43 KRW as of 2022) [25]. SAS software (version 9.4) was used for analysis. Cost simulations were performed using Microsoft Excel.