Increasing Prescription of SGLT2 Inhibitors with Expanded Indications to the Elderly Population in Japan Open Access

Sodium-glucose cotransporter-2 (SGLT2) inhibitors have been used for patients with type 2 diabetes (T2D) since 2012 in Europe, 2013 in the USA, and 2014 in Japan. Although SGLT2 inhibitors were initially developed and marketed as glucose-lowering agents, subsequent clinical trials observed that SGLT2 inhibitors improve cardiovascular and kidney outcomes in patients with T2D [1‒5]. Randomized clinical trials (RCTs) on patients with heart failure showed that dapagliflozin [6, 7] and empagliflozin [8, 9] decreased cardiovascular deaths. Likewise, RCTs on patients with chronic kidney disease (CKD) showed that canagliflozin (in patients also with T2D) [10], dapagliflozin [11], and empagliflozin [12] have kidney-protective effects. Given these findings, indications for certain SGLT2 inhibitors were expanded to include a reduction in the risk of cardiovascular outcomes in patients with heart failure since 2020 and a reduction in the risk of kidney outcomes in patients with CKD with or without T2D since 2021. Clinical guidelines recommend the use of SGLT2 inhibitors for heart failure since 2021 [13‒15] and for CKD with albuminuria since 2023 [16, 17].

Studies have shown that the number of SGLT2 inhibitor prescriptions has increased since they were released in the market [18‒22], but whether and how the demographic trends of the prescription of SGLT2 inhibitors have changed after the expansion of their indications for patients with heart failure and CKD after the year 2020 have not been studied extensively. Given that some clinical trials for SGLT2 inhibitors on patients with heart failure observed a larger share of the elderly population than other trials (online suppl. Table 1; Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000545626), the age distribution of SGLT2 inhibitor recipients might have shifted toward older ages after the expansion of their indications. Moreover, very few studies focused on the dosage of SGLT2 inhibitors prescribed to patients. Analyzing the prescription trends of SGLT2 inhibitors by their dosage may elucidate any changes in clinical practice after the expansion of their indications because the approved dosage of SGLT2 inhibitors for heart failure and CKD does not exactly coincide with that for T2D in Japan and elsewhere. Currently approved SGLT2 inhibitors, their tablet forms, and indications in Japan are listed in Table 1.

In this article, we sought to answer these questions by analyzing data from the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB), which is a government database that collects more than 95% of the total health insurance reimbursement claims in Japan [23]. Because Japan adopts a multi-payer universal health coverage system, NDB is the closest to a complete dataset of medical procedures performed in Japan. We have previously reported multiple analyses on medical procedures performed in the nation by analyzing NDB Open Data Japan [24‒28], which is an openly accessible, processed demographic dataset derived from the NDB but with no data attributable to an individual person or healthcare facility. Using this database, we analyzed whether and how the prescription pattern of SGLT2 inhibitors changed after the expansion of their indications.

This is a descriptive analysis of serial, cross-sectional data derived from NDB Open Data Japan. The numbers of SGLT2 inhibitor tablets prescribed in Japan between April 2016 and March 2023 were obtained from the 3rd to 9th NDB Open Data Japan datasets [29]. The number of prescribed tablets in each age group and sex is available from this database for the top 100 prescribed medicines (or the top 300 prescribed medicines for the 9th Open Data Japan dataset) among all oral anti-diabetic medications sold in Japan. Some numbers in the data table were hidden because numbers larger than zero and smaller than 1,000 are concealed in the tables in these datasets for the sake of privacy. The possible maximum and minimum value for each hidden number was calculated and used in subsequent analyses to evaluate the robustness of the analyses.

The number of tablets with a different dose of the same ingredient was counted and analyzed separately in most analyses because some SGLT2 inhibitors have different approved doses for different indications. For example, the use of empagliflozin 10 mg was approved for patients with heart failure with reduced ejection fraction in November 2021 (fiscal year [FY] 2021), patients with heart failure regardless of ejection fraction in April 2022 (FY 2022), and patients with CKD in February 2024 (FY 2023; prescription data not available yet). Likewise, the use of dapagliflozin 10 mg was approved for patients with heart failure with reduced ejection fraction in November 2020 (FY 2020), patients with CKD in August 2021 (FY 2021), and patients with heart failure regardless of ejection fraction in January 2023 (FY 2022). In contrast, the use of empagliflozin 25 mg and dapagliflozin 5 mg is approved only for treatments to improve glycemic control in patients with diabetes mellitus.

The population of each age and sex was obtained from the results of the Surveys of Population, Population Change and the Number of Households Based on the Basic Resident Registration, which summarizes the demographic data on all residents in Japan on January 1st every year [30].

Estimated numbers of patients with diagnosed diabetes mellitus, cardiovascular disease, and CKD in Japan were obtained from the results of the Patient Survey conducted by the Ministry of Health, Labour and Welfare of Japan in 2020 [31]. The survey collected demographic and medical data on a total of 3.1 million patients who received medical care in either outpatient or inpatient settings on a single day in hospitals and clinics which were selected by stratified random sampling. The response rate was 96.7% on an institutional basis [31].

The R system for statistical computing (version 4.4.0, R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis.

In NDB Open Data Japan datasets, the number of prescribed tablets is summarized by three different prescription categories: inpatient prescriptions, outpatient prescriptions with direct dispensing of tablets by hospitals and clinics, and outpatient prescriptions with prescription papers to be brought to pharmacies. The total number of SGLT2 inhibitor tablets prescribed between April 2022 and March 2023 was as follows: inpatient, 8,351,533 tablets (0.8%); outpatient – direct dispensing, 145,137,998 tablets (13.7%); outpatient – prescription papers, 904,598,175 tablets (85.5%). Tables for the former two categories had a substantial amount of hidden data for privacy protection. Therefore, we analyzed outpatient prescriptions with prescription papers in this study.

Figure 1 shows the number of SGLT2 inhibitor-containing tablets prescribed with prescription papers in outpatient settings in Japan from FY 2015 to 2022. FY runs from April 1 to March 31 in Japan. All SGLT2 inhibitors observed a sustained increase in the number of prescribed tablets. The total number of prescribed tablets increased from 577,996,158 tablets in FY 2020 to 904,598,175 tablets in FY 2022.

Changes in the prescriptions of SGLT2 inhibitors between FY 2020 and 2022 are shown in Figure 2. Among all SGLT2 inhibitors, tablets with expanded indications for heart failure and/or CKD with or without T2D — namely, empagliflozin 10 mg and dapagliflozin 10 mg tablets — showed the largest increase in the number of prescribed tablets between FY 2020 and 2022. The number increased from 123,211,555 to 208,013,330 tablets for empagliflozin 10 mg tablet, and from 11,323,118 to 94,070,148 tablets for dapagliflozin 10 mg tablet. In addition, the latter, which expanded its indication the earliest, showed the largest percentage increase in the number of prescribed tablets during this period. In contrast, other dosage forms of these two agents — namely, empagliflozin 25 mg and dapagliflozin 5 mg tablets — are approved only for treatments to improve glycemic control in patients with diabetes mellitus and did not observe a comparable increase in the number of prescribed tablets.

In FY 2022, the main target of prescription was between the ages of 40 and 90 (Fig. 3a), which accounted for 96.9% of the total prescription of SGLT2 inhibitors. Patients aged between 70 and 74 years received the largest number of SGLT2 inhibitor tablets among all 5-year age groups (Fig. 3a). The tablet that expanded its indications for patients with heart failure and CKD the earliest — namely, dapagliflozin 10 mg tablet — was notable for substantially increasing its share as the age increases from 70 years (Fig. 3b). Figure 3c shows the number of prescribed tablets divided by the population of each age group between FY 2020 and 2022. The age of patients with the highest number of prescribed tablets per population had shifted toward older ages between FY 2020 and 2022 (Fig. 3c). The share of each tablet is illustrated in online supplementary Figure 2.

The number of prescribed tablets in each FY for patients in each age group is illustrated in online supplementary Figure 3. The proportion of tablets prescribed for patients in each age group to the total prescribed tablets in each FY is presented in online supplementary Figure 4. These figures show that all SGLT2 inhibitors observe an increasing share of the elderly population in their recipients.

The proportion of SGLT2 inhibitor tablets prescribed for patients aged 75 years and older has been increasing regardless of their dosage or ingredient (Fig. 4). In total, the proportion of SGLT2 inhibitor tablets prescribed to patients aged 75 years and older increased from 20.3% in FY 2020 to 23.2% in FY 2021 and 27.8% in FY 2022. During this period, the SGLT2 inhibitor that expanded its indication for both heart failure and CKD the earliest, which was dapagliflozin 10 mg tablet, showed the largest and steepest increase in its share of patients aged 75 years and older, which increased from 18.6% in FY 2020 to 34.8% in FY 2021 and 40.2% in FY 2022 (Fig. 4). The second SGLT2 inhibitor that expanded its indication, which was empagliflozin 10 mg tablet, also showed a large increase in its share of patients aged 75 years and older, which increased from 22.7% in FY 2021 to 29.6% in FY 2022 (Fig. 4). These increases occurred after these two tablets expanded their indications (Table 1).

The number of prescribed tablets per population in men was twice as high as that in women (Fig. 5). The imbalance was most prominent in patients aged 40–54 years and has been increasing after FY 2020 in patients aged 60–84 years (Fig. 5b). According to the national Patient Survey, the prevalence of diabetes mellitus, heart failure, and CKD is estimated to be higher in men than in women in most age groups (online suppl. Fig. 5) [31].

The proportion of SGLT2 inhibitor tablets prescribed for patients aged 75 years and older was higher in women than in men (Fig. 6). During the 2-year period from 2020, the SGLT2 inhibitor that expanded its indication for both heart failure and CKD the earliest — namely, dapagliflozin 10 mg tablet — observed the largest increase in its share of patients aged 75 years and older, which was 35.9% in men and 49.4% in women in FY 2022 (Fig. 6). The second SGLT2 inhibitor that expanded its indication — namely, empagliflozin 10 mg tablet — also showed a large increase in its share of patients aged 75 years and older, which was 25.9% in men and 36.7% in woman in FY 2022, after its indication expansion (Fig. 6).

The proportion of SGLT2 inhibitor tablets prescribed for patients aged 60 years and older also showed the same trend (online suppl. Fig. 6, 7), but its change was generally smaller than the change in the proportion of tablets prescribed to patients aged 75 years and older. Patients aged 75 years and older reflected a population whose prescription status shifted with a greater extent during the observation period.

This study revealed that SGLT2 inhibitors were increasingly prescribed to patients of all ages but especially to the elderly population between FY 2020 and 2022 in Japan (Fig. 3c). Among all SGLT2 inhibitors, the SGLT2 inhibitor that expanded its indication for both heart failure and CKD the earliest showed the largest percentage increase in the number of prescribed tablets (Fig. 1, 2) and the highest share of the elderly population in its recipients in both sexes (Fig. 4, 6). This study also showed that the number of prescribed SGLT2 inhibitor tablets per population was consistently higher in men than in women between FY 2020 and 2022 (Fig. 5), which is consistent with the sex difference in the prevalence of T2D, heart failure, and CKD in Japan (online suppl. Fig. 5), as well as the likelihood of not having urinary tract infections regardless of SGLT2 inhibitor medications [1].

The sheer growth of the use of two SGLT2 inhibitor tablets with expanded indications for heart failure and/or CKD with or without T2D after FY 2020 was a change peculiar to these agents among all SGLT2 inhibitors (Fig. 2). The indication expansion may have contributed to the sheer growth of their use. The high proportion of these tablets prescribed to patients aged 75 years and older could also have been caused by the expansion in its indications, but we could not prove this directly because NDB Open Data Japan datasets did not show which indication each SGLT2 inhibitor tablet was prescribed for.

The proportion of SGLT2 inhibitor tablets prescribed to the elderly population had been consistently increasing even before the year 2020 (Fig. 4, 6), which suggests that this sustained increase is caused by factors other than the expansion in the indications for the use of SGLT2 inhibitors. The constant increase in the proportion of the population aged 75 years and older (13.1% in FY 2016 to 15.4% in FY 2022) in Japan [30] may partially explain this increase.

After the clinical efficacy of SGLT2 inhibitors became evident after multiple RCTs [1‒12, 32], the risks and benefits of SGLT2 inhibitors in the elderly population have been questioned and discussed extensively. Multiple post hoc subgroup analyses of the RCTs observed no age-dependent heterogeneity in the efficacy as well as in the frequency of adverse events that were more common in patients treated with SGLT2 inhibitors than placebo [33‒37]. Some studies found an increased risk of volume depletion [38, 39] and diabetic ketoacidosis [40] in elderly patients treated with SGLT2 inhibitors than placebo or other diabetic medications, but the absolute risk was low. Despite these latest findings, we speculated that the frequency of SGLT2 inhibitors to the elderly population was low because current official prescribing information for SGLT2 inhibitors in Japan, the USA, and elsewhere note the risk of volume depletion in elderly patients. Moreover, the Japan Diabetes Society writes in their recommendation paper that SGLT2 inhibitors should be used with caution in patients aged 75 years and older or in patients aged 65–74 years with geriatric syndrome [41]. However, this study revealed that the proportion of SGLT2 inhibitor tablets prescribed to patients aged 75 years and older increased from 20.3% in FY 2020 to 27.8% in FY 2022. Among all SGLT2 inhibitors, this proportion was the highest in the SGLT2 inhibitor that expanded its indication for both heart failure and CKD the earliest; patients aged 75 years and older accounted for 35.9% of prescription to men and 49.4% of prescription to women in FY 2022 (Fig. 6).

The strength of this study is that the datasets used are comprehensive real-world data that cover more than 95% of the total health insurance reimbursement claims in Japan [23], which makes it the largest and most reliable source for analyzing prescription trends of SGLT2 inhibitors. The limitations of this study include the lack of information on the indications for which the medications were used.

In conclusion, prescription of SGLT2 inhibitors has increased in patients of all ages but especially in elderly patients between FY 2020 and 2022. The SGLT2 inhibitor that expanded its indications for patients with heart failure and CKD the earliest showed the largest percentage increase in the number of prescribed tablets and the highest proportion of elderly patients after indication expansion. Prescription of SGLT2 inhibitors to patients aged 75 years and older accounts for more than a quarter of the entire prescription of SGLT2 inhibitors and is no longer infrequent in Japan. These findings contribute to updating our perception of the demographics of SGLT2 inhibitor recipients.

This study protocol was reviewed and approved by the Research Ethics Committee of the University of Tokyo Graduate School of Medicine, Approval No. 11612. Written informed consent was not required as only publicly available data were used.

Y.O. is a Research Fellow of the Japan Society for the Promotion of Science; receives scholarship and research grant from the Society; and received honoraria from Kyowa Kirin and Chugai during the 3 years prior to manuscript submission. H.N. reported receiving honoraria from Astellas, AstraZeneka, MSD, Otsuka, Ono, Kyowa Kirin, Kowa, Daiichi Sankyo, Mitsubishi Tanabe, Torii, Eli Lilly, Boehringer Ingelheim, and Bayer; and research grants from Mitsubishi Tanabe, Boehringer Ingelheim, Novo Nordisk, Bayer, and Kyowa Kirin. M.S. reported receiving honoraria from Daiichi Sankyo during the 3 years prior to manuscript submission. M.N. reported receiving research funding from Kyowa Kirin, Daiichi Sankyo, Astellas, Ono, Mitsubishi Tanabe, JT, Chugai, Bayer, Torii, and Takeda; and honoraria and/or advisory fees from Kyowa Kirin, Astellas, AstraZeneca, GSK, Daiichi Sankyo, Mitsubishi Tanabe, Chugai, Torii, JT, Novo Nordisk, and Boehringer Ingelheim.

This study was not supported by any funding.

Conceptualization: H.N. Data curation: Y.O. and H.N. Supervision: H.N. and M.N. Formal analysis, investigation, visualization, and writing – original draft: Y.O. Writing – review and editing: H.N., M.S., M.O., and M.N.

The data used in this study are publicly available in the following repositories: NDB Open Data Japan at https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000177182.html, [29]; Surveys of Population at https://www.e-stat.go.jp/stat-search/files?page=1&toukei=00200241, [30]; Patient Survey at https://www.e-stat.go.jp/stat-search/files?page=1&toukei=00450022, reference number [31].