Application of Wastewater‐Based Epidemiology in Japan to Estimate the Prevalence of Substance Use Including Methamphetamine and Ephedrine

ABSTRACT

Estimating substance use in Japan traditionally depends on police seizure records and national self‐reporting surveys, which face substantial challenges due to social stigma. This study applies wastewater‐based epidemiology to overcome these challenges of monitoring illicit drug and legal stimulant use in a Japanese metropolitan area serving ∼1.8 million residents. Influent wastewater samples collected over 4 weeks in 2021 were analyzed for 29 substances. Methamphetamine was the only illicit drug consistently detected (consumption at 17.0 ± 5.1 mg/day/1000 people), reflecting its prevalence in Japan's drug scene and aligning with patterns in other East Asian countries. Ephedrine and pseudoephedrine, known precursors of methamphetamine were detected at high levels, suggesting their common use in legal medicines. Nicotine and methamphetamine consumption remained stable across weekdays and weekends. Higher ephedrine consumption during the lockdown period likely resulted from increased use of Kampo medicines to manage COVID‐19 symptoms. This study provides the first objective evidence of population‐level methamphetamine use in a large urban catchment of Japan, which is relatively low compared to other countries. The findings demonstrate the feasibility of wastewater analysis as a complementary surveillance approach to the existing system, enabling more responsive and evidence‐based drug policy.

Methamphetamine was the only illicit drug consistently detected (consumption at 17.0 ± 5.1 mg/day/1000people). High levels of its precursors, ephedrine, and pseudoephedrine were likely from legal medicines, although there is a high risk of diversion. Our first application of wastewater‐based epidemiology in Japan to monitor illicit drug use demonstrates that this is a complementary tool to the existing system for informing evidence‐based drug policy.

1. Introduction

Methamphetamine remains a significant public health concern in Japan, as it was a primary contributing factor for mental health admissions [1]. Drug trafficking in East Asia including Japan, South Korea, and China could be facilitated by geographic proximity to potential suppliers in Southeast Asia [2], while emerging supply routes from South Africa and Latin America have diversified sources of methamphetamine entering the Japanese market [3]. South Korea is an important trans‐shipment hub for illicit precursors (production approximately 38,000 kg/year), including those originating from or destined for Asian countries such as Japan [4]. Furthermore, Japan had the highest street price for methamphetamine among East and Southeast Asian countries (approximately US $600 per gram), which may make it an attractive hub for both domestic and transnational trafficking organizations [3]. The drug laws in Japan are strict, resulting in a complex landscape of illicit drug smuggling that is difficult to trace or monitor [5]. National estimates of illicit drug use primarily rely on police‐recorded arrests and self‐reporting surveys [6]. These sources have several limitations, including small or nonrepresentative samples and the influence of social stigma, which likely leads to underestimation of actual drug use prevalence [6]. Furthermore, the limited availability of comprehensive epidemiological data means that drug arrests are often used as indicators of illicit drug use, even though they mainly reflect law enforcement efforts rather than actual consumption levels [6].

Wastewater‐based epidemiology (WBE) has proven to be an effective tool in addressing the limitations of traditional methods for monitoring illicit drug consumption [7, 8]. WBE is particularly valuable in settings with strong drug‐related stigma, such as much of Asia. Although WBE has been applied previvously in Korea, China, Taiwan, and Vietnam [4, 9, 10, 11], only two studies have recently employed this approach in Japan [12, 13]. Additionally, WBE enables rapid monitoring of shifts in consumption habits during significant events, such as the COVID‐19 pandemic [14, 15].

The COVID‐19 pandemic has profoundly impacted individual psychology and behavior, with lockdowns and social isolation exacerbating mental stress, leading to increased use of stimulant substances [16]. In Japan, even without the stringent enforcement of movement restrictions, the lockdown strategies significantly disrupted daily life and contributed to sustained psychological strain [17, 18]. There have been reports about shifts in behaviors related to smoking and drinking habits in Japan [13, 19]. Furthermore, pandemic‐related restrictions appear to influence substance use patterns differently depending on cultural contexts, periods of restriction and COVID‐19 infection level [17, 20]. It would be beneficial for public health agencies to quantify such behavioral changes to prepare for intervention if needed.

In this study, we employed the WBE approach to monitor the use of illicit drugs and legal stimulants (nicotine and caffeine) in a large Japanese metropolitan area during and after a COVID‐19 restriction period. This study aimed to establish a foundation for long‐term WBE studies in Japan, providing an addition to traditional methods to inform evidence‐based policies for controlling the use of these drugs.

2. Materials and Methods

2.1. Target Analytes

The study targeted a total of 29 substances and their biomarkers including nicotine (cotinine, COT, and hydroxycotinine, OH‐COT), caffeine (paraxanthine), paracetamol, carbamazepine, ephedrine, and pseudoephedrine together with 22 illicit drugs and metabolites including amphetamine, methamphetamine, mephedrone, 3,4‐methylenedioxyamphetamine (MDA), 3,4‐methylenedioxymeth‐amphetamine (MDMA), methyldiethanolamine (MDEA), ketamine, norketamine, fentanyl, norfentanyl, morphine, oxymorphone, hydromorphone, codeine, benzoylecgonine, oxycodone, noroxycodone, cocaine, methadone, monoacetylmorphine, buprenorphine, and norbuprenorphine. High‐purity grade (> 98%) pharmaceutical standards and isotopically labeled internal standards were purchased from Sigma‐Aldrich (Darmstadt, Germany). A list of analytes and the isotopically labeled internal standards used for quantification is provided in Table S1.

2.2. Sample Collection and Preparation

Influent wastewater samples were collected from a large sewage treatment plant (STP) in the Kanto region of Japan. The treatment capacity is approximately 750,000 m³/day, serving approximately 1.8 million residents, which corresponds to 1.3% of Japan's population.

Wastewater samples were collected by autosampler in time‐proportional mode, with 40 mL collected every 30 min. Eventually, 27 composite untreated wastewater samples were collected, representative of 24‐h periods. Sampling took place over 4 weeks, including a lockdown event, from 17th September to 13th October, 2021 (2 weeks during lockdown and 2 weeks post‐lockdown). After collection, all samples were immediately acidified to pH 2.0 with 1 M HCl and filtered through a 0.2 µm regenerated cellulose syringe filter (Minisart, Sartorius, Goettingen, Germany). A 1 mL filtrate was transferred into a 1.5 mL glass amber vial, frozen at −70°C, shipped on ice to the Queensland Alliance for Environmental Health Sciences (QAEHS), The University of Queensland, and then stored at −20°C until analysis.

2.3. Sample Analysis by Liquid Chromatography–Tandem Mass Spectrometry (LC–MS/MS)

Before LC–MS/MS analysis, frozen vials were thawed at room temperature, vortexed, and spiked with 10 µL of internal‐standard mix (0.5 µg/mL). A 10 µL aliquot was then injected into a SCIEX 6500 QTrap triple‐quadrupole/linear‐ion‐trap mass spectrometer (Sciex, Concord, Ontario, Canada) equipped with an electrospray ionization source in positive‐ion mode, coupled to a Nexera UHPLC system (Shimadzu Corp., Kyoto, Japan).

Chromatographic separation was performed using a Kinetex Biphenyl column (50 × 2.1 mm, 1.7 µm) with a Kinetex EVO C18 guard column (30 × 2.1 mm, 5 µm) at 45°C. Mobile phases were (A) 5 mM ammonium acetate + 0.1% formic acid in water and (B) 5 mM ammonium acetate + 0.1% formic acid in methanol; the gradient was 0–2.5 min (5→25% B), 2.5–5.5 min (25→55% B), 5.5–6.2 min (55→95% B), 6.2–7.9 min (hold 95% B), 7.9–8.0 min (return to 5% B), with re‐equilibration to 10 min at 5% B, all at 0.45 mL/min. Data acquisition were performed using Analyst 1.7.1, and quantification was carried out in MultiQuant 3.0 using relative response factors for native analytes and their isotopically labeled internal standards. Details of the LC–MS/MS parameters for all analytes comprising precursor, product ion transitions and retention times used for quantification are provided in Table S2.

2.4. Quality Assurance and Quality Control (QA/QC)

Sample analysis adhered to strict quality assurance and quality control measures (QA/QC). Quantification employed 10‐point calibration curves (0.02–80 µg/L; Milli‐Q water, pH 2.0) with R ² > 0.99 for most analytes. Linearity was assessed using calibration curves with 10 concentration levels for all analytes. Calibration curves were injected at the beginning and end of each batch. Several QA/QC samples including wastewater samples spiked with internal standards, and procedural blanks spiked with internal standards were analyzed in each batch. Random calibration standards (0.5, 1, 5, and 20 µg/L) and a solvent blank were injected every 20 wastewater samples to monitor instrument carryover. The method showed LODs of 10–35 ng/L, accuracy of 95%–107%, precision ≤ 7% RSD, and matrix effects of −20.8% to +13.1% (Table S1).

2.5. Back Estimation of Stimulant and Illicit Drug Consumption

Substance consumption was calculated using the following formula:

$$\begin{array}{ccc}& & \text{Consumption of parent chemical (mg/day/person)}\hfill \\ & & =\frac{C_i\times F}{P\times E_i}\times \frac{M_{Par}}{M_{Met}}\hfill \end{array}$$

C_i is the concentration of the target compound i in the influent (µg/L). E_i is a chemical‐specific excretion factor listed in Table S3. F is the daily flow of the STP (m³/day) and P is the population served in the catchment (1.8 million people). $\frac{M_{Par}}{M_{Met}}$ is the ratio of molecular weights of the parent compound and its metabolite. However, the consumption of illicit drugs in this study was estimated per 1000 people (mg/day/1000 people) due to the lower prevalence of use compared to other stimulants.

2.6. Statistical Analysis

This study employed a one‐way ANOVA test to compare substance consumption between weekdays and weekends during the lockdown and post‐lockdown periods. Additionally, a two‐sample t‐test was used to examine changes in the per capita consumption of the selected substances during and after the COVID‐19 lockdown. Statistical significance was determined at a p‐value threshold of < 0.05. All statistical analyses were performed in RStudio Version 4.4 using functions from the stats package for hypothesis testing, and the ggplot2 package for data visualization.

3. Results and Discussion

3.1. Occurrence of Biomarkers in Wastewater Samples

All influent wastewater samples contained detectable levels of biomarkers for nicotine, caffeine, paracetamol, carbamazepine, ephedrine, pseudoephedrine, and methamphetamine (100% detection frequency). However, the remaining 21 targeted drugs were not detected and were thus excluded from further analysis. Caffeine metabolism was reflected by paraxanthine concentrations, which ranged from 20.2 to 50.9 µg/L. This finding was consistent with a previous WBE study across eight STPs in Japan, where the caffeine metabolites theophylline and theobromine were among the highest‐concentration substances reported in inflows [21]. Methamphetamine was consistently detected, although at much lower concentrations (0.01–0.03 µg/L). The concentrations of nicotine biomarkers were observed in the range of 0.6–1.3 µg/L for COT and 0.9–2.2 µg/L for OH‐COT, with COT to OH‐COT ratios ranging from 0.53 to 0.68. Paracetamol was also quantified, showing comparable concentrations for both COT and paracetamol with Kadokami et al. [21]. Similarly, carbamazepine concentrations in this study (ranged from 0.04–0.07 µg/L) were consistent with previous Japanese data which demonstrated close agreement between sales‐based predicted concentrations and measured wastewater levels [22]. Overall, the concordance of our results with those of two previous Japanese studies validates the analytical performance of our method and the nondetections are likely due to very low population‐level use or concentrations below method detection limits (Figure 1, Table S4).

FIGURE 1.

Comparison of concentrations of substances in this study with wastewater studies in Japan (µg/L).

3.2. Illicit Drug and Stimulant Use

Our results indicated that methamphetamine use is low but sustained in the large urban catchment in the Kanto region of Japan, and is more prevalent than other illicit drugs monitored in this study. This pattern is consistent with WBE data from other East Asia countries such as South Korea, China, and Taiwan. [4, 8, 11, 23] Specifically, the average methamphetamine mass load was 7.3 ± 2.2 mg/day/1000 people (estimated consumption: 17.0 ± 5.1 mg/day/1000 people), which was comparable to reports from Korea and China, [4, 8, 11] the two neighboring countries in North Asia. However, such a level of use is several times lower than in Vietnam and Taiwan [9, 10] and many Western countries (Figure 2, Table S5). A multi‐country WBE study over seven years revealed a picture of methamphetamine use across different regions, with hotspots reported in North America and Australia [24].

FIGURE 2.

Comparison of daily mass loads of methamphetamine between Japan and other countries (mg/day/1000 people). Note: CN = China, KO = Korean, TA = Taiwan, VN = Vietnam, FR = France, HR = Croatia, SI = Slovenia, FI = Finland, PT = Portugal, SE = Sweden, AT = Austria, PL = Poland, GB = United Kingdom, IT = Italy, CY = Cyprus, GR = Greece, US = United States, NL = Netherlands, ES = Spain, DK = Denmark, LT = Lithuania, CH = Switzerland, TR = Turkey, DE = Germany, BE = Belgium, SK = Slovakia, LV = Latvia, CZ = Czech Republic, and AU = Australia. The data were extracted from the EUDA source (https://www.euda.europa.eu); data for AU and US are from 2023 due to availability, whereas data for other countries are from 2021.

Consistent with WBE data, methamphetamine has accounted for the largest total seizure mass of any illicit drug intercepted by Japanese Customs from 2017 to 2024. [3] Although cocaine and MDMA were not detected in this study, we estimated the upper‐boundary mass loads of their biomarkers by utilizing their limit of detection (LOD) values (both < 3.43 mg/day/1000 people). These estimates confirm that cocaine and MDMA were not common in Japan and other East Asian countries [8, 11, 23] but they are more common in Western settings.

Multiple factors might be attributed to Japan's low methamphetamine prevalence, including high retail price, strict drug policy, and strong social stigma. The high price might act as a double‐edged sword—stimulating supply incentives while simultaneously restraining demand. Japan is among the most expensive drug markets globally, with methamphetamine prices more than twice those in Korea and six times those in China [3]. In addition, strict drug laws and the criminalization of even minor drug offenses, combined with intense social stigma surrounding drug use in Japan, likely contribute to reduced usage [5, 6].

Interestingly, our study detected high levels of ephedrine and pseudoephedrine, two analgesics that could be used as precursors for clandestine production of methamphetamine. The daily mass loads of ephedrine and pseudoephedrine were 66 and 113.3 mg/day/1000 people, respectively. Although these compounds are recognized precursors for methamphetamine synthesis in East and Southeast Asia [25], they are also widely used in legal pharmaceutical products. In Japan, both substances are the main active ingredients of Ephedrae herba which is an important crude drug prescribed in Kampo formulations for treating respiratory and cold‐related symptoms. A recent study reported that 1 g of powdered Ephedrae herba contained 7.20 mg of ephedrine and 3.21 mg of pseudoephedrine in all the Kampo medications used clinically in Japan [26]. Kampo medicine consumption is therefore suspected to have contributed substantially to the high ephedrine and pseudoephedrine load observed in our wastewater samples. Only a few reports have noted that trace ephedrine has occasionally been detected in high‐purity methamphetamine seizures [27, 28]. Similar to the case in China and Korea, the detected levels of ephedrine and pseudoephedrine in wastewater suggested common use of legal medicine products [4, 29] Due to the common use and the nature of being easy to divert to illicit methamphetamine production, these observations highlight the need to monitor precursors alongside methamphetamine in wastewater to assess the level of diversion although it is likely low at the moment.

The mean daily nicotine consumption was estimated at 1.2 ± 0.2 mg/day/person, which was lower than studies in China, many European countries and the United States (Table S5). Meta‐analyses of WBE data similarly show lower consumption of both substances in Asia than in Western regions [30, 31]. Japan's gradual tobacco excise tax increases between 2018 and 2020 have been associated with significant declines in combustible cigarette smoking [32].

Caffeine consumption in this study was estimated at 301.9 ± 86.4 mg/day/person which was higher than reported in WBE study from China but lower than estimates from Latvia and Austria studies (Table S5). Notably, these WBE studies in Western countries estimated caffeine intake using the parent compound as a biomarker which could lead to overestimation of consumption [33, 34]. Our finding aligns with a dietary intake study in four large Japanese cities which accounted for major caffeine‐containing beverages and foods estimating mean caffeine consumption at 256.2 mg/day for women and 268.3 mg/day for men [35]. A previous WBE study in Kyoto City reported a mass load of caffeine of 13.68 mg/day/person, which is comparable to our result of 12.9 mg/day/person [36]. Caffeine intake is highly influenced by population‐specific habits and cultural preferences, such as preferred coffee types, caffeine content, and consumption methods. In Japan, tea is a primary source of caffeine, contributing to increased consumption alongside coffee [37]. Preferences also vary by age, with younger people tending to prefer coffee and older individuals favoring green tea [37]. Notably, caffeine consumption during our period study occasionally exceeded the daily intake limit recommended by the U.S. Food and Drug Administration (400 mg/day for healthy adults) (Table S5).

3.3. Weekly Consumption Patterns of Common Stimulants

Our findings showed that nicotine and methamphetamine were consumed regularly in the Kanto region of Japan, with estimated levels consistent throughout the week during lockdown and post‐lockdown periods. The stability of nicotine consumption aligns with previous studies, suggesting that smoking is habitual and primarily driven by dependence rather than recreational use [38, 39]. A nationwide self‐reported survey in Japan found that 49.8% of current users of any tobacco product reported daily use within their homes [40]. Similarly, stable methamphetamine consumption across all days of the week during and after the COVID‐19 pandemic aligns with wastewater studies in Taiwan [10]. Other WBE studies observed higher weekend consumption only during the sixth post‐lockdown period compared to earlier phases [41]. Likewise, no clear or only weak weekly patterns were detected in WBE studies conducted before the COVID‐19 pandemic [38, 42]. Again, this trend may be attributed to methamphetamine's strong addictive properties, which often lead to substance use disorder and dependence, driving habitual consumption among regular users [43].

In contrast, caffeine consumption remained stable throughout the week during lockdown but was significantly higher on weekends than on weekdays in the post‐lockdown period (ANOVA test, p < 0.05). This shift likely reflects increased use of these substances at home, primarily on weekends after working days. The population movement dynamics in our predominantly residential catchment might contribute to this pattern. This catchment is usually characterized by large weekday commuting flows to central Tokyo, which reduce wastewater contributions from residential catchments during the working week. On weekends, population presence returns to residential areas, potentially elevating weekend consumption levels. Similar temporal shifts were observed in other WBE studies between weekdays and weekends during lockdown and post‐lockdown periods [44]. Commuting patterns likely reduced weekday consumption, as individuals had fewer opportunities to consume these substances at home.

3.4. The Effect of COVID‐19 Restrictions

Our analysis revealed significantly lower consumption levels of all stimulants studied during the post‐lockdown period compared to the restrictions period (T‐test, all p‐values < 0.001) (Figure 3). A general explanation for this reduction could be the changes in population size and variability within the catchment area [45]. The STP serves a predominantly residential urban catchment in the Kanto region, where many inhabitants commute daily to central Tokyo. The Greater Tokyo Area is characterized by a pronounced spatial separation between residential suburbs and central employment hubs [46]. After restrictions were lifted, some individuals returned to their workplaces, decreasing the number of residents contributing to wastewater resulted in lower consumption. In contrast, during the restriction period, reduced daily commuting into and out of the catchments likely increased the proportion of residents present, thereby elevating estimated substance consumption. Similar effects of mobility reduction and COVID‐19 restrictions on the number of individuals contributing to wastewater signals were reported previously [47]. These findings highlight the importance of considering population dynamics and cultural behaviors when estimating substance consumption by WBE data. To validate population stability, our study also quantified acesulfame consumption levels, a widely used artificial sweetener. Acesulfame is almost entirely excreted unchanged and exhibits high chemical stability and resistance to degradation in wastewater systems [48, 49]. The acesulfame intake trend followed a similar pattern to that of the target substances, further validating our findings. Since acesulfame levels primarily reflect the size of the contributing population, this parallel decline supports our interpretation that population changes rather than analytical variability or differential degradation, contributed substantially to the observed reduction in stimulant consumption.

FIGURE 3.

Temporal trends in estimated consumption of substances before (red) and after (blue) the COVID‐19 lockdown. Shaded areas indicate weekend days. Lines represent LOESS (locally estimated scatterplot smoothing) smoothed trends. Note: Methamphetamine and ephedrine consumption (mg/day/1000 people), nicotine and acesulfame consumption (mg/day/person).

Substance‐specific patterns also reflected lifestyle and behavioral changes during the pandemic. Ephedrine is the main ingredient in Kampo medicine products (Maoto), a traditional Japanese herbal medicine. In addition to its prescription for upper respiratory infections such as influenza, Kampo medicine has been recently reported as a safe therapeutic option for COVID‐19. Clinical evidence suggests that Kampo medicine may serve as a key pharmacological treatment for long COVID–related general fatigue [50]. It has also been proposed as preventive care before COVID‐19 infection, an adjunctive therapy during the acute and subacute phases, and as supportive treatment during recovery [51]. A nationwide study of 1798 hospitals in Japan revealed that in 2021, the median proportion of patients prescribed Kampo was 11.5% (IQR 7.3%–16.2%), showing an increase from 7.7% in 2010 [52]. The higher consumption of ephedrine observed during the lockdown period may be partly explained by the increased use of Kampo medicines to manage COVID‐19 symptoms.

Tobacco use was not directly restricted, but COVID‐19 restrictions likely facilitated increased home smoking. A prior survey on 5120 current smokers indicated that individuals working from home or living alone reported increased smoking during Japan's state of emergency [53]. Increased stress, lack of smoke‐free rules at home and intensified tobacco advertising may have contributed to increased smoking intensity and hindered cessation [54, 55]. A wastewater study in Australia similarly observed a decline in nicotine consumption after restrictions eased [56]. Nicotine consumption levels could reflect differences in restriction severity, as stricter lockdowns have been associated with increased mental health challenges, potentially leading to greater nicotine use [57].

Given that caffeine is commonly consumed in daily life, the stay‐at‐home nature of lockdowns likely led to higher intake of foods and drinks containing this substance. Our findings aligned with a study in Southwest England that found an increase in caffeine intake during the third national lockdown, followed by a reduction post‐lockdown [58]. Other WBE studies reported minimal or inconsistent changes in caffeine consumption trends during COVID‐19 restrictions [34, 59]. Such discrepancies may be driven by various factors, including lockdown‐related stress, the need to counteract fatigue, and differences in caffeine sources, such as tea, energy drinks, and chocolate.

4. Limitations

We acknowledge some limitations in this study. First, wastewater samples were collected from a single STP in the Kanto region of Japan, which serves a large and mostly residential urban catchment. This limited the extent to which our findings can be extrapolated to represent stimulant use across Japan. However, the study shows that applying WBE in Japan is feasible and emphasizes the importance of expanding surveillance to more sites for future national assessments. Second, the sampling period was limited to four weeks, which restricts our ability to assess long‐term trends in substance consumption. Moreover, nondetection of target drugs in wastewater should be interpreted with caution, as it may reflect concentrations below the LOD rather than a true absence of consumption. Consequently, the findings might indicate low‐level use within the catchment rather than no use. Lastly, changes in population mobility during the study period likely influenced the estimated consumption of the target substances [45]. To assess population stability, we used acesulfame, a persistent wastewater marker and its trend mirrored that of the target substances, supporting our results.

5. Conclusion

This study provides the objective, population‐level evidence on illicit drugs and licit stimulants in a large population catchment in Japan using wastewater analysis. Methamphetamine was the only illicit drug detected in the studied population, while the other 21 targeted illicit drugs were not, confirming its prevalence in Japan's illicit market. We detected ephedrine and pseudoephedrine in wastewater at higher levels than methamphetamine, suggesting common use in legal medicinal products which could be diverted to illicit methamphetamine production. Nicotine consumption is relatively high throughout the study period. Lockdown (or movement restriction) may have affected the consumption level measured; methamphetamine, ephedrine, nicotine and caffeine loads were significantly lower during the post‐lockdown. This decrease likely reflects changes in population size, resulting from the movement of residents out of the catchment area after restrictions were lifted. The higher consumption of ephedrine observed during the lockdown period could plausibly reflect increased reliance on Kampo products for COVID‐19 symptoms. Overall, our findings highlight the value of WBE for monitoring illicit drug consumption in Japan to overcome the social stigma that hinders traditional methods, better informing law enforcement and public health policies.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: asia70629‐sup‐0001‐SuppMat.docx.

Do Q. T. T., Ngo H. K. T., Zheng Q., Masahiro O., and Thai P. K., “Application of Wastewater‐Based Epidemiology in Japan to Estimate the Prevalence of Substance Use Including Methamphetamine and Ephedrine.” Chemistry – An Asian Journal 21, no. 3 (2026): e70629. 10.1002/asia.70629

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.