Introduction: Attention-deficit/hyperactivity disorder (ADHD) is a common childhood neurodevelopmental disorder, yet systematic studies integrating disease burden and medication utilization across provinces remain limited in China.
Aims: To evaluate temporal trends and regional differences of ADHD burden and drug utilization among Chinese children.
Methods: Using national and provincial GBD data (2000-2021) and ADHD drug procurement data (2015-2023), we analyzed temporal and provincial patterns through descriptive statistics. We calculated average monthly percent change in drug utilization via log‑linear regression, and performed cross‑correlation analyses to assess contemporaneous and lagged associations between drug utilization and ADHD prevalence.
Results: From 2000 to 2021, ADHD incidence rose from 387.38 to 461.64 per 100,000 population, while prevalence declined from 3825.65 to 3636.24 per 100,000, and DALYs rate fell from 47.10 to 44.86 per 100,000. Notable provincial heterogeneity was observed: higher rates clustered in eastern coastal provinces (e.g., Shanghai, Jiangsu), whereas larger absolute numbers were in populous provinces (e.g., Guangdong and Henan). From 2015 to 2023, drug utilization grew steadily, with an average monthly increase of 2.54% (95% CI: 2.36-2.71%). All provinces experienced rising drug utilization, though at different paces. The fastest growth occurred in Ningxia, Hainan, and Qinghai, while the highest absolute usage levels were seen in Shanghai, Beijing, and Zhejiang. No statistically significant correlation with economic level was found overall; however, regions with medium‑low and medium‑high economic levels grew faster than high‑income regions. Nationally, no significant contemporaneous or lagged association between drug utilization and ADHD prevalence was found. Provincially, patterns varied: Fujian, Guangxi, Hebei, and Shandong showed positive contemporaneous correlations, Beijing and Liaoning showed negative correlations, and Henan, Shanghai, and Zhejiang showed lead-lag patterns.
Conclusions: ADHD burden and drug utilization show significant regional disparities. Drug utilization has expanded but aligns poorly with burden across regions, highlighting needs for targeted resource allocation.

Introduction: TikTok has experienced rapid growth in Thailand, emerging as a significant platform for online commerce, including cosmetics. Skin-whitening products represent the most frequently searched cosmetic category on this platform. The proliferation of cosmetic advertisements on social media raises concerns regarding regulatory compliance, particularly with Thailand's Cosmetic Act of 2015.

Aims: This study aimed to investigate skin-whitening cosmetic advertisements on TikTok and identify factors associated with the risk of non-compliance with Thailand's Cosmetic Act of 2015.

Methods: A cross-sectional descriptive study was conducted, analyzing 214 skin-whitening cosmetic advertisements published on TikTok between January and February 202ุ6. Two independent evaluators assessed advertisement compliance with the Cosmetic Act of 2015. Data were analyzed using descriptive statistics.

Results: Advertisement components were categorized as follows: image only (n=8, 3.74%), image and sound (n=9, 4.20%), image and text (n=167, 78.04%), and a combination of image, sound, and text (n=30, 14.02%). Non-compliant advertisements were identified in 202 cases (94.39%). Among non-compliant advertisements, text-only violations were most prevalent (n=118, 58.42%), followed by image and text combinations (n=29, 14.36%), image only (n=24, 11.88%), sound and text (n=11, 5.45%), sound only (n=13, 6.44%), image and sound (n=5, 2.48%), and all components combined (n=2, 0.99%). The most common violations included images depicting skin tone comparisons before and after product use and textual claims of permanent skin tone alteration. A statistical analysis of factors associated with non-compliance risk was not feasible due to the substantial disproportion between non-compliant (94.39%) and compliant (5.61%) advertisements.

Conclusions: Skin-whitening cosmetic advertisements on TikTok demonstrate a critically high rate of non-compliance with Thailand's Cosmetic Act of 2015. Consumers should critically evaluate product claims prior to purchasing skin-whitening cosmetics through TikTok. Regulatory authorities should strengthen enforcement mechanisms for cosmetic advertising on digital platforms.

Keywords: skin-whitening cosmetics, TikTok advertising, Cosmetic Act

Introduction: Heart failure is a common reason for cardiology admissions and Local epidemiological data are essential to identify high-burden groups and plan targeted multidisciplinary care in tertiary settings.

Aims: To estimate the prevalence of heart failure among cardiology inpatients over five years and describe their clinical profile, including age, gender, heart failure subtype, comorbidities, length of stay and in-hospital mortality.

Methods: An ambispective cohort design was used, combining a five-year retrospective review (2020-2024) with nine months of prospective follow-up. Among 5,864 cardiology admissions, 1,254 adults met screening criteria and 585 had confirmed heart failure of any stage. Data on demographics, ejection-fraction-based heart failure subtype (reduced, mildly reduced and preserved), comorbidities, social history, length of stay, number of medications and in-hospital outcomes were extracted and analysed descriptively. Prevalence was calculated overall and stratified by age and gender.

Results: Heart failure was present in 10% of cardiology admissions (585/5,864) and 46.7% of screened patients (585/1,254). Prevalence increased with age, from 30%(<45 years) to 62% (>75 years). Males comprised 72% of the cohort and had higher gender-specific prevalence (50%) than females (39%). Reduced ejection fraction was most common (53%), followed by mildly reduced (38%) and preserved (8%). Multimorbidity was frequent, with 61% of patients having at least 2 comorbidities, and most common length of stay was 4-7 days (42%). During follow-up, 83 patients (14%) died. Pharmacotherapy in these inpatients broadly followed guideline directed medical therapy, with widespread use of loop diuretics, renin-angiotensin system blockersneprilysin inhibitors, beta-blockers, mineralocorticoid receptor antagonists and sodium-glucose cotransporter-2m-glucose cotransporter-2 inhibitors.

Conclusions: Nearly half of screened cardiology inpatients and one in ten admissions were living with heart failure, with greatest burden in older adults and reduced ejection fraction. High multimorbidity and polypharmacy underscore the need for integrated multidisciplinary strategies to improve outcomes.

Keywords: Heart failure; Ambispective cohort; Multimorbidity

P-CARDIAC is an AI-driven tool predicting cardiovascular disease (CVD) recurrence in Hong Kong patients with established CVD. Clinicians evaluate clinical stability, but whether ESC and ACC guidelines classifications reflect this remains unclear. This study evaluates if P-CARDIAC can simultaneously assess risk and clinical stability for secondary prevention.

To compare P-CARDIAC, ESC 2025, and ACC 2019 risk classifications against clinician-assessed clinical stability.

Forty-eight cases were randomly selected by strata from a prospective cohort of patients aged 18–80 years with prior CVD attending specialist clinics across six Hong Kong Hospital Authority hospitals. Electronic health records provided data for P-CARDIAC scoring. Anonymized clinical notes were independently reviewed by fifteen cardiologists for ESC/ACC risk category and clinical stability. Category-based net reclassification improvement (NRI) with 500 bootstrap iterations compared P-CARDIAC with ACC and ESC, using clinician-judged instability as the event.

Forty cases were evaluated by at least two clinicians. Among thirty-three consensus cases, clinicians judged 12.5% of them as unstable and 87.5% as stable. Among stable patients, 82.1% and 65.2% were classified as very high and high risk under ESC and ACC, respectively, versus 17.8% high risk by P-CARDIAC. Compared to ACC, total NRI for P-CARDIAC model was 0.02 (95% CI −0.52 to 0.71), with NRI− 0.68 (95% CI 0.43 to 0.91) among stable patients and NRI+ −0.67 (95% CI −1.00 to <0.01) among unstable patients. Compared to ESC, total NRI was approximately 0.01 (95% CI −0.52 to 0.76), with a similar pattern (NRI− 0.67, 95% CI 0.43 to 0.87; NRI+ −0.67, 95% CI −1.00 to <0.01).

ESC and ACC demonstrate limited capability in differentiating clinical stability. P-CARDIAC reduces the overclassification of stable patients into high-risk categories, showing potential for clinical triage and resource allocation. However, larger sample sizes are needed due to the few unstable cases evaluated.

Introduction:Invasive fungal infections (IFIs) are associated with significant morbidity, mortality, prolonged hospitalization, and increased healthcare costs, particularly among critically ill and immunocompromised patients. The increasing emergence of antifungal resistance among Candida and Aspergillus species has complicated treatment strategies and underscores the importance of antifungal stewardship (AFS) programs. Continuous surveillance of antifungal utilisation and susceptibility patterns is essential for optimising therapy and mitigating resistance.

Aim:Ths study evaluated antifungal utilization patterns, characterised antifungal resistance profiles, and assessed clinical outcomes among hospitalized patients receiving systemic antifungal therapy.

Methods: A retrospective stewardship and surveillance study was conducted among patients receiving systemic antifungal therapy between 2020 and 2025 in a tertiary care hospital. Data on fungal isolates, antifungal prescribing patterns, resistance profiles, and clinical outcomes were analysed.

Results: A total of 300 patients were included. The predominant isolates were Candida tropicalis (19.7%), Aspergillus fumigatus (11.7%), Candida albicans (9.7%), and Candida auris (8.7%). Empirical therapy accounted for 76.0% of prescriptions, while targeted therapy represented 20.7%. Fluconazole was the most frequently prescribed antifungal (51.7%), followed by itraconazole (17.3%) and caspofungin (15.3%). Resistance was highest to fluconazole (27.6%) and itraconazole (24.0%), whereas caspofungin demonstrated lower resistance rates (6.7%). Year-wise surveillance revealed a progressive increase in antifungal resistance from 2020 to 2025, particularly among Candida species and azole antifungals. Clinical improvement was observed in 67.0% of patients, ICU admission occurred in 64.7%, and 30-day mortality was 23.3%.

Conclusion: This six-year evaluation demonstrated increasing azole resistance, extensive empirical antifungal use, and relatively preserved echinocandin susceptibility. Strengthened antifungal stewardship, routine susceptibility-guided therapy, and continuous surveillance are essential to curb antifungal resistance and improve clinical outcomes. Broader implementation of stewardship programs may help preserve antifungal effectiveness and reduce the global burden of antifungal resistance.

Keywords: Antifungal Stewardship, Antifungal Resistance, Invasive Fungal Infections, Candida, Surveillance

Introduction: Drug-related deaths remain a major and preventable cause of mortality, and every interaction with health services provides an opportunity to identify risks, intervene, and potentially prevent deaths.

Aims: We aimed to systematically synthesise international evidence on the prevalence of health service utilisation preceding drug-related deaths.

Method: We performed a systematic review and meta-analysis. We systematically searched PubMed, EMBASE, and PsycINFO on 10/10/20 for peer-reviewed literature. We included observational studies that reported the prevalence of any health service utilisation for any reason in the period prior to drug-related deaths. The pre-mortem observation period must have been consistently defined for all study participants. We used the Joanna Briggs Institute Critical Appraisal Tool to identify the risk of bias for each study.

Results: We included 32 studies; half were from the USA, and half involved general population samples. Accessing primary care and emergency department (ED) services were more common than drug-related services. Primary care was the most utilised service within 1 month (68.3%), 1 year (93.0%), and 2 years (81.0%) prior to death, and ED was the most utilised service in the 3 years prior to death, at 83.2%. In a meta-analysis of general population samples, 62.1% of decedents visited an ED within 1 year of death (95% confidence interval, CI 54.8%–69.0%). The one-year pooled prevalence of primary care use was 86.9% (95% CI, 79.3%-92.0%), regardless of population type.

Conclusions: This is the first systematic review and meta-analysis examining health service use prior to drug-related deaths. The high engagement with ED and primary care services presents a critical opportunity to expand overdose prevention efforts within mainstream healthcare settings. Additionally, it underscores the value of integrated models that link general healthcare with harm reduction services to increase access and enable timely risk identification.

Keywords (3): Drug overdose; facilities and services utilisation; intervention opportunities.

Introduction: High-risk patients admitted to public hospitals in Hong Kong were routinely screened for carbapenem-resistant Enterobacteriaceae (CRE) colonisation. Social distancing and increased antibiotic use during the COVID-19 pandemic might have affected the incidence of CRE colonisation.

Aims: To examine the trend and duration of CRE colonisation pre-, during and after the COVID-19 pandemic.

Methods:
Design: Interrupted time series
Setting: We collected data from the Hong Kong Hospital Authority (HA) electronic health record, patients aged >18, between 01/01/2017–30/06/2023 were included. CRE laboratory tests were conducted using either genotypic (polymerase chain reaction) or phenotypic (antimicrobial susceptibility test) methods.
Statistical analysis: We examined the changes in CRE monthly incidence before, during, and after the COVID-19 pandemic (Feb 2020–Oct 2022) in an interrupted time series using quasi-Poisson regression model to estimate the incidence rate ratio (IRR). We also conducted additional analyses with a negative control outcome (appendicitis incidence), lagged effect of COVID-19 pandemic interventions by 2 months.

Results: We identified 441,426 episodes from 439,394 patients who received ≥1 CRE test during the study period. Of these, 4,638 episodes (1.05%) had ≥1 CRE positive result. The CRE test-positive incidence trend showed increasing with seasonality before the onset of the COVID-19 pandemic in February 2020. The start of the pandemic period was marked by an immediate level reduction in CRE incidence [IRR: 0.47 (0.24, 0.91)]. Subsequently, the incidence trend experienced a steeper rise after the pandemic phase (Oct 2022), with social distancing no longer mandated. No statistical significance resulted in any additional analyses.

Conclusions: The observed rising trend prior and after to pandemic may reflect increased surveillance in public hospitals. The reduction in non-urgent hospital admissions, social distancing measures, and border closures may have contributed to decreased surveillance of CRE and a subsequent decline in CRE incidence.

Keywords: antimicrobial resistance, Carbapenem-resistant Enterobacteriaceae, hospital surveillance, antibiotics use

Importance: Coronary artery disease (CAD) and metabolic dysfunction-associated steatotic liver disease (MASLD) frequently coexist. However, whether CAD genetic susceptibility is associated with the development of MASLD and progression to liver-related events, and whether shared therapeutic pathways underlie these associations, remains unclear.
Objective: To investigate whether CAD genetic risk is associated with incident MASLD and liver-related events, and whether shared proteins explain this association.
Methods: Genetic epidemiologic study integrating cross-trait genetic analyses, Mendelian randomization, proteomic analyses, and a prospective cohort analysis. Publicly available genome-wide association study summary statistics and individual-level data from the UK Biobank. The study population included individuals with genetic information, comprising both the general population without prevalent liver disease at baseline and the MASLD population, as well as corresponding subsets with available proteomic data.
Results: Overall genetic correlations were observed between CAD and both MASLD and cirrhosis, along with shared genetic loci. Mendelian randomization supported causal effects of CAD on MASLD and cirrhosis. High CAD genetic risk was associated with increased risks of incident MASLD (HR = 1.15; 95% CI = 1.05–1.25), cirrhosis (HR = 1.23; 95% CI = 1.06–1.42), and liver-related mortality (HR = 1.30; 95% CI = 1.03–1.65). These associations were amplified in individuals with established MASLD, among whom high CAD genetic risk was also associated with higher hepatocellular carcinoma risk (HR = 1.86; 95% CI = 1.11–3.12). CAD-associated proteins showed good discrimination for MASLD (AUC = 0.827) and cirrhosis (AUC = 0.916). Drug-target Mendelian randomization and colocalization analyses identified MENT as a potential therapeutic target for MASLD and both MENT and CXCL9 as potential therapeutic targets for cirrhosis.
Conclusions and relevance: CAD genetic risk offers dual risk-stratification value for both CAD and liver-related outcomes, especially among individuals with MASLD. CAD-associated proteins show discrimination for liver-related events and provide potential therapeutic targets in cardio-hepatic pathways.

Introduction: Femoral fracture is a major cause of disability, care dependence, and mortality. Malnutrition, frailty, and impaired renal function may increase fracture susceptibility. The association between Geriatric Nutritional Risk Index (GNRI)-defined nutritional risk, renal function, and femoral fracture remains unclear in populations with and without chronic kidney disease.
Aims: To examine the association between GNRI-defined nutritional risk and femoral fracture. To explore whether the eGFR-fracture relationship differs by nutritional status.
Methods: We conducted a retrospective longitudinal study using administrative claims and health examination data from April 2014 to December 2024. Eligible participants had calculable GNRI and estimated glomerular filtration rate (eGFR), with no fracture recorded during a 180-day look-back period. The outcome was first femoral fracture (ICD-10 S72). GNRI <98 defined nutritional risk. Incidence rates were calculated per 1000 person-years. Fine-Gray models estimated sHRs with death as a competing risk, adjusted for age, sex, chronic kidney disease stage, and region. Restricted cubic splines evaluated capped eGFR, referenced to 60 mL/min/1.73 m2, stratified by GNRI group.
Results: Among 102,937 participants, 622 incident femoral fractures occurred. Incidence was 0.92 per 1000 person-years with GNRI >=98 and 4.36 per 1000 person-years with GNRI <98. Nutritional risk was associated with higher femoral fracture hazard after adjustment (sHR 1.49, 95% CI 1.19 to 1.86; p<0.001). In spline analyses, lower eGFR was associated with higher femoral fracture hazard versus eGFR 60 mL/min/1.73 m2. This increase appeared more pronounced among participants with GNRI <98. Confidence intervals were wide at extreme eGFR values.
Conclusions: GNRI-defined nutritional risk was independently associated with incident femoral fracture in a large population with and without chronic kidney disease. Lower eGFR was linked to higher fracture hazard, particularly among participants with nutritional risk. Renal function and GNRI may jointly improve fracture risk assessment.

Keywords: femoral fracture; GNRI; eGFR

Introduction
Obesity is a key risk factor for cardiometabolic disease and is commonly assessed using body mass index (BMI). However, BMI does not capture body fat distribution and may inadequately reflect adiposity-related metabolic risk. The body roundness index (BRI)—derived from height and waist circumference—has been proposed as an alternative indicator of central adiposity.
Aim
To investigate the predictive performance of BRI relative to BMI in a long-term population-based setting.
Methods
We analyzed data from the Shizuoka Kokuho Database (SKDB), a large regional healthcare database in Japan covering more than 2.5 million individuals between 2012 and 2024. Participants aged 40–74 years who underwent routine health checkups in 2013 and were free of diabetes and hypertension at baseline were followed for up to 11 years. We evaluated associations of BRI and BMI with incident diabetes and hypertension and compared predictive performance using net reclassification index (NRI) and c-statistic.
Results
A total of 93,532 individuals (median age, 64 years; 72% female; median BMI, 21.5 kg/m²) were included. Over a median follow-up of 6 years, 57% had a newly-recorded diagnosis of diabetes, hypertension, or both. Compared with a base model adjusted for age and sex, adding either BRI or BMI significantly improved risk stratification to a similar degree (NRI: 12.1% [95% CI: 10.9–13.2%] for BRI and 13.2% [95% CI: 12.3–14.3%] for BMI). Both indices improved model discrimination, with c-statistic increases of 1.07% (95% CI: 1.06–1.07%) for BRI and 1.30% (95% CI: 1.30–1.31%) for BMI. BMI demonstrated modestly better discrimination than BRI, with a 0.24% (95% CI: 0.23–0.24%) higher c-statistic. Findings were consistent when diabetes and hypertension were analyzed separately.
Conclusions
Although both BRI and BMI predicted incident diabetes and hypertension and improved risk prediction beyond age and sex, BRI did not outperform BMI in the study population.

Introduction: Type 2 diabetes (T2D) guidelines emphasize individualized treatment, yet there are limited representative studies on comorbidity patterns. People with T2D have a higher and growing burden of comorbidities, including non-cardiometabolic conditions.

Aims: To identify the multimorbidity patterns of persons with pharmacologically treated T2D and assess the risk of all-cause mortality using Australian real-world data.

Methods: A cohort of adults with T2D (≥30 years; N=95,187) and receiving glucose-lowering medication(s) was identified from the Victorian Admitted Episodes Dataset linked to Pharmaceutical Benefits Scheme (07/01/2012–12/31/2014). Latent class analysis was applied separately within three age groups (30–64, 65–74, ≥75) to identify multimorbidity patterns among those with ≥2 chronic conditions, excluding T2D. Cox regression was used to assess associations between these patterns and all-cause mortality through 06/30/2018.

Results: The sample included 34,855 individuals in the 30–64-year-old (56% male), 29,667 in the 65-74-year-old (58% male), and 30,665 in the ≥75-year-old (52% male) age groups. Five multimorbidity patterns were present in both younger age groups—complex multimorbidity, osteoporosis/cancer, cardiovascular, musculoskeletal, and unspecific (i.e., no diseases overexpressed)—while the oldest age group had only three: complex multimorbidity, osteoporosis/cancer, and unspecific. Relative to the unspecific pattern (reference), the complex multimorbidity pattern was associated with the highest mortality across all age groups (30–64: adjusted hazard ratio (aHR) 5.2, 95% confidence interval (CI) 4.5–6.0; 65–74: aHR 5.2, 95%CI 4.5–6.0; ≥75: aHR 2.1, 95%CI 2.0–2.3) followed by osteoporosis/cancer (30-64: aHR 2.9, 95%CI 2.5-3.4; 65-74: aHR 3.1, 95%CI ≥75: 1.2, 95%CI 1.1-1.2) and cardiovascular (30-64: aHR 1.5, 95%CI 1.3-1.7; 65-74: 2.2, 95%CI 1.9-1.7-2.1).

Conclusions: Complex multimorbidity was associated with five-fold higher risk of death in people aged 30-74 years and two-fold higher risk of death in people aged 75 years and older. Strategies addressing complex multimorbidity in older adults with T2D are needed.

Keywords: Latent Class Analysis, multimorbidity patterns, Type 2 diabetes

Introduction: Metabolic dysfunction–associated steatotic liver disease (MASLD) is an established independent risk factor for major adverse cardiovascular events. Combined body mass index (BMI) and waist circumference (WC) categories define distinct obesity phenotypes reflecting general versus central adiposity, yet their long-term cardiovascular impact in MASLD remains unclear.

Aims: This study evaluated the long-term cardiovascular risk associated with combined BMI/WC obesity phenotypes among patients with MASLD.

Methods: This nationwide retrospective cohort study linked Taiwan's Adult Preventive Health Care Service database with the National Health Insurance Research Database. Patients with MASLD, defined as a Hepatic Steatosis Index >36 plus at least one cardiometabolic risk factor, were identified between January 2013 and December 2018 and classified by overweight status(BMI ≥23 kg/m²) and central obesity(WC ≥90 cm in men, ≥80 cm in women) into four phenotypes: non-obesity, overweight only, central obesity only, and overweight with central obesity. Primary outcomes were all-cause mortality and 5-point major adverse cardiovascular events(5pt-MACE). Multivariable Cox proportional hazards models, adjusted for demographic and clinical covariates, estimated hazard ratios(HRs) with 95% confidence intervals(CIs).

Results: Among 1,644,704 participants followed for a mean of 7.7 years, central obesity only(HR 1.15; 95% CI, 1.08–1.22) and overweight with central obesity(HR 1.22; 95% CI, 1.17–1.26) were significantly associated with increased 5pt-MACE risk, whereas overweight only was not(HR 0.99; 95% CI, 0.95–1.03). For all-cause mortality, only central obesity only showed a marginally increased risk(HR 1.06; 95% CI, 1.00–1.14), whereas overweight only were associated with significantly reduced risk(HR 0.73; 95% CI, 0.71–0.76).

Conclusions: Cardiovascular and mortality risks varied substantially across BMI/WC phenotypes in MASLD. Central obesity, alone or combined with overweight, conferred the greatest cardiovascular risk, whereas overweight only appeared inversely related to mortality. These findings underscore the clinical value of phenotype-specific cardiovascular risk stratification in MASLD management.

Keywords: metabolic dysfunction–associated steatotic liver disease, obesity phenotypes, cardiovascular risks

Abstract

Introduction: Osteoporosis is a major public health concern characterized by a high risk of debilitating fractures. While genetic predisposition is a known risk factor, modifiable characteristics like physical frailty also play a critical role. However, their combined longitudinal impact on incident osteoporosis remains poorly understood.

Aims: This study aimed to examine the longitudinal association between the physical frailty phenotype and incident osteoporosis. It additionally explored the modifying role of genetic predisposition on this relationship.

Methods: A prospective cohort study analyzed 437,045 participants (aged 37–73 years) initially free of osteoporosis. Exposures included the five-domain Fried frailty phenotype and a categorized polygenic risk score. Hazard ratios, risk advancement periods, and measures of additive interaction were estimated.

Results: Over a median 13.62 years of follow-up, 15,481 (3.5%) incident osteoporosis cases were identified. Baseline prevalence was 43.5% for pre-frailty and 8.1% for frailty. Osteoporosis risk increased with frailty severity, showing hazard ratios of 1.40 (95% CI, 1.35–1.45) for pre-frailty and 2.80 (95% CI, 2.63–2.99) for frailty. This association was strongest in younger adults (HR=4.90; 95% CI, 3.81–6.30). Jointly, high genetic risk and frailty markedly elevated risk (HR=6.40; 95% CI, 5.71–7.18), advancing the risk period by 19.98 years. A significant additive interaction existed between frailty and genetic susceptibility (RERI=1.52 for frailty).

Conclusions: Physical frailty is strongly associated with incident osteoporosis, particularly among younger populations. Integrating both frailty and genetic predisposition assessments into clinical screening could substantially enhance targeted preventive strategies. Future research should evaluate the clinical feasibility of implementing these combined screening tools in practice.

Keywords: Osteoporosis, Physical frailty, Polygenic risk score

Introduction: Persistent pain and hearing difficulty are common later-life conditions linked to reduced independence, psychosocial distress, cognitive vulnerability, and healthcare burden. Evidence suggests persistent pain is associated with hearing difficulty, but pathways remain unclear. Tracing repeatedly measured mediators may identify dynamic, modifiable pathways through which persistent pain contributes to hearing difficulty.
Aims: This study quantified path-specific indirect effects through repeatedly measured, time-varying mediators in the association between persistent pain and incident hearing difficulty.
Methods: We used 1998-2022 biennial U.S. Health and Retirement Study data. Persistent pain was moderate or severe pain in both 1998 and 2000; incident hearing difficulty was first fair or poor self-rated hearing from 2004 onward among participants with excellent to good hearing in 2002. We estimated this association using competing-risk multi-state models and applied path-specific mediation analyses with repeatedly measured mediators spanning depressive symptoms, cognition, BMI, self-rated health, ADL/IADL, physical activity, sleep disturbance, mobility and muscle limitation, social isolation/loneliness, incident morbidity, falls, and healthcare utilization.
Results: Among 12,093 participants, mean age was 63.5years (SD 9.6) and 63.2% were women. Within the persistent pain-incident hearing difficulty association, sleep disturbance showed significant indirect effects across follow-up (e.g., 0.019, 95% CI 0.014-0.025 at 12 years), with the estimated proportion mediated increasing from 41.3% at 2 years to 58.0% at 20 years. Stable indirect effects were also observed for CES-D depressive symptoms (0.017, 95% CI 0.011-0.021; 39.9%), cognition score (0.005, 95% CI 0.002-0.007; 11.7%), BMI (0.006, 95% CI 0.002-0.008; 13.4%), self-rated health (0.014, 95% CI 0.008-0.023; 35.5%), large muscle index (0.017, 95% CI 0.008-0.025; 41.3%), incident arthritis/rheumatism (0.003, 95% CI 0.002-0.004; 10.1%), and outpatient surgery (0.001, 95% CI 0.0001-0.002; 2.7%).
Conclusions: Time-varying mediators partly explained the association between persistent pain and incident hearing difficulty, highlighting modifiable pathways for preventing or delaying hearing difficulty.
Keywords: persistent pain; hearing difficulty; mediation analysis

Introduction: Type 2 Diabetes Mellitus is a leading cause of chronic kidney disease, a major global health burden affecting nearly 700 million people worldwide. SGLT2 inhibitors have demonstrated favorable renal outcomes in patients with T2DM and CKD, with landmark trials such as DAPA-CKD showing reductions in albuminuria and slowing of CKD progression. However, prospective real-world evidence on the effectiveness of SGLT2 inhibitors among Indian patients with T2DM and CKD remains limited.

Aims: To evaluate the real-world effectiveness of SGLT-2 inhibitors on renal outcomes in patients with Type-2 DM and CKD.

Methods: A prospective cohort study was conducted among 81 adults with Type-2 DM and CKD receiving SGLT-2 inhibitor therapy at a tertiary care teaching hospital in Mysuru, India. Eligible patients were followed for 6 months, and renal and glycemic parameters were assessed at baseline, 3 months, and 6 months. Changes in UACR, eGFR, serum creatinine, RBS, and HbA1c over time were analyzed using repeated-measures ANOVA.

Results: Among 81 patients with T2DM and CKD, SGLT-2 inhibitor therapy resulted in a significant reduction in UACR from 417.38 ± 119.66 mg/g to 295.05 ± 104.01 mg/g over 6 months (p < 0.001). Renal function remained stable, with no significant changes in eGFR (47.43 ± 19.60 vs. 47.51 ± 17.64 mL/min/1.73m²; p = 0.993) or serum creatinine levels (1.75 ± 0.58 vs. 1.69 ± 0.49 mg/dL; p = 0.797). Modest reductions were observed in HbA1c and RBS, although these changes were not statistically significant.

Conclusion: SGLT2 inhibitor therapy was associated with a significant reduction in albuminuria and preservation of renal function in patients with T2DM and CKD in a real-world clinical setting. These findings support the effectiveness of SGLT2 inhibitors in the management of diabetic kidney disease and reinforce their role in slowing renal disease progression.

Keywords: SGLT-2 inhibitors; CKD; Real-World Evidence

Introduction: JAK inhibitors, established therapies for rheumatoid arthritis (RA), have been proposed as drug repurposing candidates for Alzheimer's and Dementia Risk (ADRD) through modulation of the JAK/STAT pathway, but real-world evidence remains limited.

Aims: To evaluate whether JAK inhibitors reduce the risk of developing ADRD in patients with rheumatoid arthritis.

Method: We conducted a retrospective cohort study under a target trial emulation framework using data from the TriNetX US Collaborative Network. Patients aged ≥ 65 years initiating either JAK inhibitors (tofacitinib, baricitinib, or upadacitinib) or abatacept were included. We applied a one-year washout period to exclude prevalent users. Patients with other autoimmune diseases sharing treatment indications were excluded. We incorporated a 6 month induction period to account for the latency window of ADRD. Thus, the index date was defined as 6 months after study drug initiation. Participants were followed until outcome occurrence, death, or study end. We used intention-to treat analysis and 1:1 propensity score matching to emulate randomization in the target trial and ensure similar probabilities of treatment assignment. ADRD were identified using ICD-10-CM codes for Alzheimer’s disease (G30) and unspecified dementia (F03). We applied Cox models to estimate hazard ratios (HR) and 95% confidence interval (CI), using abatacept as reference. We conducted subgroup analyses by individual JAK inhibitor agents, sex and age.

Results: We included 16,606 patients (8,837 JAK inhibitors, 7,769 abatacept) respectively. The mean age was 71.4 years (SD 5.22) and 72.2 years (SD 5.22), with 22.46% and 22.33% male, respectively. We observed an association between a reduced risk of ADRD (HR=0.866, 95% CI: 0.683-1.098).

Conclusions: We observed that JAK inhibitors were associated with a reduced risk of developing ADRD compared with abatacept, suggesting a potential effect on prevention of developing ADRD that requires validation in future studies.

Keywords: JAK inhibitors, Alzheimer’s Disease, dementia, rheumatoid arthritis.

Introduction: The landmark EV-302 trial established enfortumab vedotin plus pembrolizumab (EV+P) as the standard first-line treatment for advanced urothelial carcinoma (aUC). However, real-world evidence regarding enfortumab vedotin plus immunotherapy (EV+IO) in Asian populations remains extremely limited.

Aims: To evaluate the real-world effectiveness and safety of EV+IO in patients with aUC.

Methods: This multicenter retrospective cohort study utilized electronic medical records in Taiwan. Patients with aUC initiating EV+IO (i.e., nivolumab and pembrolizumab) between August 2022 and May 2024 were included. The index date was defined as the first EV+IO prescription and was followed through May 2026. Primary outcomes of progression-free survival (PFS) and overall survival (OS) were analyzed using the Kaplan-Meier method, alongside the objective response rate (ORR) and disease control rate (DCR). Safety endpoints focused on treatment-related adverse events (TRAEs), specifically peripheral neuropathy and dermatologic toxicities, and subsequent dose modifications.

Results: A total of 33 Asian patients (median age 68 years; 66.7% metastatic) were analyzed. Notably, upfront dose reduction (<1.25 mg/kg) was implemented in 84.8% of patients. The median number of EV cycles administered was 5.0 (range, 1.0–30.0), with achieving ORR as the primary reason for discontinuation. The ORR reached 45.4% (21.2% complete, 24.2% partial), and the DCR was 60.6% with an additional 15.2% achieving stable disease. After a 16.0-month median follow-up, the median PFS was 14.5 months, while the median OS was not reached. (overall survival rate: 69.7%). Regarding safety, only 3.0% experienced grade 3–4 toxicities (maculopapular skin rash), which were managed with dose reduction.

Conclusions: EV+IO demonstrates comparable real-world effectiveness and a manageable safety profile in Asian patients, with clinical outcomes consistent with those of the landmark EV-302 trial. However, a longer follow-up study is warranted to confirm long-term survival benefits in routine clinical practice.

Keywords: Urothelial carcinoma, enfortumab vedotin, Taiwan

Introduction: Perioperative durvalumab plus gemcitabine–cisplatin has emerged as a new standard of care for Muscle-Invasive Bladder Cancer (MIBC) following the NIAGARA trial. However, real-world evidence in Asian populations remains limited. 

Aims: This study evaluated the perioperative pattern in Taiwanese MIBC patients.

Methods: This retrospective observational study used a multicenter electronic medical records database. Patients with MIBC who initiated durvalumab plus gemcitabine–cisplatin, defined as the index date, between December 2024 and February 2026 were included. Patients were followed until loss to follow-up, death, or data cutoff in May 2026. Primary outcomes included treatment patterns, effectiveness, and safety. Treatment patterns were assessed by completion of neoadjuvant therapy and receipt of curative-intent surgery. Effectiveness was assessed based on tumor response at the first available radiographic or pathological evaluation following neoadjuvant treatment, while safety was assessed according to CTCAE v5.0.

Results: A total of 25 patients with a median age of 65 years were included. Most patients were male (80.0%) and had ECOG performance status 0 (88.0%). T2N0 disease accounted for 64.0% of cases, while 12.0% had regional lymph-node involvement. Among all patients, 17 (68.0%) completed four cycles of neoadjuvant GC+D, and 5 (20.0%) subsequently underwent radical cystectomy (RC). Tumor responses included 8 complete responses, 11 partial responses, and 3 stable diseases. After a median follow-up of 9.4 months, disease progression occurred in 8 patients (disease-free survival: non-reached). Grade 3 treatment-related adverse events occurred in 7 patients, with no Grade 4–5 events. Neutropenia was the most common Grade ≥2 adverse event (44.0%). 

Conclusions: Real-world neoadjuvant GC+D showed encouraging effectiveness and acceptable tolerability among Taiwanese patients with MIBC. However, given the limited use of subsequent RC in routine practice, larger studies with longer follow-up are warranted. 

Keywords: muscle-invasive bladder cancer; durvalumab; neoadjuvant therapy

Introduction: Cervical cancer remains a leading cause of cancer-related morbidity and mortality among women worldwide, particularly in low- and middle-income countries. Although chemotherapy and concurrent chemoradiotherapy significantly improve disease control, treatment-related toxicities continue to compromise therapeutic adherence, quality of life, and clinical outcomes. Existing evidence on chemotherapy-induced adverse events in cervical cancer remains fragmented, with limited focus on standardized toxicity reporting and pharmacovigilance-based safety evaluation. Addressing these gaps is essential for optimizing supportive care and improving patient-centered treatment strategies.

Aim: To systematically evaluate the prevalence, pattern, severity, and clinical implications of chemotherapy-induced toxicities in cervical cancer patients, with emphasis on pharmacovigilance and safety outcomes.

Methods: A systematic review was performed using PubMed, Scopus, Web of Science, and Google Scholar databases for studies published between 2010 and 2025. Studies reporting chemotherapy-associated toxicities among cervical cancer patients were screened according to PRISMA guidelines. Research involving cisplatin-based therapy, combination chemotherapy, and concurrent chemoradiotherapy was included. Data regarding adverse drug reactions, toxicity grading, treatment modifications, and supportive care interventions were extracted and qualitatively synthesized.

Results: 34 studies involving over 7,500 cervical cancer patients were included. Haematological toxicities, particularly anaemia, neutropenia, and thrombocytopenia, were the most frequently reported adverse events, followed by gastrointestinal complications such as nausea, vomiting, mucositis, and diarrhoea. Cisplatin-based regimens showed increased risks of nephrotoxicity and electrolyte disturbances, while combination therapies demonstrated higher incidences of grade III/IV toxicities. Several studies emphasized that early toxicity monitoring, dose optimization, and multidisciplinary supportive care significantly reduced treatment interruption and hospitalization.

Conclusion: Chemotherapy-induced toxicities remain a critical barrier to effective cervical cancer management. This review highlights major gaps in standardized pharmacovigilance reporting and supports the integration of proactive toxicity monitoring and patient-centered safety strategies to improve therapeutic outcomes.

Keywords: Adverse drug reactions, Cervical cancer, Chemotherapy toxicity, Pharmacovigilance, Safety outcomes, Supportive care.

Introduction:Radiotherapy (RT) combined with immune checkpoint inhibitors (ICIs) has emerged as a promising treatment strategy for patients with solid tumors. Preclinical and clinical evidence suggests that RT may enhance antitumor immune responses and improve the efficacy of ICIs. However, the optimal timing of ICI administration relative to RT remains uncertain. This uncertainty may influence both treatment effectiveness and safety outcomes.

Aims: This study aimed to compare the efficacy and safety of sequential versus concurrent ICI administration in combination with RT. A systematic review and network meta-analysis were conducted to evaluate differences in survival and toxicity outcomes between treatment strategies.

Methods: PubMed, Embase, CENTRAL, and ClinicalTrials.gov were systematically searched through February 2025. Randomized controlled trials evaluating ICIs administered either during RT (concurrent strategy) or after RT completion (sequential strategy) were eligible for inclusion. Overall survival (OS) was the primary endpoint, while progression-free survival (PFS) and grade ≥3 adverse events (AEs) were secondary endpoints. A random-effects network meta-analysis was performed.

Results: Twenty-four randomized controlled trials involving 9,480 patients were included. Compared with control treatment, sequential ICI administration significantly improved OS (HR, 0.81; 95% CI, 0.72–0.92). Sequential administration also improved PFS (HR, 0.73; 95% CI, 0.64–0.84). Concurrent administration did not demonstrate statistically significant survival benefits compared with control treatment. Direct comparison showed superior PFS with sequential administration versus concurrent administration (HR, 0.81; 95% CI, 0.67–0.97). No significant difference in grade ≥3 AEs was observed between sequential and concurrent strategies (OR, 1.02; 95% CI, 0.56–1.85).

Conclusions: Sequential administration of ICIs following RT was associated with improved survival outcomes without increasing severe toxicity. These findings suggest that treatment sequencing may be an important determinant of clinical benefit when combining RT and immunotherapy in solid tumors. Further studies are needed to identify factors that influence the optimal sequencing strategy.

Keywords: Radiotherapy; Immune checkpoint inhibitors; Network meta-analysis

Introduction: Iguratimod is a csDMARD for treating rheumatoid arthritis, approved in China, Japan, South Korea, and India (through 2011-2019). Existing evidence primarily demonstrates superior efficacy of iguratimod plus methotrexate versus methotrexate monotherapy; systematic review evidence comparing iguratimod to other csDMARDs remains limited.

Aims: To evaluate the efficacy/effectiveness of iguratimod.

Methods: A systematic review was conducted in PubMed and China National Knowledge Infrastructure (CNKI) (through April 2026). Randomized controlled trials (RCTs) and observational studies which compared the efficacy/effectiveness of iguratimod versus another csDMARD were identified. Both the initial search and study selection were performed by 2 independent researchers. Study quality was assessed by Jadad scale (for RCTs) and Newcastle-Ottawa scale (for observational studies). Random-effect models were used for meta-analysis.

Results: Fifteen studies were identified (8 in English, 7 in Chinese; 10 RCTs, 5 observational). Studies mostly had treatment durations of 24 weeks (9 studies) or 52 weeks (4 studies). Three studies were rated as low quality (all were RCTs), mainly due to lack of blinding and unclear dropout reporting; the other 12 studies were rated as high quality.
Nine studies evaluated monotherapies (iguratimod vs. leflunomide, salazosulfapyridine, or methotrexate); while 6 studies compared “methotrexate plus iguratimod” versus “methotrexate plus another csDMARD” (tacrolimus, salazosulfapyridine, leflunomide, or hydroxychloroquine). Iguratimod users exhibited significant improvements from baseline in 13 studies (measured by DAS28-CRP, TJC, SJC, etc.); while the remaining 2 studies did not report changes from baseline. Iguratimod demonstrated superior efficacy/effectiveness to comparators in 4 studies, and non-inferiority in the others.
Pooled meta-analysis showed significantly higher DAS28-CRP remission for iguratimod versus comparators (pooled RR: 1.39 [1.04; 1.87], 7 studies [1184 patients]); and no significant difference for ACR20 response (pooled RR: 1.06 [0.98; 1.14], 8 studies [1674 patients]).

Conclusions: This study adds to the evidence that iguratimod is effective for rheumatoid arthritis treatment.

Keywords: Iguratimod, efficacy, meta-analysis

Introduction: Vitamin D has been linked to metabolic and vascular pathways that may influence cardiometabolic health. However, evidence regarding its role in cardiovascular disease progression among people with prediabetes remains unclear.
Aims: We aimed to investigate the associations between vitamin D supplementation and transitions across cardiometabolic disease states among individuals with prediabetes.
Methods: This prospective cohort study included UK Biobank participants with prediabetes who were free of cardiovascular disease and cancer at baseline. The exposure of interest was vitamin D supplementation. Multi-state models were used to examine six transitions from prediabetes to incident type 2 diabetes mellitus (T2DM), major adverse cardiovascular events (MACE), and death. Additional analyses separately assessed transition patterns involving heart failure, stroke, and myocardial infarction.
Results: Among 47,265 eligible participants, 7214 (15.3%) participants developed incident T2DM, 3990 (8.4%) developed MACE, and 4642 (9.8%) died during a median follow-up of 13.97 years. Vitamin D supplementation was associated with a modestly lower risk of progression from prediabetes to incident T2DM (HR = 0.93, 95% CI, 0.88 to 0.98; P = 0.012). In contrast, associations with subsequent transitions to MACE or death were closer to the null. Specific analyses for heart failure, stroke, and myocardial infarction showed patterns broadly consistent with those observed for overall MACE.
Conclusions: These findings suggest that, in prediabetes, vitamin D supplementation was more related to early metabolic progression than to later cardiovascular or mortality transitions.
Keywords: vitamin D, prediabetes, multi-state models

Background: Irritable bowel syndrome (IBS) is a common chronic gastrointestinal disorder that significantly impairs quality of life, with a prevalence of 5.5–14.4% in Vietnam. Mebeverine and otilonium bromide (OB) are commonly used first-line antispasmodics; however, comparative evidence on their effectiveness remains limited. This study aimed to compare the efficacy and safety of mebeverine and OB, with IBS-QOL as the primary outcome.
Methods: This single-center prospective cohort study (from December 2024 to July 2025) included adult IBS patients fulfilling Rome IV criteria. Treatment allocation (mebeverine 200 mg twice daily or OB 40 mg twice daily for 4 weeks) was based on the treating physician’s decision. The IBS-QOL questionnaire was interviewer-administered at baseline and at week 4. IBS-QOL total and subdomain scores were compared within and between groups. Adverse events were recorded and graded according to the Common Terminology Criteria for Adverse Events.
Results: A total of 110 patients were included in this study. Most baseline demographic and clinical characteristics, were similar between groups, although regular exercise and the use of rifaximin and trimebutine were more frequent in the mebeverine group. IBS-D was the most common subtype. IBS-QOL total scores increased from 74.3 (66.4–85.2) to 98.2 (92.6–100.0) in the mebeverine group and from 77.9 (66.0–86.80) to 97.4 (85.8–100.0) in the OB group (p < 0.05). Greater improvements favoring mebeverine were observed in the IBS-QOL total score, interference with activity, and health worry domains (p = 0.034, 0.043, and 0.029, respectively). Mild adverse events occurred in 7.7% of patients, with no significant difference between groups (p = 0.875). Concomitant sulpiride use was independently associated with greater improvement in IBS-QOL (β = 6.72; 95% CI 0.30–13.14; p = 0.040), representing a hypothesis-generating finding.
Conclusion: Both mebeverine and OB were safe and effective in improving IBS-QOL over 4 weeks, with greater benefits observed for mebeverine.

Introduction: Timely reperfusion for acute myocardial infarction is often constrained in resource-limited health systems by delayed presentation, affordability barriers, and limited catheterization capacity. Streptokinase remains a low-cost thrombolytic option, but real-world evidence on its clinical value and safety in routine care is needed. Understanding outcomes in patients who receive or do not receive streptokinase can inform pragmatic reperfusion pathways.

Aims: To compare real-world clinical outcomes of patients with acute myocardial infarction treated with streptokinase versus those not treated with streptokinase. The study assessed recovery, in-hospital mortality, complications, length of stay, follow-up need, and short-term survival.

Methods: A retrospective observational study used medical records of 100 adult patients with acute myocardial infarction admitted to coronary care units from April to June 2024. Patients were classified as streptokinase-treated (n=50) or non-streptokinase (n=50). Demographic, laboratory, treatment, and outcome variables were extracted using a structured form. Comparative analyses and Kaplan–Meier survival analysis were performed using statistical software.

Results: Streptokinase-treated patients presented earlier than non-treated patients (4.44±3.90 vs 7.04±5.08 hours; p=0.005). Despite higher admission troponin (1.85±0.84 vs 0.78±1.02 ng/mL; p<0.001) and CK-MB levels (37.50±7.43 vs 17.55±15.06 U/L; p<0.001), the streptokinase group had substantially higher full recovery (74% vs 20%). In-hospital mortality was lower with streptokinase (6% vs 28%). Complications were more frequent with streptokinase (54% vs 18%; p<0.001), mainly cardiogenic shock, post-infarction angina, arrhythmias, and thrombolysis-related bleeding. Length of stay was similar between groups (8.14±1.68 vs 8.24±2.74 days; p=0.826). Survival distributions were not significantly different during short-term follow-up (log-rank p=0.136).

Conclusions: Streptokinase was associated with improved recovery and lower in-hospital mortality despite higher manageable complications. These findings support protocol-driven patient selection, timely administration, and safety monitoring when advanced reperfusion options are limited.

Keywords: Streptokinase; drug effectiveness; acute myocardial infarction.

Introduction: Semaglutide, a glucagon-like peptide-1 receptor agonist, provides significant glycemic and weight-loss benefits in type 2 diabetes mellitus (T2DM). However, its impact on microvascular complications, particularly diabetic retinopathy, remains uncertain due to conflicting evidence.

Aims: To evaluate the association between semaglutide and microvascular complications in patients with T2DM and to assess its effects on key metabolic outcomes including HbA1c, body weight, and body mass index.

Methods: A systematic review and meta-analysis of randomized controlled trials (RCTs), registered with PROSPERO, was conducted following PRISMA 2020 guidelines. PubMed, Embase, and Scopus were searched from inception to August 2025 for English-language RCTs involving adults (≥18 years) with T2DM receiving semaglutide at any approved dose or frequency and reporting microvascular outcomes. Risk of bias was assessed using the Cochrane RoB 2 tool, and pooled risk ratios and mean differences were calculated through meta-analysis.

Results: Semaglutide did not show a significant increase in diabetic retinopathy compared with active comparators (RR 0.89, 95% CI 0.75–1.06) or placebo (RR 1.16, 95% CI 0.92–1.46), with low heterogeneity across analyses. A lower incidence of diabetic nephropathy was observed with semaglutide than placebo recipients (3.8% vs 6.1%), suggesting potential renoprotective effects. Evidence regarding diabetic neuropathy was limited and inconclusive. Semaglutide consistently improved metabolic outcomes, demonstrating HbA1c reductions of up to 2.0%, body weight reductions of up to 13.2 kg, and BMI reductions of up to 4.1 kg/m². Dose-dependent improvements were observed across multiple trials.

Conclusions: This meta-analysis of 25,442 participants provides robust semaglutide-specific evidence demonstrating no significant increase in microvascular risk while confirming substantial metabolic benefits in T2DM. These findings support the continued clinical use of semaglutide and help address ongoing concerns regarding its microvascular safety profile. Future long-term studies with dedicated microvascular endpoints are warranted.

Keywords: Semaglutide; Type 2 Diabetes Mellitus; Microvascular Complications.

Introduction: Patients undergoing revascularization require intensive lipid-lowering therapy. However, the impact of lipid management patterns on outcomes remains unclear in China.

Aims: To evaluate the association of lipid management patterns with low-density lipoprotein cholesterol (LDL-C) and major adverse cardiovascular events (MACE) following revascularization.

Methods: Retrospective cohort data were obtained from the EHR system covering > 70 secondary and tertiary hospitals in Tianjin, China. Patients undergoing revascularization who newly initiated PCSK9 inhibitors (PCSK9is) in 2022 were included. A 90-day landmark design was applied. Patients were classified into four lipid management patterns based on statin adherence in the first 6 months post-revascularization and PCSK9i persistence during the landmark period (persistent: >1 prescription; non-persistent: 1 prescription): Group 1 (adherent/persistent), Group 2 (adherent/non-persistent), Group 3 (non-adherent/persistent), and Group 4 (non-adherent/non-persistent). Multivariable linear regression and Cox models were used to evaluate the associations of lipid management patterns with LDL-C and MACE, respectively. A 180-day landmark analysis served as sensitivity analysis.

Results: Among 1,560 eligible patients (mean age 63.0 years, 69.0% male), Group 1 achieved significantly lower LDL-C levels than Group 4 in the 90-day landmark analysis (adjusted β=-0.25 mmol/L, 95% CI -0.43 to -0.07; P=0.006). Among patients with discordant lipid management patterns, Group 3 had numerically lower LDL-C levels than Group 2 in the 90-day landmark analysis (adjusted β=0.16 mmol/L, 95% CI -0.02 to 0.35; P=0.084). This association became statistically significant in the 180-day landmark analysis (adjusted β=0.26 mmol/L, 95% CI 0.04 to 0.48; P=0.023). During a median follow-up of 1.7 years, MACE showed a directionally consistent pattern, with the lowest risk in Group 1.

Conclusions: Both statin adherence and persistent PCSK9is use were associated with improved lipid outcomes following revascularization. Persistent PCSK9i use may contribute substantially to LDL-C control, even among patients with poor statin adherence at baseline.

Keywords: Lipid management; PCSK9is; Real-world evidence

Introduction: Idiopathic pulmonary fibrosis (IPF) is a progressive, chronic interstitial lung disease. Although approved antifibrotic agents (pirfenidone and nintedanib) can slow disease progression. Multiple investigational agents have entered clinical trials, a network meta-analysis (NMA) is needed to simultaneously compare their relative efficacy and safety.
Aims: To compare the efficacy and safety of antifibrotic agents and novel therapeutics for slowing disease progression in patients with IPF.
Methods: A systematic review of randomized controlled trials (RCTs) was conducted using PubMed, Scopus, and Google Scholar (inception to March 2026). Outcomes included forced vital capacity (FVC in mL and %predicted), six-minute walk test (6MWT) distance, all-cause mortality, and diarrhea. Risk of bias was assessed using RoB 2, and network consistency was evaluated using the side-splitting approach
Results: Among 44 included RCTs, four agents significantly preserved FVC (mL) versus placebo: Zinpentraxin alfa (MD 126.05; 95%CI 11.21, 240.88), Bexotegrast (MD 106.08; 95%CI 31.72, 180.45), Pirfenidone (MD 86.81; 95%CI 37.35, 136.27), and Nintedanib (MD 74.83; 95%CI 51.05, 98.60). Significant FVC %predicted improvements were observed with BMS-986020 (MD 3.41; 95%CI 0.48, 6.33), Nintedanib (MD 2.99; 95%CI 2.17, 3.80), and Pirfenidone (MD 2.51; 95%CI 1.12, 3.91). Treprostinil significantly improved 6MWT (MD 31.12; 95%CI 13.65, 48.59). For all-cause mortality, Pirfenidone showed the greatest risk reduction (RR 0.60; 95%CI 0.44, 0.80), followed by Nintedanib (RR 0.81; 95%CI 0.67, 0.97). Nintedanib carried the highest diarrhea risk (RR 3.11; 95%CI 2.91, 3.32). Most studies had low risk of bias; network consistency was confirmed for FVC mL (p=0.37) but not FVC %predicted (p=0.004), warranting cautious interpretation of those indirect comparisons.
Conclusion: Standard antifibrotic agents remain the cornerstone of IPF treatment. Based on this NMA, pirfenidone showed reduction of all-cause mortality, followed by nintedanib; both significantly preserve lung function. However, diarrhea with nintedanib, require protective management.
Keywords: Idiopathic Pulmonary Fibrosis; Antifibrotic Agents, FVC, Network Meta-Analysis

Introduction: Medication adherence is a key determinant of glycaemic control and long-term outcomes in people with diabetes. The ABC taxonomy conceptualizes adherence as three interrelated phases: initiation, implementation, and discontinuation. However, existing evidence on adherence determinants is often fragmented and does not consistently distinguish between these phases. Understanding phase-specific determinants is essential for developing targeted interventions to improve adherence.
Aims: To identify factors associated with medication adherence in diabetes and classify these determinants according to the initiation, implementation, and discontinuation phases of the ABC taxonomy.
Methods: An umbrella review was conducted using MEDLINE, EBSCO, and Scopus from database inception to May 2025. Systematic reviews examining factors associated with medication adherence among adults with type 1 or type 2 diabetes were eligible. Determinants were mapped to the ABC taxonomy and five conceptual domains. Methodological quality was assessed using the AMSTAR 2 tool.
Results: Twenty-eight systematic reviews comprising 795 primary studies were included. Most evidence focused on the implementation phase, whereas initiation and discontinuation were less frequently examined. Initiation was primarily influenced by patient-related factors, including fear of injections, health beliefs, and self-efficacy, as well as socioeconomic factors such as income, education, and social support. Implementation was associated with a broader range of determinants, including health literacy, treatment regimen complexity, adverse effects, and financial burden. In the discontinuation phase, financial constraints, insurance coverage, perceived treatment efficacy, and adverse effects were the most frequently reported determinants. The majority of included reviews were rated as moderate to high methodological quality.
Conclusions: Medication adherence in diabetes is a multifactorial and phase-dependent process characterized by distinct yet overlapping determinants across the adherence continuum. Future research should consistently apply the ABC taxonomy to generate phase-specific evidence and support the development of targeted adherence interventions.
Keywords: diabetes mellitus; medication adherence; ABC taxonomy

Introduction: Insulin prescribing has shifted with newer non-insulin agents and insulin formulations, but age‑specific trends in East Asians remain understudied despite high young-onset diabetes and beta-cell deficiency in older adults.

Aims: To examine age-specific trends in insulin use in Hong Kong (2002-2023) and project to 2030.

Methods: We conducted a serial cross-sectional analysis using Hong Kong electronic-health-records (2002-2023) including 313,861 insulin users among 1,148,281 adults with diabetes. We calculated annual age- and sex-standardized proportions, estimated annual-average-percent-change (AAPC), and used exponential smoothing to project trends to 2030.

Results: The number of adults with diabetes increased from 166,559 in 2002 to 836,976 in 2023. The proportion of insulin users decreased modestly (11.9%-to-11.2%, AAPC= -0.4 [ 95% CI: -0.5, 0]), while absolute number increased by 4.5-fold (from 20,096 to 95,623). By age-groups, insulin use declined in the 20-44 (18.7%-to-14.5%), 45-59 (12.1%-to-9.7%), and 60-74 age-groups (11.6%-to-10.6%) in 2020-2023. Across all age-groups, human insulin use decreased substantially, while insulin analogue use increased markedly, particularly among older adults (0.3%-to-3.1%, AAPC=23.2% for 60-74 years; and 0.04%-to-3.7%, AAPC=28.1% for ≥75 years). By 2030, overall insulin use is projected to reach 11.3% or 124,793 users (95% prediction intervals [PIs]: 73,231-90,297]) in Hong Kong. Insulin use will remain highest among adults aged ≥75 years at 13.0% (95% PI:10.7-15.4%), reach 14.5%(95% PI:11.9-17.1%) in the 20-44 age-groups, remain at 10.6% (95% PI: 8.5-12.6%) in the 60-74 age-group, and decline to 8.9% (95% PI: 4.9-12.8%) among the 45-59 age-group.

Conclusions: Overall, the proportion of people treated with insulin remained stable at 11%, absolute number of users increased by 4.5-fold due to rising prevalence of diabetes. Overall insulin use declined in younger and middle-aged adults but increased in the oldest group. Projections suggest overall insulin use is expected to remain relatively stable, with the highest levels in the youngest and oldest age-groups.

Introduction: In January 2021, levetiracetam and lamotrigine were listed on the Pharmaceutical Benefits Scheme (PBS) for first-line treatment of epilepsy to include women of childbearing age. This provided subsidised access to these medicines in addition to other first-line treatments for epilepsy (carbamazepine, clonazepam, ethosuximide, nitrazepam, phenobarbital, phenytoin, primidone, sulthiame, valproate).

Aims: To evaluate changes in the 1) utilisation patterns; and 2) estimated government costs for first-line antiseizure medications in Australia, between pre-policy (2017-2020) and post-policy (2021-2025).

Methods: Publicly available (2017-2025) and aggregated PBS dispensing data were used. Data were obtained in monthly intervals, stratified by medication. Corresponding Estimated Resident Populations at 30th June each year were obtained from the Australian Bureau of Statistics. Dispensing rates per 10,000 person-years were estimated, with changes compared over time and pre/post policy change.

Results: Overall, 20,181,161 (levetiracetam (31%), valproate (28%), lamotrigine (20%), and carbamazepine (11%), clonazepam (4.5%), phenytoin (3.6%), ethosuximide (0.6%), primidone (0.6%), phenobarbital (0.4%), nitrazepam (0.2%), and sulthiame (0.05%)) prescriptions were dispensed (2017-2025). Dispensing rates increased over this period, with lamotrigine (123.84 to 222.83 per 10,000 population), and levetiracetam (186.11 to 316.39 per 10,000 population). The dispensing rates of valproate increased slightly (233.15 to 241.53 per 10,000 population). Comparing the dispensing rates from pre-policy to post-policy: lamotrigine dispensing significantly increased by 68%; levetiracetam by 53%; valproate by 14%; and carbamazepine by 10%. Whereas, the dispensing rate of phenytoin decreased by 13% during post-policy compared to pre-policy period. The total government cost for subsidising first-line epilepsy medicines from 2017–2025 was AU$477,680,081. Compared to pre-policy period, the average annualised cost for lamotrigine increased by 13% (AU$915479) and levetiracetam by 29% (AU$4,402,943) after the post-policy change.

Conclusion: Dispensing patterns of levetiracetam and lamotrigine increased significantly after expansion of PBS prescribing restrictions allowing subsidy of these medicines as a first-line use.

Keywords: Antiseizure medications, utilisation, cost

Introduction: Medicines account for 30% of healthcare expenditure among people with type 2 diabetes (T2D), with the cost of prescribed medicines being a significant treatment barrier.

Aims: We evaluated the uptake of Australia’s Pharmaceutical Benefits Scheme 60-day prescription policy (PBS 60DP policy) for T2D medicines introduced in March 2024. The policy allows for double of supply per dispensing, from 30- to 60-days, with potential major cost savings.

Methods: We used national dispensing records of a 10% random sample of PBS eligible people from March/2024 to February/2025 to examine the number of people dispensed at least one 60-day prescription, and the total volume of 60-day dispensings. We calculated savings with T2D medicines using co-payment fees ($31.60 for general beneficiaries and $7.70 for people with government entitlements) and the number of 60-day dispensings observed. We estimated population-level use and savings by multiplying the results by 10.

Results: 32,098 (19.4%) people were dispensed at least one 60-day prescription,with a total of 43,907 dispensings of 60-day prescriptions for general and 57,825 for concessional beneficiaries. Uptake reached 23.5% of people dispensed T2D medicines, and 21% of all dispensings by February/2025. Uptake was highest for fixed-dose combinations, particularly dipeptidyl peptidase 4 inhibitors + sodium-glucose co-transporter 2 inhibitors, 35.9%. People who were older people, with more comorbidities, and dispensed more T2D medicines had higher usage of 60-day prescriptions. Savings reached $1,387,461 for general patients and $445,252 for people with government entitlements.

Conclusions: The uptake of the 60DP policy for T2D medicines in the first year of implementation was modest. Extrapolating our results to the Australian population, we estimate 320,980 people accessed at least one 60-day prescription for T2D, equating to $18 million in savings in co-payments. Targeted education for clinicians and patients should further bolster policy uptake.

Keywords: Prescription duration, Pharmacotherapy, Pharmacoepidemiology, Health Policy, Type 2 diabetes

Introduction: People with heart failure (HF) often have a high comorbidity burden and use multiple medicines. Many medicines are known to potentially exacerbate HF or increase the risk of harm, yet there is limited information on the scope of use.

Aim: We aimed to evaluate real-world use of potentially inappropriate medicines following HF in Australia.

Method: We performed a population-level retrospective study among adults hospitalised with HF (primary or secondary diagnosis) in NSW, 2020. We identified medicines which may potentially cause harm based on a European Society of Cardiology position paper on potentially inappropriate prescribing for HF with reduced ejection fraction. We measured prevalent use in the 120 days following first hospitalisation, and in a subset of people using medicines indicated for HF with reduced ejection fraction.

Results: We identified 21,794 people hospitalised for HF; 53.3% male, mean age 77, using a mean of 8.5 medicines. Two-thirds (65.2%) of people dispended at least one medicine of interest, with 33% having two or more, and 40.7% with a medicine considered to have a major magnitude of effect. The most prevalent classes were corticosteroids (30.7%), beta-2 agonists (22.4%) and antiarrhythmics (12.2%). Prevalent use highest in people of older ages and using multiple medicines. We found similar rates in people using medicines indicated for reduced ejection fraction (64%).

Conclusion: Two-thirds of people hospitalised with HF were using medicines which could potentially exacerbate risk. While use of these medicines may be appropriate in many cases, prescribers must remain vigilant to ensure the risks and benefits of specific therapeutics for both HF and comorbidities are carefully managed.

Keywords: heart failure, potentially inappropriate prescribing, polypharmacy

Introduction: Lifelong antiretroviral therapy (ART) requires sustained access and adherence. Modern regimens simplified treatment, but dispensing data can show periods where a complete regimen is unavailable. Cumulative proportion of days covered (PDC) can hide short gaps. Evidence on rolling adherence and prolonged gaps is limited.

Aims: To describe first observed complete ART regimens among adults from 2013 to 2023; to estimate suboptimal 90-day adherence and >90-day gaps within two years of follow-up.

Methods: We conducted a retrospective cohort study using linked national dispensing records, mortality data, and census-derived variables. Adults with complete oral ART regimens assembled within 14 days were followed for up to 730 days. Quantity dispensed and standard adult daily dosing were used to estimate days supplied; early refills were carried forward; daily complete-ART coverage required all regimen components to overlap. Outcome 1 divided follow-up into consecutive 90-day blocks, and a block was suboptimal if complete-ART PDC was <85%. Outcome 2 was a consecutive >90-day gap in complete ART coverage. Cox models adjusted for index year, age, sex, and census variables.

Results: The cohort included ~16,400 adults. Index regimens shifted from NNRTI--based regimens to INSTI + 2 NRTI. Median complete-ART PDC was 96%. Two in five patients had a suboptimal 90-day block in year 1 and more than half by two years. One in five had a >90-day gap; one in nine restarted after a >90-day gap. INSTI + 2 NRTI had similar adjusted hazards to NNRTI + 2 NRTI for either outcome. Younger age and markers of social disadvantage, especially unemployment/not in labour force, were associated with worse outcomes.

Conclusions: Despite high cumulative PDC and INSTI uptake, frequent short-term suboptimal coverage and gaps remained common. Rolling 90-day adherence and gap outcomes provide insight beyond simple regimen trends and cumulative PDC.

Keywords: antiretroviral therapy; medication adherence; suboptimal treatment

Background: Ketamine and esketamine are novel treatments for major depressive disorder (MDD). Clinical practice guidelines (CPGs) vary across jurisdictions. Clinicians and people living with MDD depend on high‑quality guidance documents to inform treatment decision‑making. However, no study has examined the quality of CPGs and consensus statements related to ketamine/esketamine.

Objective: The primary objective was to assess the quality of CPGs and consensus statements published from 2019-2025 related to ketamine/esketamine for MDD. The secondary objective was to categorise actionable statements in CPGs.

Methods: Searches were conducted in MEDLINE, EMBASE, PsycINFO, Scopus, Web of Science, the Cochrane Library and grey literature. The quality of CPGs and consensus statements was assessed using the Appraisal of Guidelines for REsearch & Evaluation II (AGREE II) and ACcurate COnsensus Reporting Document (ACCORD) checklist, respectively. Actionable statements within CPGs were extracted and mapped to the World Health Organization’s Pharmacological Treatment of Mental Disorders in Primary Health Care framework.

Results: Seven CPGs and 12 consensus statements were included. Forty-one actionable statements (12 recommendations and 29 good practice statements) related to ketamine/esketamine were identified within CPGs (n=1 – 20 per CPG). Most actionable statements provided guidance on pre-initiation steps (n=32), particularly risk/benefit assessment. Two actionable statements in one CPG focused on the post-treatment step. For CPGs, scope and purpose had the highest mean AGREE-II score (84.1%), while applicability had the lowest (48.2%). For consensus statements, the most concordant ACCORD domain was aim and scope and the least concordant domains were related to engagement of interest holders.

Discussion: Recommendations in CPGs predominantly focused on pre-initiation steps for ketamine/esketamine treatment, with few addressing post-prescribing steps. Only one consensus statement described direct input from the public/patients. Although most CPGs and consensus statements clearly described their aims and scope, few adequately addressed the values/preferences of people living with MDD.

Introduction: People with intellectual disability experience high rates of psychiatric and physical comorbidities and are frequently prescribed psychotropic medicines. Although some psychotropic use reflects greater health needs, there are concerns regarding potentially inappropriate prescribing and associated adverse outcomes.

Aims: This study aimed to describe psychotropic medicine dispensing among people with and without intellectual disability between 2014 and 2023.

Methods: This retrospective cohort study used linked disability and health datasets from New South Wales (NSW), Australia, and Pharmaceutical Benefits Scheme (PBS) dispensing records. Psychotropic medicines, classified using Anatomical Therapeutic Chemical (ATC) codes, were extracted from the PBS for people with intellectual disability and a comparison cohort without intellectual disability matched on age, sex, and area of residence. Outcomes included annual prevalence of psychotropic medicine dispensing by class and psychotropic polypharmacy (concurrent use of two or more psychotropic medicines).

Results: Antidepressants were the most commonly dispensed psychotropic medicines among people with intellectual disability, increasing from 16.7% in 2014 to 24.7% in 2023. In contrast, dispensing of hypnotics declined over the study period (from 2% to 1%), while the prevalence of most other psychotropic drug classes remained relatively stable. Throughout 2014-2023, psychotropic use was consistently higher among people with intellectual disability than the comparison cohort; antiepileptics and antipsychotics showed the greatest differences (prevalence ratios: 10.0 [95%CI 9.7-10.3] and 7.8 [95%CI 7.6-8.0] in 2023, respectively). Among psychotropic users, psychotropic polypharmacy was threefold higher among people with intellectual disability (25.4% vs 7.8% in 2023). Antidepressants and antipsychotics were the most common combination for people with intellectual disability (28.1% in 2023).

Conclusions: Persistently higher psychotropic dispensing and polypharmacy among people with intellectual disability over time warrant regular medicine reviews and closer monitoring of prescribing. Further research should examine clinical indications, long-term outcomes, and access to appropriate health supports.

Keywords: intellectual disability, psychotropic dispensing, polypharmacy

Introduction
Pharmacotherapy for type 2 diabetes(T2D) has evolved with the introduction of sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP-1 RA). However, patterns of therapy following metformin and factors influencing treatment selection in Australia remain unexamined.
Aims
To describe post-metformin glucose-lowering drug (GLD) use, examine temporal changes in therapy, and identify predictors influencing treatment selection in Australia.
Methods
Using a 10% sample of the Pharmaceutical Benefits Scheme, this population-based cohort study identified 69,857 people with T2D who initiated at least one other GLDs following metformin in 2014-2024. Treatment regimens were identified using standard covered days (SCD), allowing 2-SCD gaps. Treatment transitions before and after the Australian T2D guideline updated in 2020 were visualized using Sankey diagrams. Risk ratios (RRs) for the second GLD were estimated by age group, male sex, concessional status, and comedication use, with all covariates defined during 12 months before GLD initiation.
Results
Among 176,463 metformin users, 31,537 initiated a second GLD before 2020 and 38,320 from 2020 onward. Mean ages were 60 years, 40% were female, <50% held concessional status, and 30% used lipid-lowering, antihypertensive or antidepressant therapies. Treatment shifted from dipeptidyl peptidase-4 inhibitors (DPP-4i) and sulfonylureas (SU) toward SGLT2i and GLP-1 RA. People aged ≥ 40 were more likely to initiate DPP-4i and SU, whereas younger people were more likely to initiate GLP-1 RA and insulin. Females were more likely to initiate GLP-1 RA. From 2020, age ≥ 40, male sex, and use of heart failure medications predicted SGLT2i initiation. Sensitivity analyses using alternative SCD definitions and 60- or 90-day gap thresholds supported the main results.
Conclusion
Post-metformin GLD use in Australia has shifted toward SGLT2i and GLP-1 RA. However, suboptimal uptake of these medications persists in older adults and individuals with cardiovascular risk.
Keyword
T2D, SGLT2i, GLP-1 RA, post-metformin

Introduction: Since the introduction of DOACs in the United States in 2010, evolving cardiovascular guidelines have increasingly recommended direct oral anticoagulants (DOACs) over warfarin for eligible patients. However, trends in oral anticoagulant (OAC) use among special populations with complex anticoagulation management, including atrial fibrillation (AF), cancer, chronic kidney disease (CKD), and underweight/obesity, remain insufficiently characterized.

Aim: To evaluate trends in DOAC and warfarin use among U.S. adults and subgroups.

Methods: We conducted a cross-sectional study including U.S. adults (≥18 years) from the National Health and Nutrition Examination Survey (NHANES) 2009-March 2020 cycles (n=34,189) to assess trends in OAC use. Survey-weighted analyses were performed. Among patients with AF and a CHA₂DS₂-VA score (a stroke risk stratification tool) ≥2, survey-weighted Poisson regression with robust variance identified factors associated with OAC use.

Results: DOAC use increased by more than 40-fold from 0.03% in 2009-2012 to 1.3% in 2017-March 2020, while warfarin use decreased by nearly half from 1.6% to 0.9%. Similar trends were observed across subgroups. Among adults with AF and CHA₂DS₂-VA ≥2, DOAC use nearly doubled from 27.6% in 2013-2016 to 53.4% in 2017-March 2020. OAC use was more than threefold higher in adults with cancer versus those without cancer (p<0.001), with DOAC use increasing from 0% to 4.2%, primarily driven by apixaban. In CKD stages 3-5, DOAC use increased from 0.2% to 7.0%, with apixaban accounting for 77.4% of OAC use in stage 4 CKD during 2017-March 2020. OAC use increased among overweight/obese adults but decreased among those with normal/low body weight. Among AF patients with CHA₂DS₂-VA ≥2, higher CHA₂DS₂-VA scores were associated with greater OAC use (PR=1.08; 95% CI=1.00-1.17).

Conclusions: From 2009 to March 2020, DOAC use increased substantially among U.S. adults (>40-fold) and across chronic disease subgroups, with apixaban emerging as the predominant DOAC.

Keywords: oral anticoagulant, DOAC, NHANES

Introduction Antibiotic overuse remains a global threat to antimicrobial resistance. Despite the antimicrobial stewardship (AMS) program implemented in the institution, monitoring data on restricted antibiotics consumption and their correlation with antibiotic resistance were inadequate.

Aims This study aimed to: determine trends in restricted antibiotic consumption in the medicine ward, correlate consumption with resistance and design pharmacist interventions targeting pharmacy-related service gaps.

Methods A retrospective cross-sectional study reviewed dispensing records of a Philippine hospital ward from March 2019 to February 2020. Adult inpatients dispensed with at least one dose of intravenous restricted antibiotics were purposively sampled. Excluded were patients not exposed to IV restricted antibiotics, which covered 4th generation cephalosporins, aztreonam, carbapenems, colistin, vancomycin and linezolid. Correlations between antibiotic consumption using Defined Daily Dose (DDD) per 1000 patient-days and antibiotic resistance from hospital antibiograms were analyzed using Pearson correlation. Pharmacist interventions were designed to match gaps in dispensing practices.

Results One-year consumption data of restricted antibiotics revealed increasing trends of ceftazidime consumption (42.2 to 123.9 DDDs/1000-patient days) but decreasing trends for levofloxacin (350.2 to 156.7 DDDs/1000-patient days) and piperacillin-tazobactam (581.6 to 487 DDDs/1000-patient days). A significantly moderately positive association was observed for ceftazidime consumption and A. baumannii resistance (r = 0.61, 95% CI 0.01 – 0.86, p=0.05). Levofloxacin consumption was significantly strongly positively associated with P. aeruginosa resistance (r = 0.71, 95% CI 0.01 – 0.94, p=0.049). Pharmacist interventions proposed include stability chart, reconstitution and storage information, proper dispensing of restricted antibiotics.

Conclusions Decreasing trends in consumption of restricted antibiotics except for ceftazidime could mean that the hospital’s AMS program is working. However, monitoring antibiotic consumption is important in controlling resistance due to significant positive correlations found between antibiotic consumption for ceftazidime and levofloxacin and resistance against A. baumanii and P. aeruginosa, respectively.

Keywords antibiotic consumption, antibiotic resistance, restricted antibiotics

Introduction: Antimicrobial resistance (AMR) is a growing global health concern driven by inappropriate antibiotic use. Community pharmacies often serve as the primary point of care, yet 40-60% of improper use occurs in these outpatient settings. Such practices of misuse and overuse of antibiotics significantly contribute to resistance and are a measurable determinant of treatment failure and increased healthcare burden. Evaluating dispensing patterns is essential for strengthening antimicrobial stewardship.

Aims: The research aimed to assess antibiotic dispensing patterns, identify specific antibiotic classes dispensed, and evaluate utilization patterns within community pharmacies.

Methods: A cross-sectional study was conducted over six months in community pharmacies purposively selected from different urban regions, comprising both retail and clinic-adjacent settings, using non-randomised sampling. Data from 503 antibiotic dispensing records were collected using structured forms, supplemented by interviews with pharmacists and patients. Antibiotics were analyzed using the WHO AWaRe Classification, quantified using Defined Daily Dose (DDD) methodology and evaluated using descriptive statistics.

Results: Out of 503 antibiotics dispensed, the majority (72.56%) had valid prescriptions, while a substantial proportion (21.66%) were dispensed without prescriptions, indicating continued occurrence of non-prescription sales despite existing regulatory restrictions. Penicillins were the most commonly dispensed class (48.90%), followed by cephalosporins (19.88%). AWaRe classification identified 53.87% Access and 42.94% Watch antibiotics. No Reserve group antibiotics were observed. Systemic antibiotic DDD analysis revealed tetracyclines had the highest average value (8.05 DDDs), followed by macrolides (5.48 DDDs). Penicillins (76.51) and cephalosporins (53.92) had considerably higher usage rates and were primary contributors to overall antibiotic utilization.

Conclusions: The ongoing non-prescription dispensing of antibiotics and significant reliance on 'Watch' agents expose critical vulnerabilities in pharmacy practices. These findings necessitate strict policy enforcement and pharmacist-led stewardship interventions to optimize rational drug use and mitigate the escalating threat of antimicrobial resistance.

Keywords: Antimicrobial Stewardship, Antimicrobial Resistance, Community Pharmacy

Introduction: A notification issued in July 2015 in South Korea recommended caution while prescribing long-acting benzodiazepines (BZDs) to older patients.

Aims: The aim of this study was to determine changes in prescription volumes for older patients before and after drug utilization review (DUR) provision for 11 long-acting BZDs.

Methods: Prescription volumes of 11 long-acting BZDs in patients aged 65 years or older were analyzed from May 2011 to December 2019 using data from the National Health Insurance Service-Senior (2002–2019) (NHIS-2024-2-082). Monthly average numbers of prescriptions per 100,000 older people within 50-month periods before and after DUR provision were analyzed.

Results: In total, 728,593 older adults were observed during the study period, of whom 723,059 (99.2%) were prescribed drugs at least once and 360,717 (49.5%) were prescribed one of the 11 long-acting BZDs at least once. The monthly average number of prescriptions per 100,000 older individuals for the 11 long-acting BZDs combined decreased by 16.1% (from 6,394 to 5,362), whereas that for all medications increased by 20.8% (from 140,000 to 169,051) during the same period. The reduction rate was more pronounced in women (-18.2%) than in men (-12.1%). By institution type, primary care clinics showed the largest decrease (-22.4%). Regarding individual BZDs, prescriptions decreased for pinazepam by 100% (0.1 to 0), clorazepate by 99.8% (24.7 to 0.1), mexazolam by 40.1% (18.2 to 10.9), oral diazepam by 29.5% (5,185.9 to 3,658.4), and diazepam injections by 5.1% (288.7 to 274). Conversely, clonazepam and flunitrazepam prescriptions increased by 81.7% (549.3 to 998.1) and 54.8% (310 to 480), respectively..

Conclusions: DUR information was shown to reduce prescriptions for most long-acting BZDs in older patients, thereby improving potentially inappropriate medication use. However, for clonazepam and flunitrazepam, where the impact was less pronounced, further analysis of their prescriptions patterns is needed.

Keywords: drug utilization review, benzodiazepines, geriatrics

Introduction: Trastuzumab is a cornerstone targeted therapy for HER2-positive breast cancer and has historically represented a significant cost burden to the healthcare budget. Subsidy listing of trastuzumab biosimilar in September 2020 following health technology assessment, aimed to improve affordability while maintaining therapeutic equivalence.

Aims: To evaluate market adoption and baseline cohort comparability of originator versus biosimilar trastuzumab following its subsidy listing in Singapore.

Methods: This retrospective study integrated patient-level dispensing data (2020-2024) from national electronic health records with the national cancer registry for residents with HER2-positive breast cancer. Pre-subsidy records and patients with diagnosis-to-treatment gap over one year were excluded. The cohort was analysed using a modified intention-to-treat framework. Demographic parameters were compared across usage groups using Mann-Whitney U and Pearson’s Chi-Square tests.

Results: The cohort comprised 419 originator-initiated and 456 biosimilar-initiated patients. Post-subsidy listing, biosimilar uptake among new initiations rose from 24.1% in the first year to 63.2%, peaking at 87.7% in Year 3 before stabilizing at 71.6% in the last year. Overall, 38.4% were originator-only users, 42.4% were biosimilar-only users, and 19.2% were switchers. Baseline clinical profiles were comparable across groups, with no significant differences in cancer stage, receptor status (ER/PR), or surgery and radiotherapy utilisation. Biosimilar-initiated patients were marginally older [median 59 (IQR 51-67) vs. 57 (IQR 49-64) years; p=0.025] with slightly longer time to initiation [median 74 (IQR 46-112) vs. 66 (IQR 35.5-112.5) days; p=0.052].

Conclusions: Trastuzumab biosimilar subsidy listing drove a substantial, sustained shift in utilisation patterns with clinical decision-making remaining aligned with baseline practices, as shown by the comparable clinical characteristics across groups. Future research examining the long-term impact of the observed baseline differences in age and time-to-treatment initiation will further strengthen the evidence base for biosimilar adoption, supporting listing policies to optimise long-term expenditure on targeted therapies.

Keywords: Trastuzumab, Biosimilars, real-world outcomes

Introduction: The use of GLP-1-based therapies has expanded rapidly in the management of type 2 diabetes and obesity, yet evidence on their global distribution and the impact of price on consumption remains limited.

Aims: To investigate trends in consumption, expenditure, and price of GLP-1-based therapies stratified by income level, and to identify factors associated with consumption.

Methods: We analysed sales data for GLP-1-based therapies between 2015 and 2024 across 74 countries and regions from the IQVIA-Multinational Integrated Data Analysis System. Consumption was evaluated using defined daily dose per 1000 inhabitants per day (DDD/TID). Expenditure and price were measured in international dollars (Int$) as Int$/TID and Int$/DDD, respectively. Mixed-effects models were used to evaluate temporal trends by average annual percent change (AAPC) and to identify factors associated with consumption.

Results: Global consumption of GLP-1-based therapies increased from 0.25 to 3.49 DDD/TID between 2015 and 2024 (AAPC 42.58%). In 2024, consumption was concentrated in high-income countries (15.71 DDD/TID vs. 0.06 in lower-middle-income countries). A mismatch was observed between consumption and the burden of diabetes and obesity, with several high-burden countries recording minimal consumption. Semaglutide was the most consumed molecule in 2024 (2.23 DDD/TID), while tirzepatide showed the fastest growth following its introduction in 2022 (AAPC 417.14%). Expenditure rose from 2.45 to 52.34 Int$/TID (AAPC 41.64%). Overall price declined slightly (AAPC -0.97%); in 2024, tirzepatide (42.56 Int$/DDD) and semaglutide (13.13 Int$/DDD) were the most expensive. Consumption was negatively associated with price, but positively associated with gross national income, health expenditure, and diabetes and obesity prevalence (all p<0.01).

Conclusions: Global consumption and expenditure of GLP-1-based therapies increased substantially between 2015 and 2024 but remained concentrated in high-income countries. High prices continue to limit access in lower-middle-income countries despite substantial burdens of diabetes and obesity.

Key words: GLP-1-based therapies; Drug utilisation; Health equity

Introduction: Adherence to secondary prevention therapies after ischemic stroke (IS) or transient ischemic attack (TIA) is often suboptimal, yet the clinical implications of longitudinal adherence trajectories remain unclear.

Aims: To identify longitudinal adherence trajectories to statins and antithrombotic agents and evaluate their associations with LDL-C control and cardiovascular outcomes.

Methods: Using the Chang Gung Research Database linked to Taiwan’s population claims data, we identified patients with first-ever IS/TIA between 2012 and 2018 who survived ≥1 year and initiated secondary prevention. Monthly adherence was measured using proportion of days covered and classified by group-based multi-trajectory modeling. The primary outcome was a composite of recurrent IS, systemic embolism, and myocardial infarction; secondary outcomes included individual components and all-cause mortality.

Results: Among 13,299 eligible patients (mean age 65.6 years; 60.3% male), four adherence trajectories were identified: dual high adherence (46.7%), antithrombotic-only adherence (35.1%), early dual discontinuation (12.2%), and gradual dual decline (6.0%). Distinct trajectories demonstrated markedly different LDL-C responses and cardiovascular risks. Dual high adherence achieved the greatest LDL-C reduction (−24.3%) and lowest mortality. Compared with dual high adherence, early dual discontinuation was associated with higher risks of the composite outcome (adjusted hazard ratio [aHR], 1.57 [95% CI, 1.31–1.88]), recurrent IS/systemic embolism (aHR, 1.60 [95% CI, 1.31–1.95]), myocardial infarction (aHR, 1.48 [95% CI, 1.02–2.14]), and all-cause mortality (aHR, 1.56 [95% CI, 1.36–1.79]). Poor adherence had a greater adverse impact in patients with baseline LDL-C ≥100 mg/dL, younger adults, and men.

Conclusions: Longitudinal adherence trajectories were strongly associated with LDL-C control and major cardiovascular outcomes. Sustained dual adherence conferred substantial cardiovascular protection, whereas early discontinuation markedly increased recurrent ischemic and mortality risks. These findings highlight adherence as a modifiable determinant of cardiovascular risk and support targeted adherence-improvement strategies to optimize secondary prevention after stroke.

Keywords: ischemic stroke, medication adherence, secondary prevention

Introduction: Janus kinase inhibitors (JAKi) have expanded the therapeutic armamentarium for autoimmune diseases but may compromise immune surveillance against viral infections. Consequently, hepatitis B virus (HBV) reactivation in patients with different serological profiles remains a critical safety concern. While guidelines classify hepatitis B surface antigen (HBsAg)-positive individuals as high risk, real-world evidence comparing specific JAKi subtypes and assessing resolved HBV cases is limited.
Aims: This study aimed to evaluate the incidence and clinical characteristics of HBV reactivation among patients with varying baseline HBV serostatus receiving JAKi therapy. A secondary objective was to assess the impact of JAK1 selectivity and baseline surface antibodies on reactivation risk.
Methods: This retrospective cohort study included adult patients with autoimmune diseases treated with JAKi for at least 90 days between January 2021 and December 2025. Patients were stratified into HBsAg-positive and HBsAg-negative/core antibody (HBcAb)-positive (resolved HBV) cohorts. Incidence rates (IR) and hazard ratios (HR) for HBV reactivation, defined by standardized viral load elevations or seroreversion, were calculated using descriptive statistics and Cox regression analyses.
Results: The analysis included 194 patients, comprising 19 HBsAg-positive and 175 HBsAg-negative/HBcAb-positive individuals. Among HBsAg-positive patients without antiviral prophylaxis, HBV reactivation occurred in 26.3% (5/19), with an IR of 96.7 per 1000 person-years. Patients receiving upadacitinib and baricitinib demonstrated adjusted HRs of 4.49 (95% CI 0.67-30.02) and 4.44 (95% CI 0.36-54.89) for reactivation, respectively, compared to tofacitinib. In the resolved HBV cohort, the overall reactivation rate was 2.9% (5/175). The risk was notably higher in the anti-HBs negative subgroup (4/44 events) compared to the remaining resolved cohort (1/131 events).
Conclusions: JAKi therapy poses a risk of HBV reactivation in HBsAg-positive patients without antiviral prophylaxis, underscoring the necessity of pre-treatment antiviral coverage. For patients with resolved HBV, preemptive monitoring is generally safe, but baseline anti-HBs negativity identifies a subgroup requiring heightened vigilance.

Introduction:
Poor adherence to inhaled corticosteroids (ICS) remains a major contributor to uncontrolled asthma and increased exacerbations. In Singapore, Malay patients experience a disproportionately high burden of asthma and lower adherence to controller medications. However, limited research has explored the culturally specific determinants of ICS adherence and how these can inform intervention development.

Aims:
Therefore, this study aims to identify barriers and facilitators to ICS adherence among Malay adults and to co-design a culturally appropriate intervention with multiple stakeholders.

Methods:
This is a multi-stakeholder qualitative study conducted in Singapore using a two-round co-design approach. In Round 1, we are currently conducting semi-structured interviews among 10 Malay patients and 10 healthcare providers. Interview guides are informed by the Socio-Ecological Model and the COM-B framework. In Round 2, patients, healthcare providers, and cultural leaders will participate in co-design workshops to translate identified barriers and facilitators into intervention components.

Results:
As of 10 June, we have conducted in-depth interviews with seven healthcare providers and two patients. Our preliminary analysis identified common barriers, including limited health literacy, misconceptions regarding ICS safety and dependency, and difficulties linking controller medication use to asthma prevention. Adherence was further complicated by practical challenges such as complex inhaler techniques and cultural influences such as preference for traditional remedies, heavy family commitments in large Malay families, language barriers, and concerns about inhaler use during Ramadan. Healthcare providers also highlighted the positive and negative influences of family support in Malay culture. To improve ICS adherence, participants suggested culturally tailored educational materials, community support groups, and partnerships with religious organisations and mosque leaders to improve acceptance of ICS use.

Conclusion: Findings will directly inform the development of a stakeholder-driven educational intervention tailored to the needs of Malay patients with asthma.

Keywords: uncontrolled asthma, adherence, inhaled corticosteroids

Introduction: Insulin remains essential for diabetes management, yet prescribing patterns by insulin type and regimen have changed alongside newer glucose-lowering drugs. Severe hypoglycemia (SH) is a major and preventable complication, particularly among insulin-treated patients.

Aims: To examine territory-wide trends in human and analogue insulin use and SH incidence as well as identify associated risk factors.

Methods: We conducted a territory-wide serial cross-sectional analysis of Hong Kong electronic health records (2002-2023) among 277,641 insulin-treated adults with diabetes. Age- and sex-standardized proportions of insulin use and SH incidence (hospitalizations) were calculated annually. Generalized Estimating Equations (GEE) evaluated associations between human versus analogue insulin and recurrent SH hospitalizations, adjusting for demographic and clinical risk factors, medications, prior SH, and participation in the Risk-Assessment-and-Management-Programme-for-Diabetes-Mellitus (RAMP-DM).

Results: From 2002 to 2023, overall insulin use decreased slightly (11.9% to 11.2%). Human insulin predominated but declined (99.6% to 78.1%), while analogue use increased (0.2% to 3.5%), largely driven by long-acting analogues. Basal-only therapy remained most common but declined (6.4% to 4.2%), followed by premixed insulin (2.5% to 2.2%). SH incidence decreased markedly (17.0 to 5.6 events/100-person-years). SH incidences were lower among analogue than human insulin users in 2023 (5.0-vs-6.6 events/100-person-years, adjusted incidence rate ratio [IRR]=0.64 [0.62-0.67]). Compared with basal-only therapy, basal-bolus therapy had the highest SH incidence in 2023 (2.2-vs-12.5 events/100-person-years, IRR=3.51 [3.43-3.59]). Use of glucagon-like peptide-1 receptor agonists (GLP-1RA) (IRR=0.74 [0.68-0.80]), sodium-glucose cotransporter-2 (SGLT2) inhibitors (IRR=0.72 [0.69-0.75]), statins (IRR=0.78 [0.77-0.80]), and RAMP-DM enrolment (IRR=0.88 [0.86-0.91]) were associated with lower SH risk, whereas sulphonylurea and prior SH were associated with higher risk.

Conclusions: Human insulin continued to dominate insulin therapy compared with insulin analogues in people with diabetes, despite evidence of lower SH risk with insulin analogues. Greater uptake of analogue insulin and adjunctive therapies may further reduce SH risk.

Keywords: insulin; trends; risk factors

Introduction: Japanese Encephalitis Virus remains an important mosquito-borne infection in Asia, with continuing exposure despite vaccine availability. Population exposure varies widely across countries, suggesting that demographic and environmental factors may influence transmission.

Aim: To quantitatively synthesise evidence on demographic and environmental risk factors associated with Japanese Encephalitis Virus exposure in Asian populations.

Methods: Published population-based and clinical seroprevalence studies from Asia were included. Data were extracted from studies conducted in India, Nepal, Thailand, Taiwan, Japan, and South Korea. Reported seroprevalence, age distribution, sex, rural or urban residence, occupational exposure, agricultural setting, and diagnostic methods were reviewed. The findings were combined to identify common patterns of exposure across different populations.

Results: The included studies showed wide variation in Japanese Encephalitis Virus exposure across Asia. Seroprevalence ranged from 8.9% in Japan to 98% in South Korea. Studies from India reported seroprevalence of 14.1% in Manipur and 15.9% in Kerala, while Thailand reported around 44% seropositivity among humans living near monkey habitats. Nepal-based studies showed higher exposure, with JE forming a major proportion of viral encephalitis cases and outbreaks. Older age groups generally showed higher antibody prevalence, suggesting cumulative lifetime exposure. Rural residence, agricultural activity, rice cultivation, pig rearing, and proximity to mosquito breeding environments were repeatedly associated with increased exposure. Male predominance was observed in some studies, although this pattern was not consistent across all regions.

Conclusion: Japanese Encephalitis Virus exposure in Asia is shaped by both demographic and environmental determinants. Rural living, agricultural ecology, animal reservoirs, and increasing age appear to be important contributors to exposure. These findings support risk-based vaccination planning, targeted surveillance, and integrated vector control in high-risk Asian populations.

Keywords: Japanese Encephalitis Virus, Seroprevalence, Risk Factors

Introduction: Indonesian population is aging rapidly, with adults aged 60 years and over projected to exceed 20% by 2050. Long‑term care infrastructure in Indonesia remains fragmented, and medication use in residential aged care is still poorly studied . Prevalence of polypharmacy especially, for residential aged-care settings in Indonesia is absent.
Aims: To describe the prevalence of polypharmacy among residents in aged-care facilities in Indonesia and to explore how socio-demographic factors including age, sex, types of aged care, multimorbidity, functional status, frailty, cognition, fall risk, and recent hospitalisation are associated with polypharmacy.
Methods: A cross-sectional study using a convenience sample of 45 consenting residents aged ≥60 years in two aged care facilities in Indonesia with different medication-management protocols. Medicines history was obtained through structured review using a standardized identification form, excluding supplements, vitamins, and traditional preparations. Frailty, functional status, cognitive function, and fall risk were assessed using a validated instrument.
Results: Median age was 79 years ([IQR] 70–82); 71.1% were female and 88.9% had multimorbidity. Polypharmacy was observed in 33.3% of respondents (15/45; 95% [CI] 21.4–47.9%) with excessive polypharmacy (i.e., > 10 medicines) in 4.4%. A significant positive association was observed between active medicine count and chronic condition (Spearman ρ = 0.34, p = 0.02). Crude odds ratios for frailty (3.25), functional dependency (2.04), fall risk, hospitalisation, and female sex were directionally positive but non-significant. Nearly two-thirds (64.4%) reported concurrent use of at least one traditional or herbal preparation in addition to prescribed medicines.
Conclusions: Polypharmacy affected one in three residents (33.3%) of Indonesian aged‑care residents. Among the factors examined — age, sex, types of aged care, multimorbidity, functional status, frailty, cognition, fall risk, and recent hospitalisation — only multimorbidity showed a statistically significant association with polypharmacy.
Keywords
polypharmacy; aged-care facilities; Indonesia

Introduction:
Older adults with type 2 diabetes are at increased risk of adverse events associated with potentially inappropriate glucose-lowering medications. Post-discharge medication reviews are an Australian Government funded service to reduce medication-related harm at transitions of care.

Aims:
To investigate whether provision of timely post-discharge mediation reviews impacts the use of potentially inappropriate glucose-lowering medications in older adults aged 65 and above.

Methods:
Administrative datasets were used to identify older adults with pre-admission use of potentially inappropriate glucose-lowering medications discharged from private and public hospitals from July 2013 to December 2017 in Victoria, Australia. The exposure was completion of a comprehensive pharmacist-led medication review within 90 days of discharge. Individuals were required to survive at least 180 days post-discharge. Discontinuation was defined as no dispensing of the pre-admission medication 91-180 days post-discharge. Inverse probability of treatment weighted regression models compared discontinuation rates between groups.

Results:
Of the total 44,510 older adults, 983 (2.2%) with pre-admission use of potentially inappropriate glucose-lowering medications received a medication review within 90 days post-discharge. In the weighted analysis, adults who received a medication review were 17% more likely to discontinue a potentially inappropriate glucose-lowering medication that those who did not receive a medication review (weighted relative risk: 1.17, 95% CI 1.14-1.20). Potentially inappropriate metformin and sulfonylureas were more likely to be discontinued in those who received a medication review (weighted relative risk: 1.18, 95% CI 1.14-1.22 and 1.17 95%CI 1.14-1.21 respectively).

Conclusions:
Post-discharge medication reviews may assist in the discontinuation of potentially inappropriate glucose lowering medications, particularly metformin and sulfonylureas. However, the low uptake of medication reviews highlights the need for strategies to improve provision in this high-risk population.

Introduction: Neuropsychiatric symptoms (NPS) affect 90% of people with dementia, contributing to caregiver burden and institutionalisation. NPS co-occur in heterogeneous patterns yet are rarely characterised systematically. Identifying distinct NPS clusters may improve risk stratification and guide targeted intervention in dementia care.

Aims: To identify and characterise distinct NPS clusters, examine predictors of class membership, and assess their association with dementia progression.

Methods: Data were obtained from the National Alzheimer's Coordinating Centre database. Adults with dementia at first visit aged ≥45 from 2015, with ≥1 follow-up visit, were included (N=2,007). NPS were assessed using the Neuropsychiatric Inventory Questionnaire across 12 domains, scored ordinally as absent, mild, moderate, or severe. Dementia stage was assessed using the Clinical Dementia Rating Sum of Boxes (CDR-SB). Covariates included demographics, dementia-related factors, functional status, medication use, and comorbidities. Latent class analysis identified NPS clusters; multinomial logistic regression examined predictors of cluster membership; and linear mixed-effects models estimated CDR-SB trajectories across clusters.

Results: Three NPS clusters were identified: low-burden (50%, N=999), medium-burden (37%, N=744), and high-burden (13%, N=264) groups. The high-burden cluster featured moderate-to-severe symptoms across all domains, with the highest baseline CDR-SB and functional impairment. Frontotemporal dementia (Relative Risk Ratio [RRR]: 7.45, 95%CI: 4.23–13.1), Lewy body dementia (RRR 2.82, 95%CI: 1.31–6.08), hypertension (RRR 2.11, 95%CI: 1.23–3.64), and traumatic brain injury (RRR 1.68, 95%CI: 1.02–2.75) were associated with the high-burden cluster. White race (RRR 0.15, 95%CI: 0.07–0.31) and diuretic use (RRR 0.37, 95%CI: 0.16–0.84) were negatively associated. CDR-SB increased over time (β=1.73 points/year, p<0.001); however, the rate of progression did not differ across NPS clusters.

Conclusions: Three clinically distinct NPS clusters exist in dementia, differentiated by dementia subtype, comorbidities, and demographics. Future research should investigate whether NPS cluster-targeted interventions can improve outcomes and reduce caregiver burden in people with dementia.

Keywords: neuropsychiatric symptoms; dementia; latent class analysis.

Introduction: Anticholinergic medications, widely prescribed to older adults, have been linked to dementia risk in prior studies. However, real-world anticholinergic use often involves dynamic patterns of discontinuation and resumption that most existing studies, relying on single-time-point or cumulative-dose exposure measures, have failed to capture or to account for frailty, a key effect modifier of risk in older adults, leaving their differential association with dementia risk unclear.

Aims: We aimed to identify distinct longitudinal trajectories of anticholinergic burden in older adults and to evaluate their association with incident dementia, stratified by frailty status.

Methods: Using Taiwan's National Health Insurance Research Database, we conducted a retrospective cohort study of adults aged ≥65 years newly initiating an Anticholinergic Cognitive Burden (ACB) Scale-listed medication in 2012. Group-based trajectory modeling (zero-inflated Poisson) was applied to 3-year quarterly ACB scores. Cox proportional hazards models assessed associations between trajectories and incident dementia over an 8-year follow-up, adjusted for age, sex, multimorbidity frailty index, and dementia-related comorbidities, with subgroup analyses by frailty status.

Results: Among 63,250 older adults (mean age 76.9 [SD 6.3] years; 50.5% female), 12,996 incident dementia cases occurred during follow-up. Five distinct ACB trajectories were identified: Low(40.2%), Declining(13.5%), Increasing(12.2%), Constantly moderate(26.7%), and Constantly high (7.5%) use group. Relative to the Low use group, the Constantly high group carried the highest risk (aHR 1.32, 95% CI 1.24–1.41), with smaller but significant elevations in Declining (1.08, 1.02–1.14), Increasing (1.07, 1.02–1.14), and Constantly moderate (1.07, 1.02–1.12). Risk associated with the Constantly high trajectory remained significant across robust, mild, and moderate frailty strata but not in severe frailty (HR 1.16; 95% CI, 0.79–1.69).

Conclusions: The Constantly high trajectory showed significantly elevated dementia risk in older adults, supporting targeted anticholinergic burden reduction in patients with sustained high-burden use rather than uniform intervention.

Keywords: anticholinergic burden; group-based trajectory modeling; dementia

Introduction: Melatonin is commonly prescribed for sleep disturbance in older adults and sundowning in those with dementia. However, its use in residential aged care facilities (RACFs) remains understudied.

Aims: The objective of this study was to investigate the prevalence of melatonin use and concomitant psychotropic medication use in Australian RACFs.

Methods: A cross-sectional analysis was conducted involving 13,550 residents across 162 RACFs in New South Wales and Queensland, Australia. Records of prescribed medications (melatonin, orexin receptor antagonists [ORAs, e.g., suvorexant], benzodiazepines [e.g., temazepam, oxazepam], Z-drugs [e.g., zolpidem, zopiclone], antipsychotics, antidepressants and medications for dementia) were extracted from all electronic resident medication profiles in January 2026.

Results: The point prevalence of prescribed melatonin was 10.6% (n=1,437), including 9.2% (n=1,245) regular and 1.6% (n=220) pro re nata (PRN) use. The point prevalence of benzodiazepines, Z_drugs, and ORAs was 30.0%, 0.5%, and 0.1%, respectively. Melatonin users had significantly higher concomitant use of benzodiazepines (38.4% vs. 29.0%, P < 0.001), antipsychotics (32.8% vs. 24.9%, P < 0.001), antidepressants (61.5% vs. 50.3%, P < 0.001), and dementia medications (13.6% vs. 9.3%, P < 0.001), compared to non-users. Notably, concomitant use of three or more psychotropic medications (benzodiazepines, antipsychotics, and antidepressants) was nearly twofold higher among residents using melatonin (13.1% vs. 7.6%, P < 0.001).

Conclusions: One in ten residents were prescribed melatonin, mostly prescribed for regular dosing. Melatonin appears to be used as an adjunct to other psychotropic medications. Further research is warranted to evaluate the appropriateness and safety of melatonin use in RACFs, particularly within the context of high psychotropic burden.

Keywords: melatonin; nursing homes; psychotropic

Introduction: Efforts to reduce opioid prescribing in nursing homes may lead to substitution with other medications, such as gabapentin, cannabinoids, and non-steroidal anti-inflammatory drugs (NSAIDs) for pain symptoms, which also pose health risks among older adults.

Aims: To describe temporal changes in opioids, gabapentinoids, non-prescription analgesics, and cannabinoids in US nursing homes.

Methods: We conducted a serial cross-sectional study using electronic health record data from over 5,000 nursing homes. Each quarter between January 2016 and December 2023, we included residents aged 18-110 years with a Minimum Dataset assessment. We used the electronic medication administration record to identify administrations for opioids, over-the-counter analgesics (e.g., NSAIDs), gabapentinoids, and cannabinoids. Relative differences between Q31 and Q1 with 99% confidence intervals (CIs) summarized medication changes over time overall and by pain and cancer diagnosis.

Results: We identified 91,326 residents in Q1 (mean age = 74 years) and 656,722 residents in Q31 (mean age = 75 years). Prescription cannabinoids increased by 25% over time (Q1=0.52%, Q31=0.65%; PR=1.25 (1.10-1.41)), and dronabinol was the most common. NSAIDs and gabapentin increased by 47% (Q1=10.7%, Q31=15.7%; PR=1.47 (1.43-1.51)), and 60% (Q1=15.9%, Q31=24.4%; PR=1.60 (1.56-1.63)), respectively. Opioid use declined from 38.6% to 23.7%, PR=0.61 (0.61-0.62). Cannabinoid prevalence increased with more severe pain symptoms (e.g., among residents in Q31, mild pain=0.59%; moderate pain=0.64%; and severe pain =0.86%), and use was highest among residents with cancer (Q31=1.52%).

Conclusions: Cannabinoid use was low, but growing over time alongside increases in NSAIDs and gabapentin and decreased opioid use. Comparative safety and effectiveness research of these medications for pain in nursing homes is needed.

Key words: Cannabinoids, nursing homes, pain, opioids

Introduction: Benzodiazepines (BZDs) are commonly prescribed to older inpatients in Japan. However, their association with in-hospital fractures remains insufficiently investigated in Asian populations. Few studies have included non-users or accounted for hepatic and renal function.
Aims: To examine the association between incident BZD use and in-hospital fracture risk among older Japanese inpatients aged ≥75 years using a new-user design. Subgroup analyses by hepatic and renal function and sensitivity analyses were conducted to evaluate robustness.
Methods: A retrospective cohort study used the Medical Data Vision database covering approximately 30% of acute-care hospitals in Japan from April 2012 to March 2017. Patients aged ≥75 years at first hospitalization with a 180-day washout period were included. Incident BZD users (n = 50,493) initiated BZDs during the index hospitalization without prior use in the preceding 180 days; non-users (n = 115,293) had no BZD dispensing during hospitalization. Fractures were defined as hospital-acquired ICD-10 fracture codes with fracture-related procedures after BZD initiation, and multivariable logistic regression was used to estimate associations, adjusting for age, sex, Charlson Comorbidity Index, and hepatic (GPT/ALT) and renal (eGFR) function.
Results: Among 165,786 Japanese inpatients aged ≥75 years, 200 experienced in-hospital fractures. Incident BZD users had a significantly higher fracture rate than non-users (0.16% vs. 0.10%, p = 0.003). After adjustment, incident BZD use was independently associated with a 1.5-fold increased fracture risk (OR: 1.51, 95% CI: 1.14–2.00, p = 0.005). Five sensitivity analyses demonstrated significant associations (OR range: 1.41–2.39). Subgroup analyses by hepatic (GPT/ALT) and renal (eGFR) function showed no significant interactions (p = 0.216; p = 0.578), indicating similar fracture risk regardless of organ function.
Conclusions: Incident BZD use was significantly associated with in-hospital fracture risk among older Japanese inpatients. These findings support careful evaluation of BZD prescribing in Japanese acute hospitals.
Keywords: benzodiazepines, new-user design, in-hospital fracture

Introduction: Polypharmacy is common in older adults and is associated with an increased risk of adverse outcomes, including hospitalisation and mortality. However, the specific medication combinations contributing to polypharmacy remain poorly characterised.

Aims: To identify common medication combinations among older adults experiencing continuous polypharmacy.

Methods: We conducted a nationwide, population-based cross-sectional study in Australia. Individuals aged ≥65 years in 2022 with continuous polypharmacy were included. Continuous polypharmacy was defined as the dispensing of ≥5 unique medications during both 1 April–30 June and 1 October–31 December. Medication exposure was obtained from the Pharmaceutical Benefits Scheme (PBS) dataset. Medication classes were classified using the Anatomical Therapeutic Chemical (ATC) system, and classes with < 3% prevalence were excluded. Association rule mining was used to identify frequent medication combinations, with support, confidence and lift as key metrics.

Results: Overall, 1,449,637 older adults with continuous polypharmacy were included. The top three most frequent drug combinations were (i) lipid modifying agents with ACE inhibitor/angiotensin II receptor blockers (support: 54.7%, lift: 1.05), (ii) drugs for acid-related disorders with lipid modifying agents (support: 36.6%, lift: 1.00); and (iii) drugs for acid-related disorders with ACE inhibitor/angiotensin II receptor blockers (support: 33.9%, lift: 0.95). Combinations with the highest lift included: (i) dipeptidyl peptidase 4 (DPP-4) inhibitors, sodium-glucose co-transporter-2 (SGLT2) inhibitors, lipid modifying agents, and metformin (support: 3.0%, lift:12.2); (ii) DPP-4 inhibitors, SGLT2 inhibitors, and metformin (support: 3.4%, lift: 10.1); and (iii) opioids (ATC N02A), other analgesics (ATC N02B) and antidepressants (support: 3.2%, lift: 4.50).

Conclusion: Identified medication combinations largely reflect common multimorbidity patterns in older adults. These patterns may also highlight opportunities for regimen simplification, including the potential role of fixed-dose combination (polypill) approaches. Future research should examine temporal trends and the influence of sociodemographic factors on these patterns.

Introduction: Chronic sleep disorders and use of hypnotics, particularly benzodiazepines and z-drugs, are predisposing factors for dementia. Dual orexin receptor antagonists (DORAs) and melatonin analogues represent safer alternatives and are therefore more commonly prescribed to older patients, even for long-term use. Preclinical studies suggest that, unlike other hypnotics, DORAs may limit the development of cognitive disorders. However, because DORAs were only recently introduced (August 2014 in the US), their long-term cognitive effects remain poorly characterized.

Aims: We aimed to compare the long-term risk of incident dementia following initiation of DORAs versus ramelteon in older adults with sleep disorders using real-world data, assessing risk differences and temporal trends across varying follow-up periods.

Methods: We conducted a retrospective cohort study using electronic health records from the TriNetX Global Collaborative Network. We identified older adults (aged >60 years) with sleep disorders who initiated DORAs or ramelteon (index date), excluding those with prior records of dementia, delirium, or critical care admissions. Kaplan-Meier survival analysis and Cox proportional hazards models were used to compare survival probabilities and risk of incident dementia up to 10 years after index date. We conducted sensitivity analyses with landmark periods at 3 and 5 years after the index date, excluding the individuals who experienced the event within those time windows.

Results: Preliminary results indicate that DORA initiators exhibit a significantly lower risk of incident dementia compared to ramelteon initiators. This significant risk difference persisted in landmark analyses. There was no difference in the risk of negative control events between the two cohorts.

Conclusions: DORAs may represent a safer alternative to ramelteon for mitigating the long-term risk of incident dementia in older adults with sleep disorders. This protective effect appears to be maintained for several years post-initiation, reducing the likelihood of potential protopathic bias.

Introduction

The Aged Care National Antimicrobial Prescribing Survey (Aged Care NAPS) is a one-day voluntary point prevalence survey conducted annually in Australian long-term care facilities (LTCFs) since 2016 to monitor antimicrobial use.

Aims
To examine annual trends in systemic antibiotic use in Australian LTCFs, overall and stratified by participation in the Aged Care NAPS, 2016-2022.

Methods
This population-based retrospective cohort study analysed linked data from the Registry of Senior Australians National Historical Cohort and the Aged Care NAPS, from July 2016 to June 2022. Changes in national annual trends in systemic antibiotic use, measured in mean percentage and number of defined daily doses (DDD) dispensed per 1,000 resident-days, were analysed using linear mixed-effects models.

Results
The number of LTCFs included ranged from 2,703 to 2,714 annually, with the number of LTCFs participating in Aged Care NAPS increasing from 204 in 2016-17 to 587 in 2021-22. The crude prevalence of systemic antibiotic use decreased during the study period, from 67.8% (95%CI 67.6, 68.0) in 2016-2017 to 56.3% (95%CI 56.5, 56.9) in 2021-22 in the overall cohort. When stratified by participation in the survey, use decreased from 63.9% (95%CI 62.7, 65.1) in 2016-17 to 56.2% (95%CI 55.5, 57.0) in 2021-22 in participating LTCFs and from 66.7% (95%CI 66.3,67.0) to 55.7% (95%CI 55.3, 56.1) in non-participating LTCFs. The adjusted annual prevalence of antibiotic use decreased by 2.1% (95%CI -2.4, -1.8) in participating LTCFs and 2.3% (95%CI -2.4, -2.2) in non-participating LTCFs, with no statistically significant differences in the adjusted annual changes between participating and non-participating LTCFs (0.1%, 95%CI -0.2, 0.4). Similar trends were observed in DDD/1,000 resident-days analyses.

Conclusions
The use of systemic antibiotics decreased in Australian LTCFs during 2016-2022, but remains high regardless of participation in the survey. Concerted efforts are required to implement antimicrobial stewardship interventions in this setting.

Introduction: Chronic diseases prevalence among U.S. pregnant women has steadily increased, driving pregnancy-related medications demand. Suboptimal adherence during pregnancy can cause severe adverse perinatal outcomes. Although cost-related nonadherence to medications (CRN) remains a major public health concern, few studies have investigated specifically among pregnant women. Identifying CRN determinants is critical for optimizing maternal-fetal health.

Aims: To compare national CRN prevalence between pregnant and non-pregnant U.S. women. Additionally, to identify risk factors among pregnant women.

Methods: 2024 U.S. NHIS data were used. Pregnancy and CRN were determined via validated self-reported questionnaires. Weighted Wald chi-square test was used to evaluate prevalence differences. Multivariable logistic regression models were used to estimate adjusted odds ratios and identify CRN determinants among pregnant women.

Results: National CRN prevalence among pregnant women was 8.2% (N = 119,068), significantly lower than 16.7% (N = 7,582,790) in non-pregnant women (P < 0.01). Pregnant women exhibited a lower baseline prevalence of depression (19.3 vs. 26.8%, P < 0.05). In the overall population, multivariable regression models showed food insecurity (aOR = 3.65, 95% CI: 2.83–4.70), depression (aOR = 2.28, 95% CI: 1.84–2.83), and cardiovascular diseases (aOR = 1.96, 95% CI: 1.13–3.37) increased CRN risks. Among pregnant women, key risk factors included medical care barriers (aOR = 6.22, 95% CI: 4.94–7.84), food insecurity (aOR = 2.11, 95% CI: 1.59–2.79), and financial hardship (aOR = 2.01, 95% CI: 1.53–2.63). Additionally, bill-related anxiety (aOR = 2.08, 95% CI: 1.60–2.71), asthma (aOR = 1.55, 95% CI: 1.18–2.04), and depression (aOR = 1.77, 95% CI: 1.39–2.24) significantly increased CRN risks.

Conclusion: Pregnant U.S. women demonstrate lower CRN than non-pregnant women, suggesting Medicaid eligibility extension may alleviate the burden on pregnant women. However, clinical monitoring should prioritize pregnant women with asthma, depression, and financial hardship to ensure equitable medication access.

Keywords: Cost-related nonadherence; pregnant women; NHIS

Introduction: Biannual ultrasound (US) is the standard surveillance modality for hepatocellular carcinoma (HCC) in patients with cirrhosis. However, it has limited sensitivity for early-stage detection. Although contrast-enhanced magnetic resonance imaging (MRI) provides superior sensitivity, its use is limited by high cost and contrast exposure. Non-contrast abbreviated MRI (NC-MRI) has emerged as a promising alternative surveillance strategy, but its long-term clinical outcomes have not been evaluated.

Aims: We aimed to evaluate whether NC-MRI-based strategies detect HCC at an earlier stage than US surveillance in patients with cirrhosis. We also assessed the clinical benefits of earlier detection.

Methods: A microsimulation Markov model simulated 10,000 patients with cirrhosis aged 50 years over a 30-year horizon using 6-month cycles. Model parameters were collected by literature review. Three strategies were compared: (1) biannual NC-MRI, (2) alternating NC-MRI and US at 6-month intervals, and (3) biannual US. Outcomes included stage distribution at HCC detection, receipt of curative treatment, cumulative mortality, and life-years.

Results: The proportion of HCC detected at very-early stage was 69.28%, 55.56%, and 13.89% with NC-MRI, the alternating strategy, and US surveillance, respectively. Advanced-stage detection decreased from 44.82% with US surveillance to 12.56% with the alternating strategy and 10.78% with NC-MRI. Compared with US surveillance, NC-MRI-based strategies consistently yielded larger populations in the post-curative treatment state after very-early HCC detection, with the greatest difference observed at approximately 10 years after surveillance initiation. Introducing NC-MRI increased curative treatment rates by 10.34 percentage points, and reduced cumulative mortality by 2.20 percentage points. It also increased life expectancy by 0.56 years.

Conclusions: NC-MRI-based surveillance strategies improved the effectiveness of HCC surveillance in patients with cirrhosis by detecting earlier HCC stages. These findings support incorporating NC-MRI into HCC surveillance to enhance the clinical benefits of surveillance in high-risk patients.

Keywords: hepatocellular carcinoma; non-contrast magnetic resonance imaging; surveillance.

Introduction: Locally advanced head and neck squamous cell carcinoma (LA-HNSCC) has high recurrence rates despite standard chemoradiotherapy. Recent immunotherapy trials, including KEYNOTE-689 and NIVOPOSTOP, have shown promising outcomes with pembrolizumab and nivolumab. However, comparative real-world evidence evaluating clinical outcomes, safety, and cost-effectiveness remains limited, particularly in Low- and Middle-Income Countries (LMICs).

Aim: To compare disease-free survival (DFS), recurrence, safety, and cost-effectiveness between pembrolizumab- and nivolumab-based regimens in LA-HNSCC.

Methods: This real-world observational study included adult patients with LA-HNSCC and Eastern Cooperative Oncology Group performance status 0–1 treated between October 2018 and July 2024. Patients received cisplatin-based radiotherapy combined with either perioperative pembrolizumab 200 mg every 3 weeks or adjuvant nivolumab at 240 mg, 360 mg, or 480 mg schedules. Primary endpoints were 3-year DFS and local/distant recurrence. Secondary endpoints included grade ≥3 adverse events and incremental cost-effectiveness ratio (ICER).

Results: A total of 601 patients were included; 301 received pembrolizumab and 300 received nivolumab. Three-year DFS was 61.1% (95% CI, 52.3–66.5) with pembrolizumab and 60.2% (95% CI, 57.0–68.7) with nivolumab (p=0.08). Local or distant recurrence occurred in 19.0% and 25.5% of patients, respectively (p=0.34). Grade ≥3 adverse events were observed in 44.6% of pembrolizumab-treated patients versus 54.0% in the nivolumab group (p=0.12). Economic evaluation demonstrated limited cost-effectiveness of pembrolizumab, with an ICER of INR 48.5 lakhs for incremental improvement in 3-year DFS.

Conclusion: Pembrolizumab demonstrated comparable DFS with numerically lower severe toxicity compared with nivolumab in LA-HNSCC. However, substantial economic burden limited its cost-effectiveness in LMIC settings. Further prospective studies are needed to optimize patient selection and long-term value-based immunotherapy strategies.

Keywords: Head and Neck Squamous Cell Carcinoma, Immunotherapy, Cost-Effectiveness Analysis

Introduction: Helicobacter pylori (H. pylori) infection is the most important risk factor for gastric cancer (GC). In Korea, the National Cancer Screening Program (NCSP) recommends biennial upper endoscopy from age 40, contributing to early GC detection. However, H. pylori eradication has not been introduced into the screening program. Therefore, the added preventive benefit of integrating H. pylori eradication into the NCSP requires evaluation.

Aims: This study aimed to develop a Markov model incorporating H. pylori infection and GC natural history. Using this model, we sought to identify the optimal age for H. pylori eradication for GC prevention.

Methods: A Markov model with a one-year cycle was developed for a hypothetical cohort of 10,000 individuals aged 20 years over a 60-year horizon. Age-specific transition probabilities were estimated. Five strategies were compared: no screening; biennial endoscopy; and biennial endoscopy combined with one-time H. pylori eradication at age 20, 30, or 40. The primary outcome was cumulative incidence reduction (CIR) relative to no screening, and the GC stage distribution was also assessed.

Results: The Markov model included health states reflecting H. pylori infection status, precancerous lesion progression, and GC stage. Without screening, 490 cumulative GC cases occurred per 10,000 individuals, and regional/distant GC accounted for 55%. With biennial endoscopy screening, GC cases increased to 510 (CIR: -4.1%) but improved early-stage detection, reducing regional/distant GC to 36%. All strategies incorporating H. pylori eradication showed higher CIR (14.5%, 13.6%, and 10.5%) than biennial endoscopy alone, with earlier eradication yielding greater reduction. GC stage distribution was consistent across all eradication strategies, similar to biennial endoscopy alone.

Conclusions: Incorporating H. pylori eradication into the NCSP reduced cumulative GC incidence. Eradication at age 20 yielded the greatest reduction, supporting it as the optimal age for integration into the NCSP.

Keywords: Gastric cancer; Helicobacter pylori; Screening strategy

Introduction: Perioperative major bleeding remains a clinically important complication of hepatectomy and may increase healthcare resource utilization and costs. However, contemporary real-world evidence on its healthcare burden is limited.

Aims: To describe the incidence of perioperative major bleeding and associated healthcare resource utilization and costs among patients undergoing hepatectomy in Japan.

Methods: We conducted a descriptive study using a hospital-based administrative database from Medical Data Vision (MDV) in Japan between 2020 and 2023. Adults undergoing elective open or laparoscopic hepatectomy who received topical hemostatic materials (i.e., primary hemostatics, flowable microfibrillar collagen, fibrin sealants, and sheet-type biological tissue adhesives/sealants) were included. Major bleeding was defined as a composite of intra- or postoperative allogeneic blood transfusion of ≥400 mL or diagnoses of bleeding complications. Postoperative healthcare resource utilization and costs were summarized as medians with interquartile ranges (IQRs) by bleeding status.

Results: A total of 3,948 patients were included, comprising 3,328 open and 620 laparoscopic hepatectomies. Major bleeding occurred across all topical hemostatic material utilization patterns and was generally more frequent in open (35.2%–46.1%) than laparoscopic hepatectomy (15.5%–35.4%). Comparison between patients with and without major bleeding: In open hepatectomy, median (IQR) postoperative length of stay was 15 (11–23) versus 12 (9–16) days, ICU stay 2 (1–4) versus 1 (1–2) days, and postoperative hospitalization costs 643.6 (445.0–1,138.7) versus 460.0 (358.0–653.5) thousand Japanese yen. In laparoscopic hepatectomy, the corresponding values were length of stay 10 (7–15) versus 8 (6–11) days, ICU stay 1 (1–2) versus 1 (1–1) day, and costs 574.6 (401.6–1,036.5) versus 395.5 (313.3–503.1) thousand Japanese yen.

Conclusions: Major bleeding was associated with substantially greater healthcare resource utilization and hospitalization costs following hepatectomy. These findings provide real-world evidence regarding the healthcare burden associated with perioperative major bleeding in contemporary surgical practice.

Keywords: Major bleeding; Healthcare resource utilization; Hepatectomy

Introduction: Hypersensitivity pneumonitis (HP) is an interstitial lung disease (ILD) that may progress to fibrosis, resulting in declining lung function and increased mortality. Several pharmacological treatments are used, including corticosteroids, immunosuppressants, antifibrotic agents, biologics, and other active drugs; however, their comparative benefits remain uncertain.

Aims: To evaluate the efficacy and safety of pharmacological treatments for HP using evidence synthesis through systematic review and network meta-analysis (NMA).

Methods: MEDLINE, Embase and the Cochrane Library were searched from inception to 10 November 2025, without language restrictions. Experimental and cohort studies were eligible if they compared any pair of medications for HP and reported at least one outcome of interest, including changes in forced vital capacity (FVC), diffusing capacity for carbon monoxide (DLCO), all-cause mortality and adverse events. A frequentist NMA was performed to estimate relative treatment effects, expressed as mean differences (MDs) and risk ratios (RRs).

Results: Twenty-one studies were included. Overall, no treatment was significantly superior to placebo in slowing declines in percentage predicted FVC or DLCO. However, in the subgroup with at least 12 months of follow-up, corticosteroids significantly reduced the decline in percentage predicted FVC compared with placebo (MD 7.15; 95% confidence interval [CI] 0.59 to 13.71). For all-cause mortality, azathioprine plus prednisolone showed the greatest risk reduction compared with placebo (RR 0.42; 95% CI 0.24 to 0.74). Regarding safety, nintedanib significantly increased the risk of gastrointestinal adverse events compared with placebo (RR 2.61; 95% CI 1.77 to 3.83), whereas serious adverse events did not differ significantly between treatments.

Conclusions: There is no definitive evidence that any medication is superior to placebo in slowing lung function deterioration in HP. Corticosteroids may help preserve FVC over follow-up periods of at least 12 months, while azathioprine plus prednisolone may reduce mortality.

Keywords: hypersensitivity pneumonitis; Interstitial lung disease; Network meta-analysis

Introduction: Medicine affordability critically determines chronic disease outcomes, with India's out-of-pocket expenditure accounting for 39.4% of total health spending as of 2021-22. Existing instruments assess financial toxicity or medication adherence in isolation; none captures the integrated pathway through which economic stress drives cost-related non-adherence. This represents a measurable gap in patient-reported outcome measurement for chronic disease populations.

Aims: To develop and content-validate the Medication Economic Stress and Decision-Making Scale (MESDS) as a structured PROM, and to characterize its association with non-adherence behaviors and sociodemographic determinants in adults with chronic illnesses.

Methods: A two-phase study was conducted over six months at a tertiary hospital in south India. Phase I involved systematic literature-based item generation (267 items) and expert panel content validation using Item-level and Scale-level Content Validity Indices (I-CVI, S-CVI). Phase II was a cross-sectional analytical study among 507 adults with chronic diseases using the finalized scale; analysis included Spearman rank correlation and multiple linear regression.

Results: The 21-item MESDS achieved an S-CVI/Ave for relevance of 0.94 and for clarity of 0.937. Phase II participants had a mean age of 57.8 ± 13.3 years (67.7% male; 51.1% insured). The mean MESDS Total Score was 34.33 ± 14.10; 57.0% experienced moderate and 13.2% high medication economic stress. Insurance independently predicted a 10.8-point reduction in total score (Beta = −0.385; p < 0.001). Family income, educational qualification, and household size were additional independent predictors, with the model explaining 44.4% of score variance (Adjusted R² = 0.427). Cost-driven non-adherence showed the strongest association with the total score (rho = 0.854; p < 0.001).

Conclusions: The MESDS demonstrated strong content validity and discriminative ability across sociodemographic subgroups. It effectively distinguishes cost-driven from unintentional non-adherence, positioning it as a practical screening tool for chronic disease outpatient settings.

Keywords: Medication economic stress, cost-related non-adherence, patient-reported outcome measure

Introduction: Disseminated dermatophytosis is increasingly associated with persistence, recurrence, and variable response to therapy. Conventional itraconazole remains widely prescribed; however, concerns regarding inconsistent bioavailability have contributed to the use of super-bioavailable (SUBA) formulations. Despite growing clinical use, comparative real-world evidence evaluating their therapeutic and economic outcomes remains limited.

Aim: To compare the clinical efficacy, relapse rates, symptom reduction, and cost-effectiveness of conventional itraconazole and SUBA itraconazole in patients with disseminated dermatophytosis.

Methods: This phase 4 non-inferiority trial used real-world patient data. A 5% non-inferiority margin was pre-specified for the primary endpoint. Adults with disseminated dermatophytosis involving ≥5% body surface area received oral conventional itraconazole 200 mg once daily or SUBA itraconazole 130 mg once daily for 6 weeks, followed by 4 weeks of follow-up. The primary endpoint was clinical cure at 6 weeks. Secondary endpoints included relapse rates, reductions in Physician Global Assessment (PGA) parameters, and Incremental Cost-Effectiveness Ratio (ICER) based on direct medical costs and the percentage reduction in PGA.

Results: A total of 110 patients were enrolled, with 55 patients in each arm. At 6 weeks, clinical cure rates were 73.7% in the conventional itraconazole arm and 68.2% in the SUBA arm, i.e., a difference of 5.5 percentage points (95% CI, 8.0-19.0; p=0.61). Relapse rates were 20% in the conventional itraconazole arm and 6.6% in the SUBA arm (p=0.02). Reduction in PGA parameters after 6 weeks was 11.28% with conventional itraconazole and 13.03% with SUBA itraconazole. The ICER between groups was INR 129.71 per 1-point PGA reduction.

Conclusions: Conventional itraconazole demonstrated clinical non-inferiority to SUBA itraconazole, with slightly higher cure rates. SUBA itraconazole showed significantly lower relapse rates, greater symptom reduction, and marginal cost-effectiveness advantages.

Keywords: Disseminated dermatophytosis, Super Bioavailable Itraconazole, Cost-effectiveness analysis

Introduction: Lifestyle interventions are effective in managing prediabetes, a condition that increases the risk of developing type-2 diabetes if left unmanaged. Community pharmacies are strategically positioned to provide prediabetes management programs, and digital health tools may support their delivery. The PRIME program integrates digital health tools with community pharmacy services to promote healthier lifestyle among adults with prediabetes.

Aims: To evaluate the clinical impact of the PRIME program and understand the barriers and facilitators influencing its implementation.

Methods: PRIME was implemented across 13 community pharmacies in Selangor and Kuala Lumpur, Malaysia. Eligible participants were adults with prediabetes who were overweight or obese. A 6-month two-arm cluster randomized controlled trial was conducted, with weight loss as the primary outcome. Participants in the intervention group received in-depth pharmacist counselling, in-app prediabetes education modules, and peer support, while usual care participants received counselling based on pharmacists’ usual practice. Focus group discussions and one-on-one interviews were conducted to understand experiences, perceptions of implementation, and the influence of PRIME on lifestyle behaviour change. Using convenience sampling, participants and pharmacists involved in PRIME were interviewed. Qualitative data were analysed thematically.

Results: Ninety-one participants (intervention n = 46; usual care n = 45) were recruited. Intervention group participants experienced larger weight loss, with a mean difference of 1.26 kg (p<0.05) between groups. Key barriers were related to the use of digital health tools (technical difficulties, lacking engaging functionalities) and resources (workload management, availability of prediabetes-specific guidelines). Participants perceived the blended approach of face-to-face pharmacist support and technology-assisted interventions useful in supporting them adopt healthier lifestyles.

Conclusions: While PRIME can effectively support healthier lifestyles among adults with prediabetes, barriers related to digital tool usability and limited resources can hinder adoption. Addressing these barriers could enhance the implementation and uptake of these programs.

Keywords: prediabetes, community pharmacy, digital health.

Introduction: Ovarian cancer remains a significant health challenge for Indian women. The treatments has advanced with targeted therapies, but their high prices create issues, especially in places with limited resources. Olaparib paired with Bevacizumab shows survival benefits, but it is expensive. This study looks at how effective and affordable the Olaparib and Bevacizumab combo is versus using Bevacizumab alone.

Aim: To assess the cost-effectiveness of Olaparib plus Bevacizumab versus Bevacizumab alone in ovarian cancer patients in India.

Methods: A Markov model with one month cycles over a 20 year horizon was constructed to capture long term outcomes using incremental cost effectiveness ratio (ICER). Health states included progression free survival, disease progression, and death. Direct treatment costs and quality adjusted life years (QALYs) were derived from clinical trial data. Costs were expressed in Indian Rupees (INR) and converted to Australian Dollars (AUD) for international benchmarking.

Results: Olaparib plus Bevacizumab yielded 1.0535 QALYs at a cost of ₹4,223,900 (≈ AUD 76,030), while Bevacizumab monotherapy yielded 0.7966 QALYs at ₹229,930 (≈ AUD 4,140). Incremental analysis showed an additional cost of ₹1,930,970 (≈ AUD 34,560) with a QALY gain of 0.257. The ICER indicated a substantial rise in expenditure relative to modest QALY improvement, exceeding conventional willingness-to-pay thresholds.

Conclusion: Olaparib with Bevacizumab improves patient outcomes but at significantly higher costs. Given the current threshold of ₹150,000 (≈ AUD 2,800) per QALY in India, the combination therapy is unlikely to be considered cost-effective. These findings highlight the need for policy-level discussions on reimbursement and prioritization of economically sustainable oncology interventions.

Keywords: Ovarian cancer; cost-effectiveness; Olaparib

Introduction
Cardiovascular disease (CVD) affects over 1.3 million Australians. Current risk assessments and guidelines derive risk over 5 years and do not consider lifetime risk, or the workplace productivity impact of CVD.

Aims
To evaluate the health productivity impact of Australian and international risk thresholds for primary prevention of CVD, to determine if earlier intervention would lead to better patient outcomes, assuming cumulative lifetime risk.

Methods
We will use a previously validated microsimulation model, the health economic model for primary prevention of cardiovascular disease (HEM-PPCVD), to project individual health outcomes and CVD-related costs of 10,000 hypothetical adults (starting age, 40-69 years). HEM-PPCVD derives cumulative lifetime risk from causal inference Mendelian randomization studies. We will simulate individuals starting at age 40 under different variations of gender, systolic blood pressure, low density lipoprotein cholesterol level, diabetes and smoking status to determine the age in which the Australian guidelines recommend treatment initiation based on 5% and 10% risk thresholds and compare outcomes to lifetime risk. The model outcomes will include incident and total events prevented, mortality, costs, and productivity adjusted life years (PALY).

Results
Australian and international guidelines and thresholds will be compared to cumulative lifetime risk to determine the PALY impact of early intervention. Formal results are subject to analysis and will be presented at a later stage to this planned analysis abstract.

Conclusion
We seek to determine if rescaling risk thresholds to include patients of low and intermediate risk can improve health productivity and reduce mortality. The goal is to determine if current Australian risk equations and thresholds exclude patients when causal lifetime risk and early intervention is not considered.

Keywords
Microsimulation, Primary prevention, Productivity-adjusted life years (PALY)

Introduction: Population-based evidence on glioma burden and healthcare utilization remains scarce in China, particularly from large RWD sources that integrate claims, tumor registry, and mortality information.

Aims: To provide an exploratory description of glioma incidence, prevalence, survival, mortality, subtype distribution, and annual healthcare costs in Inner Mongolia, China.

Methods: We conducted a retrospective cohort study using 2008–2021 claims records, tumor registry data, and death registry data from Inner Mongolia. Glioma cases and subtypes were identified using diagnosis text and morphology codes. Disease burden and healthcare utilization were summarized overall and by ethnicity and subtype. Cox proportional hazards models and generalized linear models were used to explore mortality risk and annual costs.

Results: Among 21.92 million insured residents, 1,707 glioma patients [median (IQR) age, 50 (39-60) years] were identified, of whom 802 (46.98%) were male, 1,377 (80.67%) were Han, and 282 (16.52%) were Mongolian. Glioblastoma was the most common identifiable subtype, accounting for 57.84%. The 2021 point prevalence was 4.16 (95% CI: 3.89-4.44) per 100,000. The overall all-cause mortality rate was 127.44 (95% CI: 118.74–136.62) per 1,000. The 1-year and 3-year survival rates were 78.00% (95% CI: 75.91%–79.93%), 61.44% (95% CI: 58.92%–63.85%). At the patient level, the median annual total cost was $2,806.20 (839.88–8,111.61) per year. No significant Han–Mongolian differences were observed in mortality or healthcare utilization. After covariate adjustment, glioblastoma was associated with higher all-cause mortality [HR = 1.71 (95% CI: 1.36–2.13), P < 0.001] and higher patient-level annual costs [cost ratio = 2.34 (95% CI: 1.71–3.19), P < 0.001] than non-glioblastoma gliomas.

Conclusions: This large real-world cohort provides evidence on glioma burden and healthcare utilization in China. Despite potential underestimation from incomplete coverage and missing fields, the findings may inform prognostic assessment, treatment planning, and healthcare resource allocation for glioma.

Keywords: Glioma; Disease burden; Healthcare Utilization; Cohort study; Disparities

Introduction: Breast cancer remains a major health burden in Indonesia. Although cancer care is covered under Indonesia's National Health Insurance system, patients may still face direct non-medical costs. While broader financial toxicity and out-of-pocket medical costs have been studied in Indonesia, direct non-medical costs in referral-based cancer care remain poorly quantified.

Aim: This study therefore aims to quantify direct non-medical costs and describe how they vary across patient categories in a referral hospital setting.

Methods: This cross-sectional study was conducted August and October 2024 at Dr. Sardjito Hospital, a tertiary hospital in Yogyakarta, Indonesia. Of 125 patients approached, none refused or were excluded. Adults aged ≥18 years with a breast cancer diagnosis and at least three months hospital care were recruited during outpatient visits. Direct non-medical costs were collected using an interviewer-administered three-month recall questionnaire and included transportation, meals, and accommodation for patients and families. Costs were summarised descriptively, and associations with characteristics were examined using gamma generalised linear models (GLM) with log link for costs per visit; multivariable analysis adjusted for patient characteristics. Ethical approval and informed consent were obtained.

Results: Among 125 participants, 98.4% reported direct non-medical costs over three months. Mean total cost was USD 177.54 (95% CI: 130.40-224.67), median USD 80.10 (IQR: 36.41-174.76). Transportation dominated costs, with patient and family transportation contributing 55% and 27%, respectively. Follow-up visits accounted for 61.4% of 1,988 visits. In GLM analysis, log-distance was associated with higher costs per visit in univariable (coefficient 0.606; 95% CI: 0.470-0.742) and multivariable models (coefficient 0.650; 95% CI: 0.500-0.800).

Conclusions: Participants experienced substantial non-medical costs despite medical coverage. Travel-related expenses were the main burden, suggesting referral-based cancer care may create hidden financial barriers. Findings should be interpreted considering possible recall bias and selection bias from single-site outpatient recruitment.

Keywords: Breast cancer; Direct non-medical costs; Indonesia