Background:Bleeding complications are a major contributor to adverse drug events among older inpatients, particularly in those treated with antithrombotic agents. Timely and accurate detection of bleeding events is essential for improving drug safety surveillance and clinical risk management. Objective:The study aimed to develop and validate automated algorithms for detecting major bleeding (MB) and clinically relevant nonmajor bleeding (CRNMB) events from electronic medical records (EMRs) by combining structured data-based rule models and a natural language processing (NLP) approach, and to evaluate their performance and generalizability against a manually reviewed gold standard and an external dataset. Methods:We conducted a multicenter retrospective study using routinely collected EMR data from 3 Swiss university hospitals. Patients 65 years or older who received at least one antithrombotic agent and were hospitalized between January 2015 and December 2016 were included. To detect MB and CRNMB events, rule-based algorithms were developed using structured data (International Statistical Classification of Diseases, 10th Revision, German Modification [ICD-10-GM] codes, laboratory values, transfusion records, and antihemorrhagic prescriptions), with variables and cutoff values defined according to adapted International Society on Thrombosis and Haemostasis definitions and expert consensus. In parallel, a supervised NLP model was applied to discharge summaries from one hospital. A manual review of 754 EMRs served as the reference standard for internal validation, and the algorithm performance of the structured data algorithms (SDA), NLP, and their combination (SDA+NLP) was evaluated against this manually reviewed gold standard using standard performance metrics. External validation was performed on an independent dataset from the Lausanne University Hospital to assess model robustness and generalizability. Results:Among 36,039 inpatient stays, SDA identified 8.26% (n=2979) as MB and 15.04% (n=5419) as CRNMB cases. ICD-10-GM codes alone detected 28.5% (n=849) of MB and 31.48% (n=1706) of CRNMB cases, while laboratory data contributed most to event detection (n=1994, 66.94% for MB and n=3663, 67.60% for CRNMB). Integrating SDA with NLP improved detection, identifying 12.2% (920/7513) of MB and 27.4% (2062/7513) of CRNMB cases at 1 hospital. The combined model achieved the best performance (sensitivity 0.84, positive predictive value 0.51, F1-score 0.64). External validation on Lausanne University Hospital 2021-2022 data (n=24,054 stays) confirmed the algorithms' reproducibility; the prevalence of MB decreased while CRNMB increased, reflecting evolving clinical practices and antithrombotic use patterns. Conclusions:Our integrated approach, combining SDA with NLP, enhances the detection of hemorrhagic events in older hospitalized patients treated with antithrombotic agents, suggesting its potential usefulness for drug safety monitoring and clinical risk management.

IntroductionHospital admissions due to lower respiratory tract infections (LRTIs) and chronic obstructive pulmonary disease (COPD) represent a substantial healthcare challenge for Switzerland. Demographic shift is expected to exacerbate the burden. Thus, multidimensional projections of healthcare resource utilization, environmental, and (socio)economic impact are necessary to facilitate sustainable healthcare planning for these diseases.MethodsBRONCH-2035 employed historical admissions data (2015-2023) and Swiss governmental population scenario forecasts to project nationwide hospital and intensive care unit (ICU) admissions due to LRTIs and COPD between 2025 and 2035. Projections were then leveraged to estimate future inpatient healthcare expenditures, environmental and socioeconomic burden, employing historic tariff data (SwissDRG) alongside data on disease-agnostic greenhouse gas emissions, Swiss governmental employment and Organization for Economic Co-operation and Development (OECD) economic activity.ResultsCompared to 2025, 10,055 (18.4%) additional hospitalizations and an additional 488 (14.3%) ICU admissions due to LRTI and COPD were projected for 2035, requiring 338 (21.8%) additional hospital and 7 (15.9%) surplus ICU beds. Future admissions are estimated to result in an additional 100.2 million Swiss francs (CHF) (18.3%) in inpatient healthcare expenditures and 1.3 million kg (21.1%) additional CO2-equivalent greenhouse gas emissions. Over the period of 2025-2035, LRTI and COPD-related hospitalizations are projected to cause 2.94 million missed workdays, corresponding to 15,402 lost full-time workers, CHF 1.32 billion in productivity losses, 2.46 billion Purchasing Power Parity (PPP) in lost GDP and CHF 151 million in lost tax revenue.ConclusionsBRONCH-2035 is the first study to project the healthcare, environmental and socioeconomic burdens in LRTI and COPD for Switzerland. Population growth and demographic shift alone are projected to exacerbate LRTI and COPD hospitalizations, ICU admissions, inpatient healthcare expenditures, and bed requirements by 2035, alongside socioeconomic and environmental consequences. These findings provide a robust baseline for healthcare planning, highlighting the need for consistent guideline-concordant prevention and structured outpatient care.

BACKGROUND:Reporting adverse drug reactions (ADRs) is essential for drug safety. In Switzerland, healthcare professionals are legally required to report serious and unlabelled ADRs, yet under-reporting remains widespread. We tested a novel method to increase reporting of ADR-related hospitalizations. METHODS:This retrospective observational study used ADR-indicative ICD-10 codes to screen admissions to four Swiss hospitals, identify suspected drugs and send individual case safety reports (ICSRs) of confirmed cases to Swissmedic. RESULTS:Participating hospitals previously reported ~18 ICSRs annually. During the study period (7/2023-12/2023), 200 ADR-related hospitalizations were reported following a review of 814 pre-filtered admissions. Most ICSR data were available in structured format: Median age of patients was 59 (interquartile range [IQR] 42-75); 87 (44%) males; median of two comorbidities (IQR 1-3); the three most frequent ADRs were 'K52.1 Toxic gastroenteritis and colitis' (11 [6%]), 'T42.4 Poisoning by benzodiazepines' (10 [5%]) and 'R11 Nausea and vomiting' (10 [5%]). Discharge reports contained free-text information on suspected drugs: More than half of ADR-related hospitalizations were caused by antineoplastics (45 [23%]), psycholeptics (38 [19%]), opioids and other analgesics (34 [17%]). The median time to screen a case was 1 min, 8 min to collect, compile and send data of confirmed cases. CONCLUSION:The approach successfully increased reporting of serious ADRs. The time investment for creating ICSRs might soon be rendered obsolete, as most data are available in structured format, and large language models are key to identifying suspected drugs in discharge reports.

INTRODUCTION:Reporting adverse drug reactions (ADRs) is essential for detecting drug risks. Despite legal obligations in Switzerland, underreporting remains an issue. This study assessed practice, knowledge and attitudes towards the spontaneous ADR reporting system among physicians and pharmacists. METHODS:A nationwide cross-sectional survey was disseminated via professional associations to physicians and pharmacists in Switzerland. The 21-item questionnaire assessed reporting practice, knowledge, attitudes, information needs and improvement suggestions. Multivariable regression (odds ratios [ORs], 95% confidence intervals [95% CIs]) examined associations between participant characteristics and reporting habits. RESULTS:A total of 1108 participants (834 physicians, 274 pharmacists) were included. ADRs had been suspected by 589 (53.2%), and 562 (50.7%) had reported ≥1 ADR. Most participants rejected the notion that reporting is pointless (999, 90.2%). Although 716 (64.6%) were aware of reporting obligations, 477 (43.1%) perceived reporting as time-consuming and 270 (24.4%) reported legal concerns. About half indicated no lack of incentives (587, 53.0%) and no concerns regarding personal (580, 52.3%) or patient data protection (545, 49.2%). More than half desired clearer guidance on reportable ADRs (609, 55.0%) and 648 (58.5%) expressed interest in pharmacovigilance training. Reporting ADRs was independently associated with increasing age and training (ORs between 1.28 [95% CI: 1.15, 1.42] and 1.92 [1.32, 2.79]), whereas investing >10 min in reporting was associated with age ≥60 (OR 1.59 [1.07, 2.38]), training (OR 1.28 [1.14, 1.45]) and pharmacist status (OR 1.83 [1.34, 2.51]). CONCLUSION:The fundamental willingness to report ADRs despite a simultaneous lack of specific knowledge indicates a need for targeted information campaigns and training opportunities.

[This corrects the article DOI: 10.1371/journal.pone.0308256.].

OBJECTIVES:To (i) investigate the current state of depression management in Swiss primary care post-COVID-19, focusing on the utilization of guidelines or associated tools, (ii) explore potential associations with depression management, and (iii) evaluate availability of and communication with psychiatrists and psychotherapists. METHODS:A previously developed self-report questionnaire, covering screening, diagnosis, management, and interprofessional collaboration, was distributed online, with a supplementary paper version, to 168 Swiss primary care physicians (PCPs) participating in the Swiss Sentinel Surveillance System. Multivariable logistic regressions explored associations. RESULTS:Of the 168 primary care physicians invited to participate, 116 completed the survey (response rate: 69%). Among these, 61% utilized guidelines for depression management, with statistically significant associations towards increased guideline utilization for PCPs with some psychiatric training as residents (odds ratio [OR] 4.13; 95% confidence interval (95% CI) 1.27, 16.02) and decreased utilization for the age group 60-81 (OR 0.29; 95% CI 0.09, 0.93). Guideline utilization was associated with perceived competency in prescribing antidepressants (OR 3.51; 95% CI 1.21, 11.08). The majority reported difficulties in organizing therapy with mental health specialists and rarely received feedback after patient referrals. CONCLUSION:While the utilization of guidelines among participants was inconsistent, over 60% indicated using either guidelines, tools, or both. The study highlights systemic issues in interprofessional collaboration between PCPs and mental health specialists that require attention.

BACKGROUND: Digital tools are widely utilised to improve communication and information exchange among healthcare professionals. The cantonal hospital in Lucerne was the first to implement the Epic clinical information system in a German-speaking country, including information access for primary care physicians via an electronic health record portal. OBJECTIVES: This study assessed how primary care physicians perceive the communication with hospitals in the canton of Lucerne, including their preferences for discharge summary contents and experiences and utilisation of a regionally implemented electronic health record portal. METHODS: We performed an online survey among primary care physicians and contacted all 323 primary care physicians enlisted as members of the cantonal medical society in Lucerne, Switzerland. RESULTS: A total of 109 primary care physicians completed the online survey (34% response rate). Half of the primary care physicians were satisfied with hospital communication. Three-quarters (n = 83) wanted to be informed of patients’ emergency hospital admission within 48 hours, but only 30% (n = 33) reported being notified. In discharge summaries, primary care physicians expect information on the diagnosis, medication, therapies, and recommendations for follow-up care. A large portion of primary care physicians deemed the electronic health record portal beneficial for patient management. Most primary care physicians utilise the portal to retrieve patient data, but it is rarely used for patient referrals. CONCLUSION: Half of primary care physicians were satisfied with communication with regional hospitals. Primary care physicians reported a lack of timely notifications or reports about emergency admissions, in-hospital deaths, and discharges of their patients. Primary care physicians value the electronic health record portal as a supporting tool for patient management.

Background To enhance patient empowerment, the Cantonal Hospital of Lucerne launched a patient portal (MyChart) in December 2019, granting patients access to their medical records, diagnoses, and laboratory results. Months later, the first COVID-19 case was reported in Switzerland, with the pandemic dramatically affecting health care services. Objective This analysis aims to investigate how the pattern of patient portal registrations evolved during the pandemic, with reference to the spread of COVID-19, as well as local and federal policies. Methods This retrospective observational study analyzed the distribution of patient portal registrations after its introduction at the study site from December 1, 2019, until July 31, 2022. The descriptive analysis included the 7-day mean of registrations, plotted alongside the number of administered COVID-19 tests and COVID-19 vaccinations. This was analyzed concerning predefined time periods and stratified by age and gender. Additionally, an interrupted time series analysis was conducted for the different time periods. Results A total of 126,519 patients registered on the patient portal during the study period, with a slightly higher proportion of female patients (n=66,118, 52.3%) and 11.3% (n=14,259) being 65 years of age or older. The daily registration rate differed substantially over the course of the COVID-19 pandemic, whereby four peaks with >200 registrations per day were identified. The first and third peaks coincide with high COVID-19 testing rates in autumn 2020 and 2021, whereas the second and fourth peaks coincide with the release of the vaccine in spring 2021 and the booster at the end of 2021. These patterns are also reflected in the interrupted time-series analysis: for every transition from one period to the next, the immediate effect of the intervention (level change) is statistically significant with P <.05. Regarding patient portal users aged 65 years or older, only two major peaks in registrations can be identified which coincide with the release of the COVID-19 vaccine and booster. Conclusions The COVID-19 pandemic, with its disease dynamics, including testing and vaccinations, seems to have influenced the number of patient portal registrations. In addition, it appears that patients aged 65 years or older predominantly registered for COVID-19 vaccines.

Objective This scoping review aims to explore the current state of encounter notification systems (ENS) between emergency departments (EDs) and primary care providers (PCPs), focusing on their mechanisms, effectiveness, impacts, and challenges in healthcare settings. Methods A systematic search was conducted using PubMed/MEDLINE and Google Scholar to identify relevant literature on ENS between EDs and PCPs. Eligible studies were selected based on predefined criteria, and data were synthesized narratively. Results The initial search yielded 1,396 articles, with 29 included in the review. Studies highlighted the significance of encounter notifications in improving communication and care coordination between EDs and PCPs, leading to enhanced patient outcomes. However, challenges such as technological barriers, privacy concerns, and variations in healthcare settings were identified. Conclusion ENS play a crucial role in enhancing communication and care coordination between EDs and PCPs. Despite challenges, these systems offer substantial benefits and opportunities for improving patient care in the ED-primary care continuum. Future research should focus on addressing implementation barriers and evaluating long-term impacts to optimize the effectiveness of ENS in this context.

Objectives Palliative patients generally prefer to be cared for and die at home. Overly aggressive treatments place additional strain on already burdened patients and healthcare services, contributing to decreased quality of life and increased healthcare costs. This study characterises palliative inpatients, quantifies in-hospital mortality and potentially avoidable hospitalisations.Methods We conducted a multicentre retrospective analysis using the national inpatient cohort. The extracted data encompassed all inpatients for palliative care spanning the years 2012-2021. The dataset comprised information on demographics, diagnoses, comorbidities, treatments and clinical outcomes. Content experts reviewed a list of treatments for which no hospitalisation was required.Results 120 396 hospitalisation records indicated palliative patients. Almost half were women (n=59 297, 49%). Most patients were >= 65 years old. 66% had an oncologic primary diagnosis. The majority were admitted from home (82 443; 69%). The patients stayed a median of 12 days (6-20). All treatments for 25 188 patients (21%) could have been performed at home. In-hospital deaths ended 64 739 stays (54%); of note, 10% (n=6357/64 739) of in-hospital deaths occurred within 24 hours.Conclusions In this nationwide study of palliative inpatients, two-thirds were 65 years old and older. Regarding the performed treatments alone, a fifth of these hospitalisations can be considered as avoidable. More than half of the patients died during their hospital stay, and 1 in 10 of those within 24 hours.

OBJECTIVES: Due to the increasing complexity of the healthcare system, effective communication and data exchange between hospitalists (in-hospital physicians) and primary care physicians (PCPs) is both central and challenging. In Switzerland, little is known about hospitalists’ perception of their communication with PCPs. The primary objective was to assess hospitalists’ satisfaction with their communication with PCPs. Secondary objectives addressed all information about the referral process and communication with PCPs during and after the hospital encounter. Lastly, the results of a previous survey among PCPs were juxtaposed to compare their responses to similar questions. METHODS: This study surveyed hospitalists in six hospitals in the Central Switzerland region. The survey was sent via email to hospitalists from November 2021 to February 2022. The questionnaire contained 17 questions with single- and multiple-choice answers and the option of free-text entry. Exploratory multivariable logistic regression was used to analyse independent associations. RESULTS: In total, 276 of 1134 hospitalists responded (response rate 24.3%): (1) the majority of hospitalists are satisfied with the general communication (n = 162, 58.7%) as well as with referral letters (n = 145, 52.5%), (2) preferred information channels for referral letters are email (n = 212, 76.8%) and electronic portals (n = 181, 65.5%), (3) the three most important items of information in referrals are: medication list, diagnoses and reason for referral. In multivariable regression, compared to other clinicians, internists independently favoured informing PCPs of emergency admissions of their patients in a timely manner (OR 2.04; 95%CI 1.21–3.49). Comparing responses from PCPs (n = 109), the most prominent discrepancy was that 67% (n = 184) of hospitalists claimed to “always” inform after an encounter, whereas only 7% (n = 8) of PCPs agreed. CONCLUSION: Most hospitalists are satisfied with the communication with PCPs and prefer electronic communication channels. Room for improvement was found around timely transmission of patient information before and after hospital encounters.

Interprofessional collaboration in outpatient palliative care is critical to ensuring good quality of care in the home care sector. We investigated facilitators and barriers (FaBs) of interprofessional collaboration among healthcare professionals who participated in a 6-month pilot of a newly implemented specialised mobile palliative care service (SMPCS) in rural Lucerne. This study used a mixed-methods approach to collect (i) qualitative data on FaBs as perceived by nurses and primary care physicians (PCPs), and (ii) quantitative data across the entire interprofessional collaboration using a validated questionnaire expanded with 10 specific questions about the pilot. Identified facilitators of interprofessional collaboration were (i) use of standardised documents, (ii) clear allocation of responsibilities, (iii) regular exchange and clear communication and (iv) consideration of care coordination. Reported barriers were (i) a deficit of knowledge and experience of palliative care among PCPs and (ii) time constraints. This study provides valuable insights into FaBs of interprofessional collaboration in palliative care. Several recommendations can be drawn for how interprofessional collaboration may be optimised. Awareness of FaBs and their consideration in the implementation phase of new services can strengthen the foundation for a successful interprofessional collaboration.

The World Health Organization (WHO) aims to reduce HCV mortality, but estimates are difficult to obtain. We aimed to identify electronic health records of individuals with HCV infection, and assess mortality and morbidity. We applied electronic phenotyping strategies on routinely collected data from patients hospitalized at a tertiary referral hospital in Switzerland between 2009 and 2017. Individuals with HCV infection were identified using International Classification of Disease (ICD)-10 codes, prescribed medications and laboratory results (antibody, PCR, antigen or genotype test). Controls were selected using propensity score methods (matching by age, sex, intravenous drug use, alcohol abuse and HIV co-infection). Main outcomes were in-hospital mortality and attributable mortality (in HCV cases and study population). The non-matched dataset included records from 165,972 individuals (287,255 hospital stays). Electronic phenotyping identified 2285 stays with evidence of HCV infection (1677 individuals). Propensity score matching yielded 6855 stays (2285 with HCV, 4570 controls). In-hospital mortality was higher in HCV cases (RR 2.10, 95%CI 1.64 to 2.70). Among those infected, 52.5% of the deaths were attributable to HCV (95%CI 38.9 to 63.1). When cases were matched, the fraction of deaths attributable to HCV was 26.9% (HCV prevalence: 33%), whilst in the non-matched dataset, it was 0.92% (HCV prevalence: 0.8%). In this study, HCV infection was strongly associated with increased mortality. Our methodology may be used to monitor the efforts towards meeting the WHO elimination targets and underline the importance of electronic cohorts as a basis for national longitudinal surveillance.

BACKGROUND:Increasing numbers of primary care physicians (PCPs) are reducing their working hours. This decline may affect the workforce and the care provided to patients. OBJECTIVES:This scoping review aims to determine the impact of PCPs working part-time on quality of patient care. METHODS:A systematic search was conducted using the databases PubMed, CINAHL, Embase, and the Cochrane Library. Peer-reviewed, original articles with either quantitative, qualitative or mixed methods designs, published after 2000 and written in any language were considered. The search strings combined the two concepts: part-time work and primary care. Studies were included if they examined any effect of PCPs working part-time on quality of patient care. RESULTS:The initial search resulted in 2,323 unique studies. Abstracts were screened, and information from full texts on the study design, part-time and quality of patient care was extracted. The final dataset included 14 studies utilising data from 1996 onward. The studies suggest that PCPs working part-time may negatively affect patient care, particularly the access and continuity of care domains. Clinical outcomes and patient satisfaction seem mostly unaffected or even improved. CONCLUSION:There is evidence of both negative and positive effects of PCPs working part-time on quality of patient care. Approaches that mitigate negative effects of part-time work while maintaining positive effects should be implemented.

Abstract Background Up to 8% of the general population have a rare disease, however, for lack of ICD-10 codes for many rare diseases, this population cannot be generically identified in large medical datasets. We aimed to explore frequency-based rare diagnoses (FB-RDx) as a novel method exploring rare diseases by comparing characteristics and outcomes of inpatient populations with FB-RDx to those with rare diseases based on a previously published reference list. Methods Retrospective, cross-sectional, nationwide, multicenter study including 830,114 adult inpatients. We used the national inpatient cohort dataset of the year 2018 provided by the Swiss Federal Statistical Office, which routinely collects data from all inpatients treated in any Swiss hospital. Exposure: FB-RDx, according to 10% of inpatients with the least frequent diagnoses (i.e.1.decile) vs. those with more frequent diagnoses (deciles 2–10). Results were compared to patients having 1 of 628 ICD-10 coded rare diseases. Primary outcome: In-hospital death. Secondary outcomes: 30-day readmission, admission to intensive care unit (ICU), length of stay, and ICU length of stay. Multivariable regression analyzed associations of FB-RDx and rare diseases with these outcomes. Results 464,968 (56%) of patients were female, median age was 59 years (IQR: 40–74). Compared with patients in deciles 2–10, patients in the 1. were at increased risk of in-hospital death (OR 1.44; 95% CI: 1.38, 1.50), 30-day readmission (OR 1.29; 95% CI 1.25, 1.34), ICU admission (OR 1.50; 95% CI 1.46, 1.54), increased length of stay (Exp(B) 1.03; 95% CI 1.03, 1.04) and ICU length of stay (1.15; 95% CI 1.12, 1.18). ICD-10 based rare diseases groups showed similar results: in-hospital death (OR 1.82; 95% CI 1.75, 1.89), 30-day readmission (OR 1.37; 95% CI 1.32, 1.42), ICU admission (OR 1.40; 95% CI 1.36, 1.44) and increased length of stay (OR 1.07; 95% CI 1.07, 1.08) and ICU length of stay (OR 1.19; 95% CI 1.16, 1.22). Conclusion(s) This study suggests that FB-RDx may not only act as a surrogate for rare diseases but may also help to identify patients with rare disease more comprehensively. FB-RDx associate with in-hospital death, 30-day readmission, intensive care unit admission, and increased length of stay and intensive care unit length of stay, as has been reported for rare diseases.

To the Editor, Glucagonlike peptide1 (GLP1) is an incretin secreted by the intestine and the central nervous system. In asthma models involving viral or allergen challenge, GLP1 receptor agonists (GLP1RA) reduce the production of interleukin (IL)5, IL13, and IL17, inhibiting airway inflammation and implicating the GLP1 metabolic pathway in the regulation of several immune pathways relevant to asthma.1 Clinically, GLP1RAs are approved for treatment of obesity and type 2 diabetes mellitus (T2DM). Obesity and T2DM are risk factors for worse asthma outcomes, necessitating novel treatment approaches for this highrisk population. Multiple recent studies demonstrate that GLP1RAs may reduce asthma exacerbation risk2 and improve pulmonary function,3,4 yet mechanisms remain undefined. Prior studies of T2DM patients identified GLP1RA associated changes in biomarkers relevant to airway inflammation,4,5 but lacked inclusion of patients with asthma, highlighting the need to identify potential response biomarkers among patients with asthma treated with GLP1RAs. Periostin, induced by IL4, IL13, and transforming growth factorβ, is an established biomarker of treatment response in asthma clinical trials.6 Based on the preclinical literature, we hypothesized that GLP1RA use would reduce serum periostin in patients with asthma. We obtained samples from patients with asthma using GLP1RA or comparator therapies for comorbid T2DM from an institutional biobank for discovery, and samples from a randomized controlled trial (RCT) for validation of periostin as a biomarker of GLP1RA activity in patients with obesity without asthma. Serum samples (2010– 2019) collected at a single timepoint from adults with comorbid asthma and T2DM were obtained from the Mass General Brigham Biobank (“Biobank”), the health system's electronic health record (EHR)linked biorepository. Biobank participants are recruited from multiple inpatient and outpatient settings and reflect the MGB's system's demographic distribution. Participants provide consent for use of clinical, laboratory, and biological sample data. Biobank consented patients represent approximately 2% of MGB's total patient population. We identified demographics, T2DM medication prescriptions, and clinical features including comorbidities, body mass index (BMI), smoking status, asthma medications, and haemoglobin A1c (HbA1c) from patients' EHR. T2DM treatment was determined by patients' active prescriptions (GLP1RA or nonGLP1RA) at Biobank collection date; nonGLP1RA treated patients included alternative intensified diabetes therapies (basal insulin, sulfonylureas) per T2DM treatment guidelines. Asthma severity was determined by asthma medication use at the time of sample collection. Analyses accounted for concurrent metformin use, which is standard firstline T2DM therapy. Comorbid respiratory diseases, systemic prednisone use ≤30 days of collection, and concurrent dipeptidylpeptidase4 inhibitor (DPP4i) use were exclusion criteria to minimize confounding. Serum periostin levels were measured by enzymelinked immunosorbent assay (ELISA; ShinoTest). We performed additional ELISAs to interrogate selected serum biomarkers based on the preclinical literature1,3,5 including total IgE (eBiosciences), IL6 (Abcam), IL8 (Invitrogen), and sCD163 (R&D Systems). Group means were compared between GLP1RA and nonGLP1RA treated patients using the Wilcoxon rank sum test. For adjusted analyses, a propensity score was calculated for GLP1RA use. A linear regression model was fitted for each serum biomarker level with exposure (GLP1RA use) and propensity score as independent variables. Exploratory analyses included asthma severity and sex subgroups. Statistical significance was defined by a twosided pvalue of ≤.05 (SAS version 9.4). Samples from an RCT (NCT03101930) of adults with obesity (BMI ≥30 kg/m2) and prediabetes, randomized 2:1:1 to the GLP1RA liraglutide, the DDP4i sitagliptin (100 mg/day), or hypocaloric diet were obtained for secondary use.7 Eighty of the 88 RCT participants had serum samples available for periostin analysis. Asthma was an RCT exclusion diagnosis. Change in periostin from baseline to 2and 14weeks within each treatment arm was assessed using Wilcoxon signedrank test, and comparison of weight loss between the arms were made using Wilcoxon ranksum test. In the Biobank, 150 patients met inclusion criteria, including 40 GLP1RA treated individuals. Mean BMI in both groups was obese (≥30 kg/m2). Age, sex, BMI, HbA1c, asthma severity, metformin use, smoking status, and insurance coverage did not differ between treatment groups. The GLP1RA treated group had a higher percentage of White patients (p = .03). Mean periostin level across the entire Biobank cohort was 85.54 ± 47.71 ng/ml; mean periostin in the GLP1RA treated group was 70.8 ± 28.2 ng/ml. Periostin was significantly lower in GLP1RA users in crude (p = .039) and propensity score

OBJECTIVES:The scientific literature contains an abundance of prediction models for hospital readmissions. However, no review has yet synthesized their predictors across various patient populations. Therefore, our aim was to examine predictors of hospital readmissions across 13 patient populations. STUDY DESIGN AND SETTING:An overview of systematic reviews was combined with a meta-analytical approach. Two thousand five hundred four different predictors were categorized using common ontologies to pool and examine their odds ratios and frequencies of use in prediction models across and within different patient populations. RESULTS:Twenty-eight systematic reviews with 440 primary studies were included. Numerous predictors related to prior use of healthcare services (odds ratio; 95% confidence interval: 1.64; 1.42-1.89), diagnoses (1.41; 1.31-1.51), health status (1.35; 1.20-1.52), medications (1.28; 1.13-1.44), administrative information about the index hospitalization (1.23; 1.14-1.33), clinical procedures (1.20; 1.07-1.35), laboratory results (1.18; 1.11-1.25), demographic information (1.10; 1.06-1.14), and socioeconomic status (1.07; 1.02-1.11) were analyzed. Diagnoses were frequently used (in 37.38%) and displayed large effect sizes across all populations. Prior use of healthcare services showed the largest effect sizes but were seldomly used (in 2.57%), whereas demographic information (in 13.18%) was frequently used but displayed small effect sizes. CONCLUSION:Diagnoses and patients' prior use of healthcare services showed large effects both across and within different populations. These results can serve as a foundation for future prediction modeling.

In 2013, the Global Coalition for Regulatory Science Research (GCRSR) was established with members from over ten countries (www.gcrsr.net). One of the main objectives of GCRSR is to facilitate communication among global regulators on the rise of new technologies with regulatory applications through the annual conference Global Summit on Regulatory Science (GSRS). The 11th annual GSRS conference (GSRS21) focused on "Regulatory Sciences for Food/Drug Safety with Real-World Data (RWD) and Artificial Intelligence (AI)." The conference discussed current advancements in both AI and RWD approaches with a specific emphasis on how they impact regulatory sciences and how regulatory agencies across the globe are pursuing the adaptation and oversight of these technologies. There were presentations from Brazil, Canada, India, Italy, Japan, Germany, Switzerland, Singapore, the United Kingdom, and the United States. These presentations highlighted how various agencies are moving forward with these technologies by either improving the agencies' operation and/or preparing regulatory mechanisms to approve the products containing these innovations. To increase the content and discussion, the GSRS21 hosted two debate sessions on the question of "Is Regulatory Science Ready for AI?" and a workshop to showcase the analytical data tools that global regulatory agencies have been using and/or plan to apply to regulatory science. Several key topics were highlighted and discussed during the conference, such as the capabilities of AI and RWD to assist regulatory science policies for drug and food safety, the readiness of AI and data science to provide solutions for regulatory science. Discussions highlighted the need for a constant effort to evaluate emerging technologies for fit-for-purpose regulatory applications. The annual GSRS conferences offer a unique platform to facilitate discussion and collaboration across regulatory agencies, modernizing regulatory approaches, and harmonizing efforts.

We know, not just since Daniel Kahneman's book “Thinking, fast and slow”, how prone we are to coming to conclusions by simplifying information.1Kahneman D Thinking, fast and slow.http://dspace.vnbrims.org:13000/jspui/bitstream/123456789/2224/1/Daniel-Kahneman-Thinking-Fast-and-Slow-.pdfDate: 2011Date accessed: October 23, 2023Google Scholar In clinical reasoning, pattern recognition is a valid strategy. However, patterns can be challenged, especially during a pandemic. To provide the best possible care for our patients, we need to critically evaluate our judgements. On Feb 1, 2020, a Chinese man aged 29 years was referred from an outpatient clinic to the emergency department of the Lucerne Cantonal Hospital with a suspected SARS-CoV-2 infection. He had arrived in Switzerland from Portugal 3 days earlier. Since then, he had high fever, headache, muscle pain, and a sore throat. His companion was travelling with him and had no symptoms. Clinical examination confirmed his slightly reduced general condition and febrile state (38·5°C). Laboratory tests showed increased C-reactive protein (63 mg/L; typical range <5 mg/L), a mild thrombocytopenia (130 G/L; typical range 150–330 G/L), and an increased creatinine (114 μmol/L; typical range 59–104 μmol/L), presumably due to volume depletion. He met the criteria for suspicion of SARS-CoV-2 infection (ie, symptoms of a respiratory tract infection or fever) and was isolated. A nasopharyngeal swab detected enterovirus or rhinovirus, but no SARS-CoV-2. When the patient was discharged in improved condition, an outpatient follow-up was planned as thrombocytes had decreased to 19 G/L, without signs of bleeding. The next day, the patient presented again with a headache, a fever that was unresponsive to antipyretics, and vomiting. When asked, he disclosed that he had been working intermittently in Equatorial Guinea during the past 4 years until December, 2019. A blood smear finally revealed plasmodia, with a parasitaemia of 11%. The species could not be identified. A therapy with artesunate 240 mg three times per day and doxycycline 100 mg twice per day was started and given for 24 h. Therapy was changed to oral artemether 20 mg and lumefantrine 120 mg after improvement (four tablets twice per day for 5 days). The patient was discharged in good condition. At the beginning of 2020, the global COVID-19 pandemic had reached Europe.2WHOA timeline of WHO's response to COVID-19 in the WHO European region: a living document (update to version 2.0 from 31 December 2019 to 31 July 2021).https://www.who.int/europe/publications/i/item/WHO-EURO-2021-1772-41523-56652Date: 2021Date accessed: October 23, 2023Google Scholar WHO reacted with surveillance of COVID-19 on Jan 27, 2020,2WHOA timeline of WHO's response to COVID-19 in the WHO European region: a living document (update to version 2.0 from 31 December 2019 to 31 July 2021).https://www.who.int/europe/publications/i/item/WHO-EURO-2021-1772-41523-56652Date: 2021Date accessed: October 23, 2023Google Scholar and measures were implemented to control the spread of SARS-CoV-2. By prioritising the diagnosis of COVID-19 in people who were febrile, the pandemic substantially affected good clinical reasoning. In terms of our case report, relying too much on the race of the patient resulted in anchoring bias and poor diagnosis. Enterovirus or rhinovirus might have explained the symptoms, but were later identified as innocent bystanders (ie, confounders in the diagnostic process). A more comprehensive anamnesis might have revealed the correct diagnosis at the first presentation of the patient. However, labelling SARS-CoV-2 as the “China virus”, which was particularly promoted by former President of the US Donald Trump,3Vazquez M Klein B Trump again defends use of the term 'China virus’.https://edition.cnn.com/2020/03/17/politics/trump-china-coronavirus/index.htmlDate: 2020Date accessed: October 23, 2023Google Scholar might have contributed to racial heuristics. The equation of country of provenance: China in combination with a febrile illness led to misdiagnosis and delay of adequate care. Furthermore, this case report intends to raise awareness in the context of increasing use of health-care algorithms and artificial intelligence that have been reported to be prone to racial bias.4Obermeyer Z Powers B Vogeli C Mullainathan S Dissecting racial bias in an algorithm used to manage the health of populations.Science. 2019; 366: 447-453Crossref PubMed Scopus (1789) Google Scholar, 5Rajpurkar P Chen E Banerjee O Topol EJ AI in health and medicine.Nat Med. 2022; 28: 31-38Crossref PubMed Scopus (438) Google Scholar We declare no competing interests.

Background Effective delirium prevention could benefit from automatic risk stratification of older inpatients using routinely collected clinical data. Aim Primary aim was to develop and validate a delirium prediction model (DELIKT) suitable for implementation in hospitals. Secondary aim was to select an anticholinergic burden scale as a predictor. Method We used one cohort for model development and another for validation with electronically available data collected within the first 24 h of admission. Included were patients aged ≥ 65, hospitalised ≥ 48 h with no stay > 24 h in an intensive care unit. Predictors, such as administrative and laboratory variables or an anticholinergic burden scale, were selected using a combination of feature selection filter method and forward/backward selection. The final model was based on logistic regression and the DELIKT was derived from the β-coefficients. We report the following performance measures: area under the curve, sensitivity, specificity and odds ratio. Results Both cohorts were similar and included over 10,000 patients each (mean age 77.6 ± 7.6 years) with 11% experiencing delirium. The model included nine variables: age, medical department, dementia, hemi-/paraplegia, catheterisation, potassium, creatinine, polypharmacy and the anticholinergic burden measured with the Clinician-rated Anticholinergic Scale (CrAS). The external validation yielded an AUC of 0.795. With a cut-off at 20 points in the DELIKT, we received a sensitivity of 79.7%, specificity of 62.3% and an odds ratio of 5.9 (95% CI 5.2, 6.7). Conclusion The DELIKT is a potentially automatic tool with predictors from standard care including the CrAS to identify patients at high risk for delirium.