While Electronic Prescription Systems (EPS) adoption varies across EU Member States, there's a lack of comprehensive comparative analysis. Existing studies focus on single EPSs, employ diverse methodologies, and lack up-to-date data. This study fills this gap by providing a comprehensive overview of EPS development, functionalities, and usage statistics in each EU Member State. Most EU Member States widely adopted EPS by 2022, with exceptions including Germany, France, and Luxembourg, where pilot projects or just plans existed at that time. Out of the 27 EPSs, 25 employ a similar design featuring a central server and end-user software or web-based applications. Among these, 22 are structured as single national systems. The fundamental technical solution is remarkably similar across the EU. Despite these similarities, functionalities, authentication methods, prescription validity, and medication coverage differ significantly among EPSs. A multinational team, including co-authors from each EU Member State, collected data using a structured questionnaire. The study underscores the need for standardized methodologies in EPS research and emphasizes the importance of comprehensive comparative analysis to inform healthcare policies and digitalization efforts.

Monitoring adverse drug events (ADEs) is critical for pharmacovigilance and patient safety. However, identifying ADEs remains challenging, as suspected or confirmed side effects are often documented solely in the unstructured text of electronic health records (EHRs). Manually reviewing clinical notes to detect ADEs is labor-intensive and time-consuming, highlighting the need for automated methods capable of analyzing and extracting ADE-related information from clinical documentation. In this short communication, we describe our ongoing research on fine-tuning and evaluating a large language model (LLM) for the detection of ADEs in clinical notes. Preliminary descriptive result of this study indicates that ADEs are poorly documented in discharge notes, with less than 15% explicitly linking ADEs to specific drugs, which highlights the need for improved reporting practices.

Telemedicine in primary health care is expected to address many of the issues currently challenging service delivery. However, the impact and effect will depend on who will use the new technology. ObjectiveThe objective of the study was to investigate differences between users and non-users of telemedicine integrated into traditional office-based primary health care. MethodsQuantitative registry-based population study in two regions in the southeast part of Sweden (n = 73,486), comparing users with non-users of telemedicine across the variables sex, age, socioeconomic status (SES), morbidity and health care seeking behaviour (HSB). Two study periods of six months were used (September 2019-February 2020 for Region & Ouml;sterg & ouml;tland, and September 2021-February 2022 for Region Kalmar County) to collect user data. A reference period of 36 months (September 2016-August 2019) was used, to collect data on HSB. ResultsUsers were more often women under the age of 60 and had higher morbidity (measured as resource utilisation) than non-users (p < .001). In contrast, no statistically significant differences were seen between the two groups regarding SES, measured as Care Need Index (CNI). Regarding HSB, a proxy measure (health record entries) showed more entries for users than non-users. ConclusionsOur findings suggest that users are more likely to be women and below the age of 60. Likewise, users also tend to have a greater need for health care services compared to non-users, and they seek health care more often compared to non-users. No differences regarding SES were found.

Potential adverse drug events can be signaled in Clinical Decision Support Systems (CDSSs). This study validated a Swedish CDSS (Janusmed Risk Profile) by investigating associations between calculated risk classifications of drugs with QT-prolonging potential and registered related clinical outcomes. Subjects living in Kalmar County, Sweden, between 2011 and 2020 exposed to risk drugs (risk level I: somewhat increased risk, II: moderate increased risk, III: significant increased risk) were extracted from regional electronic health records and matched to controls (risk level 0: no known increased risk) by age, sex, and index date. Ventricular arrhythmia (VA), Torsade de Pointes, cardiac arrest and death were outcomes followed for one year. Logistic regression analysis was performed adjusted for age, sex, number of drugs, days in hospital and previous diagnosis. Among the 188,453 subjects, a higher proportion of those classified by the CDSS as having a risk of QT prolongation experienced VA compared to controls (risk level I = 0.26%, II = 0.34%, III = 0.71% vs risk level 0 = 0.17%). When adjusting for other risk factors, the association decreased, but risk level III remained significant with OR 2.1 (95% CI 1.6-2.9) compared to controls. Similar results were seen for the other outcomes. Although there was an association between CDSS risk classifications and clinical outcomes, only a few subjects are affected, and other factors, such as previous diagnosis, play an important role. The need for multifactorial CDSS algorithms is thus crucial to better guide prescribers in finding high-risk patients.

Background: Deprescribing (i.e., the process of discontinuing an inappropriate medication) requires time, knowledge, and care, but there is a lack of education, support, and guidelines for this important clinical task. A clinical decision support system (CDSS) aims to influence the quality of care by combining structured medical knowledge with patient-specific information to generate recommendations. Objective: The objective was to examine the need to develop a CDSS for drug deprescribing. Furthermore, this study aimed to examine the obstacles to deprescribing and potential users' requirements for a CDSS for deprescribing. Methods: The qualitative design consisted of semistructured interviews with physicians (n = 10) in Sweden from different disciplines, including geriatrics, primary care and internal medicine. The interviews were conducted using a predefined guide containing multiple questions about any challenges related to deprescribing and the perceived need for a CDSS. A qualitative content analysis was performed to analyse the empirical data. Results: The interviews provided several aspects of the difficulty of deprescribing medicines. The structure and usability of the CDSS knowledge database in clinical practice needs to be ensured from the outset. Physicians needs fast, simple and up-to-date information filtered, summarized and synthesized from reliable sources. The information should preferably be integrated into pre-existing electronic health record. Conclusion: There is a need to develop a CDSS for deprescribing. There is little, if any, guidelines or support for deprescribing, which is regarded as a large obstacle. The current findings contribute to further knowledge regarding the perspective of physicians when deprescribing medication.

Background:  Medications are essential for health care but can cause adverse drug events (ADEs), which are harmful and sometimes fatal. Detecting ADEs is a challenging task because they are often not documented in the structured data of electronic health records (EHRs) . There is a need for automatically extracting ADE-related information from clinical notes, as manual review is labor-intensive and time-consuming.

Objectives: This study aims to fine-tune the pre-trained clinical language model, SweDeClin-BERT, for medical named entity recognition (NER) and relation extraction (RE) tasks, and to implement an integrated NER-RE approach to more effectively identify ADEs in clinical notes from clinical units in Sweden. The performance of this approach is compared to our previous machine learning method, which utilized conditional random fields (CRFs) and Random Forest (RF).

Data Sources: A subset of clinical notes from the Stockholm EPR (Electronic Patient Record) Corpus, dated 2009-2010, containing suspected ADEs based on ICD-10 codes in the A.1/A.2 categories was randomly sampled. These notes were annotated by a physician with ADE-related entities and relations following the ADE annotation guidelines.

Methods: We fine-tuned the SweDeClin-BERT model for the NER and RE tasks and implemented an integrated NER-RE pipeline to extract entities and relationships from clinical notes. The models were evaluated using 395 clinical notes from clinical units in Sweden. The NER-RE pipeline was then applied to classify the clinical notes as containing or not containing ADEs. Additionally, we conducted an error analysis to better understand the model’s behavior and to identify potential areas for improvement.

Results: In total 62% of notes contained an explicit description of an ADE, indicating that an ADE-related ICD-10 code alone does not ensure detailed event documentation. The fine-tuned SweDeClin-BERT model achieved an F1-score of 0.845 for NER and 0.81 for RE task, outperforming the baseline models (CRFs for NER and Random Forests for RE). In particular, the RE task showed a 53% improvement in macro-average F1-score compared to the baseline. The integrated NER-RE pipeline achieved an overall F1-score of 0.81.

Conclusions: Utilizing a domain-specific language model like SweDeClin-BERT for detecting ADEs in clinical notes demonstrates improved classification performance (0.77 in strict and 0.81 in relaxed mode) compared to conventional machine learning models like CRFs and RF. The proposed fine-tuned ADE model requires further refinement and evaluation on annotated clinical notes from another hospital to evaluate the model’s generalizability. In addition, the annotation guidelines should be revised, as there is an overlap of words between the Finding and Disorder entity categories, which were not consistently distinguished by the annotators. Furthermore, future work should address the handling of compound words and split entities to better capture context in the Swedish language.  

One approach to preventing adverse drug events (ADEs), such as harmful drug interactions, is the implementation of clinical decision support systems (CDSS). In an ongoing project, we are investigating the accuracy of the rule-based CDSS currently utilized in Swedish healthcare for predicting ADEs and exploring whether machine learning (ML) can improve these predictions. By analyzing real-world healthcare data from a Swedish region spanning a 10-year period, we show that ML has potential to improve ADE predictions compared to existing rule-based CDSS.

Healthcare is undergoing an unprecedented technological transition to structured documentation in electronic health records (EHR), which has the potential to increase the quality of documentation. However, given the rising demand for direct transfer of data, there is a risk that requirements for more documentation will follow. This study seeks to investigate primary care nurses' experiences of structured documentation with direct transfer to a national quality registry. Nine primary care nurses using structured documentation in their management of chronic obstructive pulmonary disease (COPD) patients were recruited from different Swedish regions. The semi-structured interviews addressed experiences and work procedures when using a structured documentation template with direct data transfer to a quality register. Interviews were transcribed verbatim and analyzed using qualitative content analysis. Data were framed according to five key concepts; patient safety, time-saving work methods, quality of care, equitable care, and professional autonomy. The nurses experienced some barriers in relation to structured documentation but mainly observed benefits, raising the potential to enhance equitable care and safety for patients with COPD in primary care. Professional experience and autonomy were described as important prerequisites in achieving these benefits. The findings from this study can contribute to strengthening the documentation work procedures of nurses.

Adverse drug events (ADEs) occur frequently and are a common cause of suffering, hospitalizations, or death, and can be caused by harmful combinations of medications. One method used to prevent ADEs is by using clinical decision support systems (CDSSs). Janusmed Risk Profile is a CDSS evaluating the risk for nine common or serious ADEs resulting from combined pharmacodynamic effects. The aim of this study was to examine the prevalence of potential ADEs identified using CDSS algorithms from Janusmed Risk Profile. This retrospective, cross-sectional study covered the population of a Swedish region (n = 246,010 inhabitants in year 2020) using data on all medications dispensed and administered. More than 20% of patients had an increased risk of bleeding, constipation, orthostatism, or renal toxicity based on their medications. The proportion of patients with an increased risk varied from 3.5% to almost 30% across the nine categories of ADEs. A higher age was associated with an increased risk of potential ADEs and there were gender differences. A cluster analysis identified groups of patients with an increased risk for several categories of ADEs. This study shows that combinations of medications that could increase the risk of ADEs are common. Future studies should examine how this correlates with observed ADEs.

Sweden is in the process of implementing the National Medication List (NLL). The aim of this study was to explore the challenges with the medication management process, as well as expectation for NLL, from a human, organizational, and technology perspective. This study included interviews with prescribers, nurses, pharmacists, patients, and their relatives and was conducted during March to June 2020, before the implementation of NLL. Challenges were (1) feeling lost with several different medication lists, (2) spending time searching for information, (3) being frustrated at parallel information systems, (4) patients being the carriers of information, and (5) the feeling of being responsible in an indistinct process. The expectations for NLL in Sweden were high, but there were several fears.