BACKGROUND:Nintedanib and pirfenidone are two anti-fibrotic agents for diseases within the interstitial lung diseases (ILDs) spectrum. Here, we provide a comprehensive analysis regarding treatment persistence and adherence rates for the Greek territory. METHODS:This was a retrospective cohort study of patients initiating anti-fibrotic treatment during the period 2019-2023, utilizing data extracted from the National Electronic Prescription Database. Treatment persistence was defined as the duration from the date of the first prescription to the end of follow-up, death, or switching to another agent. Adherence was estimated based on the Medication Possession Ratio (MPR) metric. RESULTS:Overall, 2112 patients were analyzed. The majority were naive, male patients with a diagnosis of idiopathic pulmonary fibrosis (IPF). The overall median treatment persistence was 40.2 months (95% CI: 35.5-44.6). Women and treatment-naive patients demonstrated longer median treatment persistence compared to their counterparts, while older patients demonstrated the lowest median persistence rates. Adherence levels remained high across the follow-up period (90%). Diagnosis of IPF and gastrointestinal comorbidities were associated with a higher risk of discontinuation. CONCLUSIONS:We have generated novel data concerning the factors that affect patients' outcomes under anti-fibrotic therapy. These findings may provide helpful insights for the therapeutic management of ILDs.

Managing and assessing educational and research data in academic environments has become increasingly challenging due to the growing complexity and volume of the involved information. Existing approaches often fail to provide integrated, user-friendly solutions that support academic strategic planning and decision-making in universities. Academic information is typically fragmented across various systems, each serving a different purpose. However, to optimize its use, both research and educational performance must be treated as interconnected components of overall academic performance. Understanding their correlations and how they influence each other is crucial for improving institutional outcomes. This paper presents an ontology-based approach that supports data management, presentation and evaluation through visual analytics, specifically designed for academic settings. Developed using user-centered design principles, the system provides real-time monitoring, intuitive dashboards, and comprehensive cross-sectional analysis of research and educational activities. The proposed system emphasizes the value of visual analytics in academic data management and assessment and sets a foundation for future advancements in institutional digital ecosystems. By offering a complete view of performance and goal achievement, it helps academic institutions identify areas for improvement and recognize their strengths.

In the context of higher education, student engagement and personalized learning are increasingly being recognized as critical factors for academic success and effectiveness. This research explores how academic information ecosystems, comprising learning management systems, student information systems, and research analytics platforms, can be leveraged to represent, manage, and evaluate student engagement and personalization strategies. The aim is to enhance evidence-based decision making in universities by providing tools and metrics for monitoring and improving students’ learning experiences. This study contributes to the advancement of research data management practices in academic institutions, aligned with contemporary trends in open science, data-driven governance, and learning analytics. By analyzing student participation in online distance learning and examination processes, we identify key deterrents to engagement and tailor announcements, recommendations, and feedback. We utilized a hybrid modeling technique that combines individual student profiles and community-based groupings to predict student performance and failure risk with high accuracy. Our findings indicate that personalized interventions significantly improve student success rates across multiple programming courses from C and C++ to advanced mobile programming. Additionally, the research highlights students’ willingness to provide anonymous feedback on academic processes, emphasizing the value of incorporating their opinions into the system design. The proposed system facilitates continuous quality improvement in teaching and assessment, fostering a supportive learning environment and encouraging active student participation.

BACKGROUND:Limited data is available on the epidemiological burden of idiopathic pulmonary fibrosis (IPF) for countries such as Greece. As such, the objective of the current retrospective observational cohort study was to examine the temporal trends (2019-2023) in annual incidence and prevalence rates of IPF in Greece as well as demographic determinants. METHODS:Prescription data of anti-fibrotic agents, namely pirfenidone and nintedanib, executed between January 1, 2019, and December 31, 2023, were extracted from the National Electronic Prescription Database (IDIKA). The diagnosis of IPF was based on the ICD-10 codes of the anti-fibrotic prescriptions. Population data were obtained in September 2024 from the latest national census (2021). RESULTS:In a total cohort of 2583 patients with IPF diagnosis, the majority were male patients (74.2 %) above 66 years old (84.5 %). We detected increasing annual prevalence rates throughout the follow-up period, with a mean annual rate of 14.4 cases/100,000 population. The mean annual incidence rate was 4.6 cases/100,000 population. Although calendar year was not significantly associated with prevalence or incidence in unadjusted analyses, both associations became significant after adjustment for age, sex, and region. Higher prevalence and incidence were observed in older individuals, and regional disparities indicated higher rates in less urban regions. CONCLUSIONS:Collectively, this is the most representative study concerning epidemiology of IPF in Greece. These observations could be of great assistance for healthcare policy makers in terms of improving the therapeutic management of IPF as well as the development of prevention and screening strategies.

OBJECTIVES:We aimed to evaluate the effect of the implementation of electronic-prescribing protocols for infectious diseases on outpatient antimicrobial prescribing patterns in Greece. METHODS:In 2018-2019, in the context of policies to control antimicrobial use in Greece, electronic-prescribing protocols for acute pharyngitis, community-acquired pneumonia (CAP), acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and urinary tract infection (UTI) were introduced. We conducted a retrospective nationwide Interrupted Time Series analysis of monthly outpatient antimicrobial prescription counts overall and for targeted infectious syndromes during the preintervention and postintervention periods. Data were extracted from the official nationwide electronic prescription database covering July 2017 to December 2021. RESULTS:Overall reductions in median monthly antimicrobial prescription counts between the pre- and postintervention periods were 71.3% (95% CI: -78.5% to -56.4%) for pharyngitis, 90.0% (95% CI: -94.3% to -85.3%) for CAP, 92.6% (95% CI: -94.1% to -87.9%) for AECOPD, 85.8% (95% CI: -86.6% to -83.0%) for UTI. Despite these syndrome-specific reductions, the median monthly number of total outpatient antimicrobial prescriptions did not decrease (417 633 vs. 424 592 prescriptions per month). Before the intervention, the protocol-targeted syndromes collectively accounted for a median of 27.6% of total antimicrobial prescriptions per month, whereas 'other nonprotocol conditions' represented a median of 72.4%. After implementation, these proportions shifted dramatically to a median of 4.5% and 95.5% respectively (p < 0.001 for both changes), indicating a shift in diagnostic coding associated with antimicrobial prescribing. CONCLUSIONS:Although the protocols were designed to promote targeted antimicrobial use, their complex implementation may have inadvertently encouraged suboptimal prescribing practices. Improper coding and inconsistent antimicrobial prescribing data hinder antimicrobial stewardship efforts and should prompt a national health policy action plan.

The evaluation of universities has traditionally relied on metrics, such as citation counts, impact factors and participation in research projects. The results of such evaluations can significantly contribute to the academic assessment and strategic planning. Open science has introduced new dimensions of evaluation, including transparency, reproducibility, and accessibility. The paper explores the evolving landscape of academic evaluation, considering the impact of open science. We highlight the benefits and challenges of incorporating open science principles into academia, discuss existing frameworks and propose modifications to already existing evaluation methods in order to take into account and assess the adoption of open science in the academic setting. More specifically, we have adapted a dynamic multicriteria decision aiding method and the underlying ontology with the criteria identified by our research. As a direct consequence, the evaluation method reflects more accurately the current academic environment, promoting a more comprehensive and forward-looking assessment of research contributions.

Abstract:Aims: Since type 2 diabetes is one of the greatest burdens in public health over the past decade, evidence regarding its impact on the Greek population is scarce. To address this situation, the current study aims to reveal epidemiological data on type 2 diabetes in Greece by calculating the annual prevalence between 2016 and 2022 and to investigate trends regarding age and sex factors. Abstract:Methods: Real-world prescription data were used from the official database of the e-Government Center for Social Security Services. The study population comprised individuals who were entitled to type 2 diabetes care, with the relevant International Classification of Diseases, 10th Revision code. Descriptive statistics, chi-square tests and regression analysis were performed by calculating prevalence rates, using 2011 official census of Greek population. Abstract:Results: Prevalence rates of type 2 diabetes follow an increasing trend, increasing from 7.97% in 2016 to 8.65% in 2022. A similar trend was observed among male and female individuals; the male prevalence rate increased from 8% in 2016 to 8.86% in 2022, whereas the female prevalence rate increased from 7.95% in 2016 to 8.46% in 2022. The prevalence rate ratios between 2016 and 2022 underline the increasing trends of type 2 diabetes in population, as the total prevalence rate ratio was equal to 1.09 (95% confidence interval: 1.08-1.09, p < 0.001). Linear regression revealed the association between the prevalence rate and the sex or age factor. Abstract:Conclusions: Type 2 diabetes prevalence in Greece increased between 2016 and 2022, in accordance with the European and International data, with higher rates observed among males and older adults. The marked relative increase among individuals aged 18 to 29 years constitutes an important epidemiological finding and highlights the need for targeted prevention and early intervention strategies.

Graph Neural Networks (GNNs) are a class of neural networks specifically designed to process and extract information from graph-structured data by leveraging its inherent topological structure. Among these, Graph Convolutional Networks (GCNs) extend the concept of convolution—initially developed for Convolutional Neural Networks (ConvNets)—to operate on graphs. A specialized subset of GCNs, Spectral Graph Neural Networks (SGNNs), perform convolution in the spectral domain, drawing upon principles from spectral graph theory and Graph Signal Processing (GSP). In this work, we explore SGNNs by surveying key concepts in spectral graph theory and GSP that underpin their design. We also examine how the convolution operation in ConvNets generalizes to GCNs and highlight applications of GCNs along with existing libraries that implement SGNNs.

BACKGROUND:To eradicate cervical cancer, the World Health Organization (WHO) targets 90 % human papillomavirus (HPV) vaccination coverage in girls by the age of 15 until 2030. In Greece, data regarding how close the country is to meeting this target, is completely lacking. OBJECTIVES:To assess annual HPV vaccination coverage among individuals aged 9-15 years in Greece (2022-2024). METHODS:This is a retrospective, population-based cohort study using the Greek National Electronic Prescription Database to record all HPV vaccine doses dispensed from 1/1/2019 to 31/12/2024. The annual vaccination coverage was estimated as the proportion of eligible population receiving at least one dose or the full vaccination scheme from January 1st, 2019, through December 31st of the respective reference year. Full vaccination scheme by age 15 was defined - according to national recommendations - as two doses at a minimum interval of 6 months. RESULTS:From 2022 to 2024, the proportion of individuals 9-15 years old, who appropriately initiated vaccination increased from 34.7 % (2022) to 41.4 % (2024) in girls and from 10.8 % (2022) to 31.4 % (2024) in boys. The proportion of children who initiate HPV vaccination at the age of 9, increased from 3.5 % and 3.4 % in 2022 to 8.0 % and 7.4 % in 2024 in girls and boys respectively. Among girls turning 15, appropriate vaccine initiation rate marginally exceeded 63.0 % throughout the study period while full vaccination coverage increased from 47.7 % in 2022 to almost 52.5 % in 2024. CONCLUSION:Despite notable improvements in HPV vaccination uptake among adolescents in Greece between 2022 and 2024, coverage levels remain suboptimal relative to the WHO's 90 % target. Limited early initiation despite national recommendations starting at age 9, highlight the need for targeted strategies to promote timely HPV vaccination and accelerate progress toward elimination goals.

Objectives: To determine the prescribing prevalence of epidermal growth factor receptor (EGFR)- and anaplastic lymphoma kinase (ALK)-positive non-small cell lung cancer (NSCLC) patients in Greece and examine patterns of first-line tyrosine kinase inhibitor (TKI) utilization and associated treatment costs using nationwide real-world data. Methods: A retrospective analysis of the national e-prescription database was performed, identifying patients initiating first-line treatment (FLT) for EGFR- or ALK-positive NSCLC between 1 January 2020 and 31 December 2022. Demographic characteristics, prescribing prevalence data, drug utilization patterns, total annual drug expenditures, and per patient treatment costs were assessed. All statistical analyses were performed using the statistical software SPSS-v.29. Results: Overall, 1188 EGFR-positive (mean age of 70.93 ± 11.6) and 246 (mean age of 64.26 ± 12.6) ALK-positive NSCLC patients initiated FLT during the three-year study period. EGFR mutations were slightly more common in females (53%), peaking in the 70–79 age group (35%). ALK mutations were also more common among females (52%), particularly within the 60–79 age group. In EGFR-positive patients, osimertinib usage markedly increased from 41% in 2020 to 63% in 2022, primarily displacing afatinib (from 32% to 22%) and erlotinib (from 24% to 14%), with gefitinib prescriptions falling below 2%. Among ALK-positive patients, crizotinib utilization declined significantly from 60% to 16%, whereas alectinib increased to 59% by 2022. Annual EGFR-related total drug expenditures remained stable (€11.5 million in 2020 vs. €11.9 million in 2022), driven primarily by increasing osimertinib usage. Similarly, ALK-related annual drug expenditures showed stability, with costs predominantly attributed to rising alectinib utilization. Conclusions: This nationwide analysis highlights the rapid adoption of second- and third-generation TKIs for EGFR- and ALK-positive NSCLC in Greece, reflecting evolving clinical practice patterns. Although the target patient populations are relatively small, the associated economic burden is considerable. To ensure long-term sustainability of the Greek healthcare system, policymakers should critically assess the cost-effectiveness of these innovative therapies and align resource allocation with value-based care principles.

Multiple parameters define the treatment course with biologics for a psoriatic patient while treatment switches are often associated with worse prognosis. The purpose of this study was to describe the switching patterns of biologics for psoriasis in the Greek market landscape and to detect associated factors that may impact the evolvement of selected therapy. This is a retrospective cohort study using data recorded in the nationwide digital prescription database of Greece. Patients with a diagnosis for psoriasis, with or without concomitant psoriatic arthritis (PsA), who had initiated a biologic treatment between January 1st 2016 and December 31st 2020 were included. Overall, 6,772 biologic-naïve patients were included. Patients treated with infliximab demonstrated the highest switching rates while those treated with ustekinumab and secukinumab the lowest. Secukinumab and brodalumab had the lowest rates of switch or re-initiation 12 months after initiation. Switches from secukinumab to brodalumab and ustekinumab and from adalimumab to secukinumab and ustekinumab were more frequently observed. The risk was significantly higher for patients with concurrent PsA and for women, while patients treated with brodalumab, secukinumab and ustekinumab demonstrated a lower risk compared to adalimumab. Antibodies against interleukins have lower switching rates compared to more traditional biologics. The time to switch is longer for the first transition highlighting the necessity to establish long term therapeutic options early in the treatment course. Concurrent PsA or gender may have a significant impact in outcome, thus they need to be considered before the launch of a selected therapy.

BACKGROUND:Drug persistence is a crucial aspect of treatment success in psoriasis. OBJECTIVES:To record real-world evidence concerning drug survival of biologic agents used for psoriasis treatment and to detect associated modifying factors in Greece. METHODS:This was a retrospective cohort study based on data extracted from the nationwide Greek prescription system. Included patients had psoriasis, with or without concomitant psoriatic arthritis (PsA), and had initiated biologics between 1 January 2016 and 31 December 2020. RESULTS:We included 8819 patients who received 13 359 treatment lines. Among them, 76.8% of patients were biologic naive and 16.5% were diagnosed with concomitant PsA. The overall median drug survival was 34.3 months [95% confidence interval (CI) 32.6-36.5]. Drug persistence at 12, 24, 36 and 48 months of follow-up was 71.9%, 57.7%, 49.0% and 43.7%, respectively. Patients receiving brodalumab had the highest drug survival rate in the first 2 years, while secukinumab had the highest rates beyond this period. Overall, drug survival rates were higher in the first treatment line (median 51.1 months, 95% CI 47.1 to not reached) than in the second line and onwards (median 21.7 months, 95% CI 20.0-23.5). Treatment line, PsA status, age and sex were found to significantly affect drug survival rates. CONCLUSIONS:Our findings confirm previous reports regarding the importance of efficient first-line biologics and the vulnerability of patients to coexistent PsA. The use of antibodies against interleukins confers high drug survival rates. These results will assist clinical management of patients with psoriasis in Greece.

OBJECTIVES:This study aimed to investigate the prevalence and incidence of medication-treated diabetes mellitus and the evolving patterns of glucose-lowering treatments the year before and during the first two years of the coronavirus disease 2019 (COVID-19) pandemic. METHODS:Data from the Greek electronic prescription database were analyzed for 2019, 2020, and 2021. The study population included individuals with active social security numbers. Prevalence and incidence rates were calculated based on the dispensing of glucose-lowering medications according to their unique anatomical therapeutic chemical (ATC) code. RESULTS:The study population comprised 10,289,140 individuals in 2019, 10,630,726 in 2020, and 11,246,136 in 2021. Diabetes prevalence rates were 8.06%, 6.89%, and 7.91%, and incidence rates were 16.8/1000, 8.6/1000, and 13.4/1000 individuals, respectively. Metformin was the most prescribed medication, and newer classes, like sodium-glucose cotransporter-2 inhibitors 2 (SGLT-2) inhibitors and glucagon like peptide-1 (GLP-1) receptor agonists exhibited increasing trends. CONCLUSIONS:The study identified a decrease in medication-prescribed diabetes prevalence and incidence during the initial year of the COVID-19 pandemic, attributed to healthcare access restrictions. Subsequently, figures returned close to baseline levels. Glucose-lowering medication trends reflected adherence to local and international guidelines, with metformin as the cornerstone, and increasing preference for newer classes such as GLP-1 receptor agonists and SGLT-2 inhibitors.

Although several studies have explored the geoepidemiology of autoimmune rheumatic diseases (ARD), trends of their frequency overtime are under-investigated. Herein, in a nation-wide study, we examine trends in the prevalence of various ARD over-time, taking also into account the Covid-19 pandemic. In this retrospective study in the entire Greek adult population (approximately 10.000.000 people), we searched the electronic prescription database of the e-Government Centre for Social Security Services using prespecified ICD-10 codes to capture all adult patients with Psoriatic Arthritis (PsA), Rheumatoid Arthritis (RA), Ankylosing Spondylitis (AS), Systemic Lupus Erythematosus (SLE), Systemic Sclerosis (SSc) and Polymyalgia Rheumatica or Giant Cell Arteritis (PMR/GCA). Two sequential 4-year periods, namely 2016–2019 and 2020–2023 were compared. Prevalence of RA, PsA, AxSpA, SLE and SSc increased significantly during 2020–2023 compared to 2016–2019. This applies to both genders and to all age groups for RA, PsA and AxSpA, to female patients in SLE and SSc and to patients 18–39 years in SLE and ≥ 60 years in SSc. Overall, there was 47