Scientific publications

A short summary for the scientific publications of our department. The total research output can be found via this link.

The use of artificial intelligence to optimize medication alerts generated by clinical decision support systems: a scoping review

This scoping review identified studies that used Artificial Intelligence (AI) to optimize medication alerts in hospitals. The researchers found 10 relevant studies, with only 30% reporting both statistical and clinical outcomes. Alerts optimized using AI-based methods resulted in a decreased alert burden, increased identification of inappropriate or atypical prescriptions, and enabled prediction of user responses.  Two of the 10 studies implemented AI alerts in hospitals, and none underwent external validation. Want to learn more? Click here for more information.

Jetske Graafsma, Rachel Murphy et al. DOI: 10.1093/jamia/ocae076

One-year transplant-free survival following hospital discharge after ICU Admission for ACLF in the Netherlands

This study investigates the long-term outcomes of ICU patients with acute decompensated liver cirrhosis or acute-on-chronic liver failure (ACLF). In total, 47% of 3035 patients survived hospitalization, and the overall probability of 1-year transplant-free survival after hospital discharge was 0.61. The ACLF severity at ICU admission did not independently influence post-discharge survival. Click here for more information.

Jubi de Haan, Fabian Temorshuizen et al. DOI: 10.1016/j.jhep.2024.03.004

    Characteristics and outcome of COVID-19 patients admitted to the ICU: a nationwide cohort study on the comparison between the consecutive stages of the COVID-19 pandemic in the Netherlands, an update

    This study reports mortality rates among patients with COVID-19 during the first wave and compared these rates with the following waves. Using COVID-19 patient data from Dutch Intensive Care Units (ICUs), collected by Stichting Nice, between May 2020 and January 2023,  the mortality risks in the initial upsurge of the third wave was similar to the first wave, but mortality rates decreased in later periods. Curious? Click here for more information.

    Fabian Termorshuizen et al. DOI: 10.1186/s13613-023-01238-2

    The effect of computerised decision support alerts tailored to intensive care on the administration of high-risk drug combinations, and their monitoring: a cluster randomised stepped-wedge trial

    In nine Dutch intensive care units, this study assessed the impact of tailoring potential drug-drug interaction (DDI) alerts to the ICU setting. Using a cluster randomised stepped-wedge trial, a customized clinical decision support system (CDSS) only providing alerts for potential DDIs considered as high risk was implemented and evaluated. The study found a 12% decrease in the number of administered high-risk drug combinations, indicating that tailoring alerts to the ICU setting improves CDSS effectiveness and patient safety. Click here for more information.

    Tinka Bakker, Joanna Klopotowska et al. DOI: 10.1016/S0140-6736(23)02465-0

    Strain on Scarce Intensive Care Beds Drives Reduced Patient Volumes, Patient Selection, and Worse Outcome: A National Cohort Study

    This study analyzed the impact of Intensive Care Unit (ICU) resource allocation during the COVID-19 pandemic on a non-COVID-19 cohort (120.393 patients) in Dutch ICUs. Compared to the prepandemic period cohort (164.737 patients), the pandemic cohort had lower number of non-COVID patients (27% lower), fewer medical patients (3% lower), fewer comorbidities (4% lower), more vasoactive medication (6% higher), and a slightly higher case-mix adjusted hospital mortality (odds 1.08). Click here for more information.

    Sylvia Brinkman, Nicolette de Keizer et al. DOI: 10.1097/CCM.0000000000006156

    Predicting 30-day mortality in intensive care unit patients with ischaemic stroke or intracerebral haemorrhage

    This study developed and validated two predictive models to estimate the 30-day mortality for patients with a stroke admitted to intensive care units (ICUs): one model for patients with ischaemic (N=8 422) and one model for patients with haemorrhagic stroke (N=5 881). The 30-day mortality was 27% in the first group and 41% in the second group. Both models showed high discrimination (AUC of 0.85) and good calibration. Click here for more information.

    Fabian Termorshuizen et al. DOI: 10.1097/EJA.0000000000001920

          Differences in the epidemiology, management and outcomes of kidney disease in men and women

          This review reports on differences between men and women with kidney disease. Women show a higher prevalence of chronic kidney disease stages 3-5, and men a higher prevalence of albuminuria. Women are less aware of their disease, and receive less screening, nephrologist care, and face greater barriers to kidney transplantation access. Men experience faster renal decline, higher mortality, and increased cardiovascular risk. Click here for more information. 

           Nick Chesnaye et al. DOI: 10.10338/s4181-023-00784-z

          Quality improvement of Dutch ICUs from 2009 to 2021: A registry based observational study

          This study examined the quality of intensive care unit (ICU) care in the Netherlands over the years 2009-2021 using data from the Dutch National Intensive Care Evaluation (NICE) registry. In total, 705,822 admissions from 55 ICUs were included in the analyses. Results showed improvements in several indicators over time, such as shorter ICU stays and reduced in-hospital mortality. However, the COVID-19 pandemic disrupted some of these trends without significantly affecting the variance between different ICUs. Click here for more information.

          Marie-José Blom, Ferishta Raiez, Sylvia Brinkman et al. DOI: 10.1016/j.jcrc.2023.154461

          Assessing resolvability, parsability, and consistency of RDF resources: a use case in rare diseases  

          This study evaluated the quality of 16 linked resources related to rare diseases, which rely on Uniform Resource Identifiers (URIs) to connect and enrich data. Despite the promise of 'linked data’, the actual effectiveness of these URIs remained unclear. The authors applied six quality metrics to assess the resources and identified various types of quality issues, including malfunctioning URIs and undefined ontological terms. The insights gleaned from this research are set to inform the creation of guidelines aimed at enhancing the quality and interoperability of linked biomedical resources. Curious? Click here for more information.

          Shuxin Zhang, Nirupama Benis & Ronald Cornet. DOI: 10.1186/s1336-023-00299-3

          Correctly structured problem lists lead to better and faster clinical decision-making in electronic health records compared to non-curated problem lists: A single-blinded crossover randomized controlled trial

          This randomized controlled trial investigated the impact of structured problem lists in electronic health records (EHRs) on clinical decision-making, compared to information partly documented in the free text. Results show that structured problem lists lead to better and faster decision-making among healthcare providers. The findings suggest increased use of structured lists and the development of implementation policies. Click here for more information.

          Eva Klappe et al. DOI: 10.1016/j.ijmedinf.2023.105264

          A two-step approach to create and evaluate an optimization method for surgical instrument trays to reduce their environmental impact

          This study used Integer Linear Programming (ILP) and Life Cycle Assessment (LCA) to guide reduction of the carbon footprint of surgical instrument trays. By optimizing tray composition , a 46% annual reduction in emissions could be realized for radical hysterectomy procedures. The research emphasizes the need for a diverse team, including medical professionals, to implement environmentally friendly changes in surgical practices. Click here for more information.

          Nona Schmidt, Marieke Sijm-Eeken et al. DOI: 10.1016/j.cesys.2023.100154

          Increasing trends in a low 5-min Apgar score among (near) term singletons: a Dutch nationwide cohort study

          This study analyzed the Dutch Perinatal Registry (PERINED) data (2010-2019) on 1,583,188 infants born ≥35 weeks of gestation. A significant increase in the proportion of infants with low 5-minute Apgar scores (<7 and <4) was found in the period 2010-2019. Instrumental vaginal delivery and emergency caesarean sections were performed less frequently, however these intervention groups showed the highest relative increase in infants with low Apgar scores. Read more here.

          Anita CJ Ravelli et al. DOI: 10.1038/s41372-023-01786-2

          Prolonged use of closed-loop inspired oxygen support in preterm infants: a randomised controlled trial

          This study compared the effect of automated oxygen control (A-FiO2) with manual titration (M-FiO2) on non-invasive respiratory support in preterm infants. The randomized controlled trial showed that A-FiO2 was better at maintaining the oxygen saturation within target range over a 28-days period but with a potential trade-off of increased time below the target range initially. Click here to read the whole article.

          Tim M.R. Schouten, Ameen Abu-Hanna et al. DOI: 10.1136/archdischild-2023-325831

          Identifying Environmental Impact Factors for Sustainable Healthcare: A Scoping Review

          This review identified 360 impact factors to measure environmental impact in healthcare from 46 articles These factors were categorized  resulting in the Healthcare Environmental Impact Factor (HEIF) scheme. HEIF can – for example - be used to select measurable indicators for quality management and guide future research in standardizing factors for cross-organization comparisons. Click here to read the full article.

          Marieke Sijm-Eeken, Monique Jaspers and Linda Dusseljee-Peute.
          DOI: 10.3390/ijerph20186747

          Development and Internal Validation of a Prediction Model for Falls Using Electronic Health Records in a Hospital Setting

          This study developed and validated two prediction models for falls among older hospital patients (2016-2021). It included a dataset of 21,286 admissions. The first model included missing value predictors (Model-With) and the second model did not (Model-Without). Both models showed good calibration and fair discrimination. The Model-With performing better. The models use routine data and can potentially reduce nurses' workload. Click here for more information.

          Noman Dormosh, Birgit Damoiseaux-Volman et al. DOI: 10.1016/j.jamda.2023.03.006

          Acute kidney injury associated with nephrotoxic drugs in critically ill patients 

          Acute Kidney Injury (AKI) due to nephrotoxic drugs is common in ICU patients. Yet, there is a lack of large studies on the relationship between drugs and AKI. Using real ICU data (92,616 admissions), this study identified 14 drug groups linked to higher risk of AKI . The groups included aminoglycosides, opioids and sympathomimetics. The results enhances current ICU practices and guides further research on drug-related AKI. Click here to learn more about this study.

          Izak Yasrebi-de Kom et al. DOI: 10.1093/ckj/sfad160

          Temporal validation of 30-day mortality prediction models for transcatheter aortic valve implantation using statistical process control – An observational study in a national population

          Multiple prediction models for mortality after Transcatheter Aortic Valve Implantation have been developed. To better understand the predictive performance of such models, researchers explored the effect of time on two types of models (parametric and non-parametric). They used data from the Netherlands Heart Registration between 2013 and early 2019 with data from 11,291 patients. Results showed that both models performed better when refitted over time. These models were more stable and accurate than the models that were developed just at one moment and not refitted over time. Click here for more information about this study.

          Ricardo R. Lopes, Tseko Yordanov et al. DOI: 10.1016/j.heliyon.2023.e17139

          An integrated approach to geographic validation helped scrutinize prediction model performance and its variability

          This study validates a mortality prediction model for Transcatheter Aortic Valve Implantation across 16 hospitals using multicenter data. Among 11,599 patients with 3.7% early mortality, performance varied widely between hospitals, with miscalibration in seven hospitals. Notable case-mix differences and limited model adaptability were evident. A subgroup displayed significant prediction errors and poor validation center performance, emphasizing challenges in adapting multicenter models for diverse populations. For more information, click here.

          Tseko Yordanov et al. DOI: 10.1016/j.jclinepi.2023.02.021

          Prolonged use of closed-loop inspired oxygen support in preterm infants: a randomised controlled trial

          This study compared the effect of automated oxygen control (A-FiO2) with manual titration (M-FiO2) on non-invasive respiratory support in preterm infants. The randomized controlled trial showed that A-FiO2 was better at maintaining the oxygen saturation within target range over a 28-days period but with a potential trade-off of increased time below the target range initially. Click here to read the whole article.

          Tim Schouten et al. DOI: 10.1136/archdischild-2023-325831 op

          Automated identification of patient subgroups: A case-study on mortality of COVID-19 patients admitted to the ICU

          This study assessed the quality of three methods for automated splitting of the data into subgroups. This process is called “subgroup discovery”. The researchers aimed to obtain subgroups in observational data of 14,548 COVID-19 patients admitted to 73 Dutch intensive care units. The tested methods discovered 5-62 subgroups. The intensivists who participated in the study stated that the subgroups made clinical sense with some differences between the three methods. Click here to learn more about this study.

          Iacopo Vagliano et al. DOI: 10.1016/j.compbiomed.2023.107146

          Assessing the FAIRness of databases on the EHDEN portal: A case study on two Dutch ICU databases

          This study assesses the findability, accessibility, interoperability, and reusability (FAIR) of two intensive care databases that are made accessible on the EHDEN portal, an online browser for research databases. To assess these databases, two researchers applied the seventeen FAIRsFAIR metrics which were defined as minimal requirements for databases to be FAIR. The total scores achieved were 15.5 and 12.0 out of a maximum of 25. The main areas for development were the implementation of globally unique identifiers and standardizing linked metadata. Click here to learn more about this study.

          Daniel Püttmann et al. DOI: 10.1016/j.ijmedinf.2023.105104

          Drug-related causes attributed to acute kidney injury and their documentation in intensive care patients

          This study examined causes of drug-related acute kidney injury and their documentation during Intensive Care Unit (ICU) admission. Among 8,124 ICU admissions in Amsterdam UMC (2015-2020), 542 (7%) experienced acute kidney injury and 102 (19% of 542) a drug-related acute kidney injury.  Clinical notes documented all drug-related cases (100%), while the allergy module (1%) and diagnosis codes (0%) were (almost) not utilized. To gain insights in drug-related acute kidney injury, both automated identification from clinical notes and structured documentation should be encouraged. Curious about this study? Read more here.

          Rachel Murphy et al. DOI: 10.1016/j.jcrc.2023.154292

          Predicting future falls in older people using natural language processing of general practitioners' clinical notes

          This study used Electronic Health Records data of people aged 65 or older to develop three logistic regression models. One for structured data (Baseline), another included topics from doctors' notes (Topic-based), and the third combined both (Combi). The models been tested with data from 35,357 individuals, including 4,734 who fell. The results showed good performance, with AUC values of 0.709 (Baseline), 0.685 (Topic-based), and 0.718 (Combi). They were well-calibrated and effective. Click here to learn more about this study. 

          Noman Dormosh et al. DOI: 10.1093/ageing/afad046

          Age Moderates the Effect of Obesity on Mortality Risk in Critically Ill Patients With COVID-19: A Nationwide Observational Cohort Study

          This study examined the relationship between body mass index (BMI) and mortality in 15,701 critically ill COVID-19 patients. Researchers found that for patients under 45 years old, a higher BMI (30 kg/m² or more) was associated with lower hospital mortality. This "obesity paradox" effect was not seen in patients 45 years and older. Click here for more information about this study.

          Corstiaan A den Uil, Fabian Termorshuizen et al. DOI: 10.1097/CCM.0000000000005788

          External Validation of a Prediction Model for Falls in Older People Based on Electronic Health Records in Primary Care

          An accurate prediction model can detect older people with a risk of falling. In this study, researchers investigated the use of a previously developed fall prediction tool. The tool was applied in a dataset extracted from electronic health records of general practitioners(with data of 39.342 older people of which 5124 (13%) with a fall). The results demonstrate that the model, even in this dataset, can predict fall incidents [BIvd(1] and could be useful in healthcare. Curious about this study? Read more here.

          Noman Dormosh et al. DOI: 10.1016/j.jamda.2022.07.002