In this project led by dr. Jeroen Schouten (intensive care physician at Radboud UMC), we contributed by providing our expertise on real-world data and real-world data analytics. In this project, the SIMPLIFY database (containing real-world data from electronic hospital records (EHRs) of ICU patients) and data from the National Intensive Care Registry (NICE) were used to gain insight into the extent and quality of antibiotics prescribing in Dutch ICU patients.

Status: manuscript submitted

Involved PIL team members: Joanna Klopotowska, Izak Yasrebi-de Kom

In this study conducted in Amsterdam UMC, (location VUmc), we researched the feasibility of using the online patient portal for medication verification. Together with patients, physicians, hospital pharmacists and patient portal developers we created patient journeys for medication verification via patient portal. Subsequently we researched how often the patient portal is used by patients to manage their medication, we assessed usability of the portal and evaluated medication profiles created by the patients on accuracy and completeness.

Involved PIL team members : Joanna Klopotowska

Illustration by www.videologic.nl

In this study we investigated the perspectives of ICU physicians and ICU patients on using machine learning to predict ICU survival prognosis. Ethical, legal and psycho-social implications were discussed and a set of requirements for responsible use of Artificial Intelligence for supporting medical decisions in the ICU was created.

Involved PIL team members: Joanna Klopotowska

In this project led by the Department of Research Data Management of Amsterdam UMC, we contributed by providing our expertise on de-identification tooling and will organise the annotation process of clinical narratives by medical experts as part of the validation of the developed de-identification algorithm. Such algorithm is needed to remove mentions of directly identifying information about patients from the clinical narratives. Directly identifying information includes for example: surnames, birth dates, social security numbers or home addresses. This directly identifying information must be removed in order to preserve patient privacy while reusing clinical narratives for research.

Involved PIL team members: Joanna Klopotowska, Hemmik Leopold

In this project led by dr. Pauline Bollen (hospital pharmacist at Gelderse Vallei Hospital), we contributed by providing our expertise on reusing real-world data from electronic hospital records, data analytics and computerised decision support systems. Concomitant intake of food and medications may increase or decrease drug absorption and subsequently affect clinical health outcomes.

Involved PIL team members: Joanna Klopotowska