Background
The Netherlands faces a growing challenge in providing high-quality primary care due to the increasing prevalence of chronic and complex diseases, combined with a significant shortage of general practitioners (GPs). According to a report by the Dutch Ministry of Health, Welfare, and Sport (VWS), the number of GPs is expected to decline by up to 30% by 2030, while the demand for healthcare services continues to rise, particularly for older adults with multiple chronic conditions (VWS, 2020). This shortage risks compromising the delivery of effective, timely, and accessible healthcare to the population.
In response to this challenge, the Dutch healthcare system has introduced the role of the extended scope specialist (ESS), particularly in primary GP settings. These specialists, often non-medical professionals such as master educated physiotherapists, receive additional training to take on expanded roles in diagnostics, treatment, and referral processes. The introduction of ESSs aims to alleviate pressure on GPs while maintaining high standards of care and efficient referral to secondary care, particularly in specialized fields such as musculoskeletal disorders (MSDs).
Research suggests that extended scope physiotherapists (ESPs), specifically, can provide high-quality care comparable to GPs in managing MSDs. Studies have shown that ESPs exhibit similar or better diagnostic accuracy, effectiveness of care, care utilization, and cost-effectiveness when treating musculoskeletal complaints.
Despite these promising findings, there is limited evidence on the performance of ESPs in the Dutch healthcare context. Most studies have been conducted in other countries, with different healthcare systems and models of care. Therefore, further research is needed to evaluate the efficiency of ESPs in Dutch primary care settings, assess their integration into the existing healthcare infrastructure.
Approach
This project focuses on predicting repeat consultations and treatment or referral decisions using patient data from the Nederlandse Vereniging Extended Scope database.
The project will investigate AI methods that combine structured patient information with free-text clinical data to identify patterns associated with follow-up visits and referral outcomes.
The project will focus on:
- Predicting whether patients will require repeat consultations
- Predicting the most appropriate treatment plan or referral location
- Combining structured and unstructured clinical data for predictive modelling
- Exploring NLP methods for analyzing free-text consultation information
Data
This project will use pseudonymized data from more than 3,000 patients treated by extended scope specialists in the Netherlands.
The dataset includes structured information such as ICPC codes, treatment plans, supervision by GPs, and repeat consultation outcomes, as well as free-text descriptions for the reason of repeated consultations.
Requirements
- Interest in AI applications in healthcare
- Experience with Python programming
- Familiarity with machine learning and/or natural language processing
- Affinity with healthcare data analysis
Information
Project duration: 6 months
Location: Radboud University Medical Center
The student will be embedded in the IQ health department, research and education group of allied health and nursing care.
If you are interested in applying for this Master student project, please send an email to: rtcai@radboudumc.nl
