Prevalence, incidence, and mortality of inflammatory bowel disease in the Netherlands: development and external validation of machine learning models - Scorecard - MDSpire
Advertisement
Prevalence, incidence, and mortality of inflammatory bowel disease in the Netherlands: development and external validation of machine learning models
Clinical Scorecard: Epidemiological Insights into Inflammatory Bowel Disease in the Netherlands: Creation and Validation of Machine Learning Models for Prevalence, Incidence, and Mortality Analysis
At a Glance
Category
Detail
Condition
Inflammatory Bowel Disease (IBD)
Key Mechanisms
Identification of IBD cases using machine learning models applied to administrative healthcare data
Target Population
IBD patients in the Netherlands
Care Setting
Population-based and hospital-based healthcare settings using administrative data
Key Highlights
Machine learning models (random forest and gradient-boosted trees) accurately identify IBD cases, subtypes, and incidence year using administrative data.
IBD prevalence in the Netherlands was 577.6 per 100,000 in 2020, with incidence stable at 20.1 per 100,000 and mortality rising over time.
The prevalence increase is slower than expected, indicating a shift towards the epidemiological stage of Prevalence Equilibrium.
Guideline-Based Recommendations
Diagnosis
Use validated machine learning models on administrative data to identify IBD cases accurately.
Confirm IBD diagnosis through chart review and standardized diagnostic criteria.
Management
Monitor epidemiological trends to inform healthcare resource allocation and cost-effectiveness initiatives.
Consider the impact of advanced therapies on healthcare costs and patient outcomes.
Monitoring & Follow-up
Track incidence, prevalence, and mortality rates over time using validated models and large registries.
Use administrative data linked with clinical records for ongoing surveillance.
Risks
Rising mortality rates in IBD patients compared to the general population require attention.
Increasing healthcare costs due to advanced therapies and growing patient population.
Patient & Prescribing Data
IBD patients identified via administrative data in the Netherlands
Increasing use of advanced therapies contributes to healthcare costs; effectiveness and impact on work disability require further study.
Clinical Best Practices
Leverage large nationwide registries combined with machine learning for accurate epidemiological assessment of IBD.
Validate predictive models externally to ensure generalizability across different geographic regions.
Integrate multiple administrative data sources (hospital admissions, outpatient visits, medication, insurance claims, general practice, pathology) for comprehensive case identification.
Use epidemiological data to guide healthcare planning and resource allocation for IBD management.
by Reinier C A van Linschoten, Nikki van Leeuwen, David van Klaveren, Marieke J Pierik, Rob Creemers, Evelien M B Hendrix, Jan A Hazelzet, C Janneke van der Woude, Rachel L West, Desirée van Noord
Large claims analysis finds no significant differences in serious infections, blood clots, or major cardiovascular events across biologics and a Janus kinase inhibitor.
