Studies I–IV
The studies examine how EMS acts as a safety net under uncertainty, spanning dispatcher reasoning, conditional risk patterns, concordance across system
classifications, and system-level drivers of response-time variability.
Study pages
Study I — Dispatcher work under scarcity (qualitative)
Question: How do dispatchers prioritise and coordinate resource allocation when calls compete for limited units?
In writing
Study II — Breathing emergencies: conditional risk and response time
Question: How do response time, age, and sex interact in shaping conditional probability of high-risk outcomes in breathing-problem missions?
Study III — Infection presentations: dispatch, phenotype, and high-risk triage
Question: What is the concordance between dispatch suspicion (“Fever/Infection”), on-scene phenotype (ESS), and high-risk triage (RETTS)?
Study IV — System drivers of EMS response times
Question: Which operational, temporal, and geographic factors jointly shape response-time variability?
Interpretation note
Where exploratory machine learning is used, interpretability tools (e.g., partial dependence and related approaches) are emphasised to understand
non-linear patterns and heterogeneity, not to propose deployable automation.
Adaptive EMS Allocation 