Background: Multiple sclerosis (MS) is a chronic autoimmune disease of the central nervous system. Early detection of the prodromal phase could enable timely interventions to potentially modify disease progression. This study leverages longitudinal health administrative claim (HAC) data to identify patterns distinguishing the prodromal stage of MS from other neurological conditions. Methods: HAC data from the Czech Health Insurance Bureau (2017–2022) was analyzed across three cohorts: a target MS cohort with confirmed diagnoses, a control cohort with inconsistent MS suspicions, and a cohort with related disorders. For healthcare utilization and diagnostic code data representation, we employed two approaches: temporal analysis using various time windows relative to the index date (including pre- and post-index date comparisons) and a separate segment-based analysis. Features were extracted using token frequencies and word embeddings. Random forest models were evaluated using Area Under the Receiver Operating Characteristic Curve (AUC) to assess performance. Results: Each cohort included several hundred to over a thousand individuals. The models achieved AUCs around 0.9 for distinguishing the target cohort from controls, with even higher performance in differentiating pre- and post-diagnosis phases. Longer observation windows enhanced predictive accuracy, and feature extraction methods like TF-IDF and word2vec yielded the most consistent results. Segment-based analysis identified a subset of individuals for potential diagnostic reclassification. Interpretable machine learning techniques were integrated into the analysis pipeline. Conclusions: This study highlights the potential of HAC data for detecting early prodromal indicators of MS. Unlike previous research, which often focused on the volume of healthcare utilization, this work explores the informational content within diagnostic codes and healthcare utilization patterns. The findings align with existing research on early neurological condition detection, demonstrating that administrative data could support early identification and intervention in MS and possibly other diseases.

Background: Stroke remains a significant health and economic challenge both globally and in the Czech Republic. Although a structured network of specialized stroke centres exists, comparative data on patient outcomes and healthcare costs across hospital types are still lacking in the Czech context. This study analyzed real-world administrative data to assess 90-day mortality and healthcare costs after ischemic stroke, categorized by intervention and provider type. Methods: Claims data from six Czech health insurance companies, covering approximately 44% of the population, were used for the years 2017–2020. Patients aged 18 and older with a primary diagnosis of ischemic stroke (ICD-10 code I63) were included. Interventions were categorized as thrombectomy, thrombolysis, or other treatment, and providers were classified as comprehensive stroke centres (CSCs), primary stroke centres (PSCs), secondary referral hospitals (SRHs), or others. Costs were calculated from the payer perspective using administrative claims data, and standardized 90-day mortality and effective cost per survivor (ECPS) were computed. Funnel plots were used to evaluate provider variability in outcomes and costs. The analysis included 23,568 patients (47% female; mean age 70.6). Results: Thrombectomy was associated with the highest mean costs (€13,385), the highest 90-day mortality (29.3%), and the highest ECPS (€18,880). Patients receiving other treatments had the lowest costs (€2725) and lower mortality (14.4%). CSCs recorded the highest average costs (€5087) and mortality (16.7%), while SRHs had the lowest costs (€2204) and mortality (13.7%). Funnel plots revealed greater variability in costs, mainly driven by primary hospitalization, while mortality rates showed less variation. Conclusions: These findings suggest that while stroke outcomes are relatively consistent across providers, costs differ, possibly reflecting efficiency differences and case-mix severity. The study is limited by the lack of clinical severity data, highlighting the need to link administrative data with clinical registries for more comprehensive future evaluations.