Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Posts
Future Blog Post
Published:
This post will show up by default. To disable scheduling of future posts, edit config.yml and set future: false.
Blog Post number 4
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 3
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 2
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 1
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
publications
Medication adherence halves COPD patients’ hospitalization risk – evidence from Swiss health insurance data
Published in npj primary care respiratory medicine, 2009
Medication adherence is vital for patients suffering from Chronic Obstructive Pulmonary Disease (COPD) to mitigate long-term consequences. The impact of poor medication adherence on inferior outcomes like exacerbations leading to hospital admissions is yet to be studied using real-world data. Using Swiss claims data from 2015-2020, we group patients into five categories according to their medication possession ratio. By employing a logistic regression, we quantify each category’s average treatment effect of the medication possession ratio on hospitalized exacerbations. 13,557 COPD patients are included in the analysis. Patients with high medication adherence (daily medication reserve of 80% to 100%) are 51% less likely to incur exacerbation following a hospital stay than patients with the lowest medication adherence (daily medication reserve of 0% to 20%). The study shows that medication adherence varies strongly among Swiss COPD patients. Furthermore, high medication adherence immensely decreases the risk of hospitalized exacerbations.
Recommended citation: Bischof, A.Y., Cordier, J., Vogel, J. et al. Medication adherence halves COPD patients’ hospitalization risk – evidence from Swiss health insurance data. npj Prim. Care Respir. Med. 34, 1 (2024). https://doi.org/10.1038/s41533-024-00361-2
Download Paper
Price sensitivity and demand for healthcare services: Investigating demand-side financial incentives using anonymised claims data from Switzerland
Published in Working Paper Series in Health Economics, Management and Policy, 2023
Background. Compulsory healthcare insurance in Switzerland entails a deductible system for cost-sharing between insurer and insuree up to a chosen deductible1. No study has so far tested the presence of price sensitivity for healthcare resources consumption after exceeding one’s cho sen deductible. We address this research gap by focusing on the effect of exceeding the deductible on insurees’ healthcare consumption. We present three contributions: first, we determine the pres ence of price sensitivity for healthcare consumption; second, we identify whether this leads to a change in consumption for overuse-prone service groups; third, we explore whether supply side structures influence this change in consumption. Methods. For our analyses we make use of a detailed and anonymized insuree-level dataset pro vided by the Groupe Mutuel. We included data for all insurees older than 25 that exceeded their deductible in 2018 and did not give birth between 2017 and 2019. We focus our analyses on the 2,500 deductible group2 (sample size of 12,135 observations) and provide insights on the 300 de ductible group (sample size of 212,249 observations) for a comparison. Our empirical strategy included three steps. First, to control for insurees’ individual time-varying and constant charac teristics, we ran fixed effects ordinary least square regressions of weekly healthcare expenditures on insuree characteristics. Second, on the residuals obtained from the fixed effect model (i.e., the unexplained healthcare expenditures variation), we ran insuree-level regression discontinuity in time models. Finally, we aggregated the obtained parameters by simple mean. Starting from an explorative specification of the dependent variable including all service groups, we specified our dependent variable in two additional ways: first, we excluded all complex services; second, we only included services for which we could find evidence in the literature that they might be over used. We used the first specification for patient subgroup analyses and sensitivity analyses. The second specification was used to explore potential supply-side structure effects, measured via the density of medical specialists per postal code. Results. We find a positive difference between healthcare consumption before and after exceeding the deductible, however this increase in consumption is not significant. For insurees without continuous healthcare expenditures in the 12 weeks before exceeding the franchise, we find a weakly significant increase in subsequent healthcare expenditures, which is however not signifi cant at 95% confidence level. When stratifying insurees based on retirement status, premium re duction, and number of chronic illnesses, we do not find significant effects on the healthcare con sumption pattern for any of these subgroups. Finally, supply structures do not significantly influ ence healthcare consumption patterns after exceeding the deductible. Conclusions. Our results show that, while there is an overall pattern indicating a higher consump tion of healthcare resources after exceeding the deductible, this outcome is insignificant across all specifications. Our findings show that insurees are generally not price sensitive and that the de ductible system does not create significant demand-side financial incentives for the consumption of healthcare resources. As cost-sharing solutions have been introduced to curb the rise of healthcare spending, our findings suggest that the deductible system is an effective cost-sharing solution for Switzerland.
Recommended citation: Salvi, Irene; Cordier, Johannes; Kuklinski, David; Vogel, Justus; Geissler, Alexander (2023): Price sensitivity and demand for healthcare services: Investigating demand-side financial incentives using anonymised claims data from Switzerland, Schriftenreihe in Health Economics, Management and Policy, No. 2023-06, Universität St.Gallen, School of Medicine, Lehrstuhl für Management im Gesundheitswesen, St.Gallen
Download Paper
Entity embedding of high-dimensional claims data for hospitalized exacerbation prediction
Published in Health, Econometrics and Data Group Working Papers, c/o Department of Economics, University of York., 2024
This study addresses the challenges of using high-dimensional claims data, typically represented by categorical features, for prediction tasks. Traditional one-hot encoding methods lead to computational inefficiencies and sparse data issues. To overcome these challenges, we propose using entity embedding, a technique that has shown promise in natural language processing, to transform categorical claims data into dense, low-dimensional vectors as input for downstream prediction tasks. Our study focuses on predicting hospitalizations for patients with Chronic Obstructive Pulmonary Disease using the Word2Vec Continuous Bag-of-Words model. Our findings indicate that entity embedding enhances model performance, achieving an AUC of 0.92 compared to 0.91 with one-hot encoding, and improves specificity from 0.55 to 0.60 for a recall of 0.95. Additionally, entity embedding significantly reduces required computation power. These results suggest that entity embedding not only captures the dynamics of medical events more effectively but also enhances the efficiency of training predictive models, making it a valuable tool for healthcare and insurance analytics.
Recommended citation: Cordier, J., Geissler, A., & Vogel, J. (2024). Entity embedding of high-dimensional claims data for hospitalized exacerbation prediction. Health, Econometrics and Data Group (HEDG) Working Papers 24/08, HEDG, c/o Department of Economics, University of York.
Download Paper
Application of positive and unlabeled learning: A novel approach for identifying sepsis cases from hospital administrative data
Published in Working Paper Series in Health Economics, Management and Policy, 2024
Only positive instances of various events, e.g., secondary diagnoses, are actively labeled in hospital administrative data. In line with this, several studies indicate underreporting of adverse events such as sepsis. The gold standard for relabeling of uncoded sepsis cases, medical record review, is laborious, costly, and infeasible to execute for identifying sepsis in large, national datasets. We apply a positive unlabeled (PU) learner as a novel approach to identify sepsis cases from hospital administrative data. We exploit the Hospital Case Cost Statistic from the Swiss Federal Statistics Office (data years 2017 to 2019) including 72 cost attributes at case level. We hypothesize that these cost data should prove effective for learning a classification model as positive sepsis cases in the unlabeled data should exhibit similar cost patterns as labeled positive examples. We randomly draw 200,000 unlabeled examples from the full dataset and add 64,915 positive examples of sepsis labeled in the observation period for model training and evaluation. We train a robust PU learner proven in other applications, AdaSampling, with support vector machine as classification model. For model evaluation, we perform five-fold cross validation. Due to the PU setting, we can only use positive examples in the test set and estimate recall along with precision and recall at 10%, 20%, and 30% for four different evaluation scenarios, changing the coding strategy for labeling sepsis cases. Our model has a recall of 85.1% when labeling sepsis cases explicitly in the test set. Recall decreases to 55.5% when labeling sepsis cases exclusively with an implicit coding strategy. Recall at k% is highest for the evaluation scenarios focusing on implicit coding strategies, yet remains relatively low throughout all scenarios. Precision at k% is highest when only considering cases as positive examples that would be labeled according to both the explicit as well as implicit coding strategy (e.g., 92.3% for k=10%). Compared to the sensitivity of directly identifying sepsis cases from hospital administrative data reported in studies using medical record review, the recall of our model is high. We propose a two step process using PU learning for increasing the quality of hospital administrative data and performing sensitity analyses for health economic and health services research.
Recommended citation: Vogel, Justus; Cordier, Johannes (2024) : Application of positive and unlabeled learning: A novel approach for identifying sepsis cases from hospital administrative data, Working Paper Series in Health Economics, Management and Policy, No. 2024-02, University of St.Gallen, School of Medicine, Chair of Health Economics, Policy and Management, St.Gallen
Download Paper
talks
Entity embedding of high-dimensional claims data for hospitalized exacerbation prediction
Published:
teaching
Health systems, health economics and policy
Master-level course, School of Medicine, University of St. Gallen, 2021
On completion of the course, students will have a basic understanding of the Swiss healthcare system and understand the economic and political implications of medical practice. Students will gain an understanding of the roles and objectives of the players in the Swiss healthcare system. Furthermore, students will deepen their ability to present and critically analyse scientific hypotheses, results and discourses.
Case Studies in Health Economics
Master-level course, School of Medicine, University of St. Gallen, 2022
After completing the course, students will be able to categorise basic functions and selected aspects of healthcare systems internationally and compare the different framework conditions. Furthermore, students will deepen their ability to present scientific hypotheses, results and discourses in a pointed manner and to critically scrutinise them. Students also learn how to apply scientific findings in healthcare management in possible future professional fields.
Economics in Healthcare
Master level course, School of Economics and Political Science, University of St. Gallen, 2024
This course is about the application of economic theories and tools to questions in healthcare. The course builds up on prior knowledge of microeconomic theory and econometrics and translates the concepts to pressing healthcare challenges. Goals of the course: Understanding the pressing questions in healthcare and the potential empirical approaches available to tackle them by finding answers and solutions; Overview of the current state of health economic research.