Abstract
SUMMARY
Healthcare systems around the world continue to see their expenditures increase, measured as a percentage of gross domestic product. Within health economics, the need for models that can predict healthcare costs is of substantial importance, as decisions to introduce as well as to decommission healthcare services are based on these. This dissertation is an attempt to highlight the importance of epidemiological factors for health economic research on chronic diseases. Hence, the research question of interest is: how do individual epidemiological and behavioural factors impact the healthcare utilisation of patients with a chronic disease, e.g. osteoporosis?
This dissertation proposes a framework for predicting healthcare utilisation which includes four steps: familiarisation with the study population, determining the appropriate resource use, determining which predictors are important to consider, and lastly choosing the most appropriate statistical model. This framework was developed as a result of five quantitative studies, of which four were based on patient specific data from registers, and one on cost of illness theory. The framework was applied for predicting the cost for all fractures patients in one year following the fracture, i.e. the fifth study included in this dissertation. This study showed that it is not only important to understand the population of interest, as this eases the subsequent identification of potential predictors, but also the healthcare system through which these patients are treated, as different resources were affected differently by the clinical and behavioural predictors included.
In conclusion, the results from this dissertation highlight the importance being familiar with the population of interest, identifying the relevant resources, including both epidemiological and behavioural predictors, when analysing outcomes from both an epidemiological and health economic perspective, and choosing the right statistical model to analyse all this with. Both health and social scientists interested in researching utilisation of healthcare should consider these four steps.
Healthcare systems around the world continue to see their expenditures increase, measured as a percentage of gross domestic product. Within health economics, the need for models that can predict healthcare costs is of substantial importance, as decisions to introduce as well as to decommission healthcare services are based on these. This dissertation is an attempt to highlight the importance of epidemiological factors for health economic research on chronic diseases. Hence, the research question of interest is: how do individual epidemiological and behavioural factors impact the healthcare utilisation of patients with a chronic disease, e.g. osteoporosis?
This dissertation proposes a framework for predicting healthcare utilisation which includes four steps: familiarisation with the study population, determining the appropriate resource use, determining which predictors are important to consider, and lastly choosing the most appropriate statistical model. This framework was developed as a result of five quantitative studies, of which four were based on patient specific data from registers, and one on cost of illness theory. The framework was applied for predicting the cost for all fractures patients in one year following the fracture, i.e. the fifth study included in this dissertation. This study showed that it is not only important to understand the population of interest, as this eases the subsequent identification of potential predictors, but also the healthcare system through which these patients are treated, as different resources were affected differently by the clinical and behavioural predictors included.
In conclusion, the results from this dissertation highlight the importance being familiar with the population of interest, identifying the relevant resources, including both epidemiological and behavioural predictors, when analysing outcomes from both an epidemiological and health economic perspective, and choosing the right statistical model to analyse all this with. Both health and social scientists interested in researching utilisation of healthcare should consider these four steps.
| Bidragets oversatte titel | Vigtigheden af epidemiologiske prædiktorer for sundhedsforbrug for kroniske patienter: Et casestudie af osteoporotiske fraktur patienter |
|---|---|
| Originalsprog | Engelsk |
| Udgiver | |
| ISBN'er, elektronisk | 978-87 -7112-727-0 |
| DOI | |
| Status | Udgivet - 2016 |
Bibliografisk note
PhD Supervisor:Karin Dam Petersen, Associate Professor, Aalborg University
Assistant PhD Supervisor:
Peter Vestergaard, Professor, Aalborg University Hospital, Aalborg University
Lars Holger Ehlers, Professor, Aalborg University