Research, design and development of SW tools for process management in the area of e-health : projection of future number of end-stage renal disease patients in Greece

• Κύριος συγγραφέας: Ροδινά-Θεοχαράκη, Αναστασία
• Άλλοι συγγραφείς: Παλληκαράκης, Νικόλαος
• Μορφή: Thesis
• Γλώσσα: English
• Έκδοση: 2013
• Θέματα: End-stage renal disease; Renal replacement therapy; Disease projection; Markov Chain model; Markov Chain Monte Carlo; Τελικό στάδιο χρόνιας νεφρικής ανεπάρκειας; Θεραπεία υποκατάστασης νεφρικής λειτουργίας; Πρόβλεψη ασθενειών; Μοντέλο Μαρκόφ; Μοντέλο Μαρκόφ Μόντε Κάρλο; 617.461 059 2

End Stage Renal Disease (ESRD) is the irreversible loss of kidney function, which can be due to various causes. Its treatment is one of the most costly chronic disease treatments. There are now approximately one million people worldwide living with ESRD and this number is predicted to increase in the future. The main reasons for the increasing incidence of ESRD worldwide are population ageing, the rapid increase of diabetes mellitus reaching epidemic proportions, and changes in age limits for treatment initiation. In Greece, during the period 2005-2009, 74% of the ESRD patients were on hemodialysis (HD), 7% on peritoneal dialysis (PD) and 19% were living with a functioning graft. The latter percentage brings Greece in the 26th place out of 36 countries in prevalence of functioning grafts worldwide. Cost-effectiveness analyses of these treatments have shown that RTx is overall the least expensive, followed by PD, while centre HD is the most expensive. Moreover, these treatments are also listed in the exact same order concerning the quality of life they provide to patients. The main reasons for the low RTx rate in Greece are the lack of organ donation, largely due to inadequate information, the inefficient organ distribution system, a high number of private HD centers not interested in RTx, as well as social factors. The objective of the present work was to implement a model for the projection of the ESRD patients’ number by 2020 in Greece and investigate the impact of different scenarios of an increase in RTx. In addition, a cost-effectiveness analysis of the increase in RTx was performed. The projection was performed based on a Markov chain model. The Markov models are distinguished by their simplicity and their ability to accurately represent many clinical problems. A deterministic Markov chain model was implemented in order to predict the future number of prevalent ESRD patients in Greece. Monte Carlo techniques were applied in order to add robustness to the model. Thus two models of prediction were implemented, a Markov chain and a Markov Chain Monte Carlo (MCMC) model. Age-specific data (<45, 45-65, >65 age groups) on incident and prevalent ESRD patients’ number for Greece, available from the European Renal Association – European Dialysis and Transplant Association reports for the period 1998-2009, were used for the implementation. The basic component of the Markov chain is the transition matrix defining the probability for the patient to move between the four states, i.e. HD, PD, RTx and death. An iterative error minimization technique was used in defining the transition probabilities of the Markov chain, based on the data from 1998 to 2006. Both Markov chain and MCMC models were successfully validated based on data for the period from 2007 to 2009. In each model the ESRD incident patients’ number in Greece was predicted in a different way. For the Markov chain model three incidence rate scenarios were applied: low, medium and high. Additionally, two different approaches were proposed for the increase in RTx, one for each model. In the Markov chain model, two scenarios of RTx increase were applied on the number of prevalent patients. The first one was based on the assumption that the average number of transplants performed in Greece during the period 2005-2009 will double by 2020. The second one assumed that Greece will reach by 2020 the transplantation rate of Norway in 2009, the highest transplantation rate reported during that year worldwide. In the MCMC model, the increase of RTx was accomplished by increasing annually by 1% the number of incident patients receiving RTx and reducing accordingly the number of patients performing HD. The Markov chain model projected an increase in the number of prevalent patients’ in Greece by 19.3%, 24.4% and 42.2% in 2020 compared to 2009, depending on the incidence scenario applied. Similarly, the MCMC model projected a 25.0% prevalence increase. In the Markov chain model, the results of the increase in RTx indicated that in 2020 there will be a 64.6% (first scenario) or a 107.2% (second scenario) increase in the number of RTx patients compared to 2009, resulting in total saving of €50.2 and €112.37 million, respectively, for the period 2010-2020. Finally, the increase in RTx accomplished in the MCMC model indicated a 57.9% increase of patients living with a transplanted kidney, resulting in total saving of €68.2 million. The results of both models suggest that performing more kidney transplantations instead of HD would reduce the treatment costs for the country’s healthcare system, while at the same time it would improve the quality of life for a significant number of ESRD patients.