Quality and performance assessment of healthcare providers in Slovakia on the basis of administrative data

• Název práce v češtině: Hodnoceni kvality a vykonnosti poskytovatelu zdravotni pece na Slovensku na bazi administrativnich udaju
• Název v anglickém jazyce: Quality and performance assessment of healthcare providers in Slovakia on the basis of administrative data
• Klíčová slova: hodnocení nemocnic; hodnocení kvality; veřejné hodnocení; zdravotnictví; riziková struktura, Slovensko
• Klíčová slova anglicky: hospital profiling, performance and quality assessment, public performance reports, healthcare; risk-adjustment, Slovakia
• Akademický rok vypsání: 2010/2011
• Typ práce: diplomová práce
• Jazyk práce: angličtina
• Ústav: Institut ekonomických studií (23-IES)
• Vedoucí / školitel: Mgr. Henrieta Tulejová
• Řešitel: skrytý - zadáno vedoucím/školitelem
• Datum přihlášení: 10.06.2011
• Datum zadání: 10.06.2011
• Datum a čas obhajoby: 30.01.2013 00:00
• Místo konání obhajoby: IES
• Datum odevzdání elektronické podoby: 06.01.2013
• Datum proběhlé obhajoby: 30.01.2013
• Oponenti: PhDr. Lucie Bryndová

Seznam odborné literatury

1. Canadian Institute for Health Information (1991). „Technical Notes. Hospital Standardized Mortality Ratios.“
2. Cerrito, Patricia (2010). Text Mining Techniques for Healthcare Provider Quality Determination. Methods for Rank Comparisons.New York: Medical Information Science Reference.
3. Ministerstvo zdravotnictví ČR (2005). Návrh a zhodnocení ukazatelů kvality a výkonnosti akutní nemocniční péče založených na administrativních datech. Available online at <>
4. Dr. Foster Intelligence (2012). “Understanding HSMRs. A Toolkit on Hospital Standardized Mortality Ratios.Version 7.” Available online at <>
5. Centers for Medicare and Medicaid Services (2011). „Medicare Hospital Quality Chartbook. Performance Report on Readmission Measures for Acute Myocardial Infarction, Heart Failure and Pneumonia.“ Available online at <>
6. JCAHO. Developing Risk Adjustment Techniques Using the SAS Systém for Assessing Health Care Quality in the IM System. Available online at http://www2.sas.com/proceedings/sugi22/POSTERS/PAPER225.PDF
7. Shahian, David M. & Sharon-Lise T. Normand (2008). “Comparison of “Risk-Adjusted” Hospital Outcomes”. Circulation. 2008; 117: 1955-1963.

Předběžná náplň práce v anglickém jazyce

In my thesis I will address the issue of quality measurement of healthcare providers. The aim is to examine approaches of how to identify differences in the performance of individual providers.

When comparing performance of hospitals we encounter a major problem of separating various factors that affect the results they report: structure of the patients treated by the establishment, pure chance (bad luck – sometimes the patient dies no mater what you do) and finally the actual treatment provided by the healthcare provider. In order to be able to differentiate between healthcare providers and set the purchasing strategy accordingly – rewarding those providers that offer higher quality and better value health care, we need to filter away the first two influences. In particular, we need to employ risk-adjustment on the data reported to the insurance company in order to control for the particular mix of patients treated by the healthcare provider. Typically, we standardize the data for age, sex and risk factors that describe health status of the patient – diagnosis, comorbidities and/or complications at the admission. Furthermore, after dealing with the issue of risk-adjustment, it is necessary to employ statistical techniques to treat the problem of pure chance and small numbers often encountered in the health data. This is often done by calculating confidence intervals, in this particular problem Bayesian interval estimates are most commonly used.

Issues described above are well treated in Anglo-Saxon literature. Further problems, however, arise when we attempt to employ this analytical framework in Slovak hospitals. Due to the specific nature of hospitals’ reporting process to the health insurance companies, there is virtualy no information on particular procedures provided to patient, drugs administered during his stay in the hospital or the actual times between admission and treatment, etc. In other words, once the patient is admitted to the hospital, the purchaser looses track of any and all treatment provided to the patient other than getting rough information about his movement between hospital departments. Even more troublesome is the quality (or lack thereof) of those few information that are being reported to the insurance companies – namely diagnosis and comorbidities, both of which are fundamental for risk-adjustment. It will be the aim of my thesis to search for ways of how to deal with these problems and gain the best possible estimate of providers‘ quality given the data available at the moment and how it could evolve in light of envisioned introduction of DRG payment mechanism.