Napovedovanje tveganja za ponovno hospitalizacijo pacientov s sladkorno boleznijo na osnovi podatkov o bolnišničnih obravnavah

Zeme, Nives

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Napovedovanje tveganja za ponovno hospitalizacijo pacientov s sladkorno boleznijo na osnovi podatkov o bolnišničnih obravnavah
Authors:	ID Zeme, Nives (Author) ID Povalej Bržan, Petra (Mentor) More about this mentor... ID Stožer, Andraž (Comentor)
Files:	MAG_Zeme_Nives_2018.pdf (1,01 MB) MD5: BE8903355C8E1F45CA5499B78B6C45C4 PID: 20.500.12556/dkum/f474d95a-8fc1-43fb-9327-4ea57a9b7400
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FZV - Faculty of Health Sciences
Abstract:	Izhodišča: Sladkorna bolezen sodi med kronične bolezni, ki zaradi svoje velike pogostosti ter zahtevne in kompleksne obravnave predstavljajo velik javno-zdravstveni problem. Ponovni bolnišnični sprejemi odražajo pomanjkljivosti v zdravstvenem sistemu. S pomočjo metod strojnega učenja in podatkov javnega značaja lahko določimo indikatorje, ki statistično značilno vplivajo na ponovno hospitalizacijo. Na podlagi določenih indikatorjev izdelamo napovedne modele za identifikacijo bolnikov, ki so ogroženi za ponovno hospitalizacijo. Raziskovalne metode: V magistrskem delu smo za teoretična izhodišča uporabili deskriptivno metodo zbiranja podatkov. Raziskava v empiričnem delu je temeljila na deskriptivni inferenčni statistični metodi. S statističnim programskim orodjem IBM SPSS 22.0 smo zgradili napovedni model za tveganje ponovne hospitalizacije bolnikov s sladkorno boleznijo. Posamezne spremenljivke smo preverili in statistično analizirali s pomočjo testa hi-kvadrat in neparametričnega testa Mann-Whitney U. Rezultati: Ugotovili smo, da nekateri indikatorji vplivajo na tveganje za ponovno hospitalizacijo v 30 dneh po odpustu iz bolnišnice. Za statistično značilne indikatorje so se izkazali: starost bolnika 74 let [95-odstotni IZ 74,75]; p < 0,001, ležalna doba 7 dni [95-odstotni IZ 7,8]; p < 0,001, število diagnoz 6 [95-odstotni IZ 6,6]; p = 0,047, število procedur 5 [95-odstotni IZ 5,6]; p < 0,000 in mesec odpusta iz bolnišnice 5 [95-odstotni IZ 5,5]; p = 0,001. Pri napovedovanju ponovne hospitalizacije bolnikov v slovenskih in kalifornijskih bolnišnicah obstajajo enaki indikatorji, vendar se med seboj statistično značilno razlikujejo. Diskusija in zaključek: Napovedovanje tveganja za ponovno hospitalizacijo vključuje dobro poznavanje lastnosti bolnika. Te so: zdravstveno stanje, socialno-demografski dejavniki in uporaba zdravstvenih storitev. Pomembna je opredelitev in razumevanje indikatorjev (spremenljivk) ter njihovih vrednosti, ki vplivajo na ponovno hospitalizacijo bolnikov s sladkorno boleznijo.
Keywords:	sladkorna bolezen, ponovna hospitalizacija, podatki o bolnišničnih obravnavah, napovedni model, logistična regresija, odločitveno drevo
Place of publishing:	Maribor
Publisher:	[N. Zeme]
Year of publishing:	2018
PID:	20.500.12556/DKUM-70923
UDC:	616.379-008.64:004.6(043.2)
COBISS.SI-ID:	2426276
NUK URN:	URN:SI:UM:DK:M3RFUZOC
Publication date in DKUM:	27.08.2018
Views:	1915
Downloads:	240
Metadata:
Categories:	FZV
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-ND 4.0, Creative Commons Attribution-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nd/4.0/
Description:	Under the NoDerivatives Creative Commons license one can take a work released under this license and re-distribute it, but it cannot be shared with others in adapted form, and credit must be provided to the author.
Licensing start date:	19.06.2018

Secondary language

Language:	English
Title:	Predicting readmission risk for patients with diabetes mellitus based on hospital claims data
Abstract:	Theoretical background. Diabetes mellitus is a chronic disease, which due to its high frequency and demanding treatment represents a major health care problem. Hospital readmissions reflect deficiencies in health care systems. With the help of machine learning and hospital claims data, we can determine indicators that statistically influence readmissions. Based on certain indicators, we make prediction models for patients at higher risk for readmission. Research methodology. In the thesis descriptive method of data collection was used for theoretical starting points. The research in empirical part of thesis is based on a descriptive inferential statistic method. With the help of data analysis software IBM SPSS 22.0, we developed a prediction model for diabetes patients with higher risk of readmission. Individual variables were inspected and statistically analyzed using hi-square and non-parametric Mann-Whitney U test. Results. With the help of research we found out that some indicators affect the risk of readmission within 30 days of release from hospital. Indicators proven statistically significant were: age of the patient 74 years [95 % CI 74,75]; p < 0.001, lenght of hospital stay (7 days) [95 % CI 7,8]; p < 0.001, number of diagnosis 6 [95 % CI 6,6]; p = 0.047, number of procedures 5 [95 % CI 5,6]; p < 0.000 and the month of release from the hospital 5 [95 % CI 5,5]; p = 0.001. When predicting the patients readmission in Slovenian and Californian hospitals, there were identical indicators with statistically significant differences. Conclusion. Predicting the risk of readmission involves good knowledge of patients characteristics. These are: health status, social-demographic factors and use of the health services. It is important to define and understand the indicators (variables) and their values which influence the readmission of patients with diabetes mellitus.
Keywords:	diabetes mellitus, readmission, hospital claims data, prediction model, logistic regression, decision tree

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back