Prediction Models of 5-year Mortality after Psychiatric Disorders

碩士 === 高雄醫學大學 === 醫務管理暨醫療資訊學系碩士在職專班 === 102 === Three purposes of this study: 1. To evaluate the changing trends of patient characteristics and hospital characteristics during the study period; 2. To compare the performance indices between artificial neural networks (ANN), logistic regression (LR) a...

Full description

Bibliographic Details
Main Authors: Ting-Fen Huang, 黃庭芬
Other Authors: Hon-Yi Shi
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/32m84v
id ndltd-TW-102KMC05777018
record_format oai_dc
spelling ndltd-TW-102KMC057770182019-05-15T21:43:13Z http://ndltd.ncl.edu.tw/handle/32m84v Prediction Models of 5-year Mortality after Psychiatric Disorders 精神科病人初診後五年死亡預測模式之探討 Ting-Fen Huang 黃庭芬 碩士 高雄醫學大學 醫務管理暨醫療資訊學系碩士在職專班 102 Three purposes of this study: 1. To evaluate the changing trends of patient characteristics and hospital characteristics during the study period; 2. To compare the performance indices between artificial neural networks (ANN), logistic regression (LR) and Cox proportional hazards (COX) models; 3. To conduct the global sensitivity analysis in order to weight these significant predictors. Research Methods This nationwide population-base study retrospectively conducted the claims data from 1996 to 2010. Included criteria were the patients after psychiatric disorders with age larger than 18 years old; the patients with ICD-9-CM diagnosis codes of 290~319, and use patient characteristics and hospital characteristics to identify the impact factors of 5-year mortality after diagnosis of psychiatric disorders.The comparison of performance indices of ANN, LR, and COX models were employed to predict the 5-yar mortality rate. The global sensitivity analysis was also used to weight these significant predictors. Results The results showed that ANN model is better than LR and COX models in predicting these performance indices: sensitivity (40.82% vs. 31.50% vs. 36.37%), NPV (74.23% vs. 70.90% vs. 11.48%), accuracy (71.35% vs. 68.82% vs. 40.53%), area under the accept operating characteristic curve (AUROC) (85% vs. 72% vs. 65%). Overall for the 5-year mortality rate, the ANN model also showed the better performance indices than the LR and COX models. Conclusions and Suggestions The ANN model showed the better performance indices than the LR and COX models. Medical research can therefore use this model as a multi-clinical assessment and decision making, and to identify the most important predictors to improve quality of care. Psychiatric specialist hospital is major medical provider in Taiwan. It is suggested that both physical and mental examinations should be strengthened in order to achieve early detection and early treatment in order to reduce 5-year mortality and consumption of medical resources utilization. Hon-Yi Shi 許弘毅 2014 學位論文 ; thesis 86 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 高雄醫學大學 === 醫務管理暨醫療資訊學系碩士在職專班 === 102 === Three purposes of this study: 1. To evaluate the changing trends of patient characteristics and hospital characteristics during the study period; 2. To compare the performance indices between artificial neural networks (ANN), logistic regression (LR) and Cox proportional hazards (COX) models; 3. To conduct the global sensitivity analysis in order to weight these significant predictors. Research Methods This nationwide population-base study retrospectively conducted the claims data from 1996 to 2010. Included criteria were the patients after psychiatric disorders with age larger than 18 years old; the patients with ICD-9-CM diagnosis codes of 290~319, and use patient characteristics and hospital characteristics to identify the impact factors of 5-year mortality after diagnosis of psychiatric disorders.The comparison of performance indices of ANN, LR, and COX models were employed to predict the 5-yar mortality rate. The global sensitivity analysis was also used to weight these significant predictors. Results The results showed that ANN model is better than LR and COX models in predicting these performance indices: sensitivity (40.82% vs. 31.50% vs. 36.37%), NPV (74.23% vs. 70.90% vs. 11.48%), accuracy (71.35% vs. 68.82% vs. 40.53%), area under the accept operating characteristic curve (AUROC) (85% vs. 72% vs. 65%). Overall for the 5-year mortality rate, the ANN model also showed the better performance indices than the LR and COX models. Conclusions and Suggestions The ANN model showed the better performance indices than the LR and COX models. Medical research can therefore use this model as a multi-clinical assessment and decision making, and to identify the most important predictors to improve quality of care. Psychiatric specialist hospital is major medical provider in Taiwan. It is suggested that both physical and mental examinations should be strengthened in order to achieve early detection and early treatment in order to reduce 5-year mortality and consumption of medical resources utilization.
author2 Hon-Yi Shi
author_facet Hon-Yi Shi
Ting-Fen Huang
黃庭芬
author Ting-Fen Huang
黃庭芬
spellingShingle Ting-Fen Huang
黃庭芬
Prediction Models of 5-year Mortality after Psychiatric Disorders
author_sort Ting-Fen Huang
title Prediction Models of 5-year Mortality after Psychiatric Disorders
title_short Prediction Models of 5-year Mortality after Psychiatric Disorders
title_full Prediction Models of 5-year Mortality after Psychiatric Disorders
title_fullStr Prediction Models of 5-year Mortality after Psychiatric Disorders
title_full_unstemmed Prediction Models of 5-year Mortality after Psychiatric Disorders
title_sort prediction models of 5-year mortality after psychiatric disorders
publishDate 2014
url http://ndltd.ncl.edu.tw/handle/32m84v
work_keys_str_mv AT tingfenhuang predictionmodelsof5yearmortalityafterpsychiatricdisorders
AT huángtíngfēn predictionmodelsof5yearmortalityafterpsychiatricdisorders
AT tingfenhuang jīngshénkēbìngrénchūzhěnhòuwǔniánsǐwángyùcèmóshìzhītàntǎo
AT huángtíngfēn jīngshénkēbìngrénchūzhěnhòuwǔniánsǐwángyùcèmóshìzhītàntǎo
_version_ 1719119033575407616