W S Ho1, S Y Ying, A Burd. 1. Department of Surgery, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong.
Abstract
OBJECTIVE: To evaluate the outcomes of severely burned patients treated at a regional burns unit and to develop a predictive model for survival and length of hospital stay for major burn patients in Hong Kong. DESIGN: Retrospective study. SETTING: Burns unit of a regional public hospital, Hong Kong. PATIENTS: Two hundred and eighty-six severely burned patients treated from March 1993 to February 2000. MAIN OUTCOME MEASURES: Details of demographics, mechanism of burn, extent of burn, incidence of inhalation injury, length of hospital stay, and mortality rate were recorded and entered into a database. Stepwise logistic regression and linear regression were applied to develop a predictive model for mortality and morbidity, respectively. RESULTS: Of 286 major burn patients treated in this 7-year period, 25 patients died from their injuries, yielding a mortality rate of 8.7%. Stepwise logistic regression was applied to develop a predictive model for mortality. We found that inhalation injury, age, and total body surface area involvement were independent significant predictors of death. Accuracy of this predictive model reached 93%. Similarly, stepwise linear regression was used to develop a predictive model for length of hospital stay. Sex, inhalation injury, total body surface area of burn, and total body surface area(2) of burn were significant predictors of length of hospital stay (R(2)=0.2). Only three patients' duration of hospital stay was more than three standard deviations from the predicted length of hospital stay. CONCLUSION: A predictive model for mortality and length of hospital stay has been developed for major burn patients in Hong Kong. This model may help clinicians to counsel patients and relatives at an early stage of care, to provide a basis from which new treatment plans can be compared, and to facilitate efficient allocation of valuable medical resources.
OBJECTIVE: To evaluate the outcomes of severely burned patients treated at a regional burns unit and to develop a predictive model for survival and length of hospital stay for major burn patients in Hong Kong. DESIGN: Retrospective study. SETTING: Burns unit of a regional public hospital, Hong Kong. PATIENTS: Two hundred and eighty-six severely burned patients treated from March 1993 to February 2000. MAIN OUTCOME MEASURES: Details of demographics, mechanism of burn, extent of burn, incidence of inhalation injury, length of hospital stay, and mortality rate were recorded and entered into a database. Stepwise logistic regression and linear regression were applied to develop a predictive model for mortality and morbidity, respectively. RESULTS: Of 286 major burn patients treated in this 7-year period, 25 patients died from their injuries, yielding a mortality rate of 8.7%. Stepwise logistic regression was applied to develop a predictive model for mortality. We found that inhalation injury, age, and total body surface area involvement were independent significant predictors of death. Accuracy of this predictive model reached 93%. Similarly, stepwise linear regression was used to develop a predictive model for length of hospital stay. Sex, inhalation injury, total body surface area of burn, and total body surface area(2) of burn were significant predictors of length of hospital stay (R(2)=0.2). Only three patients' duration of hospital stay was more than three standard deviations from the predicted length of hospital stay. CONCLUSION: A predictive model for mortality and length of hospital stay has been developed for major burn patients in Hong Kong. This model may help clinicians to counsel patients and relatives at an early stage of care, to provide a basis from which new treatment plans can be compared, and to facilitate efficient allocation of valuable medical resources.
Authors: Sandra L Taylor; MaryBeth Lawless; Terese Curri; Soman Sen; David G Greenhalgh; Tina L Palmieri Journal: Burns Date: 2014-05-17 Impact factor: 2.744
Authors: Winston T Richards; Winston A Richards; Makeesha Miggins; Huazhi Liu; David W Mozingo; Darwin N Ang Journal: Am J Surg Date: 2012-09-25 Impact factor: 2.565
Authors: Sandra L Taylor; Soman Sen; David G Greenhalgh; MaryBeth Lawless; Terese Curri; Tina L Palmieri Journal: JAMA Surg Date: 2015-05 Impact factor: 14.766