ABSTRACT
Survival data with a multi-state structure are frequently observed in follow-up studies. An analytic approach based on a multi-state model (MSM) should be used in longitudinal health studies in which a patient experiences a sequence of clinical progression events. One main objective in the MSM framework is variable selection, where attempts are made to identify the risk factors associated with the transition hazard rates or probabilities of disease progression. The usual variable selection methods, including stepwise and penalized methods, do not provide information about the importance of variables. In this context, we present a two-step algorithm to evaluate the importance of variables formulti-state data. Three differentmachine learning approaches (randomforest, gradient boosting, and neural network) as themost widely usedmethods are considered to estimate the variable importance in order to identify the factors affecting disease progression and rank these factors according to their importance. The performance of our proposed methods is validated by simulation and applied to the COVID-19 data set. The results revealed that the proposed two-stage method has promising performance for estimating variable importance.
ABSTRACT
Survival data with a multi-state structure are frequently observed in follow-up studies. An analytic approach based on a multi-state model (MSM) should be used in longitudinal health studies in which a patient experiences a sequence of clinical progression events. One main objective in the MSM framework is variable selection, where attempts are made to identify the risk factors associated with the transition hazard rates or probabilities of disease progression. The usual variable selection methods, including stepwise and penalized methods, do not provide information about the importance of variables. In this context, we present a two-step algorithm to evaluate the importance of variables for multi-state data. Three different machine learning approaches (random forest, gradient boosting, and neural network) as the most widely used methods are considered to estimate the variable importance in order to identify the factors affecting disease progression and rank these factors according to their importance. The performance of our proposed methods is validated by simulation and applied to the COVID-19 data set. The results revealed that the proposed two-stage method has promising performance for estimating variable importance. © 2023 Tech Science Press. All rights reserved.
ABSTRACT
Background and Objective: This study aimed to compare death anxiety and happiness in nurses of Corona wards with those of nurses in other wards in educational-medical centers of Shiraz, Iran, in 2020. Materials and Methods: This descriptive cross-sectional study was conducted on 155 nurses working in Corona wards, and 155 nurses working in other wards of educational-medical centers in Shiraz, Iran. They were selected using a relatively stratified sampling method in proportion to the number of nurses in each educational and medical center. The data were collected using a three-part questionnaire (demographic characteristics, Oxford happiness, Templer death anxiety). The obtained data were then analyzed using SPSS software (version 26) through descriptive and inferential statistical methods. Results: The mean scores of death anxiety in nurses working in Corona wards and other wards were obtained at 8.34±3.06 and 7.15±3.38, respectively. There was a statistically significant difference between the two groups in terms of the mean scores of death anxiety (P<0.05). The mean scores of happiness in nurses of Corona wards and other wards were 39.77±11.14 and 45.43±10.38, respectively. There was a statistically significant difference between the two groups in terms of the mean scores of happiness (P<0.05). Moreover, an inverse relationship was observed between death anxiety and happiness (P<0.001, r=-0.482). Conclusion: It is suggested that in critical situations, such as the COVID-19 pandemic, the mental health of nurses be given more attention. © 2022 Avicenna Journal of Nursing and Midwifery Care.
ABSTRACT
Survival data with a multi-state structure are frequently observed in follow-up studies. An analytic approach based on a multi-state model (MSM) should be used in longitudinal health studies in which a patient experiences a sequence of clinical progression events. One main objective in the MSM framework is variable selection, where attempts are made to identify the risk factors associated with the transition hazard rates or probabilities of disease progression. The usual variable selection methods, including stepwise and penalized methods, do not provide information about the importance of variables. In this context, we present a two-step algorithm to evaluate the importance of variables for multi-state data. Three different machine learning approaches (random forest, gradient boosting, and neural network) as the most widely used methods are considered to estimate the variable importance in order to identify the factors affecting disease progression and rank these factors according to their importance. The performance of our proposed methods is validated by simulation and applied to the COVID-19 data set. The results revealed that the proposed two-stage method has promising performance for estimating variable importance.
ABSTRACT
Background: Identifying the potential risk factors of the length of stay in hospital (LOSH) in COVID-19 patients could help the health system meet future demand for hospital beds. Objectives: This study aimed to determine the factors affecting the length of stay in hospital in COVID-19 patients in Hamadan, the west of Iran. Methods: This cross-sectional study recruited 512 hospitalized COVID-19 patients in Hamadan city. Demographic, clinical, and medical laboratory characteristics of the patients and their survival status were assessed by a checklist. Univariate and multiple negative binomial regressions were used by Stata 12. Results: The median hospitalization length for COVID-19 patients was five days (range: 0 to 47). In the discharged patients, the adjusted incidence rate ratios (95% CI) of LOSH for females, rural residents, patients with a history of diabetes and cardiovascular disease, SPO2 less than 88%, prothrombin time higher than 13 s, platelet count lower than 130 x 1000 mu L, blood sugar higher than 105 mg/dL, and intensive care unit experience were 1.16 (1.03, 1.44), 1.22 (1.03, 1.44), 1.43 (1.07, 1.92), 1.41 (1.23, 1.61), 0.82 (0.71, 0.93), 1.32 (1.11, 1.56), 1.18 (1.03, 1.36), and 1.85 (1.59, 2.17) compared to their references, respectively. Conclusions: Our study added new insight into LOSH determining factors that could be used for future planning in combating the need for hospital beds. The present study revealed that some demographic, social, and clinical variables could increase the IRR of a more extended hospital stay.