Claim Missing Document
Check
Articles

Found 2 Documents
Search

Analysis of Predictive Factors for Cognitive Impairment in the Elderly using Logistic Regression and Decision Tree Analysis Han, Myeunghee
Nursing and Health Sciences Journal (NHSJ) Vol. 4 No. 3 (2024): September 2024
Publisher : KHD-Production

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53713/nhsj.v4i3.318

Abstract

No studies have analyzed the path of predicting the experience of cognitive dysfunction by considering various characteristics in elderly, especially focusing on sleep duration. Thus, this study aimed to predict the experience of cognitive dysfunction according to sleep duration in older individuals. This cross-sectional study used data from 3,361 older individuals from the 2021 Community Health Survey (CHS). Participants were included in two groups according to their experience of cognitive dysfunction (yes or no). Sleep duration was categorized into the following three groups: lack of sleep(<6h), normal sleep (6 to <10h), and oversleep (≥10h). Decision tree and logistic regression analyses were used to identify factors related to cognitive dysfunction in elderly. According to the decision model, those who slept for ≥10h had depression and experienced the highest rate (89.2%)of cognitive dysfunction. In contrast, people aged 65-74 years with a lack of sleep or average sleep duration and low stress levels were the least likely to experience cognitive dysfunction (63.0%). Older individuals who were asleep for ≥10h and had depression showed the highest rate of cognitive dysfunction. Community-based programs to improve cognition in the elderly or healthcare providers caring for the elderly need to continuously assess and consider their age, sleep time, and depression to prevent and manage cognition dysfunction in elderly.
Prediction Model for Non-pharmacological Treatment Implement of Hypertension based on Residential Area Kim, Sulbin; Han, Myeunghee
Nursing and Health Sciences Journal (NHSJ) Vol. 4 No. 2 (2024): June 2024
Publisher : KHD-Production

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53713/nhsj.v4i2.354

Abstract

This study aimed to construct a predictive model for the nonpharmacological treatment of hypertension according to residential area using the 2021 Community Health Survey (CHS). This cross-sectional study analyzed the data of 48,662 individuals diagnosed with hypertension. A decision tree analysis was conducted to create a predictive model. A split-sample test was conducted to verify the accuracy of the final model. Multiple logistic regression analysis was conducted to identify the factors related to the implementation of nonpharmacological treatment. The prediction model identified that subjects who lived in a “rural” area, did not complete hypertension management education, and did not respond to the written health information literacy question showed the lowest probability of performing nonpharmacological treatment at 10.2%. Conversely, those who lived in a “city”, had completed hypertension education, and had above-average life satisfaction were most likely to implement the program (45.0%). Multiple logistic regression results showed that those who live in a city, have a good subjective health level, quit smoking, a high level of understanding of written health information, participate in hypertension management education, engage in economic activities, and have a high level of education or of life satisfaction had a high possibility of implementing nonpharmacological treatment of hypertension. Providing customized hypertension management education by identifying education levels of individuals with hypertension and ensuring their comprehension of written medical information will be effective in improving the rate of nonpharmacological treatment of hypertension.