scholarly journals A Comparison of Classification and Regression Trees and Logistic Regression for Predicting Death of Older Long-Term Care Users in Japan

2015 ◽  
Vol 18 (7) ◽  
pp. A383
Author(s):  
H Lin ◽  
N Sasaki ◽  
S Kunisawa ◽  
T Otsubo ◽  
Y Imanaka
Keyword(s):  
Logistic Regression ◽  
Long Term Care ◽  
Regression Trees ◽  
Classification And Regression Trees ◽  
Term Care ◽  
Classification And Regression

scholarly journals Predicting Incidence of Long-term Care Insurance Certification in Japan with the Kihon Checklist for Frailty Screening Tool: Analysis of Local Government Survey Data

2020 ◽  
Author(s):  
Kumiko Ito ◽  
Hisashi Kawai ◽  
Harukazu Tsuruta ◽  
Shuichi Obuchi
Keyword(s):  
Logistic Regression ◽  
Local Government ◽  
Long Term Care ◽  
Multiple Logistic Regression ◽  
Short Term ◽  
Term Care ◽  
Specificity And Sensitivity ◽  
Long Term Care Insurance ◽  
Level 1

Abstract Background: Predicting incidence of long-term care insurance (LTCI) certification in the short term is of increasing importance in Japan. The present study examined whether the Kihon Checklist (KCL) can be used to predict incidence of LTCI certification (care level 1 or higher) in the short term among older Japanese persons.Methods: In 2015, the local government in Tokyo, Japan, distributed the KCL to all individuals older than 65 years who had not been certified as having a disability or who had already been certified as requiring support level 1–2 according to LTCI system. We also collected LTCI certification data within the 3 months after collecting the KCL data. The data of 17785 respondents were analyzed. First, we selected KCL items strongly associated with incidence of LTCI certification, using stepwise forward-selection multiple logistic regression. Second, we conducted receiver operating characteristic (ROC) analyses for three conditions (1: Selected KCL items, 2: The main 20 KCL items (nos. 1–20), 3: All 25 KCL items). Third, we estimated specificity and sensitivity for each condition.Results: During a 3-month follow-up, 81 (0.5%) individuals required new LTCI certification. Eight KCL items were selected by multiple logistic regression as predictive of certification. The area under the ROC curve in the three conditions was 0.92–0.93, and specificity and sensitivity for all conditions were greater than 80%.Conclusions: Three KCL conditions predicted short-term incidence of LTCI certification. This suggests that KCL items may be used to screen for the risk of incident LTCI certification.

scholarly journals Social participation and the combination of future needs for long-term care and mortality among older Japanese people: a prospective cohort study from the Aichi Gerontological Evaluation Study (AGES)

BMJ Open ◽  
2019 ◽  
Vol 9 (11) ◽  
pp. e030500 ◽  
Author(s):  
Sei Takahashi ◽  
Toshiyuki Ojima ◽  
Katsunori Kondo ◽  
Sayaka Shimizu ◽  
Shunichi Fukuhara ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Cohort Study ◽  
Prospective Cohort Study ◽  
Social Participation ◽  
Prospective Cohort ◽  
Long Term Care ◽  
Multinomial Logistic Regression ◽  
Evaluation Study ◽  
Term Care

ObjectivesOur study aimed to examine the longitudinal association between social participation and both mortality and the need for long-term care (LTC) simultaneously.DesignA prospective cohort study with 9.4 years of follow-up.SettingSix Japanese municipalities.ParticipantsThe participants were 15 313 people who did not qualify to receive LTC insurance at a baseline based on the data from the Aichi Gerontological Evaluation Study (AGES, 2003–2013). They received a questionnaire to measure social participation and other potential confounders. Social participation was defined as participating in at least one organisation from eight categories.Primary and secondary outcome measuresThe primary outcomes were classified into three categories at the end of the 9.4 years observational period: living without the need for LTC, living with the need for LTC and death. We estimated the adjusted OR (AOR) using multinomial logistic regression analyses with adjustment for possible confounders.ResultsThe primary analysis included 9741 participants. Multinomial logistic regression analysis revealed that social participation was associated with a significantly lower risk of the need for LTC (AOR 0.82, 95% CI 0.69 to 0.97) or death (AOR 0.78, 95% CI 0.70 to 0.88).ConclusionsSocial participation may be associated with a decreased risk of the need for LTC and mortality among elderly patients.

Trends in the informal and formal home-care use of older adults in the Netherlands between 1992 and 2012

2015 ◽  
Vol 36 (9) ◽  
pp. 1870-1890 ◽  
Author(s):  
JOUKJE C. SWINKELS ◽  
BIANCA SUANET ◽  
DORLY J. H. DEEG ◽  
MARJOLEIN I. BROESE VAN GROENOU
Keyword(s):  
Logistic Regression ◽  
Home Care ◽  
Informal Care ◽  
Long Term Care ◽  
Substitution Effect ◽  
Large Decrease ◽  
Term Care ◽  
Multi Level ◽  
Over Time

ABSTRACTThis study investigates trends in, and the interdependence of, the use of informal and formal home care of community-dwelling older people over the last two decades in the context of governmental reform of long-term care services and modernisation of informal relationships. Seven observations of the Longitudinal Aging Study Amsterdam covering the time span between 1992 and 2012 were analysed using multi-level logistic regression analysis. The sample entailed 9,585 observations from 3,574 respondents, aged between 65 and 85 years and living independently at each time of measurement. Measures included formal and informal care use, health, physical and cognitive limitations, socio-demographics, partner status, social network, privately paid help and sense of mastery. Results showed that between 1992 and 2012, formal home-care use increased slightly while there was a large decrease in the use of informal care. Multivariate multi-level logistic regression analyses showed a substitution effect between formal and informal care use which decreased over time. Analyses also showed improved cognitive functioning, increased partner availability and social network size, as well as increased use of privately paid care over time. Nevertheless, these positive trends did not explain the large decrease in informal care use. The results regarding informal care use suggest a societal trend of weakened informal solidarity, reflecting increased individualisation and increased availability of formal home care. The decreased substitution effect suggests that, in agreement with current reforms of long-term care, complementary or supplementary forms of care use may be more common in the near future.

scholarly journals Predicting incidence of long-term care insurance certification in Japan with the Kihon Checklist for frailty screening tool: analysis of local government survey data

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Kumiko Ito ◽  
Hisashi Kawai ◽  
Harukazu Tsuruta ◽  
Shuichi Obuchi
Keyword(s):  
Logistic Regression ◽  
Local Government ◽  
Long Term Care ◽  
Multiple Logistic Regression ◽  
Short Term ◽  
Term Care ◽  
Specificity And Sensitivity ◽  
Long Term Care Insurance ◽  
Level 1

Abstract Background Predicting incidence of long-term care insurance (LTCI) certification in the short term is of increasing importance in Japan. The present study examined whether the Kihon Checklist (KCL) can be used to predict incidence of LTCI certification (care level 1 or higher) in the short term among older Japanese persons. Methods In 2015, the local government in Tokyo, Japan, distributed the KCL to all individuals older than 65 years who had not been certified as having a disability or who had already been certified as requiring support level 1–2 according to LTCI system. We also collected LTCI certification data within the 3 months after collecting the KCL data. The data of 17,785 respondents were analyzed. First, we selected KCL items strongly associated with incidence of LTCI certification, using stepwise forward-selection multiple logistic regression. Second, we conducted receiver operating characteristic (ROC) analyses for three conditions (1: Selected KCL items, 2: The main 20 KCL items (nos. 1–20), 3: All 25 KCL items). Third, we estimated specificity and sensitivity for each condition. Results During a 3-month follow-up, 81 (0.5%) individuals required new LTCI certification. Eight KCL items were selected by multiple logistic regression as predictive of certification. The area under the ROC curve in the three conditions was 0.92–0.93, and specificity and sensitivity for all conditions were greater than 80%. Conclusions Three KCL conditions predicted short-term incidence of LTCI certification. This suggests that KCL items may be used to screen for the risk of incident LTCI certification.

scholarly journals Predicting Incidence of Long-term Care Insurance Certification in Japan with the Kihon Checklist for Frailty Screening Tool: Analysis of Local Government Survey Data

2020 ◽  
Author(s):  
Kumiko Ito ◽  
Hisashi Kawai ◽  
Harukazu Tsuruta ◽  
Shuichi Obuchi
Keyword(s):  
Logistic Regression ◽  
Local Government ◽  
Long Term Care ◽  
Multiple Logistic Regression ◽  
Short Term ◽  
Term Care ◽  
Specificity And Sensitivity ◽  
Long Term Care Insurance ◽  
Level 1

Abstract Background: Predicting incidence of long-term care insurance (LTCI) certification in the short term is of increasing importance in Japan. The present study examined whether the Kihon Checklist (KCL) can be used to predict incidence of LTCI certification (care level 1 or higher) in the short term among older Japanese persons.Methods: In 2015, the local government in Tokyo, Japan, distributed the KCL to all individuals older than 65 years who had not been certified as having a disability or who had already been certified as requiring support level 1–2 according to LTCI system. We also collected LTCI certification data within the 3 months after collecting the KCL data. The data of 17785 respondents were analyzed. First, we selected KCL items strongly associated with incidence of LTCI certification, using stepwise forward-selection multiple logistic regression. Second, we conducted receiver operating characteristic (ROC) analyses for three conditions (1: Selected KCL items, 2: The main 20 KCL items (nos. 1–20), 3: All 25 KCL items). Third, we estimated specificity and sensitivity for each condition.Results: During a 3-month follow-up, 81 (0.5%) individuals required new LTCI certification. Eight KCL items were selected by multiple logistic regression as predictive of certification. The area under the ROC curve in the three conditions was 0.92–0.93, and specificity and sensitivity for all conditions were greater than 80%.Conclusions: Three KCL conditions predicted short-term incidence of LTCI certification. This suggests that KCL items may be used to screen for the risk of incident LTCI certification.

scholarly journals COVID-19 mortality risk factors in older people in a long-term care center

2020 ◽  
Author(s):  
Eva Heras ◽  
Pablo Garibaldi ◽  
Maite Boix ◽  
Oliver Valero ◽  
Jorge Castillo ◽  
...  
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Barthel Index ◽  
Long Term Care ◽  
Male Gender ◽  
Bivariate Analysis ◽  
Multivariate Logistic Regression ◽  
Term Care ◽  
Independent Risk Factors

Abstract Objectives: Despite high rates of COVID-19 infection and increased related mortality have been reported among older adults admitted in long-term care facilities, a limited amount of information is available about the natural course of this pandemic and prognostic factors in this population. In the current study, we aimed to investigate the epidemiologic, demographics, clinical, or therapeutic factors that may predict the prognosis in a cohort of COVID-19 infected elderly in a nursing home.Methods: We conducted a retrospective analysis of all COVID-19 confirmed institutionalized elderly in a nursing home transformed into a reference intermediate healthcare facility from March 15 to June 5, 2020. Epidemiological, demographic, and frailty status before infection, and clinical, laboratory, treatment, and outcome data during infection were collected. We used bivariate analysis and multivariate logistic regression to identify risk factors for mortality.Results: The analysis comprised all 100 COVID-19 confirmed cases during the study period. The median age was 85 years; 62% were female. The case fatality rate was 20%. In the bivariate analysis, male gender, fever, respiratory symptoms, severe cognitive decline, a low Barthel index, and lymphocytopenia were significantly associated with mortality. Multivariate logistic regression analysis identified male gender, low Barthel index, no pharmacological treatment, and lymphocytopenia as independent risk factors associated with mortality.Conclusions and Implications: Male gender, low Barthel index, no pharmacological treatment and lymphocytopenia are independent risk factors for COVID-19 mortality in institutionalized elderly patients in long-term care nursing homes. Treatment with hydroxychloroquine and azithromycin was associated with lower mortality in these patients.

Sign in / Sign up

Export Citation Format

Share Document