Assessing Proportionality in the Proportional Odds Model for Ordinal Logistic Regression

Quality of life has been increasingly emphasized in public health research in recent years. Typically, the results of quality of life are measured by means of ordinal scales. In these situations, specific statistical methods are necessary because procedures such as either dichotomization or misinformation on the distribution of the outcome variable may complicate the inferential process. Ordinal logistic regression models are appropriate in many of these situations. This article presents a review of the proportional odds model, partial proportional odds model, continuation ratio model, and stereotype model. The fit, statistical inference, and comparisons between models are illustrated with data from a study on quality of life in 273 patients with schizophrenia. All tested models showed good fit, but the proportional odds or partial proportional odds models proved to be the best choice due to the nature of the data and ease of interpretation of the results. Ordinal logistic models perform differently depending on categorization of outcome, adequacy in relation to assumptions, goodness-of-fit, and parsimony.

Download Full-text

Penerapan Regresi Logistik Ordinal Proportional Odds Model pada Analisis Faktor-Faktor yang Mempengaruhi Kelengkapan Imunisasi Dasar Anak Balita di Provinsi Aceh Tahun 2015

MEDIA STATISTIKA ◽

10.14710/medstat.10.1.37-47 ◽

2017 ◽

Vol 10 (1) ◽

pp. 37

Author(s):

Budyanra Budyanra ◽

Ghaida Nasria Azzahra

Keyword(s):

Logistic Regression ◽

Odds Ratio ◽

Maternal Education ◽

Immunization Coverage ◽

Ordinal Logistic Regression ◽

Proportional Odds Model ◽

Proportional Odds ◽

Number Of Children ◽

Ordinal Logistic Regression Analysis ◽

The Status

Province of Aceh has basic immunization coverage toddler lowest in Indonesia in 2015. even though, this province has Posyandu and Puskesmas ratio per population of the highest in the western region of Indonesia. This data their concerns regarding immunization coverage has not been handled well in Aceh Province. This papers aims to identify variables that affect the status of complete basic immunization of children aged 12-59 months in Aceh by using ordinal logistic regression analysis. Ordinal logistic regression model used is proportional odds models. Data are obtained from Susenas 2015 that was held in March 2015 by BPS-Statistic of Indonesia. Based on the results of processing data, known only 37.7% of children aged 12-59 months in the province of Aceh in 2015 which gets fully immunized, the remaining 50.6% receive primary immunization but is not complete, even about 11.7% have not received basic immunization at all. From the proportional odds model results showed that the number of children born to mothers (odds ratio = 0.88), maternal age at delivery (odds ratio = 1.03), the level of maternal education (odds ratio = 1.22), and the educational level of the household (odds ratio = 1,2) have a significant impact on the status of complete basic immunization of children. Future studies are expected to include the element of timeliness and add other variables and also with other models in ordinal logistic regression.Keywords:Immunization, Ordinal Logistic Regression, Proportional Odds, Susenas

Download Full-text

A Review of Regression Procedures for Randomized Response Data, Including Univariate and Multivariate Logistic Regression, the Proportional Odds Model and Item Response Model, and Self-Protective Responses

Data Gathering, Analysis and Protection of Privacy Through Randomized Response Techniques: Qualitative and Quantitative Human Traits - Handbook of Statistics ◽

10.1016/bs.host.2016.01.016 ◽

2016 ◽

pp. 287-315 ◽

Cited By ~ 4

Author(s):

M.J.L.F. Cruyff ◽

U. Böckenholt ◽

P.G.M. van der Heijden ◽

L.E. Frank

Keyword(s):

Logistic Regression ◽

Item Response ◽

Proportional Odds Model ◽

Response Model ◽

Randomized Response ◽

Item Response Model ◽

Multivariate Logistic Regression ◽

Proportional Odds ◽

Response Data

Download Full-text

Fitting Proportional Odds Models to Educational Data with Complex Sampling Designs in Ordinal Logistic Regression

Journal of Modern Applied Statistical Methods ◽

10.22237/jmasm/1367382300 ◽

2013 ◽

Vol 12 (1) ◽

pp. 235-248 ◽

Cited By ~ 2

Author(s):

Xing Liu ◽

Hari Koirala

Keyword(s):

Logistic Regression ◽

Ordinal Logistic Regression ◽

Proportional Odds ◽

Complex Sampling ◽

Proportional Odds Models ◽

Complex Sampling Designs

Download Full-text

Ordinal Logistic Regression with Partial Proportional Odds for Depression Prediction

IEEE Transactions on Affective Computing ◽

10.1109/taffc.2020.3031300 ◽

2020 ◽

pp. 1-1

Author(s):

Sadari Jayawardena ◽

Julien Epps ◽

Eliathamby Ambikairajah

Keyword(s):

Logistic Regression ◽

Ordinal Logistic Regression ◽

Proportional Odds

Download Full-text

Associated risk factors of underweight among under-five children in Ethiopia using multilevel ordinal logistic regression model

African Health Sciences ◽

10.4314/ahs.v21i1.46 ◽

2021 ◽

Vol 21 (1) ◽

pp. 362-72

Author(s):

Nigussie Adam Birhan ◽

Denekew Bitew Belay

Keyword(s):

Risk Factors ◽

Logistic Regression ◽

Proportional Odds Model ◽

Proportional Odds ◽

Under Five ◽

Single Level ◽

Under Five Children ◽

Under Nutrition ◽

Partial Proportional Odds Model ◽

Associated Risk Factors

Background: Malnutrition is associated with both under nutrition and over nutrition which causes the body to get improp- er amount of nutrients to maintain tissues and organ function. Under nutrition is the result of insufficient intake of food, poor utilization of nutrients due to illnesses, or a combination of these factors. The purpose of this study was to identify associated risk factors and assess the variation of underweight among under-five children of different regions in Ethiopia. Methods: Ethiopian Demography and Health Survey (EDHS-2016) weight-to-age data for under-five children is used. In order to achieve the objective of this study; descriptive, single level and multilevel ordinal logistic regression analysis were used. Results: From a total of 8935 children about 8.1% were severely underweight, 17.1% were moderately underweight and 74.8% were normal. The test of heterogeneity suggested that underweight varies among region and multilevel ordinal model fit data better than single level ordinal model. Conclusion: Educational level of mother, religion, birth order, type of birth, sex of child, mother body mass index, birth size of child, existence of diarrhea for last two weeks before survey, existence of fever for last two weeks before survey, duration of breast feeding, age child and wealth index had significant effect on underweight among under-five children in Ethiopia. The finding revealed that among the fitted multilevel partial proportional odds model, the random intercept model with fixed coefficients is appropriate to assess the risk factors of underweight among under-five children in Ethiopia. The findings of this study have important policy implications. The government should work closely with both the private sector and civil society to teach women to have sufficient knowledge, awareness and mechanisms of improving under-five under- weight for children’s wellbeing. Keywords: Underweight; Partial proportional odds model; Multilevel partial proportional odds model; under-five children.

Download Full-text

Circulating soluble RAGE isoforms are attenuated in obese, impaired-glucose-tolerant individuals and are associated with the development of type 2 diabetes

AJP Endocrinology and Metabolism ◽

10.1152/ajpendo.00146.2017 ◽

2017 ◽

Vol 313 (6) ◽

pp. E631-E640 ◽

Cited By ~ 18

Author(s):

Edwin R. Miranda ◽

Vikram S. Somal ◽

Jacob T. Mey ◽

Brian K. Blackburn ◽

Edward Wang ◽

...

Keyword(s):

Type 2 Diabetes ◽

Logistic Regression ◽

Glucose Tolerance ◽

Odds Ratio ◽

Treatment Strategies ◽

Normal Glucose Tolerance ◽

Ordinal Logistic Regression ◽

Proportional Odds ◽

Soluble Rage

The soluble receptor for advanced glycation end products (sRAGE) may be protective against inflammation associated with obesity and type 2 diabetes (T2DM). The aim of this study was to determine the distribution of sRAGE isoforms and whether sRAGE isoforms are associated with risk of T2DM development in subjects spanning the glucose tolerance continuum. In this retrospective analysis, circulating total sRAGE and endogenous secretory RAGE (esRAGE) were quantified via ELISA, and cleaved RAGE (cRAGE) was calculated in 274 individuals stratified by glucose tolerance status (GTS) and obesity. Group differences were probed by ANOVA, and multivariate ordinal logistic regression was used to test the association between sRAGE isoform concentrations and the proportional odds of developing diabetes, vs. normal glucose tolerance (NGT) or impaired glucose tolerance (IGT). When stratified by GTS, total sRAGE, cRAGE, and esRAGE were all lower with IGT and T2DM, while the ratio of cRAGE to esRAGE (cRAGE:esRAGE) was only lower ( P < 0.01) with T2DM compared with NGT. When stratified by GTS and obesity, cRAGE:esRAGE was higher with obesity and lower with IGT ( P < 0.0001) compared with lean, NGT. In ordinal logistic regression models, greater total sRAGE (odds ratio, 0.91; P < 0.01) and cRAGE (odds ratio, 0.84; P < 0.01) were associated with lower proportional odds of developing T2DM. Reduced values of sRAGE isoforms observed with both obesity and IGT are independently associated with greater proportional odds of developing T2DM. The mechanisms by which each respective isoform contributes to obesity and insulin resistance may reveal novel treatment strategies for diabetes.

Download Full-text

Factors Associated with Incidental Sickness Absence among Employees in One Health Care System

American Journal of Health Promotion ◽

10.4278/ajhp.081117-quan-286 ◽

2009 ◽

Vol 24 (1) ◽

pp. 37-48 ◽

Cited By ~ 10

Author(s):

Chih-Wen Pai ◽

John Mullin ◽

Gina M. Payne ◽

Jeaneeta Love ◽

Gayle O'Connell ◽

...

Keyword(s):

Health Care ◽

Logistic Regression ◽

Health Status ◽

Sickness Absence ◽

Health Care System ◽

Health Risks ◽

One Health ◽

Binary Logistic Regression ◽

Proportional Odds Model ◽

Proportional Odds

Purpose. Assess the association of taking incidental sickness absence with health risks and health status. Design. Observational. Setting. One Midwest health care system. Subjects. Individuals who were employed for 2 years (2006–2007) and had completed at least one health risk appraisal (HRA) in 2007 (N = 3790). Measures. Outcomes were any incidental sickness absence and absence duration in 2007 measured by an absence tracking system. Health risks and health status were estimated by HRAs. Program participation was captured using 7-year HRA data and 5-year wellness data. Analysis. Multivariate, binary logistic regression for the probability of taking any absence day among the overall population as well as four demographic subgroups; proportional odds model for the probability of taking more absence days. Results. Different patterns were observed in association with taking incidental sickness absence among age and gender subgroups. Among the overall population, three health risks (smoking overweight, and use of medication for relaxation) were positively associated with taking absence (at least p <. 05 for all three health risks). Participation in a wellness program for more years was also associated with a less likelihood of taking absence (odds ratio, .72; p = .002). Results from the proportional odds model were consistent with results from the binary logistic regression. Conclusion. Sickness absence is an important productivity concern of employers. Employers may implement early interventions to focus on preventable causes. Special interventions may target absence-causing risks such as smoking behavior and excess body weight. Study limitation includes a lack of measures for psychosocial work environment.

Download Full-text