Analysis of multivariate longitudinal substance use outcomes using multivariate mixed cumulative logit model

Background Longitudinal assessments of usage are often conducted for multiple substances (e.g., cigarettes, alcohol and marijuana) and research interests are often focused on the inter-substance association. We propose a multivariate longitudinal modeling approach for jointly analyzing the ordinal multivariate substance use data. Methods We describe how the binary random slope logistic regression model can be extended to the multi-category ordinal outcomes. We also describe how the proportional odds assumption can be relaxed by allowing differential covariate effects on different cumulative logits for multiple outcomes. Data are analyzed from a P01 study that evaluates the usage levels of cigarettes, alcohol and marijuana repeatedly across 8 measurement waves during 7 consecutive years. Results 1263 subjects participated in the study with informed consent, among whom 56.6% are females. Males and females show significant differences in terms of the time trend for substance use. Specifically, males showed steeper trends on cigarette and marijuana use over time compared to females, while less so for alcohol. For all three substances, age effects appear to be different for different cumulative logits, indicating the violation of proportional odds assumption. Conclusions The multivariate mixed cumulative logit model offers the most flexibility and allows one to examine the inter-substance association when proportional odds assumption is violated.

Severity of Infestation Levels of Tunga Penetrans in Central, Kenya: A Bayesian Cumulative Logit Model

Tungiasis is a neglected parasitic disease that significantly affects communities, especially in developing countries. This study developed a Bayesian severity of the jigger infestation model and its spatial counterpart. Putative determinants leading to different levels of infestation and the most affected areas were to be identified through the model. We collected data through a cross-sectional study with a multi-stage sampling design. A structured questionnaire was administered in each household to capture variables used for modelling jigger infestations. The severity of jigger infestation categorized for each individual was modelled against all the other predictor variables. It was also integrated with spatial data to determine the spatial distribution pattern of jigger infestation. A Bayesian multinomial logistic regression model was used to assess the association between various predictors and different infestation levels. Specifically, an ordered Bayesian Severity Hierarchical (OBSH) categorical model was obtained. This model was categorical based on the Counties (1-Nyeri, 2-Murang'a and 3-Kiambu). Results from this model showed that for a one-unit decrease in the poverty index at level 1 (individuals categorized as poor) there was about a 69% increase in the severity of jigger infestation. A one-unit increase in the percentage of clay in the soil increased the odds ratio of the severity of jigger infestation by a factor of 11.21 while a high percentage of nitrogen in the soil lowered the severity of infestation.  Severity of jigger infestation reduced from the baseline, Nyeri County to Kiambu County. It also increased with increasing altitude due to a decrease in nitrogen levels.

How the Covid-19 Pandemic Is Changing Online Food Shopping Human Behaviour in Italy

The advent of the Internet has significantly changed consumption patterns and habits. Online grocery shopping is a way of purchasing food products using a web-based shopping service. The current COVID-19 pandemic is determining a rethinking of purchase choice elements and of consumers’ behavior. This work aims to investigate which characteristics can affect the decision of online food shopping during the pandemic emergency in Italy. In particular, the work aims to analyze the effects of a set of explanatory variables on the level of satisfaction for the food online shopping experience. For achieving this aim, the proportional odds version of the cumulative logit model is carried out. Data derive from an anonymous on-line questionnaire administrated during the first months of the pandemic and filled by 248 respondents. The results of this work highlight that people having familiarity with buying food online, that have a higher educational level and consider food online channels easy to use, appear more satisfied for the food online shopping experience. These findings can be crucial for the future green global challenges as online shopping may help to reach competitive advantages for company sustainability.

Suicidality Among Young Adolescents in Secondary Schools in Hong Kong

Background:Despite concerted effort in suicide research and prevention across countries, youth suicide remains a significant public health concern in Hong Kong and nationwide. This study examined the prevalence and correlates of suicidality among secondary school students in Hong Kong. Methods:Data were derived from the 2016 Youth Sexuality Survey initiated by the Family Planning Association of Hong Kong using a stratified random sample of 3,672 secondary school students aged 12 to 18 years in Hong Kong. Suicidality was measured in four progressive levels (suicidal ideation, suicidal plan, suicidal attempts and suicidal attempts requiring medical attention). Multivariable cumulative logit model analysis was used to identify significant risk factors of suicidality.Results:Female students reported more suicidality than their male counterparts (28.0% versus 17.0%, respectively). Multivariable cumulative logit model analysis showed that female gender, age, unhappy school life, disharmony with classmates, unhappy family life, living with mother, acceptance of mothers’ discipline, not sleeping for more than 9 hours on weekdays, alcohol consumption, non-suicidal self-injury, and dissatisfaction with life were significantly associated with suicidality. Limitations:Due to cross-sectional design, causality between suicidality, individual psychosocial and psychological characteristics cannot be established.Conclusions:Suicidal behaviour among Chinese young adolescents remains prevalent in Hong Kong. Family disintegration, school-related problems, and life satisfaction are significant predictors of suicidal behaviour in this study. There is a pressing need to restore optimal mental health among youth adolescents via interdisciplinary collaboration among schools, healthcare providers, stakeholders and mental health experts. The involvement of stakeholders in the community in suicide research and prevention is pivotal in mental health promotion for young adolescents.

Factors Affecting Levels of Antiretroviral Drug Adherence among Adult Patient Living with HIV in Public Health Facilities of Arba Minch Town, Southern Ethiopia: Using Cumulative Logit Model

Background: Adherence to antiretroviral therapy is essential to reduce the multiplication of the virus and improve disease outcomes. The studies have reported a range of factors influencing antiretroviral therapy adherence at various levels. Almost all studies were modeling the factors based on binary categorization of the adherence. Objective: This study intended to determine the adherence level and its associated factors to antiretroviral therapy among adult people living with human immunodeficiency virus. Methods: This study was a cross-sectional study that employed among 391 adult patients that were selected by simple random sampling. The cumulative Logit model was used to examine the associations between the outcome of antiretroviral therapy adherence and independent variables. Results: The study participants with good level of antiretroviral therapy adherence (67.77%) were approximately four times higher than study participants with fair (17.39%) and good (14.83%) adherence levels. As the duration on ART changed from ≤12 months to >12 months, the odds of high adherence/less adherence increased with approximately 61% (p = 0.0347) across the full scale of adherence levels. The estimated odds of patients with a CD4 ≥ 200cells/mm3 was 1.65 (p = 0.0279) times toward poor level of antiretroviral therapy adherence than the estimated odds of patients with CD4 < 200cells/mm3. Study participants who have single marital status tending to have more poor level of adherence to antiretroviral therapy than patients with married marital status (p = 0.0003). Conclusion: Levels of adherence to the antiretroviral therapy is significantly determined by the duration on antiretroviral therapy, the number of CD4 counts, the types of initial antiretroviral therapy regimens and the marital status of adult people living with HIV/AIDS.

Bayesian Approaches to Assessing the Parallel Lines Assumption in Cumulative Ordered Logit Models

We first review existing literature on cumulative logit models along with various ways to test the parallel lines assumption. Building on the traditional frequentist framework, we introduce a method of Bayesian assessment of null values to provide an alternative way to examine the parallel lines assumption using highest density intervals and regions of practical equivalence. Second, we propose a new hyperparameter cumulative logit model that can improve upon existing ones in addressing several challenges where traditional modeling techniques fail. We use two empirical examples from health research to showcase the Bayesian approaches.

Sufficient Sample Size and Power in Multilevel Ordinal Logistic Regression Models

For most of the time, biomedical researchers have been dealing with ordinal outcome variable in multilevel models where patients are nested in doctors. We can justifiably apply multilevel cumulative logit model, where the outcome variable represents the mild, severe, and extremely severe intensity of diseases like malaria and typhoid in the form of ordered categories. Based on our simulation conditions, Maximum Likelihood (ML) method is better than Penalized Quasilikelihood (PQL) method in three-category ordinal outcome variable. PQL method, however, performs equally well as ML method where five-category ordinal outcome variable is used. Further, to achieve power more than 0.80, at least 50 groups are required for both ML and PQL methods of estimation. It may be pointed out that, for five-category ordinal response variable model, the power of PQL method is slightly higher than the power of ML method.

Strong Consistency of Maximum Likelihood Estimators in Sequential-Cumulative Logit Model

In this article, for the sequential-cumulative logit model, we show that maximum likelihood estimates of regression parameter vector is asymptotically existence and strongly consistent under mild conditions