Search results for: covariates

Authors: Madaki Umar Yusuf, Mohd Rizam B. Abubakar

Abstract:

The survival censored data with incomplete covariate data is a common occurrence in many studies in which the outcome is survival time. With model when the missing covariates are categorical, a useful technique for obtaining parameter estimates is the EM by the method of weights. The survival outcome for the class of generalized linear model is applied and this method requires the estimation of the parameters of the distribution of the covariates. In this paper, we propose some clinical trials with ve covariates, four of which have some missing values which clearly show that they were fully censored data.

Keywords: EM algorithm, incomplete categorical covariates, ignorable missing data, missing at random (MAR), Weibull Distribution

Procedia PDF Downloads 375

Authors: Shahadut Hossain

Abstract:

Measurement errors in continuous covariates and/or misclassifications in categorical covariates are common in epidemiological studies. Regression analysis ignoring such mismeasurements seriously biases the estimated main and interaction effects of covariates on the outcome of interest. Thus, adjustments for such mismeasurements are necessary. In this research, we propose a Bayesian parametric framework for eliminating deleterious impacts of covariate mismeasurements in logistic regression. The proposed adjustment method is unified and thus can be applied to any generalized linear and non-linear regression models. Furthermore, adjustment for covariate mismeasurements requires validation data usually in the form of either gold standard measurements or replicates of the mismeasured covariates on a subset of the study population. Initial investigation shows that adequacy of such adjustment depends on the sizes of main and validation samples, especially when prevalences of the categorical covariates are low. Thus, we investigate the impact of main and validation sample sizes on the adjusted estimates, and provide a general guideline about these sample sizes based on simulation studies.

Keywords: measurement errors, misclassification, mismeasurement, validation sample, Bayesian adjustment

Procedia PDF Downloads 381

Authors: Amitha Puranik, V. S. Binu, Seena Biju

Abstract:

Missing data is a common problem in spatial analysis especially at the aggregate level. Missing can either occur in covariate or in response variable or in both in a given location. Many missing data techniques are available to estimate the missing data values but not all of these methods can be applied on spatial data since the data are autocorrelated. Hence there is a need to develop a method that estimates the missing values in both response variable and covariates in spatial data by taking account of the spatial autocorrelation. The present study aims to develop a model to estimate the missing data points at the aggregate level in spatial data by accounting for (a) Spatial autocorrelation of the response variable (b) Spatial autocorrelation of covariates and (c) Correlation between covariates and the response variable. Estimating the missing values of spatial data requires a model that explicitly account for the spatial autocorrelation. The proposed model not only accounts for spatial autocorrelation but also utilizes the correlation that exists between covariates, within covariates and between a response variable and covariates. The precise estimation of the missing data points in spatial data will result in an increased precision of the estimated effects of independent variables on the response variable in spatial regression analysis.

Keywords: spatial regression, missing data estimation, spatial autocorrelation, simulation analysis

Procedia PDF Downloads 343

Authors: Yemane Hailu Fissuh, Zhongzhan Zhang

Abstract:

An estimating equation technique is an alternative method of the widely used maximum likelihood methods, which enables us to ease some complexity due to the complex characteristics of time-varying covariates. In the situations, when both the time-varying covariates and left-truncation are considered in the model, the maximum likelihood estimation procedures become much more burdensome and complex. To ease the complexity, in this study, the modified estimating equations those have been given high attention and considerations in many researchers under semiparametric transformation model was proposed. The purpose of this article was to develop the modified estimating equation under flexible and general class of semiparametric transformation models for left-truncated and right censored survival data with time-varying covariates. Besides the commonly applied Cox proportional hazards model, such kind of problems can be also analyzed with a general class of semiparametric transformation models to estimate the effect of treatment given possibly time-varying covariates on the survival time. The consistency and asymptotic properties of the estimators were intuitively derived via the expectation-maximization (EM) algorithm. The characteristics of the estimators in the finite sample performance for the proposed model were illustrated via simulation studies and Stanford heart transplant real data examples. To sum up the study, the bias for covariates has been adjusted by estimating density function for the truncation time variable. Then the effect of possibly time-varying covariates was evaluated in some special semiparametric transformation models.

Keywords: EM algorithm, estimating equation, semiparametric transformation models, time-to-event outcomes, time varying covariate

Procedia PDF Downloads 122

Authors: Yemane Hailu Fissuh, Zhongzhan Zhang

Abstract:

In biomedical researches and randomized clinical trials, the most commonly interested outcomes are time-to-event so-called survival data. The importance of robust models in this context is to compare the effect of randomly controlled experimental groups that have a sense of causality. Causal estimation is the scientific concept of comparing the pragmatic effect of treatments conditional to the given covariates rather than assessing the simple association of response and predictors. Hence, the causal effect based semiparametric transformation model was proposed to estimate the effect of treatment with the presence of possibly time-varying covariates. Due to its high flexibility and robustness, the semiparametric transformation model which shall be applied in this paper has been given much more attention for estimation of a causal effect in modeling left-truncated and right censored survival data. Despite its wide applications and popularity in estimating unknown parameters, the maximum likelihood estimation technique is quite complex and burdensome in estimating unknown parameters and unspecified transformation function in the presence of possibly time-varying covariates. Thus, to ease the complexity we proposed the modified estimating equations. After intuitive estimation procedures, the consistency and asymptotic properties of the estimators were derived and the characteristics of the estimators in the finite sample performance of the proposed model were illustrated via simulation studies and Stanford heart transplant real data example. To sum up the study, the bias of covariates was adjusted via estimating the density function for truncation variable which was also incorporated in the model as a covariate in order to relax the independence assumption of failure time and truncation time. Moreover, the expectation-maximization (EM) algorithm was described for the estimation of iterative unknown parameters and unspecified transformation function. In addition, the causal effect was derived by the ratio of the cumulative hazard function of active and passive experiments after adjusting for bias raised in the model due to the truncation variable.

Keywords: causal estimation, EM algorithm, semiparametric transformation models, time-to-event outcomes, time-varying covariate

Procedia PDF Downloads 92

Authors: Man-Yu Wong, Siyu Zhou, Zhiqiang Cao

Abstract:

In quality of life, health service utilization is an important determinant of medical resource expenditures on Colorectal cancer (CRC) care, a better understanding of the increased utilization of health services is essential for optimizing the allocation of healthcare resources to services and thus for enhancing the service quality, especially for high expenditure on CRC care like Hong Kong region. In assessing the association between the health-related quality of life (HRQOL) and health service utilization in patients with colorectal neoplasm, count data models can be used, which account for over dispersion or extra zero counts. In our data, the HRQOL evaluation is a self-reported measure obtained from a questionnaire completed by the patients, misreports and variations in the data are inevitable. Besides, there are more zero counts from the observed number of clinical consultations (observed frequency of zero counts = 206) than those from a Poisson distribution with mean equal to 1.33 (expected frequency of zero counts = 156). This suggests that excess of zero counts may exist. Therefore, we study tests for detecting zero-inflation in models with measurement error in covariates. Method: Under classical measurement error model, the approximate likelihood function for zero-inflation Poisson regression model can be obtained, then Approximate Maximum Likelihood Estimation(AMLE) can be derived accordingly, which is consistent and asymptotically normally distributed. By calculating score function and Fisher information based on AMLE, a score test is proposed to detect zero-inflation effect in ZIP model with measurement error. The proposed test follows asymptotically standard normal distribution under H0, and it is consistent with the test proposed for zero-inflation effect when there is no measurement error. Results: Simulation results show that empirical power of our proposed test is the highest among existing tests for zero-inflation in ZIP model with measurement error. In real data analysis, with or without considering measurement error in covariates, existing tests, and our proposed test all imply H0 should be rejected with P-value less than 0.001, i.e., zero-inflation effect is very significant, ZIP model is superior to Poisson model for analyzing this data. However, if measurement error in covariates is not considered, only one covariate is significant; if measurement error in covariates is considered, only another covariate is significant. Moreover, the direction of coefficient estimations for these two covariates is different in ZIP regression model with or without considering measurement error. Conclusion: In our study, compared to Poisson model, ZIP model should be chosen when assessing the association between condition-specific HRQOL and health service utilization in patients with colorectal neoplasm. and models taking measurement error into account will result in statistically more reliable and precise information.

Keywords: count data, measurement error, score test, zero inflation

Procedia PDF Downloads 259

Authors: Yemane Hailu Fissuh, Zhongzhan Zhang

Abstract:

a systematic observation from a defined time of origin up to certain failure or censor is known as survival data. Survival analysis is a major area of interest in biostatistics and biomedical researches. At the heart of understanding, the most scientific and medical research inquiries lie for a causality analysis. Thus, the main concern of this study is to investigate the causal effect of treatment on survival time conditional to the possibly time-varying covariates. The theory of causality often differs from the simple association between the response variable and predictors. A causal estimation is a scientific concept to compare a pragmatic effect between two or more experimental arms. To evaluate an average treatment effect on survival outcome, the estimating equation was adjusted for time-varying covariates under the semi-parametric transformation models. The proposed model intuitively obtained the consistent estimators for unknown parameters and unspecified monotone transformation functions. In this article, the proposed method estimated an unbiased average causal effect of treatment on survival time of interest. The modified estimating equations of semiparametric transformation models have the advantage to include the time-varying effect in the model. Finally, the finite sample performance characteristics of the estimators proved through the simulation and Stanford heart transplant real data. To this end, the average effect of a treatment on survival time estimated after adjusting for biases raised due to the high correlation of the left-truncation and possibly time-varying covariates. The bias in covariates was restored, by estimating density function for left-truncation. Besides, to relax the independence assumption between failure time and truncation time, the model incorporated the left-truncation variable as a covariate. Moreover, the expectation-maximization (EM) algorithm iteratively obtained unknown parameters and unspecified monotone transformation functions. To summarize idea, the ratio of cumulative hazards functions between the treated and untreated experimental group has a sense of the average causal effect for the entire population.

Keywords: a modified estimation equation, causal effect, semiparametric transformation models, survival analysis, time-varying covariate

Procedia PDF Downloads 137

Authors: Xiaoqin Wang, Li Yin

Abstract:

Background and purpose: In many practices, one estimates causal effects arising from a complex stochastic process, where a sequence of treatments are assigned to influence a certain outcome of interest, and there exist time-dependent covariates between treatments. When covariates are plentiful and/or continuous, statistical modeling is needed to reduce the huge dimensionality of the problem and allow for the estimation of causal effects. Recently, Wang and Yin (Annals of statistics, 2020) derived a new general formula, which expresses these causal effects in terms of the point effects of treatments in single-point causal inference. As a result, it is possible to conduct the modeling via point effects. The purpose of the work is to study the modeling of these causal effects via point effects. Challenges and solutions: The time-dependent covariates often have influences from earlier treatments as well as on subsequent treatments. Consequently, the standard parameters – i.e., the mean of the outcome given all treatments and covariates-- are essentially all different (null paradox). Furthermore, the dimension of the parameters is huge (curse of dimensionality). Therefore, it can be difficult to conduct the modeling in terms of standard parameters. Instead of standard parameters, we have use point effects of treatments to develop likelihood-based parametric approach to the modeling of these causal effects and are able to model the causal effects of a sequence of treatments by modeling a small number of point effects of individual treatment Achievements: We are able to conduct the modeling of the causal effects from a sequence of treatments in the familiar framework of single-point causal inference. The simulation shows that our method achieves not only an unbiased estimate for the causal effect but also the nominal level of type I error and a low level of type II error for the hypothesis testing. We have applied this method to a longitudinal study of COVID-19 mortality among Scandinavian countries and found that the Swedish approach performed far worse than the other countries' approach for COVID-19 mortality and the poor performance was largely due to its early measure during the initial period of the pandemic.

Keywords: causal effect, point effect, statistical modelling, sequential causal inference

Procedia PDF Downloads 172

Authors: Ya-Mei Chang

Abstract:

This research is intended to apply spatio-temporal point process methods to the dengue fever data in Tainan. The spatio-temporal intensity function of the dataset is assumed to be separable. The kernel estimation is a widely used approach to estimate intensity functions. The intensity function is very helpful to study the relation of the spatio-temporal point process and some covariates. The covariate effects might be nonlinear. An nonparametric smoothing estimator is used to detect the nonlinearity of the covariate effects. A fitted parametric model could describe the influence of the covariates to the dengue fever. The correlation between the data points is detected by the K-function. The result of this research could provide useful information to help the government or the stakeholders making decisions.

Keywords: dengue fever, spatial point process, kernel estimation, covariate effect

Procedia PDF Downloads 323

Authors: Mario Isaza-Ruget, Claudia C. Colmenares-Mejia, Nancy Yomayusa, Camilo A. González, Andres Cely, Jossie Murcia

Abstract:

Quantification of risks associated with complications development from chronic kidney disease (CKD) through accurate survival models can help with patient management. A retrospective cohort that included patients diagnosed with CKD from a primary care program and followed up between 2013 and 2018 was carried out. Time-dependent and static covariates associated with demographic, clinical, and laboratory factors were included. Deep Learning (DL) survival analyzes were developed for three CKD outcomes: CKD stage progression, >25% decrease in Estimated Glomerular Filtration Rate (eGFR), and Renal Replacement Therapy (RRT). Models were evaluated and compared with Random Survival Forest (RSF) based on concordance index (C-index) metric. 2.143 patients were included. Two models were developed for each outcome, Deep Neural Network (DNN) model reported C-index=0.9867 for CKD stage progression; C-index=0.9905 for reduction in eGFR; C-index=0.9867 for RRT. Regarding the RSF model, C-index=0.6650 was reached for CKD stage progression; decreased eGFR C-index=0.6759; RRT C-index=0.8926. DNN models applied in survival analysis context with considerations of longitudinal covariates at the start of follow-up can predict renal stage progression, a significant decrease in eGFR and RRT. The success of these survival models lies in the appropriate definition of survival times and the analysis of covariates, especially those that vary over time.

Keywords: artificial intelligence, chronic kidney disease, deep neural networks, survival analysis

Procedia PDF Downloads 97

Authors: Cindy Davis, Samantha Rayner

Abstract:

Secondary traumatic stress (STS) is an indirect form of trauma affecting the psychological well-being of mental health workers; STS is found to be a prevalent risk in mental health occupations. Various factors impact the development of STS within the literature; including the level of trauma individuals are exposed to and their level of empathy. Research is limited on STS in mental health workers in Australia; therefore, this study examined STS and related factors of empathetic behavior and trauma caseload among mental health workers. The research utilized an online survey quantitative research design with a purposive sample of 190 mental health workers (176 females) recruited via professional websites and unofficial social media groups. Participants completed an online questionnaire comprising of demographics, the secondary traumatic stress scale and the empathy scale for social workers. A standard hierarchical regression analysis was conducted to examine the significance of covariates, traumatized clients, traumatic stress within workload and empathy in predicting STS. The current research found 29.5% of participants to meet the criteria for a diagnosis of STS. Age and past trauma within the covariates were significantly associated with STS. Amount of traumatized clients significantly predicted 4.7% of the variance in STS, traumatic stress within workload significantly predicted 4.8% of the variance in STS and empathy significantly predicted 4.9% of the variance in STS. These three independent variables and the covariates accounted for 18.5% of the variance in STS. Practical implications include a focus on developing risk strategies and treatment methods that can diminish the impact of STS.

Keywords: mental health, PTSD, social work, trauma

Procedia PDF Downloads 299

Authors: Tahmidul Islam

Abstract:

Propensity Score Matching (PSM) technique has been widely used for estimating causal effect of treatment in observational studies. One major step of implementing PSM is estimating the propensity score (PS). Logistic regression model with additive linear terms of covariates is most used technique in many studies. Logistics regression model is also used with cubic splines for retaining flexibility in the model. However, choosing the functional form of the logistic regression model has been a question since the effectiveness of PSM depends on how accurately the PS been estimated. In many situations, the linearity assumption of linear logistic regression may not hold and non-linear relation between the logit and the covariates may be appropriate. One can estimate PS using machine learning techniques such as random forest, neural network etc for more accuracy in non-linear situation. In this study, an attempt has been made to compare the efficacy of Generalized Additive Model (GAM) in various linear and non-linear settings and compare its performance with usual logistic regression. GAM is a non-parametric technique where functional form of the covariates can be unspecified and a flexible regression model can be fitted. In this study various simple and complex models have been considered for treatment under several situations (small/large sample, low/high number of treatment units) and examined which method leads to more covariate balance in the matched dataset. It is found that logistic regression model is impressively robust against inclusion quadratic and interaction terms and reduces mean difference in treatment and control set equally efficiently as GAM does. GAM provided no significantly better covariate balance than logistic regression in both simple and complex models. The analysis also suggests that larger proportion of controls than treatment units leads to better balance for both of the methods.

Keywords: accuracy, covariate balances, generalized additive model, logistic regression, non-linearity, propensity score matching

Procedia PDF Downloads 330

Authors: Alessandra Marcelletti, Giovanni Trovato

Abstract:

Corruption in public offices is found to be the reflection of country-specific features, however, the exact magnitude and the statistical significance of its determinants effect has not yet been identified. The paper aims to propose an estimation method to measure the impact of country fundamentals on corruption, showing that covariates could differently affect the extent of corruption across countries. Thus, we exploit a model able to take into account different factors affecting the incentive to ask or to be asked for a bribe, coherently with the use of the Corruption Perception Index. We assume that discordant results achieved in literature may be explained by omitted hidden factors affecting the agents' decision process. Moreover, assuming homogeneous covariates effect may lead to unreliable conclusions since the country-specific environment is not accounted for. We apply a Finite Mixture Model with concomitant variables to 129 countries from 1995 to 2006, accounting for the impact of the initial conditions in the socio-economic structure on the corruption patterns. Our findings confirm the hypothesis of the decision process of accepting or asking for a bribe varies with specific country fundamental features.

Keywords: Corruption, Finite Mixture Models, Concomitant Variables, Countries Classification

Procedia PDF Downloads 239

Authors: Nursel Koyuncu, Nihal Ata Tutkun

Abstract:

Frailty model is a survival model that takes into account the unobserved heterogeneity for exploring the relationship between the survival of an individual and several covariates. In the recent years, proposed survival models become more complex and this feature causes convergence problems especially in large data sets. Therefore selection of sample from these big data sets is very important for estimation of parameters. In sampling literature, some authors have defined new sampling schemes to predict the parameters correctly. For this aim, we try to see the effect of sampling design in semi-parametric frailty model. We conducted a simulation study in R programme to estimate the parameters of semi-parametric frailty model for different sample sizes, censoring rates under classical simple random sampling and ranked set sampling schemes. In the simulation study, we used data set recording 17260 male Civil Servants aged 40–64 years with complete 10-year follow-up as population. Time to death from coronary heart disease is treated as a survival-time and age, systolic blood pressure are used as covariates. We select the 1000 samples from population using different sampling schemes and estimate the parameters. From the simulation study, we concluded that ranked set sampling design performs better than simple random sampling for each scenario.

Keywords: frailty model, ranked set sampling, efficiency, simple random sampling

Procedia PDF Downloads 181

Authors: Souad Romdhane, Lotfi Belkacem

Abstract:

When referring to actuarial analysis of lifetime, only models accounting for observable risk factors have been developed. Within this context, Cox proportional hazards model (CPH model) is commonly used to assess the effects of observable covariates as gender, age, smoking habits, on the hazard rates. These covariates may fail to fully account for the true lifetime interval. This may be due to the existence of another random variable (frailty) that is still being ignored. The aim of this paper is to examine the shared frailty issue in the Cox proportional hazard model by including two different parametric forms of frailty into the hazard function. Four estimated methods are used to fit them. The performance of the parameter estimates is assessed and compared between the classical Cox model and these frailty models through a real-life data set from the Quebec Pension Plan and then using a more general simulation study. This performance is investigated in terms of the bias of point estimates and their empirical standard errors in both fixed and random effect parts. Both the simulation and the real dataset studies showed differences between classical Cox model and shared frailty model.

Keywords: life insurance-pension plan, survival analysis, risk factors, cox proportional hazards model, multivariate failure-time data, shared frailty, simulations study

Procedia PDF Downloads 329

Authors: Mokhtar Kouki Inès Rekik

Abstract:

This paper focuses on the assessment of the air pollution and morbidity relationship in Tunisia. Air pollution is measured by ozone air concentration and the morbidity is measured by the number of respiratory-related restricted activity days during the 2-week period prior to the interview. Socioeconomic data are also collected in order to adjust for any confounding covariates. Our sample is composed by 407 Tunisian respondents; 44.7% are women, the average age is 35.2, near 69% are living in a house built after the 1980, and 27.8% have reported at least one day of respiratory-related restricted activity. The model consists on the regression of the number of respiratory-related restricted activity days on the air quality measure and the socioeconomic covariates. In order to correct for zero-inflation and heterogeneity, we estimate several models (Poisson, Negative binomial, Zero inflated Poisson, Poisson hurdle, Negative binomial hurdle and finite mixture Poisson models). Bootstrapping and post-stratification techniques are used in order to correct for any sample bias. According to the Akaike information criteria, the hurdle negative binomial model has the greatest goodness of fit. The main result indicates that, after adjusting for socioeconomic data, the ozone concentration increases the probability of positive number of restricted activity days.

Keywords: bootstrapping, hurdle negbin model, overdispersion, ozone concentration, respiratory-related restricted activity days

Procedia PDF Downloads 228

Authors: Saul Estrin, Ute Stephan, Suncica Vujic

Abstract:

Background and research question: Gender differences in pay are present at all organisational levels, including at the very top. One possible way for women to circumvent organizational norms and discrimination is to engage in entrepreneurship because, as CEOs of their own organizations, entrepreneurs largely determine their own pay. While commercial entrepreneurship plays an important role in job creation and economic growth, social entrepreneurship has come to prominence because of its promise of addressing societal challenges such as poverty, social exclusion, or environmental degradation through market-based rather than state-sponsored activities. This opens the research question whether social entrepreneurship might be a form of entrepreneurship in which the pay of men and women is the same, or at least more similar; that is to say there is little or no gender pay gap. If the gender gap in pay persists also at the top of social enterprises, what are the factors, which might explain these differences? Methodology: The Oaxaca-Blinder Decomposition (OBD) is the standard approach of decomposing the gender pay gap based on the linear regression model. The OBD divides the gender pay gap into the ‘explained’ part due to differences in labour market characteristics (education, work experience, tenure, etc.), and the ‘unexplained’ part due to differences in the returns to those characteristics. The latter part is often interpreted as ‘discrimination’. There are two issues with this approach. (i) In many countries there is a notable convergence in labour market characteristics across genders; hence the OBD method is no longer revealing, since the largest portion of the gap remains ‘unexplained’. (ii) Adding covariates to a base model sequentially either to test a particular coefficient’s ‘robustness’ or to account for the ‘effects’ on this coefficient of adding covariates might be problematic, due to sequence-sensitivity when added covariates are correlated. Gelbach’s decomposition (GD) addresses latter by using the omitted variables bias formula, which constructs a conditional decomposition thus accounting for sequence-sensitivity when added covariates are correlated. We use GD to decompose the differences in gaps of pay (annual and hourly salary), size of the organisation (revenues), effort (weekly hours of work), and sources of finances (fees and sales, grants and donations, microfinance and loans, and investors’ capital) between men and women leading social enterprises. Database: Our empirical work is made possible by our collection of a unique dataset using respondent driven sampling (RDS) methods to address the problem that there is as yet no information on the underlying population of social entrepreneurs. The countries that we focus on are the United Kingdom, Spain, Romania and Hungary. Findings and recommendations: We confirm the existence of a gender pay gap between men and women leading social enterprises. This gap can be explained by differences in the accumulation of human capital, psychological and social factors, as well as cross-country differences. The results of this study contribute to a more rounded perspective, highlighting that although social entrepreneurship may be a highly satisfying occupation, it also perpetuates gender pay inequalities.

Keywords: Gelbach’s decomposition, gender gap, returns to social entrepreneurship, values and preferences

Procedia PDF Downloads 217

Authors: Endeshaw Assefa Derso, Maria Gabriella Campolo, Angela Alibrandi

Abstract:

Objectives:Early childhood malnutrition can have long-term and irreversible effects on a child's health and development. This study uses the Bayesian method with spatial variation to investigate the flexible trends of metrical covariates and to identify communities at high risk of injury. Methods: Cross-sectional data on underweight are collected from the 2016 Ethiopian Demographic and Health Survey (EDHS). The Bayesian geo-additive model is performed. Appropriate prior distributions were provided for scall parameters in the models, and the inference is entirely Bayesian, using Monte Carlo Markov chain (MCMC) stimulation. Results: The results show that metrical covariates like child age, maternal body mass index (BMI), and maternal age affect a child's underweight non-linearly. Lower and higher maternal BMI seem to have a significant impact on the child’s high underweight. There was also a significant spatial heterogeneity, and based on IDW interpolation of predictive values, the western, central, and eastern parts of the country are hotspot areas. Conclusion: Socio-demographic and community- based programs development should be considered compressively in Ethiopian policy to combat childhood underweight malnutrition.

Keywords: bayesX, Ethiopia, malnutrition, MCMC, semi-parametric bayesian analysis, spatial distribution, P- splines

Procedia PDF Downloads 44

Authors: Y. Sunecher, N. Mamode Khan, V. Jowaheer

Abstract:

This paper considers the modelling of a non-stationary bivariate integer-valued autoregressive moving average of order one (BINARMA(1,1)) with correlated Poisson innovations. The BINARMA(1,1) model is specified using the binomial thinning operator and by assuming that the cross-correlation between the two series is induced by the innovation terms only. Based on these assumptions, the non-stationary marginal and joint moments of the BINARMA(1,1) are derived iteratively by using some initial stationary moments. As regards to the estimation of parameters of the proposed model, the conditional maximum likelihood (CML) estimation method is derived based on thinning and convolution properties. The forecasting equations of the BINARMA(1,1) model are also derived. A simulation study is also proposed where BINARMA(1,1) count data are generated using a multivariate Poisson R code for the innovation terms. The performance of the BINARMA(1,1) model is then assessed through a simulation experiment and the mean estimates of the model parameters obtained are all efficient, based on their standard errors. The proposed model is then used to analyse a real-life accident data on the motorway in Mauritius, based on some covariates: policemen, daily patrol, speed cameras, traffic lights and roundabouts. The BINARMA(1,1) model is applied on the accident data and the CML estimates clearly indicate a significant impact of the covariates on the number of accidents on the motorway in Mauritius. The forecasting equations also provide reliable one-step ahead forecasts.

Keywords: non-stationary, BINARMA(1, 1) model, Poisson innovations, conditional maximum likelihood, CML

Procedia PDF Downloads 100

Authors: Yung-Feng Yen, Yun-Ju Lai

Abstract:

Background: Animal studies have shown that alcohol exposure may cause immunosuppression and increase the susceptibility to tuberculosis (TB) infection. However, the temporality of alcohol consumption with subsequent TB development remains unclear. This nationwide population-based cohort study aimed to investigate the impact of alcohol exposure on TB development in Taiwanese adults. Methods: We included 46 196 adult participants from three rounds (2001, 2005, 2009) of the Taiwan National Health Interview Survey. Alcohol consumption was classified into heavy, regular, social, or never alcohol use. Heavy alcohol consumption was defined as intoxication at least once/week. Alcohol consumption and other covariates were collected by in-person interviews at baseline. Incident cases of active TB were identified from the National Health Insurance database. Multivariate logistic regression was used to estimate the association between alcohol consumption and active TB, with adjustment for age, sex, smoking, socioeconomic status, and other covariates. Results: A total of 279 new cases of active TB occurred during the study follow-up period. Heavy (adjusted odds ratio [AOR], 5.21; 95% confident interval [CI], 2.41-11.26) and regular alcohol use (AOR, 1.73; 95% CI, 1.26-2.38) were associated with higher risks of incident TB after adjusting for the subject demographics and comorbidities. Moreover, a strong dose-response effect was observed between increasing alcohol consumption and incident TB (AOR, 2.26; 95% CI, 1.59-3.21; P <.001). Conclusion: Heavy and regular alcohol consumption were associated with higher risks of active TB. Future TB control programs should consider strategies to lower the overall level of alcohol consumption to reduce the TB disease burden.

Keywords: alcohol consumption, tuberculosis, risk factor, cohort study

Procedia PDF Downloads 194

Authors: Yoseph Halala Handiso, Zewdi Gebregziabher

Abstract:

Background: Malnutrition in hospitalized patients remains a major public health problem in both developed and developing countries. Despite the fact that malnourished patients are more prone to stay longer in hospital, there is limited data regarding the magnitude of malnutrition and its effect on length of stay among surgical patients in Ethiopia, while nutritional assessment is also often a neglected component of the health service practice. Objective: This study aimed to assess the prevalence of malnutrition at admission and its effect on the length of hospital stay among adult surgical patients in Wolaita Sodo University Comprehensive Specialized Hospital, South Ethiopia, 2022. Methods: A facility-based prospective cohort study was conducted among 398 adult surgical patients admitted to the hospital. Participants in the study were chosen using a convenient sampling technique. Subjective global assessment was used to determine the nutritional status of patients with a minimum stay of 24 hours within 48 hours after admission (SGA). Data were collected using the open data kit (ODK) version 2022.3.3 software, while Stata version 14.1 software was employed for statistical analysis. The Cox regression model was used to determine the effect of malnutrition on the length of hospital stay (LOS) after adjusting for several potential confounders taken at admission. Adjusted hazard ratio (HR) with a 95% confidence interval was used to show the effect of malnutrition. Results: The prevalence of hospital malnutrition at admission was 64.32% (95% CI: 59%-69%) according to the SGA classification. Adult surgical patients who were malnourished at admission had higher median LOS (12 days: 95% CI: 11-13) as compared to well-nourished patients (8 days: 95% CI: 8-9), means adult surgical patients who were malnourished at admission were at higher risk of reduced chance of discharge with improvement (prolonged LOS) (AHR: 0.37, 95% CI: 0.29-0.47) as compared to well-nourished patients. Presence of comorbidity (AHR: 0.68, 95% CI: 0.50-90), poly medication (AHR: 0.69, 95% CI: 0.55-0.86), and history of admission (AHR: 0.70, 95% CI: 0.55-0.87) within the previous five years were found to be the significant covariates of the length of hospital stay (LOS). Conclusion: The magnitude of hospital malnutrition at admission was found to be high. Malnourished patients at admission had a higher risk of prolonged length of hospital stay as compared to well-nourished patients. The presence of comorbidity, polymedication, and history of admission were found to be the significant covariates of LOS. All stakeholders should give attention to reducing the magnitude of malnutrition and its covariates to improve the burden of LOS.

Keywords: effect of malnutrition, length of hospital stay, surgical patients, Ethiopia

Procedia PDF Downloads 17

Authors: Kamyar Moradi

Abstract:

Background: There is mixed evidence regarding the association between recreational cannabinoid use and memory performance. One of the major reasons for the present controversy is different cannabinoid use-related covariates that influence the cognitive status of an individual. Adjustment of these confounding variables provides accurate insight into the real effects of cannabinoid use on memory status. In this study, we sought to investigate the association between recent recreational cannabinoid use and memory performance while correcting the model for other possible covariates such as demographic characteristics and duration, and amount of cannabinoid use. Methods: Cannabinoid users were assigned to two groups based on the results of THC urine drug screen test (THC+ group: n = 110, THC- group: n = 410). THC urine drug screen test has a high sensitivity and specificity in detecting cannabinoid use in the last 3-4 weeks. The memory domain of NIH Toolbox battery and brain MRI volumetric measures were compared between the groups while adjusting for confounding variables. Results: After Benjamini-Hochberg p-value correction, the performance in all of the measured memory outcomes, including vocabulary comprehension, episodic memory, executive function/cognitive flexibility, processing speed, reading skill, working memory, and fluid cognition, were significantly weaker in THC+ group (p values less than 0.05). Also, volume of gray matter, left supramarginal, right precuneus, right inferior/middle temporal, right hippocampus, left entorhinal, and right pars orbitalis regions were significantly smaller in THC+ group. Conclusions: this study provides evidence regarding the acute effect of recreational cannabis use on memory performance. Further studies are warranted to confirm the results.

Keywords: brain MRI, cannabis, memory, recreational use, THC urine test

Procedia PDF Downloads 157

Authors: Pius Marthin, Nihal Ata Tutkun

Abstract:

Survival analysis has become one of the paramount procedures in the modeling of time-to-event data. When we encounter complex survival problems, the traditional approach remains limited in accounting for the complex correlational structure between the covariates and the outcome due to the strong assumptions that limit the inference and prediction ability of the resulting models. Several studies exist on the deep learning approach to survival modeling; moreover, the application for the case of complex survival problems still needs to be improved. In addition, the existing models need to address the data structure's complexity fully and are subject to noise and redundant information. In this study, we design a deep learning technique (CmpXRnnSurv_AE) that obliterates the limitations imposed by traditional approaches and addresses the above issues to jointly predict the risk-specific probabilities and survival function for recurrent events with competing risks. We introduce the component termed Risks Information Weights (RIW) as an attention mechanism to compute the weighted cumulative incidence function (WCIF) and an external auto-encoder (ExternalAE) as a feature selector to extract complex characteristics among the set of covariates responsible for the cause-specific events. We train our model using synthetic and real data sets and employ the appropriate metrics for complex survival models for evaluation. As benchmarks, we selected both traditional and machine learning models and our model demonstrates better performance across all datasets.

Keywords: cumulative incidence function (CIF), risk information weight (RIW), autoencoders (AE), survival analysis, recurrent events with competing risks, recurrent neural networks (RNN), long short-term memory (LSTM), self-attention, multilayers perceptrons (MLPs)

Procedia PDF Downloads 50

Authors: Yanling Li, Linying Ji, Zita Oravecz, Timothy R. Brick, Michael D. Hunter, Sy-Miin Chow

Abstract:

Assessing several individuals intensively over time yields intensive longitudinal data (ILD). Even though ILD provide rich information, they also bring other data analytic challenges. One of these is the increased occurrence of missingness with increased study length, possibly under non-ignorable missingness scenarios. Multiple imputation (MI) handles missing data by creating several imputed data sets, and pooling the estimation results across imputed data sets to yield final estimates for inferential purposes. In this article, we introduce dynr.mi(), a function in the R package, Dynamic Modeling in R (dynr). The package dynr provides a suite of fast and accessible functions for estimating and visualizing the results from fitting linear and nonlinear dynamic systems models in discrete as well as continuous time. By integrating the estimation functions in dynr and the MI procedures available from the R package, Multivariate Imputation by Chained Equations (MICE), the dynr.mi() routine is designed to handle possibly non-ignorable missingness in the dependent variables and/or covariates in a user-specified dynamic systems model via MI, with convergence diagnostic check. We utilized dynr.mi() to examine, in the context of a vector autoregressive model, the relationships among individuals’ ambulatory physiological measures, and self-report affect valence and arousal. The results from MI were compared to those from listwise deletion of entries with missingness in the covariates. When we determined the number of iterations based on the convergence diagnostics available from dynr.mi(), differences in the statistical significance of the covariate parameters were observed between the listwise deletion and MI approaches. These results underscore the importance of considering diagnostic information in the implementation of MI procedures.

Keywords: dynamic modeling, missing data, mobility, multiple imputation

Procedia PDF Downloads 139

Authors: Santosh Sharma

Abstract:

Background: Drug abuse, needle sharing, and risky sexual behaviour are often compounded to increase the risk of HIV transmission. Injecting Drug Users are at the duel risk of needle sharing and risky sexual Behaviour, becoming more vulnerable to STI and HIV. Thus, studying the interface of injecting drug use and risky sexual behaviour is important to curb the pace of HIV epidemic among IDUs. The aim of this study is to determine the factor associated with HIV among injecting drug users in three states of India. Materials and methods: This paper analyzes covariates of multiplicity of risk behavior among injecting drug users. Findings are based on data from Integrated Behavioral and Biological Assessment (IBBA) round 2, 2010. IBBA collects the information of IDUs from the six districts. IDUs were selected on the criteria of those who were 18 years or older, who injected addictive substances/drugs for non-medical purposes at least once in past six month. A total of 1,979 in round 2 were interviewed in the IBBA. The study employs quantitative techniques using standard statistical tools to achieve the above objectives. All results presented in this paper are unweighted univariate measures. Results: Among IDUs, average duration of injecting drugs is 5.2 years. Mean duration between first drug use to first injecting drugs among younger IDUs, belongs to 18-24 years is 2.6 years Needle cleaning practices is common with above two-fifths reporting its every time cleaning. Needle sharing is quite prevalent especially among younger IDUs. Further, IDUs practicing needle sharing exhibit pervasive multi-partner behavior. Condom use with commercial partners is almost 81 %, whereas with intimate partner it is 39 %. Coexistence of needle sharing and unprotected sex enhances STI prevalence (6.8 %), which is further pronounced among divorced/separated/widowed (9.4 %). Conclusion: Working towards risk reduction for IDUs must deal with multiplicity of risk. Interventions should deal with covariates of risk, addressing youth, and risky sexual behavior.

Keywords: IDUs, HIV, STI, behaviour

Procedia PDF Downloads 253

Authors: A.H.Y. Lai, C. Teyra

Abstract:

In Taiwan, there are 16 official indigenous groups, accounting for 2.3% of the total population. Like other indigenous populations worldwide, indigenous peoples in Taiwan have poorer mental health because of their history of oppression and colonisation. Amid the negative narratives, the ethnic identity of cultural minorities is their unique psychological and cultural asset. Moreover, positive socialisation is found to be related to strong ethnic identity. Based on Phinney’s theory on ethnic identity development and social support theory, this study adopted a strength-based approach conceptualising ethnic identity as the central organising principle that linked perceived social support and mental health among indigenous adults in Taiwan. Aims. Overall aim is to examine the effect of ethnic identity and social support on mental health. Specific aims were to examine : (1) the association between ethnic identity and mental health; (2) the association between perceived social support and mental health ; (3) the indirect effect of ethnic identity linking perceived social support and mental health. Methods. Participants were indigenous adults in Taiwan (n=200; mean age=29.51; Female=31%, Male=61%, Others=8%). A cross-sectional quantitative design was implemented using data collected in the year 2020. Respondent-driven sampling was used. Standardised measurements were: Ethnic Identity Scale(6-item); Social Support Questionnaire-SF(6 items); Patient Health Questionnaire(9-item); and Generalised Anxiety Disorder(7-item). Covariates were age, gender and economic satisfaction. A four-stage structural equation modelling (SEM) with robust maximin likelihood estimation was employed using Mplus8.0. Step 1: A measurement model was built and tested using confirmatory factor analysis (CFA). Step 2: Factor covariates were re-specified as direct effects in the SEM. Covariates were added. The direct effects of (1) ethnic identity and social support on depression and anxiety and (2) social support on ethnic identity were tested. The indirect effect of ethnic identity was examined with the bootstrapping technique. Results. The CFA model showed satisfactory fit statistics: x^2(df)=869.69(608), p<.05; Comparative ft index (CFI)/ Tucker-Lewis fit index (TLI)=0.95/0.94; root mean square error of approximation (RMSEA)=0.05; Standardized Root Mean Squared Residual (SRMR)=0.05. Ethnic identity is represented by two latent factors: ethnic identity-commitment and ethnic identity-exploration. Depression, anxiety and social support are single-factor latent variables. For the SEM, model fit statistics were: x^2(df)=779.26(527), p<.05; CFI/TLI=0.94/0.93; RMSEA=0.05; SRMR=0.05. Ethnic identity-commitment (b=-0.30) and social support (b=-0.33) had direct negative effects on depression, but ethnic identity-exploration did not. Ethnic identity-commitment (b=-0.43) and social support (b=-0.31) had direct negative effects on anxiety, while identity-exploration (b=0.24) demonstrated a positive effect. Social support had direct positive effects on ethnic identity-exploration (b=0.26) and ethnic identity-commitment (b=0.31). Mediation analysis demonstrated the indirect effect of ethnic identity-commitment linking social support and depression (b=0.22). Implications: Results underscore the role of social support in preventing depression via ethnic identity commitment among indigenous adults in Taiwan. Adopting the strength-based approach, mental health practitioners can mobilise indigenous peoples’ commitment to their group to promote their well-being.

Keywords: ethnic identity, indigenous population, mental health, perceived social support

Procedia PDF Downloads 77

Authors: Entisar A. Elgmati

Abstract:

Recurrent infant diarrhea is studied using daily data collected in Salvador, Brazil over one year and three months. A logistic regression model is fitted instead of Aalen's additive model using the same covariates that were used in the analysis with the additive model. The model gives reasonably similar results to that using additive regression model. In addition, the problem with the estimated conditional probabilities not being constrained between zero and one in additive model is solved here. Also martingale residuals that have been used to judge the goodness of fit for the additive model are shown to be useful for judging the goodness of fit of the logistic model.

Keywords: additive model, cumulative probabilities, infant diarrhoea, recurrent event

Procedia PDF Downloads 603

Authors: Davod Ahmadigheidari, Isabel Alvarez, Kate Sinclair, Marnie Davidson, Patrick Cortbaoui, Hugo Melgar-Quiñonez

Abstract:

The burden of collecting water tends to disproportionately fall on women and girls in low-income countries. Specifically, women spend between one to eight hours per day fetching water for domestic use in Sub-Saharan Africa. While there has been research done on the global time burden for collecting water, it has been mainly focused on water quality parameters; leaving the relationship between water fetching and health outcomes understudied. There is little available evidence regarding the relationship between water fetching and maternal child care practices. The main objective of this study was to help fill the aforementioned gap in the literature. Data from two surveys in Ethiopia and Malawi conducted by CARE Canada in 2016-2017 were used. Descriptive statistics indicate that women were predominantly responsible for collecting water in both Ethiopia (87%) and Malawi (99%) respectively, with the majority spending more than 30 minutes per day on water collection. With regards to child care practices, in both countries, breastfeeding was relatively high (77% and 82%, respectively); and treatment for malnutrition was low (15% and 8%, respectively). However, the same consistency was not found for weighing; in Ethiopia only 16% took their children for weighting in contrast to 94% in Malawi. These three practices were summed to create one variable for regressions analyses. Unadjusted logistic regression findings showed that only in Ethiopia was time fetching water significantly associated with child care practices. Once adjusted for covariates, this relationship was no longer found to be significant. Adjusted logistic regressions also showed that the factors that did influence child care practices differed slightly between the two countries. In Ethiopia, a lack of access to community water supply (OR= 0.668; P=0.010), poor attitudes towards gender equality (OR= 0.608; P=0.001), no access to land and (OR=0.603; P=0.000), significantly decreased a women’s odd of using positive childcare practices. Notably, being young women between 15-24 years (OR=2.308; P=0.017), and 25-29 (OR=2.065; P=0.028) increased probability of using positive childcare practices. Whereas in Malawi, higher maternal age, low decision-making power, significantly decreased a women’s odd of using positive childcare practices. In conclusion, this study found that even though amount of time spent by women fetching water makes a difference for childcare practices, it is not significantly related to women’s child care practices when controlling the covariates. Importantly, women’s age contributes to child care practices in Ethiopia and Malawi.

Keywords: time fetching water, community water supply, women’s child care practices, Ethiopia, Malawi

Procedia PDF Downloads 173

Authors: Temesgen Sidamo, Prakruti S. Rao, Eleni Akllilu, Workineh Shibeshi, Yumi Park, Yong-Soon Cho, Jae-Gook Shin, Scott K. Heysell, Stellah G. Mpagama, Ephrem Engidawork

Abstract:

The fluoroquinolones (FQs) are used off-label for the treatment of multidrug-resistant tuberculosis (MDR-TB), and for evaluation in shortening the duration of drug-susceptible TB in recently prioritized regimens. Within the class, levofloxacin (LFX) and moxifloxacin (MXF) play a substantial role in ensuring success in treatment outcomes. However, sub-therapeutic plasma concentrations of either LFX or MXF may drive unfavorable treatment outcomes. To the best of our knowledge, the pharmacokinetics of LFX and MXF in Ethiopian patients with MDR-TB have not yet been investigated. Therefore, the aim of this study was to develop a population pharmacokinetic (PopPK) model of levofloxacin (LFX) and moxifloxacin (MXF) and assess the percent probability of target attainment (PTA) as defined by the ratio of the area under the plasma concentration-time curve over 24-h (AUC0-24) and the in vitro minimum inhibitory concentration (MIC) (AUC0-24/MIC) in Ethiopian MDR-TB patients. Steady-state plasma was collected from 39 MDR-TB patients enrolled in the programmatic treatment course and the drug concentrations were determined using optimized liquid chromatography-tandem mass spectrometry. In addition, the in vitro MIC of the patients' pretreatment clinical isolates was determined. PopPK and simulations were run at various doses, and PK parameters were estimated. The effect of covariates on the PK parameters and the PTA for maximum mycobacterial kill and resistance prevention was also investigated. LFX and MXF both fit in a one-compartment model with adjustments. The apparent volume of distribution (V) and clearance (CL) of LFX were influenced by serum creatinine (Scr), whereas the absorption constant (Ka) and V of MXF were influenced by Scr and BMI, respectively. The PTA for LFX maximal mycobacterial kill at the critical MIC of 0.5 mg/L was 29%, 62%, and 95% with the simulated 750 mg, 1000 mg, and 1500 mg doses, respectively, whereas the PTA for resistance prevention at 1500 mg was only 4.8%, with none of the lower doses achieving this target. At the critical MIC of 0.25 mg/L, there was no difference in the PTA (94.4%) for maximum bacterial kill among the simulated doses of MXF (600 mg, 800 mg, and 1000 mg), but the PTA for resistance prevention improved proportionately with dose. Standard LFX and MXF doses may not provide adequate drug exposure. LFX PopPK is more predictable for maximum mycobacterial kill, whereas MXF's resistance prevention target increases with dose. Scr and BMI are likely to be important covariates in dose optimization or therapeutic drug monitoring (TDM) studies in Ethiopian patients.

Keywords: population PK, PTA, moxifloxacin, levofloxacin, MDR-TB patients, ethiopia

Procedia PDF Downloads 84

Authors: Kajori Banerjee, Laxmi Kant Dwivedi

Abstract:

India is a part of the ongoing globalization, contemporary convergence, industrialization and technical advancement that is taking place world-wide. Some of the manifestations of this evolution is rapid demographic, socio-economic, epidemiological and health transition. There has been a considerable increase in non-communicable diseases due to change in lifestyle. This study aims to assess the direction of burden of disease and compare the pressure of infectious diseases against cardio-vascular, endocrine, metabolic and nutritional diseases. The change in prevalence in a ten-year period (2004-2014) is further decomposed to determine the net contribution of various socio-economic and demographic covariates. The present study uses the recent 71st (2014) and 60th (2004) rounds of National Sample Survey. The pressure of infectious diseases against cardio-vascular (CVD), endocrine, metabolic and nutritional (EMN) diseases during 2004-2014 is calculated by Prevalence Rates (PR), Hospitalization Rates (HR) and Case Fatality Rates (CFR). The prevalence of non-communicable diseases are further used as a dependent variable in a logit regression to find the effect of various social, economic and demographic factors on the chances of suffering from the particular disease. Multivariate decomposition technique further assists in determining the net contribution of socio-economic and demographic covariates. This paper upholds evidences of stagnation of the burden of communicable diseases (CD) and rapid increase in the burden of non-communicable diseases (NCD) uniformly for all population sub-groups in India. CFR for CVD has increased drastically in 2004-2014. Logit regression indicates the chances of suffering from CVD and EMN is significantly higher among the urban residents, older ages, females, widowed/ divorced and separated individuals. Decomposition displays ample proof that improvement in quality of life markers like education, urbanization, longevity of life has positively contributed in increasing the NCD prevalence rate. In India’s current epidemiological phase, compression theory of morbidity is in action as a significant rise in the probability of contracting the NCDs over the time period among older ages is observed. Age is found to play a vital contributor in increasing the probability of having CVD and EMN over the study decade 2004-2014 in the nationally representative sample of National Sample Survey.

Keywords: cardio-vascular disease, case-fatality rate, communicable diseases, hospitalization rate, multivariate decomposition, non-communicable diseases, prevalence rate

Procedia PDF Downloads 285