Weibull Regression Model Research Articles

The Weibull Regression Model Analysis of Mahakam River Water Pollution Potential

Mahakam River has a vital role in the lives of the people of the East Kalimantan province, includingproviding a raw source of clean water. The multi-activity of the Mahakam River watershed, as a watertraffic lane, mining, fisheries, hotels, restaurants, and resident houses, has the potential to produce wasteinto the water. Increasing waste in the water flow can increase the pollution potential of river water,threatening people’s health. Therefore, precaution is necessary. In this research, statistical preventionwas proposed, providing information to the East Kalimantan people regarding the factors affecting thepollution potential of the Mahakam River through Weibull regression (WR) modeling on dissolved oxygen (DO) data 2022. Research data was secondary data provided by the Life Environmental Departmentof East Kalimantan province. The WR model is a Weibull distribution that is directly influenced by covariates. WR model consists of Weibull survival regression, cumulative distribution regression, hazardregression, and Weibull mean regression. This research aims to obtain the factors affecting the pollution potential and to provide the pollution potential information of Mahakam River 2022. The researchconcluded that factors influencing the pollution potential of the Mahakam River were watercolor degreeand nitrate concentration. Applying the WR model to DO data 2022 was able to provide the pollutionpotential information of Mahakam River, namely the probability of river water isn’t polluted is 0.6555,or the probability of the polluted river water is 0.3445, the pollution rate is 6 locations are polluted forevery 10 mg/L DO, and the DO average of river water is 5.7450 mg/L. Increasing water color degreeand nitrate concentration will decrease the probability of the Mahakam River being polluted, increasethe probability of the Mahakam River being polluted, increase the pollution rate, and reduce the DO ofMahakam River water.

Fracture analysis under modes I and II of adhesive joints on CFRP in saline environment

This study analyzes the delamination behavior of adhesive joints after exposure to a saline environment for zero, one, and twelve weeks. Delamination was assessed under static and fatigue loading conditions in fracture Modes I and II, with a detailed analysis of fracture surfaces using Scanning Electron Microscopy (SEM) and Backscattered Electron (BSE) detection. The 3D images reveal significant morphological differences in fracture surfaces, showing variations in fatigue lines and the presence of impurities depending on the fracture mode. A probabilistic fatigue life analysis was performed using a Weibull regression model, showing notable changes, especially in Mode I at a high number of cycles. A chemical analysis using EDX and FTIR-ATR complemented the mechanical study, revealing an increase in sodium and chlorine concentrations with prolonged saline exposure. Oxidative degradation was also observed, with carbonyl groups increasing significantly over time, particularly in areas most exposed to the saline mist.

Left, Right, Midpoint and Random Point Imputation Techniques for Weibull Regression Model with Right and Interval-Censored Data”

The research explores several imputation techniques, namely left, right, midpoint and random imputations for the MLE of the Weibull regression model with covariate for uncensored, right, and interval-censored data. A simulation study is conducted to obtain the parameter estimates of the model with different imputation techniques, sample sizes, and censoring proportions and its performance are evaluated using bias, standard error (SE), and root mean square error (RMSE). The simulation result indicates that midpoint imputation technique outperformed other techniques based on the lowest RMSE values. Finally, the model was fit to diabetic nephropathy data were fitted to the model using selected imputation techniques. The result concluded that the Weibull regression model may provide a good fit to the data and that the covariate, gender has a significant effect on the survival time of patient kidneys.

Novel Extended Weibull Regression Model for Investigating the Survival Times of Breast Cancer Patients

The new five-parameter alpha power generalized odd generalized exponentiated Weibull distribution is introduced, and some of its structural properties are derived. Its parameters are estimated by maximum likelihood, and a simulation study examines the accuracy of the estimates. A regression model is constructed based on the logarithm of the proposed distribution to investigate the survival times of breast cancer patients in Bauchi State, Nigeria. The applicability and flexibility of the novel model is proven by means of cancer dataset.

Gamma–lindley regression cure model for corporate credit default prediction

Predicting the firm’s default over time is marked persistently with certain flaws for firms seeking to avoid their exposure to credit risk and losses induced by defaults. From this perspective, a Gamma–lindley cure model, relying on a Gamma–lindley distribution and the survival analysis framework, was introduced to model the default probability and time of corporate credits referring to their inference ability once the firm is likely to default. In addition, it extends the standard mixture cure model allowing more flexibility for modeling lifetime data through relaxing the independence assumption of the probability and the time of default. Furthermore, it handles the missing defaulting labels as a latent variable and identifies three classes of default probability distribution, namely short, medium, and long-term through its potentially accurate classification method. A relevant parameter-learning procedure was described resting on the Maximum Likelihood approach. The results of an empirical analysis on a firms dataset reveal that the advanced Gamma–lindley cure model has a rather effective performance in terms of default time prediction compared to Exponential, Gamma, Weibull, and Cox proportional hazards regression modeling specifications, which yield a meaningful solution for timely credit risk management and cluster firms defaulter in three classes, enabling a deeper analysis of firm’s defaults situation. The numerical experiments were conducted with Python language.

How to Perform a Statistical Analysis of Non-Destructive Degradation Data to Study Crack Growth in Wind Blades as a Function of the Number of Cycles

Objective: The aim of this paper is to show the application of the Exponential Regression Model to study the failure time of a system over time. The data was taken from the paper (Wang et al., 2021). Theoretical framework: The linear regression model, where the response is associated with the explanatory variables by means of a linear model, is the best known. To formulate the model, it is necessary to specify a deterministic component and a random (stochastic) component. The exponential model should be used when it is assumed that risk is constant over time. Once the model has been specified, its parameters are estimated. In the absence of normal errors and especially in the presence of censoring, a more appropriate option is the maximum likelihood method. As the equations are non-linear in their parameters and do not have an analytical solution, it is necessary to use the Newton-Raphson numerical method. Due to the simplicity of the exponential regression model, few situations in practice are adequately adjusted by this model. The Weibull regression model is widely used in survival analysis (Bastos Lyra et al., 2008) Method: Data taken from the article (Wang et al., 2021) e foi utilizado a Regressão Exponencial para modelar os dados. Final Considerations: This result indicates that Exponential Regression is a very viable option when the aim is to monitor the number of cracks as a function of time to check whether a system will fail. In this specific Case Study, it is clear that one of the blades is better than the others and that there are some Probability Distribution models that are suitable for the model. Implications of the research: The use cases of Exponential Regression are more restricted in the scientific literature and because of this, this case study is interesting to show that this model is effective for treating data in the area of reliability. Originality/value: Despite being a well-known statistical tool, Exponential Regression has a specific application in monitoring the number of cracks over time in a system.

Body Roundness Index and All-Cause Mortality Among US Adults

Obesity, especially visceral obesity, is an established risk factor associated with all-cause mortality. However, the inadequacy of conventional anthropometric measures in assessing fat distribution necessitates a more comprehensive indicator, body roundness index (BRI), to decipher its population-based characteristics and potential association with mortality risk. To evaluate the temporal trends of BRI among US noninstitutionalized civilian residents and explore its association with all-cause mortality. For this cohort study, information on a nationally representative cohort of 32 995 US adults (age ≥20 years) was extracted from the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018 and NHANES Linked Mortality File, with mortality ascertained through December 31, 2019. Data were analyzed between April 1 and September 30, 2023. Biennial weighted percentage changes in BRI were calculated. Restricted cubic spline curve was used to determine optimal cutoff points for BRI. The survival outcome was all-cause mortality. Mortality data were obtained from the Centers for Disease Control and Prevention website and linked to the NHANES database using the unique subject identifier. Weibull regression model was adopted to quantify the association between BRI and all-cause mortality. Among 32 995 US adults, the mean (SD) age was 46.74 (16.92) years, and 16 529 (50.10%) were women. Mean BRI increased gradually from 4.80 (95% CI, 4.62-4.97) to 5.62 (95% CI, 5.37-5.86) from 1999 through 2018, with a biennial change of 0.95% (95% CI, 0.80%-1.09%; P < .001), and this increasing trend was more obvious among women, elderly individuals, and individuals who identified as Mexican American. After a median (IQR) follow-up of 9.98 (5.33-14.33) years, 3452 deaths (10.46% of participants) from all causes occurred. There was a U-shaped association between BRI and all-cause mortality, with the risk increased by 25% (hazard ratio, 1.25; 95% CI, 1.05-1.47) for adults with BRI less than 3.4 and by 49% (hazard ratio, 1.49; 95% CI, 1.31-1.70) for those with BRI of 6.9 or greater compared with the middle quintile of BRI of 4.5 to 5.5 after full adjustment. This national cohort study found an increasing trend of BRI during nearly 20-year period among US adults, and importantly, a U-shaped association between BRI and all-cause mortality. These findings provide evidence for proposing BRI as a noninvasive screening tool for mortality risk estimation, an innovative concept that could be incorporated into public health practice pending consistent validation in other independent cohorts.

Restorations performance after selective caries removal to soft dentine: 18-month follow-up of a controlled clinical trial

ObjectiveThis randomized controlled clinical trial aimed to evaluate the clinical performance of composite resin restorations placed after selective caries removal to soft dentin (SCRSD) or stepwise excavation (SW) over an 18-month period. MethodsInclusion criteria were patients with permanent molars and/or premolars presenting deep caries lesions (≥50 % of the dentin thickness). Teeth were submitted to SCRSD (n = 76) or SW (n = 76). Evaluations were performed based on FDI (World Dental Federation) criterion. Survival analysis was performed to estimate the survival of restorations and its association with clinical variables and socio-demographic characteristics (adjusted Weibull regression model). The study was registered on Registro Brasileiro de Ensaios Clínicos (ReBEC 65ntbc). ResultsThe follow-up period ranged from 6 to 18 months (mean ± SD 16.3 ± 3.4 months). Patients' ages ranged from 9 to 55 years (mean ± SD 29 ± 10.5 years). A total of 135 teeth (SCRSD = 72; SW = 63) from 101 patients were evaluated. There were 4 failures in the SCRSD group (2 loss of restoration in need of replacement and 2 marginal fractures in need of repair) and 2 in the SW group (fracture in need of repair). The association between explanatory variables and restoration failure showed similar success rates for SW (99.4 %) and SCRSD (97.9 %) (p = 0.16). Patients presenting gingivitis (≥20 % of sites with gingival bleeding) had 8.50 times more risk for failure than those with <20 % of bleeding sites (p = 0.03). ConclusionThis study showed that placing a composite resin restoration over soft dentin (after SCRSD) did not affect its clinical performance after 18 months. Clinical significanceSCRSD is a feasible treatment option for the management of deep caries lesions in permanent teeth as it preserves tooth vitality and tooth structure without compromising restoration longevity.

Residual Analysis for Poisson-Exponentiated Weibull Regression Models with Cure Fraction

The use of cure-rate survival models has grown in recent years. Even so, proposals to perform the goodness of fit of these models have not been so frequent. However, residual analysis can be used to check the adequacy of a fitted regression model. In this context, we provide Cox–Snell residuals for Poisson-exponentiated Weibull regression with cure fraction. We developed several simulations under different scenarios for studying the distributions of these residuals. They were applied to a melanoma dataset for illustrative purposes.

Statistical analysis on the incidence and predictors of death among second-line ART patients in public hospitals of North Wollo and Waghemira Zones, Ethiopia, 2021

Acquired immune deficiency virus, caused by the human immunodeficiency virus, is a significant global health concern. Sub-Saharan Africa particularly Ethiopia faces a high prevalence of human immunodeficiency virus. In low-income settings like Ethiopia, early mortality rates are elevated due to severe opportunistic infections and advanced disease at Anti-retroviral treatment initiation. Despite available treatments, delayed treatment initiation among Human Immunodeficiency Virus -infected individuals in Africa, including Ethiopia, leads to disease progression and increased mortality risk. This study aimed to identify the factors contributing to the death of HIV patients under treatment at second line regimen in public hospitals of North Wollo and Waghemira Zones. A retrospective cohort study with 474 patients was conducted in selected hospitals of North Wollo and Waghemira Zones. A parametric Weibull regression model was employed, and the adjusted hazard ratio served as the measure of association. Variables significantly affected the outcome of the study was determined at a p-value < 0.05, along with a 95% confidence interval for the variables. The patients were within the average age of 38.6(standard deviation ± 12.5) years and majority (45.57%) had no formal education. The overall death incidence rate among second-line anti-retroviral treatment patients was 1.98 per 100-person years [95% CI 1.4—2.9%]. Poor adherence to antiretroviral treatment, male gender, and being underweight significantly increased the hazard of death. Conversely, increased anti-retroviral treatment duration had a significant and negative impact, reducing the hazard of death among patients. The study reveals a high incidence of death among second line anti-retroviral treatment users. Independent predictors include poor adherence, male gender, and underweight status, all significantly increasing the risk of death. On the positive side, the hazard of death decreases with longer anti-retroviral treatment duration. A critical concern and counseling should be given for better ART adherence, to change their nutritional status and for males.

An enhanced method for evaluating the effectiveness of protective devices for road safety application

This paper presents an enhanced probabilistic approach to estimate the real-world safety performance of new device concepts for road safety applications from the perspective of Powered Two-Wheeler (PTW) riders who suffer multiple injuries in different body regions. The proposed method estimates the overall effectiveness of safety devices for PTW riders by correlating computer simulations with various levels of actual injuries collected worldwide from accident databases. The study further develops the methodology initially presented by Johnny Korner in 1989 by introducing a new indicator, Global Potential Damage (GPD), that overcomes the limitations of the original method, encompassing six biomechanical injury indices estimated in five body regions. A Weibull regression model was fit to the field data using the Maximum Likelihood Method with boundaries at the 90% confidence level for the construction of novel injury risk curves for PTW riders. The modified methodology was applied for the holistic evaluation of the effectiveness of a new safety system, the Belted Safety Jacket (BSJ), in head-on collisions across multiple injury indices, body regions, vehicle types, and speed pairs without sub-optimizing it at specific crash severities. A virtual multi-body environment was employed to reproduce a selected set of crashes. The BSJ is a device concept comprising a vest with safety belts to restrict the rider’s movements relative to the PTW during crashes. The BSJ exhibited 59% effectiveness, with an undoubted benefit to the head, neck, chest, and lower extremities. The results show that the proposed methodology enables an overall assessment of the injuries, thus improving the protection of PTW users. The novel indicator supports a robust evaluation of safety systems, specifically relevant in the context of PTW accidents.

Incidence of amyotrophic lateral sclerosis in Chile

Objective This study aimed to estimate amyotrophic lateral sclerosis (ALS) incidence and survival rates in the Metropolitan region of Chile. Methods We conducted a cohort study of ALS cases in the Metropolitan Region from 2016 to 2019. A total of 219 ALS patients were recruited from Corporación ELA-Chile registry, in collaboration with neurologists from Sociedad de Neurología, Psiquiatría y Neurocirugía de Chile. We calculated incidence rates by sex and age and determined median survival from onset and diagnosis. Survival analysis used the Kaplan–Meier statistic, estimating hazard ratios for age, sex, time from symptom onset and from diagnosis using a Weibull regression model. All analyses were done using R 4.1.0. Results Overall, ALS diagnosis incidence was 0.97 cases per 100,000 inhabitants, peaking in the 70–79 age group and declining thereafter. The male-to-female ratio was 1.23. The median time to death from diagnosis was 2.3 years (95% confidence interval [CI]: 1.9–2.5), and from the first symptom, it was 3.1 years (95% CI: 2.8–3.5). Conclusions This is the first population-based study reporting ALS incidence and survival rates in Chile’s Metropolitan region. Incidence resembled other Latin American studies. Median survival from diagnosis and from the first symptom were in line with previous findings. Our results corroborated lower ALS rates in Latin America, consistent with prior research.

PERBANDINGAN ANALISIS SURVIVAL MENGGUNAKAN REGRESI COX PROPORTIONAL HAZARD DAN REGRESI WEIBULL PADA PASIEN COVID-19 DI RSUD TAMAN HUSADA BONTANG

COVID-19 is brought on by the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) and transmitted to humans through animal. SARS-CoV-2 infection affects patient's metabolism and causes hyperinflammatory. This condition affects individuals with risk factors such as age, gender, diabetes, heart disease, hypertension, Chronic Obstructive Pulmonary Disease (COPD), obesity, and Acute Respiratory Distress Syndrome (ARDS). One approach to figuring out the association between the time of an occurrence and the independent factors is the Cox Proportional Hazard Regression. The Cox PH regression is a semiparametric model because it doesn’t require a specific distribution test. There is a parametric model used in modeling and analyzing failure time data, namely Weibull regression. The case study is patients with COVID-19 at Taman Husada Bontang Regional Public Hospital who underwent hospitalization from August 2021 to September 2021 data. Based on the Cox PH Regression and Weibull Regression models, variables that affect the survival time of COVID-19 patients are heart disease and ARDS. The AIC value obtained using the Cox Proportional Hazard regression is 635.6149, this value is smaller than the Weibull regression which is 745.5509 so the use of survival analysis with the Cox Proportional Hazard regression is better than the Weibull regression in this case.

Modeling clustered count data using mixed effect discrete Weibull regression model with cubic splines

This paper investigates the use of mixed-effect discrete Weibull (DW) regression model with cubic splines for clustered count data. DW regression model can be used for both over and under-dispersed data thus, it can be an appropriate choice for researchers to use without having the risk to violate assumptions for using a particular model distribution. We used a cubic spline function to model the non-linear relationship between the outcome variable and the covariate. The proposed model is applied to a clustered dataset from the eighth Korean National Health and Nutrition Examination Survey (KNHANES VIII) from 2019 to assess the factors influencing the amount of alcohol drink in 17 different regions. We compared the results with our proposed model with those of Poisson, negative binomial, zero-inflated Poisson and zero-inflated negative binomial regression models in the presence of spline function. The results show that using DW regression model with cubic spline gives the best fit. We also held a simulation study to investigate the performance and the robustness of DW with a spline function. It turns out that DW regression model with a spline function can flexibly model clustered count data with various dispersion and data with many zeros.

Relationship between sex, body size, and cardiac resynchronization therapy benefit: A patient-level meta-analysis of randomized controlled trials

BackgroundWomen might benefit more than men from cardiac resynchronization therapy (CRT) and do so at shorter QRS durations. ObjectiveThis meta-analysis was performed to determine whether sex-based differences in CRT effects are better accounted for by height, body surface area (BSA), or left ventricular end-diastolic dimension (LVEDD). MethodsWe analyzed patient-level data from CRT trials (MIRACLE, MIRACLE ICD, MIRACLE ICD II, REVERSE, RAFT, COMPANION, and MADIT-CRT) using bayesian hierarchical Weibull regression models. Relationships between QRS duration and CRT effects were examined overall and in sex-stratified cohorts; additional analyses indexed QRS duration by height, BSA, or LVEDD. End points were heart failure hospitalization (HFH) or death and all-cause mortality. ResultsCompared with men (n = 5628), women (n = 1439) were shorter (1.62 [interquartile range, 1.57–1.65] m vs 1.75 [1.70–1.80] m; P < .001), with smaller BSAs (1.76 [1.62–1.90] m2 vs 2.02 [1.89–2.16] m2; P < .001). In adjusted sex-stratified analyses, the reduction in HFH or death was greater for women (hazard ratio, 0.54; credible interval, 0.42–0.70) than for men (hazard ratio, 0.77; credible interval, 0.66–0.89; Pinteraction = .009); results were similar for all-cause mortality even after adjustment for height, BSA, and LVEDD. Sex-specific differences were observed only in nonischemic cardiomyopathy. The effect of CRT on HFH or death was observed at a shorter QRS duration for women (126 ms) than for men (145 ms). Indexing QRS duration by height, BSA, or LVEDD attenuated sex-specific QRS duration thresholds for the effects of CRT on HFH or death but not on mortality. ConclusionAlthough body size partially explains sex-specific QRS duration thresholds for CRT benefit, it is not associated with the magnitude of CRT benefit. Indexing QRS duration for body size might improve selection of patients for CRT, particularly with a “borderline” QRS duration.

Sample size calculations for clustered count data based on zero-inflated discrete Weibull regression models

Sample size calculations for clustered count data based on zero-inflated discrete Weibull regression models

Absolute treatment effects of novel antidiabetic drugs on a composite renal outcome: meta-analysis of digitalized individual patient data

BackgroundAbsolute treatment benefits—expressed as numbers needed to treat—of the glucose lowering and cardiovascular drugs, glucagon-like peptide-1 (GLP-1) receptor agonists and sodium-glucose transporter 2 (SGLT2) inhibitors on renal outcomes remain uncertain. With the present meta-analysis of digitalized individual patient data, we aimed to display and compare numbers needed to treat of both drugs on a composite renal outcome.MethodsFrom Kaplan–Meier plots of major cardiovascular outcome trials of GLP-1 receptor agonists and SGLT2 inhibitors vs. placebo, we digitalized individual patient time-to-event information on composite renal outcomes with WebPlotDigitizer 4.2; numbers needed to treat from individual cardiovascular outcome trials were estimated using parametric Weibull regression models and compared to original data. Random-effects meta-analysis generated meta-numbers needed to treat with 95% confidence intervals (CI).ResultsTwelve cardiovascular outcome trials (three for GLP-1 receptor agonists, nine for SGLT2 inhibitors) comprising 90,865 participants were included. Eight trials were conducted in primary type 2 diabetes populations, two in a primary heart failure and two in a primary chronic kidney disease population. Mean estimated glomerular filtration rate at baseline ranged between 37.3 and 85.3 ml/min/1.73 m2. Meta-analyses estimated meta-numbers needed to treat of 85 (95% CI 60; 145) for GLP-1 receptor agonists and 104 (95% CI 81; 147) for SGLT2 inhibitors for the composite renal outcome at the overall median follow-up time of 36 months.ConclusionThe present meta-analysis of digitalized individual patient data revealed moderate and similar absolute treatment benefits of GLP-1 receptor agonists and SGLT2 inhibitors compared to placebo for a composite renal outcome.Graphical

A New Regression Model for Transmuted Weibull with Application

In this research paper, a new regression model for the transmuted Weibull distribution was proposed, and the parameters of the new model were estimated by using the Maximum likelihood method, the different types of residuals were calculated for the new model, such as Cox and Snell Residual, Pearson Residuals, Deviance component Residual, and the martingale residual. The proposed model was applied to a set of real data representing bad debit rate, and the proposed model was also compared with some other models such as the Kumaraswamy Lindley regression model and the LG Weibull regression model.

Bayesian estimation for median discrete Weibull regression model

&lt;abstract&gt;&lt;p&gt;The discrete Weibull model can be adapted to capture different levels of dispersion in the count data. This paper takes into account the direct relationship between explanatory variables and the median of discrete Weibull response variable. Additionally, it provides the Bayesian estimate of the discrete Weibull regression model using the random walk Metropolis algorithm. The prior distributions of the coefficient predictors were carried out based on the uniform non-informative, normal and Laplace distributions. The performance of the Bayes estimators was also compared with the maximum likelihood estimator in terms of the mean square error and the coverage probability through the Monte Carlo simulation study. Meanwhile, a real data set was analyzed to show how the proposed model and the methods work in practice.&lt;/p&gt;&lt;/abstract&gt;

Incidence and Predictors of Major Adverse Drug Reactions Among Human Immunodeficiency Virus–infected Children on Antiretroviral Treatment in West Amhara Comprehensive Specialized Hospitals, Northwest Ethiopia: A Multicenter Retrospective Follow-up Study

PurposeMajor adverse drug reactions (ADRs) are the leading causes of poor adherence, switching of drugs, morbidity, and mortality. A limited studies was conducted to investigate major ADR in developing countries including Ethiopia, and the purpose of this study was to assess the incidence and predictors of major ADRs among HIV-infected children receiving antiretroviral therapy (ART) in West Amhara Comprehensive Specialized Hospitals, Northwest Ethiopia. MethodsAn institutional-based retrospective follow-up study was conducted among 460 children receiving ART from January 1, 2014 to December 31, 2021. A simple random sampling technique was employed, and data were collected using Kobo Toolbox software and then deployed to STATA 14 for analysis. The Kaplan-Meier survival curve and the log-rank test were used to estimate and compare survival times. Both bivariable and multivariable Weibull regression models were fitted to identify predictors. Finally, an adjusted hazards ratio (AHR) with a 95% CI was computed, and variables with P < 0.05 were considered statistically significant predictors of major ADR. FindingsThe overall incidence rate of major ADRs was 5.8 (95% CI, 4.6–7.3) per 1000 child months. Being female (AHR, 2.71; 95% CI, 1.52–4.84), tuberculosis (TB)–HIV co-infection (AHR, 2.49; 95% CI, 1.32–4.68), World Health Organization stage (III and IV) (AHR, 2.52; 95% CI, 1.39–4.56), zidovudine-based (AHR, 2.84; 95% CI, 1.11–7.31), and stavudine-based (AHR, 5.96; 95% CI, 1.63–21.84) regimens were found to be significant predictors of major ADRs. ImplicationsThe major ADR incidence rate was high. Health professionals should employ early screening and close follow-up for children with advanced World Health Organization clinical staging, females, those with TB-HIV co-infection, and those receiving stavudine- and zidovudine-based initial regimens to reduce the incidence of major ADRs.