Introduction: The prevalence of end-stage renal disease has raised the need for renal replacement therapy over recent decades. Even though a kidney transplant offers an improved quality of life and lower cost of care than dialysis, graft failure is possible after transplantation. Hence, this study aimed to predict the risk of graft failure among post-transplant recipients in Ethiopia using the selected machine learning prediction models. Methodology: The data was extracted from the retrospective cohort of kidney transplant recipients at the Ethiopian National Kidney Transplantation Center from September 2015 to February 2022. In response to the imbalanced nature of the data, we performed hyperparameter tuning, probability threshold moving, tree-based ensemble learning, stacking ensemble learning, and probability calibrations to improve the prediction results. Merit-based selected probabilistic (logistic regression, naive Bayes, and artificial neural network) and tree-based ensemble (random forest, bagged tree, and stochastic gradient boosting) models were applied. Model comparison was performed in terms of discrimination and calibration performance. The best-performing model was then used to predict the risk of graft failure. Results: A total of 278 completed cases were analyzed, with 21 graft failures and 3 events per predictor. Of these, 74.8% are male, and 25.2% are female, with a median age of 37. From the comparison of models at the individual level, the bagged tree and random forest have top and equal discrimination performance (AUC-ROC = 0.84). In contrast, the random forest has the best calibration performance (brier score = 0.045). Under testing the individual model as a meta-learner for stacking ensemble learning, the result of stochastic gradient boosting as a meta-learner has the top discrimination (AUC-ROC = 0.88) and calibration (brier score = 0.048) performance. Regarding feature importance, chronic rejection, blood urea nitrogen, number of post-transplant admissions, phosphorus level, acute rejection, and urological complications are the top predictors of graft failure. Conclusions: Bagging, boosting, and stacking, with probability calibration, are good choices for clinical risk predictions working on imbalanced data. The data-driven probability threshold is more beneficial than the natural threshold of 0.5 to improve the prediction result from imbalanced data. Integrating various techniques in a systematic framework is a smart strategy to improve prediction results from imbalanced data. It is recommended for clinical experts in kidney transplantation to use the final calibrated model as a decision support system to predict the risk of graft failure for individual patients.

Adherence and CD4 cell count change measure the progression of the disease in HIV patients after the commencement of HAART. Lack of information about associated factors on adherence to HAART and CD4 cell count reduction is a challenge for the improvement of cells in HIV positive adults. The main objective of adopting joint modeling was to compare separate and joint models of longitudinal repeated measures in identifying long-term predictors of the two longitudinal outcomes: CD4 cell count and adherence to HAART. A longitudinal retrospective cohort study was conducted to examine the joint predictors of CD4 cell count change and adherence to HAART among HIV adult patients enrolled in the first 10 months of the year 2008 and followed-up to June 2012. Joint model was employed to determine joint predictors of two longitudinal response variables over time. Furthermore, the generalized linear mixed effect model had been used for specification of the marginal distribution, conditional to correlated random effect. A total of 792 adult HIV patients were studied to analyze the longitudinal joint model study. The result from this investigation revealed that age, weight, baseline CD4 cell count, ownership of cell phone, visiting times, marital status, residence area and level of disclosure of the disease to family members had significantly affected both outcomes. From the two-way interactions, time * owner of cell phone, time * sex, age * sex, age * level of education as well as time * level of education were significant for CD4 cell count change in the longitudinal data analysis. The multivariate joint model with linear predictor indicates that CD4 cell count change was positively correlated (p ≤ 0.0001) with adherence to HAART. Hence, as adherence to HAART increased, CD4 cell count also increased; and those patients who had significant CD4 cell count change at each visiting time had been encouraged to be good adherents. Joint model analysis was more parsimonious as compared to separate analysis, as it reduces type I error and subject-specific analysis improved its model fit. The joint model operates multivariate analysis simultaneously; and it has great power in parameter estimation. Developing joint model helps validate the observed correlation between the outcomes that have emerged from the association of intercepts. There should be a special attention and intervention for HIV positive adults, especially for those who had poor adherence and with low CD4 cell count change. The intervention may be important for pre-treatment counseling and awareness creation. The study also identified a group of patients who were with maximum risk of CD4 cell count change. It is suggested that this group of patients needs high intervention for counseling.

CD4 cells are a type of white blood cells that plays a significant role in protecting humans from infectious diseases. Lack of information on associated factors on CD4 cell count reduction is an obstacle for improvement of cells in HIV positive adults. Therefore, the main objective of this study was to investigate baseline factors that could affect initial CD4 cell count change after highly active antiretroviral therapy had been given to adult patients in North West Ethiopia. A retrospective cross-sectional study was conducted among 792 HIV positive adult patients who already started antiretroviral therapy for 1 month of therapy. A Chi square test of association was used to assess of predictor covariates on the variable of interest. Data was secondary source and modeled using generalized linear models, especially Quasi-Poisson regression. The patients’ CD4 cell count changed within a month ranged from 0 to 109 cells/mm3 with a mean of 15.9 cells/mm3 and standard deviation 18.44 cells/mm3. The first month CD4 cell count change was significantly affected by poor adherence to highly active antiretroviral therapy (aRR = 0.506, P value = 2e−16), fair adherence (aRR = 0.592, P value = 0.0120), initial CD4 cell count (aRR = 1.0212, P value = 1.54e−15), low household income (aRR = 0.63, P value = 0.671e−14), middle income (aRR = 0.74, P value = 0.629e−12), patients without cell phone (aRR = 0.67, P value = 0.615e−16), WHO stage 2 (aRR = 0.91, P value = 0.0078), WHO stage 3 (aRR = 0.91, P value = 0.0058), WHO stage 4 (0876, P value = 0.0214), age (aRR = 0.987, P value = 0.000) and weight (aRR = 1.0216, P value = 3.98e−14). Adherence to antiretroviral therapy, initial CD4 cell count, household income, WHO stages, age, weight and owner of cell phone played a major role for the variation of CD4 cell count in our data. Hence, we recommend a close follow-up of patients to adhere the prescribed medication for achievements of CD4 cell count change progression.

Background: Currently, about 165 million children are categorized under malnutrition and 51.5 million suffering from acute malnutrition in world wide. Hence, the objective of current study was to assess the recovery time and its predictors of children under five from severe acute malnutrition admitted to Therapeutic Feeding Unit at Dubti Referral Hospital, Afar region, Eastern Ethiopia. Methods: Institutional based retrospective cohort study was conducted on 650 inpatient children with SAM admitted for therapeutic feeding unit whose treatment was from March to April/2017. Results: The result in current investigation indicates that the average recovery time from SAM was found to be 21 days (95% CI; 21.23–25.77), p-value = 0.035). A Cox proportional hazard regression model revealed that Weight of a child at birth, gestational age of a child, working status of a child at admission birth order of a child, mother’s BMI, mother’s level of education, mother’s stature, mother’s occupation, mother’s age, mother’s marital status, mother’s nutritional status, house hold income in ETB, family size in HH, number of under-five children, the type of toilet used in HH, source of improved drinking water, type of cooking fuel, ownership of livestock, age and weight of a child at admission had statistically significant association with the variation of average recovery time of children from SAM. Conclusion: Male children under severe acute malnutrition, rural children, children with different additional diseases and children who did not get mothers’ breast milk at least in the first six months after birth and children who did not get vaccination are groups at risk and needs intervention and special attention to be recovered with short period of time. Children from low income family, who did not get improved drinking water, without moderate cooking fuel and a child from larger families were groups at risk in recovery time from SAM.

Background: Globally, for individuals infected with HIV, the presence of other infections including TB tends to increase the rate of HIV replication. Of the 8.8 million TB cases worldwide, an estimated 1.1 million (13%) were found to be co-infected with HIV. This research was conducted with the objective to identify potential predictors for the status of TB and CD4 cell count under PLWHIV at Felege Hiwot Specialized Hospital, North-west Ethiopia. Methods: A retrospective repeated measurement was taken from a sample of 226 HIV patients. Separate and joint models were conducted for data analysis of CD4 cell count and TB status of people living with HIV. Results: The descriptive statistics indicated that among the HIV patients receiving HAART, 26.6% had additional TB. AIDS clinical stage, weight, and hemoglobin level had a significant positive association with CD4 cell count, but a negative association with TB status. Weight and CD4 cell count have a negative relationship with the event of HIV/TB co-infection. Hence, the expected number of CD4 cell count of HIV patients who were co-infected with TB was decreased by 2.34 as compared to people living with HIV without TB. As visiting times of patients to hospitals for treatment increased by one unit, the odds of being co-infected with TB was decreased by 0.05, and the expected number of CD4 cell count was increased by 0.2. As patients’ age increased by one year, the expected number of CD4 cell count was decreased by 0.025 cells per/mm³. Conclusion: Having lower CD4 cell count, lower weight, late WHO clinical stage, being non-adherent, having opportunistic infection, having lower hemoglobin, being ambulatory and bedridden were associated with a higher risk of co-infection of HIV/TB and were indicators of progression of the disease.

Non-adherence to Highly Active Antiretroviral Therapy (HAART) is one of the factors for treatment failure in human immunodeficiency virus (HIV) infected patients in developing countries. The main objective of this study was to identify factors for treatment failure among adult HIV patients based on the assessment of first month adherence in the study area. The study was conducted using secondary data from antiretroviral unit at Felege Hiwot Teaching and Specialized Hospital. A prospective study was undertaken on 792 randomly selected adult HIV positive patients who have started HAART. The variable of interest, adherence to HAART was categorized as non-adherence if a patient had taken less than 95% of the prescribed medication and this was measured using pill counts. Descriptive statistics, Chi-square tests of association, independent samples t-test and binary logistic regression were used for data analysis. In first month therapy, 68.2% of the patients belong to adherence group to HAART. As age increases, a patient without cell phone was less likely to be adherent to HAART as compared to patients with cell phone (AOR = 0.661, 95% CI: (0.243, 0.964)). Compared to urban patients, rural patients were less likely to adhere to HAART (AOR = 0.995, 95% CI: (0.403, 0.999)). A patient who did not disclose his/her disease to families or communities had less probability to be adherent to HAART (AOR = 0.325, 95% CI: (0.01, 0.64)). Similarly, a patient who did not get social support (AOR = 0.42, 95% CI: (0,021, 0.473)) had less probability of adherence to HAART. The main reasons for patients to be non-adherent were forgetfulness, side effects, feeling sick and running out of medication. This study indentified certain groups of patients who are at higher risk and who need counseling. Such groups should be targeted and tailored for improvement of adherence to HAART among HIV positive adults. The health care providers should advise the community to provide social support to HIV positive patients whenever their disease is disclosed. On the other hand, patients should disclose their disease to community to get integrated supports. HIV infected patients who are directed to start HAART should adhere the prescribed medication. For the adherence to be effective, patients who have cell phone should use them as reminder to take pills on time.