Journal of Biostatistics and Epidemiology

Survival Analysis of Childbirth Using a Mixture Cure Frailty Model

Wed, 06 May 2026 00:00:00 +0430

Introduction: Childbirth plays a crucial role in population growth and maternal health. In recent decades, many nations,
including Iran, have experienced declining birth rates. Since childbirth is a recurrent event in a parent's life, it is useful to
analyze it through the lens of recurrent event analysis. This methodological framework, commonly employed in biomedicine,
allows for a nuanced examination of the relationship between multiple childbirth experiences and the potential for cured
subjects. This study explores childbirth rates in Hamadan province.
Methods: A total of 633 mothers who gave birth to their first child in 2012 at Fatemiyeh Hospital in Hamadan participated
in this retrospective cohort study. Both mixture cure frailty models and simple frailty models were fitted. The analyses were
conducted using the RSTAN package in RStudio version 26.2.4.
Results: In this study, we analyzed the childbearing patterns of couples and found that the majority (60.6%) had two
children. Additionally, we discovered that 49% of mothers and 55.9% of fathers had education levels below a diploma.
The Kaplan-Meier (KM) curves indicated a cure pattern for families with three or more children, revealing that only
10.6% of individuals had three children, and a mere 0.8% had four. Furthermore, results from a mixture cure frailty model
demonstrated that maternal education plays a crucial role in influencing childbirth probabilities.
Conclusion: Based on the findings of this study, we recommend utilizing mixture cure frailty models rather than simple
frailty models when the dataset contains individuals who are cured.

Algorithm-Level Data-Guided Correction for Class Imbalance in Biological Machine Learning Predictions: Protein Interactions as a Case

Ebrahim Barzegari; Parviz Abdolmaleki (Co-Corresponding Author) — Wed, 06 May 2026 00:00:00 +0430

Introduction: In real-world biomedical applications of data mining, machine learning and artificial intelligence, there are
situations where the widespread problem of class imbalance cannot be addressed by data-level methods such as over- or
under-sampling. Correct and efficient use of algorithm-level methods, on the other hand, needs paying heed to data structure
and content. This study aims to devise and examine simple methods for addressing the imbalanced class distribution issue
in predicting the protein-protein interaction (PPI) sites in membrane proteins as a biomedical case experiment.
Methods: Using an adopted dataset of membrane protein complexes and a retrieved validation set, a class-weighted
random forests (CWRF) classifier model was built for predicting interfacial residues from positional frequencies and an
evolutionary index.
Results: Among several class weighting methods, a data imbalance-emulating weighting method for the CWRF model
achieved an area under the receiver operating characteristics curve (AUC) of 0.815 (95% CI: 0.805-0.823) in the independent
test prediction and 0.802 (95% CI: 0.794-0.809) in the prediction for the external validation set, which outperformed
previous similar studies. A case prediction confirmed the practical utility of this method.
Conclusion: The proposed approach implies potential applications in other fields of biomedicine and beyond. It also
highlights the role of algorithm-data interplay in addressing the class imbalance

Determinants of MedicationAdherence in Hypertensive Patients: Clinical Evidence from Indonesian Primary Healthcare Settings

Wed, 06 May 2026 00:00:00 +0430

Introduction: Adherence to hypertension medication remains a critical challenge in healthcare management, particularly
in resource-limited settings. This study investigated the determinants of medication adherence among patients with
hypertension in Indonesian primary healthcare settings.
Methods: A cross-sectional study involving 96 hypertensive patients selected through systematic random sampling was
conducted at the Public Health Center of Tenggilis, Surabaya. Data were collected via validated questionnaires, including the
Morisky Medication Adherence Scale-8 (MMAS-8), and analyzed via multivariate logistic regression.
Results: Among the 96 hypertensive patients included in this study, the majority were aged 40–49 years (30.2%), with a
male predominance (67.7%). Most participants had a senior high school education (57.3%) and were employed as civil
servants (30.2%). Only 52.1% of patients reported consistent medication adherence, with financial barriers and knowledge
gaps identified as the primary challenges. Multivariate logistic regression analysis revealed that regular medical control
(odds ratio [OR] = 1.963, 95% CI 1.214-3.181; p = 0.006) and alternative diagnostic methods (OR = 2.326, 95% CI 1.532-
3.538, p<0.001) were significantly associated with better medication adherence. Adherence to doctors' advice (OR = 1.699,
95% CI 1.128–2.559, p = 0.012), the ability to manage medication costs (OR = 1.518, 95% CI 1.012–2.278, p = 0.044), and
routine treatment management (OR = 1.825, 95% CI 1.219–2.736, p = 0.004) were identified as key predictors of positive
medication adherence.
Conclusion: Medication adherence in patients with hypertension is influenced by multiple factors, including diagnostic
approach, healthcare access, cost management, and routine treatment compliance. These findings emphasize the need for
comprehensive interventions that address both clinical and socioeconomic barriers to improve hypertension management
in primary healthcare settings.

Estimation of Volume Under Receiver Operating Characteristic Surface and Asymptotic Variance for Diagnostic Classifier Following Log-Normal Distribution

Sahana T S, Kumarapandiyan G — Wed, 06 May 2026 00:00:00 +0430

Introduction: Clinical diagnosis highlights the essential need to assess biomarker performance for effective disease
screening and diagnosis. The Receiver Operating Characteristic (ROC) curve serves as a fundamental tool for assessing
and interpreting biomarker effectiveness. Numerous models and techniques have been developed to analyze biomarkers
in binary classification settings (Non-Diseased vs. Diseased). This research article seeks to expand the binary classification
framework to a three-class scenario, incorporating Diseased, Suspicious, and Non-Diseased categories under a Log-Normal
distribution.
Methods: It introduces a three-class Log-Normal ROC model based on a Parametric approach, deriving metrics such as
Volume Under the ROC Surface (VUS) and Asymptotic Variance, as well as an alternative Non-Parametric approach. The
model was validated using simulated data generated for the underlying distribution, and a real-life dataset was used to fit
the VUS and ROC curves.
Results: The simulation study was conducted using four sets with varying parameters. In the fourth set, the Non-Parametric
VUS (0.9966) exceeded the Parametric VUS (0.8058), though the difference was smaller compared to the other sets. The
low Standard Error (SE) (0.0472) across all sets indicates high precision in the estimates. Additionally, for the real-life (The
multiple sclerosis (ms) disease) dataset the VUS value is 0.6782 which gives moderate fit of the model.
Conclusion: In this study, we derived the asymptotic variance and VUS for the Log-Normal distribution using simulated
data with varying parameters. The analysis compares diagnostic performance across parameter sets, highlighting the
superiority of Non-Parametric VUS over Parametric VUS. Set 4 demonstrated the highest reliability with the lowest standard
error (SE = 0.0472). The real-life MS dataset provided a moderate fit to the proposed model.

Prevalence of Obesity, Overweight and other Cardiovascular Risk Factors Among Iranian Military Personnel in 2022

Wed, 06 May 2026 00:00:00 +0430

Introduction: The incidence and prevalence of cardiovascular disease (CVD) have increased in Iran, considering the
importance of documenting and generating information about the risk of CVD in the military community, the current study
aimed to measure the prevalence of risk CVD factors as well as predict the 10-year risk of CVD among the Iranian military
personnel. The FRS items include age, gender, total cholesterol, high density lipoprotein cholesterol (HDL-C), systolic blood
pressure, status of diabetes and smoking.
Methods: This cross-sectional study was conducted on 1025 male military personnel in 2022. For comparative analysis,
ANOVA or t-test, as well as the Chi-square (or Fisher's exact) test was used. All statistical analyses were conducted using
SPSS 22 software. The statistical significance level was set at 0.05.
Results: The prevalence of hypertension was 2.3 % and increased with age. The prevalence of overweight and obesity
increased with age and was 54.7% as well as 14.1%, respectively, in those 40- 45 years of age. Diabetes affected 6.2%
of the oldest group and 8.2% of participants aged 40–45 years. TC was increased in one-third of understudied cases.
The percentage of abnormal LDL-C was 57.5%. These results were accompanied by increased TG in 34.6%, low HDL-C in
36.4%, and FPG >100 mg/dl in 13.2% of subjects. Out of a total of 608 participants over 30 years of old, a low FRS (<10%)
was calculated for 571 (93.9%), the others, were classified as moderate (5.6%) and high (0.5%) risk. The prevalence of
hypertension among the high and moderate FRS risk group was higher than low-risk group older persons have a higher 10-
year CVD risk level (p < 0.001). The level of, blood pressure, FPG, LDL, TC, and TG, in the high-risk group, was significantly
higher than in the two other groups (P=0.001).
Conclusion: Although a high proportion of military personnel had a low risk of CVD in the next 10 years, the high prevalence
of overweight and other risk factors such as LDL level needs special attention.