Bayesian inference for multivariate probit model with latent envelope.

The response envelope model proposed by Cook et al. (2010) is an efficient method to estimate the regression coefficient under the context of the multivariate linear regression model. It improves estimation efficiency by identifying material and immaterial parts of responses and removing the immaterial variation. The response envelope model has been investigated only for continuous response variables. In this paper, we propose the multivariate probit model with latent envelope, in short, the probit envelope model, as a response envelope model for multivariate binary response variables. The probit envelope model takes into account relations between Gaussian latent variables of the multivariate probit model by using the idea of the response envelope model. We address the identifiability of the probit envelope model by employing the essential identifiability concept and suggest a Bayesian method for the parameter estimation. We illustrate the probit envelope model via simulation studies and real-data analysis. The simulation studies show that the probit envelope model has the potential to gain efficiency in estimation compared to the multivariate probit model. The real data analysis shows that the probit envelope model is useful for multi-label classification.

Sex, environment, and death rate in a dementia cohort: a seven-years Bayesian survival analysis using medications data from a contaminated area in Italy.

Introduction: Studies have analyzed the effects of industrial installations on the environment and human health in Taranto, Southern Italy. Literature documented associations between different variables and dementia mortality among both women and men. The present study aims to investigate the associations between sex, environment, age, disease duration, pandemic years, anti-dementia drugs, and death rate. Methods: Data from the regional medication registry were used. All women and men with an anti-dementia medication between 2015 and 2021 were included and followed-up to 2021. Bayesian mixed effects logistic and Cox regression models with time varying exposures were fitted using integrated nested Laplace approximations and adjusting for patients and therapy characteristics. Results: A total of 7,961 person-years were observed. Variables associated with lower prevalence of acetylcholinesterase inhibitors (AChEIs) medication were male sex (OR 0.63, 95% CrI 0.42-0.96), age 70-79 years (OR 0.17, 95% CrI 0.06-0.47) and ≥ 80 years (OR 0.08, 95% CrI 0.03-0.23), disease duration of 2-3 years (OR 0.43, 95% CrI 0.32-0.56) and 4-6 years (OR 0.21, 95% CrI 0.13-0.33), and pandemic years 2020 (OR 0.50, 95% CrI 0.37-0.67) and 2021 (OR 0.47, 95% CrI 0.33-0.65). Variables associated with higher mortality were male sex (HR 2.14, 95% CrI 1.75-2.62), residence in the contaminated site of national interest (SIN) (HR 1.25, 95% CrI 1.02-1.53), age ≥ 80 years (HR 6.06, 95% CrI 1.94-18.95), disease duration of 1 year (HR 1.50, 95% CrI 1.12-2.01), 2-3 years (HR 1.90, 95% CrI 1.45-2.48) and 4-6 years (HR 2.21, 95% CrI 1.60-3.07), and pandemic years 2020 (HR 1.38, 95% CrI 1.06-1.80) and 2021 (HR 1.56, 95% CrI 1.21-2.02). Variables associated with lower mortality were therapy with AChEIs alone (HR 0.69, 95% CrI 0.56-0.86) and in combination with memantine (HR 0.54, 95% CrI 0.37-0.81). Discussion: Male sex, age, disease duration, and pandemic years appeared to be associated with lower AChEIs medications. Male sex, residence in the SIN of Taranto, age, disease duration, and pandemic years seemed to be associated with an increased death rate, while AChEIs medication seemed to be associated with improved survival rate.

Identifying Bayesian optimal experiments for uncertain biochemical pathway models.

Pharmacodynamic (PD) models are mathematical models of cellular reaction networks that include drug mechanisms of action. These models are useful for studying predictive therapeutic outcomes of novel drug therapies in silico. However, PD models are known to possess significant uncertainty with respect to constituent parameter data, leading to uncertainty in the model predictions. Furthermore, experimental data to calibrate these models is often limited or unavailable for novel pathways. In this study, we present a Bayesian optimal experimental design approach for improving PD model prediction accuracy. We then apply our method using simulated experimental data to account for uncertainty in hypothetical laboratory measurements. This leads to a probabilistic prediction of drug performance and a quantitative measure of which prospective laboratory experiment will optimally reduce prediction uncertainty in the PD model. The methods proposed here provide a way forward for uncertainty quantification and guided experimental design for models of novel biological pathways.

Description of pain associated with persistent postoperative pain after total knee arthroplasty.

After total knee arthroplasty (TKA), approximately 20% of patients experience persistent postoperative pain (PPP). Although preoperative and postoperative pain intensity is a relevant factor, more detailed description of pain is needed to determine specific intervention strategies for clinical conditions. This study aimed to clarify the associations between preoperative and postoperative descriptions of pain and PPP. Fifty-two TKA patients were evaluated for pain intensity and description of pain preoperatively and 2 weeks postoperatively, and the intensities were compared. In addition, the relationship between pain intensity and PPP at 3 and 6 months after surgery was analyzed using a Bayesian approach. Descriptions of arthritis ("Throbbing" and "aching") improved from preoperative to 2 weeks postoperative. Several preoperative ("Shooting", "Aching", "Caused by touch", "Numbness") and postoperative ("Cramping pain") descriptors were associated with pain intensity at 3 months postoperatively, but only "cramping pain" at 2 weeks postoperatively was associated with the presence of PPP at 3 and 6 months postoperatively. In conclusion, it is important to carefully listen to the patient's complaints and determine the appropriate intervention strategy for the clinical condition during perioperative pain management.

A bayesian spatio-temporal dynamic analysis of food security in Africa.

Exploring the factors influencing Food Security and Nutrition (FSN) and understanding its dynamics is crucial for planning and management. This understanding plays a pivotal role in supporting Africa's food security efforts to achieve various Sustainable Development Goals (SDGs). Utilizing Principal Component Analysis (PCA) on data from the FAO website, spanning from 2000 to 2019, informative components are derived for dynamic spatio-temporal modeling of Africa's FSN Given the dynamic and evolving nature of the factors impacting FSN, despite numerous efforts to understand and mitigate food insecurity, existing models often fail to capture this dynamic nature. This study employs a Bayesian dynamic spatio-temporal approach to explore the interconnected dynamics of food security and its components in Africa. The results reveal a consistent pattern of elevated FSN levels, showcasing notable stability in the initial and middle-to-late stages, followed by a significant acceleration in the late stage of the study period. The Democratic Republic of Congo and Ethiopia exhibited particularly noteworthy high levels of FSN dynamicity. In particular, child care factors and undernourishment factors showed significant dynamicity on FSN. This insight suggests establishing regional task forces or forums for coordinated responses to FSN challenges based on dynamicity patterns to prevent or mitigate the impact of potential food security crises.

[Bayesian network prediction study on the impact of occupational health psychological factors on insomnia among thermal power generation workers].

Objective: To explore the risk factors of insomnia among employees in the thermal power generation industry and the network relationships between their interactions, and to provide scientific basis for personalized interventions for high-risk groups with insomnia. Methods: In November 2022, 860 employees of a typical thermal power generation enterprise were selected as the research subjects by cluster sampling. On-site occupational health field surveys and questionnaire surveys were used to collect basic information, occupational characteristics, anxiety, depression, stress, occupational stress, and insomnia. The interaction between insomnia and occupational health psychological factors was evaluated by using structural equation model analysis and Bayesian network construction. Results: The detection rates of anxiety, depression and stress were 34.0% (292/860), 32.1% (276/860) and 18.0% (155/860), respectively. The total score of occupational stress was (445.3±49.9) points, and 160 workers (18.6%) were suspected of insomnia, and 578 workers (67.2%) had insomnia. Structural equation model analysis showed that occupational stress had a significant effect on the occurrence of insomnia in thermal power generation workers (standardized load coefficient was 0.644), and occupational health psychology had a low effect on insomnia (standardized load coefficient was 0.065). However, the Bayesian network model further analysis found that anxiety and stress were the two parent nodes of insomnia, with direct causal relationships, the arc strength was-8.607 and -15.665, respectively. The model prediction results showed that the probability of insomnia occurring was predicted to be 0 in the cases of no stress and anxiety, low stress without anxiety, and no stress with low anxiety. When high stress with low anxiety and low stress with high anxiety occurred, the predicted probability of insomnia occurring were 0.38 and 0.47, respectively. When both high stress and high anxiety occurred simultaneously, the predicted probability of insomnia occurring was 0.51. Conclusion: Bayesian network risk assessment can intuitively reveal and predict the insomnia risk of thermal power generation workers and the network interaction relationship between the risks. Anxiety and stress are the direct causal risks of insomnia, and stress is the main risk of individual insomnia of thermal power generation workers. The occurrence of insomnia can be reduced based on scientific intervention of stress conditions.

Perceptual error based on Bayesian cue combination drives implicit motor adaptation.

The sensorimotor system can recalibrate itself without our conscious awareness, a type of procedural learning whose computational mechanism remains undefined. Recent findings on implicit motor adaptation, such as over-learning from small perturbations and fast saturation for increasing perturbation size, challenge existing theories based on sensory errors. We argue that perceptual error, arising from the optimal combination of movement-related cues, is the primary driver of implicit adaptation. Central to our theory is the increasing sensory uncertainty of visual cues with increasing perturbations, which was validated through perceptual psychophysics (Experiment 1). Our theory predicts the learning dynamics of implicit adaptation across a spectrum of perturbation sizes on a trial-by-trial basis (Experiment 2). It explains proprioception changes and their relation to visual perturbation (Experiment 3). By modulating visual uncertainty in perturbation, we induced unique adaptation responses in line with our model predictions (Experiment 4). Overall, our perceptual error framework outperforms existing models based on sensory errors, suggesting that perceptual error in locating one's effector, supported by Bayesian cue integration, underpins the sensorimotor system's implicit adaptation.

Four decades of data indicate that planted mangroves stored up to 75% of the carbon stocks found in intact mature stands.

Mangroves' ability to store carbon (C) has long been recognized, but little is known about whether planted mangroves can store C as efficiently as naturally established (i.e., intact) stands and in which time frame. Through Bayesian logistic models compiled from 40 years of data and built from 684 planted mangrove stands worldwide, we found that biomass C stock culminated at 71 to 73% to that of intact stands ~20 years after planting. Furthermore, prioritizing mixed-species planting including Rhizophora spp. would maximize C accumulation within the biomass compared to monospecific planting. Despite a 25% increase in the first 5 years following planting, no notable change was observed in the soil C stocks thereafter, which remains at a constant value of 75% to that of intact soil C stock, suggesting that planting effectively prevents further C losses due to land use change. These results have strong implications for mangrove restoration planning and serve as a baseline for future C buildup assessments.

Uncovering hidden and complex relations of pandemic dynamics using an AI driven system.

The COVID-19 pandemic continues to challenge healthcare systems globally, necessitating advanced tools for clinical decision support. Amidst the complexity of COVID-19 symptomatology and disease severity prediction, there is a critical need for robust decision support systems to aid healthcare professionals in timely and informed decision-making. In response to this pressing demand, we introduce BayesCovid, a novel decision support system integrating Bayesian network models and deep learning techniques. BayesCovid automates data preprocessing and leverages advanced computational methods to unravel intricate patterns in COVID-19 symptom dynamics. By combining Bayesian networks and Bayesian deep learning models, BayesCovid offers a comprehensive solution for uncovering hidden relationships between symptoms and predicting disease severity. Experimental validation demonstrates BayesCovid 's high prediction accuracy (83.52-98.97%). Our work represents a significant stride in addressing the urgent need for clinical decision support systems tailored to the complexities of managing COVID-19 cases. By providing healthcare professionals with actionable insights derived from sophisticated computational analysis, BayesCovid aims to enhance clinical decision-making, optimise resource allocation, and improve patient outcomes in the ongoing battle against the COVID-19 pandemic.

Jumping to attributions during social evaluation.

Social learning is crucial for human relationships and well-being. Self- and other- evaluations are universal experiences, playing key roles in many psychiatric disorders, particularly anxiety and depression. We aimed to deepen our understanding of the computational mechanisms behind social learning, which have been implicated in internalizing conditions like anxiety and depression. We built on prior work based on the Social Evaluation Learning Task (SELT) and introduced a new computational model to better explain rapid initial inferences and progressive refinement during serial social evaluations. The Social Evaluation Learning Task-Revised (SELT-R) was improved by stakeholder input, making it more engaging and suitable for adolescents. A sample of 130 adults from the UK completed the SELT-R and questionnaires assessing symptoms of depression and anxiety. 'Classify-refine' computational models were compared with previously successful Bayesian models. The 'classify-refine' models performed better, providing insight into how people infer the attributes and motives of others. Parameters of the best fitting model from the SELT-R were correlated with Anxiety factor scores, with higher symptoms associated with greater decision noise and higher (less flexible) policy certainty. Our results replicate findings regarding the classify-refine process and set the stage for future investigations into the cognitive mechanisms of self and other evaluations in internalizing disorders.

The effect of environmental variations on the production of the principal agricultural products in Colombia.

The agricultural sector of Colombia supports the national economy and food security due to the rich lands for cultivation. Although Colombia has a vast hydrological basin, climate change can impact agricultural productivity, generating economic and social adverse effects. For this, we evaluated the impact of some environmental variables on the production of the most sold crops using production, climatic, and hydrological data of the 1121 municipalities from 2007 to 2020. We modeled the production of coffee, rice, palm oil, sugarcane, and corn, adopting a Bayesian spatio-temporal model that involved a set of environmental variables: average temperature, minimum temperature, maximum temperature, evapotranspiration, precipitation, runoff, soil moisture, vapor pressure, radiation, and wind speed. We found that increases in the average temperatures can affect coffee (-0.2% per °C), rice (-3.76% per °C), and sugarcane (-0.19% per °C) production, meanwhile, these increases can boost palm oil (+2.55% per °C) and corn (+1.28% per °C) production in Colombia. This statement implies that the agricultural sector needs to substitute land use, promoting the production of palm oil and corn. Although our results did not find a significant effect of hydrological variables in any crop, suggesting that the abundance of water in Colombia might balance the impact of these variables. The increases in vapor pressure impact all the crops negatively (between -11.2% to -0.43% per kPa), except rice, evidencing that dry air conditions affect agricultural production. Colombia must manage the production location of the traditional products and implement agro-industrial technologies to avoid the climate change impact on crops.

Keel bone fractures affect laying hens' mobility, but no evidence for reciprocal effects.

Keel bone fractures (KBF) are prevalent in commercial laying hens and are considered one of the greatest welfare concerns in the egg-production industry. While clear associations exist between KBF and animal mobility, suggesting that KBF impair mobility, the effect of mobility on KBF remains unclear. We combined data from three studies that assessed keel bone fracture severity through radiographs and monitored hens' transitions between different zones of a multi-tier aviary system (the three tiers, a littered floor, and a winter garden) the week prior to radiograph. For each hen, we extracted two daily movement behaviours: the vertical distance travelled and the mean number of zones crossed within one transition; and two daily space-use behaviours: the time spent in the top tier and the unevenness of time spent across zones. We used hierarchical Bayesian continuous time dynamic modelling to estimate how a change in a behaviour predicted a later change in keel bone fracture severity, and vice versa. Increased fracture severity did not predict later changes in space-use behaviours, but it did predict changes in movement behaviours. Specifically, increased fracture severity led to decreased vertical travelled distance and a tendency to cross more zones within one transition, suggesting impaired mobility in hens with increased fracture severity. In contrast, we found no evidence that movement or space-use behaviours predict later change in fracture severity, challenging previous literature suggesting that vertical locomotion through jumping and flying may exacerbate keel bone fractures in complex three-dimensional systems due to increased risk of collisions. However, similar efforts accounting for the location of fractures on the keel could unveil the potential influence of movement and space-use behaviours in the formation and change (healing or worsening) of KBF and increase our ability to mitigate their effects.

TORONTO: A trial-oriented multidimensional psychometric testing algorithm.

Bayesian adaptive methods for sensory threshold determination were conceived originally to track a single threshold. When applied to the testing of vision, they do not exploit the spatial patterns that underlie thresholds at different locations in the visual field. Exploiting these patterns has been recognized as key to further improving visual field test efficiency. We present a new approach (TORONTO) that outperforms other existing methods in terms of speed and accuracy. TORONTO generalizes the QUEST/ZEST algorithm to estimate simultaneously multiple thresholds. After each trial, without waiting for a fully determined threshold, the trial-oriented approach updates not only the location currently tested but also all other locations based on patterns in a reference data set. Since the availability of reference data can be limited, techniques are developed to overcome this limitation. TORONTO was evaluated using computer-simulated visual field tests: In the reliable condition (false positive [FP] = false negative [FN] = 3%), the median termination and root mean square error (RMSE) of TORONTO was 153 trials and 2.0 dB, twice as fast with equal accuracy as ZEST. In the FP = FN = 15% condition, TORONTO terminated in 151 trials and was 2.2 times faster than ZEST with better RMSE (2.6 vs. 3.7 dB). In the FP = FN = 30% condition, TORONTO achieved 4.2 dB RMSE in 148 trials, while all other techniques had > 6.5 dB RMSE and terminated much slower. In conclusion, TORONTO is a fast and accurate algorithm for determining multiple thresholds under a wide range of reliability and subject conditions.

Extracorporeal cardiopulmonary resuscitation versus standard treatment for refractory out-of-hospital cardiac arrest: a Bayesian meta-analysis.

BACKGROUND: The outcomes of several randomized trials on extracorporeal cardiopulmonary resuscitation (ECPR) in patients with refractory out-of-hospital cardiac arrest were examined using frequentist methods, resulting in a dichotomous interpretation of results based on p-values rather than in the probability of clinically relevant treatment effects. To determine such a probability of a clinically relevant ECPR-based treatment effect on neurological outcomes, the authors of these trials performed a Bayesian meta-analysis of the totality of randomized ECPR evidence. METHODS: A systematic search was applied to three electronic databases. Randomized trials that compared ECPR-based treatment with conventional CPR for refractory out-of-hospital cardiac arrest were included. The study was preregistered in INPLASY (INPLASY2023120060). The primary Bayesian hierarchical meta-analysis estimated the difference in 6-month neurologically favorable survival in patients with all rhythms, and a secondary analysis assessed this difference in patients with shockable rhythms (Bayesian hierarchical random-effects model). Primary Bayesian analyses were performed under vague priors. Outcomes were formulated as estimated median relative risks, mean absolute risk differences, and numbers needed to treat with corresponding 95% credible intervals (CrIs). The posterior probabilities of various clinically relevant absolute risk difference thresholds were estimated. RESULTS: Three randomized trials were included in the analysis (ECPR, n = 209 patients; conventional CPR, n = 211 patients). The estimated median relative risk of ECPR for 6-month neurologically favorable survival was 1.47 (95%CrI 0.73-3.32) with a mean absolute risk difference of 8.7% (- 5.0; 42.7%) in patients with all rhythms, and the median relative risk was 1.54 (95%CrI 0.79-3.71) with a mean absolute risk difference of 10.8% (95%CrI - 4.2; 73.9%) in patients with shockable rhythms. The posterior probabilities of an absolute risk difference > 0% and > 5% were 91.0% and 71.1% in patients with all rhythms and 92.4% and 75.8% in patients with shockable rhythms, respectively. CONCLUSION: The current Bayesian meta-analysis found a 71.1% and 75.8% posterior probability of a clinically relevant ECPR-based treatment effect on 6-month neurologically favorable survival in patients with all rhythms and shockable rhythms. These results must be interpreted within the context of the reported credible intervals and varying designs of the randomized trials. REGISTRATION: INPLASY (INPLASY2023120060, December 14th, 2023, https://doi.org/10.37766/inplasy2023.12.0060 ).

A Hierarchical Bayesian Model of Adaptive Teaching.

How do teachers learn about what learners already know? How do learners aid teachers by providing them with information about their background knowledge and what they find confusing? We formalize this collaborative reasoning process using a hierarchical Bayesian model of pedagogy. We then evaluate this model in two online behavioral experiments (N = 312 adults). In Experiment 1, we show that teachers select examples that account for learners' background knowledge, and adjust their examples based on learners' feedback. In Experiment 2, we show that learners strategically provide more feedback when teachers' examples deviate from their background knowledge. These findings provide a foundation for extending computational accounts of pedagogy to richer interactive settings.

Global, regional, and national burdens of atopic dermatitis under 14: a trend analysis and future prediction based on the global burden of disease study 2019.

OBJECTIVE: The aim is to evaluate the global, regional, and national trends in the burden of children and adolescents under 14 from 1990 to 2019, as well as future trend predictions. METHODS: In Global Burden of Disease (GBD), we reported the incidence, prevalence rate and the years lived with disability (YLDs), the incidence per 100,000 people, and the average annual percentage change (AAPC). We further analyzed these global trends by age, gender, and social development index (SDI). We use joinpoint regression analysis to determine the year with the largest global trend change. Bayesian age-period-cohort (BAPC) was used for predictions. RESULTS: From 1990 to 2019, the incidence rate, prevalence and YLDs of AD under 14 years old showed a downward trend. The incidence rate of AD among people under 5 years old has the largest decline [AAPC: -0.13 (95% CI: -0.15 to -0.11), P < 0.001]. The incidence rate, prevalence and YLDs of AD in women were higher than those in men regardless of age group. Regional, Asia has the highest AD incidence rate in 2019. National, Mongolia has the highest AD incidence rate in 2019. The largest drop in AD incidence rate, prevalence and YLDs between 1990 and 2019 was in the United States. CONCLUSION: From 1990 to 2019, the global incidence rate of children and adolescents under 14 declined. With the emergence of therapeutic drugs, the prevalence and YLDs rate declined significantly. From 2020 to 2030, there is still a downward trend.

Exploring the structure of household social capital in rural Vietnam: Applying Bayesian network approach.

This study aims to explore the structure of the households' social capital of rural Vietnam households with secondary data from 2008 to 2018. This paper applied the fundamental theories (resource and network theories) and the Bayesian network to estimate the interaction of proxies to explore the structure of social capital. Results showed that the triangle structure in household social capital with the core point is organization participation. The connections show the tendency from organization participation, linking to household networks. Alongside that, linking social capital and Organization participation are determinants of social capital indicators (social events, social cost). Therefore, this paper suggests employing proxies such as structured indicators for integrating social capital into the livelihood papers.

Evaluation of genomic prediction considering non-additive genetic effects on fatty acid traits of Japanese Black cattle.

Genomic prediction was conducted using 2494 Japanese Black cattle from Hiroshima Prefecture and both single-nucleotide polymorphism information and phenotype data on monounsaturated fatty acid (MUFA) and oleic acid (C18:1) analyzed with gas chromatography. We compared the prediction accuracy for four models (A, additive genetic effects; AD, as for A with dominance genetic effects; ADR, as for AD with the runs of homozygosity (ROH) effects calculated by ROH-based relationship matrix; and ADF, as for AD with the ROH-based inbreeding coefficient of the linear regression). Bayesian methods were used to estimate variance components. The narrow-sense heritability estimates for MUFA and C18:1 were 0.52-0.53 and 0.57, respectively; the corresponding proportions of dominance genetic variance were 0.04-0.07 and 0.04-0.05, and the proportion of ROH variance was 0.02. The deviance information criterion values showed slight differences among the models, and the models provided similar prediction accuracy.

Unravelling factors influencing demand for modern contraception and evaluating coverage progress since 2015 in Ethiopia, Kenya, and Nigeria: insights from multilevel and geostatistical modelling.

INTRODUCTION: The United Nations established the Sustainable Development Goals (SDGs) in 2015 to enhance global development. In this study, we examine an SDG indicator: the percentage of women aged 15-49 whose family planning needs are met by modern contraception (mDFPS). We evaluate both the factors influencing its coverage and its progress since 2015. METHODS: We used nationally representative surveys data (Demographic and Health Surveys (DHS) and Performance Monitoring for Action (PMA)) from Ethiopia, Kenya, and Nigeria. We assessed predictors of mDFPS. We also computed mDFPS coverage across countries and subnational areas, assessing coverage changes from the SDGs onset to the most recent period, using a Bayesian model-based geostatistical approach. We assessed whether the subnational areas exceeded the minimum recommended WHO mDFPS coverage of 75%. RESULTS: Varied individual and community-level determinants emerged, highlighting the countries' uniqueness. Factors such as being part of a female-headed household, and low household wealth, lowered the odds of mDFPS, while rural-residence had low odds only in Ethiopia and Nigeria. The results indicate mDFPS stagnation in most administrative areas across the three countries. Geographic disparities persisted over time, favouring affluent regions. The predicted posterior proportion of mDFPS and exceedance probability (EP) for WHO target for Ethiopia was 39.85% (95% CI: [4.51, 83.01], EP = 0.08) in 2016 and 46.28% (95% CI: [7.15, 85.99], EP = 0.13) in 2019. In Kenya, the adjusted predicted proportion for 2014 was 30.19% (95% CI: [2.59, 80.24], EP = 0.06) and 44.16% (95%CI: [9.35, 80.24], EP = 0.13) in 2022. In Nigeria, the predicted posterior proportion of mDFPS was 17.91% (95% CI: [1.24, 61.29], EP = 0.00) in 2013, and it was 23.08% (95% CI: [1.80, 56.24], EP = 0.00) in 2018. None of the sub-national areas in Ethiopia and Nigeria exceeded the WHO target. While 9 out of 47 counties in Kenya in 2022 exceeded the WHO mDFPS target. CONCLUSION: The study unveils demographic, geographic, and socioeconomic mDFPS disparities, signalling progress and stagnation across administrative areas. The findings offer policymakers and governments insights into targeting interventions for enhanced mDFPS coverage. Context-specific strategies can address local needs, aiding SDG attainment.