Bayesian Hierarchical Logistic Regression Research Articles

Use of a Liaison-Mediated Referral Strategy and Participation in Cardiac Rehabilitation After Percutaneous Coronary Intervention.

Cardiac rehabilitation (CR) improves outcomes following percutaneous coronary intervention (PCI) but remains underutilized. A liaison-mediated referral (LMR), where a health care professional explains CR's benefits, addresses barriers to participation, and places a referral before discharge, may promote CR use. Our objective was to assess the impact of an LMR on CR participation after PCI. This was a retrospective study of patients who underwent PCI across 48 hospitals in Michigan between January 2021 and April 2022 and referred to CR before discharge. Clinical registry data were linked to administrative claims to identify the primary outcome, CR participation, defined as ≥1 CR session within 90 days of discharge. Bayesian hierarchical logistic regression was used to compare CR participation between patients with and without an LMR. For the secondary outcome, frailty proportional hazard modeling compared days elapsed between discharge and first CR session between liaison cohorts. Among 9023 patients referred to CR after PCI, 4323 (47.9%) underwent an LMR (mean age, 69.3 [SD=11] years; 68.3% male) and 3390 (36.7%) attended ≥1 CR session within 90 days of discharge. The LMR cohort had a higher unadjusted CR participation rate (43.1% [95% CI, 41.5%-44.6%] versus 32.4% [95% CI, 31.1%-33.8%]; P<0.001), a higher adjusted odds ratio of attending ≥1 CR session (adjusted odds ratio, 1.21; 95% credible interval, 1.07-1.38), and a shorter delay in attending the first CR session compared with the non-LMR cohort (28 [interquartile range, 19-42] versus 33 [interquartile range, 21-47] days; P<0.001). An LMR was associated with higher odds of CR participation and may mitigate delays in CR enrollment. This referral strategy may improve CR participation and patient outcomes after PCI.

Trends and predictors of unfilled emergency medicine residency positions: A comparative analysis of the 2023 and 2024 Match cycles.

The emergency medicine (EM) landscape has evolved due to the increasing number of programs paired with fewer applicants. This study analyzed the characteristics of EM residency programs associated with unfilled positions during the 2024 Match and compared them with data from the 2023 Match to identify persistent and emerging trends influencing these outcomes. In this cross-sectional, observational study, we investigated factors associated with unfilled EM residency positions in the 2024 Match. We used publicly accessible data from the National Resident Matching Program. To identify program-level predictors of unfilled positions, we constructed a Bayesian hierarchical logistic regression model, incorporating data from the 2023 Match season. In 2024, 54 out of 281 (19.2%) residency programs remained unfilled. Our Bayesian analysis reaffirmed that smaller program size, geographical location, prior osteopathic accreditation, and corporate ownership continue to be significant factors. Programs with vacancies in the previous year were also more likely to remain unfilled. Thus, several factors identified in 2023 remained associated with this year's Match outcomes, with the impact of previous unfilled positions being particularly pronounced. This study identified several factors associated with a greater likelihood of having unfilled EM residency positions, with previous unfilled positions emerging as the most significant predictor. These findings offer critical insights for residency programs and governing bodies, providing a basis for enhancing recruitment strategies, addressing the cyclical nature of unfilled positions, and tackling workforce challenges in EM.

Exploring Socioeconomic Disparities in Surgical Site Infections

Introduction: Social disparities have been shown to impact a wide variety of healthcare outcomes. Surgical site infections (SSIs) are associated with substantial patient morbidity, but studies on the intersection of social disparity and SSI are limited. We sought to evaluate the association between SSI and the Center for Disease Control and Prevention’s social vulnerability index (SVI). Methods: Patients with National Health Safety Network (NHSN) procedure codes for colon, abdominal hysterectomy, hip prosthesis, knee prosthesis and spinal fusion surgeries were retrieved from the electronic medical records of 20 hospitals across 4 geographic markets. SSIs were identified by trained infection preventionists using NHSN definitions. Descriptive statistics were used for baseline demographic and clinical characteristics. Univariate logistic regression was performed to assess the association of demographic, clinical, and procedural factors with the outcome of SSI. Further univariate subgroup analysis was completed by procedure. To account for the nonlinear relationship between the social vulnerability index and SSIs, smoothing splines were used in a Bayesian hierarchical logistic regression model, with random effects to account for the different market practices. Nonlinear effects of procedure duration were also investigated while adjusting for the patient age, procedure type, and health system market. Results: 23,864 surgical procedures among 22,319 unique patients identified between 1 August 2022 and 31 August 2023. 96 patients with infection present at time of surgery were excluded. The study population was mostly white (74%) and female (65%) (Table 1). Less than 13% of the patients had more than one procedure during this time. In a univariate analysis, we found evidence of market and procedure effects, with colon surgery being associated with the highest odds of SSI. Procedure duration was significantly associated with SSI in both univariate and multivariable models, with a drastic increase in the odds of SSI for procedures &gt; 150 mins. In the multivariable model we found that SVIs lower than 0.4 (95% CI 0.28 to 0.55) are associated with an adjusted odds ratio (aOR) &lt; 1. (Figure 1) Conclusions: Our study shows that the relationship of social vulnerability and adverse outcomes is highly complex with nonlinear dynamics at play. After adjusting for procedure type, duration, patient age, gender and health system market the odds of SSIs increase sharply in patients with higher SVI until leveling off at an elevated risk.

State and Trait Disgust Uniquely Predict Avoidance in Contamination Fear: Specificity of Disease-Specific and Nonspecific Individual Differences

Although behavioral avoidance is observed among those with heightened contamination concerns, the extent to which such avoidance is best predicted by state and/or trait characteristics is unclear. Furthermore, while disgust proneness is a disease-specific trait that has been shown to predict avoidance among those with symptoms of contamination-based obsessive-compulsive disorder (OCD), it is unclear if other disease-specific traits may also serve a similar function. In the present study, contamination-fearful participants (N = 89) first completed self-report measures of disease-specific (disgust proneness, health anxiety, perceived vulnerability to disease) and disease-nonspecific (intolerance of uncertainty, trait anxiety) traits. They then completed a 16-step behavioral approach task (BAT) with increasing likelihood of contagion in a public restroom and provided ratings of state disgust and state anxiety at each step. Bayesian hierarchical logistic regression showed that state disgust, but not state anxiety, at a given step was associated with avoidance on the next step. Furthermore, disgust proneness was the only trait associated with avoidance on the BAT. A significant interaction between disgust proneness and BAT step completed in the public restroom was also found. Examination of this interaction suggests that intermediary, but not initial and latter, steps of the BAT differentiate those high and low in disgust proneness. These findings suggest that state and trait disgust uniquely drive behavioral avoidance among those with contamination concerns commonly observed in OCD.

Domestic (re)infestation risk with the main vector Triatoma infestans increases with surrounding green vegetation and social vulnerability in the Argentine Chaco.

Chagas disease, caused by Trypanosoma cruzi, is still a public health problem in Latin America and in the Southern Cone countries, where Triatoma infestans is the main vector. We evaluated the relationships among the density of green vegetation around rural houses, sociodemographic characteristics, and domestic (re)infestation with T. infestans while accounting for their spatial dependence in the municipality of Pampa del Indio between 2007 and 2016. The study comprised sociodemographic and ecological variables from 734 rural houses with no missing data. Green vegetation density surrounding houses was estimated by the normalized difference vegetation index (NDVI). We used a hierarchical Bayesian logistic regression composed of fixed effects and spatial random effects to estimate domestic infestation risk and quantile regressions to evaluate the association between surrounding NDVI and selected sociodemographic variables. Qom ethnicity and the number of poultry were negatively associated with surrounding NDVI, whereas overcrowding was positively associated with surrounding NDVI. Hierarchical Bayesian models identified that domestic infestation was positively associated with surrounding NDVI, suitable walls for triatomines, and overcrowding over both intervention periods. Preintervention domestic infestation also was positively associated with Qom ethnicity. Models with spatial random effects performed better than models without spatial effects. The former identified geographic areas with a domestic infestation risk not accounted for by fixed-effect variables. Domestic infestation with T. infestans was associated with the density of green vegetation surrounding rural houses and social vulnerability over a decade of sustained vector control interventions. High density of green vegetation surrounding rural houses was associated with households with more vulnerable social conditions. Evaluation of domestic infestation risk should simultaneously consider social, landscape and spatial effects to control for their mutual dependency. Hierarchical Bayesian models provided a proficient methodology to identify areas for targeted triatomine and disease surveillance and control.

Disparities in clinical trial enrollment at a Canadian comprehensive cancer center: A 15-year retrospective study.

Disparities in clinical trials (CTs) enrollment perpetuate inequities in treatment access and outcomes, but there is a paucity of Canadian data. The objective of this study was to examine disparities in cancer CT enrollment at a large Canadian comprehensive cancer center. Retrospective study of CT enrollment among new patient consultations from 2006 to 2019, with follow-up to 2021 (N=154,880), with the primary outcome of enrollment as a binary variable. Factors associated with CT enrollment were evaluated using multivariable Bayesian hierarchical logistic regression with random effects for most responsible physician (MRP) and geography, adjusted for patient characteristics (sex, age, language, geography, and primary care provider [PCP]), area-level marginalization (residential instability, material deprivation, dependency, and ethnic concentration), disease (cancer site and stage), and MRP (department, sex, language, and training). A sensitivity analysis of the cumulative incidence of enrollment was conducted to account for differences in disease type and follow-up length. CT enrollment was 11.2% overall, with a 15-year cumulative incidence of 18%. Lower odds of enrollment were observed in patients who were female (adjusted odds ratio [AOR], 0.82; 95% confidence interval [CI], 0.78-0.86), ≥65 years (AOR vs.<40, 0.61; 95% CI, 0.56-0.66), non-English speakers (0.72; 95% CI, 0.67-0.77), living ≥250 km away (AOR vs.<15 km, 0.71; 95% CI, 0.62-0.80), and without a PCP. Disease characteristics accounted for the largest proportion of observed variation (20.8%), with significantly greater odds of enrollment in patients with genitourinary cancers and late-stage disease. Significant sociodemographic disparities were observed, suggesting the need for targeted strategies to increase diversity in access to cancer CTs in Canada.

Association Between Past and Present Operative Mortality in Congenital Heart Surgery: An Analysis of The Society of Thoracic Surgeons Congenital Heart Surgery Database

BackgroundMortality after congenital heart surgery is an important metric across benchmarking, quality, and reporting initiatives. All rely on estimates from prior years, and how well these reflect current outcomes is unclear. MethodsIndex operations from The Society of Thoracic Surgeons Congenital Database (2014-2019) were included. Adjusted operative mortality in “past” 4-year and 1-year intervals vs the most recent year (“present”) was evaluated using Bayesian hierarchical logistic regression with results presented as odds ratios and 95% credible intervals (CrI). ResultsAmong 115,699 operations (106 hospitals), overall present observed mortality was 2.5%. Regression to the mean was evident, and individual hospital’s present vs past adjusted mortality were only weakly correlated (−0.11 vs past 1-year data, 0.22 vs past 4-year data). A significant relationship was found between past and present mortality only for the group of hospitals in the highest mortality quartile, most prominent for past 4-year data (adjusted odds ratio vs lowest mortality quartile, 2.04; 95% CrI, 1.44-2.80). The proportion of present hospital mortality variation explained by past mortality quartile was 52% (95% CrI, 20%-90%) using past 4-year data and only 27% (95% CrI, 92%-83%) using past 1-year data. Overall 66% of hospitals changed mortality quartiles from past to present (30% by ≥2 quartiles). ConclusionsPast mortality relates to present primarily for groups of hospitals at the extremes, with past 4-year data more informative than past 1-year data. For individual hospitals, past may differ from present, regression to the mean is common, and many change quartiles. Past mortality should be used thoughtfully and not as the sole factor informing present decision making.

Frailty’s influence on 30-day mortality in old critically ill ICU patients: a bayesian analysis evaluating the clinical frailty scale

IntroductionFrailty is widely acknowledged as influencing health outcomes among critically ill old patients. Yet, the traditional understanding of its impact has predominantly been through frequentist statistics. We endeavored to explore this association using Bayesian statistics aiming to provide a more nuanced understanding of this multifaceted relationship.MethodsOur analysis incorporated a cohort of 10,363 older (median age 82 years) patients from three international prospective studies, with 30-day all-cause mortality as the primary outcome. We defined frailty as Clinical Frailty Scale ≥ 5. A hierarchical Bayesian logistic regression model was employed, adjusting for covariables, using a range of priors. An international steering committee of registry members reached a consensus on a minimal clinically important difference (MCID).ResultsIn our study, the 30-day mortality was 43%, with rates of 38% in non-frail and 51% in frail groups. Post-adjustment, the median odds ratio (OR) for frailty was 1.60 (95% CI 1.45–1.76). Frailty was invariably linked to adverse outcomes (OR > 1) with 100% probability and had a 90% chance of exceeding the minimal clinically important difference (MCID) (OR > 1.5). For the Clinical Frailty Scale (CFS) as a continuous variable, the median OR was 1.19 (1.16–1.22), with over 99% probability of the effect being more significant than 1.5 times the MCID. Frailty remained outside the region of practical equivalence (ROPE) in all analyses, underscoring its clinical importance regardless of how it is measured.ConclusionsThis research demonstrates the significant impact of frailty on short-term mortality in critically ill elderly patients, particularly when the Clinical Frailty Scale (CFS) is used as a continuous measure. This approach, which views frailty as a spectrum, enables more effective, personalized care for this vulnerable group. Significantly, frailty was consistently outside the region of practical equivalence (ROPE) in our analysis, highlighting its clinical importance.

Potential Impact of Higher-Valency Pneumococcal Conjugate Vaccines Among Adults in Different Localities in Germany.

Next-generation pneumococcal conjugate vaccines (PCVs) have been approved for use. The serotype distribution of pneumococcal isolates can vary between regions. To understand the potential impacts of new PCVs, we evaluated trends in invasive pneumococcal disease (IPD) among adults in Germany at a local level using Bayesian hierarchical logistic regression. There was little spatial variation in IPD cases caused by 13-valent PCV serotypes, which dropped from 60% of IPD cases in 2006 to 30% in 2018. More than half of IPD cases in 2018 were attributable to serotypes covered by new PCVs (15-valent and 20-valent PCVs), which suggests they could further reduce the burden of IPD.

Abstract 15107: Dramatic Impact of a Liaison-Mediated Referral Strategy on Cardiac Rehabilitation Participation After Percutaneous Coronary Intervention: Insights From the Blue Cross Blue Shield of Michigan Cardiovascular Consortium

Background: While cardiac rehabilitation (CR) improves outcomes after percutaneous coronary intervention (PCI) it remains underutilized. A liaison-mediated referral, where a healthcare professional explains CR’s benefits and places a referral before discharge, may promote CR use. Hypothesis: We hypothesize a liaison-mediated CR referral is associated with greater CR participation after PCI. Methods: This was a retrospective cohort study of patients who underwent PCI across 48 hospitals in Michigan between January 2021 and April 2022 and were referred to CR before discharge. Clinical registry data was linked to Medicare and commercial claims to identify the primary outcome of CR participation, defined as attending ≥1 CR session within 90 days of discharge from the index PCI admission. Bayesian hierarchical logistic regression modeling was used to compare CR participation between patients with and without a liaison referral, adjusting for patient characteristics. Cox-proportional hazard modeling was used to compare days elapsed between discharge and first CR session between liaison cohorts. Results: Among 9,023 patients referred to CR after PCI, 4,323 (47.9%) underwent a liaison referral and 3,390 (37.6%) attended ≥1 CR session within 90 days of discharge. The liaison and non-liaison cohorts had a mean age of 69.3 vs 69.4 years, and 68.3% vs 68.4% were male, respectively. The liaison cohort had a higher unadjusted 90-day CR participation rate compared with the non-liaison cohort (Figure). After adjustment, the liaison cohort had a significantly higher odds of attending ≥1 CR session and a shorter delay in attending the first session compared with the non-liaison cohort (28 vs 33 days, p&lt;0.001) (Figure). Conclusion: A liaison-mediated CR referral is associated with higher odds of post-PCI CR participation and may mitigate delays in CR enrollment. This may be an effective referral strategy to improve CR participation rates and patient outcomes after PCI.

Patient Preferences for Adjuvant Treatment in Muscle-Invasive Urothelial Carcinoma: A Multi-Country Discrete Choice Experiment.

The evolving treatment landscape in muscle-invasive urothelial carcinoma creates challenges for clinicians and patients in selecting the most appropriate therapy. Here, we aimed to understand adjuvant treatment preferences among patients with muscle-invasive urothelial carcinoma who underwent radical resection, including tradeoffs between efficacy outcomes and toxicity risks. An observational, cross-sectional study utilizing a discrete choice experiment was conducted across the United States, United Kingdom, Canada, France, and Germany via a web-based survey. Patients ≥18years of age who self-reported as having been diagnosed with muscle-invasive urothelial carcinoma were included. Patients indicated their preferences between hypothetical treatment profiles varying in eight attributes relating to efficacy, regimen, and side effects. Preference weights were estimated using hierarchical Bayesian logistic regression; relative attribute importance estimates were calculated. Overall, 207 patients were included (age ≥56 years, 65.7%; male, 54.1%). Patients chose adjuvant treatment 91.2% of the time vs no treatment. Prolonging overall survival from 25 to 78 months was most important, followed by reducing serious side effect risks. Increasing disease-free survival from 12 to 24 months was more important than decreasing risks of fatigue from 54% to 15% and nausea from 53% to 7%. Treatment with the shortest dosing regimen was more important for patients who received neoadjuvant chemotherapy vs patients who did not receive neoadjuvant chemotherapy; prolonging overall survival was more important than reducing the risk of a serious side effect in non-US patients; the opposite was found in the United States. Patients with muscle-invasive urothelial carcinoma who underwent radical resection preferred adjuvant treatment over no treatment regardless of side effects. Patients prioritized overall survival improvements followed by a reduced side effect profile.

Association of glycemic gap with stroke recurrence in patients with ischemic stroke.

Glycemic gap, as a novel index of acute glycemic excursion, is associated with poor prognosis of different diseases. This study aimed to explore the association of the glycemic gap with long-term stroke recurrence in patients with ischemic stroke. This study included patients with ischemic stroke from the Nanjing Stroke Registry Program. The glycemic gap was calculated by subtracting the estimated average blood glucose from the blood glucose at admission. Multivariable Cox proportional hazards regression analysis was performed to explore the association between the glycemic gap and the risk of stroke recurrence. The Bayesian hierarchical logistic regression model was used to estimate the effects of the glycemic gap on stroke recurrence stratified by diabetes mellitus and atrial fibrillation. Among 2734 enrolled patients, 381 (13.9%) patients experienced stroke recurrence during a median follow-up of 3.02 years. In multivariate analysis, glycemic gap (high group vs. median group) was associated with significantly increased risk for stroke recurrence (adjusted hazard ratio, 1.488; 95% confidence interval, 1.140-1.942; p = .003) and had varying effects on stroke recurrence depending on atrial fibrillation. The restricted cubic spline curve showed a U-shaped relationship between the glycemic gap and stroke recurrence (p = .046 for nonlinearity). Our study found that the glycemic gap was significantly associated with stroke recurrence in patients with ischemic stroke. The glycemic gap was consistently associated with stroke recurrence across subgroups and had varying effects depending on atrial fibrillation.

Disparities in clinical trial enrolment at a Canadian comprehensive cancer centre: A 15-year retrospective study.

6550 Background: Disparities in clinical trials limit the generalizability of study findings and perpetuate disparities in treatment access and outcomes, but there is a paucity of data in the Canadian context. The objective of this study was to examine disparities in cancer clinical trial enrollment at a large, comprehensive Canadian cancer center. Methods: We conducted a retrospective cohort study of clinical trial enrollment among all newly diagnosed cancer patients (N=154,880) at the Princess Margaret Cancer Centre in Toronto, Canada (2006-2021). Multivariable Bayesian hierarchical logistic regression with random effect for most responsible physician was used to examine the correlates of clinical trial enrollment, comparing the population of enrolled and non-enrolled patients. Odds ratios were adjusted for patient variables (sex, age at diagnosis, language, geography, primary care provider), and census tract-level marginalization (residential instability, material deprivation, dependency, ethnic concentration), disease variables (cancer site and disease stage at diagnosis), and provider variables (most responsible physician (MRP), and MRP’s sex, language, medical training, and department). Results: Overall, 11.2% of patients enrolled (n=17,400) in a clinical trial, with 5-, 10-, and 15-year cumulative incidences of 12%, 15%, and 18%, respectively. Small, but significant differences were observed between enrollees and the overall patient population. Lower odds of enrollment were observed in patients who were female (adjusted odds ratio [AOR], 0.82; 95% confidence interval [CI], 0.78-0.86; p&lt;.001), ≥65 years (AOR vs &lt;40, 0.61; 95% CI, 0.56-0.65; p&lt;.001), non-English language speakers (AOR vs English, 0.72; 95% CI, 0.67-0.77; p&lt;.001), lived ≥250 km away from the cancer center (AOR vs &lt;15km, 0.71; 95% CI, 0.62-0.80; p&lt;.001), or lived in areas with greater material deprivation (AOR, 0.94; 95% CI, 0.93-0.96; p&lt;.001) or higher ethnic concentration (AOR, 0.96; 95% CI, 0.94-0.98; p&lt;.001). Greater odds of enrollment were found in patients with metastatic disease (AOR, 1.19; 95% CI, 1.13-1.25; p&lt;.001) and in those with a primary care provider (AOR, 1.69; 95% CI, 1.55-1.85; p&lt;.001). Conclusions: Disparities were observed in clinical trial enrollment, despite a publicly funded health care system. While broader prospective data collection efforts are critical to better understand the influence of patient, provider and system factors on clinical trial enrollment, these findings suggest the need for targeted strategies to increase diversity in clinical trial access.

Bayesian Logistic Regression Model for Sub-Areas

Many population-based surveys have binary responses from a large number of individuals in each household within small areas. One example is the Nepal Living Standards Survey (NLSS II), in which health status binary data (good versus poor) for each individual from sampled households (sub-areas) are available in the sampled wards (small areas). To make an inference for the finite population proportion of individuals in each household, we use the sub-area logistic regression model with reliable auxiliary information. The contribution of this model is twofold. First, we extend an area-level model to a sub-area level model. Second, because there are numerous sub-areas, standard Markov chain Monte Carlo (MCMC) methods to find the joint posterior density are very time-consuming. Therefore, we provide a sampling-based method, the integrated nested normal approximation (INNA), which permits fast computation. Our main goal is to describe this hierarchical Bayesian logistic regression model and to show that the computation is much faster than the exact MCMC method and also reasonably accurate. The performance of our method is studied by using NLSS II data. Our model can borrow strength from both areas and sub-areas to obtain more efficient and precise estimates. The hierarchical structure of our model captures the variation in the binary data reasonably well.

Abstract 12623: Association Between Past and Present Operative Mortality in Congenital Heart Surgery

Introduction: Operative mortality following congenital heart surgery is a foundational metric used across a variety of benchmarking, quality improvement, and public reporting initiatives. All of these efforts rely on mortality estimates reported from prior years, and it is unclear how well these reflect current outcomes. Methods: Adjusted operative mortality among Society of Thoracic Surgeons Congenital Heart Surgery Database index operations (2014-2019) was analyzed using standard risk-models. The association of mortality in “past” 4- and 1-year time intervals vs. the most recent year in the dataset (“present”) was evaluated using Bayesian hierarchical logistic regression. Estimates and 95% credible intervals (CrI) are presented. Results: Among 115,699 operations (106 hospitals) present observed mortality ranged from 2.3%-3.1% across quartiles, with regression to the mean evident more so vs. past 1-year than past 4-year data. Odds of present mortality were 2.04 (95%CrI 1.44-2.80) times higher for those in the past worst vs. best 4-year mortality quartile; no significant association was observed for the middle quartiles. The proportion of present hospital mortality variation explained by past mortality quartile was 52% (95%CrI 20%-90%) using past 4-year and 27% (95%CrI 2%-83%) using past 1-year data. Overall 66% of hospitals changed mortality quartiles from past to present (30% by 2 or more quartiles), with lower volume hospitals most impacted. Conclusions: Past mortality relates to present mortality primarily for groups of hospitals at the extremes, with past 4-year more informative than past 1-year data. However when considering individual hospitals, past may differ from present, regression to the mean is common, and many hospitals change quartiles. Past mortality estimates should be used thoughtfully and not as the sole factor informing present decision making.

High monkeypox vaccine acceptance among male users of smartphone-based online gay-dating apps in Europe, 30 July to 12 August 2022.

We assess monkeypox vaccination acceptance among male adults in the European Region. We conducted an online survey through two dating apps targeting men who have sex with men, from 30 July to 12 August 2022. We developed Bayesian hierarchical logistic regression models to investigate monkeypox vaccination acceptance. Overall crude vaccination acceptance was 82% and higher in north-western compared to south-eastern European regions. Acceptance strongly rose with perception of increased disease severity and transmission risk, and in individuals linked to healthcare.

Adaptive spatial designs minimizing the integrated Bernoulli variance in spatial logistic regression models - with an application to benthic habitat mapping

Constructing efficient spatial sampling designs is important for monitoring, understanding and characterizing environmental processes acting over large-scale spatial domain. Multiple types of devices can accomplish informative sampling in various ways, and time and computational limitations should be considered too. In-situ sampling is often accurate, but it tends to be very sparse in the vast spatial areas considered for mapping purposes, and careful constructions of experimental sampling designs are needed. A criterion for mapping spatial presence-absence variables is presented. The expected integrated Bernoulli variance criterion is used in order to explore regions with more uncertainty about the binary outcomes. This approach uses a hierarchical Bayesian logistic regression model for the binary variable and develops approximate closed form expressions for the expected integrated Bernoulli variance. The approximations are fast to compute for many designs. The expressions are extended to find adaptive designs in the setting of sequential spatial exploration, where there are limited computation resources and operational time restrictions. In simulation studies the approximations are compared with tedious Monte Carlo sampling methods. Approximations are shown to be accurate and to work for a large spatial grid. An application in benthic habitat mapping is presented. It considers a coral dataset from Australia to demonstrate the suggested sampling design approach.

Micronutrient deficiencies among preschool-aged children and women of reproductive age worldwide: a pooled analysis of individual-level data from population-representative surveys

Micronutrient deficiencies compromise immune systems, hinder child growth and development, and affect human potential worldwide. Yet, to our knowledge, the only existing estimate of the global prevalence of micronutrient deficiencies is from over 30 years ago and is based only on the prevalence of anaemia. We aimed to estimate the global and regional prevalence of deficiency in at least one of three micronutrients among preschool-aged children (aged 6-59 months) and non-pregnant women of reproductive age (aged 15-49 years). In this pooled analysis, we reanalysed individual-level biomarker data for micronutrient status from nationally representative, population-based surveys. We used Bayesian hierarchical logistic regression to estimate the prevalence of deficiency in at least one of three micronutrients for preschool-aged children (iron, zinc, and vitamin A) and for non-pregnant women of reproductive age (iron, zinc, and folate), globally and in seven regions using 24 nationally representative surveys done between 2003 and 2019. We estimated the global prevalence of deficiency in at least one of three micronutrients to be 56% (95% uncertainty interval [UI] 48-64) among preschool-aged children, and 69% (59-78) among non-pregnant women of reproductive age, equivalent to 372 million (95% UI 319-425) preschool-aged children and 1·2 billion (1·0-1·4) non-pregnant women of reproductive age. Regionally, three-quarters of preschool-aged children with micronutrient deficiencies live in south Asia (99 million, 95% UI 80-118), sub-Saharan Africa (98 million, 83-113), or east Asia and the Pacific (85 million, 61-110). Over half (57%) of non-pregnant women of reproductive age with micronutrient deficiencies live in east Asia and the Pacific (384 million, 279-470) or south Asia (307 million, 255-351). We estimate that over half of preschool-aged children and two-thirds of non-pregnant women of reproductive age worldwide have micronutrient deficiencies. However, estimates are uncertain due to the scarcity of population-based micronutrient deficiency data. US Agency for International Development.

A Bayesian reanalysis of the Standard versus Accelerated Initiation of Renal-Replacement Therapy in Acute Kidney Injury (STARRT-AKI) trial

BackgroundTiming of initiation of kidney-replacement therapy (KRT) in critically ill patients remains controversial. The Standard versus Accelerated Initiation of Renal-Replacement Therapy in Acute Kidney Injury (STARRT-AKI) trial compared two strategies of KRT initiation (accelerated versus standard) in critically ill patients with acute kidney injury and found neutral results for 90-day all-cause mortality. Probabilistic exploration of the trial endpoints may enable greater understanding of the trial findings. We aimed to perform a reanalysis using a Bayesian framework.MethodsWe performed a secondary analysis of all 2927 patients randomized in multi-national STARRT-AKI trial, performed at 168 centers in 15 countries. The primary endpoint, 90-day all-cause mortality, was evaluated using hierarchical Bayesian logistic regression. A spectrum of priors includes optimistic, neutral, and pessimistic priors, along with priors informed from earlier clinical trials. Secondary endpoints (KRT-free days and hospital-free days) were assessed using zero–one inflated beta regression.ResultsThe posterior probability of benefit comparing an accelerated versus a standard KRT initiation strategy for the primary endpoint suggested no important difference, regardless of the prior used (absolute difference of 0.13% [95% credible interval [CrI] − 3.30%; 3.40%], − 0.39% [95% CrI − 3.46%; 3.00%], and 0.64% [95% CrI − 2.53%; 3.88%] for neutral, optimistic, and pessimistic priors, respectively). There was a very low probability that the effect size was equal or larger than a consensus-defined minimal clinically important difference. Patients allocated to the accelerated strategy had a lower number of KRT-free days (median absolute difference of − 3.55 days [95% CrI − 6.38; − 0.48]), with a probability that the accelerated strategy was associated with more KRT-free days of 0.008. Hospital-free days were similar between strategies, with the accelerated strategy having a median absolute difference of 0.48 more hospital-free days (95% CrI − 1.87; 2.72) compared with the standard strategy and the probability that the accelerated strategy had more hospital-free days was 0.66.ConclusionsIn a Bayesian reanalysis of the STARRT-AKI trial, we found very low probability that an accelerated strategy has clinically important benefits compared with the standard strategy. Patients receiving the accelerated strategy probably have fewer days alive and KRT-free. These findings do not support the adoption of an accelerated strategy of KRT initiation.

Factors That Affect Legionella Positivity in Healthcare Building Water Systems from a Large, National Environmental Surveillance Initiative.

Legionella growth in healthcare building water systems can result in legionellosis, making water management programs (WMPs) important for patient safety. However, knowledge is limited on Legionella prevalence in healthcare buildings. A dataset of quarterly water testing in Veterans Health Administration (VHA) healthcare buildings was used to examine national environmental Legionella prevalence from 2015 to 2018. Bayesian hierarchical logistic regression modeling assessed factors influencing Legionella positivity. The master dataset included 201,146 water samples from 814 buildings at 168 VHA campuses. Overall Legionella positivity over the 4 years decreased from 7.2 to 5.1%, with the odds of a Legionella-positive sample being 0.94 (0.90-0.97) times the odds of a positive sample in the previous quarter for the 16 quarters of the 4 year period. Positivity varied considerably more at the medical center campus level compared to regional levels or to the building level where controls are typically applied. We found higher odds of Legionella detection in older buildings (OR 0.92 [0.86-0.98] for each more recent decade of construction), in taller buildings (OR 1.20 [1.13-1.27] for each additional floor), in hot water samples (O.R. 1.21 [1.16-1.27]), and in samples with lower residual biocide concentrations. This comprehensive healthcare building review showed reduced Legionella detection in the VHA healthcare system over time. Insights into factors associated with Legionella positivity provide information for healthcare systems implementing WMPs and for organizations setting standards and regulations.