Mitigating selection bias in organ allocation models.
Ontology highlight
ABSTRACT: Background
The lung allocation system in the U.S. prioritizes lung transplant candidates based on estimated pre- and post-transplant survival via the Lung Allocation Scores (LAS). However, these models do not account for selection bias, which results from individuals being removed from the waitlist due to receipt of transplant, as well as transplanted individuals necessarily having survived long enough to receive a transplant. Such selection biases lead to inaccurate predictions.Methods
We used a weighted estimation strategy to account for selection bias in the pre- and post-transplant models used to calculate the LAS. We then created a modified LAS using these weights, and compared its performance to that of the existing LAS via time-dependent receiver operating characteristic (ROC) curves, calibration curves, and Bland-Altman plots.Results
The modified LAS exhibited better discrimination and calibration than the existing LAS, and led to changes in patient prioritization.Conclusions
Our approach to addressing selection bias is intuitive and can be applied to any organ allocation system that prioritizes patients based on estimated pre- and post-transplant survival. This work is especially relevant to current efforts to ensure more equitable distribution of organs.
SUBMITTER: Schnellinger EM
PROVIDER: S-EPMC8454078 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Generalized linear mixed models are a common statistical tool for the analysis of clustered or longitudinal data where correlation is accounted for through cluster-specific random effects. In practice, the distribution of the random effects is typically taken to be a Normal distribution, although if this does not hold then the model is misspecified and standard estimation/inference may be invalid. An alternative is to perform a so-called nonparametric Bayesian analyses in which one assigns a Dirichlet process (DP) prior to the unknown distribution of the random effects. In this paper we examine operating characteristics for estimation of fixed effects and random effects based on such an analysis under a range of "true" random effects distributions. As part of this we investigate various approaches for selection of the precision parameter of the DP prior. In addition, we illustrate the use of the methods with an analysis of post-operative complications among n = 18, 643 female Medicare beneficiaries who underwent a hysterectomy procedure at N = 503 hospitals in the US. Overall, we conclude that using the DP priori n modeling the random effect distribution results in large reductions of bias with little loss of efficiency. While no single choice for the precision parameter will be optimal in all settings, certain strategies such as importance sampling or empirical Bayes can be used to obtain reasonable results in a broad range of data scenarios.
| S-EPMC5624537 | biostudies-literature
Project description:BackgroundThe Lung Allocation Score (LAS) is used in the U.S. to prioritize lung transplant candidates. Selection bias, induced by dependent censoring of waitlisted candidates and prediction of posttransplant survival among surviving, transplanted patients only, is only partially addressed by the LAS. Recently, a modified LAS (mLAS) was designed to mitigate such bias. Here, we estimate the clinical impact of replacing the LAS with the mLAS.MethodsWe considered lung transplant candidates waitlisted during 2016 and 2017. LAS and mLAS scores were computed for each registrant at each observed organ offer date; individuals were ranked accordingly. Patient characteristics associated with better priority under the mLAS were investigated via logistic regression and generalized linear mixed models. We also determined whether differences in rank were explained more by changes in predicted pre- or posttransplant survival. Simulations examined how 1-year waitlist, posttransplant, and overall survival might change under the mLAS.ResultsDiagnosis group, 6-minute walk distance, continuous mechanical ventilation, functional status, and age demonstrated the highest impact on differential allocation. Differences in rank were explained more by changes in predicted pretransplant survival than changes in predicted posttransplant survival, suggesting that selection bias has more impact on estimates of waitlist urgency. Simulations suggest that for every 1000 waitlisted individuals, 12.8 (interquartile range: 5.2-24.3) fewer waitlist deaths per year would occur under the mLAS, without compromising posttransplant and overall survival.ConclusionsImplementing a mLAS that mitigates selection bias into clinical practice can lead to important differences in allocation and possibly modest improvement in waitlist survival.
| S-EPMC10167739 | biostudies-literature
Project description:IntroductionSimulation is a valuable and novel tool in the expanding approach to racism and bias education for medical practitioners. We present a simulation case focused on identifying and addressing the implicit bias of a consultant to teach bias mitigation skills and limit harm to patients and families.MethodsLearners were presented with a case of a classic toddler's fracture in an African American child. The learners interacted with an orthopedic resident who insisted on child welfare involvement, with nonspecific and increasingly biased concerns about the child/family. The learners were expected to identify that this case was not concerning for nonaccidental trauma and that the orthopedic resident was demonstrating bias. They were expected to communicate with both the resident and the parent effectively to defuse the situation and prevent harm from reaching the family. A debrief and an anonymous survey followed the case.ResultsSeventy-five learners participated, including pediatric and emergency medicine residents, fellows, attendings, and medical students. After the case, the majority of learners expressed confidence that they could recognize racial bias in the care of a patient (90%), ensure patient care was not influenced by racial bias (88%), and utilize a tool to frame a concern about bias (79%).DiscussionParticipants felt that this simulation was relevant and effective and overall left the experience feeling confident in their abilities to identify and manage racially biased patient care. This anti-racist simulation offers an important skill-building opportunity that has been well received by learners.
| S-EPMC10507144 | biostudies-literature
Project description:Student evaluations of teaching are widely believed to contain gender bias. In this study, we conduct a randomized experiment with the student evaluations of teaching in four classes with large enrollments, two taught by male instructors and two taught by female instructors. In each of the courses, students were randomly assigned to either receive the standard evaluation instrument or the same instrument with language intended to reduce gender bias. Students in the anti-bias language condition had significantly higher rankings of female instructors than students in the standard treatment. There were no differences between treatment groups for male instructors. These results indicate that a relatively simple intervention in language can potentially mitigate gender bias in student evaluation of teaching.
| S-EPMC6519786 | biostudies-literature
Project description:PCR amplification plays an integral role in the measurement of mixed microbial communities via high-throughput DNA sequencing of the 16S ribosomal RNA (rRNA) gene. Yet PCR is also known to introduce multiple forms of bias in 16S rRNA studies. Here we present a paired modeling and experimental approach to characterize and mitigate PCR NPM-bias (PCR bias from non-primer-mismatch sources) in microbiota surveys. We use experimental data from mock bacterial communities to validate our approach and human gut microbiota samples to characterize PCR NPM-bias under real-world conditions. Our results suggest that PCR NPM-bias can skew estimates of microbial relative abundances by a factor of 4 or more, but that this bias can be mitigated using log-ratio linear models.
| S-EPMC8284789 | biostudies-literature
Project description:Background and aimsPlant growth depends on carbon availability and allocation among organs. QualiTree has been designed to simulate carbon allocation and partitioning in the peach tree (Prunus persica), whereas MappleT is dedicated to the simulation of apple tree (Malus × domestica) architecture. The objective of this study was to couple both models and adapt QualiTree to apple trees to simulate organ growth traits and their within-tree variability.MethodsMappleT was used to generate architectures corresponding to the 'Fuji' cultivar, accounting for the variability within and among individuals. These architectures were input into QualiTree to simulate shoot and fruit growth during a growth cycle. We modified QualiTree to account for the observed shoot polymorphism in apple trees, i.e. different classes (long, medium and short) that were characterized by different growth function parameters. Model outputs were compared with observed 3D tree geometries, considering shoot and final fruit size and growth dynamics.Key resultsThe modelling approach connecting MappleT and QualiTree was appropriate to the simulation of growth and architectural characteristics at the tree scale (plant leaf area, shoot number and types, fruit weight at harvest). At the shoot scale, mean fruit weight and its variability within trees was accurately simulated, whereas the model tended to overestimate individual shoot leaf area and underestimate its variability for each shoot type. Varying the parameter related to the intensity of carbon exchange between shoots revealed that behaviour intermediate between shoot autonomy and a common assimilate pool was required to properly simulate within-tree fruit growth variability. Moreover, the model correctly dealt with the crop load effect on organ growth.ConclusionsThis study provides understanding of the integration of shoot ontogenetic properties, carbon supply and transport between entities for simulating organ growth in trees. Further improvements regarding the integration of retroaction loops between carbon allocation and the resulting plant architecture are expected to allow multi-year simulations.
| S-EPMC4970356 | biostudies-literature
Project description:Memory allocation refers to the process by which neurons are recruited into the encoding ensemble to store learned information. This recruitment is regulated by neuronal selection based on differences in intrinsic excitability (IE) and the expression of the transcription factor CREB. However, whether other forms of plasticity influence memory allocation remains unknown. Here, we found that chromatin compaction and histone acetylation in the mouse lateral amygdala display a high degree of heterogeneity, a prerequisite for neuronal selection. Consequently, when we increased histone acetylation by overexpressing histone acetyl transferases (HATs), neurons with elevated histone acetylation were preferentially recruited into the encoding ensemble and memory retention was enhanced, while optogenetic silencing of the epigenetically altered neurons prevented memory expression. Mechanistically, using patch-clamp recordings and single-nucleus multi-ome sequencing, we observed that HAT overexpression increased IE and epitranscriptomic changes favoring synaptic plasticity. Lastly, by merging FRET-based epigenetic beacons with calcium indicators to simultaneously record histone acetylation and neuronal dynamics in real time, we found that epigenetic heterogeneity underlies IE in cell-autonomous manner. These results identify chromatin plasticity as a key factor catalyzing memory allocation.
2024-07-26 | GSE220696 | GEO
Project description:Due to socioeconomic differences, the accuracy and extent of reporting on the occurrence of native species differs among countries, which can impact the performance of species distribution models. We assessed the importance of geographical biases in occurrence data on model performance using Hydrilla verticillata as a case study. We used Maxent to predict potential North American distribution of the aquatic invasive macrophyte based upon training data from its native range. We produced a model using all available native range occurrence data, then explored the change in model performance produced by omitting subsets of training data based on political boundaries. We also compared those results with models trained on data from which a random sample of occurrence data was omitted from across the native range. Although most models accurately predicted the occurrence of H. verticillata in North America (AUC > 0.7600), data omissions influenced model predictions. Omitting data based on political boundaries resulted in larger shifts in model accuracy than omitting randomly selected occurrence data. For well-documented species like H. verticillata, missing records from single countries or ecoregions may minimally influence model predictions, but for species with fewer documented occurrences or poorly understood ranges, geographic biases could misguide predictions. Regardless of focal species, we recommend that future species distribution modeling efforts begin with a reflection on potential spatial biases of available occurrence data. Improved biodiversity surveillance and reporting will provide benefit not only in invaded ranges but also within under-reported and unexplored native ranges.
| S-EPMC4203300 | biostudies-literature
Project description:Amplicon based metabarcoding promises rapid and cost-efficient analyses of species composition. However, it is disputed whether abundance estimates can be derived from metabarcoding due to taxon specific PCR amplification biases. PCR-free approaches have been suggested to mitigate this problem, but come with considerable increases in workload and cost. Here, we analyze multilocus datasets of diverse arthropod communities, to evaluate whether amplification bias can be countered by (1) targeting loci with highly degenerate primers or conserved priming sites, (2) increasing PCR template concentration, (3) reducing PCR cycle number or (4) avoiding locus specific amplification by directly sequencing genomic DNA. Amplification bias is reduced considerably by degenerate primers or targeting amplicons with conserved priming sites. Surprisingly, a reduction of PCR cycles did not have a strong effect on amplification bias. The association of taxon abundance and read count was actually less predictable with fewer cycles. Even a complete exclusion of locus specific amplification did not exclude bias. Copy number variation of the target loci may be another explanation for read abundance differences between taxa, which would affect amplicon based and PCR free methods alike. As read abundance biases are taxon specific and predictable, the application of correction factors allows abundance estimates.
| S-EPMC5732254 | biostudies-literature
Project description:Harsh exclusionary discipline predicts major negative life outcomes, including adult incarceration and unemployment. This breeds racial inequality because Black students are disproportionately at risk for this type of discipline. Can a combination of policy and psychological interventions reduce this kind of discipline and mitigate this inequality? Two preregistered experiments (Nexperiment1 = 246 teachers; Nexperiment2 = 243 teachers) used an established paradigm to systematically test integration of two and then three policy and psychological interventions to mitigate the consequences of bias (troublemaker labeling and pattern perception) on discipline (discipline severity). Results indicate that the integrated interventions can curb teachers' troublemaker labeling and pattern prediction toward Black students who misbehave in a hypothetical paradigm. In turn, integration of the three components reduced racial inequality in teachers' discipline decisions. This research informs scientific theory, public policy, and interventions.
| S-EPMC7567593 | biostudies-literature