On asymptotic distributions of several test statistics for familial relatedness in linear mixed models.

In this study, the asymptotic distributions of the likelihood ratio test (LRT), the restricted likelihood ratio test (RLRT), the F and the sequence kernel association test (SKAT) statistics for testing an additive effect of the expected familial relatedness (FR) in a linear mixed model are examined based on an eigenvalue approach. First, the covariance structure for modeling the FR effect in a LMM is presented. Then, the multiplicity of eigenvalues for the log-likelihood and restricted log-likelihood is established under a replicate family setting and extended to a more general replicate family setting (GRFS) as well. After that, the asymptotic null distributions of LRT, RLRT, F and SKAT statistics under GRFS are derived. The asymptotic null distribution of SKAT for testing genetic rare variants is also constructed. In addition, a simple formula for sample size calculation is provided based on the restricted maximum likelihood estimate of the effect size for the expected FR. Finally, a power comparison of these test statistics on hypothesis test of the expected FR effect is made via simulation. The four test statistics are also applied to a data set from the UK Biobank.

A unified linear mixed model for familial relatedness and population structure in genetic association studies.

Familial relatedness (FR) and population structure (PS) are two major sources for genetic correlation. In the human population, both FR and PS can further break down into additive and dominant components to account for potential additive and dominant genetic effects. In this study, besides the classical additive genomic relationship matrix, a dominant genomic relationship matrix is introduced. A link between the additive/dominant genomic relationship matrices and the coancestry (or kinship)/double coancestry coefficients is also established. In addition, a way to separate the FR and PS correlations based on the estimates of coancestry and double coancestry coefficients from the genomic relationship matrices is proposed. A unified linear mixed model is also developed, which can account for both the additive and dominance effects of FR and PS correlations as well as their possible random interactions. Finally, this unified linear mixed model is applied to analyze two study cohorts from UK Biobank.

Brain death interval and relationship to outcomes of pediatric cardiac transplantation.

BACKGROUND: We sought to analyze brain death interval and outcomes of pediatric cardiac transplantation using national registry data. METHODS: We retrospectively evaluated a pediatric cohort from the UNOS registry from 2005 to 2014. We restricted the donor cohort to those with a primary central nervous system event as the cause of hospitalization. Brain death interval (BDI) was defined as the time between hospital admission and organ procurement. Primary outcomes were recipient and graft survival time. Logistical regression modeling was used for multivariable analysis. RESULTS: The donor cohort included 2565 cases. Multivariable analysis demonstrated no relationship between BDI and recipient or graft survival time. For patient survival time, the lowest HR was 0.94 (0.63-1.39), P = 0.531; for graft survival time, the lowest HR was 0.89 (0.53-1.49), P = 0.563. We obtained similar results using a non-restricted donor cohort. CONCLUSIONS: There was no clear relationship between BDI and recipient or graft survival after pediatric cardiac transplantation.

The design and conduct of a randomized clinical trial comparing emergency department initiation of sublingual versus a 7-day extended-release injection formulation of buprenorphine for opioid use disorder: Project ED Innovation.

ED-INNOVATION (Emergency Department-INitiated bupreNOrphine VAlidaTION) is a Hybrid Type-1 Implementation-Effectiveness multisite emergency department (ED) study funded through The Helping to End Addiction Long-termSM Initiative, or NIH HEAL InitiativeSM efforts to increase access to medications for opioid use disorder (OUD). We use components of Implementation Facilitation to enhance adoption of ED-initiated buprenorphine (BUP) at approximately 30 sites. Subsequently we compare the effectiveness of two BUP formulations, sublingual (SL-BUP) and 7-day extended-release injectable (CAM2038, XR-BUP) in a randomized clinical trial (RCT) of approximately 2000 patients with OUD on the primary outcome of engagement in formal addiction treatment at 7 days. Secondary outcomes assessed at 7 and 30 days include self-reported opioid use, craving and satisfaction, health service utilization, overdose events, and engagement in formal addiction treatment (30 days) and receipt of medications for OUD (at 7 and 30 days). A sample size of 1000 per group provides 90% power at the 2-sided significance level to detect a difference in the primary outcome of 8% and accommodates a 15% dropout rate. We will compare the cost effectiveness of the two treatments on the primary outcome using the incremental cost-effectiveness ratio. We will also conduct an ancillary study in approximately 75 patients experiencing minimal to no opioid withdrawal who will undergo XR-BUP initiation. If the ancillary study demonstrates safety, we will expand the eligibility criteria for the RCT to include individuals with minimal to no opioid withdrawal. The results of these studies will inform implementation of ED-initiated BUP in diverse EDs which has the potential to improve treatment access.

Human Pelvis Bayesian Injury Probability Curves From Whole Body Lateral Impact Experiments.

Injury criteria are used in military, automotive, and aviation environments to advance human safety. While injury risk curves (IRCs) for the human pelvis are published under vertical loading, there is a paucity of analysis that describe IRCs under lateral impact. The objective of the present study is to derive IRCs under this mode. Published data were used from 60 whole-body postmortem human surrogate (PMHS) tests that used repeated testing protocols. In the first analysis, from single impact tests, all injury data points were considered as left censored and noninjury points were considered as right censored, while repeated testing results were treated as interval censored data. In the second analysis, injury data were treated uncensored. Peak force was used as the response variable. Age, total body mass, gender, and body mass index (BMI) were used as covariates in different combinations. Bayesian survival analysis model was used to derive the IRCs. Plus-minus 95% credible intervals (CI) and their normalized CI sizes (NCIS) were obtained. This is the first study to develop IRCs in whole body PMHS tests to describe the human pelvic tolerance under lateral impact using Bayesian models.

Hierarchical process using Brier Score Metrics for lower leg injury risk curves in vertical impact.

INTRODUCTION: Parametric survival models are used to develop injury risk curves (IRCs) from impact tests using postmortem human surrogates (PMHS). Through the consideration of different output variables, input parameters and censoring, different IRCs could be created. The purpose of this study was to demonstrate the feasibility of the Brier Score Metric (BSM) to determine the optimal IRCs and derive them from lower leg impact tests. METHODS: Two series of tests of axial impacts to PMHS foot-ankle complex were used in the study. The first series used the metrics of force, time and rate, and covariates of age, posture, stature, device and presence of a boot. Also demonstrated were different censoring schemes: right and exact/uncensored (RC-UC) or right and uncensored/left (RC-UC-LC). The second series involved only one metric, force, and covariates age, sex and weight. It contained interval censored (IC) data demonstrating different censoring schemes: RC-IC-UC, RC-IC-LC and RC-IC-UC-LC. RESULTS: For each test set combination, optimal IRCs were chosen based on metric-covariate combination that had the lowest BSM value. These optimal IRCs are shown along with 95% CIs and other measures of interval quality. Forces were greater for UC than LC data sets, at the same risk levels (10% used in North Atlantic Treaty Organisation (NATO)). All data and IRCs are presented. CONCLUSIONS: This study demonstrates a novel approach to examining which metrics and covariates create the best parametric survival analysis-based IRCs to describe human tolerance, the first step in describing lower leg injury criteria under axial loading to the plantar surface of the foot.

Uncertainty Evaluations for Risk Assessment in Impact Injuries and Implications for Clinical Practice.

Injury risk curves (IRCs) represent the quantification of risk of adverse outcomes, such as a bone fracture, quantified by a biomechanical metric such as force or deflection. From a biomechanical perspective, they are crucial in crashworthiness studies to advance human safety. In clinical settings, they can be used as an assistive tool to aid in the decision-making process for surgical or conservative treatment. The estimation of risk corresponding to a level of biomechanical metric is done using a regression technique, such as a parametric survival regression model. As with any statistical procedure, error measures are computed for the IRC, representing the quality of the estimated risk. For example, confidence intervals (CIs) are recommended by the International Standards Organization, and the normalized confidence interval width (NCIW) is computed based on the width of the CI. This is a surrogate for the quality of the risk curve. A 95% CI means that if the same experiment were hypothetically repeated 100 times, at least 95 of the computed CIs should contain the true risk curve. Such an interpretation is problematic in most biomechanical contexts as rarely the same experiment is repeated. The notion that a wider confidence interval implies a poorer quality risk curve can be misleading. This article considers the evaluation of CIs and its implications in biomechanical settings for safety engineering and clinical practice. Alternatives are suggested for future studies.

Human Lumbar Spine Responses from Vertical Loading: Ranking of Forces Via Brier Score Metrics and Injury Risk Curves.

This study was conducted to quantify the human tolerance from inferior to superior impacts to whole lumbar spinal columns excised from 43 post mortem human subjects. The specimens were fixed at the ends, aligned in a consistent seated posture, load cells were attached to the proximal and distal ends of the fixation, and the impact was applied using a custom accelerator device. Pretest X-rays and computed tomography (CT) scans, prepositioned X-rays, and posttest X-rays, CT scans and dissection data were used to identify injuries. Right, left, and interval censoring processes were used for the survival analysis, 16 were right censored, 24 were interval censored, and three were left censored observations. Force-based injury risk curves were developed, and the optimal metric describing the underlying response to injury was identified using the Brier score metric. Material, geometry (disc and body areas), and demographic covariates were included in the analysis. The distal force was found to be optimal metric. The bone mineral density was a significant covariate for distal and proximal forces. Both material and geometrical factors affected the transmitted force in this mode of loading. These quantified data serve as the first set of human lumbar spinal column injury risk curves.

Human lumbar spinal column injury criteria from vertical loading at the base: Applications to military environments.

The objective of this study was to determine force-based lumbar spine injury criteria due to vertical impact using Post Mortem Human Surrogate (PMHS) experiments. Mounted personnel in military vehicles sustain loads from the pelvis in combat events such as underbody blast loadings. Forty-three post mortem human subject thoracolumbar spinal columns were obtained, screened for pre-existing trauma, bone mineral densities (BMDs) were determined, pre-test radiological images were taken, fixed at the ends in polymethylmethacrylate, load cells were attached to the ends of the fixation, positioned on custom vertical accelerator device based on a military-seating posture, and impacted at the base. Posttest images were obtained, and gross dissection was done to confirm injuries, classified into single and multilevel groups, groups A and B. Axial and resultant forces at the thoracolumbar (proximal) and lumbosacral (distal) joints were used as response variables to develop lumbar spine injury risk curves using parametric survival analysis. The Brier score metric was used to rank the variables. Age, BMD, column length, and vertebral body and intervertebral disc areas were used as covariates. The optimal metric describing the underlying response to injury was the distal resultant force for group A and proximal axial force for group B specimens. Force-BMD for group A and force-body area for group B were the best combinations. The IRCs with ±95% confidence intervals and quality of risk curves are given in the paper, and they serve as lumbar spine injury criteria. The present human cadaver Injury Risk Curves (IRCs) can be used to conduct matched pair tests to obtain dummy-based injury assessment risk curves/values to predict injury. The present IRCs can be used in human body finite element models. The relationship between covariates and primary forces presented in this study contribute to a better understanding of the role of demographic, geometric, and material factors to impact acceleration loading.

Application of resampling techniques to improve the quality of survival analysis risk curves for human frontal bone fracture.

BACKGROUND: In automotive events, head injuries (skull fractures and/or brain injuries) are associated with head contact loading. While the widely-used head injury criterion is based on frontal bone fracture and linear accelerations, injury risk curves were not developed from original datasets. OBJECTIVES: Develop skull fracture-based risk curves for using previously published data and apply resampling techniques to assess their qualities. METHODS: Force, deflection, energy, and stiffness data from thirteen human cadaver head impact tests were used to develop risk curves using parametric survival analysis. Injuries occurred to all specimens. Data points were treated as uncensored. Variables were ranked, and the variable best explaining the underlying fracture response was determined using the Brier Score Metric (BSM). The qualities of the risk curves were determined using normalized confidence interval sizes. Statistical resampling methods were used to assess the quality of the risk curves and the impact of the sample size by conducting 2000 simulations. Sample sizes ranged from 13 to 26. FINDINGS: The Weibull distribution was optimal for all the response variables, except deflection (log-logistic). The quality of the risk curves was the highest for deflection. This variable best explained the underlying head injury response, based on BSM. Improvements in the quality of the risk curves were achieved with additional samples of force and deflection (<13), while energy and stiffness variables required more size. Individual risk curves are given. INTERPRETATION: These probability curves from head contact loading add to the understanding skull fractures and can be used to improve safety in injury producing environments.

Pelvis injury risk curves in side impacts from human cadaver experiments using survival analysis and Brier score metrics.

Objectives: Post Mortem Human Surrogate (PMHS) experiments are used for describing tolerance and improve safety. For nearside impacts, the United States Standard Federal Motor Vehicle Safety Standards (FMVSS-214) used PMHS tests and binary regression methods to achieve these goals. Since this promulgation, Parametric Statistical Survival Modeling (PSSM) has become a de facto standard for developing injury risk curves (IRCs). This study is focused on pelvic injuries from side impacts. The objectives are as follows. Analyze impactor-based intact PMHS tests and develop IRCs at different AIS levels using the force metric and examine the effectiveness of other force-related variables on IRCs.Methods: Impactor-driven pelvic tests conducted using whole body PMHS were selected from published studies. The dataset had 63 tests. Peak force, 3-ms clip force, and impulse were used to develop IRCs for Abbreviated Injury Scores (AIS) AIS2+ and AIS3+, i.e., groups A and B. Brier Score Metric (BSM) was used for ranking metrics. 95% confidence intervals were computed, Normalized Confidence Interval Sizes (NCIS) were determined, and quality of the IRCs were obtained.Results: Impulse best described the underlying response of the pelvis. BSMs were the lowest for the impulse for both groups. At 10% and 50% probabilities, impulses were 71 Ns and 125 Ns for group A and 79 Ns and 160 Ns for group B; peak forces were 3.8 kN and 7.1 kN and 4 kN and 10 kN for groups A and B; and clip forces were 2.7 kN and 6.5 kN and 3.6 kN and 8.6 kN, for groups A and B. NCIS at discrete probability levels, qualities of risk curves, and individual IRCs are given.Conclusion: This study underscores the importance of using impulse to describe pelvis injury criteria in lateral impacts. These findings are applicable to anthropomorphic test devices, as matched pair tests are done to determine dummy-based injury criteria/injury assessment risk curves (IARCs). Although IRCs have been developed for WorldSID, it may be appropriate to use impulse-based IARCs. Because THOR is a potential device for automated vehicle environments, it may be appropriate to develop THOR-based IARCS. The present IRCs act as fundamental human-based injury criteria. These responses can also be used in human body and subsystem computational models.

Novel learning framework (knockoff technique) to evaluate metric ranking algorithms to describe human response to injury.

OBJECTIVE: The objective of this study is to present a novel framework, termed the knockoff technique, to evaluate different metric ranking algorithms to better describe human response to injury. METHODS: Many biomechanical metrics are routinely obtained from impact tests using postmortem human surrogates (PMHS) to develop injury risk curves (IRCs). The IRCs form the basis to evaluate human safety in crashworthiness environments. The biomechanical metrics should be chosen based on some measure of their predictive ability. Commonly used algorithms for the choice of ranking the metrics include (a) areas under the receiver operating characteristic curve (AUROC), time-varying AUROC, and other adaptations, and (b) some variants of predictive squared error loss. This article develops a rigorous framework to evaluate the metric selection/ranking algorithms. Actual experimental data are used due to the shortcoming of using simulated data. The knockoff data are meshed into existing experimental data using advanced statistical algorithms. Error rate measures such as false discovery rates (FDRs) and bias are calculated using the knockoff technique. Experimental data are used from previously published whole-body PMHS side impact sled tests. The experiments were conducted at different velocities, padding and rigid load wall conditions, and offsets and with different supplemental restraint systems. The PMHS specimens were subjected to a single lateral impact loading resulting in injury and noninjury outcomes. RESULTS: A total of 25 metrics were used from 42 tests. The AUROC-type algorithms tended to have higher FDRs compared to the squared error loss-type functions (45.3% for the best AUROC-type algorithms versus 31.4% for the best Brier score algorithm). Standard errors for the Brier score algorithm also tended to be lower, indicative of more stable metric choices and robust rankings. The wide variations observed in the performance of the algorithms demonstrated the need for data set-specific evaluation tools such as the knockoff technique developed in this study. CONCLUSIONS: In the present data set, the AUROCs and related binary classification algorithms led to inflated FDRs, rendering metric selection/ranking questionable. This is particularly true for data sets with a high proportion of censoring. Squared error loss-type algorithms (such as the Brier score algorithm or its modifications) improved the performance in the metric selection process. The presented new knockoff technique may wholly change how IRCs are developed from impact experiments or simulations. At the very least, the knockoff technique demonstrates the need for evaluations among different metric ranking/selection algorithms, especially when they produce substantially different biomechanical metric choices. Without recommending the AUROC-type or Brier score-type algorithms universally, the authors suggest careful assessments of these algorithms using the proposed framework, so that a robust algorithm may be chosen, with respect to the nature of the experimental data set. Though results are given for sets from a published series of experiments, the knockoff technique is being used by the authors in tests that are applicable to the automotive, aviation, military, and other environments.

Role of disc area and trabecular bone density on lumbar spinal column fracture risk curves under vertical impact.

While studies have been conducted using human cadaver lumbar spines to understand injury biomechanics in terms of stability/energy to fracture, and physiological responses under pure-moment/follower loads, data are sparse for inferior-to-superior impacts. Injuries occur under this mode from underbody blasts. OBJECTIVES: determine role of age, disc area, and trabecular bone density on tolerances/risk curves under vertical loading from a controlled group of specimens. T12-S1 columns were obtained, pretest X-rays and CTs taken, load cells attached to both ends, impacts applied at S1-end using custom vertical accelerator device, and posttest X-ray, CT, and dissections done. BMD of L2-L4 vertebrae were obtained from QCT. Survival analysis-based Human Injury Probability Curves (HIPCs) were derived using proximal and distal forces. Age, area, and BMD were covariates. Forces were considered uncensored, representing the load carrying capacity. The Akaike Information Criterion was used to determine optimal distributions. The mean forces, ±95% confidence intervals, and Normalized Confidence Interval Size (NCIS) were computed. The Lognormal distribution was the optimal function for both forces. Age, area, and BMD were not significant (p > 0.05) covariates for distal forces, while only BMD was significant for proximal forces. The NCIS was the lowest for force-BMD covariate HIPC. The HIPCs for both genders at 35 and 45 years were based on population BMDs. These HIPCs serve as human tolerance criteria for automotive, military, and other applications. In this controlled group of samples, BMD is a better predictor-covariate that characterizes lumbar column injury under inferior-to-superior impacts.

A Novel Competing Risk Analysis Model to Determine the Role of Cervical Lordosis in Bony and Ligamentous Injuries.

OBJECTIVE: To determine role of lordosis in cervical spine injuries using a novel competing risk analysis model. METHODS: The first subgroup of published experiments (n = 20) subjected upright human cadaver head-neck specimens to impact loading. The natural lordosis was removed. The second (n = 21) and third (n = 10) subgroups of published tests subjected inverted specimens to head impact loading. Lordosis was preserved in these 2 subgroups. Using axial force and age as variables, competing risks analysis techniques were used to determine the role of lordosis in the risk of bone-only, ligament-only, and bone and ligament injuries. RESULTS: Bony injuries were focused more at 1 level to a straightened spine, and ligament injuries were spread around multiple levels. Age was not a significant (P < 0.05) covariate. A straightened spine had 3.23 times higher risk of bony injuries than a lordotic spine. The spine with maintained lordosis had 1.14 times higher risk of ligament injuries, and 2.67 times higher risk of bone and ligament injuries than a spine without lordosis (i.e., preflexed column). CONCLUSIONS: Increased risk of bony injuries in a preflexed spine and ligament injuries in a lordotic spine may have implications for military personnel, as continuous use of helmets in the line of duty affects the natural curvature; astronauts, as curvatures are less lordotic after missions; and civilian patients with spondylotic myelopathy who use head protective devices, as curvatures may change over time in addition to the natural aging process.

Foot-ankle complex injury risk curves using calcaneus bone mineral density data.

OBJECTIVE: Biomechanical data from post mortem human subject (PMHS) experiments are used to derive human injury probability curves and develop injury criteria. This process has been used in previous and current automotive crashworthiness studies, Federal safety standards, and dummy design and development. Human bone strength decreases as the individuals reach their elderly age. Injury risk curves using the primary predictor variable (e.g., force) should therefore account for such strength reduction when the test data are collected from PMHS specimens of different ages (age at the time of death). This demographic variable is meant to be a surrogate for fracture, often representing bone strength as other parameters have not been routinely gathered in previous experiments. However, bone mineral densities (BMD) can be gathered from tested specimens (presented in this manuscript). The objective of this study is to investigate different approaches of accounting for BMD in the development of human injury risk curves. METHODS: Using simulated underbody blast (UBB) loading experiments conducted with the PMHS lower leg-foot-ankle complexes, a comparison is made between the two methods: treating BMD as a covariate and pre-scaling test data based on BMD. Twelve PMHS lower leg-foot-ankle specimens were subjected to UBB loads. Calcaneus BMD was obtained from quantitative computed tomography (QCT) images. Fracture forces were recorded using a load cell. They were treated as uncensored data in the survival analysis model which used the Weibull distribution in both methods. The width of the normalized confidence interval (NCIS) was obtained using the mean and ± 95% confidence limit curves. PRINCIPAL RESULTS: The mean peak forces of 3.9kN and 8.6kN were associated with the 5% and 50% probability of injury for the covariate method of deriving the risk curve for the reference age of 45 years. The mean forces of 5.4 kN and 9.2kN were associated with the 5% and 50% probability of injury for the pre-scaled method. The NCIS magnitudes were greater in the covariate-based risk curves (0.52-1.00) than in the risk curves based on the pre-scaled method (0.24-0.66). The pre-scaling method resulted in a generally greater injury force and a tighter injury risk curve confidence interval. Although not directly applicable to the foot-ankle fractures, when compared with the use of spine BMD from QCT scans to pre-scale the force, the calcaneus BMD scaled data produced greater force at the same risk level in general. CONCLUSIONS: Pre-scaling the force data using BMD is an alternate, and likely a more accurate, method instead of using covariate to account for the age-related bone strength change in deriving risk curves from biomechanical experiments using PMHS. Because of the proximity of the calcaneus bone to the impacting load, it is suggested to use and determine the BMD of the foot-ankle bone in future UBB and other loading conditions to derive human injury probability curves for the foot-ankle complex.