Surrogate endpoints in clinical trials of p16-positive squamous cell carcinoma of the oropharynx: an individual patient data meta-analysis.

BACKGROUND: The increased incidence of human papillomavirus (HPV)-related cancers has motivated efforts to optimise treatment for these patients with excellent prognosis. Validation of surrogates for overall survival could expedite the investigation of new therapies. We sought to evaluate candidate intermediate clinical endpoints in trials assessing definitive treatment of p16-positive oropharyngeal cancer with chemotherapy or radiotherapy. METHODS: We did a retrospective review of five multicentre, randomised trials (NRG/RTOG 9003, 0129, 0234, 0522, and 1016) that tested radiotherapy with or without chemotherapy in patients (aged ≥18 years) with p16-positive localised head or neck squamous-cell carcinomas. Eight intermediate clinical endpoints were considered as potential surrogates for overall survival: freedom from local progression, freedom from regional progression, freedom from distant metastasis, freedom from locoregional progression, freedom from any progression, locoregional progression-free survival, progression-free survival, and distant metastasis-free survival. We used a two-stage meta-analytical framework, which requires high correlation between the intermediate clinical endpoint and overall survival at the patient level (condition 1), and high correlation between the treatment effect on the intermediate clinical endpoint and the treatment effect on overall survival (condition 2). For both, an r2 greater than 0·7 was used as criteria for clinically relevant surrogacy. FINDINGS: We analysed 1373 patients with oropharyngeal cancer from May 9, 2020, to Nov 22, 2023. 1231 (90%) of patients were men, 142 (10%) were women, and 1207 (88%) were White, with a median age of 57 years (IQR 51-62). Median follow-up was 4·2 years (3·1-5·1). For the first condition, correlating the intermediate clinical endpoints with overall survival at the individual and trial level, the three composite endpoints of locoregional progression-free survival (Kendall's τ 0·91 and r2 0·72), distant metastasis-free survival (Kendall's τ 0·93 and r2 0·83), and progression-free survival (Kendall's τ 0·88 and r2 0·70) were highly correlated with overall survival at the patient level and at the trial-group level. For the second condition, correlating treatment effects of the intermediate clinical endpoints and overall survival, the composite endpoints of locoregional progression-free survival (r2 0·88), distant metastasis-free survival (r2 0·96), and progression-free survival (r2 0·92) remained strong surrogates. Treatment effects on the remaining intermediate clinical endpoints were less strongly correlated with overall survival. INTERPRETATION: We identified locoregional progression-free survival, distant metastasis-free survival, and progression-free survival as surrogates for overall survival in p16-positive oropharyngeal cancers treated with chemotherapy or radiotherapy, which could serve as clinical trial endpoints. FUNDING: NRG Oncology Operations, NRG Oncology SDMC, the National Cancer Institute, Eli Lilly, Aventis, and the University of Michigan.

Model-agnostic explanations for survival prediction models.

Advanced machine learning methods capable of capturing complex and nonlinear relationships can be used in biomedical research to accurately predict time-to-event outcomes. However, these methods have been criticized as "black boxes" that are not interpretable and thus are difficult to trust in making important clinical decisions. Explainable machine learning proposes the use of model-agnostic explainers that can be applied to predictions from any complex model. These explainers describe how a patient's characteristics are contributing to their prediction, and thus provide insight into how the model is arriving at that prediction. The specific application of these explainers to survival prediction models can be used to obtain explanations for (i) survival predictions at particular follow-up times, and (ii) a patient's overall predicted survival curve. Here, we present a model-agnostic approach for obtaining these explanations from any survival prediction model. We extend the local interpretable model-agnostic explainer framework for classification outcomes to survival prediction models. Using simulated data, we assess the performance of the proposed approaches under various settings. We illustrate application of the new methodology using prostate cancer data.

Perceptions on and roadblocks to implementation of standardized nomenclature in radiation oncology: A survey from TG-263U1.

PURPOSE: AAPM Task Group No. 263U1 (Update to Report No. 263 - Standardizing Nomenclatures in Radiation Oncology) disseminated a survey to receive feedback on utilization, gaps, and means to facilitate further adoption. METHODS: The survey was created by TG-263U1 members to solicit feedback from physicists, dosimetrists, and physicians working in radiation oncology. Questions on the adoption of the TG-263 standard were coupled with demographic information, such as clinical role, place of primary employment (e.g., private hospital, academic center), and size of institution. The survey was emailed to all AAPM, AAMD, and ASTRO members. RESULTS: The survey received 463 responses with 310 completed survey responses used for analysis, of whom most had the clinical role of medical physicist (73%) and the majority were from the United States (83%). There were 83% of respondents who indicated that they believe that having a nomenclature standard is important or very important and 61% had adopted all or portions of TG-263 in their clinics. For those yet to adopt TG-263, the staffing and implementation efforts were the main cause for delaying adoption. Fewer respondents had trouble adopting TG-263 for organs at risk (29%) versus target (44%) nomenclature. Common themes in written feedback were lack of physician support and available resources, especially in vendor systems, to facilitate adoption. CONCLUSIONS: While there is strong support and belief in the benefit of standardized nomenclature, the widespread adoption of TG-263 has been hindered by the effort needed by staff for implementation.  Feedback from the survey is being utilized to drive the focus of the update efforts and create tools to facilitate easier adoption of TG-263.

Baseline Characteristics of PATHWEIGH: A Stepped-Wedge Cluster Randomized Study for Weight Management in Primary Care.

PURPOSE: To describe the characteristics of patients and practice of clinicians during standard-of-care for weight management in a large, multiclinic health system before the implementation of PATHWEIGH, a pragmatic weight management intervention. METHODS: We analyzed baseline characteristics of patients, clinicians, and clinics during standard-of-care for weight management before the implementation of PATHWEIGH, which will be evaluated for effectiveness and implementation in primary care using an effectiveness-implementation hybrid type-1 cluster randomized stepped-wedge clinical trial design. A total of 57 primary care clinics were enrolled and randomized to 3 sequences. Patients included in the analysis met the eligibility requirements of age ≥18 years and body mass index (BMI) ≥25 kg/m2 and had a weight-prioritized visit (defined a priori) during the period March 17, 2020 to March 16, 2021. RESULTS: A total of 12% of patients aged ≥18 years and with a BMI ≥25 kg/m2 seen in the 57 practices during the baseline period (n = 20,383) had a weight-prioritized visit. The 3 randomization sequences of 20, 18, and 19 sites were similar, with an overall mean patient age of 52 (SD 16) years, 58% women, 76% non-Hispanic White patients, 64% with commercial insurance, and with a mean BMI of 37 (SD 7) kg/m2. Documented referral for anything weight related was low (<6%), and 334 prescriptions of an antiobesity drug were noted. CONCLUSIONS: Of patients aged ≥18 years and with a BMI ≥25 kg/m2 in a large health system, 12% had a weight-prioritized visit during the baseline period. Despite most patients being commercially insured, referral to any weight-related service or prescription of antiobesity drug was uncommon. These results fortify the rationale for trying to improve weight management in primary care.

Baseline Characteristics of PATHWEIGH: A Stepped-Wedge Cluster Randomized Study for Weight Management in Primary Care.

Context: Despite the fact that obesity is both treatable and preventable, treating the comorbidities, rather than obesity per se remains the mainstay of therapy. Objective: To evaluate the efficacy and implementation of a pragmatic approach to weight management in primary care that prioritizes treatment of weight rather than weight-related diseases (PATHWEIGH). Study Design and Analysis: PATHWEIGH is a hybrid type 1 cluster randomized stepped wedge clinical trial. Clinics were enrolled and randomized to three sequences using covariate constrained randomization. Descriptive statistics were used to summarize clinic and patient characteristics with t-tests, Wilcoxon rank sums or Fisher's exact tests used to compare groups. Setting: Fifty-seven primary care clinics in rural, suburban and urban Colorado in a single healthcare system were utilized. Population Studied: Patients age >18 years and body mass index (BMI) >25 kg/m2 who had a weight-prioritized visit (WPV) in the prior year were enrolled. A WPV was defined as a chief complaint or reason for visit that included "weight", ICD-10 codes for weight or use of an intake questionnaire for weight. Intervention: None. This abstract describes the baseline (pre-intervention) characteristics of the clinics and patients treated with standard-of-care (SOC) for weight management. Outcome Measures: Baseline characteristics of the clinics and patients undergoing a WPV from March 17, 2020 - March 16, 2021. Results: 20,410 patients met these eligibility requirements representing 12% of patients >18 years and body mass index (BMI) >25 kg/m2 seen at the clinic during this baseline period. The three randomization sequences of 20, 18, and 19 sites were similar with an overall median age of 53 years (IQR: 39-65), 58% women, 76% non-Hispanic whites, 64% commercial insurance, and median BMI of 36 kg/m2 (IQR: 32-41). No sequence differences were seen for vital signs, relevant laboratory values, or numbers of comorbidities or medications that cause weight loss or weight gain. Referral for anything weight-related was low (<6%) and only 334 prescriptions of an anti-obesity medication were noted. Conclusions: Of patients >18 years and body mass index (BMI) >25 kg/m2 seen in the 57 primary care clinics, 12% had a weight-prioritized visit during the baseline period. Despite most being commercially insured, referral to any weight-related service or prescription of anti-obesity medication was uncommon.

Hyperglycemia and Other Glycemic Measures Throughout Therapy for Pediatric Acute Lymphoblastic Leukemia and Lymphoma.

Transient hyperglycemia during induction chemotherapy is associated with increased morbidity and mortality in patients with acute lymphoblastic leukemia (ALL). Treatment with glucocorticoids, asparaginase, and stress are the proposed causal factors. Although these risks are not exclusive to induction, glycemic control throughout the remainder of ALL/lymphoma (ALL/ALLy) therapy has not been described. Furthermore, prior research has been limited to transient hyperglycemia. This study aimed to characterize glycemic control throughout ALL/ALLy and to evaluate risk factors and outcomes associated with increased mean glucose and glucose coefficient of variation (glucose CV) during induction chemotherapy. The records for 220 pediatric/young adult patients, age 1 to 26 years, who underwent treatment for ALL/ALLy from 2010 to 2014 at Children's Hospital Colorado were retrospectively reviewed. Measures of glycemic control were calculated for each cycle. For the cycle with the highest mean glucose, induction (n=208), multivariable models were performed to identify potential risk factors and consequences of increased glucose. Highest mean glucose by cycle were induction 116 mg/dL, pretreatment 108 mg/dL, delayed intensification 96 mg/dL, and maintenance 93 mg/dL; these cycles also had the most glycemic variability. During induction, patients with Down syndrome, or who were ≥12 years and overweight/obese, had higher mean glucoses; age and overweight/obese status were each associated with increased glucose CV. In multivariable analysis, neither induction mean glucose nor glucose CV were associated with increased hazard of infection, relapse, or death.

The Impact of Altitude at Birth on Perinatal Respiratory Support for Neonates with Trisomy 21.

OBJECTIVE: Both high altitude and trisomy 21 (T21) status can negatively impact respiratory outcomes. The objective of this study was to examine the association between altitude and perinatal respiratory support in neonates with T21 compared with those without T21. STUDY DESIGN: This retrospective cohort study used the United States all-county natality files that included live, singleton, in-hospital births from 2015 to 2019. Descriptive statistics for neonates with and without the primary outcome of sustained assisted ventilation (>6 hours) were compared using t-tests and Chi-squared analyses. Multivariable logistic regression was used to determine the association between respiratory support and the presence of T21, and included an interaction term to determine whether the association between respiratory support and the presence of T21 was modified by elevation at delivery. RESULTS: A total of 17,939,006 neonates, 4,059 (0.02%) with T21 and 17,934,947 (99.98%) without, were included in the study. The odds of requiring sustained respiratory support following delivery were 5.95 (95% confidence interval [CI]: 5.31, 6.66), 4.06 (95% CI: 2.39, 6.89), 2.36 (95% CI: 1.64, 3.40), and 5.04 (95% CI: 1.54, 16.54) times as high for neonates with T21 than without T21 when born at low, medium, high, and very high elevations, respectively. The odds of requiring immediate ventilation support following delivery were 5.01 (95% CI: 4.59, 5.46), 5.90 (95% CI: 4.16, 8.36), 2.86 (95% CI: 2.15, 3.80), and 12.08 (95% CI: 6.78, 21.51) times as high for neonates with T21 than without T21 when born at low, medium, high, and very high elevation, respectively. CONCLUSION: Neonates with T21 have increased odds of requiring respiratory support following delivery when compared with neonates without T21 at all categories of altitude. However, the odds ratios did not increase monotonically with altitude which indicates additional research is critical in understanding the effects of altitude on neonates with T21. KEY POINTS: · Neonates with T21 have an increased need for perinatal respiratory support at all altitudes.. · The odds of needing perinatal respiratory support did not increase monotonically with elevation.. · Additional research is critical to understanding the effects of altitude on neonates with T21..

De-implementation of Axillary Dissection in Women With Breast Cancer is Largely Driven By Site-level Contextual Effects.

OBJECTIVE: To assess the contribution of unknown institutional factors (contextual effects) in the de-implementation of cALND in women with breast cancer. SUMMARY OF BACKGROUND DATA: Women included in the National Cancer Database with invasive breast carcinoma from 2012 to 2016 that underwent upfront lumpectomy and were found to have a positive sentinel node. METHODS: A multivariable mixed effects logistic regression model with a random intercept for site was used to determine the effect of patient, tumor, and institutional variables on the risk of cALND. Reference effect measureswere used to describe and compare the contribution of contextual effects to the variation in cALND use to that of measured variables. RESULTS: By 2016, cALND was still performed in at least 50% of the patients in a quarter of the institutions. Black race, younger women and those with larger or hormone negative tumors were more likely to undergo cALND. However, the width of the 90% reference effect measures range for the contextual effects exceeded that of the measured site, tumor, time, and patient demographics, suggesting institutional contextual effects were the major drivers of cALND de-implementation. For instance, a woman at an institution with low-risk of performing cALND would have 74% reduced odds of havinga cALND than if she was treated at a median-risk institution, while a patient at a high-risk institution had 3.91 times the odds. CONCLUSION: Compared to known patient, tumor, and institutional factors, contextual effects had a higher contribution to the variation in cALND use.

A Gaussian copula approach for dynamic prediction of survival with a longitudinal biomarker.

Dynamic prediction uses patient information collected during follow-up to produce individualized survival predictions at given time points beyond treatment or diagnosis. This allows clinicians to obtain updated predictions of a patient's prognosis that can be used in making personalized treatment decisions. Two commonly used approaches for dynamic prediction are landmarking and joint modeling. Landmarking does not constitute a comprehensive probability model, and joint modeling often requires strong distributional assumptions and computationally intensive methods for estimation. We introduce an alternative approximate approach for dynamic prediction that aims to overcome the limitations of both methods while achieving good predictive performance. We separately specify the marker and failure time distributions conditional on surviving up to a prediction time of interest and use standard variable selection and goodness-of-fit techniques to identify the best-fitting models. Taking advantage of its analytic tractability and easy two-stage estimation, we use a Gaussian copula to link the marginal distributions smoothly at each prediction time with an association function. With simulation studies, we examine the proposed method's performance. We illustrate its use for dynamic prediction in an application to predicting death for heart valve transplant patients using longitudinal left ventricular mass index information.

Survival prediction models: an introduction to discrete-time modeling.

BACKGROUND: Prediction models for time-to-event outcomes are commonly used in biomedical research to obtain subject-specific probabilities that aid in making important clinical care decisions. There are several regression and machine learning methods for building these models that have been designed or modified to account for the censoring that occurs in time-to-event data. Discrete-time survival models, which have often been overlooked in the literature, provide an alternative approach for predictive modeling in the presence of censoring with limited loss in predictive accuracy. These models can take advantage of the range of nonparametric machine learning classification algorithms and their available software to predict survival outcomes. METHODS: Discrete-time survival models are applied to a person-period data set to predict the hazard of experiencing the failure event in pre-specified time intervals. This framework allows for any binary classification method to be applied to predict these conditional survival probabilities. Using time-dependent performance metrics that account for censoring, we compare the predictions from parametric and machine learning classification approaches applied within the discrete time-to-event framework to those from continuous-time survival prediction models. We outline the process for training and validating discrete-time prediction models, and demonstrate its application using the open-source R statistical programming environment. RESULTS: Using publicly available data sets, we show that some discrete-time prediction models achieve better prediction performance than the continuous-time Cox proportional hazards model. Random survival forests, a machine learning algorithm adapted to survival data, also had improved performance compared to the Cox model, but was sometimes outperformed by the discrete-time approaches. In comparing the binary classification methods in the discrete time-to-event framework, the relative performance of the different methods varied depending on the data set. CONCLUSIONS: We present a guide for developing survival prediction models using discrete-time methods and assessing their predictive performance with the aim of encouraging their use in medical research settings. These methods can be applied to data sets that have continuous time-to-event outcomes and multiple clinical predictors. They can also be extended to accommodate new binary classification algorithms as they become available. We provide R code for fitting discrete-time survival prediction models in a github repository.

A Pipeline for the Implementation and Visualization of Explainable Machine Learning for Medical Imaging Using Radiomics Features.

Machine learning (ML) models have been shown to predict the presence of clinical factors from medical imaging with remarkable accuracy. However, these complex models can be difficult to interpret and are often criticized as "black boxes". Prediction models that provide no insight into how their predictions are obtained are difficult to trust for making important clinical decisions, such as medical diagnoses or treatment. Explainable machine learning (XML) methods, such as Shapley values, have made it possible to explain the behavior of ML algorithms and to identify which predictors contribute most to a prediction. Incorporating XML methods into medical software tools has the potential to increase trust in ML-powered predictions and aid physicians in making medical decisions. Specifically, in the field of medical imaging analysis the most used methods for explaining deep learning-based model predictions are saliency maps that highlight important areas of an image. However, they do not provide a straightforward interpretation of which qualities of an image area are important. Here, we describe a novel pipeline for XML imaging that uses radiomics data and Shapley values as tools to explain outcome predictions from complex prediction models built with medical imaging with well-defined predictors. We present a visualization of XML imaging results in a clinician-focused dashboard that can be generalized to various settings. We demonstrate the use of this workflow for developing and explaining a prediction model using MRI data from glioma patients to predict a genetic mutation.

Predicting olfactory loss in chronic rhinosinusitis using machine learning.

OBJECTIVE: Compare machine learning (ML)-based predictive analytics methods to traditional logistic regression in classification of olfactory dysfunction in chronic rhinosinusitis (CRS-OD) and identify predictors within a large multi-institutional cohort of refractory CRS patients. METHODS: Adult CRS patients enrolled in a prospective, multi-institutional, observational cohort study were assessed for baseline CRS-OD using a smell identification test (SIT) or brief SIT (bSIT). Four different ML methods were compared to traditional logistic regression for classification of CRS normosmics versus CRS-OD. RESULTS: Data were collected for 611 study participants who met inclusion criteria between 2011 April and 2015 July. Thirty-four percent of enrolled patients demonstrated olfactory loss on psychophysical testing. Differences between CRS normosmics and those with smell loss included objective disease measures (CT and endoscopy scores), age, sex, prior surgeries, socioeconomic status, steroid use, polyp presence, asthma, and aspirin sensitivity. Most ML methods performed favorably in terms of predictive ability. Top predictors include factors previously reported in the literature, as well as several socioeconomic factors. CONCLUSION: Olfactory dysfunction is a variable phenomenon in CRS patients. ML methods perform well compared to traditional logistic regression in classification of normosmia versus smell loss in CRS, and are able to include numerous risk factors into prediction models. Several actionable features were identified as risk factors for CRS-OD. These results suggest that ML methods may be useful for current understanding and future study of hyposmia secondary to sinonasal disease, the most common cause of persistent olfactory loss in the general population.

A copula-based approach for dynamic prediction of survival with a binary time-dependent covariate.

Dynamic prediction methods incorporate longitudinal biomarker information to produce updated, more accurate predictions of conditional survival probability. There are two approaches for obtaining dynamic predictions: (1) a joint model of the longitudinal marker and survival process, and (2) an approximate approach that specifies a model for a specific component of the joint distribution. In the case of a binary marker, an illness-death model is an example of a joint modeling approach that is unified and produces consistent predictions. However, previous literature has shown that approximate approaches, such as landmarking, with additional flexibility can have good predictive performance. One such approach proposes using a Gaussian copula to model the joint distribution of conditional continuous marker and survival distributions. It has the advantage of specifying established, flexible models for the marginals for which goodness-of-fit can be assessed, and has easy estimation that can be implemented in standard software. In this article, we provide a Gaussian copula approach for dynamic prediction to accommodate a binary marker using a continuous latent variable formulation. We compare the predictive performance of this approach to joint modeling and landmarking using simulations and demonstrate its use for obtaining dynamic predictions in an application to a prostate cancer study.

Random survival forests for dynamic predictions of a time-to-event outcome using a longitudinal biomarker.

BACKGROUND: Risk prediction models for time-to-event outcomes play a vital role in personalized decision-making. A patient's biomarker values, such as medical lab results, are often measured over time but traditional prediction models ignore their longitudinal nature, using only baseline information. Dynamic prediction incorporates longitudinal information to produce updated survival predictions during follow-up. Existing methods for dynamic prediction include joint modeling, which often suffers from computational complexity and poor performance under misspecification, and landmarking, which has a straightforward implementation but typically relies on a proportional hazards model. Random survival forests (RSF), a machine learning algorithm for time-to-event outcomes, can capture complex relationships between the predictors and survival without requiring prior specification and has been shown to have superior predictive performance. METHODS: We propose an alternative approach for dynamic prediction using random survival forests in a landmarking framework. With a simulation study, we compared the predictive performance of our proposed method with Cox landmarking and joint modeling in situations where the proportional hazards assumption does not hold and the longitudinal marker(s) have a complex relationship with the survival outcome. We illustrated the use of the RSF landmark approach in two clinical applications to assess the performance of various RSF model building decisions and to demonstrate its use in obtaining dynamic predictions. RESULTS: In simulation studies, RSF landmarking outperformed joint modeling and Cox landmarking when a complex relationship between the survival and longitudinal marker processes was present. It was also useful in application when there were several predictors for which the clinical relevance was unknown and multiple longitudinal biomarkers were present. Individualized dynamic predictions can be obtained from this method and the variable importance metric is useful for examining the changing predictive power of variables over time. In addition, RSF landmarking is easily implementable in standard software and using suggested specifications requires less computation time than joint modeling. CONCLUSIONS: RSF landmarking is a nonparametric, machine learning alternative to current methods for obtaining dynamic predictions when there are complex or unknown relationships present. It requires little upfront decision-making and has comparable predictive performance and has preferable computational speed.

The Experience of Primary Caregivers of Undocumented Immigrants with End-Stage Kidney Disease that Rely on Emergency-Only Hemodialysis.

BACKGROUND: Undocumented immigrants with end-stage kidney disease (ESKD) who rely on emergency-only hemodialysis (dialysis only after an emergency department evaluation) face psychosocial distress. Emergency-only hemodialysis (EOHD) is likely burdensome for primary caregivers as well. OBJECTIVE: To understand the experience of primary caregivers of undocumented immigrants with ESKD who rely on emergency-only hemodialysis. DESIGN, SETTING, AND PARTICIPANTS: A qualitative, semi-structured interview study to assess the experiences of primary caregivers of undocumented immigrants with ESKD at a safety-net hospital in Denver, Colorado from June 28 to November 15, 2018. Applied thematic analysis was used to analyze interviews. MAIN OUTCOMES AND MEASURES: Themes and subthemes. RESULTS: Twenty primary caregiver participants had a mean (SD) age of 46 (17), 13 (65%) were female, 7 (35%) were in an adult child caregiver role, and 13 (65%) were spouses. Five themes and 17 subthemes (in parentheses) were identified: (1) Caregiver role (providing emotional, physical, and economic support, advocacy and care navigation), (2) Caregiver burden (anxiety related to patient and personal death, emotional exhaustion and personal illness, struggle with finances, self-care and redefining relationship), (3) Unpredictable EOHD (acute episodes of illness that trigger emergency, stress when patient is denied dialysis, impact on work and sleep, and emotional relief after a session of EOHD), (4) Effect on children (dropping out or missing school, psychosocial distress, children assuming caregiver responsibilities, and juggling multi-generational caregiving of children), (5) Faith and appreciation (comfort in God and appreciation of healthcare). CONCLUSIONS AND RELEVANCE: Caregivers of undocumented immigrants with ESKD who rely upon EOHD experience caregiver burden and distress. The impact of EOHD on caregivers should be considered when assessing the consequences of excluding undocumented immigrants from public insurance programs.

Risk factors associated with adverse maternal outcomes following intrapartum cesarean birth: a secondary analysis of the WHO global survey on maternal and perinatal health, 2004-2008.

BACKGROUND: To identify risk factors associated with a composite adverse maternal outcomes in women undergoing intrapartum cesarean birth. METHODS: We used the facility-based, multi-country, cross-sectional WHO Global Survey of Maternal and Perinatal Health (2004-2008) to examine associations between woman-, labor/obstetric-, and facility-level characteristics and a composite adverse maternal outcome of postpartum morbidity and mortality. This analysis was performed among women who underwent intrapartum cesarean birth during the course of labor. RESULTS: We analyzed outcomes of 29,516 women from low- and middle-income countries who underwent intrapartum cesarean birth between the gestational ages of 24 and 43 weeks, 3.5% (1040) of whom experienced the composite adverse maternal outcome. In adjusted analyses, factors associated with a decreased risk of the adverse maternal outcome associated with intrapartum cesarean birth included having four or more antenatal visits (AOR 0.60; 95% CI: 0.43-0.84; p = 0.003), delivering in a medium- or high-human development index country (vs. low-human development index country: AOR 0.07; 95% CI: 0.01-0.85 and AOR 0.02; 95% CI: 0.001-0.39, respectively; p = 0.03), and malpresentation (vs. cephalic: breech AOR 0.52; CI: 0.31-0.87; p = 0.04). Women who were medically high risk (vs. not medically high risk: AOR 1.81; CI: 1.30-2.51, p < 0.0004), had less education (0-6 years) (vs. 13+ years; AOR 1.64; CI: 1.03-2.63; p = 0.01), were obstetrically high risk (vs. not high risk; AOR 3.67; CI: 2.58-5.23; p < 0.0001), or had a maternal or obstetric indication (vs. elective: AOR 4.74; CI: 2.36-9.50; p < 0.0001) had increased odds of the adverse outcome. CONCLUSION: We found reduced adverse maternal outcomes of intrapartum cesarean birth in women with ≥ 4 antenatal visits, those who delivered in a medium or high human development index country, and those with malpresenting fetuses. Maternal adverse outcomes associated with intrapartum cesarean birth were medically and obstetrically high risk women, those with less education, and those with a maternal or obstetric indication for intrapartum cesarean birth.

Comparison of joint modeling and landmarking for dynamic prediction under an illness-death model.

Dynamic prediction incorporates time-dependent marker information accrued during follow-up to improve personalized survival prediction probabilities. At any follow-up, or "landmark", time, the residual time distribution for an individual, conditional on their updated marker values, can be used to produce a dynamic prediction. To satisfy a consistency condition that links dynamic predictions at different time points, the residual time distribution must follow from a prediction function that models the joint distribution of the marker process and time to failure, such as a joint model. To circumvent the assumptions and computational burden associated with a joint model, approximate methods for dynamic prediction have been proposed. One such method is landmarking, which fits a Cox model at a sequence of landmark times, and thus is not a comprehensive probability model of the marker process and the event time. Considering an illness-death model, we derive the residual time distribution and demonstrate that the structure of the Cox model baseline hazard and covariate effects under the landmarking approach do not have simple form. We suggest some extensions of the landmark Cox model that should provide a better approximation. We compare the performance of the landmark models with joint models using simulation studies and cognitive aging data from the PAQUID study. We examine the predicted probabilities produced under both methods using data from a prostate cancer study, where metastatic clinical failure is a time-dependent covariate for predicting death following radiation therapy.

A prediction model for colon cancer surveillance data.

Dynamic prediction models make use of patient-specific longitudinal data to update individualized survival probability predictions based on current and past information. Colonoscopy (COL) and fecal occult blood test (FOBT) results were collected from two Australian surveillance studies on individuals characterized as high-risk based on a personal or family history of colorectal cancer. Motivated by a Poisson process, this paper proposes a generalized nonlinear model with a complementary log-log link as a dynamic prediction tool that produces individualized probabilities for the risk of developing advanced adenoma or colorectal cancer (AAC). This model allows predicted risk to depend on a patient's baseline characteristics and time-dependent covariates. Information on the dates and results of COLs and FOBTs were incorporated using time-dependent covariates that contributed to patient risk of AAC for a specified period following the test result. These covariates serve to update a person's risk as additional COL, and FOBT test information becomes available. Model selection was conducted systematically through the comparison of Akaike information criterion. Goodness-of-fit was assessed with the use of calibration plots to compare the predicted probability of event occurrence with the proportion of events observed. Abnormal COL results were found to significantly increase risk of AAC for 1 year following the test. Positive FOBTs were found to significantly increase the risk of AAC for 3 months following the result. The covariates that incorporated the updated test results were of greater significance and had a larger effect on risk than the baseline variables.

The new injury severity score underestimates true injury severity in a resource-constrained setting.

Background: The new injury severity score (NISS) is widely used within trauma outcomes research. NISS is a composite anatomic severity score derived from the Abbreviated Injury Scale (AIS) protocol. It has been postulated that NISS underestimates trauma severity in resource-constrained settings, which may contribute to erroneous research conclusions. We formally compare NISS to an expert panel's assessment of injury severity in South Africa. Methods: This was a retrospective chart review of adult trauma patients seen in a tertiary trauma center. Randomly selected medical records were reviewed by an AIS-certified rater who assigned an AIS severity score for each anatomic injury. A panel of five South African trauma experts independently reviewed the same charts and assigned consensus severity scores using a similar scale for comparability. NISS was calculated as the sum of the squares of the three highest assigned severity scores per patient. The difference in average NISS between rater and expert panel was assessed using a multivariable linear mixed effects regression adjusted for patient demographics, injury mechanism and type. Results: Of 49 patients with 190 anatomic injuries, the majority were male (n = 38), the average age was 36 (range 18-80), with either a penetrating (n = 23) or blunt (n = 26) injury, resulting in 4 deaths. Mean NISS was 16 (SD 15) for the AIS rater compared to 28 (SD 20) for the expert panel. Adjusted for potential confounders, AIS rater NISS was on average 11 points (95 % CI: 7, 15) lower than the expert panel NISS (p < 0.001). Injury type was an effect modifier, with the difference between the AIS rater and expert panel being greater in penetrating versus blunt injury (16 vs. 7; p = 0.04). Crush injury was not well-captured by AIS protocol. Conclusion: NISS may under-estimate the 'true' injury severity in a middle-income country trauma hospital, particularly for patients with penetrating injury.