Efficient nonparametric inference on the effects of stochastic interventions under two-phase sampling, with applications to vaccine efficacy trials.

The advent and subsequent widespread availability of preventive vaccines has altered the course of public health over the past century. Despite this success, effective vaccines to prevent many high-burden diseases, including human immunodeficiency virus (HIV), have been slow to develop. Vaccine development can be aided by the identification of immune response markers that serve as effective surrogates for clinically significant infection or disease endpoints. However, measuring immune response marker activity is often costly, which has motivated the usage of two-phase sampling for immune response evaluation in clinical trials of preventive vaccines. In such trials, the measurement of immunological markers is performed on a subset of trial participants, where enrollment in this second phase is potentially contingent on the observed study outcome and other participant-level information. We propose nonparametric methodology for efficiently estimating a counterfactual parameter that quantifies the impact of a given immune response marker on the subsequent probability of infection. Along the way, we fill in theoretical gaps pertaining to the asymptotic behavior of nonparametric efficient estimators in the context of two-phase sampling, including a multiple robustness property enjoyed by our estimators. Techniques for constructing confidence intervals and hypothesis tests are presented, and an open source software implementation of the methodology, the txshift R package, is introduced. We illustrate the proposed techniques using data from a recent preventive HIV vaccine efficacy trial.

Prediction of VRC01 neutralization sensitivity by HIV-1 gp160 sequence features.

The broadly neutralizing antibody (bnAb) VRC01 is being evaluated for its efficacy to prevent HIV-1 infection in the Antibody Mediated Prevention (AMP) trials. A secondary objective of AMP utilizes sieve analysis to investigate how VRC01 prevention efficacy (PE) varies with HIV-1 envelope (Env) amino acid (AA) sequence features. An exhaustive analysis that tests how PE depends on every AA feature with sufficient variation would have low statistical power. To design an adequately powered primary sieve analysis for AMP, we modeled VRC01 neutralization as a function of Env AA sequence features of 611 HIV-1 gp160 pseudoviruses from the CATNAP database, with objectives: (1) to develop models that best predict the neutralization readouts; and (2) to rank AA features by their predictive importance with classification and regression methods. The dataset was split in half, and machine learning algorithms were applied to each half, each analyzed separately using cross-validation and hold-out validation. We selected Super Learner, a nonparametric ensemble-based cross-validated learning method, for advancement to the primary sieve analysis. This method predicted the dichotomous resistance outcome of whether the IC50 neutralization titer of VRC01 for a given Env pseudovirus is right-censored (indicating resistance) with an average validated AUC of 0.868 across the two hold-out datasets. Quantitative log IC50 was predicted with an average validated R2 of 0.355. Features predicting neutralization sensitivity or resistance included 26 surface-accessible residues in the VRC01 and CD4 binding footprints, the length of gp120, the length of Env, the number of cysteines in gp120, the number of cysteines in Env, and 4 potential N-linked glycosylation sites; the top features will be advanced to the primary sieve analysis. This modeling framework may also inform the study of VRC01 in the treatment of HIV-infected persons.

Acute disseminated encephalomyelitis: prognostic value of early follow-up brain MRI.

Patients with acute disseminated encephalomyelitis (ADEM) are presumed to have radiological monophasic disease, but this is uncertain since follow-up brain MRI is not routinely performed. We aimed to ascertain combined radiological and clinical monophasic disease in ADEM patients and to assess whether performing early (<6 months) follow-up brain MRI has prognostic value for subsequent multiphasic disease. We retrospectively studied the medical records of patients initially diagnosed with ADEM (years 2000-2014) at the Massachusetts General Hospital, USA. A neuroimaging specialist, masked to clinical events, reviewed all available brain MRIs. We included 62 patients (25 male; 30 pediatric; median clinical follow-up 3 years) and classified them into two subgroups: (1) clinically monophasic (no new, recurrent or worsening neurological symptoms >3 months after onset) (n = 45), and (2) clinically multiphasic (clinical relapse >3 months after onset) (n = 17). All clinically monophasic patients with brain MRI follow-up (n = 30) also had radiological monophasic disease a median of 2 years after ADEM onset. New lesions (58 vs. 14%) and persistent lesions (100 vs. 18%) on early brain MRI [available in 40 patients (65%)], as well as clinical flares (53 vs. 20%), were more common in clinically multiphasic versus monophasic patients. These early follow-up data allowed us to predict multiphasic disease with reasonable accuracy in a multivariable model (AUC = 0.73). We conclude that performing early follow-up brain MRI routinely in ADEM patients would aid clinicians in predicting multiphasic disease and may stratify patients who would benefit from initiation of disease-modifying therapy for multiple sclerosis.

Acute disseminated encephalomyelitis in 228 patients: A retrospective, multicenter US study.

OBJECTIVE: To analyze the range of demographic, clinical, MRI, and CSF features of acute disseminated encephalomyelitis (ADEM), a rare, typically monophasic demyelinating disorder, and analyze long-term outcomes including time and risk factors for subsequent clinical events as well as competing diagnoses. METHODS: We performed a retrospective, multicenter study in 4 US academic medical centers of all patients clinically diagnosed with ADEM. Initial presentation of pediatric and adult ADEM and monophasic and multiphasic disease were compared. The Aalen-Johansen estimator was used to produce estimates of the probability of transitioning to a multiphasic diagnosis as a function of time since initial diagnosis, treating death and alternative diagnoses as competing risks. RESULTS: Of 228 patients (122 children, age range 1-72 years, 106 male, median follow-up 24 months [25th-75th percentile 6-67], 7 deaths), approximately one quarter (n = 55, 24%) experienced at least one relapse. Relapsing disease in children was more often diagnosed as multiphasic ADEM than in adults (58% vs 21%, p = 0.007), in whom MS was diagnosed more often. Encephalopathy at initial presentation (hazard ratio [HR] 0.383, p = 0.001), male sex (HR 0.394, p = 0.002), and increasing age at onset (HR 0.984, p = 0.035) were independently associated with a longer time to a demyelinating disease relapse in a multivariable model. In 17 patients, diagnoses other than demyelinating disease were concluded in long-term follow-up. CONCLUSIONS: Relapsing disease after ADEM is fairly common and associated with a few potentially predictive features at initial presentation. Age-specific guidelines for ADEM diagnosis and treatment may be valuable, and vigilance for other, mostly rare, diseases is imperative.

Potential Influence of Advance Care Planning and Palliative Care Consultation on ICU Costs for Patients With Chronic and Serious Illness.

OBJECTIVES: To estimate the potential ICU-related cost savings if in-hospital advance care planning and ICU-based palliative care consultation became standard of care for patients with chronic and serious illness. DESIGN AND SETTING: Decision analysis using literature estimates and inpatient administrative data from Premier. PATIENTS: Patients with chronic, life-limiting illness admitted to a hospital within the Premier network. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Using Premier data (2008-2012), ICU resource utilization and costs were tracked over a 1-year time horizon for 2,097,563 patients with chronic life-limiting illness. Using a Markov microsimulation model, we explored the potential cost savings from the hospital system perspective under a variety of scenarios by varying the interventions' efficacies and availabilities. Of 2,097,563 patients, 657,825 (31%) used the ICU during the 1-year time horizon; mean ICU spending per patient was 11.3k (SD, 17.6k). In the base-case analysis, if in-hospital advance care planning and ICU-based palliative care consultation were systematically provided, we estimated a mean reduction in ICU costs of 2.8k (SD, 14.5k) per patient and an ICU cost saving of 25%. Among the simulated patients who used the ICU, the receipt of both interventions could have resulted in ICU cost savings of 1.9 billion, representing a 6% reduction in total hospital costs for these patients. CONCLUSIONS: In-hospital advance care planning and palliative care consultation have the potential to result in significant cost savings. Studies are needed to confirm these findings, but our results provide guidance for hospitals and policymakers.

Variation in tracheal reintubations among patients undergoing cardiac surgery across Washington state hospitals.

OBJECTIVES: The objectives of this study were to examine the variation in reintubations across Washington state hospitals that perform cardiac surgery, and explore hospital and patient characteristics associated with variation in reintubation. DESIGN: Retrospective cohort study. SETTING: All nonfederal hospitals performing cardiac surgery in Washington state. PARTICIPANTS: A total of 15,103 patients undergoing coronary artery bypass grafting or valvular surgery between January 1, 2008 and September 30, 2011. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Patient and hospital characteristics were compared between hospitals that had a reintubation frequency ≥5% or<5%. Multivariate logistic regression was used to compare the odds of reintubation across the hospitals. The authors tested for heterogeneity of odds of reintubation across hospitals by performing a likelihood ratio test on the hospital factor. After adjusting for patient-level characteristics and procedure type, significant heterogeneity in reintubations across hospitals was present (p = 0.005). This exploratory analyses suggested that hospitals with lower reintubations were more likely to have more acute care days and teaching intensive care units (ICU). CONCLUSIONS: After accounting for patient and procedure characteristics, significant heterogeneity in the relative odds of requiring reintubation was present across 16 nonfederal hospitals performing cardiac surgery in Washington state. The findings suggested that greater hospital volume and ICU teaching status were associated with fewer reintubations.

End-of-life expenditure in the ICU and perceived quality of dying.

OBJECTIVE: Although end-of-life care in the ICU accounts for a large proportion of health-care costs, few studies have examined the association between costs and satisfaction with care. The objective of this study was to investigate the association of ICU costs with family- and nurse-assessed quality of dying and family satisfaction. METHODS: This was an observational study surveying families and nurses for patients who died in the ICU or within 30 h of transfer from the ICU. A total of 607 patients from two Seattle hospitals were included in the study. Survey data were linked with administrative records to obtain ICU and hospital costs. Regression analyses assessed the association between costs and outcomes assessing satisfaction with care: nurse- and family-assessed Quality of Death and Dying (QODD-1) and Family Satisfaction in the ICU (FS-ICU). RESULTS: For family-reported outcomes, patient insurance status was an important modifier of results. For underinsured patients, higher daily ICU costs were significantly associated with higher FS-ICU and QODD-1 (P < .01 and P = .01, respectively); this association was absent for privately insured or Medicare patients (P = .50 and P = .85, QODD-1 and FS-ICU, respectively). However, higher nurse-assessed QODD-1 was significantly associated with lower average daily ICU cost and total hospital cost (P < .01 and P = .05, respectively). CONCLUSIONS: Family-rated satisfaction with care and quality of dying varied depending on insurance status, with underinsured families rating satisfaction with care and quality of dying higher when average daily ICU costs were higher. However, patients with higher costs were assessed by nurses as having a poorer quality of dying. These findings highlight important differences between family and clinician perspectives and the important role of insurance status.

Transient renal enlargement in pediatric hematopoietic cell transplant recipients.

Age-dependent renal length tables are routinely used when interpreting pediatric ultrasound. Standard renal length tables may not be accurate for HCT patients due to treatment effects on kidney size. The purpose of this study was to determine whether renal size changes from expected lengths based on age after HCT in the absence of other markers of renal disease. Four hundred and fifty renal measurements were made on 101 patients who underwent HCT between 2006 and 2010. Renal length was measured at 1-90 days pre-HCT and at 0-30, 31-90, 91-180, and 181+ days post-HCT. Values were compared with normal renal length tables. Average post-HCT renal lengths were greater than established normative renal length data within every age group. Age-adjusted average renal lengths measured at 0-30 and 31-90 days post-transplantation were significantly larger than pre-HCT renal lengths, with relative increases of 6.9% (4.5, 9.4; p < 0.001) and 3.9% (1.4, 6.4; p = 0.003), respectively. Average renal length did not differ significantly after 90 days post-transplantation. HCT patients may have larger kidneys in the absence of renal disease. Awareness of the potential phenomenon of transient renal enlargement following HCT can prevent misdiagnosis and eliminate unnecessary diagnostic evaluations, interventions, anxiety, resource allocation, and financial costs.