Predictors of urine toxicology and other biologic specimen missingness in randomized trials of substance use disorders.

BACKGROUND: High levels of missing outcome data for biologically confirmed substance use (BCSU) threaten the validity of substance use disorder (SUD) clinical trials. Underlying attributes of clinical trials could explain BCSU missingness and identify targets for improved trial design. METHODS: We reviewed 21 clinical trials funded by the NIDA National Drug Abuse Treatment Clinical Trials Network (CTN) and published from 2005 to 2018 that examined pharmacologic and psychosocial interventions for SUD. We used configurational analysis-a Boolean algebra approach that identifies an attribute or combination of attributes predictive of an outcome-to identify trial design features and participant characteristics associated with high levels of BCSU missingness. Associations were identified by configuration complexity, consistency, coverage, and robustness. We limited results using a consistency threshold of 0.75 and summarized model fit using the product of consistency and coverage. RESULTS: For trial design features, the final solution consisted of two pathways: psychosocial treatment as a trial intervention OR larger trial arm size (complexity=2, consistency=0.79, coverage=0.93, robustness score=0.71). For participant characteristics, the final solution consisted of two pathways: interventions targeting individuals with poly- or nonspecific substance use OR younger age (complexity=2, consistency=0.75, coverage=0.86, robustness score=1.00). CONCLUSIONS: Psychosocial treatments, larger trial arm size, interventions targeting individuals with poly- or nonspecific substance use, and younger age among trial participants were predictive of missing BCSU data in SUD clinical trials. Interventions to mitigate missing data that focus on these attributes may reduce threats to validity and improve utility of SUD clinical trials.

Sensitivity analysis for principal ignorability violation in estimating complier and noncomplier average causal effects.

An important strategy for identifying principal causal effects (popular estimands in settings with noncompliance) is to invoke the principal ignorability (PI) assumption. As PI is untestable, it is important to gauge how sensitive effect estimates are to its violation. We focus on this task for the common one-sided noncompliance setting where there are two principal strata, compliers and noncompliers. Under PI, compliers and noncompliers share the same outcome-mean-given-covariates function under the control condition. For sensitivity analysis, we allow this function to differ between compliers and noncompliers in several ways, indexed by an odds ratio, a generalized odds ratio, a mean ratio, or a standardized mean difference sensitivity parameter. We tailor sensitivity analysis techniques (with any sensitivity parameter choice) to several types of PI-based main analysis methods, including outcome regression, influence function (IF) based and weighting methods. We discuss range selection for the sensitivity parameter. We illustrate the sensitivity analyses with several outcome types from the JOBS II study. This application estimates nuisance functions parametrically - for simplicity and accessibility. In addition, we establish rate conditions on nonparametric nuisance estimation for IF-based estimators to be asymptotically normal - with a view to inform nonparametric inference.

Temporary Financial Assistance Reduced The Probability Of Unstable Housing Among Veterans For More Than 1 Year.

The Department of Veterans Affairs (VA) aims to reduce homelessness among veterans through programs such as Supportive Services for Veteran Families (SSVF). An important component of SSVF is temporary financial assistance. Previous research has demonstrated the effectiveness of temporary financial assistance in reducing short-term housing instability, but studies have not examined its long-term effect on housing outcomes. Using data from the VA's electronic health record system, we analyzed the effect of temporary financial assistance on veterans' housing instability for three years after entry into SSVF. We extracted housing outcomes from clinical notes, using natural language processing, and compared the probability of unstable housing among veterans who did and did not receive temporary financial assistance. We found that temporary financial assistance rapidly reduced the probability of unstable housing, but the effect attenuated after forty-five days. Our findings suggest that to maintain long-term housing stability for veterans who have exited SSVF, additional interventions may be needed.

New Users of Angiotensin II Receptor Blocker-Versus Angiotensin-Converting Enzyme Inhibitor-Based Antihypertensive Medication Regimens and Cardiovascular Disease Events: A Secondary Analysis of ACCORD-BP and SPRINT.

BACKGROUND Angiotensin II receptor blockers (ARBs) and angiotensin-converting enzyme inhibitors (ACEIs) block distinct components of the renin-angiotensin system. Whether this translates into differential effects on cardiovascular disease events remains unclear. METHODS AND RESULTS We used the ACCORD-BP (Action to Control Cardiovascular Risk in Diabetes-Blood Pressure) trial and the SPRINT (Systolic Blood Pressure Intervention Trial) to emulate target trials of new users of ARBs versus ACEIs on cardiovascular disease events (primary outcome) and death (secondary outcome). We estimated marginal cause-specific hazard ratios (HRs) and treatment-specific cumulative incidence functions with inverse probability of treatment weights. We identified 3298 new users of ARBs or ACEIs (ACCORD-BP: 374 ARB versus 884 ACEI; SPRINT: 727 ARB versus 1313 ACEI). For participants initiating ARBs versus ACEIs, the inverse probability of treatment weight rate of the primary outcome was 3.2 versus 3.5 per 100 person-years in ACCORD-BP (HR, 0.91 [95% CI, 0.63-1.31]) and 1.8 versus 2.2 per 100 person-years in SPRINT (HR, 0.81 [95% CI, 0.56-1.18]). There were no appreciable differences in pooled analyses, except that ARBs versus ACEIs were associated with a lower death rate (HR, 0.56 [95% CI, 0.37-0.85]). ARBs were associated with a lower rate of the primary outcome among subgroups of male versus female participants, non-Hispanic Black versus non-Hispanic White participants, and those randomly assigned to standard versus intensive blood pressure (Pinteraction: <0.01, 0.05, and <0.01, respectively). CONCLUSIONS In this secondary analysis of ACCORD-BP and SPRINT, new users of ARB- versus ACEI-based antihypertensive medication regimens experienced similar cardiovascular disease events rates, with important subgroup differences and lower rates of death overall. REGISTRATION URL: https://www.clinicaltrials.gov; Unique identifiers: NCT01206062, NCT00000620.

Losartan in hospitalized patients with COVID-19 in North America: An individual participant data meta-analysis.

BACKGROUND: Angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers (ARBs) have been hypothesized to benefit patients with COVID-19 via the inhibition of viral entry and other mechanisms. We conducted an individual participant data (IPD) meta-analysis assessing the effect of starting the ARB losartan in recently hospitalized COVID-19 patients. METHODS: We searched ClinicalTrials.gov in January 2021 for U.S./Canada-based trials where an angiotensin-converting enzyme inhibitors/ARB was a treatment arm, targeted outcomes could be extrapolated, and data sharing was allowed. Our primary outcome was a 7-point COVID-19 ordinal score measured 13 to 16 days post-enrollment. We analyzed data by fitting multilevel Bayesian ordinal regression models and standardizing the resulting predictions. RESULTS: 325 participants (156 losartan vs 169 control) from 4 studies contributed IPD. Three were randomized trials; one used non-randomized concurrent and historical controls. Baseline covariates were reasonably balanced for the randomized trials. All studies evaluated losartan. We found equivocal evidence of a difference in ordinal scores 13-16 days post-enrollment (model-standardized odds ratio [OR] 1.10, 95% credible interval [CrI] 0.76-1.71; adjusted OR 1.15, 95% CrI 0.15-3.59) and no compelling evidence of treatment effect heterogeneity among prespecified subgroups. Losartan had worse effects for those taking corticosteroids at baseline after adjusting for covariates (ratio of adjusted ORs 0.29, 95% CrI 0.08-0.99). Hypotension serious adverse event rates were numerically higher with losartan. CONCLUSIONS: In this IPD meta-analysis of hospitalized COVID-19 patients, we found no convincing evidence for the benefit of losartan versus control treatment, but a higher rate of hypotension adverse events with losartan.

Using Trial and Observational Data to Assess Effectiveness: Trial Emulation, Transportability, Benchmarking, and Joint Analysis.

Comparisons between randomized trial analyses and observational analyses that attempt to address similar research questions have generated many controversies in epidemiology and the social sciences. There has been little consensus on when such comparisons are reasonable, what their implications are for the validity of observational analyses, or whether trial and observational analyses can be integrated to address effectiveness questions. Here, we consider methods for using observational analyses to complement trial analyses when assessing treatment effectiveness. First, we review the framework for designing observational analyses that emulate target trials and present an evidence map of its recent applications. We then review approaches for estimating the average treatment effect in the target population underlying the emulation: using observational analyses of the emulation data alone; and using transportability analyses to extend inferences from a trial to the target population. We explain how comparing treatment effect estimates from the emulation against those from the trial can provide evidence on whether observational analyses can be trusted to deliver valid estimates of effectiveness - a process we refer to as benchmarking - and, in some cases, allow the joint analysis of the trial and observational data. We illustrate different approaches using a simplified example of a pragmatic trial and its emulation in registry data. We conclude that synthesizing trial and observational data - in transportability, benchmarking, or joint analyses - can leverage their complementary strengths to enhance learning about comparative effectiveness, through a process combining quantitative methods and epidemiological judgements.

Aspirin or Low-Molecular-Weight Heparin for Thromboprophylaxis after a Fracture.

BACKGROUND: Clinical guidelines recommend low-molecular-weight heparin for thromboprophylaxis in patients with fractures, but trials of its effectiveness as compared with aspirin are lacking. METHODS: In this pragmatic, multicenter, randomized, noninferiority trial, we enrolled patients 18 years of age or older who had a fracture of an extremity (anywhere from hip to midfoot or shoulder to wrist) that had been treated operatively or who had any pelvic or acetabular fracture. Patients were randomly assigned to receive low-molecular-weight heparin (enoxaparin) at a dose of 30 mg twice daily or aspirin at a dose of 81 mg twice daily while they were in the hospital. After hospital discharge, the patients continued to receive thromboprophylaxis according to the clinical protocols of each hospital. The primary outcome was death from any cause at 90 days. Secondary outcomes were nonfatal pulmonary embolism, deep-vein thrombosis, and bleeding complications. RESULTS: A total of 12,211 patients were randomly assigned to receive aspirin (6101 patients) or low-molecular-weight heparin (6110 patients). Patients had a mean (±SD) age of 44.6±17.8 years, 0.7% had a history of venous thromboembolism, and 2.5% had a history of cancer. Patients received a mean of 8.8±10.6 in-hospital thromboprophylaxis doses and were prescribed a median 21-day supply of thromboprophylaxis at discharge. Death occurred in 47 patients (0.78%) in the aspirin group and in 45 patients (0.73%) in the low-molecular-weight-heparin group (difference, 0.05 percentage points; 96.2% confidence interval, -0.27 to 0.38; P<0.001 for a noninferiority margin of 0.75 percentage points). Deep-vein thrombosis occurred in 2.51% of patients in the aspirin group and 1.71% in the low-molecular-weight-heparin group (difference, 0.80 percentage points; 95% CI, 0.28 to 1.31). The incidence of pulmonary embolism (1.49% in each group), bleeding complications, and other serious adverse events were similar in the two groups. CONCLUSIONS: In patients with extremity fractures that had been treated operatively or with any pelvic or acetabular fracture, thromboprophylaxis with aspirin was noninferior to low-molecular-weight heparin in preventing death and was associated with low incidences of deep-vein thrombosis and pulmonary embolism and low 90-day mortality. (Funded by the Patient-Centered Outcomes Research Institute; PREVENT CLOT ClinicalTrials.gov number, NCT02984384.).

Do superficial infections increase the risk of deep infections in tibial plateau and plafond fractures?

PURPOSE: Open reduction internal fixation of tibial plateau and pilon fractures may be complicated by deep surgical site infection requiring operative debridement and antibiotic therapy. The management of superficial surgical site infection is controversial. We sought to determine whether superficial infection is associated with an increased risk of deep infection requiring surgical debridement after fixation of tibial plateau and pilon fractures. METHODS: This is a secondary analysis of data from the VANCO trial, which included 980 adult patients with a tibial plateau or pilon fracture at elevated risk of infection who underwent open reduction internal fixation with plates and screws with or without intrawound vancomycin powder. An association of superficial surgical site infection with deep surgical site infection requiring debridement surgery and antibiotics was explored after matching on risk factors for deep surgical site infection. RESULTS: Of the 980 patients, we observed 30 superficial infections (3.1%) and 76 deep infections (7.8%). Among patients who developed a superficial infection, the unadjusted incidence of developing a deep infection within 90 days was 12.8% (95% confidence interval [CI] 1.3-24.2%). However, after a 3:1 match on infection risk factors, the 90-day marginal probability of a deep surgical site infection after sustaining a superficial infection was 6.0% (95% CI - 6.5-18.5%, p = 0.35). CONCLUSION: Deep infection after superficial infection is uncommon following operative fixation of tibial plateau and pilon fractures. Increased risk of subsequent deep infection attributable to superficial infection was inconclusive in these data. LEVEL OF EVIDENCE: Prognostic Level II.

Using natural language processing to study homelessness longitudinally with electronic health record data subject to irregular observations.

The Electronic Health Record (EHR) contains information about social determinants of health (SDoH) such as homelessness. Much of this information is contained in clinical notes and can be extracted using natural language processing (NLP). This data can provide valuable information for researchers and policymakers studying long-term housing outcomes for individuals with a history of homelessness. However, studying homelessness longitudinally in the EHR is challenging due to irregular observation times. In this work, we applied an NLP system to extract housing status for a cohort of patients in the US Department of Veterans Affairs (VA) over a three-year period. We then applied inverse intensity weighting to adjust for the irregularity of observations, which was used generalized estimating equations to estimate the probability of unstable housing each day after entering a VA housing assistance program. Our methods generate unique insights into the long-term outcomes of individuals with a history of homelessness and demonstrate the potential for using EHR data for research and policymaking.

Randomized Trials With Repeatedly Measured Outcomes: Handling Irregular and Potentially Informative Assessment Times.

Randomized trials are often designed to collect outcomes at fixed points in time after randomization. In practice, the number and timing of outcome assessments can vary among participants (i.e., irregular assessment). In fact, the timing of assessments may be associated with the outcome of interest (i.e., informative assessment). For example, in a trial evaluating the effectiveness of treatments for major depressive disorder, not only did the timings of outcome assessments vary among participants but symptom scores were associated with assessment frequency. This type of informative observation requires appropriate statistical analysis. Although analytic methods have been developed, they are rarely used. In this article, we review the literature on irregular assessments with a view toward developing recommendations for analyzing trials with irregular and potentially informative assessment times. We show how the choice of analytic approach hinges on assumptions about the relationship between the assessment and outcome processes. We argue that irregular assessment should be treated with the same care as missing data, and we propose that trialists adopt strategies to minimize the extent of irregularity; describe the extent of irregularity in assessment times; make their assumptions about the relationships between assessment times and outcomes explicit; adopt analytic techniques that are appropriate to their assumptions; and assess the sensitivity of trial results to their assumptions.

Hydroxychloroquine/chloroquine for the treatment of hospitalized patients with COVID-19: An individual participant data meta-analysis.

BACKGROUND: Results from observational studies and randomized clinical trials (RCTs) have led to the consensus that hydroxychloroquine (HCQ) and chloroquine (CQ) are not effective for COVID-19 prevention or treatment. Pooling individual participant data, including unanalyzed data from trials terminated early, enables more detailed investigation of the efficacy and safety of HCQ/CQ among subgroups of hospitalized patients. METHODS: We searched ClinicalTrials.gov in May and June 2020 for US-based RCTs evaluating HCQ/CQ in hospitalized COVID-19 patients in which the outcomes defined in this study were recorded or could be extrapolated. The primary outcome was a 7-point ordinal scale measured between day 28 and 35 post enrollment; comparisons used proportional odds ratios. Harmonized de-identified data were collected via a common template spreadsheet sent to each principal investigator. The data were analyzed by fitting a prespecified Bayesian ordinal regression model and standardizing the resulting predictions. RESULTS: Eight of 19 trials met eligibility criteria and agreed to participate. Patient-level data were available from 770 participants (412 HCQ/CQ vs 358 control). Baseline characteristics were similar between groups. We did not find evidence of a difference in COVID-19 ordinal scores between days 28 and 35 post-enrollment in the pooled patient population (odds ratio, 0.97; 95% credible interval, 0.76-1.24; higher favors HCQ/CQ), and found no convincing evidence of meaningful treatment effect heterogeneity among prespecified subgroups. Adverse event and serious adverse event rates were numerically higher with HCQ/CQ vs control (0.39 vs 0.29 and 0.13 vs 0.09 per patient, respectively). CONCLUSIONS: The findings of this individual participant data meta-analysis reinforce those of individual RCTs that HCQ/CQ is not efficacious for treatment of COVID-19 in hospitalized patients.

Risk of Mild Cognitive Impairment or Probable Dementia in New Users of Angiotensin II Receptor Blockers and Angiotensin-Converting Enzyme Inhibitors: A Secondary Analysis of Data From the Systolic Blood Pressure Intervention Trial (SPRINT).

Importance: The cardiovascular and renal outcomes of angiotensin-II receptor blocker (ARB) and angiotensin-converting enzyme inhibitor (ACEI) treatment are well-known; however, few studies have evaluated initiation of these agents and cognitive impairment. Objective: To emulate a target trial to evaluate the cognitive outcomes of initiating an ARB- vs ACEI-based antihypertensive regimen in individuals at risk for mild cognitive impairment (MCI) and probable dementia (PD). Design, Setting, and Participants: Active comparator, new-user observational cohort study design using data from the Systolic Blood Pressure Intervention Trial (SPRINT), conducted November 2010 through July 2018. Marginal cause-specific hazard ratios (HRs) and treatment-specific cumulative incidence functions were estimated with inverse probability (IP) weighting to account for confounding. Participants were using neither an ARB nor ACEI at baseline. Data analysis was conducted from April 7, 2021, to April 26, 2022. Exposures: New users of ARB vs ACEI during the first 12 months of trial follow-up. Main Outcomes and Measures: Composite of adjudicated amnestic MCI or PD. Results: Of 9361 participants, 727 and 1313 new users of an ARB or ACEI, respectively, with well-balanced baseline characteristics between medication exposure groups after inverse probability weighting (mean [SD] age, 67 [9.5] years; 1291 ]63%] male; 240 [33%] Black; 89 [12%] Hispanic; 383 [53%] White; and 15 [2%] other race or ethnicity. In the primary analysis, during a median follow-up of 4.9 years, the inverse probability-weighted rate of amnestic MCI or PD was 4.3 vs 4.6 per 100 person-years among participants initiating ARB vs ACEI (HR, 0.93; 95% CI, 0.76-1.13). In subgroup analyses, new users of an ARB vs ACEI had a lower rate of amnestic MCI or PD among those in the standard systolic blood pressure treatment arm (HR, 0.61; 95% CI, 0.41-0.91) but not in the intensive arm (HR, 1.17; 95% CI, 0.90-1.52) (P = .007 for interaction). Conclusions and Relevance: In this observational cohort study of US adults at high cardiovascular disease risk, there was no difference in the rate of amnestic MCI or PD among new users of an ARB compared with ACEI, although 95% CIs were wide.

Hydroxychloroquine/chloroquine for the treatment of hospitalized patients with COVID-19: An individual participant data meta-analysis.

Background: Results from observational studies and randomized clinical trials (RCTs) have led to the consensus that hydroxychloroquine (HCQ) and chloroquine (CQ) are not effective for COVID-19 prevention or treatment. Pooling individual participant data, including unanalyzed data from trials terminated early, enables more detailed investigation of the efficacy and safety of HCQ/CQ among subgroups of hospitalized patients. Methods: We searched ClinicalTrials.gov in May and June 2020 for US-based RCTs evaluating HCQ/CQ in hospitalized COVID-19 patients in which the outcomes defined in this study were recorded or could be extrapolated. The primary outcome was a 7-point ordinal scale measured between day 28 and 35 post enrollment; comparisons used proportional odds ratios. Harmonized de-identified data were collected via a common template spreadsheet sent to each principal investigator. The data were analyzed by fitting a prespecified Bayesian ordinal regression model and standardizing the resulting predictions. Results: Eight of 19 trials met eligibility criteria and agreed to participate. Patient-level data were available from 770 participants (412 HCQ/CQ vs 358 control). Baseline characteristics were similar between groups. We did not find evidence of a difference in COVID-19 ordinal scores between days 28 and 35 post-enrollment in the pooled patient population (odds ratio, 0.97; 95% credible interval, 0.76-1.24; higher favors HCQ/CQ), and found no convincing evidence of meaningful treatment effect heterogeneity among prespecified subgroups. Adverse event and serious adverse event rates were numerically higher with HCQ/CQ vs control (0.39 vs 0.29 and 0.13 vs 0.09 per patient, respectively). Conclusions: The findings of this individual participant data meta-analysis reinforce those of individual RCTs that HCQ/CQ is not efficacious for treatment of COVID-19 in hospitalized patients.

The 1-Year Economic Impact of Work Productivity Loss Following Severe Lower Extremity Trauma.

BACKGROUND: Severe lower extremity trauma among working-age adults is highly consequential for returning to work; however, the economic impact attributed to injury has not been fully quantified. The purpose of this study was to examine work and productivity loss during the year following lower extremity trauma and to calculate the economic losses associated with lost employment, lost work time (absenteeism), and productivity loss while at work (presenteeism). METHODS: This is an analysis of data collected prospectively across 3 multicenter studies of lower extremity trauma outcomes in the United States. Data were used to construct a Markov model that accumulated hours lost over time due to lost employment, absenteeism, and presenteeism among patients from 18 to 64 years old who were working prior to their injury. Average U.S. wages were used to calculate economic loss overall and by sociodemographic and injury subgroups. RESULTS: Of 857 patients working prior to injury, 47.2% had returned to work at 1 year. The average number of productive hours of work lost was 1,758.8/person, representing 84.6% of expected annual productive hours. Of the hours lost, 1,542.3 (87.7%) were due to working no hours or lost employment, 71.1 (4.0%) were due to missed hours after having returned, and 145.4 (8.3%) were due to decreased productivity while working. The 1-year economic loss due to injury totaled $64,427/patient (95% confidence interval [CI], $63,183 to $65,680). Of the 1,758.8 lost hours, approximately 88% were due to not being employed (working zero hours), 4% were due to absenteeism, and 8% were due to presenteeism. Total productivity loss was higher among older adults (≥40 years), men, those with a physically demanding job, and the most severe injuries (i.e., those leading to amputation as well as Gustilo type-IIIB tibial fractures and type-III pilon/ankle fractures). CONCLUSIONS: Patients with severe lower extremity trauma carry a substantial economic burden. The costs of lost productivity should be considered when evaluating outcomes.

Outcomes of Patients With Large Versus Small Bone Defects in Open Tibia Fractures Treated With an Intramedullary Nail: A Descriptive Analysis of a Multicenter Retrospective Study.

OBJECTIVES: To compare outcomes in patients with open tibia shaft fractures based on defect size. DESIGN: Retrospective review. SETTING: Eighteen trauma centers. POPULATION: The study included 132 patients with diaphyseal tibia bone defects >1 cm and ≥50% cortical loss treated with intramedullary nail. OUTCOMES: The primary outcome was number of secondary surgeries to promote healing (bone graft, revision fixation, or bone transport). Additional outcomes included occurrence of secondary surgeries (bone graft, infection, amputation, and flap failure) and proportion healed at one year. Results are compared by "radiographic apparent bone gap" of <2.5 or ≥2.5 cm. RESULTS: The estimated conditional probability of bone grafting within one year given graft-free at 90 days was 44% and 47% in the <2.5 cm and ≥2.5 cm groups, respectively. An estimated infection risk of 14% was observed in both groups [adjusted hazard ratio (HR) 0.98, 95% confidence interval (CI): 0.33-2.92], estimated amputation risk was 9% (<2.5 cm) and 4% (≥2.5 cm) (unadjusted HR 0.66, 95% CI: 0.13-3.29), and estimated flap failure risk (among those with flaps) was 10% and 13%, respectively (unadjusted HR 1.71, 95% CI: 0.24-12.25). There was no appreciable difference in the proportion healed at one year between defect sizes [adjusted HR: 1.07 (95% CI, 0.63-1.82)]. CONCLUSIONS: Larger size bone defects were not associated with higher number of secondary procedures to promote healing or a lower overall one-year healing rate. LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

A Bayesian nonparametric approach for evaluating the causal effect of treatment in randomized trials with semi-competing risks.

We develop a Bayesian nonparametric (BNP) approach to evaluate the causal effect of treatment in a randomized trial where a nonterminal event may be censored by a terminal event, but not vice versa (i.e., semi-competing risks). Based on the idea of principal stratification, we define a novel estimand for the causal effect of treatment on the nonterminal event. We introduce identification assumptions, indexed by a sensitivity parameter, and show how to draw inference using our BNP approach. We conduct simulation studies and illustrate our methodology using data from a brain cancer trial. The R code implementing our model and algorithm is available for download at https://github.com/YanxunXu/BaySemiCompeting.

Global sensitivity analysis of randomized trials with nonmonotone missing binary outcomes: Application to studies of substance use disorders.

In this paper, we present a method for conducting global sensitivity analysis of randomized trials in which binary outcomes are scheduled to be collected on participants at prespecified points in time after randomization and these outcomes may be missing in a nonmonotone fashion. We introduce a class of missing data assumptions, indexed by sensitivity parameters, which are anchored around the missing not at random assumption introduced by Robins (Statistics in Medicine, 1997). For each assumption in the class, we establish that the joint distribution of the outcomes is identifiable from the distribution of the observed data. Our estimation procedure uses the plug-in principle, where the distribution of the observed data is estimated using random forests. We establish n$\sqrt {n}$ asymptotic properties for our estimation procedure. We illustrate our methodology in the context of a randomized trial designed to evaluate a new approach to reducing substance use, assessed by testing urine samples twice weekly, among patients entering outpatient addiction treatment. We evaluate the finite sample properties of our method in a realistic simulation study. Our methods have been implemented in an R package entitled slabm.

Discussion on "Causal mediation of semicompeting risks" by Yen-Tsung Huang.

Huang proposes a method for assessing the impact of a point treatment on mortality either directly or mediated by occurrence of a nonterminal health event, based on data from a prospective cohort study in which the occurrence of the nonterminal health event may be preemptied by death but not vice versa. The author uses a causal mediation framework to formally define causal quantities known as natural (in)direct effects. The novelty consists of adapting these concepts to a continuous-time modeling framework based on counting processes. In an effort to posit "scientifically interpretable estimands," statistical and causal assumptions are introduced for identification. In this commentary, we argue that these assumptions are not only difficult to interpret and justify, but are also likely violated in the hepatitis B motivating example and other survival/time to event settings as well.

A structured methodology review showed analyses of functional outcomes are frequently limited to "survivors only" in trials enrolling patients at high risk of death.

OBJECTIVE: This structured methodology review evaluated statistical approaches used in randomized controlled trials (RCTs) enrolling patients at high risk of death and makes recommendations for reporting future RCTs. STUDY DESIGN AND SETTING: Using PubMed, we searched for RCTs published in five general medicine journals from January 2014 to August 2019 wherein mortality was ≥10% in at least one randomized group. We abstracted primary and secondary outcomes, statistical analysis methods, and patient samples evaluated (all randomized patients vs. "survivors only"). RESULTS: Of 1947 RCTs identified, 434 met eligibility criteria. Of the eligible RCTs, 91 (21%) and 351 (81%) had a primary or secondary functional outcome, respectively, of which 36 (40%) and 263 (75%) evaluated treatment effects among "survivors only". In RCTs that analyzed all randomized patients, the most common methods included use of ordinal outcomes (e.g., modified Rankin Scale) or creating composite outcomes (primary: 41 of 91 [45%]; secondary: 57 of 351 [16%]). CONCLUSION: In RCTs enrolling patients at high risk of death, statistical analyses of functional outcomes are frequently conducted among "survivors only," for which conclusions might be misleading. Given the growing number of RCTs conducted among patients hospitalized with COVID-19 and other critical illnesses, standards for reporting should be created.