Robust causal inference with continuous instruments using the local instrumental variable curve

The use of pulmonary artery catheter and clinical outcomes in older adults with cardiogenic shock

BACKGROUND

Evidence is lacking regarding the benefits of pulmonary artery catheter (PAC) for cardiogenic shock (CS).

METHODS

We analyzed the data on Medicare fee-for-service beneficiaries aged 65-99 admitted with CS from 2016 to 2020 to compare outcomes of patients monitored with versus without PAC. We implemented propensity score matching weight (PSMW) analysis with hospital fixed effects (effectively comparing outcomes within the same hospital) and quasi-experimental instrumental variable (IV) analysis (accounting for potential unmeasured confounders) with the probability of using PAC for CS in the previous year as the instrument.

RESULTS

We included 4668 and 78,502 patients admitted with CS, monitored with and without PAC, respectively. We found no evidence that the use of PAC was associated with mortality either in PSMW (adjusted absolute risk difference [aRD], +0.5-percentage-points [pp]; 95 % confidence interval [CI], -1.1 to +2.1) or IV (aRD, -2.5 pp.; 95 % CI, -8.2 to +3.2) analyses. While consistent associations were not observed between the use of PAC and major bleeding and sepsis, the use of PAC was associated with a higher risk of all-bleeding (PSMW: aRD, +1.5 pp.; 95 % CI, +0.1 to +2.9; IV: +13.3 pp.; 95 % CI, +7.7 to +18.8) and longer LOS (PSMW: adjusted mean difference, +1.6 days; 95 % CI, +1.1 to +2.0; IV: +6.9 days; +4.9 to +9.0).

CONCLUSIONS

We found no evidence that the use of PAC was associated with lower mortality in patients with CS. While high-quality randomized trials are needed, providers should be careful about appropriate settings and indications of the use of PAC for the management of CS.

Estimating bidirectional effects between social connectedness and mental health in adolescent students: Addressing biases due to endogeneity

Research on the bidirectional relationship between social connectedness and health/mental health in adolescents is scarce, with most studies on adults. Some of the existing studies exploited the availability of longitudinal data to provide evidence of the existence of a causal relationship, either from social connectedness to health or establish a bidirectional relationship. There are at least two weaknesses associated with earlier research to assess the size of the effects. As acknowledged in the literature, one relates to attributing causality to empirical findings, due to well-known but inadequately addressed endogeneity biases. The other relates to failure to account for potentially important covariates, sometimes due to data limitations, or because such variables are not frequently used in sociological research. Existing research predominantly finds that the strongest path is from social connectedness to health/mental health, with effect estimates modest in size. I followed a quasi-experimental strategy by modelling adolescent students' perceptions of social connectedness and mental health perceptions as potentially endogenous variables when estimating bidirectional effects. An instrumental variables (IV) modelling approach was followed, supplemented with a recently developed alternative approach to testing the exclusion restrictions of individual instruments. I exploited the rich information available in the PISA 2018 multi-country dataset, which allows for conditioning for a wide array of information on adolescent students' personal circumstances, self-reported personality-related attributes, relationships with parents; and school characteristics. I found that (1) accounting for endogeneity biases is important; and (2) as opposed to findings reported in the literature, the dominant effect is from mental health perceptions to social connectedness for both male and female participants. The policy relevance of the findings is that adolescent mental health should be the primary focus of interventions, i.e., identifying and treating mental health symptoms as a primary intervention and as a precursor to improving the social connectedness of adolescents.

ASSESSING TIME-VARYING CAUSAL EFFECT MODERATION IN THE PRESENCE OF CLUSTER-LEVEL TREATMENT EFFECT HETEROGENEITY AND INTERFERENCE

The micro-randomized trial (MRT) is a sequential randomized experimental design to empirically evaluate the effectiveness of mobile health (mHealth) intervention components that may be delivered at hundreds or thousands of decision points. MRTs have motivated a new class of causal estimands, termed "causal excursion effects", for which semiparametric inference can be conducted via a weighted, centered least squares criterion (Boruvka et al., 2018). Existing methods assume between-subject independence and non-interference. Deviations from these assumptions often occur. In this paper, causal excursion effects are revisited under potential cluster-level treatment effect heterogeneity and interference, where the treatment effect of interest may depend on cluster-level moderators. Utility of the proposed methods is shown by analyzing data from a multi-institution cohort of first year medical residents in the United States.

Comparative Effectiveness of Percutaneous Microaxial Left Ventricular Assist Device vs Intra-Aortic Balloon Pump or No Mechanical Circulatory Support in Patients With Cardiogenic Shock

Importance

Recent studies have produced inconsistent findings regarding the outcomes of the percutaneous microaxial left ventricular assist device (LVAD) during acute myocardial infarction with cardiogenic shock (AMICS).

Objective

To compare the percutaneous microaxial LVAD vs alternative treatments among patients presenting with AMICS using observational analyses of administrative data.

Design, Setting, and Participants

This comparative effectiveness research study used Medicare fee-for-service claims of patients admitted with AMICS undergoing percutaneous coronary intervention from October 1, 2015, through December 31, 2019. Treatment strategies were compared using (1) inverse probability of treatment weighting to estimate the effect of different baseline treatments in the overall population; (2) instrumental variable analysis to determine the effectiveness of the percutaneous microaxial LVAD among patients whose treatment was influenced by cross-sectional institutional practice patterns; (3) an instrumented difference-in-differences analysis to determine the effectiveness of treatment among patients whose treatment was influenced by longitudinal changes in institutional practice patterns; and (4) a grace period approach to determine the effectiveness of initiating the percutaneous microaxial LVAD within 2 days of percutaneous coronary intervention. Analysis took place between March 2021 and December 2022.

Interventions

Percutaneous microaxial LVAD vs alternative treatments (including medical therapy and intra-aortic balloon pump).

Main Outcomes and Measures

Thirty-day all-cause mortality and readmissions.

Results

Of 23 478 patients, 14 264 (60.8%) were male and the mean (SD) age was 73.9 (9.8) years. In the inverse probability of treatment weighting analysis and grace period approaches, treatment with percutaneous microaxial LVAD was associated with a higher risk-adjusted 30-day mortality (risk difference, 14.9%; 95% CI, 12.9%-17.0%). However, patients receiving the percutaneous microaxial LVAD had a higher frequency of factors associated with severe illness, suggesting possible confounding by measures of illness severity not available in the data. In the instrumental variable analysis, 30-day mortality was also higher with percutaneous microaxial LVAD, but patient and hospital characteristics differed across levels of the instrumental variable, suggesting possible confounding by unmeasured variables (risk difference, 13.5%; 95% CI, 3.9%-23.2%). In the instrumented difference-in-differences analysis, the association between the percutaneous microaxial LVAD and mortality was imprecise, and differences in trends in characteristics between hospitals with different percutaneous microaxial LVAD use suggested potential assumption violations.

Conclusions

In observational analyses comparing the percutaneous microaxial LVAD to alternative treatments among patients with AMICS, the percutaneous microaxial LVAD was associated with worse outcomes in some analyses, while in other analyses, the association was too imprecise to draw meaningful conclusions. However, the distribution of patient and institutional characteristics between treatment groups or groups defined by institutional differences in treatment use, including changes in use over time, combined with clinical knowledge of illness severity factors not captured in the data, suggested violations of key assumptions that are needed for valid causal inference with different observational analyses. Randomized clinical trials of mechanical support devices will allow valid comparisons across candidate treatment strategies and help resolve ongoing controversies.

Instrumented difference-in-differences

Unmeasured confounding is a key threat to reliable causal inference based on observational studies. Motivated from two powerful natural experiment devices, the instrumental variables and difference-in-differences, we propose a new method called instrumented difference-in-differences that explicitly leverages exogenous randomness in an exposure trend to estimate the average and conditional average treatment effect in the presence of unmeasured confounding. We develop the identification assumptions using the potential outcomes framework. We propose a Wald estimator and a class of multiply robust and efficient semiparametric estimators, with provable consistency and asymptotic normality. In addition, we extend the instrumented difference-in-differences to a two-sample design to facilitate investigations of delayed treatment effect and provide a measure of weak identification. We demonstrate our results in simulated and real datasets.

Regression-based estimation of heterogeneous treatment effects when extending inferences from a randomized trial to a target population

Most work on extending (generalizing or transporting) inferences from a randomized trial to a target population has focused on estimating average treatment effects (i.e., averaged over the target population's covariate distribution). Yet, in the presence of strong effect modification by baseline covariates, the average treatment effect in the target population may be less relevant for guiding treatment decisions. Instead, the conditional average treatment effect (CATE) as a function of key effect modifiers may be a more useful estimand. Recent work on estimating target population CATEs using baseline covariate, treatment, and outcome data from the trial and covariate data from the target population only allows for the examination of heterogeneity over distinct subgroups. We describe flexible pseudo-outcome regression modeling methods for estimating target population CATEs conditional on discrete or continuous baseline covariates when the trial is embedded in a sample from the target population (i.e., in nested trial designs). We construct pointwise confidence intervals for the CATE at a specific value of the effect modifiers and uniform confidence bands for the CATE function. Last, we illustrate the methods using data from the Coronary Artery Surgery Study (CASS) to estimate CATEs given history of myocardial infarction and baseline ejection fraction value in the target population of all trial-eligible patients with stable ischemic heart disease.

Simultaneous Maximum Likelihood Estimation for Piecewise Linear Instrumental Variable Models

Analysis of instrumental variables is an effective approach to dealing with endogenous variables and unmeasured confounding issue in causal inference. We propose using the piecewise linear model to fit the relationship between the continuous instrumental variable and the continuous explanatory variable, as well as the relationship between the continuous explanatory variable and the outcome variable, which generalizes the traditional linear instrumental variable models. The two-stage least square and limited information maximum likelihood methods are used for the simultaneous estimation of the regression coefficients and the threshold parameters. Furthermore, we study the limiting distribution of the estimators in the correctly specified and misspecified models and provide a robust estimation of the variance-covariance matrix. We illustrate the finite sample properties of the estimation in terms of the Monte Carlo biases, standard errors, and coverage probabilities via the simulated data. Our proposed model is applied to an education-salary data, which investigates the causal effect of children's years of schooling on estimated hourly wage with father's years of schooling as the instrumental variable.

Local Instrumental Variable Methods to Address Confounding and Heterogeneity when Using Electronic Health Records: An Application to Emergency Surgery

Background

Electronic health records (EHRs) offer opportunities for comparative effectiveness research to inform decision making. However, to provide useful evidence, these studies must address confounding and treatment effect heterogeneity according to unmeasured prognostic factors. Local instrumental variable (LIV) methods can help studies address these challenges, but have yet to be applied to EHR data. This article critically examines a LIV approach to evaluate the cost-effectiveness of emergency surgery (ES) for common acute conditions from EHRs.

Methods

This article uses hospital episodes statistics (HES) data for emergency hospital admissions with acute appendicitis, diverticular disease, and abdominal wall hernia to 175 acute hospitals in England from 2010 to 2019. For each emergency admission, the instrumental variable for ES receipt was each hospital’s ES rate in the year preceding the emergency admission. The LIV approach provided individual-level estimates of the incremental quality-adjusted life-years, costs and net monetary benefit of ES, which were aggregated to the overall population and subpopulations of interest, and contrasted with those from traditional IV and risk-adjustment approaches.

Results

The study included 268,144 (appendicitis), 138,869 (diverticular disease), and 106,432 (hernia) patients. The instrument was found to be strong and to minimize covariate imbalance. For diverticular disease, the results differed by method; although the traditional approaches reported that, overall, ES was not cost-effective, the LIV approach reported that ES was cost-effective but with wide statistical uncertainty. For all 3 conditions, the LIV approach found heterogeneity in the cost-effectiveness estimates across population subgroups: in particular, ES was not cost-effective for patients with severe levels of frailty.

Conclusions

EHRs can be combined with LIV methods to provide evidence on the cost-effectiveness of routinely provided interventions, while fully recognizing heterogeneity.

Highlights

A National Study to Compare Effective Management of Constipation in Children Receiving Concurrent Versus Standard Hospice Care

Constipation is a distressing and uncomfortable symptom children experience at end of life. There is a gap in knowledge about how different approaches to hospice care delivery might improve pediatric symptom management of constipation. The purpose of this study was to evaluate the effectiveness of pediatric concurrent hospice versus standard hospice care to manage constipation. Medicaid data (2011-2013) were analyzed. Children who were younger than 21 years enrolled in hospice care and had a hospice enrollment between January 1, 2011, and December 31, 2013, were included. Instrumental variable analysis was used to test the effectiveness of concurrent versus standard hospice care. Among the 18 152 children, approximately 14% of participants were diagnosed or treated for constipation from a nonhospice provider during hospice enrollment. A higher proportion of children received nonhospice care for constipation in concurrent hospice care, compared with standard hospice (19.5% vs 13.2%), although this was not significant (β = .22, P < .05) after adjusting for covariates. The findings demonstrated that concurrent care was no more effective than standard hospice care in managing pediatric constipation. Hospice and nonhospice providers may be doing a sufficient job ordering bowel regimens before constipation becomes a serious problem for children at end of life.

Effectiveness of Pediatric Concurrent Hospice Care to Improve Continuity of Care

BACKGROUND

The 2010 Patient Protection and Affordable Care Act (ACA) mandated landmark hospice care legislation for children at end of life. Little is known about the impact of pediatric concurrent hospice care.

OBJECTIVE

The purpose of this study was to examine the effect of pediatric concurrent vs standard hospice care on end-of-life care continuity among Medicaid beneficiaries.

METHODS

Using national Medicaid data, we conducted a quasi-experimental designed study to estimate the effect of concurrent vs standard hospice care to improve end-of-life care continuity for children. Care continuity (i.e., hospice length of stay, hospice disenrollment, emergency room transition, and inpatient transition) was measured via claims data. Exposures were concurrent hospice vs standard hospice care. Using instrumental variable analysis, the effectiveness of exposures on care continuity was compared.

RESULTS

Concurrent hospice care affected care continuity. It resulted in longer lengths of stays in hospice (β = 2.76, P < .001) and reduced hospice live discharges (β = -2.80, P < .05), compared to standard hospice care. Concurrent care was not effective at reducing emergency room (β = 2.09, P < .001) or inpatient care (β = .007, P < .05) transitions during hospice enrollment.

CONCLUSION

Our study provides critical insight into the quality of care delivered for children at end of life. These findings have policy implications.

Doubly robust nonparametric instrumental variable estimators for survival outcomes

Instrumental variable (IV) methods allow us the opportunity to address unmeasured confounding in causal inference. However, most IV methods are only applicable to discrete or continuous outcomes with very few IV methods for censored survival outcomes. In this article, we propose nonparametric estimators for the local average treatment effect on survival probabilities under both covariate-dependent and outcome-dependent censoring. We provide an efficient influence function-based estimator and a simple estimation procedure when the IV is either binary or continuous. The proposed estimators possess double-robustness properties and can easily incorporate nonparametric estimation using machine learning tools. In simulation studies, we demonstrate the flexibility and double robustness of our proposed estimators under various plausible scenarios. We apply our method to the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial for estimating the causal effect of screening on survival probabilities and investigate the causal contrasts between the two interventions under different censoring assumptions.

Efficient Nonparametric Causal Inference with Missing Exposure Information

Missing exposure information is a very common feature of many observational studies. Here we study identifiability and efficient estimation of causal effects on vector outcomes, in such cases where treatment is unconfounded but partially missing. We consider a missing at random setting where missingness in treatment can depend not only on complex covariates, but also on post-treatment outcomes. We give a new identifying expression for average treatment effects in this setting, along with the efficient influence function for this parameter in a nonparametric model, which yields a nonparametric efficiency bound. We use this latter result to construct nonparametric estimators that are less sensitive to the curse of dimensionality than usual, e. g. by having faster rates of convergence than the complex nuisance estimators they rely on. Further we show that these estimators can be root-n consistent and asymptotically normal under weak nonparametric conditions, even when constructed using flexible machine learning. Finally we apply these results to the problem of causal inference with a partially missing instrumental variable.