Aaron Sarvet

Aaron Sarvet

Harvard University

H-index: 24

North America-United States

About Aaron Sarvet

Aaron Sarvet, With an exceptional h-index of 24 and a recent h-index of 23 (since 2020), a distinguished researcher at Harvard University, specializes in the field of Causal inference, epidemiology, statistics.

His recent articles reflect a diverse array of research interests and contributions to the field:

Improved bounds and inference on optimal regimes

Rejoinder to" Perspectives onharm'in personalized medicine--an alternative perspective"

Grace periods in comparative effectiveness studies of sustained treatments

Longitudinal incremental propensity score interventions for limited resource settings

Causal effects of intervening variables in settings with unmeasured confounding

Conditional separable effects

Mats Stensrud, Vanessa Didelez and Aaron Sarvet's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to …

Separable effects for adherence

Aaron Sarvet Information

University

Harvard University

Position

PhD Candidiate in Epidemiology

Citations(all)

5664

Citations(since 2020)

5127

Cited By

2530

hIndex(all)

24

hIndex(since 2020)

23

i10Index(all)

30

i10Index(since 2020)

30

Email

University Profile Page

Harvard University

Aaron Sarvet Skills & Research Interests

Causal inference

epidemiology

statistics

Top articles of Aaron Sarvet

Improved bounds and inference on optimal regimes

Authors

Julien D Laurendeau,Aaron L Sarvet,Mats J Stensrud

Journal

arXiv preprint arXiv:2404.11510

Published Date

2024/4/17

Point identification of causal effects requires strong assumptions that are unreasonable in many practical settings. However, informative bounds on these effects can often be derived under plausible assumptions. Even when these bounds are wide or cover null effects, they can guide practical decisions based on formal decision theoretic criteria. Here we derive new results on optimal treatment regimes in settings where the effect of interest is bounded. These results are driven by consideration of superoptimal regimes; we define regimes that leverage an individual's natural treatment value, which is typically ignored in the existing literature. We obtain (sharp) bounds for the value function of superoptimal regimes, and provide performance guarantees relative to conventional optimal regimes. As a case study, we consider a commonly studied Marginal Sensitivity Model and illustrate that the superoptimal regime can be identified when conventional optimal regimes are not. We similarly illustrate this property in an instrumental variable setting. Finally, we derive efficient estimators for upper and lower bounds on the superoptimal value in instrumental variable settings, building on recent results on covariate adjusted Balke-Pearl bounds. These estimators are applied to study the effect of prompt ICU admission on survival.

Rejoinder to" Perspectives onharm'in personalized medicine--an alternative perspective"

Authors

Aaron L Sarvet,Mats J Stensrud

Journal

arXiv preprint arXiv:2403.14869

Published Date

2024/3/21

In our original article (Sarvet & Stensrud, 2024), we examine twin definitions of "harm" in personalized medicine: one based on predictions of individuals' unmeasurable response types (counterfactual harm), and another based solely on the observations of experiments (interventionist harm). In their commentary, Mueller & Pearl (2024) (MP) read our review as an argument that "counterfactual logic should [...] be purged from consideration of harm and benefit" and "strongly object [...] that a rational decision maker may well apply the interventional perspective to the exclusion of counterfactual considerations." Here we show that this objection is misguided. We analyze MP's examples and derive a general result, showing that determinations of harm through interventionist and counterfactual analyses will always concur. Therefore, individuals who embrace counterfactual formulations and those who object to their use will make equivalent decisions in uncontroversial settings.

Grace periods in comparative effectiveness studies of sustained treatments

Authors

Kerollos Nashat Wanis,Aaron L Sarvet,Lan Wen,Jason P Block,Sheryl L Rifas-Shiman,James M Robins,Jessica G Young

Journal

Journal of the Royal Statistical Society Series A: Statistics in Society

Published Date

2024/1/22

Researchers are often interested in estimating the effect of sustained use of a treatment on a health outcome. However, adherence to strict treatment protocols can be challenging for individuals in practice and, when non-adherence is expected, estimates of the effect of sustained use may not be useful for decision making. As an alternative, more relaxed treatment protocols which allow for periods of time off treatment (i.e. grace periods) have been considered in pragmatic randomized trials and observational studies. In this article, we consider the interpretation, identification, and estimation of treatment strategies which include grace periods. We contrast natural grace period strategies which allow individuals the flexibility to take treatment as they would naturally do, with stochastic grace period strategies in which the investigator specifies the distribution of treatment utilization. We estimate the effect of initiation of …

Longitudinal incremental propensity score interventions for limited resource settings

Authors

Aaron L Sarvet,Kerollos N Wanis,Jessica G Young,Roberto Hernandez‐Alejandro,Mats J Stensrud

Journal

Biometrics

Published Date

2023/12

Many real‐life treatments are of limited supply and cannot be provided to all individuals in the population. For example, patients on the liver transplant waiting list usually cannot be assigned a liver transplant immediately at the time they reach highest priority because a suitable organ is not immediately available. In settings with limited supply, investigators are often interested in the effects of treatment strategies in which a limited proportion of patients receive an organ at a given time, that is, treatment regimes satisfying resource constraints. Here, we describe an estimand that allows us to define causal effects of treatment strategies that satisfy resource constraints: incremental propensity score interventions (IPSIs) for limited resources. IPSIs flexibly constrain time‐varying resource utilization through proportional scaling of patients' natural propensities for treatment, thereby preserving existing propensity rank ordering …

Causal effects of intervening variables in settings with unmeasured confounding

Authors

Lan Wen,Aaron L Sarvet,Mats J Stensrud

Journal

arXiv preprint arXiv:2305.00349

Published Date

2023/4/29

We present new results on average causal effects in settings with unmeasured exposure-outcome confounding. Our results are motivated by a class of estimands, e.g., frequently of interest in medicine and public health, that are currently not targeted by standard approaches for average causal effects. We recognize these estimands as queries about the average causal effect of an intervening variable. We anchor our introduction of these estimands in an investigation of the role of chronic pain and opioid prescription patterns in the opioid epidemic, and illustrate how conventional approaches will lead unreplicable estimates with ambiguous policy implications. We argue that our altenative effects are replicable and have clear policy implications, and furthermore are non-parametrically identified by the classical frontdoor formula. As an independent contribution, we derive a new semiparametric efficient estimator of the frontdoor formula with a uniform sample boundedness guarantee. This property is unique among previously-described estimators in its class, and we demonstrate superior performance in finite-sample settings. Theoretical results are applied with data from the National Health and Nutrition Examination Survey.

Conditional separable effects

Authors

Mats J Stensrud,James M Robins,Aaron Sarvet,Eric J Tchetgen Tchetgen,Jessica G Young

Journal

Journal of the American Statistical Association

Published Date

2023/10/2

Researchers are often interested in treatment effects on outcomes that are only defined conditional on posttreatment events. For example, in a study of the effect of different cancer treatments on quality of life at end of follow-up, the quality of life of individuals who die during the study is undefined. In these settings, naive contrasts of outcomes conditional on posttreatment events are not average causal effects, even in randomized experiments. Therefore, the effect in the principal stratum of those who would have the same value of the posttreatment variable regardless of treatment (such as the survivor average causal effect) is often advocated for causal inference. While principal stratum effects are average causal effects, they refer to a subset of the population that cannot be observed and may not exist. Therefore, it is not clear how these effects inform decisions or policies. Here we propose the conditional separable …

Mats Stensrud, Vanessa Didelez and Aaron Sarvet's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to …

Authors

Mats J Stensrud,Vanessa Didelez,Aaron L Sarvet

Journal

Journal of the Royal Statistical Society Series A: Statistics in Society

Published Date

2023/4

Despite an increasing reliance on fully-automated algorithmic decision-making in our day-to-day lives, humans still make consequential decisions. While the existing literature focuses on the bias and fairness of algorithmic recommendations, an overlooked question is whether they improve human decisions. We develop a general statistical methodology for experimentally evaluating the causal impacts of algorithmic recommendations on human decisions. We also examine whether algorithmic recommendations improve the fairness of human decisions and derive the optimal decision rules under various settings. We apply the proposed methodology to the first-ever randomized controlled trial that evaluates the pretrial Public Safety Assessment in the United States criminal justice system. Our analysis of the preliminary data shows that providing the PSA to the judge has little overall impact on the judge’s decisions …

Separable effects for adherence

Authors

Kerollos Nashat Wanis,Mats Julius Stensrud,Aaron Leor Sarvet

Journal

arXiv preprint arXiv:2309.13751

Published Date

2023/9/24

Comparing different medications is complicated when adherence to these medications differs. We can overcome the adherence issue by assessing effectiveness under sustained use, as in the usual causal `per-protocol' estimand. However, when sustained use is challenging to satisfy in practice, the usefulness of this estimand can be limited. Here we propose a different class of estimands: separable effects for adherence. These estimands compare modified medications, holding fixed a component responsible for non-adherence. Under assumptions about treatment components' mechanisms of effect, the separable effects estimand can eliminate differences in adherence. These assumptions are amenable to interrogation by subject-matter experts and can be evaluated using causal graphs. We describe an algorithm for constructing causal graphs for separable effects, illustrate how these graphs can be used to reason about assumptions required for identification, and provide semi-parametric weighted estimators.

Aaron Sarvet and Mats Stensrud's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety …

Authors

Aaron L Sarvet,Mats J Stensrud

Journal

Journal of the Royal Statistical Society Series A: Statistics in Society

Published Date

2023/4

Despite an increasing reliance on fully-automated algorithmic decision-making in our day-to-day lives, humans still make consequential decisions. While the existing literature focuses on the bias and fairness of algorithmic recommendations, an overlooked question is whether they improve human decisions. We develop a general statistical methodology for experimentally evaluating the causal impacts of algorithmic recommendations on human decisions. We also examine whether algorithmic recommendations improve the fairness of human decisions and derive the optimal decision rules under various settings. We apply the proposed methodology to the first-ever randomized controlled trial that evaluates the pretrial Public Safety Assessment in the United States criminal justice system. Our analysis of the preliminary data shows that providing the PSA to the judge has little overall impact on the judge’s decisions …

Discussion of ‘Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment’, part 2

Authors

Mats Julius Stensrud,Vanessa Didelez,Aaron Leor Sarvet

Journal

Journal of the Royal Statistical Society Series A: Statistics in Society

Published Date

2023/2/14

Imai et al.(IJGHS) have conducted a timely experiment on evaluating a decision support algorithm. However, we are concerned by their choice of estimands which, even if they appear plausible at first, rely on notions and assumptions for which we cannot ever obtain empirical evidence. We suggest an alternative.

Cannabis use, cannabis use disorder and mental health disorders among pregnant and postpartum women in the US: A nationally representative study

Authors

Qiana L Brown,Dvora Shmulewitz,Aaron L Sarvet,Kelly C Young-Wolff,Tyriesa Howard,Deborah S Hasin

Journal

Drug and Alcohol Dependence

Published Date

2023/7/1

BackgroundCannabis use and cannabis use disorder (CUD) are associated with mental health disorders, however the extent of this matter among pregnant and recently postpartum (e.g., new moms) women in the US is unknown. Associations between cannabis use, DSM-5 CUD and DSM-5 mental health disorders (mood, anxiety, personality and post-traumatic stress disorders) were examined among a nationally representative sample of pregnant and postpartum women.MethodsThe 2012–2013 National Epidemiologic Survey on Alcohol and Related Conditions–III was used to examine associations between past-year cannabis use, CUD and mental health disorders. Weighted logistic regression models were used to estimate unadjusted and adjusted odds ratios (aORs). The sample (N=1316) included 414 pregnant and 902 postpartum women (pregnant in the past year), aged 18–44 years old.ResultsThe …

Perspectives on harm in personalized medicine

Authors

Aaron L Sarvet,Mats J Stensrud

Journal

arXiv preprint arXiv:2302.01371

Published Date

2023/2/2

Avoiding harm is an uncontroversial aim of personalized medicine and other epidemiologic initiatives. However, the precise mathematical translation of "harm" is disputable. Here we use a formal causal language to study common, but distinct, definitions of "harm". We clarify that commitment to a definition of harm has important practical and philosophical implications for decision making. We relate our practical and philosophical considerations to ideas from medical ethics and legal practice.

Incidence and outcomes of non–ventilator-associated hospital-acquired pneumonia in 284 US hospitals using electronic surveillance criteria

Authors

Barbara E Jones,Aaron L Sarvet,Jian Ying,Robert Jin,McKenna R Nevers,Sarah E Stern,Aileen Ocho,Caroline McKenna,Laura E McLean,Matthew A Christensen,Russell E Poland,Jeffrey S Guy,Kenneth E Sands,Chanu Rhee,Jessica G Young,Michael Klompas

Journal

JAMA Network Open

Published Date

2023/5/1

ImportanceNon–ventilator-associated hospital-acquired pneumonia (NV-HAP) is a common and deadly hospital-acquired infection. However, inconsistent surveillance methods and unclear estimates of attributable mortality challenge prevention.ObjectiveTo estimate the incidence, variability, outcomes, and population attributable mortality of NV-HAP.Design, Setting, and ParticipantsThis cohort study retrospectively applied clinical surveillance criteria for NV-HAP to electronic health record data from 284 US hospitals. Adult patients admitted to the Veterans Health Administration hospital from 2015 to 2020 and HCA Healthcare hospitals from 2018 to 2020 were included. The medical records of 250 patients who met the surveillance criteria were reviewed for accuracy.ExposuresNV-HAP, defined as sustained deterioration in oxygenation for 2 or more days in a patient who was not ventilated concurrent with abnormal …

Interpretational errors in statistical causal inference

Authors

Aaron L Sarvet,Mats J Stensrud,Lan Wen

Journal

arXiv preprint arXiv:2312.07610

Published Date

2023/12/11

We formalize an interpretational error that is common in statistical causal inference, termed identity slippage. This formalism is used to describe historically-recognized fallacies, and analyse a fast-growing literature in statistics and applied fields. We conducted a systematic review of natural language claims in the literature on stochastic mediation parameters, and documented extensive evidence of identity slippage in applications. This framework for error detection is applicable whenever policy decisions depend on the accurate interpretation of statistical results, which is nearly always the case. Therefore, broad awareness of identity slippage will aid statisticians in the successful translation of data into public good.

Mats J Stensrud and Aaron L. Sarvet's contribution to the Discussion of'Assumption-lean inference for generalised linear model parameters' by Vansteelandt and Dukes

Authors

Mats Julius Stensrud,Aaron Leor Sarvet

Journal

Journal Of The Royal Statistical Society Series B-Statistical Methodology

Published Date

2022/7/1

Inference for the parameters indexing generalised linear models is routinely based on the assumption that the model is correct and a priori specified. This is unsatisfactory because the chosen model is usually the result of a data-adaptive model selection process, which may induce excess uncertainty that is not usually acknowledged. Moreover, the assumptions encoded in the chosen model rarely represent some a priori known, ground truth, making standard inferences prone to bias, but also failing to give a pure reflection of the information that is contained in the data. Inspired by developments on assumption-free inference for so-called projection parameters, we here propose novel nonparametric definitions of main effect estimands and effect modification estimands. These reduce to standard main effect and effect modification parameters in generalised linear models when these models are correctly specified, but have the advantage that they continue to capture respectively the (conditional) association between two variables, or the degree to which two variables interact in their association with outcome, even when these models are misspecified. We achieve an assumption-lean

Without commitment to an ontology, there could be no causal inference

Authors

Aaron L Sarvet,Mats J Stensrud

Journal

Epidemiology

Published Date

2022/5/1

We interpret methods development in causal inference as a process of constructing tools for answering internal questions (following Carnap 1). Here,“internal” means that these questions can only be understood with respect to an ontology (otherwise their meaning is unclear). Specifically, we argue that epidemiologists will be aided by an understanding of ontologies when faced with the difficult tasks of estimand selection and statistical model specification. To illustrate our points, we refer to the article by Davis-Plourde et al. 2 concerning causal inference methods in dementia research.

Comparison of rates of type 2 diabetes in adults and children treated with anticonvulsant mood stabilizers

Authors

Jenny W Sun,Jessica G Young,Aaron L Sarvet,L Charles Bailey,William J Heerman,David M Janicke,Pi-I Debby Lin,Sengwee Toh,Jason P Block

Journal

JAMA Network Open

Published Date

2022/4/1

ImportanceAnticonvulsant mood stabilizer treatment is associated with an increased risk of weight gain, but little is known about the risk of developing type 2 diabetes (T2D).ObjectiveTo evaluate the comparative safety of anticonvulsant mood stabilizers on risk of T2D in adults and children by emulating a target trial.Design, Setting, and ParticipantsThis observational cohort study used data from IBM MarketScan (2010-2019), with a 5-year follow-up period. The nationwide sample of US commercially insured patients included children (aged 10-19 years) and adults (aged 20-65 years) who initiated anticonvulsant mood stabilizer treatment. Data were analyzed from August 2020 to May 2021.ExposuresInitiation and continuation of carbamazepine, lamotrigine, oxcarbazepine, or valproate.Main Outcomes and MeasuresOnset of T2D during follow-up. Weighted pooled logistic regression was used to estimate the …

Optimal regimes for algorithm-assisted human decision-making

Authors

Mats J Stensrud,Julien Laurendeau,Aaron L Sarvet

Journal

arXiv preprint arXiv:2203.03020

Published Date

2022/3/6

We introduce optimal regimes for algorithm-assisted human decision-making. Such regimes are decision functions of measured pre-treatment variables and enjoy a "superoptimality" property whereby they are guaranteed to outperform conventional optimal regimes currently considered in the literature. A key feature of these superoptimal regimes is the use of natural treatment values as input to the decision function. Importantly, identification of the superoptimal regime and its value require exactly the same assumptions as identification of conventional optimal regimes in several common settings, including instrumental variable settings. As an illustration, we study superoptimal regimes in an example that has been presented in the optimal regimes literature.

Causal Inference with Limited Resources

Authors

Aaron L Sarvet

Published Date

2022

Constraints on treatment resources present problems in many practical settings. For example, during the current coronavirus disease 2019 pandemic, several health care systems have experienced shortages of ventilators, protective equipment and personnel. Similarly, patients with organ failures are allocated to waiting lists because the number of organ transplants are scarce. Investigators are often interested in causal effects of different treatment strategies in these settings, but classical causal inference methods often fail to explicitly consider the resource constraints. This thesis develops a suite of methods for resource limited settings, focusing on the tasks of defining and identifying new causal estimands. In Chapters 1 and 2, this thesis develops a general framework for evaluating the effect of counterfactual treatment allocation and prioritization regimes for limited resource settings. Crucially, this thesis represents …

Association of cannabis use–related predictor variables and self-reported psychotic disorders: US adults, 2001–2002 and 2012–2013

Authors

Ofir Livne,Dvora Shmulewitz,Aaron L Sarvet,Melanie M Wall,Deborah S Hasin

Journal

American journal of psychiatry

Published Date

2022/1

Objective The authors sought to determine the association of cannabis indicators with self-reported psychotic disorders in the U.S. general population. Methods Participants were from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC; 2001–2002; N=43,093) and NESARC-III (2012–2013; N=36,309). Logistic regression was used to estimate standardized prevalences of past-year self-reported psychotic disorders within each survey and to evaluate the association of past-year self-reported psychotic disorders with indicators of nonmedical cannabis use (any use; frequent use [at least three times/week], daily/near-daily use, and DSM-IV cannabis use disorder) compared with those with no past-year nonmedical cannabis use. Whether the strength of associations differed between surveys was indicated by difference-in-difference tests (between-survey contrasts) and ratios of odds ratios …

See List of Professors in Aaron Sarvet University(Harvard University)

Aaron Sarvet FAQs

What is Aaron Sarvet's h-index at Harvard University?

The h-index of Aaron Sarvet has been 23 since 2020 and 24 in total.

What are Aaron Sarvet's top articles?

The articles with the titles of

Improved bounds and inference on optimal regimes

Rejoinder to" Perspectives onharm'in personalized medicine--an alternative perspective"

Grace periods in comparative effectiveness studies of sustained treatments

Longitudinal incremental propensity score interventions for limited resource settings

Causal effects of intervening variables in settings with unmeasured confounding

Conditional separable effects

Mats Stensrud, Vanessa Didelez and Aaron Sarvet's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to …

Separable effects for adherence

...

are the top articles of Aaron Sarvet at Harvard University.

What are Aaron Sarvet's research interests?

The research interests of Aaron Sarvet are: Causal inference, epidemiology, statistics

What is Aaron Sarvet's total number of citations?

Aaron Sarvet has 5,664 citations in total.

    academic-engine

    Useful Links