Estimating optimal dynamic treatment strategies under resource constraints using dynamic marginal structural models

Murray, Eleanor; Hernán, Miguel A.; Shahn, Zach

doi:10.1002/sim.9107

Cited by 3 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an alternative, they propose to estimate the regime which maximises the average conditional median

normal𝔼 ({}, Median ({}, Y^{*} false(d false) false| X))

. Other criterion used to define an optimal regime include maximising efficacy subject to a constraint on side‐effects (Linn et al , 2015; Laber et al , 2018; Wang et al , 2018) or resources (Caniglia et al , 2021; Luedtke & van der Laan, 2016c); maximising choice (Ertefaie et al , 2016; Fard & Pineau, 2011; Laber et al , 2014; Lizotte & Laber, 2016; Wu, 2016); and maximising patient (latent) utility (Butler et al , 2018; Luckett et al , 2020).…”

Section: Setup and Problem Formulationmentioning

confidence: 99%

Optimal Treatment Regimes: A Review and Empirical Comparison

Chen

Laber

et al. 2023

Int Statistical Rev

View full text Add to dashboard Cite

SummaryA treatment regime is a sequence of decision rules, one per decision point, that maps accumulated patient information to a recommended intervention. An optimal treatment regime maximises expected cumulative utility if applied to select interventions in a population of interest. As a treatment regime seeks to improve the quality of healthcare by individualising treatment, it can be viewed as an approach to formalising precision medicine. Increased interest and investment in precision medicine has led to a surge of methodological research focusing on estimation and evaluation of optimal treatment regimes from observational and/or randomised studies. These methods are becoming commonplace in biomedical research, although guidance about how to choose among existing methods in practice has been somewhat limited. The purpose of this review is to describe some of the most commonly used methods for estimation of an optimal treatment regime, and to compare these estimators in a series of simulation experiments and applications to real data. The results of these simulations along with the theoretical/methodological properties of these estimators are used to form recommendations for applied researchers.

show abstract

“…As an alternative, they propose to estimate the regime which maximises the average conditional median

normal𝔼 ({}, Median ({}, Y^{*} false(d false) false| X))

Section: Setup and Problem Formulationmentioning

confidence: 99%

Optimal Treatment Regimes: A Review and Empirical Comparison

Chen

Laber

et al. 2023

Int Statistical Rev

View full text Add to dashboard Cite

show abstract

“…Much has been written about random DTRs, but almost exclusively with the goal of statistical inference in the study population, and with treatments that are either discrete (e.g., when to start therapy) or one‐dimensional (e.g., minutes exercised). There is also a recent, relevant literature on optimal DTRs under resource constraints, but these methods tend to assume that the amount of available treatment is fixed in advance (Caniglia et al., 2021), rather than arriving randomly as with deceased‐donor organs, or that the dependence of random treatment assignment on confounders is fixed (Sarvet et al., 2020), which likely does not hold when transporting inference to a new allocation policy era.…”

Section: Introductionmentioning

confidence: 99%

Transportability of Causal Inference under Random Dynamic Treatment Regimes for Kidney–Pancreas Transplantation

et al. 2023

View full text Add to dashboard Cite

A difficult decision for patients in need of kidney–pancreas transplant is whether to seek a living kidney donor or wait to receive both organs from one deceased donor. The framework of dynamic treatment regimes (DTRs) can inform this choice, but a patient‐relevant strategy such as “wait for deceased‐donor transplant” is ill‐defined because there are multiple versions of treatment (i.e., wait times, organ qualities). Existing DTR methods average over the distribution of treatment versions in the data, estimating survival under a “representative intervention.” This is undesirable if transporting inferences to a target population such as patients today, who experience shorter wait times thanks to evolutions in allocation policy. We, therefore, propose the concept of a generalized representative intervention (GRI): a random DTR that assigns treatment version by drawing from the distribution among strategy compliers in the target population (e.g., patients today). We describe an inverse‐probability‐weighted product‐limit estimator of survival under a GRI that performs well in simulations and can be implemented in standard statistical software. For continuous treatments (e.g., organ quality), weights are reformulated to depend on probabilities only, not densities. We apply our method to a national database of kidney–pancreas transplant candidates from 2001–2020 to illustrate that variability in transplant rate across years and centers results in qualitative differences in the optimal strategy for patient survival.

show abstract

“…However, methods for point treatment settings (Athey & Wager, 2021; Luedtke & van der Laan, 2016) are of limited utility when patients have sequential opportunities for treatment and constraints delay, rather than prevent, treatment. Additionally, more general constrained optimization strategies (e.g., Caniglia et al., 2021) will identify DTRs that contradict extant prioritization strategies (e.g., waiting lists) that a policy‐maker wishes to preserve. In contrast, Boatman and Vock (2018) consider estimating the value of regimes for patients who are offered organs (lungs) on a transplant waiting list where the structure of the waiting list is left unmanipulated, but suggest intensive simulation‐based estimators with unknown statistical properties and require that the waiting list algorithm of the observed data‐generating mechanism is precisely known to the analyst.…”

Section: Introductionmentioning

confidence: 99%

Longitudinal Incremental Propensity Score Interventions for Limited Resource Settings

et al. 2023

View full text Add to dashboard Cite

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 compared to the status quo. We derive a simple class of inverse‐probability‐weighted estimators, and we apply one such estimator to evaluate the effect of restricting or expanding utilization of “increased risk” liver organs to treat patients with end‐stage liver disease.

show abstract

Estimating optimal dynamic treatment strategies under resource constraints using dynamic marginal structural models

Cited by 3 publications

References 26 publications

Optimal Treatment Regimes: A Review and Empirical Comparison

Optimal Treatment Regimes: A Review and Empirical Comparison

Transportability of Causal Inference under Random Dynamic Treatment Regimes for Kidney–Pancreas Transplantation

Longitudinal Incremental Propensity Score Interventions for Limited Resource Settings

Contact Info

Product

Resources

About