Modified treatment policy effect estimation with weighted energy distance

Recently, the International Conference on Harmonisation finalized an estimand framework for randomized trials that was adopted by regulatory bodies worldwide. The framework introduced five strategies for handling post-randomization events; namely the treatment policy, composite variable, while on treatment, hypothetical and principal stratum estimands. We describe an illustrative example to elucidate the difference between these five strategies for handling intercurrent events and provide an estimation technique for each. Specifically, we consider the intercurrent event of treatment discontinuation and introduce potential outcome notation to describe five estimands and corresponding estimators: (1) an intention-to-treat estimator of the total effect of a treatment policy; (2) an intention-to-treat estimator of a composite of the outcome and remaining on treatment; (3) a per-protocol estimator of the outcome in individuals observed to remain on treatment; (4) a g-computation estimator of a hypothetical scenario that all individuals remain on treatment; and (5) a principal stratum estimator of the treatment effect in individuals who would remain on treatment under the experimental condition. Additional insight is provided by defining situations where certain estimands are equal, and by studying the while on treatment strategy under repeated outcome measures. We highlight relevant causal inference literature to enable adoption in practice.

The authors develop a simulation model to predict the effects on quit rates and cost-effectiveness of different smoking treatment policies. A decision theoretic model of quit behavior is first developed that incorporates the decision to quit and the choice of treatment. A policy model then examines the effect on quit attempts and quit rates of policies to cover the costs of different combinations of treatments and to require health care providers to conduct brief interventions. The model incorporates substitution between treatments and effects of policies on treatment effectiveness. The cost per quit is also calculated for each policy. The model of quit behavior predicts a 1-year quit rate of 4.5% for the population of smokers. The policy model predicts a 37% increase in quit rates from a policy that combines mandated brief interventions with coverage of all proven tobacco treatments. Smaller effects are predicted from policies that provide more restricted coverage of treatments, especially those limited to behavioral treatment. Payments for brief interventions alone increase quit rates by about 7%. Brief intervention and behavioral therapy policies had lower costs per quit but yield substantially fewer additional quits than policies that cover pharmacotherapy. There is, however, considerable variation around these estimates depending on assumptions about the effects of policy on treatment use, substitution between treatments, and treatment effectiveness. Tobacco treatment policies, especially those with broad and flexible coverage, have the potential to substantially increase smoking quit rates. However, further research is needed on the effect of payment policies on the use and effectiveness of tobacco treatments.

Reinforcement learning (RL) demonstrates promise for developing effective treatment policies in critical care settings. However, existing RL methods often require large and comprehensive patient data sets and do not readily lend themselves to settings in which certain patient subpopulations are severely underrepresented. In this study, we develop a new method, noisy Bayesian policy updates (NBPU), for selecting high-performing reinforcement learning–based treatment policies for underrepresented patient subpopulations using limited observations. Our method uses variational inference to learn a probability distribution over treatment policies based on a reference patient subpopulation for which sufficient data are available. It then exploits limited data from an underrepresented patient subpopulation to update this probability distribution and adapts its recommendations to this subpopulation. We demonstrate our method’s utility on a data set of ICU patients receiving intravenous blood anticoagulant medication. Our results show that NBPU outperforms state-of-the-art methods in terms of both selecting effective treatment policies for patients with nontypical clinical characteristics and predicting the corresponding policies’ performance for these patients.

The paper presents an exploratory study of a road safety policy index developed for Norway. The index consists of ten road safety measures for which data on their use from 1980 to 2021 are available. The ten measures were combined into an index which had an initial value of 50 in 1980 and increased to a value of 185 in 2021. To assess the application of the index in evaluating the effects of road safety policy, negative binomial regression models and multivariate time series models were developed for traffic fatalities, fatalities and serious injuries, and all injuries. The coefficient for the policy index was negative, indicating the road safety policy has contributed to reducing the number of fatalities and injuries. The size of this contribution can be estimated by means of at least three estimators that do not always produce identical values. There is little doubt about the sign of the relationship: a stronger road safety policy (as indicated by index values) is associated with a larger decline in fatalities and injuries. A precise quantification is, however, not possible. Different estimators of effect, all of which can be regarded as plausible, yield different results.

Summary We provide adaptive inference methods, based on $\ell _1$ regularization, for regular (semiparametric) and nonregular (nonparametric) linear functionals of the conditional expectation function. Examples of regular functionals include average treatment effects, policy effects, and derivatives. Examples of nonregular functionals include average treatment effects, policy effects, and derivatives conditional on a covariate subvector fixed at a point. We construct a Neyman orthogonal equation for the target parameter that is approximately invariant to small perturbations of the nuisance parameters. To achieve this property, we include the Riesz representer for the functional as an additional nuisance parameter. Our analysis yields weak ‘double sparsity robustness’: either the approximation to the regression or the approximation to the representer can be ‘completely dense’ as long as the other is sufficiently ‘sparse’. Our main results are nonasymptotic and imply asymptotic uniform validity over large classes of models, translating into honest confidence bands for both global and local parameters.

Modeling the Impact of Smoking-Cessation Treatment Policies on Quit Rates

We used organizational justice theory to explore reactions to employer-sponsored alcohol testing and alcohol treatment policies among a sample (N = 1,777) of the employed public in a western state. Level of alcohol use and safety-sensitivity of the job were related to the perceived fairness of alcohol testing. In addition, voluntary treatment policies were rated more positively than coerced or monitored policies in terms of fairness and organizational attractiveness. Alcohol use moderated the effects of treatment policy on perceived fairness and organizational attractiveness, although the effect sizes were small. These results support the use of organizational justice theory to explain reactions to organizational alcohol testing and treatment and provide a basis for future research in this area.

This paper develops a simple model of the war against illegal drugs in producer and consumer countries. Our analysis shows how the equilibrium quantity of illegal drugs, as well as their price, depends on key parameters of the model, among them the price elasticity of demand, and the effectiveness of the resources allocated to enforcement and prevention and treatment policies. Importantly, this paper studies the trade-off faced by drug consumer country's government between prevention policies (aimed at reducing the demand for illegal drugs) and enforcement policies (aimed at reducing the production and trafficking of illegal drugs in producer countries). We use available data for the war against cocaine production and trafficking in Colombia, and that against consumption in the U.S. in order to calibrate the unobservable parameters of the model. Among these are the effectiveness of prevention and treatment policies in reducing the demand for cocaine; the relative effectiveness of interdiction efforts at reducing the amount of cocaine reaching consumer countries; and the cost of illegal drug production and trafficking activities in producer countries.

This paper develops a simple model of the war against illegal drugs in producer and consumer countries. Our analysis shows how the equilibrium quantity of illegal drugs, as well as their price, depends on key parameters of the model, among them the price elasticity of demand, and the effectiveness of the resources allocated to enforcement and prevention and treatment policies. Importantly, this paper studies the trade-off faced by drug consumer country`s government between prevention policies (aimed at reducing the demand for illegal drugs) and enforcement policies (aimed at reducing the production and trafficking of illegal drugs in producer countries). We use available data for the war against cocaine production and trafficking in Colombia, and that against consumption in the U.S. in order to calibrate the unobservable parameters of the model. Among these are the effectiveness of prevention and treatment policies in reducing the demand for cocaine; the relative effectiveness of interdiction efforts at reducing the amount of cocaine reaching consumer countries; and the cost of illegal drug production and trafficking activities in producer countries.

This study examines the effect of past tobacco control policies and projects the effect of future policies on smoking and snus use prevalence and associated premature mortality in Sweden. The established SimSmoke model was adapted with population, smoking rates and tobacco control policy data from Sweden. SimSmoke evaluates the effect of taxes, smoke-free air, mass media, marketing bans, warning labels, cessation treatment and youth access policies on smoking and snus prevalence and the number of deaths attributable to smoking and snus use by gender from 2010 to 2040. Sweden SimSmoke estimates that significant inroads to reducing smoking and snus prevalence and premature mortality can be achieved through tax increases, especially when combined with other policies. Smoking prevalence can be decreased by as much as 26% in the first few years, reaching a 37% reduction within 30 years. Without effective tobacco control policies, almost 54 500 lives will be lost in Sweden due to tobacco use by the year 2040. Besides presenting the benefits of a comprehensive tobacco control strategy, the model identifies gaps in surveillance and evaluation that can help better focus tobacco control policy in Sweden.

Objectives: To develop a simulation model to predict the effects of different smoking treatment policies on quit rates, smoking rates, and smoking attributable deaths. Methods: We first develop a decision...

The ICH E9(R1) addendum provides guidelines on accounting for intercurrent events in clinical trials using the estimands framework. However, there has been limited attention to the estimands framework for meta-analysis. Using treatment switching, a well-known intercurrent event that occurs frequently in oncology, we conducted a simulation study to explore the bias introduced by pooling together estimates targeting different estimands in a meta-analysis of randomized clinical trials (RCTs) that allowed treatment switching. We simulated overall survival data of a collection of RCTs that allowed patients in the control group to switch to the intervention treatment after disease progression under fixed effects and random effects models. For each RCT, we calculated effect estimates for a treatment policy estimand that ignored treatment switching, and a hypothetical estimand that accounted for treatment switching either by fitting rank-preserving structural failure time models or by censoring switchers. Then, we performed random effects and fixed effects meta-analyses to pool together RCT effect estimates while varying the proportions of trials providing treatment policy and hypothetical effect estimates. We compared the results of meta-analyses that pooled different types of effect estimates with those that pooled only treatment policy or hypothetical estimates. We found that pooling estimates targeting different estimands results in pooled estimators that do not target any estimand of interest, and that pooling estimates of varying estimands can generate misleading results, even under a random effects model. Adopting the estimands framework for meta-analysis may improve alignment between meta-analytic results and the clinical research question of interest.

Stage I-II Hodgkin's Disease: Current Therapeutic Options and Recommendations

Tobacco use has shifted increasingly from cigarettes to other products. While the focus has been mostly on cigarette-oriented policies, it is important to gauge the effects of policies targeting other products. We review and critique the literature on how policies affect smokeless tobacco (ST). We conducted a search of the literature on tobacco control policies as they relate to ST use, focusing on tobacco taxes, smoke-free air laws, media campaigns, advertising restrictions, health warnings, cessation treatment policies, and youth access policies. Findings from 78 total studies are summarized. ST taxes, media campaigns, health warnings, and cessation treatment policies were found to be effective tools in reducing ST use. Evidence on the effects of current youth access policies is less strong. Studies have not yet been conducted on marketing or product content restrictions, but the literature indicates that product marketing, through advertising, packaging, flavorings, and extension of cigarette brands, plays an important role in ST use. Although the evidence base is less established for ST policies than for cigarette policies, the existing literature indicates ST use responds to tobacco control policies. Policies should be structured in a way that aims to reduce all tobacco use while at the same time increasing the likelihood that continuing tobacco users use the least risky products. Studies find that policies targeting smoking and policies targeting smokeless products affect smokeless use, but studies are needed to examine the effect of policies on the transitions between cigarette and smokeless use.

Michigan has implemented several of the tobacco control policies recommended by the World Health Organization MPOWER goals. We consider the effect of those policies and additional policies consistent with MPOWER goals on smoking prevalence and smoking-attributable deaths (SADs). The SimSmoke tobacco control policy simulation model is used to examine the effect of past policies and a set of additional policies to meet the MPOWER goals. The model is adapted to Michigan using state population, smoking, and policy data starting in 1993. SADs are estimated using standard attribution methods. Upon validating the model, SimSmoke is used to distinguish the effect of policies implemented since 1993 against a counterfactual with policies kept at their 1993 levels. The model is then used to project the effect of implementing stronger policies beginning in 2014. SimSmoke predicts smoking prevalence accurately between 1993 and 2010. Since 1993, a relative reduction in smoking rates of 22 % by 2013 and of 30 % by 2054 can be attributed to tobacco control policies. Of the 22 % reduction, 44 % is due to taxes, 28 % to smoke-free air laws, 26 % to cessation treatment policies, and 2 % to youth access. Moreover, 234,000 SADs are projected to be averted by 2054. With additional policies consistent with MPOWER goals, the model projects that, by 2054, smoking prevalence can be further reduced by 17 % with 80,000 deaths averted relative to the absence of those policies. Michigan SimSmoke shows that tobacco control policies, including cigarette taxes, smoke-free air laws, and cessation treatment policies, have substantially reduced smoking and SADs. Higher taxes, strong mass media campaigns, and cessation treatment policies would further reduce smoking prevalence and SADs.