Developing an individualized treatment rule for Veterans with major depressive disorder using electronic health records.

Efforts to develop an individualized treatment rule (ITR) to optimize major depressive disorder (MDD) treatment with antidepressant medication (ADM), psychotherapy, or combined ADM-psychotherapy have been hampered by small samples, small predictor sets, and suboptimal analysis methods. Analyses of large administrative databases designed to approximate experiments followed iteratively by pragmatic trials hold promise for resolving these problems. The current report presents a proof-of-concept study using electronic health records (EHR) of n = 43,470 outpatients beginning MDD treatment in Veterans Health Administration Primary Care Mental Health Integration (PC-MHI) clinics, which offer access not only to ADMs but also psychotherapy and combined ADM-psychotherapy. EHR and geospatial databases were used to generate an extensive baseline predictor set (5,865 variables). The outcome was a composite measure of at least one serious negative event (suicide attempt, psychiatric emergency department visit, psychiatric hospitalization, suicide death) over the next 12 months. Best-practices methods were used to adjust for nonrandom treatment assignment and to estimate a preliminary ITR in a 70% training sample and to evaluate the ITR in the 30% test sample. Statistically significant aggregate variation was found in overall probability of the outcome related to baseline predictors (AU-ROC = 0.68, S.E. = 0.01), with test sample outcome prevalence of 32.6% among the 5% of patients having highest predicted risk compared to 7.1% in the remainder of the test sample. The ITR found that psychotherapy-only was the optimal treatment for 56.0% of patients (roughly 20% lower risk of the outcome than if receiving one of the other treatments) and that treatment type was unrelated to outcome risk among other patients. Change in aggregate treatment costs of implementing this ITR would be negligible, as 16.1% fewer patients would be prescribed ADMs and 2.9% more would receive psychotherapy. A pragmatic trial would be needed to confirm the accuracy of the ITR.

Confidence intervals for policy evaluation in adaptive experiments.

Adaptive experimental designs can dramatically improve efficiency in randomized trials. But with adaptively collected data, common estimators based on sample means and inverse propensity-weighted means can be biased or heavy-tailed. This poses statistical challenges, in particular when the experimenter would like to test hypotheses about parameters that were not targeted by the data-collection mechanism. In this paper, we present a class of test statistics that can handle these challenges. Our approach is to adaptively reweight the terms of an augmented inverse propensity-weighting estimator to control the contribution of each term to the estimator's variance. This scheme reduces overall variance and yields an asymptotically normal test statistic. We validate the accuracy of the resulting estimates and their CIs in numerical experiments and show that our methods compare favorably to existing alternatives in terms of mean squared error, coverage, and CI size.

High-dimensional regression adjustments in randomized experiments.

We study the problem of treatment effect estimation in randomized experiments with high-dimensional covariate information and show that essentially any risk-consistent regression adjustment can be used to obtain efficient estimates of the average treatment effect. Our results considerably extend the range of settings where high-dimensional regression adjustments are guaranteed to provide valid inference about the population average treatment effect. We then propose cross-estimation, a simple method for obtaining finite-sample-unbiased treatment effect estimates that leverages high-dimensional regression adjustments. Our method can be used when the regression model is estimated using the lasso, the elastic net, subset selection, etc. Finally, we extend our analysis to allow for adaptive specification search via cross-validation and flexible nonparametric regression adjustments with machine-learning methods such as random forests or neural networks.

Estimated Average Treatment Effect of Psychiatric Hospitalization in Patients With Suicidal Behaviors: A Precision Treatment Analysis.

Importance: Psychiatric hospitalization is the standard of care for patients presenting to an emergency department (ED) or urgent care (UC) with high suicide risk. However, the effect of hospitalization in reducing subsequent suicidal behaviors is poorly understood and likely heterogeneous. Objectives: To estimate the association of psychiatric hospitalization with subsequent suicidal behaviors using observational data and develop a preliminary predictive analytics individualized treatment rule accounting for heterogeneity in this association across patients. Design, Setting, and Participants: A machine learning analysis of retrospective data was conducted. All veterans presenting with suicidal ideation (SI) or suicide attempt (SA) from January 1, 2010, to December 31, 2015, were included. Data were analyzed from September 1, 2022, to March 10, 2023. Subgroups were defined by primary psychiatric diagnosis (nonaffective psychosis, bipolar disorder, major depressive disorder, and other) and suicidality (SI only, SA in past 2-7 days, and SA in past day). Models were trained in 70.0% of the training samples and tested in the remaining 30.0%. Exposures: Psychiatric hospitalization vs nonhospitalization. Main Outcomes and Measures: Fatal and nonfatal SAs within 12 months of ED/UC visits were identified in administrative records and the National Death Index. Baseline covariates were drawn from electronic health records and geospatial databases. Results: Of 196 610 visits (90.3% men; median [IQR] age, 53 [41-59] years), 71.5% resulted in hospitalization. The 12-month SA risk was 11.9% with hospitalization and 12.0% with nonhospitalization (difference, -0.1%; 95% CI, -0.4% to 0.2%). In patients with SI only or SA in the past 2 to 7 days, most hospitalization was not associated with subsequent SAs. For patients with SA in the past day, hospitalization was associated with risk reductions ranging from -6.9% to -9.6% across diagnoses. Accounting for heterogeneity, hospitalization was associated with reduced risk of subsequent SAs in 28.1% of the patients and increased risk in 24.0%. An individualized treatment rule based on these associations may reduce SAs by 16.0% and hospitalizations by 13.0% compared with current rates. Conclusions and Relevance: The findings of this study suggest that psychiatric hospitalization is associated with reduced average SA risk in the immediate aftermath of an SA but not after other recent SAs or SI only. Substantial heterogeneity exists in these associations across patients. An individualized treatment rule accounting for this heterogeneity could both reduce SAs and avert hospitalizations.

Low-intensity fires mitigate the risk of high-intensity wildfires in California's forests.

The increasing frequency of severe wildfires demands a shift in landscape management to mitigate their consequences. The role of managed, low-intensity fire as a driver of beneficial fuel treatment in fire-adapted ecosystems has drawn interest in both scientific and policy venues. Using a synthetic control approach to analyze 20 years of satellite-based fire activity data across 124,186 square kilometers of forests in California, we provide evidence that low-intensity fires substantially reduce the risk of future high-intensity fires. In conifer forests, the risk of high-intensity fire is reduced by 64.0% [95% confidence interval (CI): 41.2 to 77.9%] in areas recently burned at low intensity relative to comparable unburned areas, and protective effects last for at least 6 years (lower bound of one-sided 95% CI: 6 years). These findings support a policy transition from fire suppression to restoration, through increased use of prescribed fire, cultural burning, and managed wildfire, of a presuppression and precolonial fire regime in California.

Prevalence and Predictors of Malnutrition among Guatemalan Children at 2 Years of Age.

OBJECTIVE: To identify the prevalence and predictors of malnutrition among 2-year old children in the Western Highlands of Guatemala. METHODS: Prospective cohort of 852 Guatemalan children in San Lucas Toliman, Guatemala followed from birth to age 2 from May 2008 to December 2013. Socio-demographic, anthropometric, and health data of children was collected at 2 month intervals. RESULTS: Among the 402 males and 450 females in the cohort, mean weight-for-age Z-score (WAZ) declined from -0.67 ± 1.01 at 1 year to -1.07 ± 0.87 at 2 years, while mean height-for-age Z-score (HAZ) declined from -1.88 ± 1.19 at 1 year to -2.37 ± 0.99 at 2 years. Using multiple linear regression modeling, number of children <5 years old, vomiting in the past week, fever in the past week, and WAZ at 1 year were significant predictors of WAZ at 2 years. Significant predictors of HAZ at 2 years included household size, number of children <5 years old, diarrhea in the past week, WAZ at 1 year, and HAZ at 1 year. Vomiting in the past week and WAZ at 1 year were significant predictors of weight-for-height z-score (WHZ) at 2 years. CONCLUSIONS: Number of children <5 years old, symptoms such as vomiting or diarrhea in the previous week, and prior nutritional status were the most significant predictors of malnutrition in this cohort. Future research may focus on the application of models to develop predictive algorithms for mobile device technology, as well as the identification of other predictors of malnutrition that are not well characterized such as the interaction of environmental exposures with protein consumption and epigenetics.

Confidence Intervals for Random Forests: The Jackknife and the Infinitesimal Jackknife.

We study the variability of predictions made by bagged learners and random forests, and show how to estimate standard errors for these methods. Our work builds on variance estimates for bagging proposed by Efron (1992, 2013) that are based on the jackknife and the infinitesimal jackknife (IJ). In practice, bagged predictors are computed using a finite number B of bootstrap replicates, and working with a large B can be computationally expensive. Direct applications of jackknife and IJ estimators to bagging require B = Θ(n1.5) bootstrap replicates to converge, where n is the size of the training set. We propose improved versions that only require B = Θ(n) replicates. Moreover, we show that the IJ estimator requires 1.7 times less bootstrap replicates than the jackknife to achieve a given accuracy. Finally, we study the sampling distributions of the jackknife and IJ variance estimates themselves. We illustrate our findings with multiple experiments and simulation studies.

Quantifying and exploiting the age dependence in the effect of supplementary food for child undernutrition.

Motivated by the lack of randomized controlled trials with an intervention-free control arm in the area of child undernutrition, we fit a trivariate model of weight-for-age z score (WAZ), height-for-age z score (HAZ) and diarrhea status to data from an observational study of supplementary feeding (100 kCal/day for children with WAZ [Formula: see text]) in 17 Guatemalan communities. Incorporating time lags, intention to treat (i.e., to give supplementary food), seasonality and age interactions, we estimate how the effect of supplementary food on WAZ, HAZ and diarrhea status varies with a child's age. We find that the effect of supplementary food on all 3 metrics decreases linearly with age from 6 to 20 mo and has little effect after 20 mo. We derive 2 food allocation policies that myopically (i.e., looking ahead 2 mo) minimize either the underweight or stunting severity - i.e., the sum of squared WAZ or HAZ scores for all children with WAZ or HAZ [Formula: see text]. A simulation study based on the statistical model predicts that the 2 derived policies reduce the underweight severity (averaged over all ages) by 13.6-14.1% and reduce the stunting severity at age 60 mo by 7.1-8.0% relative to the policy currently in use, where all policies have a budget that feeds [Formula: see text]% of children. While these findings need to be confirmed on additional data sets, it appears that in a low-dose (100 kCal/day) supplementary feeding setting in Guatemala, allocating food primarily to 6-12 mo infants can reduce the severity of underweight and stunting.