Association of statin use with outcomes of patients admitted with COVID-19: an analysis of electronic health records using superlearner

IMPORTANCE

Statin use prior to hospitalization for Coronavirus Disease 2019 (COVID-19) is hypothesized to improve inpatient outcomes including mortality, but prior findings from large observational studies have been inconsistent, due in part to confounding. Recent advances in statistics, including incorporation of machine learning techniques into augmented inverse probability weighting with targeted maximum likelihood estimation, address baseline covariate imbalance while maximizing statistical efficiency.

OBJECTIVE

To estimate the association of antecedent statin use with progression to severe inpatient outcomes among patients admitted for COVD-19.

DESIGN, SETTING AND PARTICIPANTS

We retrospectively analyzed electronic health records (EHR) from individuals ≥ 40-years-old who were admitted between March 2020 and September 2022 for ≥ 24 h and tested positive for SARS-CoV-2 infection in the 30 days before to 7 days after admission.

EXPOSURE

Antecedent statin use-statin prescription ≥ 30 days prior to COVID-19 admission.

MAIN OUTCOME

Composite end point of in-hospital death, intubation, and intensive care unit (ICU) admission.

RESULTS

Of 15,524 eligible COVID-19 patients, 4412 (20%) were antecedent statin users. Compared with non-users, statin users were older (72.9 (SD: 12.6) versus 65.6 (SD: 14.5) years) and more likely to be male (54% vs. 51%), White (76% vs. 71%), and have ≥ 1 medical comorbidity (99% vs. 86%). Unadjusted analysis demonstrated that a lower proportion of antecedent users experienced the composite outcome (14.8% vs 19.3%), ICU admission (13.9% vs 18.3%), intubation (5.1% vs 8.3%) and inpatient deaths (4.4% vs 5.2%) compared with non-users. Risk differences adjusted for labs and demographics were estimated using augmented inverse probability weighting with targeted maximum likelihood estimation using Super Learner. Statin users still had lower rates of the composite outcome (adjusted risk difference: - 3.4%; 95% CI: - 4.6% to - 2.1%), ICU admissions (- 3.3%; - 4.5% to - 2.1%), and intubation (- 1.9%; - 2.8% to - 1.0%) but comparable inpatient deaths (0.6%; - 1.3% to 0.1%).

CONCLUSIONS AND RELEVANCE

After controlling for confounding using doubly robust methods, antecedent statin use was associated with minimally lower risk of severe COVID-19-related outcomes, ICU admission and intubation, however, we were not able to corroborate a statin-associated mortality benefit.

Comparing g-computation, propensity score-based weighting, and targeted maximum likelihood estimation for analyzing externally controlled trials with both measured and unmeasured confounders: a simulation study

OBJECTIVES

To have confidence in one's interpretation of treatment effects assessed by comparing trial results to external controls, minimizing bias is a critical step. We sought to investigate different methods for causal inference in simulated data sets with measured and unmeasured confounders.

METHODS

The simulated data included three types of outcomes (continuous, binary, and time-to-event), treatment assignment, two measured baseline confounders, and one unmeasured confounding factor. Three scenarios were set to create different intensities of confounding effect (e.g., small and blocked confounding paths, medium and blocked confounding paths, and one large unblocked confounding path for scenario 1 to 3, respectively) caused by the unmeasured confounder. The methods of g-computation (GC), inverse probability of treatment weighting (IPTW), overlap weighting (OW), standardized mortality/morbidity ratio (SMR), and targeted maximum likelihood estimation (TMLE) were used to estimate average treatment effects and reduce potential biases.

RESULTS

The results with the greatest extent of biases were from the raw model that ignored all the potential confounders. In scenario 2, the unmeasured factor indirectly influenced the treatment assignment through a measured controlling factor and led to medium confounding. The methods of GC, IPTW, OW, SMR, and TMLE removed most of bias observed in average treatment effects for all three types of outcomes from the raw model. Similar results were found in scenario 1, but the results tended to be biased in scenario 3. GC had the best performance followed by OW.

CONCLUSIONS

The aforesaid methods can be used for causal inference in externally controlled studies when there is no large, unblockable confounding path for an unmeasured confounder. GC and OW are the preferable approaches.

Estimating the effect of donor sex on red blood cell transfused patient mortality: A retrospective cohort study using a targeted learning and emulated trials-based approach

BACKGROUND

Observational studies determining the effect of red blood cell (RBC) donor sex on recipient mortality have been inconsistent. Emulating hypothetical randomized target trials using large real-world data and targeted learning may clarify potential adverse effects.

METHODS

In this retrospective cohort study, a RBC transfusion database from the Capital Region of Denmark comprising more than 900,000 transfusion events defined the observational data. Eligible patients were minimum 18 years, had received a leukocyte-reduced RBC transfusion, and had no history of RBC transfusions within the past year at baseline. The doubly robust targeted maximum likelihood estimation method coupled with ensembled machine learning was used to emulate sex-stratified target trials determining the comparative effectiveness of exclusively transfusing RBC units from either male or female donors. The outcome was all-cause mortality within 28 days of the baseline-transfusion. Estimates were adjusted for the total number of transfusions received on each day k, hospital of transfusion, calendar period, patient age and sex, ABO/RhD blood group of the patient, Charlson comorbidity score, the total number of transfusions received prior to day k, and the number of RBC units received on each day k from donors younger than 40 years of age.

FINDINGS

Among 98,167 adult patients who were transfused between Jan. 1, 2008, and Apr. 10, 2018, a total of 90,917 patients (54.6% female) were eligible. For male patients, the 28-day survival was 2.06 percentage points (pp) (95 % confidence interval [CI]: 1.81-2.32, P<0.0001) higher under treatment with RBC units exclusively from male donors compared with exclusively from female donors. In female patients, exclusively transfusing RBC units from either male or female donors increased the 28-day survival with 0.64pp (0.52-0.76, P<0.0001), and 0.62pp (0.49-0.75, P<0.0001) compared with the current practice, respectively. No evidence of a sex-specific donor effect was found for female patients (0.02pp [-0.18-0.22]). The sensitivity analyses showed that a large unknown causal bias would have to be present to affect the conclusions.

INTERPRETATION

The results suggest that a sex-matched transfusion policy may benefit patients. However, a causal interpretation of the findings relies on the assumption of no unmeasured confounding, treatment consistency, positivity, and minimal model misspecifications.

FUNDING

Novo Nordisk Foundation and the Innovation Fund Denmark.

Effects of changes in depressive symptoms on handgrip strength in later life: A four-year longitudinal study in England

BACKGROUND

Few studies have examined the associations between changes in depressive symptoms and handgrip strength in older people. This study aimed to examine the magnitude of the association between depressive symptoms over 2 years and weak handgrip strength on the 4 years of follow-up.

METHODS

We conducted a longitudinal study using data from the English Longitudinal Study of Aging, a nationally representative panel survey of older adults in England. Data were derived from waves 4 (2008-2009), 5 (2010-2011), and 6 (2012-2013). A total of 5,080 participants were included in the analysis. Depressive symptoms were dichotomized using the eight-item Center for Epidemiological Studies Depression Scale in waves 4 and 5 and were regarded as the exposure. Handgrip strength measurements objectively measured in wave 6 were dichotomized according to the 25th percentile of the British norm and used as the outcome. The targeted maximum likelihood estimation model was utilized to assess time-variant depressive symptoms on handgrip strength, adjusted for time-variant and time-invariant covariates.

RESULTS

The maintenance of non-depressive symptoms (relative risk [RR], 0.72; 95% confidence interval [CI], 0.59-0.87) or improvement of depressive symptoms (RR, 0.71; 95% CI, 0.55-0.91) had a protective effect on weak handgrip strength compared with the persistence of depressive symptoms.

LIMITATION

This study is not a randomized control trial but a longitudinal observational study, indicating that our study finding may still have been affected by unknown confounders.

CONCLUSION

Our findings suggest the importance of managing depressive symptoms to prevent poor physical function.

Recombinant hepatitis B vaccine uptake and multiple sclerosis risk: A marginal structural modeling approach

BACKGROUND

There has been an ongoing debate regarding the association between uptake of recombinant vaccine against hepatitis B virus (HBV) and multiple sclerosis (MS) risk. This case-control study tested the hypothesis whether recombinant HBV vaccination status is causally associated with MS risk using targeted maximum likelihood estimation (TMLE) technique.

METHODS

Confirmed 110 MS cases and age (± 5 years), sex and nativity matched (1:1) 110 controls were enrolled. Data were collected on sociodemographics, environmental factors, history of vaccinations and past morbidities through face-to-face interview both from cases and controls. To estimate the causal parameters including marginal odds ratio (OR), causal relative risk (RR), causal risk difference (RD) and their 95% confidence intervals (CIs), we implemented case-control-weighted TMLE for a matched design that uses data-adaptive flexible stacked ensemble-based machine learning system namely Super Learner. Additionally, population preventable fraction (PPF) of MS risk was computed.

RESULTS

This study demonstrated a significant nonspecific protective effect of HBV vaccination against MS risk (marginal OR 0.44; 95% CI: 0.19-0.68; p = 0.006; causal RR 0.64, 95% CI: 0.46-0.89; p = 0.004). The significant causal RD showed that among the vaccinated 19% fewer MS cases occurred owing to their HBV vaccination (causal RD -0.19; 95% CI: -0.32 - -0.06; p = 0.014). In the source population, vaccination against HBV led to 17.4% reduced MS risk (PPF = 17.4%; 95% CI: 3.8%, 36.3%).

CONCLUSION

The results suggest a significant nonspecific protective effect of recombinant HBV vaccine against MS risk. Future studies may contemplate to confirm these results.

The unintended impact of helmet use on bicyclists' risk-taking behaviors

INTRODUCTION

Safety is a critical factor in promoting sustainable urban non-motorized travel modes like bicycles. Helmets have shown to be effective in reducing injury severity in bicycle crashes, however, their effects on bicyclists' behaviors still requires deeper understanding, especially amid the emerging trend of using shared bicycles. Risk compensation effects suggest that bicyclists may offset perceived gains in safety from wearing a helmet by increasing risk-taking behaviors. A better understanding of these compensation effects can be useful in assessing various bicycle safety related programs.

METHOD

Using a sample of 131 bicyclists from the San Francisco Bay area, this research studies how bicyclists respond with respect to risk-taking behaviors under various urban-street conditions, as a function of helmet use. Study participants were each shown 12 videos, shot in Berkeley, California, from the perspective of a bicyclist riding behind another bicyclist. A fractional factorial experiment design was used to systematically vary contextual attributes (e.g., speed, bike lane facilities, on-street parking, passing vehicles) across the videos. After each video, participants were asked to indicate if they would overtake the bicyclist in the video. With the help of data adaptive estimation techniques, targeted maximum likelihood estimation (TMLE) was applied to estimate the average risk difference between helmeted users and non-users, controlling for self-selection effects. Individual-based nonparametric bootstrap was performed to assess the uncertainty associated with the estimator.

RESULTS

Our findings suggest, on average, individuals more likely to wear a helmet are 15.6% more likely to undertake a risky overtaking maneuver. Practical Applications: This study doesn't try to oppose mandatory helmet laws, but rather serves as a cautionary warning that road safety programs may need to consider strategies in which unintended impact of bicycle helmet use can be mitigated. Moreover, our findings also provide additional evaluation component when it comes to the cost-benefit assessment of helmet-related laws.

Effect of perinatal depression on risk of adverse infant health outcomes in mother-infant dyads in Gondar town: a causal analysis

BACKGROUND

Approximately one-third of pregnant and postnatal women in Ethiopia experience depression posing a substantial health burden for these women and their families. Although associations between postnatal depression and worse infant health have been observed, there have been no studies to date assessing the causal effects of perinatal depression on infant health in Ethiopia. We applied longitudinal data and recently developed causal inference methods that reduce the risk of bias to estimate associations between perinatal depression and infant diarrhea, Acute Respiratory Infection (ARI), and malnutrition in Gondar Town, Ethiopia.

METHODS

A cohort of 866 mother-infant dyads were followed from infant birth for 6 months and the cumulative incidence of ARI, diarrhea, and malnutrition were assessed. The Edinburgh Postnatal Depression Scale (EPDS) was used to assess the presence of maternal depression, the Integrated Management of Newborn and Childhood Illnesses (IMNCI) guidelines were used to identify infant ARI and diarrhea, and the mid upper arm circumference (MUAC) was used to identify infant malnutrition. The risk difference (RD) due to maternal depression for each outcome was estimated using targeted maximum likelihood estimation (TMLE), a doubly robust causal inference method used to reduce bias in observational studies.

RESULTS

The cumulative incidence of diarrhea, ARI and malnutrition during 6-month follow-up was 17.0% (95%CI: 14.5, 19.6), 21.6% (95%CI: 18.89, 24.49), and 14.4% (95%CI: 12.2, 16.9), respectively. There was no association between antenatal depression and ARI (RD = - 1.3%; 95%CI: - 21.0, 18.5), diarrhea (RD = 0.8%; 95%CI: - 9.2, 10.9), or malnutrition (RD = -7.3%; 95%CI: - 22.0, 21.8). Similarly, postnatal depression was not associated with diarrhea (RD = -2.4%; 95%CI: - 9.6, 4.9), ARI (RD = - 3.2%; 95%CI: - 12.4, 5.9), or malnutrition (RD = 0.9%; 95%CI: - 7.6, 9.5).

CONCLUSION

There was no evidence for an association between perinatal depression and the risk of infant diarrhea, ARI, and malnutrition amongst women in Gondar Town. Previous reports suggesting increased risks resulting from maternal depression may be due to unobserved confounding.

Effect modification of general and central obesity by sex and age on cardiovascular outcomes: Targeted maximum likelihood estimation in the atherosclerosis risk in communities study

BACKGROUND AND AIMS

To elucidate the effect modification of general and central obesity by sex and age on the risk of cardiovascular events.

METHODS

The analysis included 14,983 males and females aged 45-75 years from the Atherosclerosis Risk in Communities study. Obesity was defined with body mass index (BMI), waist-to-hip ratio (WHR) and body shape index (BSI) which categorized the participants as obese and non-obese. Targeted maximum likelihood estimation was used to estimate the risk ratio (RR) with the tmle package in R software.

RESULTS

After adjustment, the strongest effect of BMI on CHD was in females (RR (95%CI): 1.26 (1.11, 1.42)) and in age>54 (RR (95%CI): 1.16 (1.06, 1.27)) and for HF it was in age>54 (RR (95%CI): 1.18 (1.10, 1.26)) and in females (RR (95%CI): 1.17 (1.08, 1.28)). Regarding central obesity, WHR (RR (95%CI): 1.19 (1.05, 1.34)) had the strongest effects on CHD for males and BSI (RR (95%CI): 1.140 (1.02, 1.26)) for age ≤ 54, and for HF the WHR (RR (95%CI): 1.22 (1.10, 1.36)) and BSI (RR (95%CI): 1.18 (1.07, 1.30)) had the strongest effects for age≤54, respectively.

CONCLUSION

Among males and age≤54, WHR index was associated with a higher risk of CHD and HF while BMI was so for females and age>54.

Maternal folate levels during pregnancy and offspring brain development in late childhood

BACKGROUND

Cumulative evidence shows that low maternal folate levels during pregnancy are associated with offspring neuropsychiatric disorders even in the absence of neural tube defects. However, the relationship between prenatal exposure to folate and brain development in late childhood has been rarely investigated.

METHODS

In 2095 children from a prospective population-based cohort in Rotterdam, the Netherlands, we examined the association of maternal folate levels during pregnancy with downstream brain development in offspring. Maternal folate concentrations were measured from venous blood in early gestation. Child structural neuroimaging data were measured at age 9-11 years. In addition, measures of child head circumference using fetal ultrasound in the third trimester and total brain volume using magnetic resonance imaging at age 6-8 years were used for analyses with repeated assessments of brain development.

RESULTS

Maternal folate deficiency (i.e., <7 nmol/L) during pregnancy was associated with smaller total brain volume (B = -18.7 cm³, 95% CI -37.2 to -0.2) and smaller cerebral white matter (B = -7.2 cm³, 95% CI -11.8 to -2.6) in children aged 9-11 years. No differences in cortical thickness or surface area were observed. Analysis of the repeated brain assessments showed that children exposed to deficient folate concentrations in utero had persistently smaller brains compared to controls from the third trimester to childhood (β = -0.4, 95% CI -0.6 to -0.1).

CONCLUSIONS

Low maternal folate levels during pregnancy are associated with altered offspring brain development in childhood, suggesting the importance of essential folate concentrations in early pregnancy.

Metalworking Fluids and Colon Cancer Risk: Longitudinal Targeted Minimum Loss-based Estimation

Metalworking fluids (MWFs) are a class of complex mixtures of chemicals and oils, including several known carcinogens that may pose a cancer hazard to millions of workers. Reports on the relation between MWFs and incident colon cancer have been mixed.

METHODS

We investigated the relation between exposure to straight, soluble, and synthetic MWFs and the incidence of colon cancer in a cohort of automobile manufacturing industry workers, adjusting for time-varying confounding affected by prior exposure to reduce healthy worker survivor bias. We used longitudinal targeted minimum loss-based estimation (TMLE) to estimate the difference in the cumulative incidence of colon cancer comparing counterfactual outcomes if always exposed above to always exposed below an exposure cutoff while at work. Exposure concentration cutoffs were selected a priori at the 90th percentile of total particulate matter for each fluid type: 0.034, 0.400, and 0.003 for straight, soluble, and synthetic MWFs, respectively.

RESULTS

The estimated 25-year risk differences were 3.8% (95% confidence interval [CI] = 0.7, 7.0) for straight, 1.3% (95% CI = -2.3, 4.8) for soluble, and 0.2% (95% CI = -3.3, 3.7) for synthetic MWFs, respectively. The corresponding risk ratios were 2.39 (1.12, 5.08), 1.43 (0.67, 3.04), and 1.08 (0.51, 2.30) for straight, soluble, and synthetic MWFs, respectively.

CONCLUSIONS

By controlling for time-varying confounding affected by prior exposure, a key feature of occupational cohorts, we were able to provide evidence for a causal effect of straight MWF exposure on colon cancer risk that was not found using standard analytical techniques in previous reports.

An educational intervention to improve knowledge about prevention against occupational asthma and allergies using targeted maximum likelihood estimation

PURPOSE

Occupational asthma and allergies are potentially preventable diseases affecting 5-15% of the working population. However, the use of preventive measures is often insufficient. The aim of this study was to estimate the average treatment effect of an educational intervention designed to improve the knowledge of preventive measures against asthma and allergies in farm apprentices from Bavaria (Southern Germany).

METHODS

Farm apprentices at Bavarian farm schools were asked to complete a questionnaire evaluating their knowledge about preventive measures against occupational asthma and allergies (use of personal protective equipment, personal and workplace hygiene measures). Eligible apprentices were randomized by school site to either a control or an intervention group. The intervention consisted of a short educational video about use of preventive measures. Six months after the intervention, subjects were asked to complete a post-intervention questionnaire. Of the 116 apprentices (70 intervention group, 46 control group) who answered the baseline questionnaire, only 47 subjects (41%; 17 intervention group, 30 control group) also completed the follow-up questionnaire. We, therefore, estimated the causal effect of the intervention using targeted maximum likelihood estimation. Models were controlled for potential confounders.

RESULTS

Based on the targeted maximum likelihood estimation, the intervention would have increased the proportion of correct answers on all six preventive measures by 18.4% (95% confidence interval 7.3-29.6%) had all participants received the intervention vs. had they all been in the control group.

CONCLUSIONS

These findings indicate the improvement of knowledge by the educational intervention.

Causal Methods for Observational Research: A Primer

The goal of many observational studies is to estimate the causal effect of an exposure on an outcome after adjustment for confounders, but there are still some serious errors in adjusting confounders in clinical journals. Standard regression modeling (e.g., ordinary logistic regression) fails to estimate the average effect of exposure in total population in the presence of interaction between exposure and covariates, and also cannot adjust for time-varying confounding appropriately. Moreover, stepwise algorithms of the selection of confounders based on P values may miss important confounders and lead to bias in effect estimates. Causal methods overcome these limitations. We illustrate three causal methods including inverse-probability-of-treatment-weighting (IPTW) and parametric g-formula, with an emphasis on a clever combination of these 2 methods: targeted maximum likelihood estimation (TMLE) which enjoys a double-robust property against bias.

Double robust and efficient estimation of a prognostic model for events in the presence of dependent censoring

In longitudinal data arising from observational or experimental studies, dependent subject drop-out is a common occurrence. If the goal is estimation of the parameters of a marginal complete-data model for the outcome, biased inference will result from fitting the model of interest with only uncensored subjects. For example, investigators are interested in estimating a prognostic model for clinical events in HIV-positive patients, under the counterfactual scenario in which everyone remained on ART (when in reality, only a subset had). Inverse probability of censoring weighting (IPCW) is a popular method that relies on correct estimation of the probability of censoring to produce consistent estimation, but is an inefficient estimator in its standard form. We introduce sequentially augmented regression (SAR), an adaptation of the Bang and Robins (2005. Doubly robust estimation in missing data and causal inference models. Biometrics 61, 962-972.) method to estimate a complete-data prediction model, adjusting for longitudinal missing at random censoring. In addition, we propose a closely related non-parametric approach using targeted maximum likelihood estimation (TMLE; van der Laan and Rubin, 2006. Targeted maximum likelihood learning. The International Journal of Biostatistics 2 (1), Article 11). We compare IPCW, SAR, and TMLE (implemented parametrically and with Super Learner) through simulation and the above-mentioned case study.

Modeling the impact of hepatitis C viral clearance on end-stage liver disease in an HIV co-infected cohort with targeted maximum likelihood estimation

Despite modern effective HIV treatment, hepatitis C virus (HCV) co-infection is associated with a high risk of progression to end-stage liver disease (ESLD) which has emerged as the primary cause of death in this population. Clinical interest lies in determining the impact of clearance of HCV on risk for ESLD. In this case study, we examine whether HCV clearance affects risk of ESLD using data from the multicenter Canadian Co-infection Cohort Study. Complications in this survival analysis arise from the time-dependent nature of the data, the presence of baseline confounders, loss to follow-up, and confounders that change over time, all of which can obscure the causal effect of interest. Additional challenges included non-censoring variable missingness and event sparsity. In order to efficiently estimate the ESLD-free survival probabilities under a specific history of HCV clearance, we demonstrate the double-robust and semiparametric efficient method of Targeted Maximum Likelihood Estimation (TMLE). Marginal structural models (MSM) can be used to model the effect of viral clearance (expressed as a hazard ratio) on ESLD-free survival and we demonstrate a way to estimate the parameters of a logistic model for the hazard function with TMLE. We show the theoretical derivation of the efficient influence curves for the parameters of two different MSMs and how they can be used to produce variance approximations for parameter estimates. Finally, the data analysis evaluating the impact of HCV on ESLD was undertaken using multiple imputations to account for the non-monotone missing data.

Targeted maximum likelihood estimation in safety analysis

OBJECTIVES

To compare the performance of a targeted maximum likelihood estimator (TMLE) and a collaborative TMLE (CTMLE) to other estimators in a drug safety analysis, including a regression-based estimator, propensity score (PS)-based estimators, and an alternate doubly robust (DR) estimator in a real example and simulations.

STUDY DESIGN AND SETTING

The real data set is a subset of observational data from Kaiser Permanente Northern California formatted for use in active drug safety surveillance. Both the real and simulated data sets include potential confounders, a treatment variable indicating use of one of two antidiabetic treatments and an outcome variable indicating occurrence of an acute myocardial infarction (AMI).

RESULTS

In the real data example, there is no difference in AMI rates between treatments. In simulations, the double robustness property is demonstrated: DR estimators are consistent if either the initial outcome regression or PS estimator is consistent, whereas other estimators are inconsistent if the initial estimator is not consistent. In simulations with near-positivity violations, CTMLE performs well relative to other estimators by adaptively estimating the PS.

CONCLUSION

Each of the DR estimators was consistent, and TMLE and CTMLE had the smallest mean squared error in simulations.