Switching antiretroviral regimens when human immunodeficiency virus (HIV) viremia is controlled for a new regimen is challenging when there is the potential for prior nucleoside reverse-transcriptase inhibitor (NRTI) resistance. The objective was to study virologic outcomes after switching to dolutegravir compared with remaining on a boosted protease inhibitor (protease inhibitor/ritonavir [PI/r]) regimen in people with HIV (PWH) with prior documented virologic failure and/or exposure to mono/dual NRTIs.We used the Quebec HIV Cohort including 10 219 PWH whose data were collected at 4 sites in Montreal, Canada. We included all PWH with documented virologic failure or exposure to mono/dual NRTI therapy who were virologically suppressed on a PI/r-based regimen for at least 6 months on or after January 1, 2014 (n = 532). A marginal structural Cox model analysis was used to estimate the effect of the switch to dolutegravir on virologic outcome compared with remaining on PI/r. The outcome was defined as 2 consecutive viral loads (VLs) >50 copies/mL or 1 VL >50 copies/mL if it occurred at the last VL available.Among 532 eligible participants, 216 (40.6%) had their regimen switched to dolutegravir with 2 NRTIs, whereas 316 (59.4%) remained on the PI/r with 2 NRTIs. The weighted hazard ratio for the effect of dolutegravir switch on virologic failure compared with patients whose regimen remained on PI/r was 0.57 (95% confidence interval, 0.21-1.52).We did not find evidence of an increased risk for virologic failure after switching to dolutegravir from PI/r among patients with previous virologic failure or prior exposure to mono/dual NRTI.
Many studies seek to evaluate the effects of potentially harmful pregnancy exposures during specific gestational periods. We consider an observational pregnancy cohort where pregnant individuals can initiate medication usage or become exposed to a drug at various times during their pregnancy. An important statistical challenge involves how to define and estimate exposure effects when pregnancy loss or delivery can occur over time. Without proper consideration, the results of standard analysis may be vulnerable to selection bias, immortal time-bias, and time-dependent confounding. In this study, we apply the “target trials” framework of Hernán and Robins in order to define effects based on the counterfactual approach often used in causal inference. This effect is defined relative to a hypothetical randomized trial of timed pregnancy exposures where delivery may precede and thus potentially interrupt exposure initiation. We describe specific implementations of inverse probability weighting, G-computation, and Targeted Maximum Likelihood Estimation to estimate the effects of interest. We demonstrate the performance of all estimators using simulated data and show that a standard implementation of inverse probability weighting is biased. We then apply our proposed methods to a pharmacoepidemiology study to evaluate the potentially time-dependent effect of exposure to inhaled corticosteroids on birthweight in pregnant people with mild asthma.
Liver transplantation is a high-risk surgery associated with important perioperative bleeding and transfusion needs. Uncertainties remain on the association between preoperative fibrinogen level and bleeding in this population.We conducted a cohort study that included all consecutive adult patients undergoing a liver transplantation for end-stage liver disease in 1 center. We analyzed the association between the preoperative fibrinogen level and bleeding-related outcomes. Our primary outcome was intraoperative blood loss, and our secondary outcomes were estimated perioperative blood loss, intraoperative and perioperative red blood cell transfusions, reinterventions for bleeding and 1-y graft and patient survival. We estimated linear regression models and marginal risk models adjusted for all important potential confounders. We used restricted cubic splines to explore potential nonlinear associations and reported dose-response curves.We included 613 patients. We observed that a lower fibrinogen level was associated with a higher intraoperative blood loss, a higher estimated perioperative blood loss and a higher risk of intraoperative and perioperative red blood cell transfusions (nonlinear effects). Based on an exploratory analysis of the dose-response curves, these effects were observed below a threshold value of 3 g/L for these outcomes. We did not observe any association between preoperative fibrinogen level and reinterventions, 1-y graft survival or 1-y patient survival.This study suggests that a lower fibrinogen level is associated with bleeding in liver transplantation. The present results may help improving the selection of patients for further studies on preoperative fibrinogen administration in liver transplant recipients with end-stage liver disease.
Abstract Many clinical and epidemiological applications of survival analysis focus on interval‐censored events that can be ascertained only at discrete times of clinic visits. This implies that the values of time‐varying covariates are not correctly aligned with the true, unknown event times, inducing a bias in the estimated associations. To address this issue, we adapted the simulation‐extrapolation (SIMEX) methodology, based on assessing how the estimates change with the artificially increased time between clinic visits. We propose diagnostics to choose the extrapolating function. In simulations, the SIMEX‐corrected estimates reduced considerably the bias to the null and generally yielded a better bias/variance trade‐off than conventional estimates. In a real‐life pharmacoepidemiological application, the proposed method increased by 27% the excess hazard of the estimated association between a time‐varying exposure, representing the 2‐year cumulative duration of past use of a hypertensive medication, and the hazard of nonmelanoma skin cancer (interval‐censored events). These simulation‐based and real‐life results suggest that the proposed SIMEX‐based correction may help improve the accuracy of estimated associations between time‐varying exposures and the hazard of interval‐censored events in large cohort studies where the events are recorded only at relatively sparse times of clinic visits/assessments. However, these advantages may be less certain for smaller studies and/or weak associations.
Abstract Causal inference methods have been developed for longitudinal observational study designs where confounding is thought to occur over time. In particular, one may estimate and contrast the population mean counterfactual outcome under specific exposure patterns. In such contexts, confounders of the longitudinal treatment‐outcome association are generally identified using domain‐specific knowledge. However, this may leave an analyst with a large set of potential confounders that may hinder estimation. Previous approaches to data‐adaptive model selection for this type of causal parameter were limited to the single time‐point setting. We develop a longitudinal extension of a collaborative targeted minimum loss‐based estimation (C‐TMLE) algorithm that can be applied to perform variable selection in the models for the probability of treatment with the goal of improving the estimation of the population mean counterfactual outcome under a fixed exposure pattern. We investigate the properties of this method through a simulation study, comparing it to G‐Computation and inverse probability of treatment weighting. We then apply the method in a real‐data example to evaluate the safety of trimester‐specific exposure to inhaled corticosteroids during pregnancy in women with mild asthma. The data for this study were obtained from the linkage of electronic health databases in the province of Quebec, Canada. The C‐TMLE covariate selection approach allowed for a reduction of the set of potential confounders, which included baseline and longitudinal variables.
In longitudinal settings, causal inference methods usually rely on a discretization of the patient timeline that may not reflect the underlying data generation process. This article investigates the estimation of causal parameters under discretized data. It presents the implicit assumptions practitioners make but do not acknowledge when discretizing data to assess longitudinal causal parameters. We illustrate that differences in point estimates under different discretizations are due to the data coarsening resulting in both a modified definition of the parameter of interest and loss of information about time-dependent confounders. We further investigate several tools to advise analysts in selecting a timeline discretization for use with pooled longitudinal targeted maximum likelihood estimation for the estimation of the parameters of a marginal structural model. We use a simulation study to empirically evaluate bias at different discretizations and assess the use of the cross-validated variance as a measure of data support to select a discretization under a chosen data coarsening mechanism. We then apply our approach to a study on the relative effect of alternative asthma treatments during pregnancy on pregnancy duration. The results of the simulation study illustrate how coarsening changes the target parameter of interest as well as how it may create bias due to a lack of appropriate control for time-dependent confounders. We also observe evidence that the cross-validated variance acts well as a measure of support in the data, by being minimized at finer discretizations as the sample size increases.
Tofacitinib is the first oral Janus kinase inhibitor approved for the treatment of rheumatoid arthritis (RA). We compared the effectiveness and safety of tofacitinib, disease-modifying antirheumatic drugs (DMARDs), tumor necrosis factor inhibitors (TNFi), and non-TNF biologics in patients with RA previously treated with methotrexate. We used MarketScan® databases (2011–2014) to study methotrexate-exposed patients with RA who were newly prescribed tofacitinib, DMARDs other than methotrexate, and biologics. The date of first prescription was defined as the cohort entry. The therapy was considered effective if all of the following criteria from a claims-based algorithm were achieved at the first year of follow-up: high adherence, no biologic or tofacitinib switch or addition, no DMARD switch or addition, no increase in dose or frequency of index drug, no more than one glucocorticoid joint injection, and no new/increased oral glucocorticoid dose. The safety outcome was serious infections requiring hospitalization. Non-TNF biologics comprised the reference group. We included 21,832 patients with RA, including 0.8% treated with tofacitinib, 24.7% treated with other DMARDs, 61.2% who had started therapy with TNFi, and 13.3% treated with non-TNF biologics. The rates of therapy effectiveness were 15.4% for tofacitinib, 11.1% for DMARDs, 18.6% for TNFi, and 19.8% for non-TNF biologics. In adjusted analyses, tofacitinib and non-TNF biologics appeared to have similar effectiveness rates, whereas DMARD initiators were less effective than non-TNF biologics. We could not clearly establish if tofacitinib was associated with a higher rate of serious infections. In patients with RA previously treated with methotrexate, our comparisons of tofacitinib with non-TNF biologics, though not definitive, did not demonstrate differences with respect to hospitalized infections or effectiveness.
