Time-dependent Treatment Effects Research Articles

Management and Risk Reduction of Rheumatoid Arthritis in Individuals with Obstructive Sleep Apnea: A Nationwide Population-Based Study in Taiwan.

To explore associations between obstructive sleep apnea (OSA) and autoimmune diseases and evaluate whether OSA management reduces the incidence of autoimmune diseases. This was a retrospective cohort study using nationwide database research. The data was from 105,846 adult patients in whom OSA was diagnosed and recorded in the Taiwan National Health Insurance Research Database between 2002 and 2011 were the patients were analyzed retrospectively. Patients with antecedent autoimmune diseases were excluded. A comparison cohort of 423,384 participants without OSA served as age- and sex-matched controls. Multivariable Cox regression analysis was performed on both cohorts to compute risk of autoimmune diseases during follow-up. Time-dependent OSA treatment effect was analyzed among patients with OSA. There were no interventions. Among patients with OSA, overall risk for incident autoimmune diseases was significantly higher than that in controls (adjusted hazard ratio [HR] = 1.95, 95% confidence interval [CI] = 1.66-2.27). Risk for individual autoimmune diseases, including rheumatoid arthritis (RA), Sjögren syndrome (SS), and Behçet disease, was significantly higher in patients with OSA than in controls (HRs [95% CI]: RA 1.33 [1.03-1.72, SS 3.45 [2.67-4.45] and Behçet disease 5.33 [2.45-12.66]). Increased risk for systemic lupus erythematosus (HR 1.00 [0.54-1.84]) and systemic sclerosis (HR 1.43 [0.51-3.96]) did not reach statistical significance. Patients with OSA receiving treatment had an overall reduced risk of RA and other autoimmune diseases (time-dependent HRs [95% CI]: 0.22 [0.05-0.94] and 0.51 [0.28-0.92], respectively). Patients with OSA are associated with higher risk for developing RA, SS, and Behçet disease. OSA management is associated with reduced risk of RA.

Knee osteoarthritis functional classification scheme – Validation of time dependent treatment effect. One year follow-up of 518 patients

Objective: The purpose of the current study was to validate time dependent changes of a novel functional classification for patients with knee osteoarthritis (KOA), following a home-based biomechanical treatment (HBBT).Methods: A retrospective analysis of 518 patients with KOA was conducted. All patients were classified using a novel knee osteoarthritis functional grade (KOFG) classification for KOA, based on spatio-temporal gait analysis. Patients were re-classified after 3 months and 1 year of HBBT to examine and validate this classification using timedependant changes. The time dependent changes in the classification were compared to gold-standard selfassessment questionnaires, WOMAC and short form 36 (SF-36).Results: The changes in KOFG were demonstrated over time, with most changes occurring after 3 months of treatment with consolidation of the effect at 12 months. For example, of 427 patients that were classified in KOFG 2-4 grade at baseline, 44.9% and 51.5% had lower (better) KOFG grades at 3 and 12 months of treatment, respectively. The changes in KOFG were validated with WOMAC and SF-36 questionnaires showing a significant correlation between KOFG changes and changes in WOMAC and SF-36. SF-36 pain sub-scale showed an improvement of 33.0% and 38.0% following 3 months and 12 months of treatment, respectively (p values <0.0001).Conclusions: The results of the current study validate the knee osteoarthritis functional grade classification scheme as a tool to assess time dependant changes in KOA as well as its sensitivity to assess treatment effect. The KOFG can offer a more robust mode of reporting clinical results in describing the natural history and time-dependent treatment results of patients suffering from knee OA and should be considered as an additional outcome measure in future studies.

The longitudinal relationship between quality of life and survival in advanced stage cancer

Quality of Life (QoL) at baseline is frequently found to be a prognostic factor in cancer studies. However, little is known about the relationship of the trajectory of QoL and survival in patients with advanced cancer. This study evaluates the effects of both level and change of QoL on survival to explore the potential of utilizing longitudinal information of QoL for prognosis. A series of joint models were used in a sample (N = 512) of patients diagnosed with advanced cancer (sample consisted of nine different cancer sites) with assessments of QoL across six time points and with survival information recorded up to 28 months after diagnosis. We used FACT-G as the QoL measure, and we evaluated the effects of change in QoL controlling for the time-dependent effects of chemotherapy and radiation. The median survival for patients was 14.2 months, and 10% of the sample had survived beyond 28 months after the diagnosis of advanced cancer. The effect of change of QoL on survival was significant (hazard ratio = 0.98; p < 0.001) controlling for time-dependent treatment effects. Also, the slope of the trajectory in QoL was found to be a significant predictor of survival (hazard ratio = 0.18; p < 0.001). These preliminary findings suggest that the patient's longitudinal experience in QoL may be a significant prognostic factor of survival, a novel finding with potentially important implications in medical decision making. Longitudinal information on QoL can be used for updating the patient's prognosis of survival.

Assessing potentially time-dependent treatment effect from clinical trials and observational studies for survival data, with applications to the Women's Health Initiative combined hormone therapy trial.

For risk and benefit assessment in clinical trials and observational studies with time-to-event data, the Cox model has usually been the model of choice. When the hazards are possibly non-proportional, a piece-wise Cox model over a partition of the time axis may be considered. Here, we propose to analyze clinical trials or observational studies with time-to-event data using a certain semiparametric model. The model allows for a time-dependent treatment effect. It includes the important proportional hazards model as a sub-model and can accommodate various patterns of time-dependence of the hazard ratio. After estimation of the model parameters using a pseudo-likelihood approach, simultaneous confidence intervals for the hazard ratio function are established using a Monte Carlo method to assess the time-varying pattern of the treatment effect. To assess the overall treatment effect, estimated average hazard ratio and its confidence intervals are also obtained. The proposed methods are applied to data from the Women's Health Initiative. To compare the Women's Health Initiative clinical trial and observational study, we use the propensity score in building the regression model. Compared with the piece-wise Cox model, the proposed model yields a better model fit and does not require partitioning of the time axis.

Adjusting for treatment switching in randomised controlled trials - A simulation study and a simplified two-stage method.

Estimates of the overall survival benefit of new cancer treatments are often confounded by treatment switching in randomised controlled trials (RCTs) - whereby patients randomised to the control group are permitted to switch onto the experimental treatment upon disease progression. In health technology assessment, estimates of the unconfounded overall survival benefit associated with the new treatment are needed. Several switching adjustment methods have been advocated in the literature, some of which have been used in health technology assessment. However, it is unclear which methods are likely to produce least bias in realistic RCT-based scenarios. We simulated RCTs in which switching, associated with patient prognosis, was permitted. Treatment effect size and time dependency, switching proportions and disease severity were varied across scenarios. We assessed the performance of alternative adjustment methods based upon bias, coverage and mean squared error, related to the estimation of true restricted mean survival in the absence of switching in the control group. We found that when the treatment effect was not time-dependent, rank preserving structural failure time models (RPSFTM) and iterative parameter estimation methods produced low levels of bias. However, in the presence of a time-dependent treatment effect, these methods produced higher levels of bias, similar to those produced by an inverse probability of censoring weights method. The inverse probability of censoring weights and structural nested models produced high levels of bias when switching proportions exceeded 85%. A simplified two-stage Weibull method produced low bias across all scenarios and provided the treatment switching mechanism is suitable, represents an appropriate adjustment method.

Modeling nonlinear time-dependent treatment effects: an application of the generalized time-varying effect model (TVEM).

The goal of this article is to introduce to social and behavioral scientists the generalized time-varying effect model (TVEM), a semiparametric approach for investigating time-varying effects of a treatment. The method is best suited for data collected intensively over time (e.g., experience sampling or ecological momentary assessments) and addresses questions pertaining to effects of treatment changing dynamically with time. Thus, of interest is the description of timing, magnitude, and (nonlinear) patterns of the effect. Our presentation focuses on practical aspects of the model. A step-by-step demonstration is presented in the context of an empirical study designed to evaluate effects of surgical treatment on quality of life among early stage lung cancer patients during posthospitalization recovery (N = 59; 61% female, M age = 66.1 years). Frequency and level of distress associated with physical symptoms were assessed twice daily over a 2-week period, providing a total of 1,544 momentary assessments. Traditional analyses (analysis of covariance [ANCOVA], repeated-measures ANCOVA, and multilevel modeling) yielded findings of no group differences. In contrast, generalized TVEM identified a pattern of the effect that varied in time and magnitude. Group differences manifested after Day 4. Generalized TVEM is a flexible statistical approach that offers insight into the complexity of treatment effects and allows modeling of nonnormal outcomes. The practical demonstration, shared syntax, and availability of a free set of macros aim to encourage researchers to apply TVEM to complex data and stimulate important scientific discoveries.

Olanzapine-activated AMPK signaling in the dorsal vagal complex is attenuated by histamine H1 receptor agonist in female rats.

Weight gain and its related metabolic disorders are major side effects associated with second generation antipsychotic drug treatment. The dorsal vagal complex (DVC) and AMP-activated protein kinase (AMPK) are implicated in the regulation of food intake and body weight. Blocking the histamine H1 receptor contributes to antipsychotic-induced weight gain. The present study investigated the time-dependent effect of olanzapine treatment (8, 16, and 36 d) on DVC AMPK signaling in olanzapine-induced weight gain and whether these changes are associated with olanzapine-induced H1 receptor antagonism. During the 8-day olanzapine treatment, the rats were hyperphagic and rapidly gained weight. The phosphorylation of AMPK (pAMPK) (activated AMPK) as well as its directly downstream phospho-acetyl-coenzyme A carboxylase was significantly increased. The pAMPK/AMPK ratio, an indicator of AMPK activity, was significantly positively correlated with feeding efficiency and weight gain. As treatment was prolonged (16 and 36 d of olanzapine treatment), the rats were no longer hyperphagic, and there were no longer any changes in DVC AMPK signaling. Although the DVC H1 receptor protein expression was not significantly altered by olanzapine, the pAMPK expression was significantly positively correlated with the H1 receptor level after the 8-, 16-, and 36-day olanzapine treatments. Moreover, we showed that an H1 receptor agonist, 2-(3-trifluoromethylphenyl) histamine, significantly inhibited the olanzapine-induced hyperphagia and DVC AMPK activation in a dose-dependent manner. These results suggest a time-dependent role of DVC AMPK in olanzapine-induced obesity. Thus, olanzapine-induced DVC AMPK activation may be at least partially related to olanzapine's antagonistic effect on the H1 receptor.

Impact of duration and dosage of statin treatment and epicardial fat thickness on the recurrence of atrial fibrillation after electrical cardioversion

The purpose of this study was to investigate the time-dependent effect of statin treatment and echocardiographic epicardial fat thickness (EFT) on the maintenance of sinus rhythm (SR) in atrial fibrillation (AF) patients after electrical cardioversion (EC). One hundred sixty-three AF patients without previous statin treatment who underwent EC were consecutively enrolled. The maintenance rate of SR after EC (1, 3, 6, and 12 months) as documented by electrocardiogram and EFT were compared between patients with statin treatment (statin group, n = 63) and those without (no statin group, n = 100). There was no significant difference in the maintenance rate of SR between the groups soon after EC (statin group; 85.7 % vs. no statin; 84.8%, p = 0.535), after 1 month (71.0 vs. 59.1%, p = 0.091), and after 3 months (63.2 vs. 50.0%, p = 0.086). However, the maintenance rate of SR was significantly higher in the statin group compared to no statin group (61.8 vs. 42.9%, p = 0.024) after 6 months, and this significant difference persisted up to 12 months of follow up (60.1 vs. 36.4%, p = 0.001). Patients with recurrence showed higher baseline EFT (7.4 ± 2.7 vs. 8.5 ± 3.0 mm, p = 0.014). Multivariate linear regression analysis indicated that EFT, left atrial diameter, high-density lipoprotein cholesterol, statin treatment, and dose were the significant contributors to the maintenance of SR for all periods after EC. Statin treatment and low EFT were associated with a higher maintenance rate of SR in AF patients after EC. Significant benefit of statin was realized 6 months after EC, and this benefit was shown to be maintained over time.

Semiparametric inference on the absolute risk reduction and the restricted mean survival difference

For time-to-event data, when the hazards are non-proportional, in addition to the hazard ratio, the absolute risk reduction and the restricted mean survival difference can be used to describe the time-dependent treatment effect. The absolute risk reduction measures the direct impact of the treatment on event rate or survival, and the restricted mean survival difference provides a way to evaluate the cumulative treatment effect. However, in the literature, available methods are limited for flexibly estimating these measures and making inference on them. In this article, point estimates, pointwise confidence intervals and simultaneous confidence bands of the absolute risk reduction and the restricted mean survival difference are established under a semiparametric model that can be used in a sufficiently wide range of applications. These methods are motivated by and illustrated for data from the Women's Health Initiative estrogen plus progestin clinical trial.

Targeted maximum likelihood estimation for marginal time-dependent treatment effects under density misspecification

Targeted maximum likelihood methods have been proposed to estimate treatment effects for longitudinal data in the presence of time-dependent confounders. This class of methods has been mathematically proven to be doubly robust and to optimize the asymptotic estimating efficiency among the class of regular, semi-parametric estimators when all estimated density components are correctly specified. We show that methods previously proposed to build a one-step estimator with a logistic loss function generalize to a generalized linear loss function, and so may be applied naturally to an outcome that can be described by any exponential family member. We evaluate several methods for estimating unstructured marginal treatment effects for data with two time intervals in a simulation study, showing that these estimators have competitively low bias and variance in an array of misspecified situations, and can be made to perform well under near-positivity violations. We apply the methods to the PROmotion of Breastfeeding Intervention Trial data, demonstrating that longer term breastfeeding can protect infants from gastrointestinal infection.

Extending time-dependent instrumental variable estimation to estimate the causal effect of time-dependent exposure

Objectives: The purpose of the study was to compare the traditional models with the instrumental variable estimation method to estimate the causal effect of time-dependent exposure on the time-dependent outcome in the presence of unmeasured confounders for this effect. Methods: In recent studies, the instrumental variable (IV) estimation method of exposure effect frequently was used to collect information on potentially confounding variables. This is the case in particular where the time-dependent exposure is affected by the pervious events. We develop IV estimation method to estimate the effect of time-dependent exposure or treatment on timedependent outcome. This estimation method is designed for settings in which the exposure in time t is influenced by the exposure and outcome at time t − 1. Specifically, we extend IV estimation method in 2 stages in to a situation where exposure and outcome vary over the time. We show that our approach estimation in nonlinearity/additivity model is more consistent in comparison with the classical regression estimation method. This is done through simulation studies. Results: Our estimation strategy has larger mean square error (MSE) for the effect estimate in spite of having smaller bias in comparison with the tradition regression estimation method. However, this result might be the weak correlation between the time-dependent of instrumental variable and treatment. The case where the degree of correlation between time-depended instrumental variable and treatment at time t is stronger, we have more consistent estimate of causal effect. That is, our effect estimate tended to have less bias and less MSE, and more likely to have better confidence intervals and coverage probabilities. In the simulations study, we added interaction effect between exposures and measured confounder (nonadditivity model). In this model, by changing the association between outcome and confounders through different covariates, we could obtain more consistent estimate of the causal effect through our method than the classical method. The same results have been observed in case where we consider nonlinearity relationship between confounders and treatment. Alternatively, in some cases, we have better MSE for the classical estimation method, but we must note that its confidence intervals would not include the null causal effect value. Conclusions:We sketched a model for estimating the effect of timedependent treatment in settings in which classical methods may fail because of unmeasured confounding variables. We represented time-dependent instrumental variable to control the impact of measured and unmeasured confounders. We showed, after controlling confounding variables effect, that the effect of the previous treatment of interest that we wish to estimate could play the role of confounding variables under our assumptions and should be adjusted in the analysis. We showed that our approach has less bias than the classical regression models. Moreover, the mean square error of our method is large in some cases; this might be because our method performs in 2 steps. Not so surprisingly, strong association between instrument and treatment in each time reduces bias and the mean square error of the causal effect estimator. We showed also that our estimator is more efficient in coverage probabilities and confidence intervals in nonadditivity and nonlinearity/additivity models.

Response to “traits and stress: keys to identify community effects of low levels of toxicants in test systems” by Liess and Beketov (2011)

Response to “traits and stress: keys to identify community effects of low levels of toxicants in test systems” by Liess and Beketov (2011)

Time-Dependent Beneficial Effect of Chronic Polyphenol Treatment with Catechin on Endothelial Dysfunction in Aging Mice

A controlled redox environment is essential for vascular cell maturation and function. During aging, an imbalance occurs, leading to endothelial dysfunction. We hypothesized that, according to the concept of hormesis, exposure to physiologic oxidative stress during the maturation phase of the endothelium will activate protective pathways involved in stress resistance. C57Bl/6 mice were treated with the polyphenol catechin for the last 3 (post-maturation) or 9 months prior study at 12 months of age. Endothelial dysfunction, assessed by acetylcholine-induced dilations of isolated renal arteries, was present at 12 months (P<0.05). Only the 3-month treatment with catechin fully prevented the decline in efficacy and sensitivity to acetylcholine (P<0.05). Splenocytes adhesion to the native endothelium, expression of CD18 and shedding of CD62L and PSGL-1 augmented in 12 months old mice (P<0.05): only 3-month catechin fully normalized adhesion and prevented the expression of adhesion molecules on splenocytes (P<0.05). Aging was associated with vascular gene alterations, which were prevented by 3-month catechin treatment (P<0.05). In contrast, 9-month catechin further increased COX-2, p22(phox) and reduced MnSOD (P<0.05). In conclusion, we demonstrate a pivotal role of cellular redox equilibrium: exposure to physiologic oxidative stress during the maturation phase of the endothelium is essential for its function.

Extracorporeal Shock Wave Therapy (ESWT) Minimizes Ischemic Tissue Necrosis Irrespective of Application Time and Promotes Tissue Revascularization by Stimulating Angiogenesis

To assess the time-dependent treatment effects of extracorporeal shock wave therapy (ESWT) in a standard rodent ischemic epigastric flap model. ESWT has been shown to accelerate tissue repair in acute and chronic wounds and improve graft survival, but the mechanism remains incompletely understood. Shock waves at 0.1 mJ/mm and 5 impulses/s (total 300 impulses) were applied to the epigastric flap ischemic region at various times pre-, immediately and 24 hours postischemic insult. Flap survival; vascular perfusion; vessel number; von Willebrand factor and smooth muscle actin protein expression as well as in vivo vascular endothelial growth factor receptor 2 expression were evaluated at 1, 3, and 7 days postoperatively in ESWT-treated and untreated controls. Flap perfusion, microvessel number, and survival (through reduced flap contraction and necrosis) were significantly enhanced in the treated groups compared with controls, irrespective of timing of shock wave treatment (preischemia vs. postischemia). Vascular endothelial growth factor receptor 2 expression was dynamically upregulated in response to ESWT. Shock wave preconditioning and treatment postischemic insult improves skin flap survival through neovascularization and early upregulation of angiogenesis-related growth factors.

Time-Dependent Activation of c-fos in Limbic Brain Areas by Ocular Administration of Nerve Growth Factor in Adult Rats

The ocular administration of the neurotrophin nerve growth factor (NGF) has been successfully used in humans to recover damaged ocular tissues. Studies on animal models have demonstrated the ability of ocular applied NGF to reach the retina and the optic nerve and affect brain visual areas. The aim of this study was to examine whether the ocular application of NGF as eye drops might affect brain areas other than the primary visual centers. Two drops (10 μL) of NGF solution (200 μg/mL) or saline were applied as collyrium to both eyes of adult male Sprague-Dawley rats. The animals were sacrificed at 4, 8, or 24 h after treatment and the brains were fixed through intracardiac perfusion. Coronal brain sections were cut with a cryostat and used for immunohistochemical time series and double immunofluorescence studies using c-fos and NeuN as markers for neuronal activation. The immunohistochemical studies show a time-dependent effect of NGF eye drop treatment. At 4 h after NGF ocular administration, the increase in c-fos immunoreactivity is mainly observed in areas belonging to the central visual system, such as the lateral geniculate nucleus and visual cortex. At 8 h posttreatment, c-fos expression is enhanced in several limbic structures, including the frontal cortex, hippocampus, amygdala, and hypothalamus. The effects of NGF on c-fos distribution persist at 24 h postadministration. Specificity of NGF-induced c-fos in brain was confirmed using inactivated NGF. The neuronal nature of the NGF-activated cells was demonstrated by confocal microscopy observation of c-fos and NeuN colocalization. This study demonstrates that NGF, when applied on ocular surface, is able to activate c-fos in several areas of the limbic system in a time-dependent manner. These findings suggest that the effects of NGF eye drops are not restricted to the primary visual areas, but are extended to all the retinal central targets, including the forebrain structure. Based on these data, the use of NGF eye drops as a strategy to produce NGF-mediated protective and reparative actions in brain is hypothesizable.

Using selective herbicides to manage beneficial and rare weed species in winter wheat

Weed species within arable farming systems act as a source of food and provide habitat for various taxa at higher trophic levels. Using field experiments in winter wheat, we investigated whether selective herbicides can be used as a potential tool for sustaining beneficial, less competitive weed species and rare species of high conservation value while controlling highly competitive weed species. We tested three selective herbicides with different active ingredients (amidosulfuron + iodosulfuron, fluroxypyr, mecoprop-P) at two application rates. We analysed cover of selected focal species (Centaurea cyanus and Papaver rhoeas) as well as weed species richness, winter wheat yield and weed community composition as influenced by herbicide treatments. Herbicide treatments involving fluroxypyr generally allowed for the selective retention of the regionally threatened weed species C. cyanus and P. rhoeas as a beneficial species of high value for associated trophic level and provided considerable control of the competitive species G. aparine. Herbicide treatments significantly affected weed species composition. The PRC method clearly showed the time-dependent herbicide treatment effects on weed community composition and allowed an estimate of the variance in species composition that is explained by the herbicide treatments. Our results indicated that applying selective herbicides can provide a feasible option for in-crop management of weed diversity by retention of beneficial or rare weed species occurring at moderate densities and control of highly-competitive weeds. However, weed control strategies must be tailored to site-specific conditions to account for increased abundance of competitive species.

Double Inverse-Weighted Estimation of Cumulative Treatment Effects Under Nonproportional Hazards and Dependent Censoring

In medical studies of time-to-event data, nonproportional hazards and dependent censoring are very common issues when estimating the treatment effect. A traditional method for dealing with time-dependent treatment effects is to model the time-dependence parametrically. Limitations of this approach include the difficulty to verify the correctness of the specified functional form and the fact that, in the presence of a treatment effect that varies over time, investigators are usually interested in the cumulative as opposed to instantaneous treatment effect. In many applications, censoring time is not independent of event time. Therefore, we propose methods for estimating the cumulative treatment effect in the presence of nonproportional hazards and dependent censoring. Three measures are proposed, including the ratio of cumulative hazards, relative risk, and difference in restricted mean lifetime. For each measure, we propose a double inverse-weighted estimator, constructed by first using inverse probability of treatment weighting (IPTW) to balance the treatment-specific covariate distributions, then using inverse probability of censoring weighting (IPCW) to overcome the dependent censoring. The proposed estimators are shown to be consistent and asymptotically normal. We study their finite-sample properties through simulation. The proposed methods are used to compare kidney wait-list mortality by race.

Duration of Symptoms Is Not Always the Key Modulator of the Choice of Reperfusion for ST-Elevation Myocardial Infarction

Fibrinolytic therapy (FT) and primary percutaneous coronary intervention (PCI) are both well-accepted reperfusion therapies in ST-segment elevation myocardial infarction (STEMI). The evidence of randomized clinical trials indicates a relatively modest difference in 30-day mortality (≈1%) in favor of primary PCI over fibrin-specific FT and was based on very timely primary PCI (ie, a primary PCI–related delay of 40 minutes [door-to-balloon less door-to-needle time]).1 Longer delays to primary PCI, which are far more frequent in clinical practice,2 are associated with attenuated benefit or no benefit at all, particularly when compared with fibrin-specific FT.3,4 The benefit of timely primary PCI over FT is likely to especially apply to higher-risk patients.5,6 Irrespective of the method of reperfusion, the potential for myocardial salvage and better clinical outcome is inversely proportional to ischemic time or its only available clinical surrogate, symptom duration.7–12 These considerations underpin the notion expressed in the American College of Cardiology/American Heart Association (ACC/AHA) guidelines on the treatment of STEMI that timely reperfusion therapy is likely more important in determining outcome than whether FT or primary PCI is the chosen reperfusion method.13 Response by Armstrong et al p 1310 The ACC/AHA STEMI guidelines highlight the time point of 3 hours of symptom duration in guiding the choice of reperfusion therapy. They state that if symptom duration is <3 hours, no preference exists between FT and primary PCI provided that treatment is timely (for FT, door-to-needle time <30 minutes; for primary PCI, door-to-balloon time <90 minutes and ≤60 minutes between estimated needle time and estimated balloon time). However, if symptom duration exceeds 3 hours, these guidelines favor primary PCI over FT, again provided that primary PCI can be performed in a timely fashion.13 This article reexamines the evidence that may or may not be the basis …

Estimation of treatment effect adjusting for treatment changes using the intensity score method: Application to a large primary prevention study for coronary events (MEGA study)

The MEGA study was a prospective, randomized, open-labeled, blinded-endpoints study conducted in Japan to evaluate the primary preventive effect of pravastatin against coronary heart disease (CHD), in which 8214 subjects were randomized to diet or diet plus pravastatin. The intention-to-treat (ITT) analysis showed that pravastatin reduced the incidence of CHD (hazard ratio=0.67; 95 per cent confidence interval (CI): 0.49-0.91) and of stroke events, which was the secondary endpoint in the MEGA study (hazard ratio=0.83; 95 per cent CI: 0.57-1.21). Owing to considerable treatment changes, it is also of interest to estimate the causal effect of treatment that would have been observed had all patients complied with the treatment to which they were assigned. In this paper, we present an intensity score method developed for clinical trials with time-to-event outcomes that correct for treatment changes during follow-up. The proposed method can be easily extended to the estimation of time-dependent treatment effects, where the technique of g-estimation has been difficult to apply in practice. We compared the performances of the proposed method with other methods (as-treated, ITT, and g-estimation analysis) through simulation studies, which showed that the intensity score estimator was unbiased and more efficient. Applying the proposed method to the MEGA study data, several prognostic factors were associated with the process of treatment changes, and after adjusting for these treatment changes, larger treatment effects for pravastatin were observed for both CHD and stroke events. The proposed method provides a valuable and flexible approach for estimating treatment effect adjusting for non-random non-compliance.

Bootstrap confidence intervals for principal response curves

The principal response curve (PRC) model is of use to analyse multivariate data resulting from experiments involving repeated sampling in time. The time-dependent treatment effects are represented by PRCs, which are functional in nature. The sample PRCs can be estimated using a raw approach, or the newly proposed smooth approach. The generalisability of the sample PRCs can be judged using confidence bands. The quality of various bootstrap strategies to estimate such confidence bands for PRCs is evaluated. The best coverage was obtained with BC a intervals using a non-parametric bootstrap. The coverage appeared to be generally good, except for the case of exactly zero population PRCs for all conditions. Then, the behaviour is irregular, which is caused by the sign indeterminacy of the PRCs. The insights obtained into the optimal bootstrap strategy are useful to apply in the PRC model, and more generally for estimating confidence intervals in singular value decomposition based methods.