How to account for early overly small risk sets in the analysis of pregnancy outcome data?-Comparison of different methods for stabilizing the Aalen-Johansen estimator.

PURPOSE: In analyzing pregnancy data concerning drug exposure in the first trimester, the risk of spontaneous abortions is of primary interest. For estimating the cumulative incidence function, the Aalen-Johansen estimator is typically used, and competing risks such as induced abortion and livebirth are considered. However, the delayed study entry can lead to overly small risk sets for the first events. This results in large jumps in the estimated cumulative incidence function of spontaneous abortions or induced abortions using the Aalen-Johansen estimator, and consequently in an overestimation of the probability. METHODS: Several approaches account for early overly small risk sets. The first approach is conditioning on the event time being greater than the event time causing the large jump. Second, the events can be ignored by censoring them. Third, the events can be postponed until a large enough number is at risk. These three approaches are compared. RESULTS: All approaches are applied using data of 54 lacosamide-exposed pregnancies. The Aalen-Johansen estimate of the probability of spontaneous abortion is 22.64%, which is relatively large for only three spontaneous abortions in the dataset. The conditional approach and the ignore approach have an estimated probability of 7.17%. In contrast, the estimate of the postpone approach is 16.45%. In this small sample, bootstrapped confidence intervals seem more accurate. CONCLUSIONS: In the analyses of pregnancy data with rare events, the postpone approach is favorable as no events are excluded. However, the approach that ignores early events has the narrowest confidence interval.

Comparison of nonparametric estimators of the expected number of recurrent events.

We compare the performance of nonparametric estimators for the mean number of recurrent events and provide a systematic overview for different recurrent event settings. The mean number of recurrent events is an easily interpreted marginal feature often used for treatment comparisons in clinical trials. Incomplete observations, dependencies between successive events, terminating events acting as competing risk, or gaps between at risk periods complicate the estimation. We use survival multistate models to represent different complex recurrent event situations, profiting from recent advances in nonparametric estimation for non-Markov multistate models, and explain several estimators by using multistate intensity processes, including the common Nelson-Aalen-type estimators with and without competing mortality. In addition to building on estimation of state occupation probabilities in non-Markov models, we consider a simple extension of the Nelson-Aalen estimator by allowing for dependence on the number of prior recurrent events. We pay particular attention to the assumptions required for the censoring mechanism, one issue being that some settings require the censoring process to be entirely unrelated while others allow for state-dependent or event-driven censoring. We conducted extensive simulation studies to compare the estimators in various complex situations with recurrent events. Our practical example deals with recurrent chronic obstructive pulmonary disease exacerbations in a clinical study, which will also be used to illustrate two-sample-inference using resampling.

Adverse events after left ventricular assist device implantation linked to psychosocial risk in women and men.

BACKGROUND: Reasons for women's increased probability to experience adverse events (AEs) after left ventricular assist device (LVAD) implantation compared with men's remain uncertain. We explored the role of psychosocial risk in the experience of AEs in women and men. METHODS: INTERMACS patients receiving a primary continuous-flow LVAD between July 2006 and December 2017, median follow-up 13.6 months, were included (n = 20,123, 21.3% women). Time-to-event was calculated with cumulative incidence functions for 10 types of AEs separately (e.g., infection, device malfunction), each time accounting for the competing outcomes death, heart transplant, and device explant due to recovery. Event-specific Cox proportional hazard models were run with a binary psychosocial risk variable (including substance abuse, psychiatric diagnosis, limited social support, limited cognition, repeated noncompliance), controlled for covariates. RESULTS: Psychosocial risk was more prevalent in men than in women (21.4% vs 17.5%, p < 0.001). Seven out of 10 AEs were more likely in women than in men (e.g., infection 44.5% vs 39.2%, p < 0.001). The association of psychosocial risk with each AE was either stronger in women than in men (e.g., device malfunction HRadj 1.29, 95% confidence interval (CI) (1.06-1.56) vs HRadj 1.10, 95% CI (0.97-1.25); rehospitalization HRadj 1.15, 95% CI (1.02-1.29) vs HRadj 1.03, 95% CI (0.97-1.10) or similar between sexes. CONCLUSIONS: Independent of clinical parameters, the presence of psychosocial risk is associated with increases in AEs. This suggests that early modification of psychosocial risk factors may have the potential to lower the risk for AEs in this patient population.

Increasing use of newer antiseizure medication during pregnancy: An observational study with special focus on lacosamide.

INTRODUCTION: Epilepsy is a common neurological disease requiring long-term therapy also during pregnancy. Most studies on pregnancy outcomes in women with epilepsy are based on antiseizure medication (ASM) in monotherapy. However, about 20-30% of epilepsy patients require polytherapy and newer ASMs are an option, when seizure control is not achieved with first line ASMs. METHODS: Observational study evaluating the use of newer ASMs with marketing authorization since 2005 reported to the Embryotox Center of Clinical Teratology and Drug Safety in Pregnancy between 2004 and 2019. In addition, course and outcome of lacosamide exposed pregnancies were analysed. RESULTS: Our study confirms the increasing use of newer ASMs also in pregnant women. This is especially true for lacosamide, eslicarbazepine and brivaracetam with rising numbers of exposed pregnancies soon after marketing authorization. Analysis of 55 prospectively and 10 retrospectively ascertained lacosamide exposed pregnancies does not indicate increased risks of major birth defects or spontaneous abortion. However, bradycardia observed in 3 neonates might be related to prenatal lacosamide exposure. CONCLUSION: Available data do not support the assumption of lacosamide being a major teratogen. The increasing use of newer ASMs during pregnancy underscores the need for more studies to guide preconception counselling, especially for lacosamide, eslicarbazepine and brivaracetam.

Modeling unmeasured baseline information in observational time-to-event data subject to delayed study entry.

Unmeasured baseline information in left-truncated data situations frequently occurs in observational time-to-event analyses. For instance, a typical timescale in trials of antidiabetic treatment is "time since treatment initiation", but individuals may have initiated treatment before the start of longitudinal data collection. When the focus is on baseline effects, one widespread approach is to fit a Cox proportional hazards model incorporating the measurements at delayed study entry. This has been criticized because of the potential time dependency of covariates. We tackle this problem by using a Bayesian joint model that combines a mixed-effects model for the longitudinal trajectory with a proportional hazards model for the event of interest incorporating the baseline covariate, possibly unmeasured in the presence of left truncation. The novelty is that our procedure is not used to account for non-continuously monitored longitudinal covariates in right-censored time-to-event studies, but to utilize these trajectories to make inferences about missing baseline measurements in left-truncated data. Simulating times-to-event depending on baseline covariates we also compared our proposal to a simpler two-stage approach which performed favorably. Our approach is illustrated by investigating the impact of baseline blood glucose levels on antidiabetic treatment failure using data from a German diabetes register.

A connection between survival multistate models and causal inference for external treatment interruptions.

Recently, treatment interruptions such as a clinical hold in randomized clinical trials have been investigated by using a multistate model approach. The phase III clinical trial START (Stimulating Targeted Antigenic Response To non-small-cell cancer) with primary endpoint overall survival was temporarily placed on hold for enrollment and treatment by the US Food and Drug Administration (FDA). Multistate models provide a flexible framework to account for treatment interruptions induced by a time-dependent external covariate. Extending previous work, we propose a censoring and a filtering approach both aimed at estimating the initial treatment effect on overall survival in the hypothetical situation of no clinical hold. A special focus is on creating a link to causal inference. We show that calculating the matrix of transition probabilities in the multistate model after application of censoring (or filtering) yields the desired causal interpretation. Assumptions in support of the identification of a causal effect by censoring (or filtering) are discussed. Thus, we provide the basis to apply causal censoring (or filtering) in more general settings such as the COVID-19 pandemic. A simulation study demonstrates that both causal censoring and filtering perform favorably compared to a naïve method ignoring the external impact.

General independent censoring in event-driven trials with staggered entry.

Randomized clinical trials with time-to-event endpoints are frequently stopped after a prespecified number of events has been observed. This practice leads to dependent data and nonrandom censoring, which can in general not be solved by conditioning on the underlying baseline information. In case of staggered study entry, matters are complicated substantially. The present paper demonstrates that the study design at hand entails general independent censoring in the counting process sense, provided that the analysis is based on study time information only. To illustrate that the filtrations must not use abundant information, we simulated data of event-driven trials and evaluated them by means of Cox regression models with covariates for the calendar times. The Breslow curves of the cumulative baseline hazard showed considerable deviations, which implies that the analysis is disturbed by conditioning on the calendar time variables. A second simulation study further revealed that Efron's classical bootstrap, unlike the (martingale-based) wild bootstrap, may lead to biased results in the given setting, as the assumption of random censoring is violated. This is exemplified by an analysis of data on immunotherapy in patients with advanced, previously treated nonsmall cell lung cancer.

Long-Term Mortality of Patients With Osteoarthritis After Joint Replacement: Prognostic Value of Preoperative and Postoperative Pain and Function.

OBJECTIVE: To investigate whether osteoarthritis (OA)-specific assessment values (i.e., Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]) and generic pain and function (visual analog scale, Hanover Functionality Status Questionnaire) measured before and 12 months after arthroplasty are associated with the risk of long-term mortality in a cohort of patients with advanced OA of the hip or knee. METHODS: The Ulm Osteoarthritis Study was a prospective cohort study of OA patients with unilateral total hip or knee replacement between January 1995 and December 1996. Correlation coefficients were calculated to describe the agreement between the different assessments. Mortality was assessed during the follow-up period (last update July 2019). Cox proportional regression models were used to estimate hazard ratios (HRs) for mortality after adjusting for covariates. RESULTS: Arthroplasty was accompanied by a clear reduction in pain and improved function throughout all assessments in the 706 included patients. The results of the adjusted Cox models showed no relationship between baseline and follow-up joint-specific WOMAC assessments and long-term mortality. However, an independent increased risk of mortality was found with generic function assessments. In the final adjusted model, the HR for the 12-month follow-up value was 1.79 (95% confidence interval 1.24-2.60) in the group with clinically relevant impairment versus the reference group. CONCLUSION: Poor function based on the generic assessment was associated with increased long-term mortality, suggesting that functional impairments in daily life activities may be more important for long-term survival than OA-specific impairments in this patient group.

Recurrent disability progression endpoints in multiple sclerosis clinical trials.

BACKGROUND: The current standard endpoint to assess disability accumulation in multiple sclerosis (MS) clinical trials is the time to the first confirmed disability progression, which excludes subsequent progression events. Including recurrent progression events may permit a more comprehensive assessment of treatment effects on disability progression. OBJECTIVE: To propose a definition of recurrent disability progression events and to compare time-to-first and recurrent event analysis. METHODS: Recurrent disability progression events were defined by expanding the recommended first event definition. Marginal recurrent event methods (negative binomial model, Lin-Wei-Yang-Ying model) were compared with Cox regression in data from three randomized controlled trials in relapsing multiple sclerosis (RMS) and primary progressive multiple sclerosis (PPMS), and in simulated randomized controlled trial data. RESULTS: The recurrent event analyses included a substantially larger number of progression events compared with the time-to-first-event analyses (+7.5% and +9.9% in the RMS trials and +22.7% in the PPMS trial). The increase in the number of events resulted in more precise treatment effect estimates and a corresponding gain in statistical power. CONCLUSION: Our results support the use of recurrent event data analysis, especially in progressive MS trials, to improve estimates of treatment effects, increase statistical power, and better capture the clinically meaningful long-term disability progression experience.

Sex Differences in Recovery and Device Replacement After Left Ventricular Assist Device Implantation as Destination Therapy.

Background The relevance of sex and preimplant factors for clinical outcomes among patients with left ventricular assist devices intended for destination therapy is unclear. Methods and Results INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support) data (2006-2017) from 6771 men and 1690 women with left ventricular assist devices as destination therapy were analyzed to evaluate the contribution of preimplant clinical, demographic, and clinically judged psychosocial characteristics to time until death, heart transplant, device explant due to recovery, or complication-related device replacement. Associations of sex with time until each competing outcome were evaluated using cumulative incidence functions and event-specific Cox proportional hazards models. Women were younger, more likely to have nonischemic diagnoses, and reported less substance abuse but were more likely to be unmarried, not working for an income, overweight, and depressed than men. After 2 years, women had higher probabilities for recovery (3.7% versus 1.6%, P<0.001) and device replacement (12.1% versus 10%, P=0.019) than men but not for death and transplant (P>0.12). The sex differences remained after controlling for covariates (adjusted hazard ratio [HRadj] recovery, 1.85; 95% CI, 1.30-2.70; P<0.001; HRadj device replacement, 1.22; 95% CI, 1.04-1.33; P=0.015). Female-specific diagnoses (eg, postpartum heart failure) contributed to women's enhanced rate of recovery. Demographic and psychosocial factors were unrelated to women's increased event rates. Conclusions In destination therapy, women have higher rates of device replacement and recovery than men. The latter was partly explained by female-specific diagnoses. Standardized assessments of psychosocial characteristics are needed to elucidate their association with sex differences in outcomes.

Evolution of disability in spinocerebellar ataxias type 1, 2, 3, and 6.

OBJECTIVE: The aim was to study the evolution of disability in spinocerebellar ataxias (SCAs) type 1, 2, 3, and 6 (SCA1, 2, 3, 6). METHODS: We analyzed data of two longitudinal cohorts (RISCA, EUROSCA) which recruited ataxic and non-ataxic SCA1, SCA2, SCA3, and SCA6 mutation carriers. To study disability, we used a five-stage system for ataxia defined by walking ability (stages 0-3) and death (stage 4). Transitions were analyzed using a multi-state model with proportional transition hazards. Based on the hazard estimates, transition probabilities and the expected lengths of stay in each stage were calculated. We further studied the effect of sex and CAG repeat length on progression. RESULTS: Data of 3138 visits in 677 participants were analyzed. Median SARA scores for SCA1, SCA2, SCA3, and SCA6 ranged from 1.5 (interquartile range [IQR] = 0.0-3.5) to 3.5 (IQR = 1.4-6.1) in stage 0, 11.5 (IQR = 9.6-14.0) to 13.8 (IQR = 11.0-16.0) in stage 1, 19.0 (IQR = 17.0-21.0) to 23.8 (IQR = 19.5-27.0) in stage 2, and 28.5 (IQR = 26.0-32.5) to 34.0 (IQR = 32.6-37.1) in stage 3. Modeling allowed to calculate the subtype-specific probability to be in a certain stage at a given age and duration of each stage. CAG repeat length was associated with faster progression in SCA1 (HR, 95% CI: 1.1, 1.1-1.2), SCA2 (1.2, 1.1-1.3), and SCA3 (1.1, 1.0-1.2). In SCA6, female sex was associated with faster progression (1.7, 1.1-2.6). INTERPRETATION: Our data are important for counselling of patients, assessment of the relevance of outcome markers, and design of clinical trials.

Design aspects of COVID-19 treatment trials: Improving probability and time of favorable events.

As a reaction to the pandemic of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a multitude of clinical trials for the treatment of SARS-CoV-2 or the resulting corona disease 2019 (COVID-19) are globally at various stages from planning to completion. Although some attempts were made to standardize study designs, this was hindered by the ferocity of the pandemic and the need to set up clinical trials quickly. We take the view that a successful treatment of COVID-19 patients (i) increases the probability of a recovery or improvement within a certain time interval, say 28 days; (ii) aims to expedite favorable events within this time frame; and (iii) does not increase mortality over this time period. On this background, we discuss the choice of endpoint and its analysis. Furthermore, we consider consequences of this choice for other design aspects including sample size and power and provide some guidance on the application of adaptive designs in this particular context.

Trueness of full-arch IO scans estimated based on 3D translational and rotational deviations of single teeth-an in vitro study.

OBJECTIVES: To three-dimensionally evaluate deviations of full-arch intraoral (IO) scans from reference desktop scans in terms of translations and rotations of individual teeth and different types of (mal)occlusion. MATERIALS AND METHODS: Three resin model pairs reflecting different tooth (mal)positions were mounted in the phantom head of a dental simulation unit and scanned by three dentists and three non-graduate investigators using a confocal laser IO scanner (Trios 3®). The tooth-crown surfaces of the IO scans and reference scans were superimposed by means of best-fit alignment. A novel method comprising the measurement of individual tooth positions was used to determine the deviations of each tooth in the six degrees of freedom, i.e., in terms of 3D translation and rotation. Deviations between IO and reference scans, among tooth-(mal)position models, and between dentists and non-graduate investigators were analyzed using linear mixed-effects models. RESULTS: The overall translational deviations of individual teeth on the IO scans were 76, 32, and 58 µm in the lingual, mesial, and intrusive directions, respectively, resulting in a total displacement of 114 µm. Corresponding rotational deviations were 0.58° buccal tipping, 0.04° mesial tipping, and 0.14° distorotation leading to a combined rotation of 0.78°. These deviations were the smallest for the dental arches with anterior crowding, followed by those with spacing and those with good alignment (p < 0.05). Results were independent of the operator's level of education. CONCLUSIONS: Compared to reference desktop scans, individual teeth on full-arch IO scans showed high trueness with total translational and rotational deviations < 115 µm and < 0.80°, respectively. CLINICAL RELEVANCE: Available confocal laser IO scanners appear sufficiently accurate for diagnostic and therapeutic orthodontic applications. Results indicate that full-arch IO scanning can be delegated to non-graduate dental staff members.

Psychosocial Risk and Health Behaviors as Predictors of Clinical Events in Patients Wait-Listed for a New Heart: Results from 7 Years of Follow-Up.

We examined the long-term relationship of psychosocial risk and health behaviors on clinical events in patients awaiting heart transplantation (HTx). Psychosocial characteristics (e.g., depression), health behaviors (e.g., dietary habits, smoking), medical factors (e.g., creatinine), and demographics (e.g., age, sex) were collected at the time of listing in 318 patients (82% male, mean age = 53 years) enrolled in the Waiting for a New Heart Study. Clinical events were death/delisting due to deterioration, high-urgency status transplantation (HU-HTx), elective transplantation, and delisting due to clinical improvement. Within 7 years of follow-up, 92 patients died or were delisted due to deterioration, 121 received HU-HTx, 43 received elective transplantation, and 39 were delisted due to improvement. Adjusting for demographic and medical characteristics, the results indicated that frequent consumption of healthy foods (i.e., foods high in unsaturated fats) and being physically active increased the likelihood of delisting due improvement, while smoking and depressive symptoms were related to death/delisting due to clinical deterioration while awaiting HTx. In conclusion, psychosocial and behavioral characteristics are clearly associated with clinical outcomes in this population. Interventions that target psychosocial risk, smoking, dietary habits, and physical activity may be beneficial for patients with advanced heart failure waiting for a cardiac transplant.

Survival analysis for AdVerse events with VarYing follow-up times (SAVVY)-estimation of adverse event risks.

BACKGROUND: The SAVVY project aims to improve the analyses of adverse events (AEs), whether prespecified or emerging, in clinical trials through the use of survival techniques appropriately dealing with varying follow-up times and competing events (CEs). Although statistical methodologies have advanced, in AE analyses, often the incidence proportion, the incidence density, or a non-parametric Kaplan-Meier estimator are used, which ignore either censoring or CEs. In an empirical study including randomized clinical trials from several sponsor organizations, these potential sources of bias are investigated. The main purpose is to compare the estimators that are typically used to quantify AE risk within trial arms to the non-parametric Aalen-Johansen estimator as the gold-standard for estimating cumulative AE probabilities. A follow-up paper will consider consequences when comparing safety between treatment groups. METHODS: Estimators are compared with descriptive statistics, graphical displays, and a more formal assessment using a random effects meta-analysis. The influence of different factors on the size of deviations from the gold-standard is investigated in a meta-regression. Comparisons are conducted at the maximum follow-up time and at earlier evaluation times. CEs definition does not only include death before AE but also end of follow-up for AEs due to events related to the disease course or safety of the treatment. RESULTS: Ten sponsor organizations provided 17 clinical trials including 186 types of investigated AEs. The one minus Kaplan-Meier estimator was on average about 1.2-fold larger than the Aalen-Johansen estimator and the probability transform of the incidence density ignoring CEs was even 2-fold larger. The average bias using the incidence proportion was less than 5%. Assuming constant hazards using incidence densities was hardly an issue provided that CEs were accounted for. The meta-regression showed that the bias depended mainly on the amount of censoring and on the amount of CEs. CONCLUSIONS: The choice of the estimator of the cumulative AE probability and the definition of CEs are crucial. We recommend using the Aalen-Johansen estimator with an appropriate definition of CEs whenever the risk for AEs is to be quantified and to change the guidelines accordingly.

Motor learning might contribute to a therapeutic anterior shift of the habitual mandibular position-An exploratory study.

BACKGROUND: Passive mandibular advancement with functional appliances is commonly used to treat juvenile patients with mandibular retrognathism. OBJECTIVE: The aim of this study was to investigate whether active repetitive training of the mandible into an anterior position would result in a shift of the habitual mandibular position (HMP). METHODS: Twenty adult healthy subjects were randomly assigned to one of two groups: a training group receiving six supervised functional training sessions of 10 min each and a control group without training. Bonded lateral biteplates disengaged occlusion among both groups throughout the 15-day experiment. Customised registration-training appliances consisted of a maxillary component with an anterior plane and a mandibular component with an attached metal sphere. Training sessions consisted of repeated mouth-opening/closing cycles (frequency: 30/min) to hit an anteriorly positioned hemispherical target notch with this metal sphere. The HMP was registered at defined times during the experiment. RESULTS: The HMP in the training group showed a statistically significant anterior shift of 1.6 mm (interquartile range [IQR]: 1.2 mm), compared with a significant posterior shift of -0.8 mm (IQR: 2.8 mm) in the control group (p < .05). Although the anterior shift among the training group showed a partial relapse 4 days after the first training block, it then advanced slightly in the 4-day interval after the second training block, which might indicate neuroplasticity of the masticatory motor system. CONCLUSIONS: Motor learning by repetitive training of the mandible into an anterior position might help to improve the results of functional appliance therapy among patients with mandibular retrognathism.

Estimating and comparing adverse event probabilities in the presence of varying follow-up times and competing events.

Safety analyses of adverse events (AEs) are important in assessing benefit-risk of therapies but are often rather simplistic compared to efficacy analyses. AE probabilities are typically estimated by incidence proportions, sometimes incidence densities or Kaplan-Meier estimation are proposed. These analyses either do not account for censoring, rely on a too restrictive parametric model, or ignore competing events. With the non-parametric Aalen-Johansen estimator as the "gold standard", that is, reference estimator, potential sources of bias are investigated in an example from oncology and in simulations, for both one-sample and two-sample scenarios. The Aalen-Johansen estimator serves as a reference, because it is the proper non-parametric generalization of the Kaplan-Meier estimator to multiple outcomes. Because of potential large variances at the end of follow-up, comparisons also consider further quantiles of the observed times. To date, consequences for safety comparisons have hardly been investigated, the impact of using different estimators for group comparisons being unclear. For example, the ratio of two both underestimating or overestimating estimators may not be comparable to the ratio of the reference, and our investigation also considers the ratio of AE probabilities. We find that ignoring competing events is more of a problem than falsely assuming constant hazards by the use of the incidence density and that the choice of the AE probability estimator is crucial for group comparisons.

Assessment of assumptions of statistical analysis methods in randomised clinical trials: the what and how.

When analysing and presenting results of randomised clinical trials, trialists rarely report if or how underlying statistical assumptions were validated. To avoid data-driven biased trial results, it should be common practice to prospectively describe the assessments of underlying assumptions. In existing literature, there is no consensus on how trialists should assess and report underlying assumptions for the analyses of randomised clinical trials. With this study, we developed suggestions on how to test and validate underlying assumptions behind logistic regression, linear regression, and Cox regression when analysing results of randomised clinical trials.Two investigators compiled an initial draftbased on a review of the literature. Experienced statisticians and trialists from eight different research centres and trial units then participated in a anonymised consensus process, where we reached agreement on the suggestions presented in this paper.This paper provides detailed suggestions on 1) which underlying statistical assumptions behind logistic regression, multiple linear regression and Cox regression each should be assessed; 2) how these underlying assumptions may be assessed; and 3) what to do if these assumptions are violated.We believe that the validity of randomised clinical trial results will increase if our recommendations for assessing and dealing with violations of the underlying statistical assumptions are followed.

Survival analysis for AdVerse events with VarYing follow-up times (SAVVY): Rationale and statistical concept of a meta-analytic study.

The assessment of safety is an important aspect of the evaluation of new therapies in clinical trials, with analyses of adverse events being an essential part of this. Standard methods for the analysis of adverse events such as the incidence proportion, that is the number of patients with a specific adverse event out of all patients in the treatment groups, do not account for both varying follow-up times and competing risks. Alternative approaches such as the Aalen-Johansen estimator of the cumulative incidence function have been suggested. Theoretical arguments and numerical evaluations support the application of these more advanced methodology, but as yet there is to our knowledge only insufficient empirical evidence whether these methods would lead to different conclusions in safety evaluations. The Survival analysis for AdVerse events with VarYing follow-up times (SAVVY) project strives to close this gap in evidence by conducting a meta-analytical study to assess the impact of the methodology on the conclusion of the safety assessment empirically. Here we present the rationale and statistical concept of the empirical study conducted as part of the SAVVY project. The statistical methods are presented in unified notation, and examples of their implementation in R and SAS are provided.

Five years' trajectories of functionality and pain in patients after hip or knee replacement and association with long-term patient survival.

To describe the 5 years' trajectories in functionality and pain of patients with hip or knee osteoarthritis and arthroplasty and analyze the association of these with long-term patients survival. Patients with OA receiving total hip or knee arthroplasty were recruited and completed two sets of standardized questionnaires for functionality and pain 6, 12, and 60 months postoperatively. Multivariate mixed models were conducted to assess trajectories over time and the resulting improvement per month during the last time period was included in a landmark-model to estimate adjusted hazard ratios for mortality. In total 809 patients with joint replacement were included (mean age 65.0 years, 62.2% female), 407 patients died (median follow-up 18.4 years). Both instruments of functionality and pain showed extensive improvement during the first 6 months. Baseline and change in functionality (both p < 0.001) and pain (p = 0.02) during the first 6 months were associated with mortality. Better values in functionality corresponded with improved survival whereas the association with the pain scores was inverse. In patients with hip and knee OA, an explicit improvement in function is seen within the first 6 months after arthroplasty. In addition, especially the functionality scores at baseline as well as their improvement showed an association with long-term patient survival.