Flexible competing risks regression modeling and goodness-of-fit

Poor Olfaction and Risk of Stroke in Older Adults: The Atherosclerosis Risk in Communities Study

BACKGROUND

Poor olfaction may be associated with adverse cerebrovascular events, but empirical evidence is limited. We aimed to investigate the association of olfaction with the risk of stroke in the Atherosclerosis Risk in Communities Study.

METHODS

We included 5799 older adults with no history of stroke at baseline from 2011 to 2013 (75.5±5.1 years, 59.0% female and 22.2% Black). Olfaction was assessed by the 12-item Sniffin' Sticks odor identification test and defined as poor (number correct ≤8), moderate (9-10), or good (11-12). Participants were followed from baseline to the date of the first stroke, death, last contact, or December 31, 2020, whichever occurred first. We used the discrete-time subdistribution hazard model to estimate the marginal cumulative incidence of stroke across olfactory statuses and adjusted risk ratios, accounting for covariates and competing risk of death.

RESULTS

After up to 9.6 years of follow-up, we identified 332 incident stroke events. The adjusted marginal cumulative incidence of stroke at 9.6-year follow-up was 5.3% (95% CI, 4.2%-6.3%), 5.9% (95% CI, 4.8%-7.1%), and 7.7% (95% CI, 6.5%-9.1%) for good, moderate, and poor olfaction, respectively. Compared with good olfaction, poor olfaction was significantly associated with higher stroke risk throughout follow-up, albeit the association modestly attenuated after 6 years. Specifically, the adjusted risk ratios were 2.14 (95% CI, 1.22-3.94) at year 2, 1.98 (95% CI, 1.43-3.02) at year 4, 1.91 (95% CI, 1.43-2.77) at year 6, 1.49 (95% CI, 1.17-2.00) at year 8, and 1.45 (95% CI, 1.16-1.95) at year 9.6. Results were robust in multiple subgroup and sensitivity analyses.

CONCLUSIONS

In older adults, poor olfaction assessed by a single olfaction test was associated with the higher risk of stroke in the next 10 years.

Estimands and Cumulative Incidence Function Regression in Clinical Trials: Some New Results on Interpretability and Robustness

Regression analyses based on transformations of cumulative incidence functions are often adopted when modeling and testing for treatment effects in clinical trial settings involving competing and semi-competing risks. Common frameworks include the Fine-Gray model and models based on direct binomial regression. Using large sample theory we derive the limiting values of treatment effect estimators based on such models when the data are generated according to multiplicative intensity-based models, and show that the estimand is sensitive to several process features. The rejection rates of hypothesis tests based on cumulative incidence function regression models are also examined for null hypotheses of different types, based on which a robustness property is established. In such settings supportive secondary analyses of treatment effects are essential to ensure a full understanding of the nature of treatment effects. An application to a palliative study of individuals with breast cancer metastatic to bone is provided for illustration.

Early, Intensive and Persistent Lipid-Lowering Therapy for Secondary Prevention of Acute Coronary Syndrome

AIM

Early and intensive low-density lipoprotein (LDL-C)-lowering therapy plays important roles in secondary prevention of acute coronary syndrome (ACS), but the treatment period for further clinical benefit remains undefined. This single-center, retrospective study explored LDL-C trajectory after ACS and its associations with subsequent cardiovascular events (CVE).

METHODS

In 831 patients with ACS, we evaluated LDL-C reduction during the first 2 months post-ACS as an index of early intervention and the area over the curve for LDL-C using 70 mg/dl as the threshold in the next 6 months (AOC-70) as a persistent intensity index. Patients were followed for a median of 3.0 (1.1-5.2) years for CVE, defined as the composite of cardiovascular death, non-fatal myocardial infarction, angina pectoris requiring revascularization, cerebral infarction, and coronary bypass grafting.

RESULTS

LDL-C decreased from baseline to 2 months post-ACS (107±38 mg/dl to 78±25 mg/dl, p＜0.001) through high-intensity statin prescription (91.8%), while achieving rates of LDL-C ＜70 mg/dl at 2 months remained only 40.2% with no significant changes thereafter. During the follow-up period, CVE occurred in 200 patients. LDL-C reduction during the first 2 months and AOC-70 in the next 6 months were both associated with subsequent CVE risk (sub-HR [hazard ratio] [95% confidence interval]: 1.48 [1.16-1.89] and 1.22 [1.05-1.44]). Furthermore, early intervention followed by persistently intensive LDL-C-lowering therapy resulted in further CVE risk reduction.

CONCLUSIONS

The present study observed that achieving early and intensive LDL-C reduction within the first two months after ACS and maintaining it for the next six months suppressed subsequent CVE risk, suggesting the importance of early, intensive, and persistent LDL-C-lowering therapy in the secondary prevention of ACS.

Alcohol use disorder and risk of specific methods of suicide death in a national cohort

INTRODUCTION

Alcohol use disorder (AUD) is among the strongest correlates of suicide death, but it is unclear whether AUD status is differentially associated with risk of suicide by particular methods.

METHODS

The authors used competing risks models to evaluate the association between AUD status and risk of suicide by poisoning, suffocation, drowning, firearm, instruments, jumping, or other means in a large Swedish cohort born 1932-1995 (total N = 6,581,827; 48.8% female). Data were derived from Swedish national registers, including the Cause of Death Register and a range of medical registers.

RESULTS

After adjusting for sociodemographic factors and familial liability to suicidal behavior, AUD was positively associated with risk of suicide for each method evaluated (cumulative incidence differences: 0.006-1.040 for females, 0.046-0.680 for males), except the association with firearm suicide in females. AUD was most strongly associated with risk of suicide by poisoning. Sex differences in the effects of AUD and family liability were observed for some, but not all, methods. Furthermore, high familial liability for suicidal behavior exacerbated AUD's impact on risk for suicide by poisoning (both sexes) and suffocation and jumping (males only), while the inverse interaction was observed for firearm suicide (males only).

CONCLUSIONS

AUD increases risk of suicide by all methods examined and is particularly potent with respect to risk of suicide by poisoning. Differences in risk related to sex and familial liability to suicidal behavior underscore AUD's nuanced role in suicide risk. Future research should investigate targeted means restriction effectiveness among persons with AUD.

Longer-Term All-Cause and Cardiovascular Mortality With Intensive Blood Pressure Control: A Secondary Analysis of a Randomized Clinical Trial

Importance

The Systolic Blood Pressure Intervention Trial (SPRINT) showed that intensive blood pressure control reduced cardiovascular morbidity and mortality. However, the legacy effect of intensive treatment is unknown.

Objective

To evaluate the long-term effects of randomization to intensive treatment with the incidence of cardiovascular and all-cause mortality approximately 4.5 years after the trial ended.

Design, Setting, and Participants

In this secondary analysis of a multicenter randomized clinical trial, randomization began on November 8, 2010, the trial intervention ended on August 20, 2015, and trial close-out visits occurred through July 2016. Patients 50 years and older with hypertension and increased cardiovascular risk but without diabetes or history of stroke were included from 102 clinic sites in the US and Puerto Rico. Analyses were conducted between October 2021 and February 2022.

Interventions

Randomization to systolic blood pressure (SBP) goal of less than 120 mm Hg (intensive treatment group; n = 4678) vs less than 140 mm Hg (standard treatment group; n = 4683).

Main Outcomes and Measures

Extended observational follow-up for mortality via the US National Death Index from 2016 through 2020. In a subset of 2944 trial participants, outpatient SBP from electronic health records during and after the trial were examined.

Results

Among 9361 randomized participants, the mean (SD) age was 67.9 (9.4) years, and 3332 (35.6%) were women. Over a median (IQR) intervention period of 3.3 (2.9-3.9) years, intensive treatment was beneficial for both cardiovascular mortality (hazard ratio [HR], 0.66; 95% CI, 0.49-0.89) and all-cause mortality (HR, 0.83; 95% CI, 0.68-1.01). However, at the median (IQR) total follow-up of 8.8 (8.3-9.3) years, there was no longer evidence of benefit for cardiovascular mortality (HR, 1.02; 95% CI, 0.84-1.24) or all-cause mortality (HR, 1.08; 95% CI, 0.94-1.23). In a subgroup of participants, the estimated mean outpatient SBP among participants randomized to intensive treatment increased from 132.8 mm Hg (95% CI, 132.0-133.7) at 5 years to 140.4 mm Hg (95% CI, 137.8-143.0) at 10 years following randomization.

Conclusions and Relevance

The beneficial effect of intensive treatment on cardiovascular and all-cause mortality did not persist after the trial. Given increasing outpatient SBP levels in participants randomized to intensive treatment following the trial, these results highlight the importance of consistent long-term management of hypertension.

Trial Registration

ClinicalTrials.gov Identifier: NCT01206062.

Kidney Transplantation After Rescue Allocation-the Eurotransplant Experience: A Retrospective Multicenter Outcome Analysis

BACKGROUND

At Eurotransplant (ET), kidneys are transferred to "rescue allocation" (RA), whenever the standard allocation (SA) algorithms Eurotransplant Kidney Allocation System (ETKAS) and Eurotransplant Senior Program (ESP) fail. We analyzed the outcome of RA.

METHODS

Retrospective patient clinical and demographic characteristics association analyses were performed with graft outcomes for 2422 recipients of a deceased donor renal transplantation (DDRT) after RA versus 25 481 after SA from 71 centers across all ET countries from 2006 to 2018.

RESULTS

Numbers of DDRTs after RA increased over the time, especially in Germany. RA played a minor role in ESP versus ETKAS (2.7% versus 10.4%). RA recipients and donors were older compared with SA recipients and donors, cold ischemia times were longer, waiting times were shorter, and the incidence of primary nonfunction was comparable. Among ETKAS recipients, HLA matching was more favorable in SA (mean 3.7 versus 2.5). In multivariate modeling, the incidence of graft loss in ETKAS recipients was reduced in RA compared with SA (subdistribution hazard ratio, 0.80; 95% confidence interval [0.70-0.91], P < 0.001), whereas other outcomes (mortality, death with functioning graft (DwFG)) were not significantly different. None of the 3 outcomes were significantly different when comparing RA with SA within the ESP program.

CONCLUSIONS

Facing increased waiting times and mortality on dialysis due to donor shortage, this study reveals encouragingly positive DDRT outcomes following RA. This supports the extension of RA to more patients and as an alternative tool to enable transplantation in patients in countries with prohibitively long waiting times or at risk of deterioration.

Competing risks regression models with covariates-adjusted censoring weight under the generalized case-cohort design

A generalized case-cohort design has been used when measuring exposures is expensive and events are not rare in the full cohort. This design collects expensive exposure information from a (stratified) randomly selected subset from the full cohort, called the subcohort, and a fraction of cases outside the subcohort. For the full cohort study with competing risks, He et al. (Scand J Stat 43:103-122, 2016) studied the non-stratified proportional subdistribution hazards model with covariate-dependent censoring to directly evaluate covariate effects on the cumulative incidence function. In this paper, we propose a stratified proportional subdistribution hazards model with covariate-adjusted censoring weights for competing risks data under the generalized case-cohort design. We consider a general class of weight functions to account for the generalized case-cohort design. Then, we derive the optimal weight function which minimizes the asymptotic variance of parameter estimates within the general class of weight functions. The proposed estimator is shown to be consistent and asymptotically normally distributed. The simulation studies show (i) the proposed estimator with covariate-adjusted weight is unbiased when the censoring distribution depends on covariates; and (ii) the proposed estimator with the optimal weight function gains parameter estimation efficiency. We apply the proposed method to stem cell transplantation and diabetes data sets.

Semiparametric copula-based regression modeling of semi-competing risks data

Semi-competing risks data often arise in medical studies where the terminal event (e.g., death) censors the non-terminal event (e.g., cancer recurrence), but the non-terminal event does not prevent the subsequent occurrence of the terminal event. This article considers regression modeling of semi-competing risks data to assess the covariate effects on the respective non-terminal and terminal event times. We propose a copula-based framework for semi-competing risks regression with time-varying coefficients, where the dependence between the non-terminal and terminal event times is characterized by a copula and the time-varying covariate effects are imposed on two marginal regression models. We develop a two-stage inferential procedure for estimating the association parameter in the copula model and time-varying regression parameters. We evaluate the finite sample performance of the proposed method through simulation studies and illustrate the method through an application to Surveillance, Epidemiology, and End Results-Medicare data for elderly women diagnosed with early-stage breast cancer and initially treated with breast-conserving surgery.

Conducting Real-world Evidence Studies on the Clinical Outcomes of Diabetes Treatments

Real-world evidence (RWE), the understanding of treatment effectiveness in clinical practice generated from longitudinal patient-level data from the routine operation of the healthcare system, is thought to complement evidence on the efficacy of medications from randomized controlled trials (RCTs). RWE studies follow a structured approach. (1) A design layer decides on the study design, which is driven by the study question and refined by a medically informed target population, patient-informed outcomes, and biologically informed effect windows. Imagining the randomized trial we would ideally perform before designing an RWE study in its likeness reduces bias; the new-user active comparator cohort design has proven useful in many RWE studies of diabetes treatments. (2) A measurement layer transforms the longitudinal patient-level data stream into variables that identify the study population, the pre-exposure patient characteristics, the treatment, and the treatment-emergent outcomes. Working with secondary data increases the measurement complexity compared to primary data collection that we find in most RCTs. (3) An analysis layer focuses on the causal treatment effect estimation. Propensity score analyses have gained in popularity to minimize confounding in healthcare database analyses. Well-understood investigator errors, like immortal time bias, adjustment for causal intermediates, or reverse causation, should be avoided. To increase reproducibility of RWE findings, studies require full implementation transparency. This article integrates state-of-the-art knowledge on how to conduct and review RWE studies on diabetes treatments to maximize study validity and ultimately increased confidence in RWE-based decision making.

A Novel Score for Predicting Alzheimer's Disease Risk from Late Life Psychopathological and Health Risk Factors

With the increasing size of the aging population, dementia risk reduction has become a main public health concern. Dementia risk models or indices may help to identify individuals in the community at high risk to develop dementia. We have aimed to develop a novel dementia risk index focused on the late-life (65 years or more) population, that addresses risk factors for Alzheimer's disease (AD) easily identifiable at primary care settings. These risk factors include some shown to be associated with the risk of AD but not featured in existing indices, such as hearing loss and anxiety. Our index is also the first to account for the competing risk of death. The Zaragoza Dementia and Depression Project (ZARADEMP) Alzheimer Dementia Risk Score predicts an individual´s risk of developing AD within 5 years. The probability of late onset AD significantly increases in those with risk scores between 21 and 28 and, furthermore, is almost 4-fold higher for those with risk scores of 29 or higher. Our index may provide a practical instrument to identify subjects at high risk of AD and to design preventive strategies targeting the contributing risk factors.

Paracetamol use and lowered risk of acute kidney injury in patients with rhabdomyolysis

BACKGROUND

Mortality with rhabdomyolysis-associated acute kidney injury can be as high as 80%. Experimental data from mouse models of rhabdomyolysis showed that paracetamol reduces the expected increase in serum creatinine level. We aimed to assess the association between paracetamol use and the need for starting renal replacement therapy (RRT).

METHODS

We conducted a propensity score-matched cohort study in Orléans Hospital, France (a 1136-bed, public, university-affiliated and teaching hospital). All patients with serum creatine phosphokinase (CK) level > 5000 IU/L between January 1st, 2008 and December 31st, 2017 were included. A propensity score was calculated for each included patient by using multivariable logistic regression and all available baseline characteristics. The main outcome was the incidence of RRT initiation from day 1 to day 28 in the propensity score-matched cohort between patients exposed and unexposed to paracetamol.

RESULTS

Over the study period, 1065 patients with at least one CK level measurement > 5000 IU/L were included; 40 (3.8%) had at least one RRT session. Among the 343 matched pairs, 10 (2.9%) exposed and 24 (7.0%) unexposed patients underwent RRT before day 28 (P = 0.021). Primary time-to-event analysis showed that exposure to paracetamol was significantly associated with reduced absolute risk of RRT: absolute risk difference = - 3.18% (95% CI - 5.23 to - 1.20, P = 0.001). All secondary analyses showed a significantly reduced absolute risk of RRT in patients exposed to paracetamol.

CONCLUSION

Our study showed a significant association between paracetamol exposure and reduced incidence of RRT among patients with rhabdomyolysis.

Ventilation practices in burn patients—an international prospective observational cohort study

Background

It is unknown whether lung-protective ventilation is applied in burn patients and whether they benefit from it. This study aimed to determine ventilation practices in burn intensive care units (ICUs) and investigate the association between lung-protective ventilation and the number of ventilator-free days and alive at day 28 (VFD-28).

Methods

This is an international prospective observational cohort study including adult burn patients requiring mechanical ventilation. Low tidal volume (VT) was defined as VT ≤ 8 mL/kg predicted body weight (PBW). Levels of positive end-expiratory pressure (PEEP) and maximum airway pressures were collected. The association between VT and VFD-28 was analyzed using a competing risk model. Ventilation settings were presented for all patients, focusing on the first day of ventilation. We also compared ventilation settings between patients with and without inhalation trauma.

Results

A total of 160 patients from 28 ICUs in 16 countries were included. Low VT was used in 74% of patients, median VT size was 7.3 [interquartile range (IQR) 6.2–8.3] mL/kg PBW and did not differ between patients with and without inhalation trauma (p = 0.58). Median VFD-28 was 17 (IQR 0–26), without a difference between ventilation with low or high VT (p = 0.98). All patients were ventilated with PEEP levels ≥5 cmH2O; 80% of patients had maximum airway pressures &lt;30 cmH2O.

Conclusion

In this international cohort study we found that lung-protective ventilation is used in the majority of burn patients, irrespective of the presence of inhalation trauma. Use of low VT was not associated with a reduction in VFD-28.

Trial registration

Clinicaltrials.gov NCT02312869. Date of registration: 9 December 2014.

Semiparametric mixture cure model analysis with competing risks data: Application to vascular access thrombosis data

The article is motivated by a nephrology study in Taiwan, which enrolled hemodialysis patients who suffered from vascular access thrombosis. After treatment, some patients were cured of thrombosis, while some may experience recurrence of either type (acute or nonacute) of vascular access thrombosis. Our major interest is to estimate the cumulative incidence probability of time to the first recurrence of acute thrombosis after therapy. Since the occurrence of one type of vascular access thrombosis precludes occurrence of the other type, patients are subject to competing risks. To account for the presence of competing risks and cured patients, we develop a mixture model approach to the regression analysis of competing-risks data with a cure fraction. We make inference about the effects of factors on both the cure rate and cumulative incidence function (CIF) for a failure of interest, which are separately specified in the logistic regression model and semiparametric regression model with time-varying and time-invariant effects. Based on two-stage method, we develop novel estimation equations using the inverse probability censoring weight techniques. The asymptotic properties of the estimators are rigorously studied and the plug-in variance estimators can be obtained for constructing interval estimators. We also propose a lack-of-fit test for assessing the adequacy of the proposed model and several tests for time-varying effects. The simulation studies and vascular access thrombosis data analysis are conducted to illustrate the proposed method.

Cultural Engagement Is a Risk-Reducing Factor for Frailty Incidence and Progression

Objectives

Given that frailty is a multifaceted health condition of increasing importance to policy-makers and care providers, it is relevant to consider whether multimodal interventions could provide combined psychophysiological support. As studies have demonstrated the beneficial effects of cultural engagement (including visiting museums/theatre/cinema) for many of the components of frailty, this study sought to explore whether community cultural engagement is associated both with a reduced risk of becoming frail and a slower trajectory of frailty progression in older adults.

Methods

We used data from the English Longitudinal Study of Ageing to measure frequency of cultural engagement and both incident frailty and frailty progression over the following 10 years in 4,575 adults.

Results

Our analyses used competing risks regression models and multilevel growth curve models adjusting for socioeconomic, health behaviors, social confounders, and subthreshold symptoms of frailty. There was a dose–response relationship between increasing frequency of cultural engagement and both incidence and progression of frailty (attendance every few months or more: incidence subhazard ratio = 0.79, 95% confidence interval [CI] = 0.63 to 0.996; trajectory coefficient = –0.0039, 95% CI = –0.0059 to –0.0019).

Discussion

Older adults who engaged in cultural activities every few months or more had a reduced risk of becoming frail and a slower progression of frailty over time. Findings are in line with current calls for multimodal, multifactor, community approaches to support health in older age.

Excess cumulative incidence estimation for matched cohort survival studies

We suggest a regression approach to estimate the excess cumulative incidence function (CIF) when matched data are available. In a competing risk setting, we define the excess risk as the difference between the CIF in the exposed group and the background CIF observed in the unexposed group. We show that the excess risk can be estimated through an extended binomial regression model that actively uses the matched structure of the data, avoiding further estimation of both the exposed and the unexposed CIFs. The method naturally deals with two time scales, age and time since exposure and simplifies how to deal with the left truncation on the age time-scale. The model makes it easy to predict individual excess risk scenarios and allows for a direct interpretation of the covariate effects on the cumulative incidence scale. After introducing the model and some theory to justify the approach, we show via simulations that our model works well in practice. We conclude by applying the excess risk model to data from the ALiCCS study to investigate the excess risk of late events in childhood cancer survivors.

Compatibility at amino acid position 98 of MICB reduces the incidence of graft-versus-host disease in conjunction with the CMV status

Graft-versus-host disease (GVHD) and cytomegalovirus (CMV)-related complications are leading causes of mortality after unrelated-donor hematopoietic cell transplantation (UD-HCT). The non-conventional MHC class I gene MICB, alike MICA, encodes a stress-induced polymorphic NKG2D ligand. However, unlike MICA, MICB interacts with the CMV-encoded UL16, which sequestrates MICB intracellularly, leading to immune evasion. Here, we retrospectively analyzed the impact of mismatches in MICB amino acid position 98 (MICB98), a key polymorphic residue involved in UL16 binding, in 943 UD-HCT pairs who were allele-matched at HLA-A, -B, -C, -DRB1, -DQB1 and MICA loci. HLA-DP typing was further available. MICB98 mismatches were significantly associated with an increased incidence of acute (grade II-IV: HR, 1.20; 95% CI, 1.15 to 1.24; P < 0.001; grade III-IV: HR, 2.28; 95% CI, 1.56 to 3.34; P < 0.001) and chronic GVHD (HR, 1.21; 95% CI, 1.10 to 1.33; P < 0.001). MICB98 matching significantly reduced the effect of CMV status on overall mortality from a hazard ratio of 1.77 to 1.16. MICB98 mismatches showed a GVHD-independent association with a higher incidence of CMV infection/reactivation (HR, 1.84; 95% CI, 1.34 to 2.51; P < 0.001). Hence selecting a MICB98-matched donor significantly reduces the GVHD incidence and lowers the impact of CMV status on overall survival.

Direct adjusted survival and cumulative incidence curves for observational studies

SERIES EDITORS' NOTE

Large randomized clinical trials testing the impact of subject-, disease- and transplant-related co-variates on outcomes amongst recipients of haematopoietic cell transplants are uncommon. For example, who is the best donor, which is the best pretransplant conditioning regimen or the best regimen to prevent or treat acute and/or chronic graft-versus-host disease. To answer these questions we often rely on analyses of data from large observational datasets such as those of the Center for International Blood and Marrow Transplant Research (CIBMTR) and the European Society for Blood and Marrow Transplantation (EBMT). Such analyses have proved extremely important in advancing the field. However, in contrast to randomized trials, we cannot be certain potentially important prognostic or predictive co-variates are balanced between cohorts selected for comparison from an observational dataset, a limitation which can lead to incorrect conclusions. In the typescript which follows the authours describe a method to adjust for known imbalances in co-variates and get a closer approximation of the truth. They give two examples, the impact of a new pretransplant conditioning regimen on disease-free survival (DFS) in subjects with Ewing sarcoma and the impact of donor-type on treatment-related mortality (TRM) and leukaemia relapse in subjects with acute leukaemia. Direct adjusted survival and cumulative incidence function (CIF) analyses are an important step forward. These analyses can be done using available statistical packages and we encourage readers to use them rather than reporting unadjusted analyses. Finally, we must emphasize direct adjustment can only be done for know prognostic or predictive co-variates, not unknown co-variates. Unknown co-variates will be balanced in randomized trials which is why we do them. So direct adjustment is an important step forward but not a perfect substitute for randomized trials. But any step forward is important. To quote Laozi: (A journey of a thousand miles begins with a single step).

On evaluating how well a biomarker can predict treatment response with survival data

One of the objectives of personalized medicine is to take treatment decisions based on a biomarker measurement. Therefore, it is often interesting to evaluate how well a biomarker can predict the response to a treatment. To do so, a popular methodology consists of using a regression model and testing for an interaction between treatment assignment and biomarker. However, the existence of an interaction is not sufficient for a biomarker to be predictive. It is only necessary. Hence, the use of the marker-by-treatment predictiveness curve has been recommended. In addition to evaluate how well a single continuous biomarker predicts treatment response, it can further help to define an optimal threshold. This curve displays the risk of a binary outcome as a function of the quantiles of the biomarker, for each treatment group. Methods that assume a binary outcome or rely on a proportional hazard model for a time-to-event outcome have been proposed to estimate this curve. In this work, we propose some extensions for censored data. They rely on a time-dependent logistic model, and we propose to estimate this model via inverse probability of censoring weighting. We present simulations results and three applications to prostate cancer, liver cirrhosis, and lung cancer data. They suggest that a large number of events need to be observed to define a threshold with sufficient accuracy for clinical usefulness. They also illustrate that when the treatment effect varies with the time horizon which defines the outcome, then the optimal threshold also depends on this time horizon.

Comparison of the marginal hazard model and the sub-distribution hazard model for competing risks under an assumed copula

For the analysis of competing risks data, three different types of hazard functions have been considered in the literature, namely the cause-specific hazard, the sub-distribution hazard, and the marginal hazard function. Accordingly, medical researchers can fit three different types of the Cox model to estimate the effect of covariates on each of the hazard function. While the relationship between the cause-specific hazard and the sub-distribution hazard has been extensively studied, the relationship to the marginal hazard function has not yet been analyzed due to the difficulties related to non-identifiability. In this paper, we adopt an assumed copula model to deal with the model identifiability issue, making it possible to establish a relationship between the sub-distribution hazard and the marginal hazard function. We then compare the two methods of fitting the Cox model to competing risks data. We also extend our comparative analysis to clustered competing risks data that are frequently used in medical studies. To facilitate the numerical comparison, we implement the computing algorithm for marginal Cox regression with clustered competing risks data in the R joint.Cox package and check its performance via simulations. For illustration, we analyze two survival datasets from lung cancer and bladder cancer patients.

Antibiotics with activity against intestinal anaerobes and the hazard of acquired colonization with ceftriaxone-resistant Gram-negative pathogens in ICU patients: a propensity score-based analysis

Background

Intestinal colonization resistance is mainly exerted by commensal anaerobes.

Objectives

To assess whether exposure to non-carbapenem antibiotics with activity against intestinal anaerobes (namely, piperacillin/tazobactam, amoxicillin/clavulanate and metronidazole) may promote the acquisition of gut colonization with ceftriaxone-resistant Gram-negative bacteria (CFR-GNB) in ICU patients.

Patients and methods

All patients with a first stay >3 days in a single surgical ICU over a 30 month period were retrospectively included. Rectal carriage of CFR-GNB (i.e. ESBL-producing Enterobacteriaceae, AmpC-hyperproducing Enterobacteriaceae, Pseudomonas aeruginosa, Stenotrophomonas maltophilia and CFR Acinetobacter baumannii) was routinely screened for at admission then weekly. The impact of anti-anaerobe antibiotics was investigated in propensity score (PS)-matched cohorts of patients exposed and not exposed to these drugs and through PS-based inverse probability of treatment weighting on the whole study cohort, treating in-ICU death or discharge as competing risks for CFR-GNB acquisition.

Results

Among the 352 included patients [median ICU stay 16 (9–30) days, in-ICU mortality 12.2%], 120 (34.1%) acquired one or more CFR-GNB, mostly AmpC-hyperproducing Enterobacteriaceae (17.6%) and P. aeruginosa (14.8%). Exposure to anti-anaerobe antibiotics was the main predictor of CFR-GNB acquisition in both the PS-matched cohorts [adjusted HR (aHR) 3.92, 95% CI 1.12–13.7, P = 0.03] and the whole study cohort (aHR 4.30, 95% CI 1.46–12.63, P = 0.01). Exposure to other antimicrobials—especially ceftriaxone and imipenem/meropenem—exerted no independent impact on the likelihood of CFR-GNB acquisition.

Conclusions

Exposure to non-carbapenem antibiotics with activity against intestinal anaerobes may predispose to CFR-GNB acquisition in ICU patients. Restricting the use of these drugs appears to be an antibiotic stewardship opportunity.

Clinically significant anxiety as a risk factor for dementia in the elderly community

OBJECTIVE

To evaluate whether clinically significant anxiety is an independent risk factor for dementia, taking into account both depression among potentially confounding factors and the competing risk of death.

METHOD

During the Zaragoza Dementia and Depression (ZARADEMP) study, a random sample of community dwellers aged 55 years or older was assessed (n = 4803), and a two-wave, 4.5-year follow-up was completed. Geriatric Mental State (GMS)-AGECAT criteria were used to diagnose anxiety and DSM-IV criteria were applied to diagnose incident dementia. The multivariate Fine and Gray regression model was implemented to calculate dementia risk.

RESULTS

Compared with non-cases (GMS-AGECAT criteria), the incidence rate of dementia was significantly higher in subcases of anxiety, and particularly significant in the cases of anxiety (incidence rate ratio (IRR): 2.77; P = 0.010). Cases of anxiety, but not subcases, at baseline were significantly associated with dementia risk (adjusted subdistribution hazard ratio (SHR): 2.7; P = 0.019).

CONCLUSION

Clinically significant anxiety is associated with an almost threefold increase in the risk of dementia in the population, even when controlling for depression and considering mortality in the competing risks model.

Subdistribution hazard models for competing risks in discrete time

A popular modeling approach for competing risks analysis in longitudinal studies is the proportional subdistribution hazards model by Fine and Gray (1999. A proportional hazards model for the subdistribution of a competing risk. Journal of the American Statistical Association94, 496–509). This model is widely used for the analysis of continuous event times in clinical and epidemiological studies. However, it does not apply when event times are measured on a discrete time scale, which is a likely scenario when events occur between pairs of consecutive points in time (e.g., between two follow-up visits of an epidemiological study) and when the exact lengths of the continuous time spans are not known. To adapt the Fine and Gray approach to this situation, we propose a technique for modeling subdistribution hazards in discrete time. Our method, which results in consistent and asymptotically normal estimators of the model parameters, is based on a weighted ML estimation scheme for binary regression. We illustrate the modeling approach by an analysis of nosocomial pneumonia in patients treated in hospitals.

Frailty is an independent predictor of incident dementia: Evidence from the English Longitudinal Study of Ageing

The aim of this study was to determine whether frailty in older adults is associated with the risk of subsequent dementia. A total of 8,722 older adults from the English Longitudinal Study of Ageing were followed-up every two years until they reported a diagnosis of dementia, died, or were right censored. Frailty was defined using a frailty index comprised of 47 health deficits. To test if cognitive function influences the relationship between frailty and incident dementia, the analyses were repeated according to lower or upper three quartiles of baseline cognitive function. Competing risks regression and Cox proportional hazard models were used to evaluate whether the degree of baseline frailty was associated with incident dementia. Compared with non-frail participants, pre-frail (HR: 1.51 95%CI [1.12-2.02]) and frail participants (HR: 1.73 95%CI [1.22-2.43]) had a higher risk of developing dementia, after adjustment for covariates. The association between frailty and incident dementia was significant for adults in the upper three quartiles of global cognitive function (HR: 3.48 95%CI [1.98-6.12]), but not for adults who were in the lowest quartile of cognitive function (HR: 1.13 95%CI [0.74-1.71]). Frailty should be monitored alongside cognitive functioning when assessing risk factors for dementia in older adults.

Instrumental variable with competing risk model

In this paper, we discuss causal inference on the efficacy of a treatment or medication on a time-to-event outcome with competing risks. Although the treatment group can be randomized, there can be confoundings between the compliance and the outcome. Unmeasured confoundings may exist even after adjustment for measured covariates. Instrumental variable methods are commonly used to yield consistent estimations of causal parameters in the presence of unmeasured confoundings. On the basis of a semiparametric additive hazard model for the subdistribution hazard, we propose an instrumental variable estimator to yield consistent estimation of efficacy in the presence of unmeasured confoundings for competing risk settings. We derived the asymptotic properties for the proposed estimator. The estimator is shown to be well performed under finite sample size according to simulation results. We applied our method to a real transplant data example and showed that the unmeasured confoundings lead to significant bias in the estimation of the effect (about 50% attenuated). Copyright © 2017 John Wiley & Sons, Ltd.

Staging cognitive impairment and incidence of dementia

AIMS

In a background of interest in staging models in psychiatry, we tested the validity of a simple staging model of cognitive impairment to predict incident dementia.

METHOD

A large community sample of adults aged ≥55 years (N = 4803) was assessed in the baseline of a longitudinal, four-wave epidemiological enquiry. A two-phase assessment was implemented in each wave, and the instruments used included the Mini-Mental Status Examination (MMSE); the History and Aetiology Schedule and the Geriatric Mental State-AGECAT. For the standardised degree of cognitive impairment Perneczky et al's MMSE criteria were applied. A panel of psychiatrists diagnosed cases of dementia according to DSM-IV criteria, and cases and sub-cases of dementia were excluded for the follow-up waves. Competing risk regression models, adjusted by potential confounders, were used to test the hypothesised association between MMSE levels and dementia risk.

RESULTS

Out of the 4057 participants followed up, 607 (14.9%) were classified as 'normal' (no cognitive impairment), 2672 (65.8%) as 'questionable' cognitive impairment, 732 (18.0%) had 'mild' cognitive impairment, 38 (0.9%) had 'moderate' cognitive impairment and eight (0.2%) had 'severe' impairment. Cognitive impairment was associated with risk of dementia, the risk increasing in parallel with the level of impairment (hazard ratio: 2.72, 4.78 and 8.38 in the 'questionable', 'mild' and 'moderate' level of cognitive impairment, respectively).

CONCLUSIONS

The documented gradient of increased risk of dementia associated with the severity level of cognitive impairment supports the validity of the simple staging model based on the MMSE assessment.

Matching for the nonconventional MHC-I MICA gene significantly reduces the incidence of acute and chronic GVHD

Graft-versus-host disease (GVHD) is among the most challenging complications in unrelated donor hematopoietic cell transplantation (HCT). The highly polymorphic MHC class I chain-related gene A, MICA, encodes a stress-induced glycoprotein expressed primarily on epithelia. MICA interacts with the invariant activating receptor NKG2D, expressed by cytotoxic lymphocytes, and is located in the MHC, next to HLA-B Hence, MICA has the requisite attributes of a bona fide transplantation antigen. Using high-resolution sequence-based genotyping of MICA, we retrospectively analyzed the clinical effect of MICA mismatches in a multicenter cohort of 922 unrelated donor HLA-A, HLA-B, HLA-C, HLA-DRB1, and HLA-DQB1 10/10 allele-matched HCT pairs. Among the 922 pairs, 113 (12.3%) were mismatched in MICA MICA mismatches were significantly associated with an increased incidence of grade III-IV acute GVHD (hazard ratio [HR], 1.83; 95% confidence interval [CI], 1.50-2.23; P < .001), chronic GVHD (HR, 1.50; 95% CI, 1.45-1.55; P < .001), and nonelapse mortality (HR, 1.35; 95% CI, 1.24-1.46; P < .001). The increased risk for GVHD was mirrored by a lower risk for relapse (HR, 0.50; 95% CI, 0.43-0.59; P < .001), indicating a possible graft-versus-leukemia effect. In conclusion, when possible, selecting a MICA-matched donor significantly influences key clinical outcomes of HCT in which a marked reduction of GVHD is paramount. The tight linkage disequilibrium between MICA and HLA-B renders identifying a MICA-matched donor readily feasible in clinical practice.

Quantifying intrinsic and extrinsic control of single-cell fates in cancer and stem/progenitor cell pedigrees with competing risks analysis

The molecular control of cell fate and behaviour is a central theme in biology. Inherent heterogeneity within cell populations requires that control of cell fate is studied at the single-cell level. Time-lapse imaging and single-cell tracking are powerful technologies for acquiring cell lifetime data, allowing quantification of how cell-intrinsic and extrinsic factors control single-cell fates over time. However, cell lifetime data contain complex features. Competing cell fates, censoring, and the possible inter-dependence of competing fates, currently present challenges to modelling cell lifetime data. Thus far such features are largely ignored, resulting in loss of data and introducing a source of bias. Here we show that competing risks and concordance statistics, previously applied to clinical data and the study of genetic influences on life events in twins, respectively, can be used to quantify intrinsic and extrinsic control of single-cell fates. Using these statistics we demonstrate that 1) breast cancer cell fate after chemotherapy is dependent on p53 genotype; 2) granulocyte macrophage progenitors and their differentiated progeny have concordant fates; and 3) cytokines promote self-renewal of cardiac mesenchymal stem cells by symmetric divisions. Therefore, competing risks and concordance statistics provide a robust and unbiased approach for evaluating hypotheses at the single-cell level.

Penalized estimation for competing risks regression with applications to high-dimensional covariates

High-dimensional regression has become an increasingly important topic for many research fields. For example, biomedical research generates an increasing amount of data to characterize patients' bio-profiles (e.g. from a genomic high-throughput assay). The increasing complexity in the characterization of patients' bio-profiles is added to the complexity related to the prolonged follow-up of patients with the registration of the occurrence of possible adverse events. This information may offer useful insight into disease dynamics and in identifying subset of patients with worse prognosis and better response to the therapy. Although in the last years the number of contributions for coping with high and ultra-high-dimensional data in standard survival analysis have increased (Witten and Tibshirani, 2010. Survival analysis with high-dimensional covariates. Statistical Methods in Medical Research19(1), 29–51), the research regarding competing risks is less developed (Binder and others, 2009. Boosting for high-dimensional time-to-event data with competing risks. Bioinformatics25(7), 890–896). The aim of this work is to consider how to do penalized regression in the presence of competing events. The direct binomial regression model of Scheike and others (2008. Predicting cumulative incidence probability by direct binomial regression. Biometrika95(1), 205–220) is reformulated in a penalized framework to possibly fit a sparse regression model. The developed approach is easily implementable using existing high-performance software to do penalized regression. Results from simulation studies are presented together with an application to genomic data when the endpoint is progression-free survival. An R function is provided to perform regularized competing risks regression according to the binomial model in the package timereg (Scheike and Martinussen, 2006. Dynamic Regression models for survival data. New York: Springer), available through CRAN.

Efficient Estimation of Semiparametric Transformation Models for the Cumulative Incidence of Competing Risks

The cumulative incidence is the probability of failure from the cause of interest over a certain time period in the presence of other risks. A semiparametric regression model proposed by Fine and Gray (1999) has become the method of choice for formulating the effects of covariates on the cumulative incidence. Its estimation, however, requires modeling of the censoring distribution and is not statistically efficient. In this paper, we present a broad class of semiparametric transformation models which extends the Fine and Gray model, and we allow for unknown causes of failure. We derive the nonparametric maximum likelihood estimators (NPMLEs) and develop simple and fast numerical algorithms using the profile likelihood. We establish the consistency, asymptotic normality, and semiparametric efficiency of the NPMLEs. In addition, we construct graphical and numerical procedures to evaluate and select models. Finally, we demonstrate the advantages of the proposed methods over the existing ones through extensive simulation studies and an application to a major study on bone marrow transplantation.

Joint Inference for Competing Risks Survival Data

This article develops joint inferential methods for the cause-specific hazard function and the cumulative incidence function of a specific type of failure to assess the effects of a variable on the time to the type of failure of interest in the presence of competing risks. Joint inference for the two functions are needed in practice because (i) they describe different characteristics of a given type of failure, (ii) they do not uniquely determine each other, and (iii) the effects of a variable on the two functions can be different and one often does not know which effects are to be expected. We study both the group comparison problem and the regression problem. We also discuss joint inference for other related functions. Our simulation shows that our joint tests can be considerably more powerful than the Bonferroni method, which has important practical implications to the analysis and design of clinical studies with competing risks data. We illustrate our method using a Hodgkin disease data and a lymphoma data. Supplementary materials for this article are available online.

A competing risk analysis for hospital length of stay in patients with burns

IMPORTANCE

Current outcome predictors for illness and injury are measured at a single time point-admission. However, patient prognosis often changes during hospitalization, limiting the usefulness of those predictions. Accurate depiction of the dynamic interaction between competing events during hospitalization may enable real-time outcome assessment.

OBJECTIVE

To determine how the effects of burn outcome predictors (ie, age, total body surface area burn, and inhalation injury) and the outcomes of interest (ie, mortality and length of stay) vary as a function of time throughout hospitalization.

DESIGN, SETTING, AND PARTICIPANTS

In this retrospective study, we used the American Burn Association's National Burn Repository, containing outcomes and patient and injury characteristics, to identify 95 579 patients admitted with an acute burn injury to 80 tertiary American Burn Association burn centers from 2000 through 2009. We applied competing risk statistical methods to analyze patient outcomes.

MAIN OUTCOMES AND MEASURES

We estimated the cause-specific hazard rates for death and discharge to assess how the instantaneous risk of these events changed across time. We further evaluated the varying effects of patient age, total body surface area burn, and inhalation injury on the probability of discharge and death across time.

RESULTS

Maximum length of stay among patients who died was 270 days and 731 days among those discharged. Total body surface area, age, and inhalation injury had significant effects on the subdistribution hazard for discharge (P < .001); these effects varied across time (P < .002). Burn size (coefficient -0.046) determined early outcomes, while age (coefficient -0.034) determined outcomes later in the hospitalization. Inhalation injury (coefficient -0.622) played a variable role in survival and hospital length of stay.

CONCLUSIONS AND RELEVANCE

Real-time measurement of dynamic interrelationships among burn outcome predictors using competing risk analysis demonstrated that the key factors influencing outcomes differed throughout hospitalization. Further application of this analytic technique to other injury or illness types may improve assessment of outcomes.

Checking Fine and Gray subdistribution hazards model with cumulative sums of residuals

Recently, Fine and Gray (J Am Stat Assoc 94:496-509, 1999) proposed a semi-parametric proportional regression model for the subdistribution hazard function which has been used extensively for analyzing competing risks data. However, failure of model adequacy could lead to severe bias in parameter estimation, and only a limited contribution has been made to check the model assumptions. In this paper, we present a class of analytical methods and graphical approaches for checking the assumptions of Fine and Gray's model. The proposed goodness-of-fit test procedures are based on the cumulative sums of residuals, which validate the model in three aspects: (1) proportionality of hazard ratio, (2) the linear functional form and (3) the link function. For each assumption testing, we provide a p-values and a visualized plot against the null hypothesis using a simulation-based approach. We also consider an omnibus test for overall evaluation against any model misspecification. The proposed tests perform well in simulation studies and are illustrated with two real data examples.

Depression and incident Alzheimer disease: the impact of disease severity

OBJECTIVES

To test the hypothesis that clinically significant depression (particularly severe depression) increases the risk of Alzheimer's disease (AD).

METHODS

A longitudinal, three-wave epidemiologic study was implemented in a sample of individuals aged 55 years and older (n = 4,803) followed up at 2.5 years and 4.5 years. This was a population-based cohort drawn from the Zaragoza Dementia and Depression (ZARADEMP) Project, in Zaragoza, Spain. Participants included individuals cognitively intact at baseline (n = 3,864). The main outcome measures were depression as assessed by using the diagnostic interview Geriatric Mental State- Automated Geriatric Examination for Computer Assisted Taxonomy package; and AD diagnosed by a panel of research psychiatrists according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, criteria. The Fine and Gray multivariate regression model was used in the analysis, accounting for mortality.

RESULTS

At baseline, clinically significant depression was diagnosed in 452 participants (11.7%); of these, 16.4% had severe depression. Seventy incident cases of AD were found at follow-up. Compared with nondepressed individuals, the incidence rate of AD was significantly higher in the severely depressed subjects (incidence rate ratio: 3.59 [95% confidence interval: 1.30-9.94]). A consistent, significant association was observed between severe depression at baseline and incident AD in the multivariate model (hazard ratio: 4.30 [95% CI: 1.39-13.33]). Untreated depression was associated with incident AD in the unadjusted model; however, in the final model, this association was attenuated and nonsignificant.

CONCLUSIONS

Severe depression increases the risk of AD, even after controlling for the competing risk of death.

Semicompeting risks in aging research: methods, issues and needs

A semicompeting risks problem involves two-types of events: a nonterminal and a terminal event (death). Typically, the nonterminal event is the focus of the study, but the terminal event can preclude the occurrence of the nonterminal event. Semicompeting risks are ubiquitous in studies of aging. Examples of semicompeting risk dyads include: dementia and death, frailty syndrome and death, disability and death, and nursing home placement and death. Semicompeting risk models can be divided into two broad classes: models based only on observables quantities (class [Formula: see text]) and those based on potential (latent) failure times (class [Formula: see text]). The classical illness-death model belongs to class [Formula: see text]. This model is a special case of the multistate models, which has been an active area of methodology development. During the past decade and a half, there has also been a flurry of methodological activity on semicompeting risks based on latent failure times ([Formula: see text] models). These advances notwithstanding, the semicompeting risks methodology has not penetrated biomedical research, in general, and gerontological research, in particular. Some possible reasons for this lack of uptake are: the methods are relatively new and sophisticated, conceptual problems associated with potential failure time models are difficult to overcome, paucity of expository articles aimed at educating practitioners, and non-availability of readily usable software. The main goals of this review article are: (i) to describe the major types of semicompeting risks problems arising in aging research, (ii) to provide a brief survey of the semicompeting risks methods, (iii) to suggest appropriate methods for addressing the problems in aging research, (iv) to highlight areas where more work is needed, and (v) to suggest ways to facilitate the uptake of the semicompeting risks methodology by the broader biomedical research community.

Weighted comparison of two cumulative incidence functions with R-CIFsmry package

In this paper we propose a class of flexible weight functions for use in comparison of two cumulative incidence functions. The proposed weights allow the users to focus their comparison on an early or a late time period post treatment or to treat all time points with equal emphasis. These weight functions can be used to compare two cumulative incidence functions via their risk difference, their relative risk, or their odds ratio. The proposed method has been implemented in the R-CIFsmry package which is readily available for download and is easy to use as illustrated in the example.

Comparative benefit from small tumour size and adjuvant chemotherapy: clues for explaining breast cancer mortality decline

Background

Breast cancer mortality steadily declined from the 1990s and this has been attributed to early detection and/or to improvements in therapy. Which of those two has had the greater impact is a subject of contention.

Methods

A database of 386 patients, enrolled in a randomized clinical trial on the effect of adjuvant chemotherapy (CMF), was analysed. The probabilities of recurrence and death were estimated by the Fine and Gray’s model and by the Cox model. Time dependent covariate and interaction effects were investigated by additive models. Absolute risk reductions (ARR) related to adjuvant treatment or to tumour size [diameter ≤ 2 cm (T1) or >2 cm (T2/T3)] were estimated.

Results

CMF-related reduction in recurrence emerges early, reaches a maximum level at 3 years and persists at a constant level thereafter. Tumour-size-related recurrence reduction, after a maximum at 3 years, displays a progressive regular reduction approaching zero. Patients with any tumour size, when given CMF, exhibit mortality reduction that displays an early regular increase and continues to a persistent plateau. In contrast, tumour-size-related mortality reduction reaches a maximum at 5–7 years and then regularly drops to very low values for patients of both trial arms.

Conclusions

Findings reveal that there is a different time-dependent benefit from chemotherapy and from smaller tumour size at diagnosis. The benefit from adjuvant chemotherapy is long-lasting for patients with any tumour size while the early benefit of diagnosing smaller tumours substantially decreases afterwards. Treatment improvements have probably had greater impact on the mortality reduction than mammography screening.

Generalized random sign and alert delay models for imperfect maintenance

This paper considers the modelling of the process of Corrective and condition-based Preventive Maintenance, for complex repairable systems. In order to take into account the dependency between both types of maintenance and the possibility of imperfect maintenance, Generalized Competing Risks models have been introduced in "Doyen and Gaudoin (J Appl Probab 43:825-839, 2006)". In this paper, we study two classes of these models, the Generalized Random Sign and Generalized Alert Delay models. A Generalized Competing Risks model can be built as a generalization of a particular Usual Competing Risks model, either by using a virtual age framework or not. The models properties are studied and their parameterizations are discussed. Finally, simulation results and an application to real data are presented.

The British antibiotic and silver-impregnated catheters for ventriculoperitoneal shunts multi-centre randomised controlled trial (the BASICS trial): study protocol

Background

Insertion of a ventriculoperitoneal shunt (VPS) for the treatment of hydrocephalus is one of the most common neurosurgical procedures in the UK, but failures caused by infection occur in approximately 8% of primary cases. VPS infection is associated with considerable morbidity and mortality and its management results in substantial cost to the health service. Antibiotic-impregnated (rifampicin and clindamycin) and silver-impregnated VPS have been developed to reduce infection rates. Whilst there is some evidence showing that such devices may lead to a reduction in VPS infection, there are no randomised controlled trials (RCTs) to support their routine use.

Methods/design

Overall, 1,200 patients will be recruited from 17 regional neurosurgical units in the UK and Ireland. Patients of any age undergoing insertion of their first VPS are eligible. Patients with previous indwelling VPS, active and on-going cerebrospinal fluid (CSF) or peritoneal infection, multiloculated hydrocephalus requiring multiple VPS or neuroendoscopy, and ventriculoatrial or ventriculopleural shunt planned will be excluded. Patients will be randomised 1:1:1 to either standard silicone (comparator), antibiotic-impregnated, or silver-impregnated VPS. The primary outcome measure is time to VPS infection. Secondary outcome measures include time to VPS failure of any cause, reason for VPS failure (infection, mechanical failure, or patient failure), types of bacterial VPS infection (organism type and antibiotic resistance), and incremental cost per VPS failure averted.

Discussion

The British antibiotic and silver-impregnated catheters for ventriculoperitoneal shunts multi-centre randomised controlled trial (the BASICS trial) is the first multi-centre RCT designed to determine whether antibiotic or silver-impregnated VPS reduce early shunt infection compared to standard silicone VPS. The results of this study will be used to inform current neurosurgical practice and may potentially benefit patients undergoing shunt surgery in the future.

Trial registration

International Standard Randomised Controlled Trial Number: ISRCTN49474281.

Comparison of competing risks models based on cumulative incidence function in analyzing time to cardiovascular diseases

BACKGROUND

Competing risks arise when the subject is exposed to more than one cause of failure. Data consists of the time that the subject failed and an indicator of which risk caused the subject to fail.

METHODS

With three approaches consisting of Fine and Gray, binomial, and pseudo-value, all of which are directly based on cumulative incidence function, cardiovascular disease data of the Isfahan Cohort Study were analyzed. Validity of proportionality assumption for these approaches is the basis for selecting appropriate models. Such as for the Fine and Gray model, establishing proportionality assumption is necessary. In the binomial approach, a parametric, non-parametric, or semi-parametric model was offered according to validity of assumption. However, pseudo-value approaches do not need to establish proportionality.

RESULTS

Following fitting the models to data, slight differences in parameters and variances estimates were seen among models. This showed that semi-parametric multiplicative model and the two models based on pseudo-value approach could be used for fitting this kind of data.

CONCLUSION

We would recommend considering the use of competing risk models instead of normal survival methods when subjects are exposed to more than one cause of failure.

Goodness-of-fit test for proportional subdistribution hazards model

This paper concerns using modified weighted Schoenfeld residuals to test the proportionality of subdistribution hazards for the Fine-Gray model, similar to the tests proposed by Grambsch and Therneau for independently censored data. We develop a score test for the time-varying coefficients based on the modified Schoenfeld residuals derived assuming a certain form of non-proportionality. The methods perform well in simulations and a real data analysis of breast cancer data, where the treatment effect exhibits non-proportional hazards.

Competing risks with missing covariates: effect of haplotypematch on hematopoietic cell transplant patients

In this paper we consider a problem from hematopoietic cell transplant (HCT) studies where there is interest on assessing the effect of haplotype match for donor and patient on the cumulative incidence function for a right censored competing risks data. For the HCT study, donor's and patient's genotype are fully observed and matched but their haplotypes are missing. In this paper we describe how to deal with missing covariates of each individual for competing risks data. We suggest a procedure for estimating the cumulative incidence functions for a flexible class of regression models when there are missing data, and establish the large sample properties. Small sample properties are investigated using simulations in a setting that mimics the motivating haplotype matching problem. The proposed approach is then applied to the HCT study.

Applying competing risks regression models: an overview

In many clinical research applications the time to occurrence of one event of interest, that may be obscured by another--so called competing--event, is investigated. Specific interventions can only have an effect on the endpoint they address or research questions might focus on risk factors for a certain outcome. Different approaches for the analysis of time-to-event data in the presence of competing risks were introduced in the last decades including some new methodologies, which are not yet frequently used in the analysis of competing risks data. Cause-specific hazard regression, subdistribution hazard regression, mixture models, vertical modelling and the analysis of time-to-event data based on pseudo-observations are described in this article and are applied to a dataset of a cohort study intended to establish risk stratification for cardiac death after myocardial infarction. Data analysts are encouraged to use the appropriate methods for their specific research questions by comparing different regression approaches in the competing risks setting regarding assumptions, methodology and interpretation of the results. Notes on application of the mentioned methods using the statistical software R are presented and extensions to the presented standard methods proposed in statistical literature are mentioned.

Absolute risk regression for competing risks: interpretation, link functions, and prediction

In survival analysis with competing risks, the transformation model allows different functions between the outcome and explanatory variables. However, the model's prediction accuracy and the interpretation of parameters may be sensitive to the choice of link function. We review the practical implications of different link functions for regression of the absolute risk (or cumulative incidence) of an event. Specifically, we consider models in which the regression coefficients β have the following interpretation: The probability of dying from cause D during the next t years changes with a factor exp(β) for a one unit change of the corresponding predictor variable, given fixed values for the other predictor variables. The models have a direct interpretation for the predictive ability of the risk factors. We propose some tools to justify the models in comparison with traditional approaches that combine a series of cause-specific Cox regression models or use the Fine-Gray model. We illustrate the methods with the use of bone marrow transplant data.

Durability of pericardial versus porcine bioprosthetic heart valves

OBJECTIVES

To compare the probability, and modes, of explantation for Carpentier-Edwards pericardial versus porcine valves.

METHODS

Our porcine series began in 1974 and our pericardial series in 1991, with annual prospective follow-up. We used the Kaplan-Meier method and Cox regression for estimation and analysis of patient mortality, and the cumulative incidence function and competing risks regression for estimation and analysis of valve durability.

RESULTS

Through the end of 2010, we had implanted 506 porcine and 2449 pericardial aortic valves and 181 porcine and 163 pericardial mitral valves. The corresponding total and maximum follow-up years were 3471 and 24, 11,517 and 18, 864 and 22, and 645 and 9. The corresponding probabilities (cumulative incidence function) of any valve explant were 7%, 8%, 22%, and 8%, and of explant for structural valve deterioration were 4%, 5%, 16%, and 5% at 15 years for the first 3 series and at 8 years for the fourth (pericardial mitral valve) series. Using competing risks regression for structural valve deterioration explant, with age, gender, valve size, and concomitant coronary bypass surgery as covariates, a slight (subhazard ratio, 0.79), but nonsignificant, protective effect was found for the pericardial valve in the aortic position and a greater (subhazard ratio, 0.31) and almost significant (P = .08) protective effect of the pericardial valve in the mitral position. Leaflet tear was responsible for 61% of the structural valve deterioration explants in the porcine series and 46% in the pericardial series.

CONCLUSIONS

Using competing risks regression, the pericardial valve had a subhazard ratio for structural valve deterioration explant of less than 1 in both positions, approaching statistical significance in the mitral position. The mode of structural valve deterioration was predominantly leaflet tear for porcine valves and fibrosis/calcification for pericardial valves.

Patterns of dental service utilization following nontraumatic dental condition visits to the emergency department in Wisconsin Medicaid

OBJECTIVES

To examine patterns of dental service utilization for adult Medicaid enrollees in Wisconsin following nontraumatic dental condition (NTDC) visits to the emergency department (ED).

METHODS

This is a retrospective, observational study of claims for NTDC visits to the ED and dental service encounters from the Wisconsin Medicaid Evaluation and Decision Support database (2001-2009). We used competing risk models to predict probabilities of returning to the ED versus obtaining follow-up care from a dentist.

RESULTS

We observed a 43 percent increase in the rate of NTDC visits to the ED, with most of this increase occurring from 2001 to 2005. Within 30 days of an NTDC visit to the ED, ∼29.6 percent of enrollees will first visit a dentist office, while ∼9.9 percent will return to the ED. Young to middle-aged adults (18 to <50 years) and enrollees living in counties with a lower supply of dental providers were more likely to return to the ED following a NTDC visit. Among the enrollees that first visited a dental office following an ED visit, 37.6 percent had an extraction performed at this visit.

CONCLUSIONS

Almost one in five adult Medicaid enrollees will subsequently return to the ED following a previous NTDC visit. The provision of definitive care for these individuals appears to primarily consist of extractions.

Incidence and lifetime risk of dementia and Alzheimer's disease in a Southern European population

OBJECTIVE

To calculate both the incidence rates and the lifetime risk (LTR) of dementia and Alzheimer's disease (AD).

METHODS

A two-phase case-finding procedure was implemented in a cohort of 4057 cognitively intact individuals 55+ years of age living in Zaragoza, Spain, and followed-up at 2.5 and 4.5 years. Age- and sex-specific incidence rates were calculated. A mortality-adjusted, multivariate model was used to document LTRs.

RESULTS

The incidence rate of dementia continued to rise after the age of 90 years, but was slightly lower than in North and West European studies. Only a tendency for an increased LTR with age was observed. Thus, LTR was 19.7% for a 65-year-old woman and 20.4% at the age of 85 years, the corresponding figures for AD being 16.7% and 17.6%. The LTR of AD was higher in women and was about twice as high among illiterate individuals when compared with individuals with higher educational levels.

CONCLUSIONS

The incidence rate of dementia in this Southern European city was slightly lower than in previous studies in North-West Europe. LTR of dementia and AD seems to be slightly increased with age. The association of illiteracy with higher LTR of AD is intriguing.