Estimation of treatment effects in early-phase randomized clinical trials involving external control data.

There are good reasons to perform a randomized controlled trial (RCT) even in early phases of clinical development. However, the low sample sizes in those settings lead to high variability of the treatment effect estimate. The variability could be reduced by adding external control data if available. For the common setting of suitable subject-level control group data only available from one external (clinical trial or real-world) data source, we evaluate different analysis options for estimating the treatment effect via hazard ratios. The impact of the external control data is usually guided by the level of similarity with the current RCT data. Such level of similarity can be determined via outcome and/or baseline covariate data comparisons. We provide an overview over existing methods, propose a novel option for a combined assessment of outcome and baseline data, and compare a selected set of approaches in a simulation study under varying assumptions regarding observable and unobservable confounder distributions using a time-to-event model. Our various simulation scenarios also reflect the differences between external clinical trial and real-world data. Data combinations via simple outcome-based borrowing or simple propensity score weighting with baseline covariate data are not recommended. Analysis options which conflate outcome and baseline covariate data perform best in our simulation study.

Dosing Characteristics of Recombinant Human Luteinizing Hormone or Human Menopausal Gonadotrophin-Derived LH Activity in Patients Undergoing Ovarian Stimulation: A German Fertility Database Study.

OBJECTIVES: The aim of the study was to evaluate dosing of recombinant human luteinizing hormone (r-hLH) or human menopausal gonadotrophin (hMG)-derived medications with LH activity in ovarian stimulation (OS) cycles for in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI). DESIGN: A non-interventional study was performed to analyse data from the German RecDate database (January 2007-December 2011). PARTICIPANTS/MATERIALS, SETTING, METHODS: Starting/total r-hLH/hMG dose, OS duration/cycle number, r-hLH/hMG initiation day (first day of administration), and population/cycle characteristics were assessed in women (≥18 years) undergoing OS for IVF/ICSI using r-hLH or hMG-derived medications (excluding corifollitropin alfa, clomiphene citrate, letrozole, mini/micro-dose human chorionic gonadotrophin, and urofollitropin alone). Data were summarized descriptively. RESULTS: 67,858 identified cycles utilized medications containing r-hLH (10,749), hMG (56,432), or both (677). Mean (standard deviation) OS duration with r-hLH and hMG was 10.1 (4.43) and 9.8 (6.16) days, respectively. Median (25th-75th percentile) r-hLH starting dose (75.0 [75.0-150.0] IU) was consistent across patients regardless of age, infertility diagnosis, or gonadotrophin-releasing hormone (GnRH) protocol. Median (25th-75th percentile) hMG-derived LH activity starting dose was 225.0 (150.0-300.0) IU, regardless of GnRH protocol, but was lower in women aged <35 years and those with ovulation disorders/polycystic ovary syndrome. Median (25th-75th percentile) total dose for r-hLH (750.0 [337.5-1,125.0] IU) and hMG-derived LH activity (1,575.0 [750.0-2,625.0] IU) varied according to patients' age, infertility diagnosis, cycle number, and r-hLH/hMG initiation day. GnRH antagonist use resulted in a numerically higher median total hMG-derived LH activity dose than GnRH agonist use. LIMITATIONS: The data used in this study were taken from electronic medical records relating to a specific timeframe (2007-2011) and therefore may not accurately reflect current clinical practice; however, it is likely that the differences between the two compounds would be maintained. Additionally, secondary data sources may suffer from uniformity and quality issues. CONCLUSIONS: The standard of care for OS cycles is described with respect to IVF/ICSI treatment including an LH component in Germany during the specified timeframe.

Real-world outcomes among patients with advanced or metastatic biliary tract cancers initiating second-line treatment.

BACKGROUND: Limited data are available regarding second-line (2 L) treatment for advanced or metastatic biliary tract cancers (BTC) in the US real-world setting. This study explores the rapidly evolving and growing treatment landscape in the 2 L setting for advanced or metastatic BTC with a large cohort of patients treated in a community oncology setting. METHODS: Adult patients with BTC initiating 2 L treatment after a platinum-containing first-line between 1/1/10- and 6/30/19 were identified from the US Oncology Network electronic healthcare record database and followed through 12/31/19. Baseline patient and treatment characteristics were analyzed descriptively, including overall response rate (ORR) in the real-world clinical setting. Kaplan-Meier methods were used to measure duration of response, progression-free survival (PFS), and overall survival (OS). RESULTS: The overall population (N = 160) included 74 patients (46.3%) with intrahepatic cholangiocarcinoma, 41 (25.6%) with extrahepatic cholangiocarcinoma, and 45 (28.1%) with gallbladder cancer. Thirty unique 2 L regimens were recorded for the study population, with folinic acid, fluorouracil and oxaliplatin (FOLFOX, 34.4%) and capecitabine monotherapy (20.0%) being the most common. ORR was 7.5% (95% CI, 3.9%-12.7%). From 2 L initiation, median PFS was 2.8 months (95% CI, 2.4-3.3 months), and median OS was 5.2 months (95% CI, 4.2-6.7 months). CONCLUSION: Results from this study provide real-world evidence that although patients treated in the community oncology setting receive a wide variety of 2 L treatments, the regimens are consistent with those recommended by guidelines. Although responses are observed with 2 L treatment, duration is brief and associated with poor OS in patients with advanced or metastatic disease.

Real-World Effectiveness of Beta-Blockers versus Other Antihypertensives in Reducing All-Cause Mortality and Cardiovascular Events.

Aim: The aim of the study was to compare the effectiveness of beta-blockers with other antihypertensive classes in reducing all-cause mortality, cardiovascular-related mortality and the risk of cerebrocardiovascular events. Methods: This noninterventional study was conducted within the UK Clinical Practice Research Datalink. Hypertensive patients who initiated antihypertensive monotherapy were allocated to one of five cohorts: beta-blockers; angiotensin-converting enzyme inhibitors (ACEi); angiotensin II receptor blockers (ARB); calcium channel blockers (CCB); and diuretics. Differences in outcomes were assessed using Cox proportional hazard models with competing risks. Results: A total of 44,404 patients were prescribed beta-blockers (75% atenolol), 132,545 ACEi, 12,018 ARB, 91,731 CCB, and 106,547 diuretics. At baseline, patients in the beta-blocker cohort presented more frequently with angina, arrhythmia, and atrial fibrillation. The risk of all-cause mortality was lower for those treated with ACEi, ARB, and CCB, and no difference was observed compared with diuretics (adjusted hazard ratio versus beta-blockers (98.75% CI), for ACEi 0.71 (0.61, 0.83), ARB 0.67 (0.51, 0.88), CCB 0.76 (0.66, 0.88), diuretics 1.06 (0.93, 1.22)). No differences were seen in the risk of cardiovascular mortality for patients treated with beta-blockers, ARB, CCB, and diuretics, while a lower risk in patients treated with ACEi was observed (ACEi 0.63 (0.43, 0.91), ARB 0.64 (0.32, 1.28), CCB 0.71 (0.49, 1.03), diuretics 0.97 (0.69, 1.37)). Conclusions: These data add to the limited pool of evidence from real-world studies exploring the effectiveness of beta-blockers versus other antihypertensive classes. Discrepancies to previously published studies might be partly explained by differences in the selected populations and in the follow-up time.

Cost-effectiveness analysis of recombinant human follicle-stimulating hormone alfa(r-hFSH) and urinary highly purified menopausal gonadotropin (hMG) based on data from a large German registry.

This was a retrospective real-world evidence analysis of the costs per live birth for reference recombinant human follicle-stimulating hormone alfa (r-hFSH-alfa) versus highly purified urinary human menopausal gonadotropin (hMG-HP), based on data from a German in vitro fertilization registry (RecDate). Pregnancy and live birth rates from the RecDate real-world evidence study over three complete assisted reproductive technology (ART) cycles using the same gonadotropin drug were used as clinical inputs. Costs related to ART treatment and to drugs were obtained from public sources. Treatment with r-hFSH-alfa resulted in higher adjusted cumulative live birth rates versus hMG-HP after one (25.3% vs. 22.3%), two (30.9% vs. 27.5%), and three (31.9% vs. 28.6%) ART cycles. Costs per live birth were lower with r-hFSH-alfa versus hMG-HP after one (€17,938 vs. €20,054), two (€18,251 vs. €20,437), and three (€18,473 vs. €20,680) ART cycles. r-hFSH-alfa was found to be a cost-effective strategy compared with hMG-HP over three cycles.

Effectiveness of bisoprolol versus other ß-blockers and other antihypertensive classes: a cohort study in the Clinical Practice Research Datalink.

Aim: To compare blood pressure (BP) and safety outcomes in patients with hypertension initiating bisoprolol, versus other ß-blockers, angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers, calcium channel blockers or diuretics. Materials & methods: New user cohort study. Patients initiating bisoprolol were matched with up to four patients, in each comparator cohort using propensity score. BP outcomes were compared using linear mixed models and safety outcomes using Cox proportional hazards. Results: Differences in average systolic and diastolic BP variation were ≤3 mmHg between bisoprolol versus the compared classes. No difference was observed in risk of diabetes, obesity or erectile dysfunction. An increased dyslipidemia risk was only observed versus diuretics (hazard ratio: 0.76; 98.75% CI: 0.58, 0.99). Conclusion: No differences in BP variation and safety outcomes.

An adaptive design for early clinical development including interim decision for single-arm trial with external controls or randomized trial.

In early clinical development, randomized controlled trials (RCT) or single-arm trials with external controls (SATwEC) are design options, which allow adjustment for confounding: RCT via design, SATwEC via analysis using propensity score methods. SATwEC requires less investment than RCT. However, if the confounder space substantially differs between the experimental and external control group, the SATwEC might lead to inappropriate decisions for further development. We develop an adaptive two-stage design (ATD) for early clinical development that reduces the risk of unreliable decision-making at the end of a SATwEC. In Stage I, subjects are solely assigned to the experimental group. If at the interim the propensity score distributions of internal and external data are comparable based on the preference score, the subjects in stage II will again be solely assigned to the experimental arm; if not, a randomized stage II will be conducted. In a simulation study guided by a motivating example, data is generated using a time-to-event model with observable and unobservable confounders. The confounder space is varied to investigate the impact on false go/stop probabilities as well as a loss function, which reflects the quality of treatment effect estimates and decision-making. The proposed ATD provides a compromise between optimizing quality (as expressed by false go/stop probabilities and the loss function) and investment (defined by sample size and trial duration).

Effectiveness, Healthcare Resource Utilization and Adherence to Subcutaneous Interferon Beta-1a According to Age in Patients With Multiple Sclerosis: A Cohort Study Using a US Claims Database.

Background: It is thought that older patients with multiple sclerosis (MS) may present with a different clinical disease phenotype, and therefore respond to subcutaneous interferon beta-1a (sc IFN ß-1a) differently to younger patients. However, few real-world data are available concerning the effectiveness of sc IFN ß-1a according to age. Using data from US claims databases, this cohort analysis aimed to determine the differences in relapse rates, healthcare utilization, treatment adherence, and discontinuation according to pre-defined age groups. Methods: Patient data were pooled from the IBM® MarketScan® Commercial Claims Database and Medicare Supplemental Database. Patients with a confirmed MS diagnosis who initiated treatment with sc IFN ß-1a between July 01, 2010 and December 31, 2015, along with at least 6 months continuous enrolment in a healthcare plan, were followed from first prescription (index date) until date of discontinuation, treatment switch, or end of observation period (1 year after index date). Results: Of the 5,340 patients included in the analysis, there was a high proportion of patients free from relapse across all age groups (range: 94.1-95.4%), with a numerical decrease in the number of MRI performed by age (mean: 0.25, 18-30 years; 0.20, 31-40 years; 0.16, 41-50 years; 0.14, ≥51 years). Adherence (≥80%) was seen to increase with age (77.6%, 18-30 years; 79.6%, 31-40 years; 81.3%, 41-50 years; 84.0%, ≥51 years), at the same time as a non-significant decrease in discontinuation (incidence rate: 79.91, 73.01, 71.75, 68.71%). Conclusion: The effectiveness of sc IFN ß-1a does not appear reduced as a consequence of age in this real-world setting. Older patients had lower discontinuation rates and reduced disease activity, reflected in lower relapse rates and fewer MRI scans compared with younger patients.

Comparative effectiveness of recombinant human follicle-stimulating hormone alfa (r-hFSH-alfa) versus highly purified urinary human menopausal gonadotropin (hMG HP) in assisted reproductive technology (ART) treatments: a non-interventional study in Germany.

BACKGROUND: This study compared the effectiveness of recombinant human follicle-stimulating hormone alfa (r-hFSH-alfa; GONAL-f®) with urinary highly purified human menopausal gonadotropin (hMG HP; Menogon HP®), during assisted reproductive technology (ART) treatments in Germany. METHODS: Data were collected from 71 German fertility centres between 01 January 2007 and 31 December 2012, for women undergoing a first stimulation cycle of ART treatment with r-hFSH-alfa or hMG HP. Primary outcomes were live birth, ongoing pregnancy and clinical pregnancy, based on cumulative data (fresh and frozen-thawed embryo transfers), analysed per patient (pP), per complete cycle (pCC) and per first complete cycle (pFC). Secondary outcomes were pregnancy loss (analysed per clinical pregnancy), cancelled cycles (analysed pCC), total drug usage per oocyte retrieved and time-to-live birth (TTLB; per calendar week and per cycle). RESULTS: Twenty-eight thousand six hundred forty-one women initiated a first treatment cycle (r-hFSH-alfa: 17,725 [61.9%]; hMG HP: 10,916 [38.1%]). After adjustment for confounding variables, treatment with r-hFSH-alfa versus hMG HP was associated with a significantly higher probability of live birth (hazard ratio [HR]-pP [95% confidence interval (CI)]: 1.10 [1.04, 1.16]; HR-pCC [95% CI]: 1.13 [1.08, 1.19]; relative risk [RR]-pFC [95% CI]: 1.09 [1.05, 1.15], ongoing pregnancy (HR-pP [95% CI]: 1.10 [1.04, 1.16]; HR-pCC [95% CI]: 1.13 [1.08, 1.19]; RR-pFC [95% CI]: 1.10 [1.05, 1.15]) and clinical pregnancy (HR-pP [95% CI]: 1.10 [1.05, 1.14]; HR-pCC [95% CI]: 1.14 [1.10, 1.19]; RR-pFC [95% CI]: 1.10 [1.06, 1.14]). Women treated with r-hFSH-alfa versus hMG HP had no statistically significant difference in pregnancy loss (HR [95% CI]: 1.07 [0.98, 1.17], were less likely to have a cycle cancellation (HR [95% CI]: 0.91 [0.84, 0.99]) and had no statistically significant difference in TTLB when measured in weeks (HR [95% CI]: 1.02 [0.97, 1.07]; p = 0.548); however, r-hFSH-alfa was associated with a significantly shorter TTLB when measured in cycles versus hMG HP (HR [95% CI]: 1.07 [1.02, 1.13]; p = 0.003). There was an average of 47% less drug used per oocyte retrieved with r-hFSH-alfa versus hMG HP. CONCLUSIONS: This large (> 28,000 women), real-world study demonstrated significantly higher rates of cumulative live birth, cumulative ongoing pregnancy and cumulative clinical pregnancy with r-hFSH-alfa versus hMG HP.

Cerebral Autoregulation and Neurovascular Coupling after Craniospinal Irradiation in Long-Term Survivors of Malignant Pediatric Brain Tumors of the Posterior Fossa.

INTRODUCTION: Long-term survivors of craniospinal irradiation have an increased risk for stroke which increases with radiation dose and follow-up time. Radiotherapy induces structural changes of the cerebral vasculature, affecting both, large, and small vessels. It is unknown how these structural changes affect functional mechanisms of cerebral blood flow regulation such as cerebral autoregulation and neurovascular coupling. METHODS: Using the transcranial Doppler, we compared dynamic cerebral autoregulation and neurovascular coupling of 12 patients after long-term survival of craniospinal irradiation due to a malignant pediatric brain tumor of the posterior fossa and 12 age- and sex-matched healthy patients. Mean arterial blood pressure and cerebral blood flow velocities in the middle and posterior cerebral artery were recorded at rest during normal breathing to assess cerebral autoregulation (transfer function parameters phase and gain, as well as the correlation coefficient indices Mx, Sx, and Dx), and during 10 cycles of a visual task to assess neurovascular coupling (parameters time delay, natural frequency, gain, attenuation, and rate time). RESULTS: Parameters of cerebral autoregulation showed a consistent trend toward reduced cerebral autoregulation in patients that did not reach statistical significance. Neurovascular coupling was not altered after craniospinal irradiation. CONCLUSION: In this pilot study, we demonstrated a trend toward reduced cerebral autoregulation, and no alteration of neurovascular coupling after irradiation in long-term survivors of malignant pediatric brain tumors of the posterior fossa.

Estimating cumulative incidence functions in competing risks data with dependent left-truncation.

Both delayed study entry (left-truncation) and competing risks are common phenomena in observational time-to-event studies. For example, in studies conducted by Teratology Information Services (TIS) on adverse drug reactions during pregnancy, the natural time scale is gestational age, but women enter the study after time origin and upon contact with the service. Competing risks are present, because an elective termination may be precluded by a spontaneous abortion. If left-truncation is entirely random, the Aalen-Johansen estimator is the canonical estimator of the cumulative incidence functions of the competing events. If the assumption of random left-truncation is in doubt, we propose a new semiparametric estimator of the cumulative incidence function. The dependence between entry time and time-to-event is modeled using a cause-specific Cox proportional hazards model and the marginal (unconditional) estimates are derived via inverse probability weighting arguments. We apply the new estimator to data about coumarin usage during pregnancy. Here, the concern is that the cause-specific hazard of experiencing an induced abortion may depend on the time when seeking advice by a TIS, which also is the time of left-truncation or study entry. While the aims of counseling by a TIS are to reduce the rate of elective terminations based on irrational overestimation of drug risks and to lead to better and safer medical treatment of maternal disease, it is conceivable that women considering an induced abortion are more likely to seek counseling. The new estimator is also evaluated in extensive simulation studies and found preferable compared to the Aalen-Johansen estimator in non-misspecified scenarios and to at least provide for a sensitivity analysis otherwise.

Bootstrapping complex time-to-event data without individual patient data, with a view toward time-dependent exposures.

We consider nonparametric and semiparametric resampling of multistate event histories by simulating multistate trajectories from an empirical multivariate hazard measure. One advantage of our approach is that it does not necessarily require individual patient data, but may be based on published information. This is also attractive for both study planning and simulating realistic real-world event history data in general. The concept extends to left-truncation and right-censoring mechanisms, nondegenerate initial distributions, and nonproportional as well as non-Markov settings. A special focus is on its connection to simulating survival data with time-dependent covariates. For the case of qualitative time-dependent exposures, we demonstrate that our proposal gives a more natural interpretation of how such data evolve over the course of time than many of the competing approaches. The multistate perspective avoids any latent failure time structure and sampling spaces impossible in real life, whereas its parsimony follows the principle of Occam's razor. We also suggest empirical simulation as a novel bootstrap procedure to assess estimation uncertainty in the absence of individual patient data. This is not possible for established procedures such as Efron's bootstrap. A simulation study investigating the effect of liver functionality on survival in patients with liver cirrhosis serves as a proof of concept. Example code is provided.

Effect of carbapenem resistance on outcomes of bloodstream infection caused by Enterobacteriaceae in low-income and middle-income countries (PANORAMA): a multinational prospective cohort study.

BACKGROUND: Low-income and middle-income countries (LMICs) are under-represented in reports on the burden of antimicrobial resistance. We aimed to quantify the clinical effect of carbapenem resistance on mortality and length of hospital stay among inpatients in LMICs with a bloodstream infection due to Enterobacteriaceae. METHODS: The PANORAMA study was a multinational prospective cohort study at tertiary hospitals in Bangladesh, Colombia, Egypt, Ghana, India, Lebanon, Nepal, Nigeria, Pakistan, and Vietnam, recruiting consecutively diagnosed patients with carbapenem-susceptible Enterobacteriaceae (CSE) and carbapenem-resistant Entero-bacteriaceae (CRE) bloodstream infections. We excluded patients who had previously been enrolled in the study and those not treated with curative intent at the time of bloodstream infection onset. There were no age restrictions. Central laboratories in India and the UK did confirmatory testing and molecular characterisation, including strain typing. We applied proportional subdistribution hazard models with inverse probability weighting to estimate the effect of carbapenem resistance on probability of discharge alive and in-hospital death, and multistate modelling for excess length of stay in hospital. All patients were included in the analysis. FINDINGS: Between Aug 1, 2014, and June 30, 2015, we recruited 297 patients from 16 sites in ten countries: 174 with CSE bloodstream infection and 123 with CRE bloodstream infection. Median age was 46 years (IQR 15-61). Crude mortality was 20% (35 of 174 patients) for patients with CSE bloodstream infection and 35% (43 of 123 patients) for patients with CRE bloodstream infection. Carbapenem resistance was associated with an increased length of hospital stay (3·7 days, 95% CI 0·3-6·9), increased probability of in-hospital mortality (adjusted subdistribution hazard ratio 1·75, 95% CI 1·04-2·94), and decreased probability of discharge alive (0·61, 0·45-0·83). Multilocus sequence typing showed various clades, with marginal overlap between strains in the CRE and CSE clades. INTERPRETATION: Carbapenem resistance is associated with increased length of hospital stay and mortality in patients with bloodstream infections in LMICs. These data will inform global estimates of the burden of antimicrobial resistance and reinforce the need for better strategies to prevent, diagnose, and treat CRE infections in LMICs. FUNDING: bioMérieux.

Impaired dynamic cerebral autoregulation in patients with cerebral amyloid angiopathy.

Cerebral amyloid angiopathy (CAA) might disturb the sensitive mechanism of cerebral pressure autoregulation. This study examines whether dynamic cerebral autoregulation (CA) is impaired in the posterior or anterior circulation of CAA patients. Fifteen patients with known CAA on magnetic resonance imaging (MRI) and 14 age-matched controls were examined with transcranial Doppler. Dynamic CA was assessed in the middle (MCA) and posterior cerebral artery (PCA) via transfer function phase and gain during respiratory-induced 0.1 Hz oscillations of arterial pressure. Within the patient group, 4 patients showed additional microbleeds in the basal ganglia on the MRI performed within the study (pure CAA vs mixed microbleeds). PCA phase was significantly lower in patients compared with controls (p = 0.018), particularly in patients with pure CAA (p = 0.0034). MCA values showed a similar but non-significant trend towards lower phase in patients with pure CAA. Poorer phase was associated with a higher number of microbleeds on MRI (MCA r = -0.57, p = 0.027; PCA r = -0.52, p = 0.098) and superficial cortical siderosis (PCA: p = 0.0025). In conclusion, dynamic cerebral autoregulation is impaired in patients with CAA. The degree of impairment is associated with the extent of cerebral microbleeds.

A note on pseudo-observations and left-truncation.

Pseudo-observations have been introduced as a way to perform regression analysis of a mean value parameter related to a right-censored time-to-event outcome, such as the survival probability or the restricted mean survival time. Since the introduction of the approach there have been several extensions from the original setting. However, the proper definition and performance of pseudo-observations under left-truncation has not yet been addressed. Here, we look at two types of pseudo-observations under right-censoring and left-truncation. We explored their performance in a simulation study and applied them to data on diabetes patients with left-truncation.

Stabilizing cumulative incidence estimation of pregnancy outcome with delayed entries.

A pregnancy may end up with (at least) three possible events: live birth, spontaneous abortion, or elective termination, yielding a competing risks issue when studying an association between a risk factor and a pregnancy outcome. Cumulative incidences (probabilities to end up with the different outcomes depending on gestational age) can be estimated via the Aalen-Johansen estimate. Another issue is that women are usually not entering such an observational study from the first day of pregnancy, resulting in delayed entries. As in traditional survival analysis, this can be solved by considering "at risk" at a given gestational age only for those women who entered the study before that age. However, the number of women at risk at an early gestational age might be extremely low, such that the estimates of cumulative incidence may increase exaggeratedly at that age because of a single event. One solution to reduce the problem has been recently proposed in the literature, which is to ignore simply those early events, creating a small mean bias but enhancing stability of estimates. In the present paper, we propose an alternative computationally simple approach to tackle this problem that consists to postpone to later gestational ages (rather than to ignore) those early events. The two approaches are compared with respect to bias, stability, and sensitivity on the smoothing parameter via simulations reproducing realistic pregnancy scenarios, and are illustrated with data from a study on the effects of statins on pregnancy outcomes. We also outline that all three approaches are asymptotically equivalent.

Estimation of adjusted expected excess length-of-stay associated with ventilation-acquired pneumonia in intensive care: A multistate approach accounting for time-dependent mechanical ventilation.

The expected excess length-of-stay is an established concept to assess the health and economic impact of nosocomial, that is, hospital-acquired infections such as ventilation-acquired pneumonia in intensive care. Estimation must account for the timing of infection as in a multistate perspective, because common retrospective comparisons yield inflated estimates due to time-dependent bias. Since occurrence of ventilation-acquired pneumonia is closely linked to ventilation status, we suggest a multistate model incorporating time-dependent mechanical ventilation as additional states. The appeal is that the expected excess length-of-stay decomposes into extra days spent under ventilation and not under ventilation. This is not only highly relevant from a patient's perspective regarding quality of life, but also from an economic point of view, because ventilation is a major cost driver. The challenge is that estimation involves complex functionals of the matrix of transition probabilities, which in turn are based on the transition hazards. To address heterogeneity between patients, which is a common phenomenon in observational hospital epidemiology, we apply pseudovalue regression to adjust the ventilation-specific quantities for baseline confounding. The performance of our proposal is assessed by simulation and the methods are illustrated on data provided by 12 French intensive care units. Preliminary results indicate that the expected excess length-of-stay associated with ventilation-acquired pneumonia is mainly triggered by extra days spent under mechanical ventilation, and that the excess is most pronounced for intensive care patients with fewer comorbidities at baseline. We also find that such a decomposition is challenging for early times. Example code is provided.

No long-term impairment of cerebral autoregulation after preeclampsia.

INTRODUCTION: Preeclampsia is a pregnancy-related hypertensive disorder with strongly impaired cerebral autoregulation in the acute stage. A history of preeclampsia is an independent cardiovascular and cerebrovascular risk factor. It is unclear whether impaired cerebral autoregulation persists after preeclampsia and thus contributes to the known increased cerebrovascular morbidity. METHODS: Using transcranial Doppler, we compared cerebral hemodynamics and dynamic cerebral autoregulation of 25 women with a history of severe preeclampsia and 25 healthy mothers, on average 2-3 years postpartum. Mean arterial blood pressure (MAP) and cerebral blood flow velocities (CBFV) in the middle and posterior cerebral artery were recorded at rest, dynamic cerebral autoregulation was assessed via transfer function phase and gain between oscillations of CBFV and MAP during regular breathing at 0.1 Hz. RESULTS: MAP and body mass index were higher in former preeclamptic women compared with healthy mothers (p-value <0.001 and 0.006, respectively). CBFV in the middle cerebral artery was slightly increased in former preeclamptic women compared with healthy mothers (p-value 0.004), intima-media thickness (IMT) of the common carotid artery was higher by trend (p-value 0.065). Dynamic cerebral autoregulation was not impaired in women with a history of preeclampsia, phase even tended to be higher than in healthy mothers. CONCLUSION: Dynamic cerebral autoregulation is not persistently impaired in women after severe preeclampsia. Long-term cerebrovascular changes rather result from a higher incidence of cerebrovascular risk factors in women with a history of preeclampsia.

Disturbance of a rare seabird by ship-based tourism in a marine protected area.

Managers of marine protected areas (MPAs) must often seek ways to allow for visitation while minimizing impacts to the resources they are intended to protect. Using shipboard observers, we quantified the "zone of disturbance" for Kittlitz's and marbled murrelets (Brachyramphus brevirostris and B. marmoratus) exposed to large cruise ships traveling through Glacier Bay National Park, one of the largest MPAs in North America. In the upper reaches of Glacier Bay, where Kittlitz's murrelets predominated, binary logistic regression models predicted that 61% of all murrelets within 850 m perpendicular distance of a cruise ship were disturbed (defined as flushing or diving), whereas in the lower reaches, where marbled murrelets predominated, this percentage increased to 72%. Using survival analysis, murrelets in both reaches were found to react at greater distances when ships approached indirectly, presumably because of the ship's larger profile, suggesting murrelets responded to visual rather than audio cues. No management-relevant covariates (e.g., ship velocity, route distance from shore) were found to be important predictors of disturbance, as distance from ship to murrelet accounted for > 90% of the explained variation in murrelet response. Utilizing previously published murrelet density estimates from Glacier Bay, and applying an average empirical disturbance probability (68%) out to 850 m from a cruise ship's typical route, we estimated that a minimum of 9.8-19.6% of all murrelets in Glacier Bay are disturbed per ship entry. Whether these disturbance levels are inconsistent with Park management objectives, which include conserving wildlife as well as providing opportunities for visitation, depends in large part on whether disturbance events caused by cruise ships have impacts on murrelet fitness, which remains uncertain.

The number of primary events per variable affects estimation of the subdistribution hazard competing risks model.

OBJECTIVES: To examine the effect of the number of events per variable (EPV) on the accuracy of estimated regression coefficients, standard errors, empirical coverage rates of estimated confidence intervals, and empirical estimates of statistical power when using the Fine-Gray subdistribution hazard regression model to assess the effect of covariates on the incidence of events that occur over time in the presence of competing risks. STUDY DESIGN AND SETTING: Monte Carlo simulations were used. We considered two different definitions of the number of EPV. One included events of any type that occurred (both primary events and competing events), whereas the other included only the number of primary events that occurred. RESULTS: The definition of EPV that included only the number of primary events was preferable to the alternative definition, as the number of competing events had minimal impact on estimation. In general, 40-50 EPV were necessary to ensure accurate estimation of regression coefficients and associated quantities. However, if all of the covariates are continuous or are binary with moderate prevalence, then 10 EPV are sufficient to ensure accurate estimation. CONCLUSION: Analysts must base the number of EPV on the number of primary events that occurred.