Dynamic predictions with time-dependent covariates in survival analysis using joint modeling and landmarking.

A key question in clinical practice is accurate prediction of patient prognosis. To this end, nowadays, physicians have at their disposal a variety of tests and biomarkers to aid them in optimizing medical care. These tests are often performed on a regular basis in order to closely follow the progression of the disease. In this setting, it is of interest to optimally utilize the recorded information and provide medically relevant summary measures, such as survival probabilities, which will aid in decision making. In this work, we present and compare two statistical techniques that provide dynamically updated estimates of survival probabilities, namely landmark analysis and joint models for longitudinal and time-to-event data. Special attention is given to the functional form linking the longitudinal and event time processes, and to measures of discrimination and calibration in the context of dynamic prediction.

Bayesian hierarchical modeling of longitudinal glaucomatous visual fields using a two-stage approach.

The Bayesian approach has become increasingly popular because it allows to fit quite complex models to data via Markov chain Monte Carlo sampling. However, it is also recognized nowadays that Markov chain Monte Carlo sampling can become computationally prohibitive when applied to a large data set. We encountered serious computational difficulties when fitting an hierarchical model to longitudinal glaucoma data of patients who participate in an ongoing Dutch study. To overcome this problem, we applied and extended a recently proposed two-stage approach to model these data. Glaucoma is one of the leading causes of blindness in the world. In order to detect deterioration at an early stage, a model for predicting visual fields (VFs) in time is needed. Hence, the true underlying VF progression can be determined, and treatment strategies can then be optimized to prevent further VF loss. Because we were unable to fit these data with the classical one-stage approach upon which the current popular Bayesian software is based, we made use of the two-stage Bayesian approach. The considered hierarchical longitudinal model involves estimating a large number of random effects and deals with censoring and high measurement variability. In addition, we extended the approach with tools for model evaluation. Copyright © 2017 John Wiley & Sons, Ltd.

Dealing with missing covariates in epidemiologic studies: a comparison between multiple imputation and a full Bayesian approach.

Incomplete data are generally a challenge to the analysis of most large studies. The current gold standard to account for missing data is multiple imputation, and more specifically multiple imputation with chained equations (MICE). Numerous studies have been conducted to illustrate the performance of MICE for missing covariate data. The results show that the method works well in various situations. However, less is known about its performance in more complex models, specifically when the outcome is multivariate as in longitudinal studies. In current practice, the multivariate nature of the longitudinal outcome is often neglected in the imputation procedure, or only the baseline outcome is used to impute missing covariates. In this work, we evaluate the performance of MICE using different strategies to include a longitudinal outcome into the imputation models and compare it with a fully Bayesian approach that jointly imputes missing values and estimates the parameters of the longitudinal model. Results from simulation and a real data example show that MICE requires the analyst to correctly specify which components of the longitudinal process need to be included in the imputation models in order to obtain unbiased results. The full Bayesian approach, on the other hand, does not require the analyst to explicitly specify how the longitudinal outcome enters the imputation models. It performed well under different scenarios. Copyright © 2016 John Wiley & Sons, Ltd.

Global Visit Effects in Point-Wise Longitudinal Modeling of Glaucomatous Visual Fields.

PURPOSE: One of the difficulties in modeling visual field (VF) data is the sometimes large and correlated measurement errors in the point-wise sensitivity estimates. As these errors affect all locations of the same VF, we propose to model them as global visit effects (GVE). We evaluate this model and show the effect it has on progression estimation and prediction. METHODS: Visual field series (24-2 Full Threshold; 15 biannual VFs per patient) of 125 patients with primary glaucoma were included in the analysis. The contribution of the GVE was evaluated by comparing the fitting and predictive ability of a conventional model, which does not contain GVE, to such a model that incorporates the GVE. Moreover, the GVE's effect on the estimated slopes was evaluated by determining the absolute difference between the slopes of the models. Finally, the magnitude of the GVE was compared with that of other measurement errors. RESULTS: The GVE model showed a significant improvement in both the model fit and predictive ability over the conventional model, especially when the number of VFs in a series is limited. The average absolute difference in slopes between the models was 0.13 dB/y. Lastly, the magnitude of the GVE was more than three times larger than the measureable factors combined. CONCLUSIONS: By incorporating the GVE in the longitudinal modeling of VF data, better estimates may be obtained of the rate of progression as well as of predicted future sensitivities.

Reduced trigeminovascular cyclicity in patients with menstrually related migraine.

OBJECTIVE: A case-control study to investigate the effect of the menstrual cycle on trigeminal nerve-induced vasodilation in healthy women and patients with menstrually related migraine (MRM). METHODS: Using a laser-Doppler imager, we compared the vasodilator effects of capsaicin application and electrical stimulation (ES) on the forehead skin, a trigeminal nerve-innervated dermatome, in premenopausal patients with MRM (n = 22), healthy controls (n = 20), and postmenopausal women without migraine (n = 22). Blood samples were collected for female sex hormone measurements. RESULTS: Dermal blood flow (DBF) responses to capsaicin were higher in controls during days 1-2 than during days 19-21 of their menstruation cycle (mean Emax ± SEM: 203 ± 28 AU vs 156 ± 27 AU [p = 0.031] for 0.06 mg/mL capsaicin and 497 ± 25 AU vs 456 ± 24 AU [p = 0.009] for 6.0 mg/mL capsaicin). In contrast, patients with MRM demonstrated DBF responses without significant cycle-dependent variability (days 1-2 vs days 19-21: Emax 148 ± 20 AU vs 154 ± 20 AU [p = 0.788] for 0.06 mg/mL capsaicin and 470 ± 17 AU vs 465 ± 20 AU [p = 0.679] for 6.0 mg/mL capsaicin). DBF responses to ES were not different between either patients with MRM or controls, at either occasion. Estradiol levels on days 19-21 of the menstrual cycle were higher in healthy controls (mean ± SEM: 75 ± 8 pg/mL) than in patients with MRM (52 ± 4 pg/mL, p = 0.014). In postmenopausal women, DBF responses to capsaicin and ES, as well as estradiol levels at both visits, were all significantly reduced compared to patients with MRM and controls (in all cases, p < 0.05). CONCLUSIONS: Our study provides evidence for a reduced menstrual cyclicity of both estradiol levels and the trigeminovascular vasodilator system in patients with MRM.

Criminal victimisation in people with severe mental illness: a multi-site prevalence and incidence survey in the Netherlands.

BACKGROUND: Although crime victimisation is as prevalent in psychiatric patients as crime perpetration (and possibly more so), few European figures for it are available. We therefore assessed its one-year prevalence and incident rates in Dutch severely mentally ill outpatients, and compared the results with victimisation rates in the general population. METHOD: This multisite epidemiological survey included a random sample of 956 adult severely mentally ill outpatients. Data on victimisation were obtained using the victimisation scale of the Dutch Crime and Victimisation Survey, which assesses crime victimisation over the preceding 12 months. Comparison data were derived from the nationwide survey on safety and victimisation in the Netherlands. Prevalence and incident rates were weighted for sex, age, ethnicity and socioeconomic status, and compared with a general population sample matched by region (N = 38,227). RESULTS: In the past year, almost half of the severely mentally ill outpatients (47%) had been victim of a crime. After control for demographic differences, prevalence rates of overall and specific victimisation measures were significantly higher in severely mentally ill outpatients than in the general population. The relative rates were especially high for personal crimes such as violent threats (RR = 2.12, 95% CI: 1.72-2.61), physical assaults (RR = 4.85, 95% CI: 3.69-6.39) and sexual harassment and assaults (RR = 3.94, 95% CI: 3.05-5.09). In concordance, severely mentally ill outpatients reported almost 14 times more personal crime incidents than persons from the general population (IRR = 13.68, 95% CI: 12.85-14.56). CONCLUSION: Crime victimisation is a serious problem in Dutch severely mentally ill outpatients. Mental-healthcare institutions and clinicians should become aware of their patients' victimisation risk, and should implement structural measures to detect and prevent (re-)victimisation.

Optimizing structure-function relationship by maximizing correspondence between glaucomatous visual fields and mathematical retinal nerve fiber models.

PURPOSE: To introduce a method to optimize structural retinal nerve fiber layer (RNFL) models based on glaucomatous visual field data and to show how such an optimized model can be used to reduce noise in visual fields while probably preserving clinically important features. METHODS: Correlation coefficients between age-adjusted deviation values of pairs of visual field test locations were calculated from 103 visual fields of eyes with moderate glaucomatous damage. Distances between those test locations were defined for various parameters of a mathematical RNFL model. Then, the correspondence between the structural and functional data was defined by the spread, or variance, of the correlation coefficients for all distances. The model parameters that minimized this spread constituted the optimized model. To reduce noise in visual fields, the optimized model was used to smooth visual field data according to the RNFL's structure. The resulting fields were compared with visual fields that were smoothed based on the regular testing grid. RESULTS: The optimal parameters for the RNFL model reduced the variance of the correlation coefficients by 78% and were well within the range of parameters previously determined from fundus photographs. Smoothing the visual fields based on the optimized RNFL model strongly reduced noise while keeping important features. CONCLUSIONS: Mathematic RNFL models can be optimized based on visual field data, resulting in a strong structure-function relationship. Taking the RNFL's shape, as defined by such an optimized model, into account when smoothing visual fields results in better noise reduction while preserving important details.

A systematic review and meta-analyses show that carbapenem use and medical devices are the leading risk factors for carbapenem-resistant Pseudomonas aeruginosa.

A systematic review and meta-analyses were performed to identify the risk factors associated with carbapenem-resistant Pseudomonas aeruginosa and to identify sources and reservoirs for the pathogen. A systematic search of PubMed and Embase databases from 1 January 1987 until 27 January 2012 identified 1,662 articles, 53 of which were included in a systematic review and 38 in a random-effects meta-analysis study. The use of carbapenem, use of fluoroquinolones, use of vancomycin, use of other antibiotics, having medical devices, intensive care unit (ICU) admission, having underlying diseases, patient characteristics, and length of hospital stay were significant risk factors in multivariate analyses. The meta-analyses showed that carbapenem use (odds ratio [OR] = 7.09; 95% confidence interval [CI] = 5.43 to 9.25) and medical devices (OR = 5.11; 95% CI = 3.55 to 7.37) generated the highest pooled estimates. Cumulative meta-analyses showed that the pooled estimate of carbapenem use was stable and that the pooled estimate of the risk factor "having medical devices" increased with time. We conclude that our results highlight the importance of antibiotic stewardship and the thoughtful use of medical devices in helping prevent outbreaks of carbapenem-resistant P. aeruginosa.

Robust and censored modeling and prediction of progression in glaucomatous visual fields.

PURPOSE: Classic regression is based on certain assumptions that conflict with visual field (VF) data. We investigate and evaluate different regression models and their assumptions in order to determine point-wise VF progression in glaucoma and to better predict future field loss for personalised clinical glaucoma management. METHODS: Standard automated visual fields of 130 patients with primary glaucoma with a minimum of 6 years of follow-up were included. Sensitivity estimates at each VF location were regressed on time with classical linear and exponential regression models, as well as different variants of these models that take into account censoring and allow for robust fits. These models were compared for the best fit and for their predictive ability. The prediction was evaluated at six measurements (approximately 3 years) ahead using varying numbers of measurements. RESULTS: For fitting the data, the classical uncensored linear regression model had the lowest root mean square error and 95th percentile of the absolute errors. These errors were reduced in all models when increasing the number of measurements used for the prediction of future measurements, with the classical uncensored linear regression model having the lowest values for these errors irrespective of how many measurements were included. CONCLUSIONS: All models performed similarly. Despite violation of its assumptions, the classical uncensored linear regression model appeared to provide the best fit for our data. In addition, this model appeared to perform the best when predicting future VFs. However, more advanced regression models exploring any temporal-spatial relationships of glaucomatous progression are needed to reduce prediction errors to clinically meaningful levels.

Neutrophil gelatinase-associated lipocalin at ICU admission predicts for acute kidney injury in adult patients.

RATIONALE: Measured at intensive care unit admission (ICU), the predictive value of neutrophil gelatinase-associated lipocalin (NGAL) for severe acute kidney injury (AKI) is unclear. OBJECTIVES: To assess the ability of plasma and urine NGAL to predict severe AKI in adult critically ill patients. METHODS: Prospective-cohort study consisting of 632 consecutive patients. MEASUREMENTS AND MAIN RESULTS: Samples were analyzed by Triage immunoassay for NGAL expression. The primary outcome measure was occurrence of AKI based on Risk-Injury-Failure (RIFLE) classification during the first week of ICU stay. A total of 171 (27%) patients developed AKI. Of these 67, 48, and 56 were classified as RIFLE R, I, and F, respectively. Plasma and urine NGAL values at ICU admission were significantly related to AKI severity. The areas under the receiver operating characteristic curves for plasma and urine NGAL were for RIFLE R (0.77 ± 0.05 and 0.80 ± 0.04, respectively), RIFLE I (0.80 ± 0.06 and 0.85 ± 0.04, respectively), and RIFLE F (0.86 ± 0.06 and 0.88 ± 0.04, respectively) and comparable with those of admission estimated glomerular filtration rate (eGFR) (0.84 ± 0.04, 0.87 ± 0.04, and 0.92 ± 0.04, respectively). Plasma and urine NGAL significantly contributed to the accuracy of the "most efficient clinical model" with the best four variables including eGFR, improving the area under the curve for RIFLE F prediction to 0.96 ± 0.02 and 0.95 ± 0.01. Serial NGAL measurements did not provide additional information for the prediction of RIFLE F. CONCLUSIONS: NGAL measured at ICU admission predicts the development of severe AKI similarly to serum creatinine-derived eGFR. However, NGAL adds significant accuracy to this prediction in combination with eGFR alone or with other clinical parameters and has an interesting predictive value in patients with normal serum creatinine.