Statistical methods and strategies for working with large data bases

High-Content Imaging to Phenotype Human Primary and iPSC-Derived Cells

Increasingly powerful microscopy, liquid handling, and computational techniques have enabled cell imaging in high throughput. Microscopy images are quantified using high-content analysis platforms linking object features to cell behavior. This can be attempted on physiologically relevant cell models, including stem cells and primary cells, in complex environments, and conceivably in the presence of perturbations. Recently, substantial focus has been devoted to cell profiling for cell therapy, assays for drug discovery or biomarker identification for clinical decision-making protocols, bringing this wealth of information into translational applications. In this chapter, we focus on two protocols enabling to (1) benchmark human cells, in particular human endothelial cells as a case study and (2) extract cells from blood for follow-up experiments including image-based drug testing. We also present concepts of high-content imaging and discuss the benefits and challenges, with the aim of enabling readers to tailor existing pipelines and bring such approaches closer to translational research and the clinic.

An observational study on the rate of reporting of adverse event on healthcare staff in a mental health setting: An application of Poisson expectation maximisation analysis on nurse staffing data

Evidence highlights the intrinsic link between nurse staffing and expertise, and outcomes for service users of healthcare, and that workforce retention is linked to the clinical and organisational experiences of employees. However, this understanding is less well established in mental health. This study comprises a retrospective observational study carried out on routinely collected data from a large mental healthcare provider. Two databases comprising nurse staffing levels and adverse events were modelled using latent variable methods to account for the presence of multiple underlying behaviours. The analysis reveals a strong dependence of the rate of adverse events on the location and perceived clinical demand of the wards, and a reduction in adverse events where registered nurses exceed ‘clinically required levels’. In the first study of its kind, these findings present significant implications for nursing workforce policy and present an opportunity to not only improve safety but potentially impact nurse retention.

Risk adjustment and outcome research. Part I

OBJECTIVE

The increasing demand for comparative evaluation of outcomes requires the development and diffusion of epidemiologic research, the ability to correctly formulate hypotheses, to conduct analyses and to interpret the results. The purpose of this paper is to provide a detailed but easy-reading review of epidemiologic methods to compare healthcare outcomes, particularly risk-adjustment methods.

METHODS

The paper is divided into three parts. Part I describes confounding in observational studies, the ways confounding is identified and controlled (propensity adjustment and risk adjustment), and the methods for constructing the severity measures in risk-adjustment procedures.

CONCLUSIONS

It is becoming increasingly important for policy makers and planners to identify which factors may improve or worsen the effectiveness of treatments and services and to compare the performances of providers. Politicians, managers, epidemiologists, and clinicians should make their decisions based on the validity and precision of study results, by using the best scientific knowledge available. The statistical methods described in this review cannot measure 'reality' as it 'truly' is, but can produce 'images' of it, defining limits and uncertainties in terms of validity and precision. Studies that use credible risk-adjustment strategies are more likely to yield reliable and applicable findings.

Neurological Complications After Coronary Artery Bypass Grafting Related to the Performance of Cardiopulmonary Bypass

STUDY OBJECTIVES

Neurologic disorders belong among the most serious complications of cardiac surgery. We tested the hypothesis that combinations of hemodynamic events from apparently normal cardiopulmonary bypass (CPB) procedures are related to the development of postoperative neurologic complications and affect the impact of common clinical risk factors.

DESIGN

Retrospective study.

SETTING

Cardiothoracic surgery department in a university hospital.

METHODS AND PATIENTS

A multivariate statistical procedure (ie, cluster analysis) was applied to a data set of automatically recorded perfusions from 1,395 patients who had undergone coronary artery bypass grafting. One-way analysis of variance was used to select five parameters with the strongest significant correlation to postoperative neurologic complications for further cluster analysis. The dependencies in the clusters were tested against common clinical risk factors. To our knowledge, this is the first study of its kind.

RESULTS

The following five parameters emerged for cluster analysis: mean arterial pressure (MAP); dispersion of MAP; dispersion of systemic vascular resistance; dispersion of arterial pulse pressure; and the maximum value of mixed venous saturation. Using these parameters, we found four clusters that were significantly different by CPB performance (first cluster, 389 patients; second cluster, 431 patients; third cluster; and fourth cluster, 229 patients). The frequency of postoperative neurologic complications was 0.3% in the first cluster and increased to 3.9% in the fourth cluster. Importantly, the impact of common clinical risk factors for postoperative neurologic complications was affected by the performance of the CPB procedure. For example, the frequency of neurologic complications among patients with cerebrovascular disease in their medical history was 22% in the fourth cluster, whereas it was zero in the second cluster.

CONCLUSIONS

This study shows that apparently normal CPB procedures affect the impact of common clinical risk factors on postoperative neurologic complications. Patients who underwent CPB procedures with large fluctuations in hemodynamic parameters particularly showed an increased risk for the development of postoperative neurologic complications.

Relationship Between Processes of Care and Coronary Bypass Operative Mortality and Morbidity

BACKGROUND

Information is limited regarding the effects of processes of care on cardiac surgical outcomes. Correspondingly, many recommended cardiac surgical processes of care are derived from animal experiments or clinical judgment. This report from the VA Cooperative Study in Health Services, "Processes, Structures, and Outcomes of Cardiac Surgery," focuses on the relationships between 3 process groups (preoperative evaluation, intraoperative care, and supervision by senior physicians) and a composite outcome, perioperative mortality and morbidity.

METHODS

Data on 734 risk, process, and structure variables were collected prospectively on 3,988 patients who underwent coronary artery bypass grafting at 14 VA medical centers between 1992 and 1996. Data reduction was accomplished by examining data completeness and variation across sites and surgeon, using previously published data and clinical judgment. We then applied multivariable logistic regression to the 39 remaining processes of care to determine which were related to the composite outcome after adjusting for 17 patient-related risk factors and controlling for intraoperative complications.

RESULTS

Our first analysis showed several measures of operative duration, the use of inotropic agents, transesophageal echo, lowest systemic temperature, and hemoconcentration/ultrafiltration, to be powerful predictors of the composite outcome. Because the use of inotropic agents and operative duration may be related to an intermediate outcome (eg, intraoperative complications), we performed a second analysis omitting these processes. The use of intraoperative transesophageal echo and hemoconcentration/ultrafiltration remained significantly associated with an increased risk of an event (odds ratios 1.60 and 1.36, respectively).

CONCLUSIONS

Our results viewed in the context of past studies suggest the possibility that inotropic use, TEE, and hemoconcentration/ultrafiltration may have adverse effects on operative outcome. Further evaluation of these processes of care using observational data, as well as randomized trials when feasible, would be of interest.

A way of obtaining isoresource consumption care episodes

The purpose of this study is to identify groups of care episodes that involve pneumologic and cardiologic problems and that exhibit homogeneous patterns of frequentation and diagnostic test consumption in an ambulatory consulting room. A 1-year prospective study of care content was done, and the episodes were retrospectively analyzed 2 years later. Data were collected in an ambulatory cardiologic and pneumologic consulting room in Spain. Nonlinear principal components analysis was applied before cluster analysis. Five typologies with a homogeneous pattern of resource consumption were obtained: three related to acute episodes of care and two to chronic ones.

Health services research and the endocrinologist

This article highlights the importance of health services research to endocrinologists. The content and goals of health services research are defined, and, with examples related to endocrinology, the field's focus and key themes are described and its methods and sources of data delineated. Considerations that informed readers should keep in mind when reading this literature are illustrated, with a recent example that has important implications for the role of endocrinologists in the management of diabetic patients.