Identifying temporal patterns in patient disease trajectories using dynamic time warping: A population-based study

Time is a crucial parameter in the assessment of comorbidities in population-based studies, as it permits to identify more complex disease patterns apart from the pairwise disease associations. So far, it has been, either, completely ignored or only, taken into account by assessing the temporal directionality of identified comorbidity pairs. In this work, a novel time-analysis framework is presented for large-scale comorbidity studies. The disease-history vectors of patients of a regional Spanish health dataset are represented as time sequences of ordered disease diagnoses. Statistically significant pairwise disease associations are identified and their temporal directionality is assessed. Subsequently, an unsupervised clustering algorithm, based on Dynamic Time Warping, is applied on the common disease trajectories in order to group them according to the temporal patterns that they share. The proposed methodology for the temporal assessment of such trajectories could serve as the preliminary basis of a disease prediction system.

Incidence, predictors, and impact on survival of left ventricular systolic dysfunction and recovery in advanced cancer patients

Although left ventricular (LV) dysfunction occurs not uncommonly in the course of cancer therapy, little is known about its natural history and prognostic impact on patients. To investigate the incidence, predictors, and impact on survival of LV systolic dysfunction and recovery during cancer therapy, we conducted a retrospective cohort observational study over 1 year at the University of Texas MD Anderson Cancer Center. We enrolled patients with a decrease in ejection fraction by echocardiography to <50% while undergoing cancer therapy from January 2009 to December 2009. We collected and analyzed their chart data. Of 7,648 patients with echocardiograms in 2009, 366 (4.8%) had ejection fraction <50% and 104 met study criteria. LV systolic dysfunction was associated with cardiotoxic therapy in 53 patients (51%). Recovery occurred in 57 patients (55%) and was independently predicted by younger age, smaller left atrial volume index, and lower B-type natriuretic peptide. At last follow-up, 69 patients (66%) were dead, and 35 (34%) were alive. There was a 20% advantage in 2-year survival among patients with LV systolic recovery compared with those without (95% confidence interval 4% to 41%, p = 0.02). In this retrospective study, LV systolic dysfunction recovery occurred in over half of the patients, appeared independent of cardiotoxic etiology, and associated with a 20% survival benefit at 2 years. Multivariable predictors of recovery are younger age, a small left atrial volume index, and lower B-type natriuretic peptide.

Pragmatic measures: what they are and why we need them

Pragmatic measures are important to facilitate implementation and dissemination, address stakeholder issues, and drive quality improvement. This paper proposes necessary and recommended criteria for pragmatic measures, provides examples of projects to develop and identify such measures, addresses potential concerns about these recommendations, and identifies areas for future research and application. Key criteria for pragmatic measures include importance to stakeholders in addition to researchers, low burden, broad applicability, sensitivity to change, and being actionable. Examples of pragmatic measures are provided, including ones for different settings (e.g., primary care, hospital) and levels (e.g., individual, practitioner, setting) that illustrate approaches to produce broad-scale dissemination and the development of brief, standardized measures for use in pragmatic studies. There is an important need for pragmatic measures to facilitate pragmatic research, guide quality improvement, and inform progress on public health goals, but few examples are currently available. Development and evaluation of pragmatic measures and metrics would provide useful resources to advance science, policy, and practice.