Background: In classification and diagnostic testing, the receiver-operator characteristic (ROC) plot and the area under the ROC curve (AUC) describe how an adjustable threshold causes changes in two types of error: false positives and false negatives. Only part of the ROC curve and AUC are informative however when they are used with imbalanced data. Hence, alternatives to the AUC have been proposed, such as the partial AUC and the area under the precision-recall curve. However, these alternatives cannot be as fully interpreted as the AUC, in part because they ignore some information about actual negatives. Methods: We derive and propose a new concordant partial AUC and a new partial c statistic for ROC data-as foundational measures and methods to help understand and explain parts of the ROC plot and AUC. Our partial measures are continuous and discrete versions of the same measure, are derived from the AUC and c statistic respectively, are validated as equal to each other, and validated as equal in summation to whole measures where expected. Our partial measures are tested for validity on a classic ROC example from Fawcett, a variation thereof, and two real-life benchmark data sets in breast cancer: the Wisconsin and Ljubljana data sets. Interpretation of an example is then provided. Results: Results show the expected equalities between our new partial measures and the existing whole measures. The example interpretation illustrates the need for our newly derived partial measures. Conclusions: The concordant partial area under the ROC curve was proposed and unlike previous partial measure alternatives, it maintains the characteristics of the AUC. The first partial c statistic for ROC plots was also proposed as an unbiased interpretation for part of an ROC curve. The expected equalities among and between our newly derived partial measures and their existing full measure counterparts are confirmed. These measures may be used with any data set but this paper focuses on imbalanced data with low prevalence. Future work: Future work with our proposed measures may: demonstrate their value for imbalanced data with high prevalence, compare them to other measures not based on areas; and combine them with other ROC measures and techniques.

Background Studies have shown the effectiveness and user acceptance of mobile health (mHealth) technologies in managing patients with chronic kidney disease (CKD). However, incorporating mHealth technology into the standard care of patients with CKD still faces many challenges. To our knowledge, there are no reviews on mHealth interventions and their assessments concerning the management of patients undergoing dialysis. Objective This study provided a scoping review on existing apps and interventions of mHealth technologies in adult patients undergoing chronic dialysis and identified the gaps in patient outcome assessment of mHealth technologies in the literature. Methods We systematically searched PubMed (MEDLINE), Scopus, and the Cumulative Index to Nursing and Allied Health Literature databases, as well as gray literature sources. Two keywords, “mHealth” and “dialysis,” were combined to address the main concepts of the objectives. Inclusion criteria were as follows: (1) mHealth interventions, which are on a smartphone, tablet, or web-based portals that are accessible through mobile devices; and (2) adult patients (age ≥18 years) on chronic dialysis. Only English papers published from January 2008 to October 2018 were included. Studies with mHealth apps for other chronic conditions, based on e-consultation or videoconferencing, non-English publications, and review papers were excluded. Results Of the 1054 papers identified, 22 met the inclusion and exclusion criteria. Most studies (n=20) were randomized controlled trials and cohort studies. These studies were carried out in 7 countries. The main purposes of these mHealth interventions were as follows: nutrition or dietary self-monitoring (n=7), remote biometric monitoring (n=7), web-based portal (n=4), self-monitoring of in-session dialysis-specific information (n=3), and self-monitoring of lifestyle or behavioral change (n=1). The outcomes of the 22 included studies were organized into five categories: (1) patient satisfaction and acceptance, (2) clinical effectiveness, (3) economic assessment, (4) health-related quality of life, and (5) impact on lifestyle or behavioral change. The mHealth interventions showed neutral to positive results in chronic dialysis patient management, reporting no to significant improvement of dialysis-specific measurements and some components of the overall quality of life assessment. Evaluation of these mHealth interventions consistently demonstrated evidence in patients’ satisfaction, high level of user acceptance, and reduced use of health resources and cost savings to health care services. However, there is a lack of studies evaluating safety, organizational, sociocultural, ethical, and legal aspects of mHealth technologies. Furthermore, a comprehensive cost-effectiveness and cost-benefit analysis of adopting mHealth technologies was not found in the literature. Conclusions The gaps identified in this study will inform the creation of health policies and organizational support for mHealth implementation in patients undergoing dialysis. The findings of this review will inform the development of a comprehensive service model that utilizes mHealth technologies for home monitoring and self-management of patients undergoing chronic dialysis.

BackgroundResearch on factors associated with dialysis withdrawal is scarce. This study examined the predictors that might influence rate of dialysis withdrawal. Existing literature is summarized, analyzed and synthesized to identify gaps in the literature with regard to the factors associated with dialysis withdrawal.MethodsThis scoping review used a systematic search to synthesize research findings related to dialysis withdrawal and identified gaps in the literature. The search strategy was developed and applied using PubMed, EMBASE and CINHAL databases. The selection criteria included articles written in English and published between 1997 and 2016 that examined dialysis withdrawal and associated factors in patients with any modality of renal dialysis.. Case reports and studies only including renal transplant patients were excluded. Fifteen articles were selected in accordance with these selection criteria.ResultsThe literature review revealed a scarcity of research on dialysis withdrawal and associated factors. Furthermore, the study findings were inconsistent and inconclusive. Authors have defined dialysis withdrawal in terms of dialysis discontinuation, withholding, death, withdrawal, treatment refusal/cessation, or technique failure. Authors have selected homogeneous patient population on either hemodialysis (HD) or peritoneal dialysis (PD) patients, thus making comparisons of studies and generalization of findings difficult.ConclusionFuture studies should explore the influence of both HD and PD on patient-elected dialysis withdrawal using a large a priori calculated sample size.Electronic supplementary materialThe online version of this article (10.1186/s12882-018-0894-5) contains supplementary material, which is available to authorized users.

No abstract

Purpose: Cardiovascular rehabilitation programs (CRPs) are effective in secondary stroke prevention, yet the enrollment rate is suboptimal. This study aims to identify demographic and clinical factors and patient-reported reasons for non-enrollment in a center-based outpatient CRP among patients with transient ischemic attack (TIA) or mild stroke. Methods: This mixed-method retrospective chart review was conducted in an outpatient CRP affiliated with a tertiary care hospital in Canada from January 2009 to October 2017. A total of 621 patients with TIA or mild stroke were included. Multiple logistic regression was used to determine the relationship between demographic and clinical predictors with non-enrollment. A thematic analysis of multidisciplinary progress notes was done for the non-enrollment subgroup of patients to understand the patient-reported reasons. Results: The non-enrollment rate was 42%. Travel distance to CRP (OR = 1.024; 95% CI, 1.010-1.038), age (OR = 1.023; 95% CI, 1.004-1.042), and current smoking status (OR = 1.935; 95% CI, 1.230-3.042) were associated with non-enrollment. The patient-reported reasons for non-enrollment were occurrence of new medical events and comorbidities, their perceptions of health and CRP, transportation, work/time conflict, and distance. Conclusions: This study found that patients with TIA or mild stroke who were older, lived farther from the CRP center, or were current smokers were less likely to enroll in a CRP. The present findings may help clinicians identify patients unlikely to enroll in a CRP and allow the implementation of interventions focused on health education and physical activity to improve enrollment. Future research should validate these factors in multiple settings using prospective mixed methods so that interventions can be developed to address non-enrollment in the CRP.