Optimized approach to decision fusion of heterogeneous data for breast cancer diagnosis.

As more diagnostic testing options become available to physicians, it becomes more difficult to combine various types of medical information together in order to optimize the overall diagnosis. To improve diagnostic performance, here we introduce an approach to optimize a decision-fusion technique to combine heterogeneous information, such as from different modalities, feature categories, or institutions. For classifier comparison we used two performance metrics: The receiving operator characteristic (ROC) area under the curve [area under the ROC curve (AUC)] and the normalized partial area under the curve (pAUC). This study used four classifiers: Linear discriminant analysis (LDA), artificial neural network (ANN), and two variants of our decision-fusion technique, AUC-optimized (DF-A) and pAUC-optimized (DF-P) decision fusion. We applied each of these classifiers with 100-fold cross-validation to two heterogeneous breast cancer data sets: One of mass lesion features and a much more challenging one of microcalcification lesion features. For the calcification data set, DF-A outperformed the other classifiers in terms of AUC (p < 0.02) and achieved AUC=0.85 +/- 0.01. The DF-P surpassed the other classifiers in terms of pAUC (p < 0.01) and reached pAUC=0.38 +/- 0.02. For the mass data set, DF-A outperformed both the ANN and the LDA (p < 0.04) and achieved AUC=0.94 +/- 0.01. Although for this data set there were no statistically significant differences among the classifiers' pAUC values (pAUC=0.57 +/- 0.07 to 0.67 +/- 0.05, p > 0.10), the DF-P did significantly improve specificity versus the LDA at both 98% and 100% sensitivity (p < 0.04). In conclusion, decision fusion directly optimized clinically significant performance measures, such as AUC and pAUC, and sometimes outperformed two well-known machine-learning techniques when applied to two different breast cancer data sets.

Establishing a regional, multisite database for quality improvement and service planning in community-based palliative care and hospice.

BACKGROUND: Outpatient palliative care, an evolving delivery model, seeks to improve continuity of care across settings and to increase access to services in hospice and palliative medicine (HPM). It can provide a critical bridge between inpatient palliative care and hospice, filling the gap in community-based supportive care for patients with advanced life-limiting illness. Low capacities for data collection and quantitative research in HPM have impeded assessment of the impact of outpatient palliative care. APPROACH: In North Carolina, a regional database for community-based palliative care has been created through a unique partnership between a HPM organization and academic medical center. This database flexibly uses information technology to collect patient data, entered at the point of care (e.g., home, inpatient hospice, assisted living facility, nursing home). HPM physicians and nurse practitioners collect data; data are transferred to an academic site that assists with analyses and data management. Reports to community-based sites, based on data they provide, create a better understanding of local care quality. CURRENT STATUS: The data system was developed and implemented over a 2-year period, starting with one community-based HPM site and expanding to four. Data collection methods were collaboratively created and refined. The database continues to grow. Analyses presented herein examine data from one site and encompass 2572 visits from 970 new patients, characterizing the population, symptom profiles, and change in symptoms after intervention. CONCLUSION: A collaborative regional approach to HPM data can support evaluation and improvement of palliative care quality at the local, aggregated, and statewide levels.

Low demographic variability in wild primate populations: fitness impacts of variation, covariation, and serial correlation in vital rates.

In a stochastic environment, long-term fitness can be influenced by variation, covariation, and serial correlation in vital rates (survival and fertility). Yet no study of an animal population has parsed the contributions of these three aspects of variability to long-term fitness. We do so using a unique database that includes complete life-history information for wild-living individuals of seven primate species that have been the subjects of long-term (22-45 years) behavioral studies. Overall, the estimated levels of vital rate variation had only minor effects on long-term fitness, and the effects of vital rate covariation and serial correlation were even weaker. To explore why, we compared estimated variances of adult survival in primates with values for other vertebrates in the literature and found that adult survival is significantly less variable in primates than it is in the other vertebrates. Finally, we tested the prediction that adult survival, because it more strongly influences fitness in a constant environment, will be less variable than newborn survival, and we found only mixed support for the prediction. Our results suggest that wild primates may be buffered against detrimental fitness effects of environmental stochasticity by their highly developed cognitive abilities, social networks, and broad, flexible diets.

Does improved access to diagnostic imaging results reduce hospital length of stay? A retrospective study.

BACKGROUND: One year after the introduction of Information and Communication Technology (ICT) to support diagnostic imaging at our hospital, clinicians had faster and better access to radiology reports and images; direct access to Computed Tomography (CT) reports in the Electronic Medical Record (EMR) was particularly popular. The objective of this study was to determine whether improvements in radiology reporting and clinical access to diagnostic imaging information one year after the ICT introduction were associated with a reduction in the length of patients' hospital stays (LOS). METHODS: Data describing hospital stays and diagnostic imaging were collected retrospectively from the EMR during periods of equal duration before and one year after the introduction of ICT. The post-ICT period was chosen because of the documented improvement in clinical access to radiology results during that period. The data set was randomly split into an exploratory part used to establish the hypotheses, and a confirmatory part. The data was used to compare the pre-ICT and post-ICT status, but also to compare differences between groups. RESULTS: There was no general reduction in LOS one year after ICT introduction. However, there was a 25% reduction for one group - patients with CT scans. This group was heterogeneous, covering 445 different primary discharge diagnoses. Analyses of subgroups were performed to reduce the impact of this divergence. CONCLUSION: Our results did not indicate that improved access to radiology results reduced the patients' LOS. There was, however, a significant reduction in LOS for patients undergoing CT scans. Given the clinicians' interest in CT reports and the results of the subgroup analyses, it is likely that improved access to CT reports contributed to this reduction.

Postoperative statin use and risk of biochemical recurrence following radical prostatectomy: results from the Shared Equal Access Regional Cancer Hospital (SEARCH) database.

OBJECTIVE: To investigate the effect of statin use after radical prostatectomy (RP) on biochemical recurrence (BCR) in patients with prostate cancer who never received statins before RP. PATIENTS AND METHODS: We conducted a retrospective analysis of 1146 RP patients within the Shared Equal Access Regional Cancer Hospital (SEARCH) database. Multivariable Cox proportional hazards analyses were used to examine differences in risk of BCR between post-RP statin users vs nonusers. To account for varying start dates and duration of statin use during follow-up, post-RP statin use was treated as a time-dependent variable. In a secondary analysis, models were stratified by race to examine the association of post-RP statin use with BCR among black and non-black men. RESULTS: After adjusting for clinical and pathological characteristics, post-RP statin use was significantly associated with 36% reduced risk of BCR (hazard ratio [HR] 0.64, 95% confidence interval [CI] 0.47-0.87; P = 0.004). Post-RP statin use remained associated with reduced risk of BCR after adjusting for preoperative serum cholesterol levels. In secondary analysis, after stratification by race, this protective association was significant in non-black (HR 0.49, 95% CI 0.32-0.75; P = 0.001) but not black men (HR 0.82, 95% CI 0.53-1.28; P = 0.384). CONCLUSION: In this retrospective cohort of men undergoing RP, post-RP statin use was significantly associated with reduced risk of BCR. Whether the association between post-RP statin use and BCR differs by race requires further study. Given these findings, coupled with other studies suggesting that statins may reduce risk of advanced prostate cancer, randomised controlled trials are warranted to formally test the hypothesis that statins slow prostate cancer progression.

Pharmacogenomics in early-phase clinical development.

Pharmacogenomics (PGx) offers the promise of utilizing genetic fingerprints to predict individual responses to drugs in terms of safety, efficacy and pharmacokinetics. Early-phase clinical trial PGx applications can identify human genome variations that are meaningful to study design, selection of participants, allocation of resources and clinical research ethics. Results can inform later-phase study design and pipeline developmental decisions. Nevertheless, our review of the clinicaltrials.gov database demonstrates that PGx is rarely used by drug developers. Of the total 323 trials that included PGx as an outcome, 80% have been conducted by academic institutions after initial regulatory approval. Barriers for the application of PGx are discussed. We propose a framework for the role of PGx in early-phase drug development and recommend PGx be universally considered in study design, result interpretation and hypothesis generation for later-phase studies, but PGx results from underpowered studies should not be used by themselves to terminate drug-development programs.

Variation in Inpatient Rehabilitation Utilization After Hospitalization for Burn Injury in the United States.

Approximately 45,000 individuals are hospitalized annually for burn treatment. Rehabilitation after hospitalization can offer a significant improvement in functional outcomes. Very little is known nationally about rehabilitation for burns, and practices may vary substantially depending on the region based on observed Medicare post-hospitalization spending amounts. This study was designed to measure variation in rehabilitation utilization by state of hospitalization for patients hospitalized with burn injury. This retrospective cohort study used nationally collected data over a 10-year period (2001 to 2010), from the Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SIDs). Patients hospitalized for burn injury (n = 57,968) were identified by ICD-9-CM codes and were examined to see specifically if they were discharged immediately to inpatient rehabilitation after hospitalization (primary endpoint). Both unadjusted and adjusted likelihoods were calculated for each state taking into account the effects of age, insurance status, hospitalization at a burn center, and extent of burn injury by TBSA. The relative risk of discharge to inpatient rehabilitation varied by as much as 6-fold among different states. Higher TBSA, having health insurance, higher age, and burn center hospitalization all increased the likelihood of discharge to inpatient rehabilitation following acute care hospitalization. There was significant variation between states in inpatient rehabilitation utilization after adjusting for variables known to affect each outcome. Future efforts should be focused on identifying the cause of this state-to-state variation, its relationship to patient outcome, and standardizing treatment across the United States.

Variation in pediatric traumatic brain injury outcomes in the United States.

OBJECTIVE: To ascertain the degree of variation, by state of hospitalization, in outcomes associated with traumatic brain injury (TBI) in a pediatric population. DESIGN: A retrospective cohort study of pediatric patients admitted to a hospital with a TBI. SETTING: Hospitals from states in the United States that voluntarily participate in the Agency for Healthcare Research and Quality's Healthcare Cost and Utilization Project. PARTICIPANTS: Pediatric (age ≤ 19 y) patients hospitalized for TBI (N=71,476) in the United States during 2001, 2004, 2007, and 2010. INTERVENTIONS: None. MAIN OUTCOME MEASURES: Primary outcome was proportion of patients discharged to rehabilitation after an acute care hospitalization among alive discharges. The secondary outcome was inpatient mortality. RESULTS: The relative risk of discharge to inpatient rehabilitation varied by as much as 3-fold among the states, and the relative risk of inpatient mortality varied by as much as nearly 2-fold. In the United States, approximately 1981 patients could be discharged to inpatient rehabilitation care if the observed variation in outcomes was eliminated. CONCLUSIONS: There was significant variation between states in both rehabilitation discharge and inpatient mortality after adjusting for variables known to affect each outcome. Future efforts should be focused on identifying the cause of this state-to-state variation, its relationship to patient outcome, and standardizing treatment across the United States.

Things learned in early adulthood are remembered best.

Evidence is reviewed that for older adults the period from 10 to 30 years of age produces recall of the most autobiographical memories, the most vivid memories, and the most important memories. It is the period from which peoples' favorite films, music, and books come and the period from which they judge the most important world events to have originated. Factual, semantic, general-knowledge, multiple-choice questions about the Academy Awards, the World Series, and current events from this period were answered more accurately by two different groups of 30 older adults tested 10 years apart. A cognitive theory based on the importance of transitions and several noncognitive theories are considered as explanations of this pervasive phenomenon.

Finding our way through phenotypes.

Despite a large and multifaceted effort to understand the vast landscape of phenotypic data, their current form inhibits productive data analysis. The lack of a community-wide, consensus-based, human- and machine-interpretable language for describing phenotypes and their genomic and environmental contexts is perhaps the most pressing scientific bottleneck to integration across many key fields in biology, including genomics, systems biology, development, medicine, evolution, ecology, and systematics. Here we survey the current phenomics landscape, including data resources and handling, and the progress that has been made to accurately capture relevant data descriptions for phenotypes. We present an example of the kind of integration across domains that computable phenotypes would enable, and we call upon the broader biology community, publishers, and relevant funding agencies to support efforts to surmount today's data barriers and facilitate analytical reproducibility.

Overview of FEED, the feeding experiments end-user database.

The Feeding Experiments End-user Database (FEED) is a research tool developed by the Mammalian Feeding Working Group at the National Evolutionary Synthesis Center that permits synthetic, evolutionary analyses of the physiology of mammalian feeding. The tasks of the Working Group are to compile physiologic data sets into a uniform digital format stored at a central source, develop a standardized terminology for describing and organizing the data, and carry out a set of novel analyses using FEED. FEED contains raw physiologic data linked to extensive metadata. It serves as an archive for a large number of existing data sets and a repository for future data sets. The metadata are stored as text and images that describe experimental protocols, research subjects, and anatomical information. The metadata incorporate controlled vocabularies to allow consistent use of the terms used to describe and organize the physiologic data. The planned analyses address long-standing questions concerning the phylogenetic distribution of phenotypes involving muscle anatomy and feeding physiology among mammals, the presence and nature of motor pattern conservation in the mammalian feeding muscles, and the extent to which suckling constrains the evolution of feeding behavior in adult mammals. We expect FEED to be a growing digital archive that will facilitate new research into understanding the evolution of feeding anatomy.

Mapping the semantic structure of cognitive neuroscience.

Cognitive neuroscience, as a discipline, links the biological systems studied by neuroscience to the processing constructs studied by psychology. By mapping these relations throughout the literature of cognitive neuroscience, we visualize the semantic structure of the discipline and point to directions for future research that will advance its integrative goal. For this purpose, network text analyses were applied to an exhaustive corpus of abstracts collected from five major journals over a 30-month period, including every study that used fMRI to investigate psychological processes. From this, we generate network maps that illustrate the relationships among psychological and anatomical terms, along with centrality statistics that guide inferences about network structure. Three terms--prefrontal cortex, amygdala, and anterior cingulate cortex--dominate the network structure with their high frequency in the literature and the density of their connections with other neuroanatomical terms. From network statistics, we identify terms that are understudied compared with their importance in the network (e.g., insula and thalamus), are underspecified in the language of the discipline (e.g., terms associated with executive function), or are imperfectly integrated with other concepts (e.g., subdisciplines like decision neuroscience that are disconnected from the main network). Taking these results as the basis for prescriptive recommendations, we conclude that semantic analyses provide useful guidance for cognitive neuroscience as a discipline, both by illustrating systematic biases in the conduct and presentation of research and by identifying directions that may be most productive for future research.

Road Traffic Injury Prevention Initiatives: A Systematic Review and Metasummary of Effectiveness in Low and Middle Income Countries.

BACKGROUND: Road traffic injuries (RTIs) are a growing but neglected global health crisis, requiring effective prevention to promote sustainable safety. Low- and middle-income countries (LMICs) share a disproportionately high burden with 90% of the world's road traffic deaths, and where RTIs are escalating due to rapid urbanization and motorization. Although several studies have assessed the effectiveness of a specific intervention, no systematic reviews have been conducted summarizing the effectiveness of RTI prevention initiatives specifically performed in LMIC settings; this study will help fill this gap. METHODS: In accordance with PRISMA guidelines we searched the electronic databases MEDLINE, EMBASE, Scopus, Web of Science, TRID, Lilacs, Scielo and Global Health. Articles were eligible if they considered RTI prevention in LMICs by evaluating a prevention-related intervention with outcome measures of crash, RTI, or death. In addition, a reference and citation analysis was conducted as well as a data quality assessment. A qualitative metasummary approach was used for data analysis and effect sizes were calculated to quantify the magnitude of emerging themes. RESULTS: Of the 8560 articles from the literature search, 18 articles from 11 LMICs fit the eligibility and inclusion criteria. Of these studies, four were from Sub-Saharan Africa, ten from Latin America and the Caribbean, one from the Middle East, and three from Asia. Half of the studies focused specifically on legislation, while the others focused on speed control measures, educational interventions, enforcement, road improvement, community programs, or a multifaceted intervention. CONCLUSION: Legislation was the most common intervention evaluated with the best outcomes when combined with strong enforcement initiatives or as part of a multifaceted approach. Because speed control is crucial to crash and injury prevention, road improvement interventions in LMIC settings should carefully consider how the impact of improvements will affect speed and traffic flow. Further road traffic injury prevention interventions should be performed in LMICs with patient-centered outcomes in order to guide injury prevention in these complex settings.

The accuracy and completeness for receipt of colorectal cancer care using Veterans Health Administration administrative data.

BACKGROUND: The National Comprehensive Cancer Network and the American Society of Clinical Oncology have established guidelines for the treatment and surveillance of colorectal cancer (CRC), respectively. Considering these guidelines, an accurate and efficient method is needed to measure receipt of care. METHODS: The accuracy and completeness of Veterans Health Administration (VA) administrative data were assessed by comparing them with data manually abstracted during the Colorectal Cancer Care Collaborative (C4) quality improvement initiative for 618 patients with stage I-III CRC. RESULTS: The VA administrative data contained gender, marital, and birth information for all patients but race information was missing for 62.1% of patients. The percent agreement for demographic variables ranged from 98.1-100%. The kappa statistic for receipt of treatments ranged from 0.21 to 0.60 and there was a 96.9% agreement for the date of surgical resection. The percentage of post-diagnosis surveillance events in C4 also in VA administrative data were 76.0% for colonoscopy, 84.6% for physician visit, and 26.3% for carcinoembryonic antigen (CEA) test. CONCLUSIONS: VA administrative data are accurate and complete for non-race demographic variables, receipt of CRC treatment, colonoscopy, and physician visits; but alternative data sources may be necessary to capture patient race and receipt of CEA tests.