The publication data currently available has been vetted by Vanderbilt faculty, staff, administrators and trainees. The data itself is retrieved directly from NCBI's PubMed and is automatically updated on a weekly basis to ensure accuracy and completeness.
If you have any questions or comments, please contact us.
Obstetric care refers to the care provided to patients during ante-, intra-, and postpartum periods. Predicting length of stay (LOS) for these patients during their hospitalizations can assist healthcare organizations in allocating hospital resources more effectively and efficiently, ultimately improving maternal care quality and reducing costs to patients. In this paper, we investigate the extent to which LOS can be forecast from a patient's medical history. We introduce a machine learning framework to incorporate a patient's prior conditions (e.g., diagnostic codes) as features in a predictive model for LOS. We evaluate the framework with three years of historical billing data from the electronic medical records of 9188 obstetric patients in a large academic medical center. The results indicate that our framework achieved an average accuracy of 49.3%, which is higher than the baseline accuracy 37.7% (that relies solely on a patient's age). The most predictive features were found to have statistically significant discriminative ability. These features included billing codes for normal delivery (indicative of shorter stay) and antepartum hypertension (indicative of longer stay).
Patients have diverse health information needs, and secure messaging through patient portals is an emerging means by which such needs are expressed and met. As patient portal adoption increases, growing volumes of secure messages may burden healthcare providers. Automated classification could expedite portal message triage and answering. We created four automated classifiers based on word content and natural language processing techniques to identify health information needs in 1000 patient-generated portal messages. Logistic regression and random forest classifiers detected single information needs well, with area under the curves of 0.804-0.914. A logistic regression classifier accurately found the set of needs within a message, with a Jaccard index of 0.859 (95% Confidence Interval: (0.847, 0.871)). Automated classification of consumer health information needs expressed in patient portal messages is feasible and may allow direct linking to relevant resources or creation of institutional resources for commonly expressed needs.
Ecologists are increasingly using statistical models to predict animal abundance and occurrence in unsampled locations. The reliability of such predictions depends on a number of factors, including sample size, how far prediction locations are from the observed data, and similarity of predictive covariates in locations where data are gathered to locations where predictions are desired. In this paper, we propose extending Cook's notion of an independent variable hull (IVH), developed originally for application with linear regression models, to generalized regression models as a way to help assess the potential reliability of predictions in unsampled areas. Predictions occurring inside the generalized independent variable hull (gIVH) can be regarded as interpolations, while predictions occurring outside the gIVH can be regarded as extrapolations worthy of additional investigation or skepticism. We conduct a simulation study to demonstrate the usefulness of this metric for limiting the scope of spatial inference when conducting model-based abundance estimation from survey counts. In this case, limiting inference to the gIVH substantially reduces bias, especially when survey designs are spatially imbalanced. We also demonstrate the utility of the gIVH in diagnosing problematic extrapolations when estimating the relative abundance of ribbon seals in the Bering Sea as a function of predictive covariates. We suggest that ecologists routinely use diagnostics such as the gIVH to help gauge the reliability of predictions from statistical models (such as generalized linear, generalized additive, and spatio-temporal regression models).
OBJECTIVES - The goal of this study was to examine 2006 to 2010 emergency department (ED) admission rates, hospital procedures, lengths of stay, and costs for acute heart failure (AHF).
BACKGROUND - Patients with AHF are often admitted and are associated with high readmissions and cost.
METHODS - We utilized Nationwide Emergency Department Sample AHF data from 2006 to 2010 to describe admission proportion, hospital length of stay (LOS), and ED charges as a surrogate for resource utilization. Results were compared across U.S. regions, patient insurance status, and hospital characteristics.
RESULTS - There were 958,167 mean yearly ED visits for AHF in the United States. Fifty-one percent of the patients were female, and the median age was 75.1 years (interquartile range [IQR]: 62.5 to 83.7 years). Overall, 83.7% (95% confidence interval: 83.1% to 84.2%) were admitted; the median LOS was 3.4 days (IQR: 1.9 to 5.8 days). Comparing 2006 with 2010, there was a small decrease in median LOS (0.09 days), but the proportion admitted did not change. Odds of admission, adjusting for age, sex, hospital characteristic (academic and safety net status), and insurance (Medicare, Medicaid, private, self-pay/no charge) were highest in the Northeast. Median ED charges were $1,075 (IQR: $679 to $1,665) in 2006 and $1,558 (IQR: $1,018 to $2,335) in 2010. Patients without insurance were more likely to be discharged from the ED, but when admitted, were more likely to receive a major diagnostic or therapeutic procedure.
CONCLUSIONS - A very high proportion of ED patients with AHF are admitted nationally, with significant variation in disposition and procedural decisions based on region of the country and type of insurance, even after adjusting for potential confounding.
Copyright © 2014 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.
OBJECT - Recent legislation and media coverage have heightened awareness of concussion in youth sports. Previous work by the authors' group defined significant variation of care in management of children with concussion. To address this variation, a multidisciplinary concussion program was established based on a uniform management protocol, with emphasis on community outreach via traditional media sources and the Internet. This retrospective study evaluates the impact of standardization of concussion care and resource utilization before and after standardization in a large regional pediatric hospital center.
METHODS - This retrospective study included all patients younger than 18 years of age evaluated for sports-related concussion between January 1, 2007, and December 31, 2011. Emergency department, sports medicine, and neurosurgery records were reviewed. Data collected included demographics, injury details, clinical course, Sports Concussion Assessment Tool-2 (SCAT2) scores, imaging, discharge instructions, and referral for specialty care. The cohort was analyzed comparing patients evaluated before and after standardization of care.
RESULTS - Five hundred eighty-nine patients were identified, including 270 before standardization (2007-2011) and 319 after standardization (2011-2012). Statistically significant differences (p < 0.0001) were observed between the 2 groups for multiple variables: there were more girls, more first-time concussions, fewer initial presentations to the emergency department, more consistent administration of the SCAT2, and more consistent supervision of return to play and return to think after adoption of the protocol.
CONCLUSIONS - A combination of increased public awareness and legislation has led to a 5-fold increase in the number of youth athletes presenting for concussion evaluation at the authors' center. Establishment of a multidisciplinary clinic with a standardized protocol resulted in significantly decreased institutional resource utilization and more consistent concussion care for this growing patient population.
BACKGROUND - Extensive patient and family education is required at the time of a new diagnosis of pediatric cancer yet little data exist regarding the availability and linguistic competency of new cancer diagnosis education provided by pediatric oncology institutions.
PROCEDURE - Using the American Society of Pediatric Hematology/Oncology (ASPHO) membership list, a web-based survey was conducted among a cohort of pediatric oncologists to determine pediatric oncologists' assessment of institutional resources for new cancer diagnosis education and the availability of linguistically appropriate education.
RESULTS - Of 1,294 ASPHO members sent email survey invitations, 573 (44.3%) responded with 429 meeting eligibility criteria. Oncologists at academic institutions reported their institutions had more availability of resources for new diagnosis education compared with those from non-academic institutions (mean 78.6 vs. 74.3; 0 [not at all]-100 [well equipped]; P = 0.05). The mean score increased with volume of new cancer diagnoses/year: small (<75) = 73.4; medium (75-149) = 76.7; large (>150) = 84.5 (P < 0.001). Oncologists at large volume institutions reported more availability of an established patient education protocol (50.8% vs. 38.1%, P < 0.001) and increased use of dedicated non-physician staff (79.9% vs. 66.1%, P = 0.02), but less use of websites for patient education (17.2% vs. 33.3%, P = 0.001). Availability of linguistically appropriate education improved with increasing institution size: small (76.4), medium (82.3), and large (84.0) patient volume (P < 0.011).
CONCLUSION - According to pediatric oncologists, a disparity in educational and linguistic resources for new pediatric cancer diagnosis education exists depending on institution type and size.
© 2013 Wiley Periodicals, Inc.
Population declines due to amphibian chytridiomycosis among selected species of ranid frogs from western North America have been severe, but there is evidence that the Oregon spotted frog, Rana pretiosa Baird and Girard, 1853, displays resistance to the disease. Norepinephrine-stimulated skin secretions were collected from a non-declining population of R. pretiosa that had been exposed to the causative agent Batrachochytrium dendrobatidis. Peptidomic analysis led to identification and isolation, in pure form, of a total of 18 host-defense peptides that were characterized structurally. Brevinin-1PRa, -1PRb, -1PRc, and -1PRd, esculentin-2PRa and -PRb, ranatuerin-2PRa, -2PRb, -2PRc, and -2PRe, temporin-PRb and -PRc were identified in an earlier study of skin secretions of frogs from a different population of R. pretiosa known to be declining. Ranatuerin-2PRf, -2PRg, -2PRh, temporin-PRd, -PRe, and -PRf were not identified in skin secretions from frogs from the declining population, whereas temporin-PRa and ranatuerin-2PRd, present in skin secretions from the declining population, were not detected in the current study. All purified peptides inhibited the growth of B. dendrobatidis zoospores. Peptides of the brevinin-1 and esculentin-2 families displayed the highest potency (minimum inhibitory concentration = 6.25-12.5 μM). The study provides support for the hypothesis that the multiplicity and diversity of the antimicrobial peptide repertoire in R. pretiosa and the high growth-inhibitory potency of certain peptides against B. dendrobatidis are important in conferring a measure of resistance to fatal chytridiomycosis.
OBJECTIVE - Major amputation is associated with increased short-term healthcare resource utilization (RU), early mortality, and socioeconomic status (SES) disparities. Our objective is to study patient-specific and SES-related predictors of long-term RU and survival after amputation.
METHODS - This retrospective analysis identified 364 adult patients who underwent index major amputation for critical limb ischemia from January 1995 through December 2000 at two tertiary centers with outcomes through December 2010. Age, gender, SES (race, income, insurance, and marital status), comorbidities (congestive heart failure [CHF], diabetes, diabetes with complications, and renal failure [RF]), subsequent procedures, cumulative length of stay (cLOS), and mortality were analyzed. Bivariate and multivariate Poisson regression for subsequent procedures and cLOS and Cox proportional hazard modeling for all-cause mortality were undertaken.
RESULTS - During a mean follow-up of 3.25 years, amputation patients had mean cLOS of 71.2 days per person-year (median, 17.6), 19.5 readmissions per person-year (median, 2.1), 0.57 amputation-related procedures (median, 0), and 0.31 cardiovascular procedures (median, 0). Below-knee amputation as the index procedure was performed in 70% of patients, and 25% had additional amputation procedures. Of readmissions at ≤ 30 days, 52% were amputation-related. Overall mortality during follow-up was 86.9%; 37 patients (10.2%) died within 30 days. Among patients surviving >30 days, multivariate Poisson regression demonstrated that younger age (incidence rate ratio [IRR], 0.98), public insurance (IRR, 1.63), CHF (IRR, 1.60), and RF (IRR, 2.12) were associated with increased cLOS. Diabetes with complications (IRR, 1.90) and RF (IRR, 2.47) affected subsequent amputation procedures. CHF (IRR, 1.83) and RF (IRR, 3.67) were associated with a greater number of cardiovascular procedures. Cox proportional hazard modeling indicated older age (hazard ratio [HR], 1.04), CHF (HR, 2.26), and RF (HR, 2.60) were risk factors for decreased survival. Factors associated with SES were not significantly related to the outcomes.
CONCLUSIONS - This study found that RU is high for amputees, and increased RU persists beyond the perioperative period. Results were similar across SES indices, suggesting higher SES may not be protective against poor outcomes when limb salvage is no longer attainable. These findings support the hypothesis that SES disparities may be more modifiable during earlier stages of care for critical limb ischemia.
Copyright © 2013. Published by Mosby, Inc.
Despite available demographic data on the factors that contribute to breast cancer mortality in large population datasets, local patterns are often overlooked. Such local information could provide a valuable metric by which regional community health resources can be allocated to reduce breast cancer mortality. We used national and statewide datasets to assess geographical distribution of breast cancer mortality rates and known risk factors influencing breast cancer mortality in middle Tennessee. Each county in middle Tennessee, and each ZIP code within metropolitan Davidson County, was scored for risk factor prevalence and assigned quartile scores that were used as a metric to identify geographic areas of need. While breast cancer mortality often correlated with age and incidence, geographic areas were identified in which breast cancer mortality rates did not correlate with age and incidence, but correlated with additional risk factors, such as mammography screening and socioeconomic status. Geographical variability in specific risk factors was evident, demonstrating the utility of this approach to identify local areas of risk. This method revealed local patterns in breast cancer mortality that might otherwise be overlooked in a more broadly based analysis. Our data suggest that understanding the geographic distribution of breast cancer mortality, and the distribution of risk factors that contribute to breast cancer mortality, will not only identify communities with the greatest need of support, but will identify the types of resources that would provide the most benefit to reduce breast cancer mortality in the community.
Groesbeck Parham and colleagues describe their Cervical Cancer Prevention Program in Zambia, which has provided services to over 58,000 women over the past five years, and share lessons learned from the program's implementation and integration with existing HIV/AIDS programs.