Practical development and operationalization of a 12-hour hospital census prediction algorithm.
Ontology highlight
ABSTRACT:
SUBMITTER: Ryu AJ
PROVIDER: S-EPMC8344501 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Fatigue-related impairments in the nursing workforce contribute to a multitude of health, safety, and economic consequences at the individual, organizational and societal levels. Long and compressed work schedules are commonly worked in the healthcare industry, but more research is needed to understand the cumulative effects of multiple work shifts on physiology-based performance outcomes in nurses. The purpose of this study was to compare the effects of a single nursing work shift versus three compressed (one every 24 hours) 12 hour shifts on performance-based fatigue in nurses and aides. Twenty-six fulltime hospital working nurses and aides (age = 36.1 ± 13.3 years) reported to the lab for testing before, immediately after working a single 12 hour shift, and after working three 12 hour shifts in a 72 hour period. Outcome measures included vigilance-based reaction time, lapses of attention, and muscle function assessments (lower and upper body muscle strength, explosive strength and vertical jump performance). All variables except hand grip strength showed a significant decline following the three work shifts. The psychomotor vigilance reaction time and lapses of attention variables also generally showed a significant decline from the end of shift one to the end of shift three, indicting an accumulation of fatigue in these metrics with increasing number of shifts worked. Muscle function variables responded early in the duty cycle, showing a significant decline after a single work shift, but did no further decline by the end of the third shift. These findings use objective measures to substantiate that fatigue impairments occur from working a single 12 hour shift, and in several instances, increase further with more successive work shifts. Caution should be employed by personnel and administrators with work schedules involving multiple compressed 12 hour shifts. Fatigue management strategies may be used to improve risks and consequences from fatigue-related mishaps, and this study reports several variables that appear sensitive to identifying and tracking fatigue in this population.
| S-EPMC6366767 | biostudies-literature
Project description:Objectives12-hour shifts worked by nurses on acute hospital wards have been associated with increased rates of missed care reported by nurses. This study aimed to measure the association between nurses working shifts of at least 12 hours and an objective measure of missed care: vital signs observations taken on time according to an acuity-based surveillance protocol.DesignA retrospective observational study using routinely collected data from March 2012 to March 2015.Setting32 general inpatient wards at a large acute hospital in England.Participants658 628 nursing shifts nested in 24 069 ward days.Outcome measuresThe rate of daily delayed and missed vital signs observations. We focused on situations where vital signs observations were required at least every 4 hours and measured the number of instances where observations were delayed or missed, per 24-hour period. For each ward and each day, shift patterns were characterised in terms of proportion of care hours per patient day deriving from 'long' shifts (≥12 hours) for both registered nurses and healthcare assistants.ResultsOn 99 043 occasions (53%), observations were significantly delayed, and on 81 568 occasions (44%), observations were missed. Observations were more likely to be delayed when a higher proportion of the hours worked by healthcare assistants were part of long shifts (IRR=1.05; 95% CI 1.00 to 1.10). No significant association was found in relation to the proportion of hours registered nurses worked as long shifts.ConclusionOn days when a higher proportion of hours worked by healthcare assistants are from long shifts, the risk of delaying vital signs observations is higher, suggesting lower job performance. While longer shifts are thought to require fewer staff resources to maintain nurse-to-patient ratios, any benefits may be lost if staff become less productive.
| S-EPMC6361343 | biostudies-literature
Project description:ObjectiveSepsis remains a costly and prevalent syndrome in hospitals; however, machine learning systems can increase timely sepsis detection using electronic health records. This study validates a gradient boosted ensemble machine learning tool for sepsis detection and prediction, and compares its performance to existing methods.Materials and methodsRetrospective data was drawn from databases at the University of California, San Francisco (UCSF) Medical Center and the Beth Israel Deaconess Medical Center (BIDMC). Adult patient encounters without sepsis on admission, and with at least one recording of each of six vital signs (SpO2, heart rate, respiratory rate, temperature, systolic and diastolic blood pressure) were included. We compared the performance of the machine learning algorithm (MLA) to that of commonly used scoring systems. Area under the receiver operating characteristic (AUROC) curve was our primary measure of accuracy. MLA performance was measured at sepsis onset, and at 24 and 48 h prior to sepsis onset.ResultsThe MLA achieved an AUROC of 0.88, 0.84, and 0.83 for sepsis onset and 24 and 48 h prior to onset, respectively. These values were superior to those of SIRS (0.66), MEWS (0.61), SOFA (0.72), and qSOFA (0.60) at time of onset. When trained on UCSF data and tested on BIDMC data, sepsis onset AUROC was 0.89.Discussion and conclusionThe MLA predicts sepsis up to 48 h in advance and identifies sepsis onset more accurately than commonly used tools, maintaining high performance for sepsis detection when trained and tested on separate datasets.
| S-EPMC6556419 | biostudies-literature
Project description:BackgroundThe ability to accurately forecast census counts in hospital departments has considerable implications for hospital resource allocation. In recent years several different methods have been proposed forecasting census counts, however many of these approaches do not use available patient-specific information.MethodsIn this paper we present an ensemble-based methodology for forecasting the census under a framework that simultaneously incorporates both (i) arrival trends over time and (ii) patient-specific baseline and time-varying information. The proposed model for predicting census has three components, namely: current census count, number of daily arrivals and number of daily departures. To model the number of daily arrivals, we use a seasonality adjusted Poisson Autoregressive (PAR) model where the parameter estimates are obtained via conditional maximum likelihood. The number of daily departures is predicted by modeling the probability of departure from the census using logistic regression models that are adjusted for the amount of time spent in the census and incorporate both patient-specific baseline and time varying patient-specific covariate information. We illustrate our approach using neonatal intensive care unit (NICU) data collected at Women & Infants Hospital, Providence RI, which consists of 1001 consecutive NICU admissions between April 1st 2008 and March 31st 2009.ResultsOur results demonstrate statistically significant improved prediction accuracy for 3, 5, and 7 day census forecasts and increased precision of our forecasting model compared to a forecasting approach that ignores patient-specific information.ConclusionsForecasting models that utilize patient-specific baseline and time-varying information make the most of data typically available and have the capacity to substantially improve census forecasts.
| S-EPMC3680345 | biostudies-literature
Project description:Genomic prediction is an effective way for predicting complex traits, and it is becoming more essential in horticultural crop breeding. In this study, we applied genomic prediction in the breeding of cucumber plants. Eighty-one cucumber inbred lines were genotyped and 16,662 markers were identified to represent the genetic background of cucumber. Two populations, namely, diallel cross population and North Carolina II population, having 268 combinations in total were constructed from 81 inbred lines. Twelve cucumber commercial traits of these two populations in autumn 2018, spring 2019, and spring 2020 were collected for model training. General combining ability (GCA) models under five-fold cross-validation and cross-population validation were applied to model validation. Finally, the GCA performance of 81 inbred lines was estimated. Our results showed that the predictive ability for 12 traits ranged from 0.38 to 0.95 under the cross-validation strategy and ranged from -0.38 to 0.88 under the cross-population strategy. Besides, GCA models containing non-additive effects had significantly better performance than the pure additive GCA model for most of the investigated traits. Furthermore, there were a relatively higher proportion of additive-by-additive genetic variance components estimated by the full GCA model, especially for lower heritability traits, but the proportion of dominant genetic variance components was relatively small and stable. Our findings concluded that a genomic prediction protocol based on the GCA model theoretical framework can be applied to cucumber breeding, and it can also provide a reference for the single-cross breeding system of other crops.
| S-EPMC8421847 | biostudies-literature
Project description:BackgroundFoodborne diseases, as a type of disease with a high global incidence, place a heavy burden on public health and social economy. Foodborne pathogens, as the main factor of foodborne diseases, play an important role in the treatment and prevention of foodborne diseases; however, foodborne diseases caused by different pathogens lack specificity in clinical features, and there is a low proportion of clinically actual pathogen detection in real life.ObjectiveWe aimed to analyze foodborne disease case data, select appropriate features based on analysis results, and use machine learning methods to classify foodborne disease pathogens to predict foodborne disease pathogens that have not been tested.MethodsWe extracted features such as space, time, and exposed food from foodborne disease case data and analyzed the relationship between these features and the foodborne disease pathogens using a variety of machine learning methods to classify foodborne disease pathogens. We compared the results of 4 models to obtain the pathogen prediction model with the highest accuracy.ResultsThe gradient boost decision tree model obtained the highest accuracy, with accuracy approaching 69% in identifying 4 pathogens including Salmonella, Norovirus, Escherichia coli, and Vibrio parahaemolyticus. By evaluating the importance of features such as time of illness, geographical longitude and latitude, and diarrhea frequency, we found that they play important roles in classifying the foodborne disease pathogens.ConclusionsData analysis can reflect the distribution of some features of foodborne diseases and the relationship among the features. The classification of pathogens based on the analysis results and machine learning methods can provide beneficial support for clinical auxiliary diagnosis and treatment of foodborne diseases.
| S-EPMC7872834 | biostudies-literature
Project description:ObjectiveTo examine the effect of census caps and unit-based admissions on resident workload, conference attendance, duty hour compliance, and patient safety.Participants and methodsWe implemented a census cap of 14 patients on 6 Mayo Clinic internal medicine resident hospital services and a unit-based admissions process in which patients and care teams were consolidated within hospital units. All 280 residents and 15,926 patient admissions to resident and nonresident services 1 year before the intervention (September 1, 2006, through August 31, 2007) and 1 year after the intervention (May 1, 2008, through April 30, 2009) were included. Residents' workload, conference attendance, and duty hours were tracked electronically. Patient safety variables including Rapid Response Team and cardiopulmonary resuscitation events, intensive care unit transfers, Patient Safety Indicators, and 30-day readmissions were compared preintervention and postintervention.ResultsAfter the intervention, residents' mean (SE) ratings of workload appropriateness improved (3.10 [0.08] vs 3.87 [0.08] on a 5-point scale; P<.001), as did conference attendance (1523 [56. 8%] vs 1700 [63.5%] conferences attended; P<.001). Duty hour violations for working more than 30 consecutive hours and not having 10 hours off between duty periods decreased from 77 of 9490 possible violations (0.81%) to 27 (0.28%) and from 70 (0.74%) to 14 (0.15%) violations, respectively (both, P<.001). Thirty-day readmissions to resident services decreased (1010 [18.14%] vs 682 [15. 37%]; P<.001). All other patient safety measures remained unchanged. After adjustment for illness severity, there were no significant differences in patient outcomes between resident and nonresident services.ConclusionCensus caps and unit-based admissions were associated with improvements in resident workload, conference attendance, duty hour compliance, and readmission rates while patient outcomes were maintained.
| S-EPMC3538463 | biostudies-literature
Project description:Interventions: None.
Primary outcome(s): The aim of the study is to create a serum CEA-measurement based algorithm of CEA rising. If the algorithm proves higher sensitivity of detecting recurrence, and therefore higher curable treatment, it can result in an application which helps professionals in health care, mainly general practitioners, with decision making. One should think of decisions like closer monitoring of CEA measurements, additional imaging based on serum CEA measurements (CEA-triggered imaging”) or referring to a specialist
Study Design: Non-randomized controlled trial, Open (masking not used), N/A , unknown, Other
| 2442417 | ecrin-mdr-crc
Project description:BackgroundThe detection of wheezes as an exacerbation sign is important in certain respiratory diseases. However, few highly accurate clinical methods are available for automatic detection of wheezes in children. This study aimed to develop a wheeze detection algorithm for practical implementation in children.MethodsA wheeze recognition algorithm was developed based on wheezes features following the Computerized Respiratory Sound Analysis guidelines. Wheezes can be detected by auscultation with a stethoscope and using an automatic computerized lung sound analysis. Lung sounds were recorded for 30 s in 214 children aged 2 months to 12 years and 11 months in a pediatric consultation room. Files containing recorded lung sounds were assessed by two specialist physicians and divided into two groups: 65 were designated as "wheeze" files, and 149 were designated as "no-wheeze" files. All lung sound judgments were agreed between two specialist physicians. We compared wheeze recognition between the specialist physicians and using the wheeze recognition algorithm and calculated the sensitivity, specificity, positive predictive value, and negative predictive value for all recorded sound files to evaluate the influence of age on the wheeze detection sensitivity.ResultsThe detection of wheezes was not influenced by age. In all files, wheezes were differentiated from noise using the wheeze recognition algorithm. The sensitivity, specificity, positive predictive value, and negative predictive value of the wheeze recognition algorithm were 100%, 95.7%, 90.3%, and 100%, respectively.ConclusionsThe wheeze recognition algorithm could identify wheezes in sound files and therefore may be useful in the practical implementation of respiratory illness management at home using properly developed devices.
| S-EPMC7544038 | biostudies-literature
Project description:Despite recent efforts to develop automated protein structure determination protocols, structural genomics projects are slow in generating fold assignments for complete proteomes, and spatial structures remain unknown for many protein families. Alternative cheap and fast methods to assign folds using prediction algorithms continue to provide valuable structural information for many proteins. The development of high-quality prediction methods has been boosted in the last years by objective community-wide assessment experiments. This paper gives an overview of the currently available practical approaches to protein structure prediction capable of generating accurate fold assignment. Recent advances in assessment of the prediction quality are also discussed.
| S-EPMC1074308 | biostudies-literature