Hui Chen, Lei Shu, Jianying Zhang Jongwoo Song, William S. Cleveland Department of Statistics Purdue University Mummoorthy Murugesan, Yang Wang Christopher.

Slides:



Advertisements
Similar presentations
Quality control tools
Advertisements

Recent Emergency Study in Palestine Triage System At Al – Makassed Hospital Emergency Department Jamal Al-Wahadneh 2009.
Slide 1 Bayesian Model Fusion: Large-Scale Performance Modeling of Analog and Mixed- Signal Circuits by Reusing Early-Stage Data Fa Wang*, Wangyang Zhang*,
Peterson-Kaiser Health System Tracker How do health expenditures vary across the population?
Energy-Efficient Computing for Wildlife Tracking: Design Tradeoffs and Early Experiences with ZebraNet Presented by Eric Arnaud Makita
September 2000Department of Statistics Kansas State University 1 Statistics and Design of Experiments: Role in Research George A. Milliken, PhD Department.
Health Outcomes Research and Policy Center Joseph Thomas III, M.S., Ph.D., FAPhA.
1 Electrical and Computer Engineering Drebin Rescuing Firefighters in Distress FPR Team Ganz: Jonathan Bruso Michael Carney Daniel Fortin James Schafer.
How Do I Evaluate Workflow?
Worm Origin Identification Using Random Moonwalks Yinglian Xie, V. Sekar, D. A. Maltz, M. K. Reiter, Hui Zhang 2005 IEEE Symposium on Security and Privacy.
Introduction to Control Charts.
ANALYZING STORAGE SYSTEM WORKLOADS Paul G. Sikalinda, Pieter S. Kritzinger {psikalin, DNA Research Group Computer Science Department.
1 Graphical Models in Data Assimilation Problems Alexander Ihler UC Irvine Collaborators: Sergey Kirshner Andrew Robertson Padhraic Smyth.
Problem Solving Tools INSY 3021 Auburn University Spring 2008.
Results 2 (cont’d) c) Long term observational data on the duration of effective response Observational data on n=50 has EVSI = £867 d) Collect data on.
MEDEMAS –Medical Device Management and Maintenance System Architecture
Lean Training Standard Work. Agenda What is it? What’s it for? How does it work? When do you use it? What’s an example?
Robin McDougall, Ed Waller and Scott Nokleby Faculties of Engineering & Applied Science and Energy Systems & Nuclear Science 1.
Cost Program: Barbados Experience
Home Health Care and Assisted Living John Stankovic, Sang Son, Kamin Whitehouse A.Wood, Z. He, Y. Wu, T. Hnat, S. Lin, V. Srinivasan AlarmNet is a wireless.
New Jersey Transit Fatigue Risk Report Assignments for 27 th October December
Presenter: Shant Mandossian EFFECTIVE TESTING OF HEALTHCARE SIMULATION SOFTWARE.
Dual Prediction-based Reporting for Object Tracking Sensor Networks Yingqi Xu, Julian Winter, Wang-Chien Lee Department of Computer Science and Engineering,
Medical Statistics (full English class) Ji-Qian Fang School of Public Health Sun Yat-Sen University.
Verification & Validation
Kaz Sobczak 1, Joyce Nacario 2, Ho Lom Lee 3 1 University of California Medical Center, San Francisco, CA (USA) 2 University of California Medical Center,
FREQUENCY ANALYSIS.
ALIP: Automatic Linguistic Indexing of Pictures Jia Li The Pennsylvania State University.
The Complete Medication Safety and Policy Solution for Care Homes.
Clinical Decision Support Systems Paula Coe MSN, RN, NEA-BC NUR 705 Informatics and Technology for Improving Outcomes in Advanced Practice Nursing Dr.
 Polar Move  Polar Activity measurement  E600  Computer  PE Manager Software.
Problem Refinement and Possible Solutions Too many different devices are used to input data  Team: Grace (CS) Ran (ID) Busra (HS) Musheer(CS) Class: COA-8823.
September 18-19, 2006 – Denver, Colorado Sponsored by the U.S. Department of Housing and Urban Development Conducting and interpreting multivariate analyses.
CCR = Connectivity Residue Ratio = Pr. [ node pair connected by an edge are together in a common page on computer disk drive.] “U of M Scientists were.
Results Background This quality improvement study objectively quantified time spent on tasks for physician extender staff. Physician extender types included.
Department of Patient RelationsMeasuring to Achieve Patient Safety Safety Observer’s Orientation.
Regulations 201: Thorny Issues What is Research? Exempt and Expedited Reviews.
The Practice of Statistics, 5th Edition Starnes, Tabor, Yates, Moore Bedford Freeman Worth Publishers CHAPTER 12 More About Regression 12.1 Inference for.
Title : How The Mind Works (Ergonomic-Stress in Nursing)
Status Reports: Measuring against Mission National Institute of Standards and Technology U.S. Department of Commerce 1 Technology Program Evaluation: Methodologies.
Olubukola Olanrewaju Maria Popel Karley Mullikin David Ramnarine University of Tampa EME 644.
Emergency Medical Services Department 2014 Budget Overview.
The Practice of Statistics, 5th Edition Starnes, Tabor, Yates, Moore Bedford Freeman Worth Publishers CHAPTER 12 More About Regression 12.1 Inference for.
Physiological Data Analysis of Neuro-Critical Patients Using Markov Models By Shashwat Bhoop sb3758.
Privacy Vulnerability of Published Anonymous Mobility Traces Chris Y. T. Ma, David K. Y. Yau, Nung Kwan Yip (Purdue University) Nageswara S. V. Rao (Oak.
EMT/ Paramedic 8.1 Research Paramedic as a career.
T Relationships do matter: Understanding how nurse-physician relationships can impact patient care outcomes Sandra L. Siedlecki PhD RN CNS.
Electronic Information Management
8 Principles of Effective Documentation.
Governing Body QAPI 2013 Update for ASC
Overview Modern chip designs have multiple IP components with different process, voltage, temperature sensitivities Optimizing mix to different customer.
DATA COLLECTION METHODS IN NURSING RESEARCH
Final Presentation Presented By: IOE 481 Team #3
Exploring Nursing Work in the Emergency Department
How do health expenditures vary across the population?
Mobile Computing for Healthcare
Worm Origin Identification Using Random Moonwalks
Confidence Intervals.
Bringing Pharmaceutical Care to the Child’s Bedside
Dr. Cleveland, This file contents 4 tracks poster slides
Critical Path Method Farrokh Alemi, Ph.D.
CHAPTER 12 More About Regression
Summer Training Industrial Engineering Department
How Do I Evaluate Workflow?
How do health expenditures vary across the population?
ILLINOIS Visualizing Graphs Distributed Across Multiple Processes
Human Factors Assessment of the Emergency Department
Module 6 Part 3 Choosing the Correct Type of Control Chart Limits
MECH 3550 : Simulation & Visualization
Presentation transcript:

Hui Chen, Lei Shu, Jianying Zhang Jongwoo Song, William S. Cleveland Department of Statistics Purdue University Mummoorthy Murugesan, Yang Wang Christopher W. Clifton Department of Computer Science Purdue University  During catastrophic events, medical system should be able to provide timely and needed care to a sudden increase of hundreds of patients at all levels and by all means. This is often hindered by the limitation of available resources. Therefore efficient planning/management of hospital resources will become very critical.  Our study aims to better understand the nurse workload and work environment by learning about how nurses spend their time and energy, and how they travel on/off the shifts. This knowledge will enable improvements in hospital design, policies, and procedures that improve the nurse working efficiency and making best use of the nurse resources, resulting in improved healthcare, better job satisfaction, and more effective use of highly skilled healthcare professionals.  Results will be useful in planning out a hospital layout in the wake of a massive emergency situation, given the availability of nurses, and the expected number of patients to be treated. Track A: Documentation Time Track B: Nurse Work Sampling. Wireless Receivers Data downloaded to laptop Data: checked for Quality and loaded into Oracle DB R Objects generationR Statistical Software Graphs and Reports Secure data transfer to 24x7 Purdue Server Medical Management for Catastrophic Events Nurse Time, Motion, and Physiologic Response Project Study Methodology Schematic Preliminary Result  “Meds Paperwork”, “Assessment”, “Other” are the most frequent documentation activities, which also have the largest variations among the units. Other 5 documentation activities are relatively less frequent with smaller variations.  Durations for different documentation activities follow log-normal distributions with different parameters.  There is a “time of day” effect on duration, but no “day of week” effect. Future Work The posterior means of the Bayesian models will provide estimates of the activity probabilities and the parameters of the duration marginal distributions, and the overall posterior distribution will provide information about the uncertainty of the estimates. Goal Provide a characterization of the activity probabilities for each unit. Methodology  An algorithm was developed to assign the shift number to each nurse.  Visualization tools were developed to:  Study activity variability across units, nurses, shifts and time.  Identify the distribution of the activity-duration.  Show the dependency of the stochastic activity- state/activity-duration process  Bayesian model will be built up to get the characterization. Track A has each nurse record all documentation activities-the beginning time, ending time, and the category of a documentation activity. Background and Goals Sponsored by Ascension Health and Kaiser Permanente Track D: BodyMedia Armband. Track C: Nurses carry locating RFID tags

Track B has each nurse carry a PDA that vibrates randomly during the day. When the PDA vibrates, the nurse enters the location and the activity category in the PDA. Goal Provide a characterization of the activity parameters for each unit – the probabilities of the 76 activities and how they relate to the unit environmental variables. Methodology  Visualization tools were developed to Study:  Activity variability across nurses, time, and units  Determine the exact form of the model  Model checking  Bayesian model with hierarchical priors will be built up to provide the characterization Preliminary Result  Nurses work in the Nurse Station and Patient Room most often, then on the unit and off the unit.  Direct Care, Indirect Care and Documentation are the most significant activity groups compared to Personal, Other, Waste, Administrative.  60% of the nurses' activities are value added. 17% are non-value added. 23% are necessary. Future Work The posterior means of the Bayesian model will provide estimates of the activity probabilities and the overall posterior distribution will provide information about the uncertainty of the estimates. Goal  Learn how nurses travel during their shifts, such as heavily traveled paths, transition probabilities, and total travel distance.  Learn how nurses spend their time at different locations, such as the nurse station and the patient room.  Learn about the relationship between nurse motion and travel distance and unit environmental variables such as the architectural layout Methodology  An algorithm was developed to determine the location and stay duration of a nurse by studying reports from the four RFID tags  Visualization tools (trellis display) were developed to display the large volume of spatial data and for data exploration.  Animation tools were used to show the movement of a nurse vividly. Preliminary Result  Nurses spend most of their time (40-60%) at the nurse station, 17-37% in the patient room (see the first figure).  Most frequently traveled area is the nurse station and the surrounding area. Future Work Statistical Bayesian models will be built to characterize the nurse motion, such as frequent path patterns and the travel distance. This model will be based on the architectural layout of locations. Track C uses Indoor Positioning System (IPS) to collect nurse motion data. Nurses carry four tracking tags (RFID) that report their locations and timestamps regularly. Colored nodes are IPS receivers. Green nodes are for hallway receivers, pink for rooms, mainly, patient rooms, blue for nurse stations. A Link between two nodes represents the path from one node to another. The two numbers on each edge are the frequencies that the nurse travels from one node to another. Interesting Result  The travel distance during a specific nurse shift can vary from almost 0 km to 11km. (Some nurses walk really a lot!)  The walking speed which is measuring the motion intensity of nurse varies across hospitals, the hospital which has the smallest travel speed happens to be the circle design!  On average, the nurse’s heart rate beating per minute in their working duration is about 72~90.  On average, the nurses’ energy expenditure in one minute is about 1~3 kcal, but if look at the total energy spent on a shift, the maximum can be 2500kcal! Track D has volunteer nurses from any track wear the Bodymedia armband to measure the physical impact of workload and stress on the nurses. The Bodymedia armband measures the physiological variables, such as kilo calories burnt per minute, steps taken per minute, etc. Goal  Evaluate and compare nurse working intensity and nurse motion intensity on/off the shift  Compare nurse’s working intensity and motion intensity among hospitals which could be related to hospital’s design. Methodology Created plots for visualizing data and based on the plots, built statistical model.