Contribute to project narrative here

Introduction

Given evidence of potentially severe unintended consequences upon system/ technology implementation in health care settings, adequate analysis of workflow and exploration of potential unintended consequences is essential prior to implementation of any system or technology. 1 However, standard analysis is limited to observation of pre-implementation workflow. Additionally, systems and technologies chosen for broad implementation in tertiary care facilities may not fit specialized settings such as emergency departments. Agent-based modeling enables exploration of system behavior over a broad range of parameters, and so may provide crucial pre-implementation insight into the impact of a new system/technology on workflow and/or outcomes.

Purpose To explore via agent based modeling, the affect of the number and placement of provider computer work stations on emergency department workflow and related patient.

Specific Aims

Specific Aim 1: Develop an agent-based model simulating patient flow and provider workflow in an emergency department setting, relative to the number and placement of terminals in an emergency department.

Specific Aim 2: Through agent-based modeling experiments, characterize the behavior of the emergency care delivery system relative to the number and placement of workstations over a wide range of parameters.

Background and Significance

Adequate analysis and consideration of workflow in relation to new systems/ technologies is essential in health care settings. A 2005 study by Han et al found increased mortality after implementation of CPOE (computerized provider order entry) in the Pittsburgh Children’s Hospital emergency department and PICU (pediatric intensive care unit). 1 A subsequent study by DelBaccoro and colleagues yielded contradictory findings, showing no change in mortality rate in a PICU. 2 Experts attributed these contradictory findings to substantial qualitative differences in the process of CPOE implementation in the two settings. 3 In the case of Pittsburgh Children’s Hospital, multiple unintended consequences resulting from inadequate pre-implementation analysis and planning led to delays in care. One of the unintended consequences was a severe mismatch between workflow imposed by CPOE and existent workflow in the care setting. In essence, the unique characteristics of workflow were not adequately considered relative to the system/ technology, prior to implementation.

In the Han et al study, the emergency department encountered multiple unintended consequences of CPOE implementation. Unique characteristics of emergency department settings, relative to most other care settings in the hospital, include high patient acuity, rapid turnover of patients, and high intensity of nursing and medical care. Unexpected alterations in workflow related to the introduction of technology/ systems in these settings have demonstrated potential to adversely affect patient outcomes. 1 Additionally, expected alterations in workflow caused by the introduction of technologies/systems may have more severe effects in the emergency department, due to high patient acuity coupled with the necessity of rapid, highly coordinated care. Clearly, a careful consideration of potential unintended consequences in the specialized emergency department setting is desirable pre-implementation, in order to inform design and implementation processes.

One basic design/ implementation decision, particularly in the case of electronic health record (EHR) and CPOE systems, is the number and placement of computer terminals. Health care providers, generally nurses or physicians, must physically access a terminal to access information (i.e. health history, recent vital signs) or to input information (i.e. documentation, orders). Poissant and colleagues (2005) reviewed studies of time efficiency in EHR documentation. Their comparisons of PDA, bedside, and central terminal placement evidence no clear advantage to any particular distribution of terminals. 4 The comparison was complicated by the time point at which data was collected in the individual studies. Some studies collected data during the first three months post-implementation, when there is a generally observed increase in documentation time, while others collected data at later timepoints. Also, the studies differ dramatically in setting. As a result, there is no clear empiric evidence to guide optimal number and placement of terminals.

Theoretical Framework Systems Development Life Cycle (SDLC)(10) is a software development process comprised of multipel steps including: 1) Identifying and defining a need for the new system, 2) Analyzing the information needs of the end users, 3) Creating a blueprint for the design including specifications related to hardware, software, human and data resources 4) Coding and debugging the system and 5) System testing-which involves the evaluation of the system's actual versus intended functionality. Our study addresses step 5, System Testing. We employ an agent-based modeling framework to test the functionality of varying numbers and placements of CPOE workstations in an emergency department setting.

Methods

Methods for Specific Aim 1:

Develop an agent-based model simulating patient flow and provider workflow in an emergency department setting, relative to the number and placement of terminals in an emergency department.

Agent-based modeling. Agent based models are increasingly acknowledged as powerful tools in studying social systems and decision-making. 5 Agent based models are useful for revealing emergent properties of systems by allowing individual agents within the system to interact and respond heterogeneously to different information, different scenarios and different decision rules. 5 The models show properties of the overall system that are not properties of the individual agents themselves. 6 The flexible, adaptive, dynamic and heterogeneous properties of the agents allow for analysis in simulations that cannot be carried out with traditional methods. 7 The usefulness for agent based models to study flow patterns and organizational design has been noted. 8

Creation of Agent-Based Model (ABM)

Software: All programs were written in NetLogo version 4.0Beta. NetLogo is an open source software package which provides a computing environment which facilitates the modeling and visualization of complex natural and social phenomena. NetLogo allows for the creation of individual interacting agents to whom rules and behaviors can be assigned. These agents move within a 1-dimentional environment (represented as patches in NetLogo). Emergent macro-level phenomena resulting from agent interactions with each other and the environment can then be explored.

The Basic Model

Agents: Providers: Provider agents were assigned the overall goal of checking for new patients, treating patients, discharging patients, checking in at the nurses station, and entering and/or retrieving information from a CPOE which was currently not in use. Provider agents vary in number present in the emergency department environment (0-10), time waiting for patient (distributed either random uniform or random normal), minutes spent with each patient (0-60), time at CPOE (0-20). Physician Agents were required to visit the terminal after each (k) patient (s).

Patients: Patient agents present in the waiting room. Patient agents vary in acuity (non-urgent, urgent, emergency status) whether they have been seen (yes/no), their arrival process (frequency and distribution of arrivals: either random uniform or Poisson) arrival probability (0-1) and patient queue.

Environment: CPOE workstations: Are placed in the emergency department environment for provider agent use. Terminals vary in number and spatial location within the emergency department. Nurses Station: The nursing station served as checkpoint for provider agents between patients encounters. The nursing station patch does not vary.

Waiting Room: Patients agents appeared and remained in the waiting room until a physician was available to see the patient.

Outcomes/Emergent Macro-Level phenomenon: Average patient wait time: The average patient wait time may vary by agent behavior and environmental characteristics. Average patient wait time was monitored as an emergent phenomena resulting from agent-agent, agent environment interactions.

Please find NetLogo code in Appendix 1.

Methods for Specific Aim 2:

Through agent-based modeling experiments, characterize the behavior of the emergency care delivery system relative to the number and placement of workstations over a wide range of parameters.

Initial conditions: ______________________

ABM Experiments

The effects on changes in system characteristics over time on emergent phenomena are explored by conducting experiments in NetLogo. Time in NetLogo is represented by ticks. For the current model, 25 ticks represent 1 minute. The following experiments were run for a minimum of ____ ticks (___ minutes) or until patterns of emergent phenomena in question were identified.

The number of CPOE workstations was the unit of experimentation. CPOE workstations were configured as: 1) clustered, 2) distributed (linearly) across the emergency department and 3) at the patient bedside. Please see Figures 1, 2 and 3 for visual of terminal formation. Emergency department workflow was investigated under each CPOE workstation configuration. Specifically, under each CPOE configuration, we investigated the dependence of patient wait time on settings of X parameters:

1) Number of providers present in the emergency department 2) Patient arrival distribution 3) Time spent with each patient 4) Time providers spent at CPOE 5) Patient acuity 6) Length of patient queue 7) More?

All data was exported to an ASCII file for additional statistical analysis

Statistical Analysis of AMB experiments:

Plots depicting the trends and relationships between ranges of parameters representing emergency department workflow and patient will be created for each CPOE configuration. SAS© statistical analysis software was employed to describe and visualize results of ABM experiments.

Results

I. Workflow behavior under CPOE clustered formation II. Workflow behavior under CPOE distributed formation III. Workflow behavior under CPOE patient bedside formation

Discussion

References

1. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics. Dec 2005;116(6):1506-1512.
2. Del Beccaro MA, Jeffries HE, Eisenberg MA, Harry ED. Computerized provider order entry implementation: no association with increased mortality rates in an intensive care unit. Pediatrics. Jul 2006;118(1):290-295.
3. Ammenwerth E, Talmon J, Ash JS, et al. Impact of CPOE on mortality rates--contradictory findings, important messages. Methods Inf Med. 2006;45(6):586-593.
4. Poissant L, Pereira J, Tamblyn R, Kawasumi Y. The impact of electronic health records on time efficiency of physicians and nurses: a systematic review. J Am Med Inform Assoc. Sep-Oct 2005;12(5):505-516.

--Mpoynton 12:05, 21 June 2007 (MDT)

1. Lempert R. Agent-based modelling as organizational and public policy simulators. Proc Natl Acad Sci

U S A. 2002;99(suppl 3):7195–7196.

1. Flake GW. The Computational Beauty of Nature.

Cambridge, Mass: MIT Press, 1998.

1. Miller JH and Page SE. Complex Adaptive Systems - An introduction to computational models of social life. Princeton, NJ: Princeton University Press, 2007.
2. Bonabeau E. Agent-based modeling: methods and

techniques for simulating human systems. Proc Natl Acad Sci U S A. 2002;99(suppl 3):7280–7287

1. Kushniruk A. Evaluation in the design of health information systems: application of approaches emerging from usability engineering. Comput Biol Med. 2002 May;32(3):141-9

--Rhonda 23:11, 21 June 2007 (MDT)