Friday, June 22, 2007

EMR and Device Integration

Biomedical/Bedside/ICU Device Integration
In the words of the guru Tim Gee - Medical Connectivity Consulting "Medical device integration is a critical (and an often overlooked) part of EMR planning. To be successful, any plan must take into account many more considerations beyond getting an HL7 (or CCR) feed into the EMR. Multiple stakeholders including nursing and clinical/biomedical engineering must be engaged. Putting together a successful long term plan requires negotiations across traditional hospital silos and an in depth understanding of point-of-care workflows, medical device connectivity and device vendor offerings and product strategies".

The benefits of automatic data collection (heart rate, invasive/non-invasive blood pressure, respiration rate, oxygen saturation, blood glucose, etc.) from acute care monitoring devices have become so obvious that all hospitals now require that their clinical information system (CIS), anesthesia information management system (AIMS), electronic medical records (EMR), electronic patient record system (EPR or PHR), or other hospital/healthcare information system (HIS) provide interfacing capabilities to biomedical devices - in order to ensure that key vital signs are stored in the Centralized Data Repository (CDR) - to track patient progress over time.

Here is the HIMSS guide to the EMR Adoption Model.

Patient monitoring systems are among the first to be integrated; because each HIS requires at least basic patient vital sign collection. Integration with anesthesia devices is a must for any AIMS. Data collection from ventilator systems is required in most cases for ICU systems. Infusion device data integration is becoming increasingly requested in cases where CPOE systems are implemented.

But connecting to bedside medical devices and collecting data in your CIS or EPR is not as simple as it may seem. Device interface development is a specialized task that consumes resources and diverts attention away from core competencies. Competitive issues make obtaining device protocols difficult and sometimes impossible. Incomplete connectivity results in frustration and decreased efficiency of the hospital.

The various questions you need to have when integrating devices with a HIS are as below:
Categories of Medical Devices for Integration:

  • Vital Signs or Diagnostics devices
  • Infusion Pumps
  • Dialysis devices
  • Anesthesia machines
  • EKG and EEG devices
  • Endoscopy devices
  • Glucometers and Lab Devices
  • Urimeters
  • Bedside devices
  • Oximeters with Patient Monitoring and Alarm Systems
  • Ventilators
  • Ultrasound devices
  • Stress testing devices
  • Home care/monitoring devices

Type of Device Connectivity to the HIS
  • Wireless/Mobile
  • Fixed/"Wired Network"
Format of Message feed from Device(s) to the HIS
  • HL7(or CCR) format result messages with possible Images links, etc across TCP/IP
  • Proprietary format messages across TCP/IP
  • Binary format data across USB or others
Format of Message feed to Device(s) from the HIS
  • HL7(or CCR) format ADT messages across TCP/IP
  • Proprietary format messages across TCP/IP
  • Binary format data across USB or others
Frequency and Location of Device Data Feed to the HIS
  • Continuous (Periodic) Real-time – 1 message per minute
  • Manual (Aperiodic) or on-demand
  • Server-based - with storage for real-time data and polling options
  • Location: ICU or PACU or sleep clinic
  • Timing Synchronization among all the connected systems is important
Grouping of Device Data in the HIS is based on:
  • Patient Chart sections
  • Department needs and Security Roles
  • Common Device Parameters
  • Dependent Device Parameters
  • Device Monitoring and Asset Tracking
  • Display and Storage of the data - claims, clinical encounters, medication/pharmacy orders-results, lab orders-results and images - captured and mapped to a common (XML) format, possibly ASTM’s Continuity of Care Record (CCR or CCD) [providing for inter-operability between multiple HIS and Regional EMR systems].
Security Issues:
  • Caregivers need access to validate device data onto the patient chart at point of care
  • Unique Patient Identification (EMPI).
  • Audit trail, secure error handling and enterprise timestamps on device data.
  • High speed secure network with firewalls to protect ePHI
  • FDA guidelines compliance
  • HIPAA guidelines compliance
  • JCAHO guidelines compliance
  • SSO and CCOW support 
  • Legal guidelines compliance
  • Anonymization or De-Identification for research patient data
Examples:
  • Vital Signs mobile devices feed patient data to the EMR and a senior RN can review results before they are attached to the patients’ charts.
  • Infusion Pumps drug/fluid delivery tracking in EMR for long term critical care.
  • Enabling medical devices, such as infusion pumps, ECG machines and Glucometers, to wirelessly send data from the ICU/system to a patient’s medical record or to a physician
  • Home care and chronic disease monitoring systems that provide patient feedback, patient monitoring and alerts (to both patients and physicians) to the EMR.
Software for Device Integration with the HIS:
  • Capsule Technologie’ DataCaptor is a generic, third-party software + hardware suite that provides the most complete biomedical device connectivity solution available on the market. DataCaptor has the largest library of supported devices - more than 250 diverse bedside devices, advanced features, and easy integration with hospital information systems.
  • Stinger Medical - Integriti - provides a secure and mobile method of transmitting patient vital signs wirelessly to the EMR.
  • Current Capsule Technologie - DataCaptor - OEM partners include (among other HIS vendors of all sizes)

Benefits of Device Integration:
As in several hospitals; the reasons for integrating devices is to automate the flow of data and interface it to the HIS application:
  • To reduce transcription/documentation errors. Currently, nurses manually transcribe the data from scratch pads or from the devices onto the patient report resulting in problems like indecipherable handwriting, data in the wrong chart, vital signs written on scraps of paper (hands, scrub suits, etc.) that get forgotten, and then there is sometimes considerable lag between readings and reporting.
  • To decrease documentation time. Significant increases in productivity can be gained by an interface that allows the nurse to validate rather than enter the data.
  • To support quality data collection (charts, images, vitals) and to provide increased surveillance for critical patients - even when the care-provider is not present at the bedside. This allows for safe collection of data over time, thus providing a more accurate and valid history of patient progress.
  • To increase patient safety. Safety is enhanced by decreasing data entry errors, and by allowing the nurse to review data collected when he/she is not present at the bedside. In addition the data can be captured at an increased frequency creating a more accurate depiction of the patient’s condition.
  • To enable research and quality control. Data can be collected for future analysis by de-identifying or anonymizing patient demographics. 
  • To provide better patient care and more physician – patient contact time. A silent factor of a hospital's revenue is quality of patient care. One of the chief drivers of quality of patient care is the quality of information provided efficiently to the Physicians though which they can make those critical decisions.
  • To securely and quickly share assessment, diagnosis, treatment and patient data across facilities/HIE's or Health Care Information Exchanges/RHIO (regions)/states thereby enabling the patient to be provided the best care anywhere.
  • To reduce patient, physician and nurse stress and legal issues.
  • To provide complete and comprehensive data on patient charts.
  • To enable future devices to seamlessly connect to the existing EMR.
  • To prevent errors in diagnosis, prescription and medication, by basing decisions on the entire patient history/allergies, the latest medications and the latest technology that is available to the patient and the care provider.
  • Clinical (or Diagnostic) Decision Support Systems [CDSS] and Best Practice systems are more effective with comprehensive and secure digital files (historical patient charts).
  • To provide for Business Intelligence searches/reports.
  • To increase security and prevent tampering of Patient Records - since all data is digital and secured via layers of Role based security, by HIPAA and by Digital laws - the security is much more comprehensive than a system with voluminous paper records and difficult audit trails.
  • Finally, to improve overall hospital throughput and patient hospital-visit time, success ratios and Improving Patient Efficiency Throughput.
I’ve linked the Capsule Technologie-DataCaptor web site below to show their method of ensuring data flow between DataCaptor (the server), Concentrator (the ‘router’ or Terminal box), the bedside devices and the HIS and other systems.
Note: This article is based on personal experiences and public information gathered from websites including Medical Connectivity Consulting and Capsule Technologies and other medical device manufacturer’s web-sites. Thanks to these companies for this public information and this document is intended solely for personal reading and understanding of this technology and is not for any commercial gain.

Since PACS is a type of "Device Integrator", the following is an addition to the above article:

Radiology RIS, Cardiology CIS, PACS and HIS/EMR Integration


The PACS - Picture Archiving and Communication System - is a filmless method of communicating and storing X-rays, CT/MRI/NM scans, other radiographs, Cardiology Cath/Echo images, etc - that are verified by Radiologists or Cardiologists - after being acquired by the X-ray, CT/MRI/NM, Cath, Echo machines and other variants used in the Radiology/Cardiology Departments. Images may be acquired from a patient in slices and with 3D or 4D image reconstruction - the entire patients' full body scan may be visualized on diagnostic quality workstations. Key images, Radiology reports and low resolution non-diagnostic images are provided for viewing on any screen - securely across the internet and for viewing in the HIS/EMR. If bandwidth permits - in certain cases - entire diagnostic quality images may be viewable, securely across the internet using network-aware compression technologies.
The RIS/CIS - Radiology/Cardiology Information System - enables "Radiology/Cardiology" patient scheduling, reporting/dictation, and image tracking to ensure that the PACS and the Radiology/Cardiology modalities are effectively utilized and the patients' structured reports are immediately available.
The EMR - Electronic Medical Records System or HISHospital Information System - provide a "global" view or patient historical report folder of the patient's visits or encounters with his/her care providers. From a "Radiology/Cardiology" perspective - the EMR sends ADT/schedules/orders to a RIS/CIS and receives results including patient image links and data from the PACS (via RIS/CIS) - thus enabling access to that patients Structured Reports in a single uniform location in the EMR. Thus, images can be integrated with the radiology report, and with other patient information systems' (such as laboratory, pharmacy, cardiology, and nursing) reports, thereby providing a comprehensive folder on the patient. Charges can also be sent from these systems to the HIS or automatically done inside the HIS - from where the final comprehensive claim can be sent to the insurance provider.

Key Features of a good PACS System are:
  • Modules for comparison study of prior patient images, along with similar cases
  • Modules for Computer Aided Detection using Clinical Decision Support Systems and Key Facets
  • Excellent Data Compression Techniques ensure effective network utilization and high speed transfers of quality images to workstations and other web based viewing systems.
  • Excellent EMR Integration based on IHE Integration Profiles, standard HL7 / CCR, standard DICOM and the support for secure, high-speed access to patient images via internet
  • Standard Security Features along with audit trails and Integration with RIS/CIS and EMR security.
  • Modules for 3D and 4D reconstruction of CT slices, Image Enhancement and Quality Printing
  • Immediate availability of Images on network or CD/DVD for quick diagnosis and review by remote Radiologists/experts.
  • Excellent Short Term Storage with very low retrieval time/latencies.
  • Excellent Long Term Storage with decent retrieval time/latencies and predictable data recovery.
  • Excellent RIS/CIS Integration.
  • Extensively tested and successfully working in other hospitals for 2 years at least.
  • Multiple vendor modality Integration features.
  • Downtime plan with Disaster Recovery Support.
  • Easy Upgrade-ability of hardware/storage to ensure almost infinite storage based on hospital need
  • Support for Research Patient data Anonymization or De-Identification
  • Support for Reporting off the PACS/RIS/CIS for Business Intelligence reports and data analysis.
Now that you have (selected) the PACS and RIS/CIS systems, here is the list of questions you should have regarding integration with the EMR:

EMR and RIS/CIS/PACS Integration Issues:
  • RIS/CIS/PACS features and limitations
  • Modality support for DMWL (Digital Modality Worklist - ensuring correct patient scans at modality)
  • Key Data Mappings between the RIS, CIS, PACS and EMR (eg. Study-DateTime, PatientID, ProviderID, CPT code, procedure code, Study Status, Accession number, etc.)
  • Department Workflow changes (Types of Orders, Downtime Orders, Unsolicited Results, Billing, etc.)
  • Data being displayed in the Modality Worklist and when does this worklist get updated?
  • Historical data import, cut-off dates, access policies to legacy data, etc
  • Security, User access and integrating the PACS/RIS/CIS users with the EMR users to enable secure web access to images.
  • De-Identification or Anonymization for Research patient studies.
  • Business Intelligence reporting and Regional Patient Record Integration.
Note: The above article is based on personal experience and is not for any commercial gain.