TrakCare Solution Guide
Section 4: TrakCare System Architecture and Technology
Unlike other systems, TrakCare was built as a connected solution, and provides unparalleled interoperability with other systems. This architecture allows TrakCare to serve as the EPR while new functionality is implemented over time.
Much of the power of TrakCare derives from the fact that all of its modules share the same data model and a single high-performance database, InterSystems Caché. So TrakCare’s modules do not require integration with each other, saving time and expense. Plus, healthcare information systems built on Caché are known for their high performance and scalability.
This unified information environment provides several benefits: simplified sharing of information, faster implementation, lower costs, and easier maintenance. In addition, the InterSystems Ensemble rapid integration platform embedded in TrakCare ensures interoperability of its modules with your other applications.
TrakCare uses the browser as the client front end and requires little in the way of bandwidth or power at the desktop. The use of the browser enables the system to be used with full functionality over lowspeed connections without the need for expensive thin-client technologies. This gives ubiquitous but secure access via the Internet to appropriate information wherever and whenever it is needed.
Since Ensemble is embedded in TrakCare, TrakCare allows organizations to quickly build and deploy new solutions that leverage the functionality of existing applications, orchestrate new business processes, and integrate data from across an entire facility, outsourced service providers, and community health services.
TrakCare Connects Via the Internet to Other Systems
4.1 TrakCare Technical Features
TrakCare supports a number of technical features that differentiate it from many competitive systems, including:
- Mobile Computing offers features that differ
substantially in both capability and flexibility.
The capabilities of the application on the various
platforms will be directly related to the capabilities
of the device.
TrakCare has been developed to utilize the latest technology, with minimal or no modifications to the software required. In broad terms, it can be accessed from any device that runs Internet Explorer Version 6.0 or higher. In the acute environment, this allows pen-based tablets and laptop computers connected in a wireless network to be used to deliver patient information at the bedside. When staff work away from the acute environment, such as in a community environment, TrakCare provides remote access to patient information.
- Security allows the organization to establish its
own desired security levels, departments, and roles
in the system, and to determine which functions
and screens should be secured and at what security
level. TrakCare also provides full data security, tied
to user log-in names and associated passwords that
limit access to the database.
TrakCare allows for log-on security to be unified with Windows log-on authentication, utilizing pass-through authentication of the Windows logon. This reduces the time it takes for users to get into the application and the need to have multiple log-ons for both Windows and TrakCare.
Specific security features available include the following:
- Access to TrakCare is user-name and password protected.
- Restrictions can be placed on the actions a user can perform.
- Users can be required to change their passwords on a regular basis.
- Access to TrakCare workflows that is restricted by the role the user is assigned (in addition to password protection).
TrakCare has additional safeguards to maintain the confidentiality of the data by associating patient records with treating doctors on a “need to know” basis. Only the treating doctor, or other doctors permitted by the treating doctor, can access such records.
- Reporting provides users with extensive facilities to manipulate data in a quick and easy manner. Since organizations are dynamic environments, the information needs of clinicians vary considerably. TrakCare supports this requirement with the following features:
- Each module of TrakCare comes with a number of standard reports for operational, managerial, and statistical purposes. Users can control the selection and sort criteria of these reports. These reports can also be modified by the organization.
- Reports may be viewed on-screen, printed, or imported to other office automation products, including Microsoft Excel and Word and email systems. TrakCare can provide a seamless interface between its database and these products, and provides an interface to all Microsoft and open-system-compliant desktop packages.
- The organization can develop its own ad-hoc reports using ODBC-compatible reporting tools.
- TrakCare Analytics embeds capabilities for more informed decision-making.
- TrakCare Analytics provides real-time access to management information through reports, graphs, and dashboards. TrakCare Analytics includes predefined data models and key performance indicators for clinical and administrative data throughout the TrakCare system.
- TrakCare can also be linked to the organization’s existing business intelligence tools, allowing users to make ad-hoc queries on the TrakCare database.
- Auditing includes the following facilities:
- Audit Trails keep a record of all changes made to a patient’s record, including the date, time of change, user making the change, and details of the change.
Features include: - Configurable
- Detail of changes made to the database
- Time and date stamp
- Audit Footprint keeps a record of any user who has accessed the patient’s record even if no changes were made.
Features include: - Configurable
- Detail of who views the patient record
- Time and date stamp
- Code Table Audit keeps a record of all changes made to code table records, including the date, time of change, user making the change, and details of the change.
- Security Violations keep a record of any security violation attempt.
4.2 Application Architecture
TrakCare implementations are created with the philosophy that you must use the right tool for the job. Solutions are built using a unique Component Framework, enabling rapid Web development, extraordinary transaction processing speed, and massive scalability with minimal maintenance requirements, resulting in a high-performance, highly flexible EPR application.
The TrakCare component model enables rapid application development, a consistent user interface, a consistent and inherent model for system tailoring, and considerable reuse of system components.
TrakCare can be quickly and easily tailored to an organization’s requirements without the expense, delay, complexity, or risk of custom development. The system can optionally be tailored across some or all modules through configurable workflows and configurable views of data to best serve differing needs within organizations and groups, and even among individuals.
TrakCare provides support for a range of secure multi-organization configurations. Solutions can be configured to meet specific requirements.
Typical model configurations include:
- Multiple instances of the application on a single server
- A single database with multiple hospitals in a multi-hospital configuration
- A multi-organization configuration with independent databases and a single EPR summary
- A multi-region, multi-organization, central EPR configuration with effective replication of the EPR with the patient’s home EPR
4.3 Operating System Environments
TrakCare is certified to run on a range of commodity operating system environments, including but not limited to Microsoft Windows 2000, XP Pro, Server 2003 for x86-32, Microsoft Windows Vista for x86-32, Microsoft Windows Server 2003 for x86-64, and a range of other operating systems such as OS X, HP-UX, OpenVMS, Tru64, AIX, Red Hat, Solaris, and SUSE Linux. TrakCare can run in a mixed-platform environment.
4.4 Logical Model
TrakCare is logically designed as a multi-tier system. The components in the different tiers are responsible for different tasks and can be logically grouped based on the type of work they perform. These tiers may be physically configured across one or more servers to provide scalability and resilience.
The four logical tiers are:
- Data Tier
- Application Tier
- Web Tier/File and Print Tier
- Client Tier
TrakCare is a Multi-Tier System
4.4.1 Data Tier
The Data Tier is made up of a database containing TrakCare transaction data and database schema. Typically both a primary and a secondary database server are active at all times. The primary database server is used for back-end transactions. The secondary (shadow) database is a mirror image of the primary database, which is kept current by copying transaction journals from the primary database and replaying the journals into the shadow database. The shadow database is ideally located in an offsite location and is used for quick recovery in the event of a data center failure or other disaster.
Application and Data Tiers communicate over standard TCP/IP network configurations.
4.4.2 Application Tier
The Application Tier is the workhorse of the TrakCare application, containing TrakCare application programs and data.
It is responsible for dynamically generating TrakCare application Web pages in response to client requests. One or more Web Tier servers connect to the Application Tier via a Web gateway. The application server constructs an HTML representation of the data, which is passed back to the Web server via the gateway.
Application services can be located on servers separate from the Data Tier, ensuring that if they contain high-use static data, the database server is only accessed when dynamic transaction data is required.
When the Application Server needs to access a remote data server, a protocol is used to access the remote database and cache data on the application server so future related data requests can often be satisfied without using network resources.
4.4.3 Web Tier/File and Print Tier
The Web Tier/File and Print Tier typically unify with existing infrastructure.
TrakCare utilizes the TrakCare Reporting Model, which supports both centralized (system) printing and distributed (client preview) printing models.
TrakCare Print Service (TPS) supports a centralized queued model to manage system-generated printing, faxing, and emailing. Multiple print services can be run simultaneously, with each instance aware of the total number of print processes currently running to effectively control licensing costs.
ODBC-compatible reporting tools can be used to generate dynamic HTML or HTML, or display results via an ActiveX viewer. The preview mechanism offers the flexibility to print to a locally defined printer and the functionality to export reports in various formats.
The Web Tier serves TrakCare application Web pages – created dynamically by the Application Tier – to client PCs and returns client transactions back to the Application Tier.
TrakCare utilizes InterSystems Caché Server Pages (CSP) technology, which allows the building and deployment of high-performance, highly scalable Web applications. CSP enables Web pages to be generated at runtime on the application server, allowing content (HTML, XML, style sheets, images, and other types) to be served dynamically.
The client connects to the Web server via a Web browser using the standard HTTP protocol. The HTTP client (the Web browser) requests a page from the Web Tier using HTTP. The Web server recognizes this as a CSP request and forwards it to the Application Tier using a fast server application programming interface. The CSP Server in the Application Tier processes the request and returns a page to the Web Tier, which in turn sends it to the browser. CSP manages communications between the Web Tier and Application Tier and invokes application code to generate the page.
CSP supports HTML, XML, WML, and other Web-oriented mark-up languages. CSP uses a dynamic server page technology based on the Document Object Model.
4.4.4 Client Tier
The Client Tier is the only part of the TrakCare system that end users interact with directly. This tier can be any low-maintenance, light-footprint client device that supports Microsoft Internet Explorer 6.0 or higher. It provides a single interface between the user and the components of TrakCare, including Foundation CIS, Foundation PAS, departmental and add-on modules, system administration, report viewing, and non-TrakCare applications.
In addition to the server-side code that executes on the Application Tier, some code, typically JavaScript, is run on the client browser, usually to support advanced functionality such as data validation and reformatting, or invoking server-side code. ActiveX and Java controls are also used to execute client-side applets to preview complex images and standard reports and to modify TrakCare page layouts. Use of the capabilities can be restricted to specific users or groups of users and may be subject to license restrictions.
4.4.5 Standards and Protocols
The TrakCare solution utilizes the latest standards and protocols to enable connectivity and interoperability with other systems. TrakCare supports the following key technologies and standards, among others:
TrakCare Connects with Other Systems
4.5 System Architecture Configurations
The innovative design of TrakCare, which uses logical tiers, results in considerable flexibility to tailor a solution to an organization’s geographical and technological requirements. Multi-tier, multiserver solutions have the added advantage of providing outstanding performance, massive scalability, realtime data analytics, and robust reliability. A solution can be designed to accommodate several hundred or thousands of users, in a single location or over many sites. An entry-level configuration can be extended to support additional users as an organization’s use of TrakCare grows, without the need to rebuild or redesign from scratch.
4.5.1 Single-Server Configuration
In the simplest configuration, a single server is used for all the logical tiers.
The entry-level option is designed for deployment at a single site or for a limited implementation of TrakCare. Although the single-server configuration does not immediately provide the necessary infrastructure to support future growth, additional servers can be added as needs change.
A high-availability environment is possible by adding additional hardware. For example, a clustered or multi-tier configuration would require at least one more server.
TrakCare Single-Server Configuration
4.5.2 Multi-Tier Configuration
The multi-tier, multi-server configuration uses distributed database communications to enable a higher system capacity and greater system redundancy, allowing far more clients to connect to the system through the use of multiple application servers.
For sites requiring high capacity and high availability, it is recommended that the TrakCare Application and Database Tiers be distributed across multiple servers. The Web Tier should also be a separate tier. In addition, report services may be on separate servers depending on reporting requirements.
In the simplest multi-tier setup, two or more systems, acting as Application Tier servers, are placed between the central Data Tier server and the Web Tier servers. In this configuration, the Application Tier servers host processes that perform work for the client's benefit, off-loading the CPU load from the database server. This type of configuration scales best for Web applications such as TrakCare, which offers a significant level of data read access. More complex configurations, with multiple data servers as well as data stored on application server machines, are also possible.
Typically sites use multi-tier configurations for scaling as well as for providing high availability, with application servers serving as hot-standby systems.
For entry-level high availability, only a duplicate system with the same specifications as the database server is required. Shadow and failover database servers are similarly sized.
TrakCare Multi-Tier Configuration
4.5.3 Scalability and High Availability
A multi-tier architecture allows the system components to be spread across two or more data centers connected by high-capacity network connections.
The architecture does not alter significantly when scaling up, although the capacity of the components can be increased as additional sites and modules are brought online.
Acceptance, system testing, and training environments are located on their own pool of servers.
A multi-tier, multi-server system architecture is characterized by:
- Individual servers for the database and application servers
- High-availability clustering of the database servers
- Redundant Web, file, and print servers
- Disaster tolerance through geographically separate data centers and remote mirroring
- Private gigabit network between servers
- Use of SAN storage for all disk requirements (except operating system and system booting if required)
The architecture is designed to scale through the addition of:
- Application servers
- Web servers
- Scaled increments of hardware (e.g., memory, CPUs)
Unix systems are preferred for database and application servers for larger installations. However, Microsoft Windows Servers can be used in multi-tier configurations.
The following table compares single-server configuration with multi-tier configuration.
Table: Comparison of Different Options
| Option One Single-Server Configuration |
Option Two Multi-Tier Configuration |
|
|---|---|---|
| Hardware Costs | Lower start-up hardware costs | Higher start-up hardware costs |
| Hardware Efficiency | Seldom-used redundant hardware if high availability is employed | All [multiple] application servers for high availability are fully utilised |
| Scalability | Limited, and would require significant downtime to upgrade to a multi-tier environment | No interruption to services as additional server hardware can dynamically be brought online |
4.5.4 Client Architecture
The supported client operating systems and minimum recommended hardware configuration of PCs to be used as client workstations for the TrakCare client installation and operating system are:
| Client Operating System | Processor | Minimum Memory | Minimum Disk | Screen Resolution | |
|---|---|---|---|---|---|
| Minimum | Windows XP Professional SP2 with Internet Explorer 6.0 or 7.0 | 1 GHz 32-bit (x86) |
256MB | 10GB | 1,024 x 768 |
| Recommended | Windows XP Professional SP3 with Internet Explorer 7.0 | 2 GHz 32-bit (x86) |
512MB | 20GB | 1,024 x 768 |
| Recommended | Vista Business SP1 with Internet Explorer 7.0 | 2 GHz 32-bit (x86) |
1GB | 40GB including 15GB of free space | 1,024 x768 Vista requires DirectX 9 graphics 128MB memory |
The main client browser application used in the TrakCare solution is Internet Explorer 6.0 or higher.
Minimum required additional components for a typical TrakCare client workstation include:
| Application | Purpose | Hard Disk Requirements |
|---|---|---|
| Java Runtime Environment (JAI) | Required for the TrakCare Image Viewer used for viewing images and annotations. | 80MB |
| Java Advanced Imaging (JAI) | Required for the TrakCare Image Viewer used for viewing images and annotations. | 5MB |
| SVG Viewer | Used for charting and graphing. | 5MB |
| Adobe Acrobat Reader | Used for viewing PDFs such as Clinical Summaries and saved reports. | 60MB |
InterSystems certifies these environments with the TrakCare solution.
The additional software components required for administration application maintenance or configuration include:
- Visual Basic application for code table maintenance
- Ensemble client installation and ActiveX controls for editing layouts
While TrakCare does not require Microsoft Word, where Word integration capabilities are required, Word must be available to the client PC in addition to ODBC.
Users who wish to develop or maintain other ODBC report systems need to have a report designer and an Ensemble ODBC driver.
4.6 Other Architecture Requirements
Several environments should be maintained as part of the entire healthcare information system infrastructure. It is expected that there will be several TrakCare instances on the servers supporting different versions of the TrakCare system.
4.6.1 Production
The production system is used for live system activity.
4.6.2 System Test
An environment for vendors to place and run initial quality assurance tests. This environment can also be used for:
- Performance and stress testing
- Investigating the effect of alternative or emerging infrastructure components on application performance or availability
- Other testing (e.g., start-up, shutdown, backup, and recovery)
4.6.3 Training
Generally, TrakCare used for training is a copy of the production release (with full functionality), but newer versions could be used if training is required for a new release. During planned training sessions, the database may be refreshed prior to commencement. The training environment is often used to acceptance-test the latest releases prior to their deployment into the production system.
Server hardware for acceptance testing ideally would be the pre-production equipment. In large installations, it may be necessary to maintain a replica of the production systems for the purpose of continual acceptance testing; however, this would be rare. Disk requirements may be scaled down to suit the expectation of the level of acceptance testing.
4.7 Additional Peripherals Supported
The TrakCare solution has the capability to support a broad range of commodity peripherals, including tablet PCs, mobile devices, scanners, and barcode readers.
4.7.1 Tablet PCs
TrakCare supports the use of wireless tablet devices for deployment; these provide the full functionality of the desktop environment.
4.7.2 Scanners
TWAIN-compatible scanners can be interfaced into the TrakCare application. The interface is limited only by the TWAIN driver; however, the client must have an associated TWAIN driver. The interface is called using an ActiveX control via the browser.
4.7.3 Barcode Readers
The solution supports rapid and error-free search and selection through the use of barcode scanners. Typical uses include laboratory specimens and medical records.
TrakCare Solution Guide
 |
Next Page |
 |
