Evaluate how the design of a database can affect data quality. Discuss the role of a data dictionary in ensuring both the quality of enterprise-wide data and data within a specific database application. Discuss how to ensure the integrity and security
Godsgift1
chapter_14_reading.docxChapter 14: Fundamentals of Electronic Information Systems
Martin J. Smith, MEd, RHIT, CCA
Learning Objectives
Discuss the major components of an information system
Identify principal activities of information systems
Describe the major types of information systems and give an example of each
Distinguish between the purposes and functions of MIS, DSS, ES, and KMS
Differentiate between the various types of computers
Describe the steps in the systems development life cycle
Explain the vendor selection process
Identify the three main types of system software and provide an example of each
Discuss the purpose of an electronic database
Describe the purpose of a database management system (DBMS)
Explain the use and components of a data dictionary (DD)
Summarize the purpose and benefits of a data warehouse
Understand the functions of a communications system’s components
Compare and contrast clients and servers
Describe the general practices used to ensure data and computer security
Compare and contrast local-area networks, wide-area networks, intranets, extranets, and the Internet
Become acquainted with the policies that must be incorporated into the use of an information system
Describe functions associated with the management of computer systems
Identify new and emerging personal productivity software
Key Terms
Application service provider (ASP)
Application software
Artificial intelligence
Assembler
Assembly language
Barcode
Bus topology
Business process
Business-to-business (B2B)
Business-to-customer (B2C)
Chief information officer (CIO)
Client
Cloud computing
Column/field
Communications technology
Dashboard
Data dictionary
Data mining
Data type
Data warehouse
Database
Database management system (DBMS)
Decision support system (DSS)
Direct cutover conversion
E-health
Encoder
Enterprise-wide system
Executive information system (EIS)
Expert system (ES)
Extranet
Foreign key
Graphical user interface (GUI)
Hardware
Health information exchange (HIE)
Information system (IS)
Information technology (IT)
Inputs
Integrity constraints
Internet
Intranet
Key field
Knowledge management system (KMS)
Language translator
Local-area network (LAN)
Machine language
Mainframe
Management information systems (MIS)
Management support information system
Mesh topology
Minicomputer
Natural language
Network
Network protocol
Object
Object-oriented database
Object-relational database
Operating system
Operation support systems
Outputs
Parallel conversion
Peripheral
Phased conversion
Physical topology
Primary key
Productivity software
Programming language
Relational database
Ring topology
Row/record
Screen prototype
Secondary storage
Sequence diagram
Server
Software
Specialty software
Star topology
Strategic information systems planning
Structured query language (SQL)
Supercomputer
System software
Telecommunications
Transaction-processing system (TPS)
Turnkey systems
Unified modeling language (UML)
Use case diagram
Utility program
Wide-area network (WAN)
Wireless network
Workstation
Introduction
The computer is now an essential component for the collection and management of health information. While the planning for the widespread use of electronic health records has covered several decades, recent federal initiatives and legislation has ensured that the use of computers in healthcare delivery is now a requirement for healthcare providers.
Computers comprise a major portion of an organization’s information system (IS). The IS helps healthcare organizations meet the growing demands of patients, providers, and payers for electronic information. Efficient management and distribution of information are indispensable to a healthcare organization’s operation. Because of the principal role that computers play in an organization’s healthcare information system, it is vital that HIT professionals understand what makes up a well-constructed computer IS.
It is also essential that health information technicians (HITs) understand their evolving roles and responsibilities in working in this new e-health environment. To be effective workers, in addition to knowing the basic structure of an electronic information system and how it is applied to the healthcare environment, HITs must also understand the language of IS. This chapter introduces the primary components of an IS and discusses its essential elements, including data, people, processes, and computer systems. The chapter serves as a foundation for chapters 15 and 16 of this text.
Theory into Practice
Every healthcare provider must adopt an electronic health record no matter what setting they practice medicine in. Although the following example refers to the physician office setting, the challenges faced apply to all settings. This case study from the August 2009 edition of the Journal of AHIMA discusses a systematic planning process for successful EHR implementation in physician practices (Gaudreau and Palermo 2009, 80).
One consistent theme emerges from EHR implementations in physician practices: almost everyone underestimates the complexity, time, and effort required.
Implementing ambulatory EHRs includes all the challenges of acute EHR implementations with the potential for many more issues. Delivering an EHR program to scores of ambulatory practices requires enterprise-grade IT scaled down to the practice level at a reasonable cost. The sweep of EHRs includes many aspects of the way healthcare is delivered, affecting processes developed informally over decades. To translate so many manual processes into a tightly defined EHR is a major undertaking.
To deal with the complexity, time, and costs of EHR implementation within a short time period, a solid planning method must be in place. A rapid implementation planning process, called EHR RIP, can focus organizations and guide them in developing a plan that will help bring success.
RIP is a systematic approach that helps organizations establish their EHR objectives, translate them to functional and technical requirements, and identify all resources needed to implement an EHR. RIP also helps to establish realistic budgets, define roles and responsibilities, identify additional resources that need to be hired or contracted, and schedule every element in a detailed timeline.
RIP is a workshop-based process that unfolds in four to six weeks. The working sessions focus on gaining agreement on the strategy, defining functional requirements, and setting the scope of the program. As the plan takes shape, the process delves into the stated and implied assumptions, bringing them to the surface to ensure they will be addressed during the detailed planning.
The process begins with stakeholder executives and the implementation team at a kickoff meeting that establishes overall program objectives at the 40,000-foot level. The first working session descends to 10,000 feet to translate strategic objectives into strategic requirements that will fulfill each high-level goal grouped into program phases. The focus then descends to treetop level in subsequent workshops to map strategic requirements to specific functional requirements for data, access, and security. The rollout’s scope is defined in terms of the number of practices, providers, and users; their locations; and the interfaces they will use. The process drills down to illuminate all required resources and identify their likely costs.
RIP contains several important steps. These include: planning, design, building, and implementation. Each of these steps is successively broken down into activities. The following are examples of activities within each step:
Planning:
— Validation of existing EHR planning against best practices
— Establish project governance
— Develop high-level budget
— Establish project management
Design:
— High-level technology and requirements design
— Clinical standards design
— Budget refinement
— Implementation plan finalization
Building:
— Clinical buildout
— Integration and testing
— Implementation budget finalization
Implementation:
— Pilot rollout
— Pilot assessment
— Refinement
— Implementation and design changes
— Budget review
— Implementation
Basic Concepts of an Information System
An IS is the integration of several elements of a business process to achieve a specific outcome. The system receives and processes input and provides output. For instance, a physician ordering a lab test is a business process. The input is the order for the lab test. The output is a report from the laboratory. An IS is essential in integrating all the elements of the process so that the laboratory test is performed. For the system to work, the physician must provide the laboratory with specific information about what test is to be performed, when it needs to be performed, and on whom it is to be performed. The laboratory must schedule the requested test, collect the sample, analyze the sample (usually with the aid of a computer system), and finally report the test results to the physician. As you can see, the IS integrates all of the steps of the test-ordering process.
A well-designed and well-managed IS is critical for supporting the purposes of the health record. Accurate data can provide physicians, nurses, and administrators with information to make sound decisions. Bad data, on the other hand, can create bad decisions that can affect the health of patients. In Computers, Communications & Information, Hutchinson and Sawyer (2000, 9.2) define an information system “as a collection of related components that interact to perform a task in order to accomplish a goal.” All organizations generate information that must be managed and used in ways that allow them to accomplish their goals. “In addition to coordination among elements, systems must also be adaptive. That is, they must be able to respond to environmental changes by self-correcting the relationships among their internal elements” (Johns 1999, 368). Computers often are used to assist organizations with this challenging task.
Information System Components
An IS consists of data, people, and work processes and a combination of hardware (machines and media), software (computer programs), and communications technology (computer networks) known as information technology (IT).
Data, or raw facts, are provided to the IS by end users. These raw facts (also called inputs) have no meaning of their own. The IS refines them into meaningful information. For example, an end user may input the number 103 into the IS. The raw fact of 103 does not have any specific meaning of its own. However, when the IS system associates the fact as a patient’s temperature and compares the fact to the normal temperature range of 98.6 degrees Fahrenheit, the raw fact is transformed into meaningful information, in this case, indicating that the patient’s temperature is above normal.
In addition to data, an IS also consists of people. Users work with the IS to accomplish a variety of tasks. In healthcare, users include providers such as doctors and nurses, health data managers, technicians, therapists, oncology registrars, unit secretaries, case managers, chief information officers, and others who add data to the patient’s health record. Because the IS must satisfy their needs and solve their problems, users should be a part of the team that designs the IS. Their tasks and goals should be included in the system’s initial concept, and the users should be able to suggest changes as the system is being developed.
The last component of an IS is work business processes. Processes are the policies and procedures that users must follow to do their work. An IS automates many processes that are performed by end users. For example, procedures that are followed to assign an ICD code to a diagnosis would be considered a process. A coding encoder automates many of the procedures used to assign an ICD code number.
Information System Activities
An IS performs five specific activities: input, processing, output, storage, and controlling. Typically, the first activity an IS performs is accepting input (for example, a patient’s identification, temperature, blood pressure, and heart rate).
The second activity is processing the input. Processing can include performing calculations, making comparisons, or selecting alternative actions. For instance, the patient’s temperature and blood pressure readings may be compared against the normal values to determine if his or her vital signs are within normal range.
The third activity is producing meaningful output (such as providing a report of a patient’s vital signs). The output of an IS is usually considered information to be used in making decisions.
In addition to accepting input and processing it into meaningful output, an IS performs storage and control activities. In an EHR, the computer would store a patient’s physical exam on specific storage media such as a hard disk so that many users can retrieve it or so that it can be accessed at a later time.
The final activity of an IS is to control its own performance. For example, a hospital administrator might discover that the daily census output does not add up to the correct monthly census. This may indicate that data-entry or data-processing procedures need to be corrected.
Check Your Understanding 14.1
Instructions: Choose the correct answer for the following questions.
1. What is an information system?
A. A collection of related components that interact to perform a task in order to accomplish a goal
B. The integration of several elements in a business process to effect a specific outcome
C. A process that refines raw facts into meaningful information
D. All of the above
2. What are the components of an information system?
A. Computer servers, networks, and wiring along with personal computers
B. A combination of hardware, software, and communications technology
C. Data, people, and processes and a combination of hardware, software, and communications technology
D. Collecting, maintaining, analyzing, and disseminating information
3. What is the difference between data and information?
A. Data are composed of numbers, and information is composed of words.
B. Data represent raw facts and figures, and information represents the meaningful interpretation of data.
C. There is no difference between data and information.
D. All of the above.
4. Which of the following is an information system activity?
A. privacy
B. security
C. quality
D. input
5. A physician ordering a lab test is an example of:
A. business process
B. input
C. output
D. storage
Types of Information Systems
In healthcare organizations, computer-based information systems have historically been used to help managers at different levels to do their work. Physicians used specialized computer systems called decision support systems (DDSs) to help them make important decisions about the care of patients. The managers of the nursing, physical therapy, and health information management (HIM) departments often used computer systems called management information systems (MISs) to manage budgets, create work schedules, perform employee evaluations, and so on.
Coders, on the other hand, still use transaction systems. Every record is coded and entered into the database as part of an ongoing, daily activity. Eventually, the billing department uses the codes to submit bills to insurance companies and patients. The computer system stores this vast amount of data and makes them available for reuse for a variety of purposes.
Information systems usually fall into one of several categories. These include operation support systems, enterprise-wide systems, management support information systems, expert systems, and knowledge management systems. With the advance of sophisticated information technologies some of these boundaries are becoming blurred. However, the categorization is still useful in identifying the primary functions of specific systems. The following sections examine the different types of information systems in detail.
Operation Support Systems
When ISs are used to process data created and used by business operations, they are referred to as operation support systems. The role of an operations support system is to efficiently process business transactions, support communication and collaboration among business units, and update business databases (O’Brien 2002, 26). A registration, admission, discharge, and transfer (R-ADT) system is an example of an operation support system. It is used to process data created and used for the registration, admission, discharge, and transfer of patients. Outputs of an R-ADT system would include the daily admission, census, and discharge reports.
Transaction-Processing System
A transaction-processing system (TPS) is an example of an operations support system. A TPS manages the different kinds of transactions that occur in a healthcare facility. Patient admissions, employee time cards, and supply purchases are examples of transactions that take place in a healthcare facility.
The characteristics of a TPS include the following:
Inputs and outputs: Examples include patient admissions, discharges, and transfers.
Users: The users of a TPS are mainly lower-level managers who make daily operational decisions.
Products: These are detailed reports on transactions. A dictation-monitoring system reports the number of dictated events by type and by physician. A transcription supervisor can see how much work is left to do, what each transcriptionist has done, and other important data that can be used to determine how many transcriptionists are needed, how much overtime is required to handle the workload, and so on.
Support MISs and DSSs (discussed later): For example, the database for the R-ADT transaction system in a hospital supports higher-level decision-making support systems. The R-ADT system can send its data through an interface to an MIS software application. The MIS software turns the data into information such as average daily census, average length of stay, percentage of occupancy, and bed turnover rates and provides this information to management decision makers.
Enterprise Collaboration Systems
Enterprise collaboration systems are another type of operations support system. These systems typically enhance teamwork and are sometimes called office automation systems. Examples include electronic mail and appointment scheduling, project management software to coordinate tasks and schedules, and videoconferencing systems to hold electronic meetings.
Enterprise-wide System
An enterprise-wide system, sometimes referred to as an enterprise resource planning (ERP) system, is a large IS that manages data for an entire healthcare business. It helps automate information at the point of service and analyzes business and clinical practices for outcome and cost improvements. It is intended to solve enterprise-wide, rather than department, problems. Enterprise-wide systems are products that allow an organization to perform activities such as:
Human resource management
Financial management
Customer relationship management (CRM)
Supply chain planning
Sales and logistics
These systems enable a hospital to gather and manage data from across a health system providing administrators with consolidated data to evaluate operations by a variety of measures such as patient volumes, procedures, care sites, and other factors. This type of information helps administrators make decisions about standardization, productivity improvement, and cost reductions.
It is interesting to note that the electronic health record is a network connected enterprise-wide information system. It is the traditional record transformed into digital format and can be shared across different healthcare settings, because it is embedded in this type of system. Records may include a whole range of data including demographics, medical history, medication and allergies, laboratory test results, radiology images, and all the billing information for any particular patient.
Management Support Information Systems
Information systems that provide information primarily to support manager decision making are called management support information systems. The three major information systems fall into these categories: management information systems (MIS), decision support systems (DSS), and executive information systems (EIS).
Management Information System
An MIS is supported by a Transaction Processing System (TPS) which collects, stores, modifies, and retrieves the transactions of an organization to help middle managers make decisions about their departments’ objectives. MISs are usually specialized and designed to support a particular area of the business. For example, there are accounting management information systems, financial management information systems, marketing management information systems, and so on. The features of an MIS include the following:
Transaction data input
The users of an MIS include middle managers, such as directors of HIM departments.
The products of the system include summary, exception, periodic, and on-demand reports. An exception report might be a monthly report that lists the percentage of incomplete records by clinical department specialty.
In an HIM MIS, for example, input data might include admission, discharge, and transfer data; and data on the number of dictated reports, coded records, filed records, and incomplete records. Examples of the outputs would include structured reports, production schedules, and productivity analysis so that the HIM director can make management decisions.
Decision Support System
A DSS provides information to help users make accurate decisions. To the healthcare provider, this means using a product that goes beyond supplying facts about a patient’s medical condition. One example of a clinical DSS (CDSS) is a special type of DSS that helps a provider make decisions about patient care. It may alert the physician when a lab result is outside the normal range, for instance.
Moreover, reminders help physicians comply with clinical practice guidelines in the management of certain disease processes, such as diabetes mellitus. In addition, DSSs are used to link physicians via the World Wide Web to clinical knowledge databases. One such database is the National Library of Medicine, which enables physicians to search the literature to learn about the latest research. Top management also uses DSSs in planning for the future.
The general characteristics of a DSS include the following:
The inputs and outputs of a DSS include summarized reports, transaction data, and perhaps external data such as the ORYX performance measurement data developed by The Joint Commission (TJC).
The users of a DSS include top and middle managers and clinicians. Information from the system can help users make decisions about unexpected, and sometimes isolated, problems.
The products of a DSS include analytic models. Analytic models are mathematical interpretations of real systems such as pharmacy drug inventory systems. They help users manage inventory, predict customer needs, and make informed business decisions.
Executive Information System
An executive information system (EIS) is a type of management information system intended to facilitate and support the information and decision-making needs of senior executives by providing easy access to both internal and external information relevant to meeting the strategic goals of an organization. For example, an EIS can assist marketing executives in making effective marketing decisions as it provides an approach to sales forecasting, which can allow the market executive to compare a sales forecast with past sales.
The user interface is usually a series of interactive, interconnected, and cascading dashboards providing information that is relevant to specific executives and managers. For top executives the dashboard allows access to high-level information in addition to the ability to drill down to departmental-level data. See chapter 10 for additional information on dashboards.
Expert System
An expert system (ES) is a knowledge system built from a set of rules applied to specific problems. It can take the place of a human expert when it comes to problem solving. The system simulates the reasoning process of human experts in certain well-defined areas. “Knowledge engineers interview the expert or experts and determine the rules and knowledge that must go into the system. Programs incorporate not only surface knowledge (‘textbook knowledge’) but also deep knowledge (‘tricks of the trade’)” (Hutchinson and Sawyer 2000, 11.19).
Dr. Larry Weed’s knowledge coupler system (Weed 1991) is an example of an expert system applied to the practice of internal medicine. Much of the knowledge built into the system comes from his experience as a practicing physician as well as from common medical theory. Expert systems are also discussed in chapter 15.
Knowledge Management System
A knowledge management system (KMS) is a more recent type of information system that has the potential to increase work effectiveness. This type of system supports the creation, organization, and dissemination of business or clinical knowledge and expertise to providers, employees, and managers throughout the healthcare enterprise.
A KMS is usually composed of an electronic library or central repository of best practices that offers enhanced search capabilities. Information is organized by specific business domain in the electronic library. Most IT-based KMS in healthcare are web-based technologies. Employees can access the electronic library through the Internet or an extranet to search for information. KMSs also use special software that enables employees to collaborate in teams to use or add to the knowledge in the electronic library.
An example of a KMS is one developed at Partners Healthcare in Boston. This system integrates the clinical database and patient electronic records and embeds drug information into the physician order-entry process. Using this type of system, the physician is alerted to potential drug interactions before an order is processed. This system has reportedly reduced medication errors by 55 percent (Davenport and Glaser 2002).
Check Your Understanding 14.2
Instructions: Match the type of information system with the scenario in which it would be used.
1. Dr. J is treating a patient with a rare disease. He enters the patient’s signs and symptoms into a computer program that indicates the probability of the correct diagnosis with its treatment regime. The doctor then uses this information to determine the best treatment protocol for the patient.
2. Every week, an HIM department director receives statistical information on the number of incomplete medical records for discharged patients. Summary reports show totals and trends by the physician responsible for completing each patient record.
3. Dr. J orders a series of lab tests on his patient with a rare disease. The computer-generated results signal several abnormal values. This prompts the physician to add three new medications to the patient’s treatment protocol. The computer then reminds Dr. J that serious drug interactions could occur when the new medications are combined with the original drug protocol. In response to this reminder, he stops the order for the new medications.
4. The HIM department director receives daily reports on the number of new admissions to and discharges from the hospital.
A. Transaction-processing system
B. Management information system
C. Executive information system
D. Expert system
E. Decision support system
Development of Information Systems
Information systems must be created in a logical manner. The system development life cycle (SDLC) is the traditional way to plan and implement an IS in an organization. The major phases of the cycle are planning, analysis, design, implementation, and maintenance. Figure 14.1 shows the five phases of the system development life cycle. The case presented in Theory into Practice at the beginning of this chapter is a modification of the traditional system development life cycle.
Figure 14.1. Phases in the system development life cycle
1. Planning
2. Analysis
3. Design
4. Implementation
5. Maintenance and evaluation
Planning Phase
Given the size and complexity of healthcare organizations and the number of emerging technologies, strategic information systems planning is an essential first step in adopting new IS technology. Strategic information systems planning is the process of identifying and assigning priorities to the various upgrades and changes that might be made in an organization’s ISs. Its goal is to ensure that all changes contribute to the achievement of the organization’s strategic goals and objectives by developing a strategic information systems plan.
The financial and human resources of most healthcare organizations are limited. Different functional areas may want or need to upgrade or replace current ISs with new ones. All the competing requests may be valid and appropriate. For example, the pharmacy director may want to replace an outdated pharmacy IS with state-of-the-art handheld devices designed to allow electronic prescribing. At the same time, the laboratory director may request funds to upgrade the laboratory IS to the latest version. The health information manager may wish to implement a document-imaging system to address space constraints within the health information services department. The vice president for finance may want to install a decision support system on the desktop of every administrator. The whole organization might also need to install a facility-wide electronic healthcare system. How can the organization decide which of these requests is the most important to implement?
The IS strategic plan can be thought of as something like a blueprint for remodeling a house. A blueprint that incorporated all the wishes and needs of the family members—enlarged kitchen for mom, new garage for dad, additional bedrooms for the kids, new roof, third bathroom—probably would be too expensive to build and the resulting house might not function well. Family members would need to consider their long-term priorities before deciding how best to invest their limited resources. Similarly, an IS strategic plan should be based on the needs of the organization as a whole and its long-term priorities.
In the earlier example, implementing all the requests for a pharmacy IS, an administrative decision support system and so on probably would be too expensive and might not support the organization’s long-term goals. If the hospital’s long-range strategy were to implement a nonproprietary EHR system within two years, each director’s request should be considered within the context of how each application would contribute to that goal. Priority would be given to those projects that support the future implementation of an enterprise-wide EHR system.
Adopting new information technology represents an enormous investment in terms of staff time, hardware and software costs, and consulting fees. Establishing enterprise-wide priorities for IS development is not an easy task. As stated earlier, every area of the organization may be able to make a compelling case for why a new IS or new technology is needed and how a new system would help the area contribute to the organization’s strategic goals. Many factors come into play. And priorities do change in response to political pressure, patient safety issues, new federal regulations, and other factors.
One approach is to compile a comprehensive list of all of the proposed changes to the organization’s ISs. The list would include a description of every proposed change along with information on the new technology’s availability and ease of implementation, expected benefits and cost savings, and estimated acquisition and maintenance costs. The list then could be distributed to a broad sample of clinicians, managers, and end users. They would be asked to rank the different systems. The steering committee might oversee such a process under the leadership of the chief information officer (CIO) or some other top-level clinician or administrator. For clinical systems, it is critical that the initiative be clinician driven, not IS driven.
The strategic IS plan should serve as a guide to making difficult decisions about where, when, and how an institution should allocate resources for the management of information. The interdisciplinary IS steering committee and its role in setting priorities and managing projects are discussed in more detail later in this chapter.
Analysis Phase
Analysis phase of the SDLC is usually initiated by the submission of a project requisition or request from a department for the development, modification or purchase of an information system. The request typically includes an overview of the system purpose, desired functions, anticipated benefits, and costs. The request is generally reviewed by an organization-wide information systems planning or steering committee that determines how the proposed system supports the organization’s information systems strategic plan. Before approval additional information may also be obtained, including detailed cost estimates, resource requirements, and timeline.
Once the request is approved a systems analysis is performed. It is very likely that the HIM professional will be involved in the analysis phase of systems development. HITs may participate in this phase as end users who identify necessary functions of a new or existing IS for a particular HIM function. If the system is part of the HIM department, the HIT may be a member of the HIM team that determines the costs and benefits of the new system and prepares a report that justifies proceeding with analysis and design of the new system. Therefore, it is important that the HIT be familiar with the various steps in the systems analysis process.
The systems analysis defines the whats of the proposed IS project. The whats are based on asking some of the following questions:
What are the current healthcare business practices the IS will support?
What areas of the healthcare organization will be using the IS?
What are the IS project’s schedule and budgetary constraints?
What training methods are used currently in the healthcare organization?
What are the users’ needs for this system?
What system interfaces will be needed for this IS (for example, user interfaces, system-to-system interfaces, and so on)?
What legal issues are involved with this IS project (for example, patient confidentiality, physicians’ signatures, and so on)?
How do systems analysts collect information about the requirements for an IS project? Several methods are available to them. One very useful method for collecting data is to review existing documentation, forms, and databases. In the healthcare industry, paper forms are vital tools in understanding healthcare processes.
Another useful method is to do research and make site visits. Systems analysts must observe the work environment and learn how existing computer screens, data, and forms are used in the organization. The users of an IS can be invaluable to analysts. Users know their jobs well and can help analysts determine the system’s requirements. Users can show analysts how they do the jobs in question. This enables the analysts to understand the processes firsthand.
Conducting a joint application development (JAD) session is a valuable technique used to identify the goals, objectives, and required functions of a proposed IS. A JAD session is made up of a group of end users, system analysts, and technical development professionals who are brought together to analyze the strengths and weaknesses of the current IS and to propose functionalities for the new system. It is very likely that an HIT will participate as an end user in a JAD session.
A trained facilitator conducts the JAD session, which usually spans a period of several days. The session is held away from the organizational campus in a specially prepared meeting room so that participants will not be distracted or interrupted. At the conclusion of the JAD session, the essential functions are identified. The strength of the JAD method is that end users, analysts, and developers are brought together to collaborate in analysis of the IS. This allows for the free exchange and input of information among all concerned groups. The premise underlying JAD is that a group of individuals working together at the same time can perform an analysis faster and better than individuals working independently.
Prototyping is another analysis technique. A prototype is a model or example of what a completed IS may look like. Prototyping a system allows for maximum end-user input while speeding up the analysis and development process by simulating potential end versions of the system. End users and analysts work together to develop the external features of the IS, such as input screens and reports. These external features provide the look and feel of the proposed system but do not include actual program application codes that would make the IS work. The strength of the prototyping method is that it allows the end user to critique the functionalities of the system before time and expense are put into programming efforts. When the basic functionality of the system is prototyped, developers and computer programmers can develop application program codes and databases to support the new system.
As the systems analyst collects data for the IS project, he or she needs a way to document the system’s requirements in an easy-to-use graphical manner. Many graphical techniques are available to assist systems analysts. One technique is the unified modeling language (UML). The UML is an object-oriented modeling language that assists in the documentation of a software project by specifying, visualizing, modifying, constructing, and documenting the artifacts of a system under development.
An important part of the documentation process is the use case diagram. A use case documents the functions of a system from the user’s point of view. The use case is similar to a scenario or story that describes the functions of the system and is often used with a sequence diagram and sequence table. Figure 14.2 shows a complete use case. In this example the use case shows the interaction between a physician end user and the IS function of searching for a patient. The sequence table in the use case shows how the interaction between the physician and system unfolds (that is, similar to a story). Other documentation may include process flowcharts which are charts that are graphical in nature and provide a symbolic representation of the processing activities performed on the work piece; data flow diagrams (DFD) that are a graphical representation of the flow of data through an information system, modeling its process aspects; and conceptual data models that are maps of concepts and their relationships.
Figure 14.2. Example of a complete use case with sequence diagram and sequence table
The output of the analysis phase usually results in identification of what the system is to accomplish; its functions; models describing process flow, data flow, and entity-relationship diagrams (ERD), a database modeling method used to produce a type of conceptual schema or semantic data model of a system; and its requirements in a top-down fashion that identifies data items to be included in the IS.
Design Phase
When the analysis phase has been completed and IS functions identified, the system must be designed. The systems design phase specifies the functions of the system and provides the design or blueprint of the proposed system. The design phase describes the system’s hows. The hows of the project are based on asking some of the following questions:
How do the users interact with the system? What should the user interfaces look like?
How do the data identified for the IS relate to each other? What are the logical and physical data models for this project?
How do the pieces of the system interact with each other? What does the solutions model look like?
How will this system be programmed? What languages, databases, and tools will be used for this system?
System designers work with the system’s requirements and the systems analysts to determine the most appropriate design model for the problem being analyzed. They may use several models, including an object design or logical design, a physical design, and screen prototypes. Screen prototypes are full-color prototypes that illustrate the visual design of various page templates in a system.
The choice of design depends on the information gathered earlier in the analysis phase. The documentation developed in the analysis phase provides a function-by-function analysis of the system and may include entity-relationship diagrams, storyboards, and data dictionaries. Data dictionaries are discussed later in this chapter.
The IS design can be compared to a blueprint of a house. The blueprint tells the contractor how the house should be built. The IS design, including all the models developed by the systems analysts, tells the programmers and database administrators how to develop the information system.
In-House Development or Purchase
During the design phase it is decided how the new system will be developed. Will the new system be built in-house? Will the organization contract with an outside developer to build the system? Will the organization purchase a generic system from a vendor? The advantages of a ready-built system are timing of implementation and also the planning stage since the system is already built and has been tested. Some of the disadvantages of using a ready-made system are there might be many different functionalities of that system that are of no use in the setting, such as multiple physician capabilities when there are only three physicians in a particular practice. Or the ability to input x-rays into the system when the facility has no equipment to take x-rays.
Because of time, cost, and staffing constraints, most healthcare organizations coordinate the design process by looking at systems that are already available on the market. If a decision is made to purchase an already developed system, then the organization must go through a vendor analysis and selection process. Figure 14.3 describes the vendor analysis and selection process.
Figure 14.3. Vendor selection process
Identify Vendors
The selection process begins with the organization identifying a number of vendors that can meet the requirements of a new system. Information about IS vendors is available from a number of sources such as exhibits at professional association conferences, publications, consulting firms, and professional colleagues. Often the organization sends a request for information (RFI) to a list of vendors known to offer products or systems that may meet its needs. The RFI asks for general product information and is a good tool for prescreening vendors. Responses to the RFI can be used to narrow the list of potential vendors who will receive a request for proposal (RFP).
Request for Proposal
Next the organization issues a request for proposal (RFP) to the narrowed down list of vendors. The RFP is a written document that generally includes a detailed description of the requirements for the system and gives guidelines for vendors to follow in bidding for the contract. For example, the RFP might request the vendor provide the following (Wager, Lee, and Glaser 2005, 156):
• Vendor qualifications: general background of vendor, experience, number of installations, financial stability, list of current clients, standard content, and implementation plan.
• Proposed solutions: how vendor believes its product meets the goals and needs of the healthcare organization. Vendor may include case studies, results from system analysis projects, and other evidence of the benefits of its proposed solution.
• General contractual requirements: such as warranties, payment schedule, penalties for failure to meet schedules specified in contract, vendor responsibilities, and so forth.
• Pricing and support: quote on cost of system, using standardized terms and forms. Vendors are usually given a specified time in which to respond to the RFP.
Evaluation Vendors
After the RFP has been distributed to the vendors, the organization may hold a bidders’ conference to answer vendors’ questions about the RFP. This gives all the vendors the opportunity to receive the same information.
The organization should assess the reliability, cost, and projected benefits of each product. However, evaluation of the vendor and its products may not depend solely on the vendor’s response to the RFP. Other formal and informal mechanisms may be used to assess the vendor’s fit with the organization and the product’s capabilities. For example, it is a good idea to hold vendor presentations, attend user group meetings, and make site visits to other facilities that use the product. The purpose of these activities is to gain as much relevant information as possible. Clinicians and other end users should participate throughout the vendor selection process.
Contract Negotiations
After the top two or three vendors have been identified, the contract negotiation process can begin. The contract generally addresses numerous technical issues, everything from when the system is to be delivered and installed to who is responsible for ensuring that the product works with the organization’s other ISs. Internal legal counsel should review the contract carefully before it is signed and binding agreements are made.
Implementation
Once contract negotiations are completed, system implementation follows specified steps and a timeline, stipulated in the contract. Typically, the organization designates an interdisciplinary implementation team led by a project manager to develop a plan for implementing the new system. Implementation of a vendor system should include the usual implementation steps including coding, loading of data, testing, and user training.
Cloud Computing
The expense barrier in the 1960s and 1970s led to the emergence of so-called shared systems. Shared systems were developed by data-processing companies to provide computing power simultaneously to several healthcare organizations within a local or regional area. Healthcare organizations were charged for using the shared-system company’s data-processing centers and, usually, their applications. Like the mainframe systems used by hospitals that could afford them, most shared-system products began as administrative and financial systems. During the 1970s, the systems gradually migrated toward departmentlevel clinical systems such as laboratory, radiology, and pharmacy systems. Shared systems came to be known as application service providers (ASPs). Today this is known as cloud computing. Cloud computing “denotes the use of cloud- or Internet-based computers for a variety of services” (Shimrat 2009, 27). The Internet is used to access systems such as EHR, financial information systems, CPOEs, and other healthcare information systems software that are located at a remote site.
Turnkey Systems
Turnkey systems also began to emerge in the 1970s. Turnkey systems were actually developed by IS vendors and installed on the hospitals’ computers. Essentially the healthcare institution purchased a complete system of hardware and software that was installed by the vendor and ready to implement. Typically, the hospital’s data-processing staff maintained the systems (Johns 1997; Wager et al. 2005). However, most turnkey systems could not be modified.
Implementation Phase
The implementation phase is a complex undertaking and includes the development of the computer programs, testing of the system, and development of system documentation, user training, and system conversion. Typically, the organization designates an interdisciplinary implementation team led by a project manager to develop a plan for implementing the new system. Selecting an effective project manager is critical to successful system implementation. Ideally, the project manager should be a knowledgeable individual with past experience in similar implementation projects. The project manager should be someone who is highly respected by others in the organization. Some of the typical functions of an implementation project manager include overseeing the entire project from conception to completion, controlling the budget, resources, timelines, and coordinating project status meetings. Finally, he or she should have strong organizational and communication skills and the political influence and authority necessary to get the project done.
A well-executed implementation does not guarantee that users will accept a new system. Problems during implementation can lead to user frustration, dissatisfaction, and disillusionment. Indeed, some organizations never recover fully from disastrous system implementations. Some of the potential problems facing the implementation are overspending of the budget, delays affecting the installation of equipment, and clients requesting last minute changes. Of course, the larger the healthcare entity, the higher the stakes for a successful implementation.
The project manager and the implementation team should begin the implementation process by making a list of the tasks to be completed. The scope and complexity of the process depend on the type of system, the number of users, and the complexity of the conversion process. Typical milestones in the implementation process for healthcare information systems include:
Preparation of the site (for example, remodeling work spaces and installing telephone lines, computer cables, and electrical power lines)
Installation of hardware and software
Preparation of data tables
Construction of system interfaces
Development of a network infrastructure to support the system
Training for managers, technical staff, and other end users who will test the new system
Testing of the new system to identify and correct problems
Preparation of support documentation (for example, procedure manuals)
Development of backup and disaster recovery procedures
Training for other end users
Data conversion
Conversion to the new system
Many implementation activities take place simultaneously. Others need to be completed before other activities can begin. It is generally a good idea for the project manager to use a Gantt chart or another project management tool to schedule and track the milestones in the implementation process. The tool should list the activities to be completed along with the estimated start and completion dates for each, the names of the individuals responsible for each activity, and the resources needed to complete each task. Project management software (for example, Microsoft Project) is useful for creating Gantt charts and tracking project resources (for example, staff, equipment) and expenditures.
All the implementation tasks are important. However, three are especially critical to the success of the project: testing the new system, training the end users, and converting to the new system.
Testing of the New System
Thorough testing of new systems (hardware and/or software) before the actual conversion date is critical. Systems testers test the cases developed in the design phase against the system’s requirements. These testers should be users of the system. If a requirement fails, the tester reports the problem to the technical staff. The technical staff then fixes the problem and completes their report. Based on this report, the testers retest the requirement until it passes.
Testing should be conducted using actual patient data, not sample data the vendor has provided or the organization has created for training purposes. Correcting a problem in the test mode is often easier than correcting it after the system is fully operational. Even though it is nearly impossible to identify and correct every potential problem before a new system goes live, it is essential to identify and correct as many of them as possible. It is very unlikely that any errors will be found once the system has been through the testing phase. Problems that are not identified and resolved prior to implementation could end up with the system causing major problems and even could result in having to take the system down if the problems are significant.
It is very likely that an HIT professional will participate in systems testing, particularly when the IS is part of the HIM department or electronic health record. For example, the HIT professional may be asked to put test data into the new IS, such as an electronic master patient index, and determine whether the system provides the expected output.
Training for End Users
It is critical to provide adequate training for the end users of the new system. For any new system to be successful, it must be accepted and used by the staff. This includes physicians and other caregivers in the healthcare setting. When staff is not thoroughly trained, the result may be low morale and dissatisfaction with the system. Some of the resources required for training will be system manuals, training software (potentially online through the vendor), and customer service support. It is important to have policies and procedures and other documentation available during training so that staff can be trained on the entire process—not just the computer system.
One common approach to staff training is a train-the-trainer program. In this approach, key people in the various functional areas (for example, nursing, laboratory, and billing) are identified and trained first. They then train the other users in their area. This approach is effective because the trainers are still available to help the staff after the system has been installed and the vendor is no longer on-site to answer questions.
It is equally critical to allow adequate time for training. Staff should not have to squeeze in training during their lunch breaks. They will need time to practice using the new system. Just as it is important to use actual data to test the system, it is also important for staff to practice using the new system with actual data on real patients. One other method to ensure staff receive the necessary training is to negotiate a training package with the vendor who is supplying the hardware and/or software.
Conversion to the New System
Conversion to a new IS often requires significant changes in workflow and organizational structure. The process also demands a lot of staff time and disrupts productivity and business processes if not properly executed and resourced. Adequate technical support staff must be available to assist managers and end users as needed.
Several different approaches may be used to make the conversion from an old IS to a new one. In deciding which approach to use, it is important to consider the nature of the application, the risks and costs associated with each alternative, and the resources available. The most common approaches are parallel, phased, and direct cutover conversions (Wold and Shriver 1994).
The parallel approach involves running both the old and the new systems until the managers and staff are confident that the new system works. This approach is costly and can be confusing to staff, but it ensures that a backup system would be available if needed, which makes it the safest implementation approach.
One type of phased approach involves implementing portions of the new system over time instead of installing the entire system all at once. Another type of phased approach involves implementing systems in selected locations instead of at all locations at the same time. For example, a document-imaging system might be implemented in one clinic as a pilot site, with future plans to deploy the system to other clinics within the facility. Converting in phased stages helps the staff gain confidence in the new system, helps the implementation team learn from the experience at the pilot site, and ensures that sufficient time is allowed to make the transition smoothly.
Finally, with the direct cutover approach, the organization stops using the old system and starts the new one on a specified date. This approach is risky, but can work effectively when sufficient testing was done and adequate backup procedures are in place. In fact, it may make more sense to convert via the direct cutover method when the old system is quite different from the new one.
Maintenance and Evaluation Phase
The last phase of the systems development life cycle is system maintenance and evaluation. Maintenance and evaluation activities ensure both the short- and long-term success of the information system. Problems inevitably show up after a new IS is put into operation. Adequate IS support staff must be available to identify potential problems and take steps to correct them.
Technical support for critical systems should be available 24 hours a day, 7 days a week. The patient care IS is probably the most vital system in a healthcare organization. Sufficient technical staff also should be available to oversee the following activities:
System backups
Software upgrades
Equipment maintenance and replacement
Ongoing user training and assistance
Disaster recovery
Some IS experts estimate that at least 25 percent of the technical staff’s time should be devoted to maintenance activities (Austin and Boxerman 2003). System maintenance can be either performed by full-time employees or contracted out to the system vendor.
Every organization that maintains electronic health information is required by federal law to develop emergency backup procedures and a disaster recovery plan. (See the discussion of the Health Insurance Portability and Accountability Act in chapters 13 and 17.) Staff training in emergency procedures also is required. Written backup and emergency procedures should be made readily available to staff.
Maintaining and supporting new systems is not enough, however. The effectiveness of every IS should be evaluated on a continuous basis. Continuous evaluation activities ensure that the organization’s ISs support its overall mission and goals and meet users’ needs.
Healthcare administrators today demand information on the organization’s return on investment (ROI) when new technological systems are implemented. The ability to measure ROI is increasingly important as healthcare institutions struggle to manage limited resources more effectively. Consequently, as a part of the system evaluation process, organizations are looking at the organizational, technological, and economic impact of ISs on the enterprise as a whole (Friedman and Wyatt 1997; Anderson and Aydin 2005).
Evaluation measures how well the original ambitions of the new system (i.e. the logical design laid down during the analysis phase) have been achieved. Evaluation doesn’t really serve to improve the system that is being evaluated; it serves to improve the next system that will be worked on. Different systems will have different criteria that demonstrate their effectiveness or efficiency. For example, if you buy a cow, you don’t evaluate its speed, and if you buy a greyhound, you don’t evaluate its milk output.
Once you know what criteria you need to evaluate, you need to devise a way to evaluate them. There are two main ways: objective and subjective. Objective evaluation involves collecting facts, figures, and measurements. Subjective evaluation involves finding out opinions. Where possible, objective measurements are preferred because they are more reliable.
So how long after implementation should you start the evaluation phase? If you wait too long, users remember less about the development process and how it might be improved, but if you start too soon, users have insufficient time to assess system strengths and weaknesses. A good rule of thumb is six months, but pressure to finish sooner often exists. A client should wait until the system is “bedded down” and users are familiar with it, then evaluate whether it has solved the original problem. Has it achieved the goal specified in the Problem Analysis? If not, it should be fixed as soon as possible. Don’t sit around with a new system that is as bad, maybe worse, than the old system.
Ideally, the post-implementation evaluation should be performed by people who were not involved in the development process. External auditors often are involved, since they are impartial and don’t have a stake in the success or failure of the system.
Check Your Understanding 14.3
Instructions: Choose the best answer to complete the following statements.
1. The systems development life cycle (SDLC):
A. Is the traditional way to plan and implement an IS
B. Takes at least three years to complete
C. Is primarily transaction oriented
D. Is a never-ending process
2. The first phase of the SDLC is the _____ phase.
A. Design
B. Planning
C. Implementation
D. Maintenance
3. End-user requirements are identified in the _____ phase of the SDLC.
A. Design
B. Analysis
C. Implementation
D. Maintenance
4. The method(s) used to identify requirements for a proposed IS is (are):
A. Conducting site visits
B. Conducting a JAD
C. Developing a prototype
D. All of the above
5. Testing the new IS system is part of the ______ phase.
A. Design
B. Planning and analysis
C. Implementation
D. Maintenance
6. Instructions: Following is a list of the different tasks or activities that might need to be done to implement an automated coding system. Number the tasks in the order in which they should be carried out. Assign the same number to any tasks that can be done simultaneously.
____ a. Establish an implementation team and develop a project timeline
____ b. Assess the coders’ needs
____ c. Determine whether the proposed coding program is in line with the organization’s strategic plan and mission
____ d. Hold bidders’ conferences for potential vendors
____ e. Prepare a list of users’ specifications
____ f. Negotiate a contract with the system vendor
____ g. Prepare documentation and procedures to support the new system
____ h. Determine the date of implementation
____ i. Submit an RFI to potential vendors
____ j. Train staff on how to use the new system
____ k. Test the new system
____ l. Submit an RFP to potential vendors
____ m. Identify problems with current coding processes and opportunities for improvement
____ n. Contact other sites that use the products or systems being considered
____ o. Determine conversion plans
____ p. Conduct cost-benefit analysis of the different products or systems being considered
Information Architecture
Information architecture (IA) is the art and science of organizing and labeling websites, intranets, online communities and software to support usability. As information proliferates exponentially, usability is becoming the critical success factor for websites and software applications. Good IA lays the necessary groundwork for an information system that makes sense to users. Information technology (IT) encompasses a wide range of topics. This section limits examination to the basic functions of the hardware and software used in information systems.
Categories of Computers
Even though they perform basically the same functions, computers come in various size categories. The different sizes of computers are discussed in the following subsections.
Supercomputers
Supercomputers are the fastest and highest-capacity machines built today. They can cost millions of dollars and are used in large-scale activities such as weather forecasting and mathematical research.
Mainframe Systems
Mainframes were the only computers available until the late 1960s. They can perform millions of instructions per second, are designed to connect input/output devices over long distances, and can handle hundreds or thousands of users at the same time. Mainframe systems vary in size from middle to large capacity, depending on the number of concurrent stations they serve. They are generally used in healthcare to handle input/output-intensive transactions.
Most hospitals use mainframes to store payroll, personnel, billing, and accounting data. The challenge for today’s systems engineers is to interface newer technologies for clinical information systems with the older legacy systems stored within the mainframes.
Midrange Systems
There are two types of midrange systems: minicomputers and workstations. Minicomputers can support hundreds of connected users at the same time via terminals consisting of a keyboard and a video screen. They are cheaper than mainframes.
A more recent introduction in the 1980s was the workstation. A workstation is a very powerful desktop computer used mainly by power users such as graphics specialists for multimedia production. Workstations are comparable to midsize mainframes but sit on a desktop. They also may be used as servers to microcomputers connected through a network.
Microcomputers
Microcomputers, also called personal computers (PCs), were introduced in the early 1970s and are the fastest-growing type of computer today. They come in a variety of sizes, including desktop, laptop, palmtop, and pen-based. Microcomputers can be used in a stand-alone environment or connected to a network.
Handheld Devices
These devices are lightweight mobile devices that provide special functions such as taking notes, organizing telephone numbers and addresses, and calendaring. Other application software such as word processing and spreadsheets are now widely available. Today’s devices, such as iPads and Kindles include thousands of applications that enable users to perform job functions, search the Internet, and can act as a truly portable computer. Importantly, these devices can exchange data with a desktop computer and can be used alone or in a networked environment.
The vast majority of these devices have web-based capabilities allowing them to send or receive e-mails and access the Internet. Similarly, some devices also provide video conferencing capabilities. Today’s devices have a user-friendly interface that includes touch screen capabilities and a stylus input device. The stylus is used to navigate the PDA and write notes.
The use of these devices in healthcare is growing rapidly and computer vendors are now producing them specifically for the healthcare market. These devices may allow clinicians to:
Track patients and write, upload, and retrieve clinical notes
Enter laboratory orders and check results
Access reference materials to perform medical calculations, such as determining drug dosages
Write prescriptions
Record and verify charge capture
Specific privacy and security policies and procedures should be in place for control of protected health information that may reside on a particular device.
Cellular Phones
A cellular phone can be a web-based telephone having features of analog and digital devices, also referred to as a smart phone. In addition to basic phone capabilities, a smart phone also provides the functions to receive and send e-mails and faxes and access the Internet. Cell phones allow patients to look up information on conditions, schedule appointments, communicate with healthcare providers, and more. Cell phones can also have special applications. Most recently MIT Media Lab’s NextLab program is developing cell phone applications geared toward the developing world. The software would allow patients to transmit their health information, including pictures, to a doctor or nurse in a remote location in order to receive a preliminary diagnosis and to find out whether the condition warrants a trip to the clinic (Merrill 2009).
Types of Computers
Computers can also be divided into three categories depending upon their instruction and form of input data that they accept and process. These are analog computers, digital computers, and hybrid computers.
Analog Computers
The word analog means continuously varying in quantity. Analog computers accept input in continuous analog signal form, and output is obtained in the form of scaled graphs. Voltage, current, sound, speed, temperature, and pressure values are examples of analog data. These values continuously increase and decrease. The thermometer is an example of an analog device because it measures continuously the length of a mercury column. Another example is the analog clock because it measures the time by means of the distance continuously covered by the needle around a dial. Similarly speedometer and tire-pressure gauges are also examples of analog devices.
Analog computers have low memory size and have fewer functions. They are very fast in processing, but output return is not very accurate. They are used in industrial units to control various processes and also used in some fields of engineering.
Digital Computers
The word digital means discrete. It refers to the binary system, which consists of only two digits (that is, 0 and 1). Digital data consists of binary data represented by OFF (low) and ON (high) electrical pulses. These pulses are increased and decreased in discontinuous form rather than in continuous form.
In digital computers, quantities are counted rather than measured. A digital computer operates by counting numbers or digits and gives output in digital form. A digital computer represents the data in digital signals 0 and 1 and then processes it using arithmetic and logical operations. Examples of digital devices are calculators, personal computers, digital watches, digital thermometers, and so forth. Today, most of the computers used in offices and homes are digital computers.
The main features of these computers are that they give accurate results, they possess high-speed data processing, they can store large amounts of data, they are easy to program and use, and finally, they consume low energy.
Hybrid Computers
The hybrid computers have the best features of both analog and digital computers. These computers contain both the digital and analog components. In hybrid computers, the users can process both the continuous (analog) and discrete (digital) data. These are special purpose computers and are very fast and accurate. They are used in scientific fields and in hospitals, and are used to watch the patient’s health condition in the ICU (intensive care unit). These are also used in telemetry, spaceships, missiles, and so forth.
Computer Peripherals
The different pieces of hardware that are connected to central processing units (CPUs) to make them more functional and user-friendly are called peripherals. Peripherals are usually described in terms of the five basic computer functions: input, processing and memory, output, storage, and communications.
Input Devices
Input devices include keyboards; microphones; scanners; pointing devices such as mice, trackballs, light pens, and intelligent tablets; sensors; and biometrics such as fingerprints, handprints, and iris scans. Sensors are important in healthcare because they can read a patient’s physiological data and transmit them to a computer database. A common example is the use of monitoring systems in critical care units.
The selection of the appropriate type of input device for an IS depends on the user’s workflow. If the user moves around a lot (as a nurse does, for example), an input device that allows data entry while moving from place to place is ideal. Headsets with speech input are often used in a mobile work environment.
Barcodes are being used more frequently in healthcare for data capture. Barcodes are machine-readable representations of data, typically dark ink on a light background. Barcodes are recognized by scanners as narrow and wide, or 0 or 1. Using appropriate barcode software, these data are then interpreted into meaningful information. In HIM department electronic document management systems, barcodes are used to eliminate the manual indexing of document type, patient name, provider name, medical record number, and other information, as well as medical record separator sheets during the digital scanning process (Kohn 2009).
Output Devices
An output device is any piece of computer hardware equipment used to communicate the results of data processing carried out by an information-processing system (such as a computer) to the outside world. The most common output device is the screen of the visual display unit. The processor is continually addressing the screen to send it signals whenever the results of an instruction have to be communicated to the user. Other output devices include printers, faxes, and speakers. The function of these devices is to translate the machine’s response to a usable form for the user.
Data Storage Devices
Secondary storage devices include a flash drive, hard disk drive, magnetic tape, and an optical disk drive. The drives may be internal or external. Digital data can be stored permanently on any of these media, although the life span of each medium varies. This type of device has come into notoriety since they can be easily lost or misplaced.
Optical storage allows for extremely large quantities of data to be stored on one CD. This medium is very useful for storing image and sound data and has the longest life span of secondary storage media.
Processors
A processor is a microchip implanted in a CPU’s hard drive that processes instructions sent to it by the computer and software programs. Processors come in a number of sizes. The greater the gigahertz capacity of the processor, the quicker the computer will be able to process instructions sent to it. Some corporations have developed a dual processor, which allows one processor to process multiple instructions at the same time without slowing down performance.
Communications Devices
Communications devices are used to assist communications among different computers. Modems translate digital data into analog data so that the data can be transmitted over telephone lines and received by a remote computer. At the receiving end, the modems reverse the process by turning the analog signal back to digital.
Transmission speeds are very important to the transmission of data. They are expressed in terms of bits per second (bps) or kilobits per second (kbps). Many phone lines max out at 28.8 kbps. Thus, even though the user may have a 56K modem, the rate of transmission will still be 28.8K. Newer technologies in communications include integrated services digital network (ISDN) lines that allow digital data to be transmitted through copper wire telephone lines. This is a dedicated line that can be very costly compared to normal telephone service, but it is five times faster than the fastest modem.
Another technology is the asymmetric digital subscriber line (ADSL). This is thought to be the successor to the ISDN and also functions on standard telephone lines. Cable modems are connected through TV cable lines. This technology is faster than ADSL. Cable modems can download data at a million bits per second (mbps).
Finally, satellite dishes, a wireless connection, offer the fastest transmission. Clients subscribe through a service provider. These are entities that provide web services but can also provide communications, storage, or processing services. One example is an Internet Service Provider or ISP.
The advances in the portfolio of communication technologies (that is, faster broadband wireless networks, smart phones, and data compression techniques) have made it possible to put together a suite of telemedicine applications. For example, communications technologies support emergency medical services (EMS) in a variety of ways including transmission of real-time physiologic data and real-time video from ambulance to emergency room. Communication technologies also support telemedicine services, connecting rural health clinics to clinical specialists and services at urban centers.
Computer Programming Languages and Software
To make all these devices work together, the equipment must use a set of instructions called software. The software programs (or directs) the hardware components to perform the tasks required of them. Software is an ordered sequence of instructions and is developed by using programming languages.
An overview of programming languages and software categories is provided in this section.
Programming Languages
Software programmers who write instructions to the computer use highly specialized languages. The relative ease with which people can interact with and instruct computers has progressed dramatically—from very difficult hardwiring instructions to the use of natural languages.
Since they were introduced in 1945, programming languages have gone through five generations of development, including:
1. Machine languages: The first generation of programming languages, machine language, consists of ones and zeros.
2. Assembly languages: The second generation, assembly language, uses a standard set of abbreviations to replace some of the ones and zeros of machine language. Assembly languages are usually defined by the hardware manufacturer and therefore are not portable to different computers.
3. High-level languages: High-level languages use words and arithmetic phrases to construct programs. Examples of third-generation languages include COBOL and BASIC.
4. Very high-level languages: The fourth generation of programming languages includes report generators, query languages, data management languages, and application generators. These languages were developed to reduce programming effort, time, and costs. They are easier to use and require fewer commands to execute programs.Structured query language (SQL), Report Builder, and SPPS are just a few examples of a fourth-generation language.
5. Natural languages: Natural programming languages allow users to speak in a more conversational way with the computer and are part of the expanding field of artificial intelligence (AI). “One of the most successful natural language systems is LUNAR, developed to help users analyze the rocks brought back from the moon” (Hutchinson and Sawyer 2000, 13.14). AI is used widely to develop expert systems that help physicians manage patient care, telemedicine, teleradiography, and other e-health and telemedicine activities.
Software
Software can be categorized into three major classes based on the software’s function. These broad classes include: system software, computer programming software, and application software, although the boundaries between them at times are often fuzzy.
System Software
In an orchestra, the different musical instruments are played in complete synchronization. How does this happen? A composer writes a set of instructions for each instrument using the language of music. In a sense, a computer programmer is a composer except that the programmer writes a set of instructions for computer hardware. Like the conductor guiding the orchestra through a musical composition, the system software acts as the conductor for all the hardware components and the application software. It essentially tells the computer when to start, what to do, and when to stop. Without system software, nothing would happen.
System software has three basic pieces. These are discussed in the following subsections.
Operating System
The operating system “consists of the master programs, called the supervisor, that manage the basic operations of the computer. These programs reside in RAM [random access memory] while the computer is on and provide resource management services of many kinds; for example, they run and store other programs and store and process data. The operating system includes BIOS (the basic input/output system), which manages the essential peripherals such as the keyboard, screen disk drives, and parallel and serial ports” (Hutchinson and Sawyer 2000, 5.4–5.5).
Common microcomputer operating systems include DOS (disk operating system), Windows, and earlier versions—OS/2, Unix, and Macintosh.
User interfaces are an important software support for an operating system because they determine how the user communicates with the computer. The DOS system, for example, uses a command-driven interface that requires instructions to be typed in at a command prompt. A second type of interface is the menu-driven approach. This approach allows the user to select choices using a mouse or some other non-keyboard pointer. Presently, the graphical user interface (GUI) is the standard microcomputer interface. It operates on the basis of icons that represent different computer tasks and programs. The GUI allows for keyboard as well as point, click, and drag functions.
a. Utility Programs
Utility programs “are generally used to support, enhance, or expand existing programs in a computer system” (Hutchinson and Sawyer 2000, 5.8). Examples of utilities include:
Backup processes that store data in more than one location
Virus protection that protects computer programs and data
Data recovery that prevents the loss of data in the event of physical or software accidents
b. Language Translator
The language translator “is software that translates a program written by a programmer in a language such as C++ into machine language, which the computer can understand. All system software and applications software must be turned into machine language for execution by the computer” (Hutchinson and Sawyer 2000, 5.12). According to Hutchinson and Sawyer (2000, 10.15–10.16), language translators include assemblers, compilers, and interpreters.
Application Software
Application software can be further categorized into four different types:
Productivity software products are used in almost all businesses to assist with word processing, accounting, database management, graphics presentations, scheduling, e-mail, time management, and other functions performed in offices and homes.
Specialty software includes programs designed for a specific industry. For example, in health information management, encoders are designed to accelerate the medical coding process. Other types of specialty software in other industries include multimedia authoring, desktop publishing, and CAD/CAM products.
Education and reference software includes encyclopedias, anatomy atlases, and library searches. One very successful library search program is Medline, which is offered through the Internet site of the National Library of Medicine.
Entertainment software includes games and audio/video entertainment.
Check Your Understanding 14.4
Instructions: Indicate whether the following statements are true or false (T or F).
1. ____ Computer peripherals such as mice, printers, and monitors control the computer’s operating system.
2. ____ Common utility programs include virus checkers, backup, and recovery.
3. ____ DOS stands for disk operating standard.
4. ____ The user communicates with the computer through an interface (Windows, for example).
5. ____ Machine languages use a standard set of abbreviations to replace some of the ones and zeros used in assembly language.
6. ____ The increasing use of wireless technology is one of the hottest technology trends today.
7. ____ Artificial intelligence is one type of natural programming language.
8. ____ Education and reference software is one category of application software.
9. ____ The three basic pieces of system software are the operating system, the language translator, and the utility program.
Databases
The most critical resource in healthcare is patient data. An argument can be made that data is actually more valuable than money since you can only use money once before it is gone whereas data can be used over and over again. The most important functions of any healthcare information system involve being able to create, modify, delete, and view patient data. And the most important storage mechanism used to perform these functions is a database. As discussed in chapter 4, databases are essential in the development of EHR systems.
A database is an organized collection of data saved as a binary-type file on a computer. Users cannot read a binary-type file because it contains only ones and zeros. A database management system provides the ability to perform the functions mentioned earlier. Many different kinds of database management system vendors are available in the marketplace, including Oracle, SQL Server, Sybase, and Access.
Database Approach
There are three popular types of databases: relational databases, object-oriented databases, and object-relational databases.
A relational database stores data in predefined tables that contain rows and columns similar to a spreadsheet. The kinds of data that can be stored in a relational database are currency, real numbers, integers, and strings (characters of data). They are used for the storage of information in databases used for financial records, manufacturing and logistical information, personnel data, and much more. This type of database is a popular model used in healthcare applications.
An object-oriented database stores objects of data. An object is a discrete or abstract thing such as a car or a line at the grocery store. Data objects can model relational data or advanced data types such as graphics, movies, and audio. Objects are used in object-oriented languages such as Smalltalk, C++, Java, and others
Finally, the object-relational database combines the best of the relational and object-oriented databases. It uses both traditional data types (such as currency, integers, and strings) and advanced data types (such as graphics, movies, audio, and so on). They can be used to build database management system and can connect to a company’s website and allow updates to items such as their inventory records.
Database Purpose and Activities
The purpose of a database is to store and retrieve data. A popular common language called structured query language (SQL) is used to store and retrieve data in relational databases. SQL gives the information system the ability to query and report on data and to insert, update, and delete data from the database.
A database management system (DBMS) is a collection of computer programs that controls the creation, maintenance, and use of a database. The DBMS allows different application programs to access the database. An important purpose of a DBMS is to maintain the data definitions (data dictionary) for all the data elements in the database. It also enforces data integrity and security constraints.
Relational Database
The relational database, as stated earlier, is one of the most popular database architectures used in healthcare. Relational databases perform the following activities:
Store data in tables consisting of rows and columns. Figure 14.4 shows the hierarchy of data in a relational database. A column consists of a name and a data type (the kind of data being stored in the column). A row in a table consists of data values in the various columns. A row would be the values of these columns, such as “John” for the first name and “Smith” for the last name.
Figure 14.4. Storage hierarchy of a relational database
Retrieve and store data in tables using SQL to insert, update, delete, and query data from tables.
Provide a level of security, usually by user and by table and column that each user is allowed to access. In healthcare systems, access to health records must be limited to only certain users in order to protect patient privacy.
A table is a two-dimensional structure made up of rows and columns. Table 14.1 shows an example of a table that stores information about a patient.
Table 14.1. Example of a PATIENTS table
PATIENT_ID
LAST_NAME
FIRST_NAME
DOB
RACE
DOCTOR_ID
1
Smith
Keith
2/13/70
White
1
2
Roberts
Debbie
7/30/70
Black
1
3
Morrison
Rebecca
2/4/99
White
1
4
James
Sally
7/29/50
Black
1
A column/field is a basic fact such as LAST_NAME, FIRST_NAME, DOB, RACE. A row/record is a set of columns or a collection of related data items. An example of a row in the PATIENTS table is: Smith, Keith, 2/13/70, White.
A key field uniquely identifies each row in a table. There are two types of keys:
Primary keys ensure that each row in a table is unique. A primary key must not change in value. Typically, a primary key is a number that is a one-up counter or a randomly generated number in large databases. A number is used because a number processes faster than an alphanumeric character. In large tables, this makes a difference. In the PATIENTS table, the PATIENT_ID is the primary key. It is good programming practice to create a primary key that is independent of the data in a table.
A foreign key is a column of one table that corresponds to a primary key in another table. Together, they allow two tables to join together. For example, say we have two tables, a CUSTOMER table that includes all customer data, and an ORDERS table that includes all customer orders. The intention here is that all orders must be associated with a customer that is already in the CUSTOMER table. To do this, we will place a foreign key in the ORDERS table and have it relate to the primary key of the CUSTOMER table.
The cancer registry is an example of a health record database that can be computerized. A manual registry is already organized logically by site of the neoplasm. Imagine that a two-drawer file cabinet is being used for the cancer registry. The entire file cabinet is considered the database. The first drawer contains a set of abstracts by site of the neoplasm. Within each file folder are the abstracts or records kept on each patient. Each record contains a collection of data fields or facts about the patient.
The second file drawer is organized by month. Each month is one file folder. Within each month’s folder are the cases that need to be followed for that month. Each case is considered a record. Within each record are the necessary pieces of information, or data fields, required to contact the patient and update the medical data.
In a computerized IS, each case is stored in one table and each patient is stored in another table. Case and patient have to be related to each other so that the user can look up the case and also locate the patient information. For example, if the cases were stored in a CASES table (with a primary key of CASE_ID) and the patients were stored in a PATIENTS table (with a primary key of PATIENT_ID), a foreign key would be needed to relate the case to the patient. If the patient were to have many cases, a foreign key of PATIENT_ID would have to be located in the CASES table in order to create the one-to-many relationship. Thus, when a user queries a case, he or she can easily use the foreign key PATIENT_ID to find the patient information. (Relational databases are also covered in chapter 16.)
Data Models
Data models provide a contextual framework and graphical representation that aid in the definition of data elements. In a relational database, the data model lays the foundation for the database and identifies important entities, their attributes, and the relationships among entities (Johns 2002). An entity is anything about which data can be stored and can be a concept, person, place, thing, or event. In the cancer registry example above, “PATIENT” and “CASES” would both be entities. Data models are usually constructed by data administrators after a thorough analysis of the business processes has been performed. A sample data model for the entity “Person” is shown in figure 14.5.
Figure 14.5. Sample data model for the entity “person”
Source: Shakir 1999.
Data Dictionary
Databases are only as good as the data they contain. Without a mutually agreed upon set of data elements with clearly defined names and definitions, the validity and reliability of the data contained in a system are suspect at best and must be discounted at worst. The data dictionary is a central building block that supports communication across business processes (AHIMA e-HIM Workgroup on EHR Data Content 2006). A data dictionary improves data validity and reliability within, across, and outside the enterprise because it ensures that each piece of data can only mean one thing. It also improves communication in clinical treatment, research, and business processes because terms are defined and used the same way by everyone in the organization. Standardization provides developers with a common road map to promote consistency across applications. For example, the data element “PATIENT” would have the same field length and definition across all applications in the organization.
Lack of a sound data dictionary can cause problems within and across organizations. For example, in an organization one department may use the term “PATIENT” while another department may use the term “CLIENT” and yet another department may use the term “CUSTOMER” to define the same entity. Clearly this situation makes it difficult for an organization to collect all of the information it needs, and it may make it impossible to combine or map data across systems because the definitions are not identical. A worse possibility is that an organization may combine data elements it believes to be equivalent and draw incorrect inferences from the invalid data. Multiple users entering data may have different definitions or perceptions of what goes into a data field, thereby confounding the data. For example, are “reason for visit” and “chief complaint” the same or different?
Table 14.2 provides an example of a data dictionary. Maintaining the data dictionary frequently is an HIM function. A data dictionary is essential in ensuring consistent definitions of what data names mean and making sure that data are accurate. A data dictionary is critical in the development of EHR systems.
Table 14.2. Sample data dictionary
Table Name: PATIENT
Field Name
Field Type
Field Size
Definition
Allowable Values/Edits
Patient_ID (Primary Key)
Number
8
Unique patient number automatically generated (autonumber) on first patient encounter
0-9
Gender
Character
1
Gender of the patient
M = Male
F = Female
U = Unknown
PT_LN
Alphanumeric
50
Patient’s legal surname
A-Z
PT_FN
Alphanumeric
50
Patient’s legal first name
A-Z
PT_DOB
Alphanumeric
8
Patient’s date of birth
DD-MM-YYYY
PayType
Alphanumeric
25
Patient’s primary source of payment
BC/BS
CHAMPUS
Medicaid
Medicare
Self-pay
Other
PT_Race
Alphanumeric
25
Patient’s declared race
American Indian
Alaskan Native
Asian
African American
White
The content structure of a data dictionary may vary among organizations. Data dictionary content depends on the use of the dictionary and the specific notation method used to develop the dictionary. A good dictionary defines each data field or column according to the following information:
Name of field
Type of data field
Length of data field
Edits placed on the data field
Values allowed to be placed in the data field
A clear definition of each value
The data fields themselves usually evolve from a predetermined data set, such as the data sets discussed in chapter 4.
Data sets are typically developed by groups of people who have a legitimate use for collecting the data. The purpose of carefully defining a data set is to help ensure the accuracy of the data collected and the usefulness of the statistics obtained from the collection of the data.
HIT professionals can play a vital role in the development of data sets and data dictionaries because they possess the knowledge and terminology that are recognized throughout their particular healthcare setting. Their involvement will help healthcare organizations to create clear and nonredundant electronic patient records in the future. Additional information on the data dictionary use in the EHR is provided in chapter 16.
Techniques for Database Integrity
Because databases are the backbone of an EHR, careful attention must be given to ensure that data stored in a database have the data quality characteristics described in chapter 2. These include accuracy, accessibility, comprehensiveness, consistency, currency, definition, granularity, precision, relevancy, and timeliness. The following discussion is a guide for achieving these characteristics in a database.
Good Database Design
Good database design is a process that involves a team of individuals who have good relational database knowledge and extensive technical database design expertise. In addition, the database development team needs to include end users who have knowledge of the real world or domain, which the database will describe. It is very rare to find one person with knowledge in both technical database design and the described domain.
Usually the end user describes the domain, its elements, and the data that needs to be stored to the database designer. This step is very important as it defines the goals and the requirements of the database. Because of their experience and knowledge with healthcare information, HIM professionals should take the lead in being the interface between the domain and technical experts in documenting the data dictionary.
Good database design relies heavily on development of accurate data models (figure 14.5). Data models, as stated earlier, are the blueprint of the database and describe database entities and their relationships. Data models are usually developed by individuals with expertise in data modeling who identify entities and their relationships with help of the individual who has the domain knowledge.
A good database design implements exactly the data requirements of the end users (the ones with the domain knowledge). A proper database design ensures that there are minimal redundant data in the database and that the data structure is flexible enough to easily handle change in the database requirements.
Integrity Constraints
Using integrity constraints is extremely important. Integrity constraints provide a way of ensuring that data that are entered or updated in a database by authorized users do not result in a loss of data quality. Integrity constraints could either be a specification of uniqueness for values of a column (for example, only allowing the input of “M” for male or “F” for female in a gender field) or validation for values of a column (for example, allowing only a specified range of values for a field that records a patient’s temperature). Referential integrity is another strategy that is used to ensure data quality in relational databases. Referential integrity ensures that there is consistency between tables that are linked in the relational database. HIM professionals are frequently involved in helping to develop integrity constraints in a database.
Database Management Systems
Database management systems (DBMSs) fall into two categories: a personal DBMS, which runs on a client; and a server-based DBMS, which runs on a server.
Personal DBMS
A personal DBMS is used for small projects such as storing contact information (for example, in a personal address book). It should not be used for systems that require large amounts of storage and 24/7 access. Most healthcare systems would not use a personal DBMS. Personal DBMSs are tempting to use because (1) they are inexpensive and (2) most people already have access to one with an office software package such as Microsoft Office. Microsoft Access only supports a few gigabytes of storage. Today, databases span terabytes of storage. These types of databases require the use of a server-based DBMS.
Server-Based DBMS
A server-based database management system runs on a server computer. It also runs as a separate application from a personal system. This allows the DBMS to run faster and more efficiently than the personal DBMS. A question might take minutes to answer on a personal DBMS, but only seconds on a server-based DBMS. Examples of a server-based DBMS are Oracle, SQL Server, Sybase, and Informix. A server-based DBMS also allows data to be retrieved by multiple end users. Such a database system is essential in an EHR system.
Server-based database management systems provide support for 24/7 access with large amounts of storage in the terabyte range or higher. They are very expensive compared to personal database management systems.
Data Warehouses
Today, most organizations use multiple databases in their daily business operations. Many of these are separated from each other and the data are not available in a consolidated form to help managers and others make decisions. A data warehouse is a special type of database that alleviates this problem by consolidating and storing data from various databases throughout the enterprise. Data warehouses are designed to perform data analysis rather than to support routine operations (figure 14.6).
Figure 14.6. Example of a data warehouse
The principal purpose of the data warehouse is to provide data for improved decision support. A data warehouse usually contains historical data that are derived from operation (transaction) systems and is optimized for speed of retrieval of data. “A data warehouse enables users to tap into knowledge hidden in the massive amounts of data to understand business trends and make timely strategic decisions. The data warehouse allows decision makers to perform simple reporting, complicated analysis, multidimensional analysis, population analysis, physician analysis and benchmarking, clinical protocol development, reimbursement/cost analysis, and much more” (Farishta 2001, 28–32).
The benefits of a data warehouse include better customer service, lower production costs, increased profitability, and quicker turnaround times for making decisions. One of the most powerful applications of a data warehouse is to engage in data mining.
Data Mining
Data mining is associated with data warehouses. Data mining is a process that identifies patterns and relationships by searching through large amounts of data. Because data warehouses contain large amounts of data, data mining processes are frequently used to systematically analyze these data. In healthcare, data mining is used to identify methods for cutting healthcare costs, suggest more appropriate medical treatments, and predict medical outcomes. It has also been used by the federal government to help identify fraud and abuse practices and evaluate and reduce Medicare payment errors.
Typically, data mining is performed by individuals with a background in statistical analysis. An evolving role for the HIM professional is in data analytics. A certified health data analyst (CHDA) manages, analyzes, interprets, and transforms data into accurate, consistent, and timely information.
Telecommunications
Telecommunications is the term used to describe voice and data communications within an organization. It is a central and critical aspect of any healthcare organization. Telecommunications provides the technology infrastructure that enables people in any type of organization to talk with each other using a telephone, access a database, and send e-mail to each other using a computer.
In a healthcare organization, the efficiency of the technology infrastructure sometimes means life or death for patients. Thus, a properly designed technology infrastructure is needed to provide instant and highly reliable access to data with 24/7 security.
Health information exchanges (HIEs) are now beginning to radically transform healthcare delivery. HIEs (see chapter 16) are starting to connect providers, patients, payers, and other stakeholders through telecommunications and other technologies. Connected by computer networks, patient care data are starting to be seamlessly transferred to where they are needed, at the time they are needed, and to who needs them.
HIM professionals are involved in the development of HIEs and will continue to play an important role in clinical and health data exchange for years to come (Durkin and Just 2008). An understanding of telecommunication infrastructure and tools will help the HIM professional in this role.
Network Fundamentals
A computer network, often simply referred to as a network, is a collection of hardware components and computers interconnected by communication channels that allow sharing of resources and information. Where at least one process in one device is able to send and receive data to and from at least one process residing in a remote device, then the two devices are said to be in a network. They may be classified according to a wide variety of characteristics such as the medium used to transport the data, communications protocol used, scale, topology, and organizational scope.
To understand the nature of a computer network, it is important to first understand the nature of communications systems. As shown in figure 14.7, a communications system is made up of four components:
Figure 14.7. Components of a communications system
The transmitter is the device that sends information.
The receiver is the device that receives information.
The medium is the mechanism that connects the transmitter to the receiver. It may be a cable (for example, the twisted-pair cable commonly used to connect workstations in an office building) or the air (in the case of cellular phone transmissions).
Data create the message that is transferred from the transmitter to the receiver as electrical pulses.
One very common communications system is the telephone. The handset of a typical telephone acts as both transmitter and receiver. The cables and telephone wires that connect one telephone to another are the media. Voice signals, such as words and other sounds, are the data.
It is possible to create a communication system by connecting one computer to another to form a network. The users of computers in the network then can share information and resources. In a computer network, a computer functions as both transmitter and receiver.
The specific type of communication that takes place on a computer network is often called data communication. Data communication involves transferring information in binary form, which is the electronic equivalent of ones and zeros. In addition to computers, a network can incorporate a variety of computer devices (called nodes) such as printers, fax modems, scanners, CD-ROM drives, tape backup units, and plotters. The collection of computers and devices that make up a network are sometimes called the network’s resources.
The purpose of a network is to allow users to share its resources easily and efficiently. For example, any network user can access the network printer, not just the user sitting behind the computer to which the printer is attached.
Clients, Peers, and Servers
Network computers play one of two roles: client or server. Clients are computers that access shared resources. Servers are computers that share resources such as printers or hard-disk space across the network. For example, a user interacts with the client computer to request something from the network (for example, e-mail messages). The server responds by supplying the e-mail messages the user requested. People often refer to this interaction as the client-server relationship (figure 14.8).
Figure 14.8. Client-server network relationship
The term services refers to tasks that a network server performs, such as facilitating e-mail, web, Internet, and printer connections; providing database access; performing backups; providing network communication; coordinating security; and managing files. The various services a server provides include:
Application services that allow different pieces of application software to reside on a specialized type of server called an application server
Communication services that allow different systems to communicate with each other
E-mail services that permit the exchange of correspondence electronically
Internet services that let users transfer files and other information via the Internet
Web services that allow users to use a browser and a server to access the World Wide Web (a system of Internet servers that support specially formatted documents called web pages, although some Internet servers are not part of the web)
Printer services that enable users to access a printer from anywhere on the network
Database services that store and retrieve information such as accounting and financial data
Backup services that routinely create backup copies of information
Network management services that make it possible to manage network resources from a central location
Security services that prevent unauthorized users from accessing the network
File services that allow anyone with access to a server to store and retrieve files
Because network computers can act in only one of two roles, the peer-to-peer relationship is an extension of the client/server model. In a peer-to-peer relationship, a networked computer acts as both client and server.
Network Architectures
The way a computer network is set up is its basic design or architecture. There are three main types of network architectures: local-area networks, wide-area networks, and wireless networks.
Local-Area Networks
The local-area network (LAN) connects computers in a relatively small area (for example, within one room or one building). It can take the form of a client/server network, a peer-to-peer network, or a hybrid network. A hybrid network is a mixture of client/server and peer-to-peer networks.
One special form of a LAN is an intranet. An intranet is a specialized client/server network that uses Internet technologies. An intranet provides information within the organization. This information can include policies and procedures, cafeteria menus, forms, training, and more.
Intranets let corporations supply Internet services over their LANs. Essentially, intranets are private Internets with the security required to protect a corporation’s assets. An extranet, on the other hand, provides network connectivity between suppliers to allow direct connection to each other’s network.
Wide-Area Networks
The wide-area network (WAN) connects devices across a large geographical area (for example, across a state or even the world). WANs can take several forms but often simply consist of two or more LANs connected by telephone lines.
The world’s largest public WAN is the Internet. The Internet consists of hundreds of thousands, if not millions, of interconnected LANs around the world. By connecting private LANs to the Internet, individuals and corporations can create a relatively inexpensive WAN. However, the Internet is a very insecure environment. Thus, LANs connected via the Internet may be subject to intrusion from unauthorized users.
Wireless Networks
The wireless network refers to any type of computer network that is not connected by cables of any kind. It is a method by which homes, telecommunication networks and enterprise (business) installations avoid the costly process of introducing cables into a building, or as a connection between various equipment locations.
Network Topologies
Computer networks consist of a configuration of cables, computers, and other devices. How data flow through a network is referred to as physical topology. For example, in star topology, data transferred between one computer and another must first flow through a central connection point called a hub. There are four types of physical network topologies that are commonly used. Each of these, described below, has its strengths and weaknesses. Selection of a topology is based on the purpose of the network, size of the network, and cost, among other considerations. The main types of physical topologies are star, ring, bus, and mesh. Figure 14.9 shows three of these topologies.
Figure 14.9. Common network topologies
Source: Durkin and Just 2008.
Bus Topology
Bus technology is an older topology where each computer is connected to a common backbone or trunk through some kind of connector. A signal or message from the source computer travels along the backbone in both directions to all machines. If the machine address (that is, MAC or IP address) does not match the intended address for the data, the machine ignores the data. Where the machine address matches, the data are accepted. Bus topology is inexpensive to implement and works best with a limited number of devices. A disadvantage of this topology is that the central cable is a single point of failure. If this cable fails for any reason then the entire network goes down.
Star Topology
In a star topology each machine is connected to a central hub. All of the data on the network have to pass through the hub, which forwards those data to the correct destination. The star topology is easy to design and install, and adding nodes (computers or other devices) to the network is less difficult than other topologies. Like the bus topology, the star topology has a single point of failure, which is the hub. If the hub fails, the entire network would go down.
Ring
In the ring topology each device is connected to the network in a closed loop or ring. Each machine is identified by a unique address. The signal passes through each machine or computer connected to the ring in one direction. Ring topology usually uses a token passing scheme. A token is essentially an empty container to transport data. To use the network a machine has to capture an empty token and insert its message. The token is then passed from machine to machine sequentially on the network until it reaches its correct destination. Only one machine can transmit on the network at a time. Ring topology is good for small networks and not costly to install, and it is easy to add new nodes to the ring. The primary disadvantage of ring topology is the failure of one machine will cause the entire network to fail.
Mesh
A mesh topology combines characteristics of bus, ring, and star topologies, but allows for redundant routes for data transfer. Essentially each node in a mesh topology is interconnected to all the other nodes. Even if one node fails on the network, the network will not go down because there is more than one path from one node to another. Mesh topologies allow for the expansion of an existing network, and enable companies to configure a network to meet their needs. The Internet is the most famous example of a mesh topology. The disadvantages include the overall length of each segment is limited by the type of cabling used; if the backbone line breaks, the entire segment goes down; and they are more difficult to configure and wire than other topologies.
Data Transfer
Network protocols enable computers on the network to communicate with each other. The computers have to use the same language, just as people have to speak the same language in order to understand each other. Network protocols provide computers with a common language.
The variety of network types, computers, operating systems, and browsers available today greatly complicates the exchange of information between computers. To facilitate this exchange, networks rely on some basic protocols, or rules, to govern the organization and transmission of data. Most protocols apply to three distinct phases of the data exchange process:
1. Connection setup phase: Just as a telephone uses ringing to request a connection to another telephone, a protocol must first initiate a connection between computers on the network.
2. Data transfer phase: After setting up the connection, the protocol allows the computers to transfer data. After the data are transmitted, the receiving computer decides whether to accept or reject them.
3. Connection release phase: After the data transfer is complete, the protocol allows the computers to terminate the connection.
The following are common network protocols:
Transmission Control Protocol/Internet Protocol (TCP/IP) is the basic communication language or protocol of the Internet. It is also used as a protocol in a private network such as an intranet or extranet.
File Transfer Protocol (FTP) is a protocol used to exchange and manipulate files over a TCP/IP network.
Hypertext Transfer Protocol (HTTP) is a protocol used to transfer and display information in the form of web pages on browsers.
Simple Mail Transfer Protocol (SMTP) is an Internet standard for electronic mail transmission across Internet Protocol (IP) networks.
HL7
Health Level Seven (HL7), founded in 1987, is a not-for-profit, ANSI-accredited standards developing organization that provides comprehensive standards for the exchange, integration, sharing, and retrieving of electronic health information that supports patient care. The HL7 standard allows exchange of data between common systems that make up the EHR such as radiology, laboratory, pharmacy, and other systems.
As described in chapter 16, Health Level Seven (HL7) is a family of standards that aid the exchange of data among hospital systems and, more recently, physician practices and other types of provider systems. HL7 is used by almost every EHR vendor in the United States and its version 3 is web-based and has been adopted in several other countries around the world.
Check Your Understanding 14.5
Instructions: Match the following terms with their definitions.
1. Intranet
2. World Wide Web
3. Extranet
4. LAN
5. WAN
6. Data warehouse
7. Clients
8. Telecommunications
9. Relational database
10. Data dictionary
a. Type of database that stores data in predefined tables made up of rows and columns
b. Storage device for multiple databases that can be accessed via a single question-and-report interface
c. Type of network that works as a private Internet
d. Computer network that connects devices in a small geographical area
e. System of Internet servers that supports specially formatted documents
f. Description of the structure of a specific database
g. Type of network that allows the networks of separate organizations to communicate with one another
h. Term used to describe a system of voice and data transmission
i. Computers that are used to access shared resources in a network
j. Computer network that connects devices across a large geographical area
Electronic Commerce in Healthcare
“E-commerce is a euphemism for conducting business via Internet tools, especially the World Wide Web, and to a lesser extent, e-mail” (McLendon 2000, 22). Electronic commerce (e-commerce) has revolutionized healthcare. E-commerce uses both the web and electronic data interchange (EDI) as the means for conducting business. Traditionally a web-based platform has been used for business-to-customer (B2C) transactions while EDI has been used for business-to-business (B2B) transactions. However, with advances in web and related technologies, a web-based platform is being used increasingly for both B2C and B2B transactions.
A special task force (AHIMA e-Health Task Force 2002) predicted that e-commerce would have a tremendous impact on healthcare delivery. Specifically the task force provided the following examples:
Increased numbers of consumers will access the Internet for information about healthcare providers, treatment options, and their own personal health information.
Health websites will provide consumers with tools to develop and maintain their own online health records.
Consumers and health providers will correspond via e-mail.
Businesses and consumers will purchase supplies and equipment over the Internet.
Health information management (HIM) business processes will use the Internet for off-site transaction processing.
All of the above examples are common today. B2B e-commerce is routinely conducted between healthcare facilities, vendors, suppliers, insurance companies, and others. B2B e-commerce focuses on improving the efficiency of providing products, services, and information among businesses. Some examples include web-enabled online claims handling, online medical supply purchasing, web-enabled medical record systems for individual physicians, and remote HIM support services such as transcription, coding, release of information, and billing to healthcare providers or organizations. Streamlining the processes involved in the ordering, purchasing, and delivery of products and services through the use of e-commerce can save millions of dollars.
B2C e-commerce today includes a variety of applications to help improve the participation of the patient in the delivery of care. Among the many healthcare examples are websites that educate consumers on disease prevention and wellness promotion, portals for communities of interest, online provider directories, customer service portals (that is, making an appointment with a healthcare provider), and portals for managing individual health information (that is, personal health records).
While the media and means of maintaining and transporting health information and services may have changed, the challenges of delivering quality information and protecting the confidentiality and security of information have not changed. HIM professionals should be in the forefront of developing standards of practice that address the security, privacy, and quality standards of health information in an e-commerce environment.
Management of Information Technology
Information technology by itself cannot fulfill its potential unless it is appropriately organized and managed as a system to meet the goals of the organization. Formal organizational structures for the planning and management of information systems are required. Modern healthcare organizations recognize that information is a critical resource to be managed as carefully as human, financial, and capital resources. In large part, the organization’s ability to provide high-quality and cost-effective healthcare services depends on the availability of accurate information.
The specific organizational structure for managing information resources varies among organizations. For example, a small community-based clinic would have a different organizational structure for information systems management than a large urban acute care facility. Regardless of size, there is a set of functions that be must organized and executed to ensure information resources management. Among these are:
Information systems strategic planning and enterprise-wide oversight of the organization’s information systems functions. In smaller organizations, this responsibility might be assumed by the top manager or director with assistance from a consultant. In larger organizations, this role might be filled by a chief information officer (CIO).
Management of the technical infrastructure and staff to support an organization’s information systems including such areas as data administration, network administration, database management, and end-user support. This responsibility usually is assumed by an information systems director or department. In smaller organizations a manager may oversee these responsibilities in conjunction with a consultant or outsourcing company.
Management of security and privacy of healthcare and other information. In larger organizations, these responsibilities are usually assumed by individuals with titles of chief security officer (CSO) and a chief privacy officer (CPO). In smaller organizations, the functions may be assumed by management staff with the assistance of consultants or professional health information managers.
Health record information management functions such as managing health record content, retention, retrieval, diagnostic and procedural coding, and compliance with accreditation and legal standards. Management of these functions should be performed by qualified health information professionals.
Additional information on the management and organization of information system resources is provided in chapter 17.
Personal Productivity Software
Personal productivity software is software applications designed for individual use on personal computers to improve the efficiency and productivity of work. The variety and sophistication of personal productivity software has grown over the past few years. Just a decade ago these systems principally consisted of word-processing, spreadsheet, and simple database and project management applications. Today, however, productivity software consists of powerful applications brought to a user’s desktop that frequently come in a bundle, called an application suite.
Many of these applications are used to support health information management functions. Therefore the HIT professional should know the purpose of these software applications and have a working knowledge of when and how to use them. Some of the more popular of these are described below.
Database Systems
Database applications designed for personal computer systems have the same purpose as the large database systems described earlier in this chapter. The principal functions of a database system are to organize and store data so that they can be readily retrieved in a meaningful way by end users.
Personal databases usually are complete database management systems and provide functions for the creation, storage, and use of data. These systems are fairly sophisticated with user-friendly interfaces that allow for easy and quick defining and development of database structures such as tables and relationships. These systems also provide functions for the development of forms, reports, screens, and queries for the input and output of data, and can be used for the development of smaller, complete applications.
These database applications are appropriate for individual, small group, or team level work. The drawback of such systems is that they do not have the flexibility, security, or performance of larger and more sophisticated systems. HITs may use this type of database application for small departmental or research projects. One of the most popular examples of this type of system is Microsoft Access.
Spreadsheets
Spreadsheets are computer applications designed for use on personal computer systems and networks and are used for business analysis, planning, and modeling. A spreadsheet is an electronic worksheet of rows and columns that simulates a paper accounting worksheet. The user designs the format of the spreadsheet by identifying variables and entering associated data. Associations between cells are defined through formulas. For example, a column of numbers can be related by a formula to sum (add) all the numbers in that column.
Most spreadsheet applications include functions for statistical analyses and for developing displays of spreadsheet data such as charts and graphs. The strength of spreadsheet software is that numerical data can be modeled and analyzed. HITs may use spreadsheets to collect and analyze data for smaller research projects or to analyze business processes associated with HIM functions. Popular spreadsheets include Microsoft Excel, IBM Lotus 1-2-3, and Apple Numbers.
Presentation Software
Presentation software is used to support a public presentation by creating multiple computer screens that display text and graphical information. Presentation software usually has functions that allow a user to insert and format text, graphics, and sound or video clips. The software allows the user to navigate from one screen to the next, in effect creating a slide show to support a presentation. Using specialized projection equipment, the slide show can be projected on a screen to accommodate large audiences. Slide shows can be uploaded as web pages and be viewed on the web. Electronic presentations can be used to supplement or replace traditional presentation materials such as chalkboards, overhead transparencies, and flip charts. HITs are likely to use presentation software in conducting privacy and security or other HIM-related training or in providing information to healthcare organization staff and management. Microsoft PowerPoint is one of the most popular presentation software packages.
Project Management Software
Project management software provides functions for the planning, organizing, and managing of resources to complete a project. Most project management software includes applications for planning, scheduling, and tracking tasks through the use of automated GANTT or PERT charts. More sophisticated packages add tools to simulate the project and can readjust schedules, budget, and personnel assignments as well as provide sophisticated tracking and analysis of project progress. HITs frequently work with project teams whose management is supported by project management software. Microsoft Project is one of the most popular products for small- to medium-sized projects.
Groupware
Groupware is another category of productivity software. It helps increase efficiency within work groups and teams by facilitating the exchange of information and supporting realtime collaboration. The software can support teams who are both located onsite and offsite. Groupware can include functions that allow users to share documents, post notes, receive notifications, plan meetings and events in a group calendar, access e-mail, track schedules, and store document versions. Many groupware applications include discussion boards and chat rooms. Lotus Notes is one of the more popular groupware packages.
Real-World Case
This case is based on article in AHIMA’s 79th National Convention and Exhibit Proceedings (Nugent et al. 2007).
Developing and Implementing Telehealth Program: The VA’s Story
The VA’s Office of Care Coordination’s (OCC) Home Telehealth (HT) Program is currently putting medical devices in patient homes to improve the quality of care and standard of living for veterans throughout the United States. Veterans use these home devices to capture various measurements including blood glucose, blood pressure, pulse from the blood pressure device, temperature, weight, pain, and pulse from the pulse oximetry device. All these measurements are obtained from measurement devices (except pain, which is subjective) that are connected to the home device. The home device sends the measurements to a national vendor server located within the VA network, and it sends those measurements to VHA systems where their VA providers can view them.
Summary
Well-managed information systems are even more critical to the success of today’s healthcare organizations than ever before. Information systems are the integration of people, data, processes, and technology to affect specific outcomes. These systems can be categorized as operation support, enterprise-wide, and management systems. The traditional way to manage information systems is through the systems development life cycle. The SDLC is a four-step process that takes the organization from system planning and design through implementation and maintenance. As the seismic shift for health information rapidly continues to change from paper to electronic, the health information management professional is at the forefront in the planning, design, development, and implementation of information systems. A variety of computer hardware and software components are available to ensure the effective storage, management, and transmission of healthcare data. Moreover, advances in programming languages are enabling users to communicate with and through computers with greater ease and efficiency.
The database is the primary foundation of the electronic health record. The most important functions of any healthcare information system involve being able to create, modify, delete, and view patient data. And the most important storage mechanism used to perform these functions is a database. The relational database is the most common type of database system. Most healthcare organizations maintain several different databases to serve different functions in their day-to-day operation. These databases are managed by a collection of computer programs called a database management system. The DBMS allows different application programs to access the database. An important purpose of a DBMS is to maintain the data definitions (data dictionary) for all the data elements in the database. It also enforces data integrity and security constraints. Increasingly, healthcare organizations are using data warehousing technology to access data from different databases for analyzing patient care delivery and business processes. The health information professional provides key support in the development and maintenance of databases and data warehouses.
Healthcare organizations must also have a technology infrastructure in place that provides the users of healthcare and business data with a communications system that enables them to share information. One example of such a system is a computer network in which computers are either clients or servers. The two main types of computer network setups are local-area networks (LANS) and wide-area networks (WANs).
Electronic commerce is revolutionizing healthcare in many ways through both business-to-customer and business-to-business transactions. The impact of e-commerce growth is reflected in increased numbers of consumers accessing the Internet for information, growth of health websites, increased use of portals by healthcare organizations providing a variety of healthcare services, and increased management of the supply chain through business-to-business applications. Health information management professionals play a principal role in planning, designing, and implementing e-health applications using electronic commerce technologies.
References
AHIMA e-Health Task Force. 2002. Report on the roles and functions of e-health information management. Chicago: AHIMA.
AHIMA e-HIM Work Group on EHR Data Content. 2006. Guidelines for developing a data dictionary. Journal of AHIMA 77(2):64A–64D.
Anderson, J.G., and C. Aydin. 2005. Evaluating the Organizational Impact of Health Care Information Systems, 2nd ed. New York: Springer.
Austin, C.J., and S.B. Boxerman. 2003. Information Systems for Health Services Administration, 6th ed. Chicago: Health Administration Press.
Davenport, T.H., and J. Glaser. 2002. Just-in-time delivery comes to knowledge management. Harvard Business Review 80(7).
Durkin, S., and B. Just. 2008. An IT primer for health information exchange. Journal of AHIMA 79(1):38–42.
Farishta, M. 2001. More than a database: Mining your data for decision-making success. Journal of AHIMA 72(10):28–32.
Friedman, C.P., and J.C. Wyatt. 1997. Evaluation Methods in Medical Informatics. New York: Springer.
Gaudreau, E., and D. Palermo. 2009. EHR fast track: Fundamentals of rapid implementation planning. Journal of AHIMA 80(8):40–43.
Hutchinson, S.E., and S.C. Sawyer. 2000. Computers, Communications & Information. Boston: Irwin McGraw-Hill.
Johns, M. 1999. The electronic health record process for health services delivery. In Electronic Health Records: Changing the Vision, edited by G.F. Murphy, M.A. Hanken, and K.L. Waters. Philadelphia: W.B. Saunders.
Johns, M.L. 1997. Information Management for Health Professions. Albany, NY: Delmar.
Johns, M.L. 2002. Information Management for Health Professions, 2nd ed. Albany, NY: Delmar Thomson Learning.
Kohn, D. 2009. How information technology supports virtual HIM departments. Journal of AHIMA 80(3):Web extra.
McLendon, K. 2000. E-commerce and HIM: Ready or not, here it comes. Journal of AHIMA 71(1):22–23.
Merrill, M. 2009. MIT ventures explore cell phone use in developing countries. Healthcare IT News. http://www.healthcareitnews.com/news/mit-ventures-explore-cell-phone-use-developing-countries
Nugent, L., M. Johnson, D. Maloney, and P. Goyal. 2007. Developing and implementing telehealth program: The VA’s story. AHIMA’s 79th National Convention and Exhibit Proceedings.
O’Brien, J.A. 2002. Management Information Systems: Managing Information Technology in the E-Business Enterprise. New York: McGraw-Hill.
Shakir, A.M. 1999. Tools for defining data. Journal of AHIMA 70(8):48–53.
Shimrat, O. 2009. Cloud computing and healthcare: Bad weather or sunny forecast. http://www.himss.org/content/files/Code%2093_Shimrat_CloudComputingandHealthcare_2009.pdf
Wager, K.A., F.W. Lee, and J.P. Glaser. 2005. Managing Health Care Information Systems: A Practical Approach for Health Care Executives. San Francisco: Jossey-Bass.
Weed, L. 1991. Knowledge Coupling. New York: Springer-Verlag.
Wold, G.H., and R.F. Shriver. 1994. The Healthcare Systems Planning Manual: Evaluating, Selecting, and Implementing Information Systems throughout the Organization. Chicago: Probus Publishing.
Additional Resources
Amatayakul, M. 1999. The Role of Health Information Managers in CPR Projects. Chicago: AHIMA.
Andrews, D., and S. Stalick. 2000. BPR project management: A radical approach to project team organization. American Programmer 8(6).
Briggs, B. 1999. 1999 Comprehensive Guide to Electronic Health Records. Edited by B. Hoehn. New York: Faulkner and Gray.
Burke, L., and B. Weill. 2000. Information Technology for the Health Professions. New York: Prentice-Hall.
DeJesus, E. 2000. The wireless e-connection. Healthcare Informatics 2:67–70.
Duggan, C.M. 2004. The constancy of change. Journal of AHIMA 75(1):40.
Goodman, P.S. 2000. Hello, Internet? Talking Web sites next challenge. Washington Post.
Hoehn, B, ed. 1999. 1999 Comprehensive Guide to Electronic Health Records. New York: Faulkner and Gray.
Loshin, D. 2004. Issues and opportunities in data quality management coordination. DM Review Magazine.
Morneau, K. 2000. MCSD Guide to Solutions Architectures. Boston: Thompson Course Technology.
Nakamoto, G. 2000. Building a legacy system. Health Management Technology 21(10):56.
Rollins, G. 2009. ARRA and the HIM workforce. Journal of AHIMA 80(10):26–30.
Welch, J.J. 1999. CPR systems: Which one is right for your organization? Journal of AHIMA
