Introduction
Decision support systems (DSS) are
automated tools designed to support decision-making activities and improve the
decision-making process and decision outcomes.
Clinical decision support systems (CDSS) is
designed to support healthcare providers in making decisions about the delivery
and management of patient care. A CDSS program’s goals may include patient
safety and improved outcomes for specific patient populations as well as
compliance with clinical guidelines, standards of practice and regulatory requirements.
The primary goal of CDSS is the optimization of both efficiency and
effectiveness with which clinical decisions are made and care is delivered (Tan
and Sheps, 1998).
Clinical tasks to which CDSS may be applied
include:
·
Alerts and reminders
·
Diagnostic assistance
·
Therapy critiques and plans
·
Medication orders
·
Image recognition and
interpretation
·
Information retrieval
Nursing decision support systems (NDSS) are
tools that help nurses improve their effectiveness, identify appropriate
interventions, determine areas in need of policy or protocol development and
support patient safety initiatives and quality improvement activities.
Accuracy, timeliness, availability, format and reliability of information are
just as important to nursing as they are to other healthcare providers. Nurses
as knowledge workers need access to current knowledge at the point-of-care
delivery where it is useful.
Definition
A CDSS may be defined as any computer
program that helps health professionals make clinical decisions.
·
Johnsons et al. (1994) defined
it as “computer software employing knowledge base designed for use by a
clinician involved in patient care as a direct aid to clinical
decision-making”.
·
Sims et al. (2001) broadened
the definition to “CDSS are software designed to be a direct aid to clinical
decision-making, in which the characteristics of an individual patient are
matched to a computerized clinical knowledge base and patient-specific
assessments or recommendations are then presented to the clinician or the patient
for a decision”.
·
Coiera (1994) discussed the
role of CDSS as augmenting human performance and providing assistance for
healthcare providers especially for tasks subject to human error.
·
Berrier (1999) healthcare
provider is being transformed through information and knowledge management and
technology is being used to “tame data and transform information”.
Expanded Use of CDSS
DSS integrate and capitalize on the
strengths of both human and computer. Thus, three main purposes of a DSS are to
1.
Assist in problem solving with
semi-structured problems.
2.
Support, not replace, the
judgement of a manager or clinician.
3.
Improve the effectiveness of
the decision-making process.
History of CDSS
One of the earliest known CDSS designed to
support diagnosis of acute abdominal pain was developed by de Dombal in 1972 at
Leeds University. This system used Bayesian theory to predict the probability
that a given patient, based on symptoms had one of seven possible conditions.
The system was correct in its diagnosis in 91% of the cases compared to the
physicians whose accuracy ranged from 65% to 80% depending on their experience
in practice (Musen, Shahar and Shortliffe 2001).
By 1971, INTERNIST I was developed at the
University of Pittsburgh to support the diagnostic process in general internal
medicine by linking diseases with symptoms.
MYCIN, a rule-based expert system to
diagnose and recommend treatment for certain blood infections was functional in
1976.
In nursing, two early and well known systems, COMMES (Creighton online multiple modular expert system) (Evans, 1990) and computer aided nursing diagnosis and intervention (CANDI) (Chang, 1988) were developed to assist nurses with care planning and nursing diagnosis.
A series of systems addressing specific
clinical issues were developed next. These included
·
ONCOCIN developed for oncology
protocol management at Stanford
·
CASNET developed at Rutgers
University for diagnosis and treatment of glaucoma
·
ABEL an expert system developed
at MIT that used causal reasoning to manage acid-base and electrolyte imbalance
Types and characteristics of DSS
Types of DSS
1.
Administrative and
Organizational System. Included in the field of healthcare decision support are
systems that support organizational, executive/managerial, financial and
clinical decisions. Administrative systems, including those designed for
finance or quality monitoring, generally support the business decision-making
process. These systems encompasses decision processes other than direct patient
care delivery and even in clinical in nature, such as quality improvement
systems, are mainly used for strategic planning, budgeting, financial analysis,
quality management, continuous process improvement and clinical bencmarking.
2.
Integrated systems. More
recently, healthcare agencies have begun to understand that combination systems
offer optimal value to the organization. Such systems are able to support
outcomes performance management by integrating operating data (the business
side) – budgeting, executive decision-making, financial analysis, quality
management, and strategic planning data, with clinical data (the clinical side)
– clinical event tracking, results reporting, pharmaceutical ordering and
dispensing, differential diagnoses, real-time clinical pathways, literature
research and clinical alerts.
Characteristics of DSS
These systems can be studied based on their
structure, their organization, their content or their purpose. Shortliffe
(1990) uses function mode of advice, consultation style, underlying
decision-science methodology and user-computer interactions to categorize
systems. Teich and Wrinn (2000) examine DSS from the aspects of functional and logical classes and structural
elements.
Functional class includes feedback provided to clinician, the organization of the data, the extent of proactive information provided, the intelligent actions of the system and the communication method.
Logical class includes substitute therapy
alerts, drug family checking, structured entry, consequent actions, parameter
checking, redundant utilization checking, relevant information display,
time-based checks, templates and order sets and profile display and analysis,
rule-based event detection and aggregate data trending.
Structural elements according to Teich and Wrinn include triggering, dispatching, rule logic, process control, notification/acknowledgement, action choices, action execution, and rule editor.
Key CDSS functions
Perreault (1999) organized key CDSS functions as:
- Administrative – support for clinical coding and documentation
- Management of clinical complexity and details – keeping patients on research and chemotherapy and protocols, tracking orders, referrals, follow-up and preventive care
- Cost control – monitoring medication orders and avoiding duplicate or unnecessary tests
- Decision support – supporting clinical diagnostic and treatment plan processes promotion of best practices, use of condition-specific guidelines, and population-based management.
From ontologic perspective (Tan and Sheps,
1998), DSSs could be divided into data-based (population-based), model-based
(case-based). Knowledge-based (rule-based) and graphics-based systems.
- Data-based systems capitalize on the fundamental input into any intelligent system data. They provide decision support with a population perspective and use routinely collected longitudinal, cohort and cross-sectional databases. Population-based information is used to enhance clinical decision-making, “funnel” patients to medical care and enhance medical practice.
- Model-based DSSs are driven by access to and manipulation of a statistical, financial, optimization, and/or simulation model. The data in this instance are compared to various decision-making and analytic models. A model is a generalization that can be used to describe the relationships among a number of observations to represent a perception of how things fit together.
- Knowledge-based systems rely on expert knowledge that is either embedded in the system or accessible from another source and uses some type of knowledge acquisition process to understand and capture the cognitive processes of healthcare providers. Much of what we consider evidence-based practice (EBP) refers to knowledge-based decision support
- Graphics-based systems take advantage of the user interface to support decisions by providing decision “cues” to the user in the form of color, graphical representation options and data visualization.
Examples of CDSS applications:
1.
Reminders and alerts which are
computer tools for focusing attention such as “flags” for abnormal values.
2.
Therapy critiquing and planning
as well as care, maps, guidelines, protocols and so on.
3.
Diagnostic assistance providing
patient-specific consultations using diagnostic or management tools, such as
Problem Knowledge Couplers (PKC) (Weed, 1991).
4.
Lab systems with interpretation
of measured values and automated preparation of reports as well as physician
guidance as to which tests to order.
5.
Prescribing decision support
such as drug advisory systems used for advising on drug-drug interactions, side
effects, selecting most cost-effective drug.
CDSS
Impact on Clinicians and Clinical Decisions
Need for Evidence-based Practice
The use
of CDSS to find and prevent errors related to gaps between optimal and
actual practice can result in improved quality of care (Bates et al., 2001).
Applications of CDSS suggest the ability to lessen the incidence of adverse
drug events, nosocomial infections, and the inappropriate use of antibiotics.
Prevention of prescription errors is seen as one of the most valuable and
widely used functions of CDSS (Lippman, 2000). Bates et al. (2003) have
determined that effective clinical decision support depends on CDSS speed,
anticipation of information needs, real-time delivery, usability, simplicity
and the maintenance of the knowledge-based system.
Barriers
to the Use of CDSS Systems
Barriers identified included
·
lack of noticeable benefits
·
insufficient cost benefits
·
inadequate staff training
·
lack of system support
·
system costs
·
lack of exposure to technology
Coiera (2003) cites lack of congruence with
work flow and the additional effort required by the user to fit the system into
the usual care process as a barrier to use of CDSSs. The involvement of
healthcare professionals in CDSS
selection is essential to system acceptance. It is also important to consider
how a CDSS will affect organizational culture, practice and personnel
attitudes.
Evaluation
of CDSS
There is a need to evaluate CDSS with sound
studies. The effect of CDSS on clinical outcomes currently remains uncertain
without valid and generalizable findings. Increasing pressure to deliver
quality care at the lowest possible cost has led to the implementation of CDSS
to drive appropriate process improvement activities needed to achieve
successful care outcomes (Healthcare Financial Management). Three key areas to
assess in CDSS development are Cost
Management, Quality Management and Information
Management.
One consideration in the evaluation of a
system is the assessment of how system handles continuous variables, like age.
Needed is a series of probabilities outlining the risks at a variety of
decision points.
Sittig (1999) cites the following five
elements as necessary, but not sufficient for a real-time clinical decision
support system.
1.
Integrated real-time patient database which
combines patient data from multiple sources, lab, radiology, pharmacy,
admissions, nursing notes, and so on. This is needed to provide context for
results interpretation.
2.
Data-drive mechanism that allows event
triggers to go into effect and activate alerts and reminders automatically.
3.
Knowledge engineer who can translate the
knowledge representation scheme used in the system so that the clinical
knowledge in the system can be extracted ad translated into machine executable
logic.
4.
Time-driven mechanism to permit
automatic execution of programs at a specific time to alert provider to carry
out a specific action or insure that the action had been completed.
5.
Long-term clinical data repository – data
collected over time form a variety of sources allowing a longitudinal patient
record.
Responsibility
of User: Ethical and Legal Issues
CDSS are considered similar to medical
devices but the legal responsibility for treatment and advice given to a
patient rests with the clinician regardless of whether a CDSS is used (Hunt et
al., 1998). Still unknown are the legal ramifications of not following CDSS
advice. There seems to be no major adverse effect from the use of a CDSS;
however, such systems must be developed with high standards of quality. The
provision of erroneous information and/or incorrect guidance does have the
potential for harmful impact. CDSS will be expected to comply with a “duty of
care” if it is to become safely integrated into routine patient care. The
knowledge base of healthcare changes frequently and often past practices are
proven ineffective and perhaps even hazardous. Healthcare knowledge may also be
based on professional judgement without objective scientific evidence.
Therefore, the knowledge base of a CDSS must be as reflective as possible of
the current state of professional and scientific opinion and evidence and must
draw on traditional knowledge sources such as journals and texts to maintain
currency.
Safe use of CDSS may include such
techniques as limiting access, developing audit trails, monitoring use and
clinical hazard alerts. It is possible that a CDSS may be considered a
specialized piece of clinical equipment requiring certification for use.
Implications
for Future Uses of CDSS in Nursing
Increasing Inclusion of Patients
CDSS systems of the future may also allow
patient access to the knowledge base of the system. The computer can become a
patient health medium with reference databases, library access for healthcare
information, drug and disease management information, self-help programs and
advice about prevention available.
CDSS applications can be developed for a
spectrum of clinical users including nurses. No specific CDSS for nursing
currently exists but the potential for future nursing applications is vast if
certain basic issues are resolved. This include definition of the nursing
knowledge base and how nursing knowledge can be represented and how structure
can be added to nursing knowledge.
Dual purpose of documentation
