A clinical decision support system (CDSS) is an application that analyzes data to help healthcare providers make decisions and improve patient care. It is a variation of the decision support system (DSS) commonly used to support business management. A CDSS focuses on using knowledge management to get clinical advice based on multiple factors of patient-related data. Clinical decision support systems enable integrated workflows, provide assistance at the time of care and offer care plan recommendations.
When using clinical decision support systems, data mining may be conducted to examine a patient's medical history in conjunction with relevant clinical research. Such analysis can then help predict potential events, such as drug interactions, or flag disease symptoms.
Purpose of CDSS
The purpose of a clinical decision support system is to assist healthcare providers, enabling an analysis of patient data and using that information to aid in formulating a diagnosis. A CDSS offers information to clinicians and primary care providers to improve the quality of the care their patients receive.
CDSS tools can, for example, offer reminders for preventive care, give alerts about potentially dangerous drug interactions and alert clinicians to possible redundant testing their patient has been scheduled to undergo. As such, using a CDSS can lower costs and increase efficiency.
Some providers deploy a CDSS to flag patients who were improperly diagnosed or either missed or was given the wrong dosage of medication. These errors are added to problem lists and are included in population health management (PHM) reports that can serve as the basis for improvement initiatives.
Why healthcare professionals use a CDSS
Clinicians use a CDSS to diagnose and improve care by eliminating unnecessary testing, enhancing patient safety and avoiding potentially dangerous and costly complications.
The use of clinical decision support systems increased after passage of the HITECH Act (Health Information Technology for Economic and Clinical Health Act), which stipulated that providers had to demonstrate the meaningful use of health IT by 2015 or face a reduction in Medicare reimbursements the following year.
Under meaningful use, providers must implement one clinical decision support rule, including diagnostic test ordering and the ability to track compliance with that rule. That rule, furthermore, should apply to a specialty or high-priority condition.
Some physicians may prefer to avoid overconsulting their CDSS -- instead, relying on their professional experience to determine the best course of care.
Knowledge-based vs. non-knowledge-based
There are two main types of clinical decision support systems. One type of CDSS, which uses a knowledge base, applies rules to patient data using an inference engine and then displays the results. Most knowledge-based CDSSes consist of a data repository, an inference engine and a mechanism to communicate, and they commonly operate under if-then rules.
For example, if the knowledge-based CDSS is trying to assess potential drug interactions, then a rule might be that if drug A is taken and drug B is prescribed, then an alert should be issued.
A CDSS without a knowledge base, on the other hand, relies on machine learning to analyze clinical data. An example of a non-knowledge-based CDSS is an artificial neural network, which learns how to perform certain tasks by considering specific examples, usually without being programmed with if-then or other task-specific rules. The artificial neural network instead analyzes patterns found in patient data to determine relationships between symptoms and a diagnosis.
CPOE with clinical decision support systems
Computerized physician order entry (CPOE) refers to a variety of systems to automate the medication ordering process. CPOEs ensure legible, complete orders by only accepting them in a standardized format, and CPOE systems can interface with CDSSes, further enhancing the efficacy and safety of patient care.
Per the analysis of a patient's information, clinical decision support systems can make suggestions on drug doses and frequencies and perform drug allergy checks, while also providing guidelines and reminders regarding corollary prescription orders -- e.g., glucose testing to coincide with an order for insulin.
Clinical decision support systems and electronic health records (EHRs) are often integrated to streamline workflows and make use of existing data sets. There are a growing number of CDSS functions that are built into EHR systems. Before purchasing a stand-alone CDSS, providers should plan for and eliminate any overlapping alerts that it might create while working alongside their EHR systems.
Benefits and drawbacks of a CDSS
Despite the benefits, there are also cons to implementing clinical decision support systems. The first challenge is that a CDSS must integrate with a healthcare organization's clinical workflow, which is often already complex. Some clinical decision support systems are stand-alone products that lack interoperability with reporting and EHR software. Also, the number of clinical research and medical trials being published on an ongoing basis makes it difficult to incorporate the resulting data into CDSSes in a timely manner. Furthermore, incorporating large amounts of data into existing systems places significant strain on application and infrastructure maintenance.
Another potential problem with a CDSS is alert fatigue for clinicians. The alerts triggered by a CDSS can overwhelm caretakers who also receive prompts from other technology systems. A study on the effectiveness of a CDSS, commissioned by the Agency for Healthcare Research and Quality (AHRQ), concluded that improperly using a CDSS can be more harmful than not deploying a CDSS at all.
This was last updated in July 2018
Continue Reading About clinical decision support system (CDSS)
- How software helps avoid clinical decision errors
- Medical imaging data part of clinical decision process
- AI in healthcare can support clinical conclusions
- Health informaticians and the value of clinical data
Dig Deeper on Healthcare IT systems and applications
-
Eli Lilly data strategy paves way for AI in drug discovery
By: John Moore
- decision support system (DSS)
By: TechTarget Contributor
-
Oracle builds CoVPN volunteer screening tool for NIH
By: Makenzie Holland
-
Prognos and Datavant partner to improve health data analytics
By: Sabrina Polin