Innovative technology in clinical settings remains one of the primary factors for both increased efficiency and rising costs in the United States health care system. Practitioners and industry alike have critical roles in the development, implementation, and utilization of modern technologies. However, while there are federal agencies that monitor aspects of medical treatment and medication, healthcare technology remains mostly unregulated and its assessment practices are limited. In the modern economy, healthcare technology innovation has been left in the hands of free-market forces and the private sector influences. Clinical decision support system (CDSS) is one of the primary healthcare technologies that is experiencing widespread adoption and has shown to be efficient at improving the quality of care.
Clinical decision support systems are technological computer systems that help medical staff to make choices about the treatment of individual patients based on the variety of available information and evidence-based practice. CDSS can vary significantly in function and design, ranging from providing clinical information to complex analyses of administrative data. However, recently, systems have become sophisticated enough to perform a combination of functions and analyses (Berner & La Lande, 2016).
CDSS frameworks vary based on the timing and activity of the provided decision support. While some provide preliminary data before diagnosis and passively responds to input, another system may issue real-time alerts and actively guide treatment. CDSS consist of three parts: the knowledge base, reasoning mechanism, and communication mechanism. The knowledge base uses compiled information based on if-then algorithms (each action results in a specific reaction from the system). The reasoning mechanism is known as the inference engine is the framework for combining practitioner input with patient data and knowledge bases. Finally, the communication mechanisms are the method which the system uses to display information and alert the personnel who will be making the decision (Berner & La Lande, 2016).
Measurement and Evaluation
The assessment and evaluation of health care technologies such as the CDSS are based on the need to understand the value and financial impact on clinical settings. Evaluation of CDSS can be approached from a variety of angles and consists of the systemic collection of data about one of the system’s aspects and performance. It helps to gain insight into the system’s performance and function as well as influence the future development, design, and utilization of CDSS. The evaluation consists of a combination of empirical and observational data. For example, a management-based model focuses on the effects of CDSS on the clinical workflow and efficiency of the organization. It is a context-oriented approach that studies disruption, modifications to workflow processes, and impact on productivity. Meanwhile, a participant-oriented model focuses on the user experience and interface design aspects of the system. This type of evaluation is based on user feedback through interviews and discussion (Lobach, 2016).
Despite the critical nature of evaluations in assessing CDSS systems and determining their return on investment, these studies have unique challenges. Since CDSS are most often components of a larger hospital and technological system, it is difficult to establish the necessary level of control over the tools in order to limit access or isolate a specific component. Integration with electronic health records (EHR) and computerized physician order entry (CPOE) creates the necessity to ensure that CDSS is a well-functioning healthcare technology (Lobach, 2016). The three elements used to evaluate the user-technology interface of CDSS are:
- Mobility through consumer electronic devices,
- Dynamic checklist design,
- Rapid Response Team (RRT) integration and communication checklists.
The elements of CDSS outlined in the previous section, which has been found critical for the success of this health care technology, will be analyzed to determine their impact on functionality within the context of clinical support. Mobility through the use of consumer-level devices is critical to addressing numerous challenges. Mobile and tablet devices available to consumers are much cheaper, more intuitive, and ergonomic than specialized medical equipment. Handheld devices have experienced increased use over the last decade at the point-of-care in order to obtain guidance, information on medications, perform necessary calculations, and assess other clinical data. Handheld devices would require little formal training and can be implemented by frontline health workers who have shown to improve the quality of care through prevention, diagnosis, and treatment (Agarwal et al., 2018). Therefore, it would be significantly advantageous for clinical decision-support systems to be compatible with mobile devices and be linked with other databases such as EHR and COEP.
Another measurement of functionality is a dynamic checklist design. A typical support tool system uses decision trees or free-form search which are difficult to operate, slow on various mobile devices, and ineffective in emergency situations. Meanwhile, dynamic checklists help reduce medical and cognitive errors, allow prioritization and highly intuitive use, and can be easily integrated into clinical workflow. Dynamic checklists provide a degree of freedom and speed, with an easily adaptable interface that can be used to aid in the decision-making process. Finally, RRT integration and communication checklists allow the CDSS to be connected to the hospital communication system. This allows for efficient correspondence and receiving alerts by personnel who are working with the patient. RRT helps to quickly notify relevant staff in cases of emergency which increases response times. Meanwhile, communication checklists allow for standardized and efficient communication between nurses and doctors as guidelines are provided for how to prepare a patient for the physician’s treatment (Yuan, Finley, Long, Mills, & Johnson., 2013).
Suggestions for Improvement
In order to design efficient clinical decision support systems, the primary challenge remains to improve the human-computer interface. Research has shown that the user interface directly impacted the success of CDSS implementation in clinical settings. The system has to consider a variety of factors in a multi-dimensional approach to become applicable in a working hospital environment with a complex evaluation of clinical outcomes and overcoming barriers to implementation (Murphy, 2014).
CDSS interface must fit into the clinical workflow, practitioner needs, and optimization issues. These may include speed, intuitive design, availability or display of content and data, and point of care delivery. The best method of improving user interface designs is to conduct trials in real-life settings and receive comprehensive feedback from the staff which can indicate any issues with design, including interpretability, understandability, readability, and prioritization of information on the display. Furthermore, the decision support in terms of alerts and recommendations must be intuitive and easily comprehensible at a basic level of understanding (Murphy, 2014). The process of improvement and optimization for CDSS will take time before the full penetration of clinical settings. In addition, to overcome design and technical challenges, practitioners must transition to CDSS for practice guidelines. Since CDSS offers support based on evidence-based practice, it is critical to ensure adherence to recommendations, which remains a significant barrier to implementation.
Agarwal, S., Tamrat, T., Glenton, C., Lewin, S., Henschke, N., Maayan, N.,… Mehl, G. L. (2018). Decision-support tools via mobile devices to improve quality of care in primary healthcare settings. Cochrane Database of Systematic Reviews, 2. Web.
Berner, E.S., & La Lande, T.J. (2016). Overview of clinical decision support systems. In E. Berner (Ed.) Clinical decision support systems. Health informatics (pp. 1-18). Cham, Switzerland: Springer.
Lobach, D.F. (2016). Evaluation of clinical decision support. In E. Berner (Ed.) Clinical decision support systems. Health informatics (pp. 147-162). Cham, Switzerland: Springer.
Murphy, E. V. (2014). Clinical decision support: Effectiveness in improving quality processes and clinical outcomes and factors that may influence success. The Yale Journal of Biology and Medicine, 87(2), 187–197.
Yuan, M. J., Finley, G. M., Long, J., Mills, C., & Johnson, R. K. (2013). Evaluation of user interface and workflow design of a bedside nursing clinical decision support system. Interactive Journal of Medical Research, 2(1). Web.