[[HOPES Project Index]] [[Introduction to Clinical Playbook]] --- Reading - https://topflightapps.com/ideas/clinical-decision-support-system-implementation/ - CDSS have the following key elements: - **Fundamental**: - Order sets - The ability to recommend/suggest treatment/medication/tests based on patient's condition and medical history - Clinical Guidelines - The ability to provide guidelines on managing that specific condition. - Drug interaction Alerts - Dosage Reminders - based on patient's weight, age, kidney function - Diagnostic Support - Provide differential diagnoses that clinicians might consider - **Add On Features✱ - Patient safety alerts - alerts for allergies, contraindications, duplicate therapies - Predictive Analytics - Predict outcomes and suggest preventive measures - Documentation Support - Population Health Management - managing and analyzing health data of populations looking for trends, allocate resources, plan interventions Different Types of CDSS 1. Knowledge Based 2. AI Driven Based Different Form: 1. Standalone Software Suite -- 2. Integrated Module -- within electronic health records Different Mode 1. Passive -- only when clinicians opt to review at their discretion 2. Active - pushes alerts directly into clinical workflow Target User 1. Clinician Focused CDSS - 2. Non-Clinician CDSS -- tools to help patient manage chronic condition. personalised health advice based on real time data. --- First step is to identify the need for a CDSS. Why do we need one? ---- https://www.england.nhs.uk/long-read/supporting-clinical-decisions-with-health-information-technology/ - 6 Is - inclusive, intelligent, interpretable, integrated, impactful, interoperable - 5 core principles - providing the right information, to the right person, in the right format, through the right channel, at the right time. - "**Right information:** evidence or guideline based CDS that incorporates stakeholder inputs and aligns with current improvement initiatives. - **Right person:** CDS must meet the specific needs of everyone in the multidisciplinary clinical team. Users should be aware of the purpose and limitations of the CDS, and how to report issues with its advice. - **Right format:** CDS formats must present information to clinicians in a manner that complements workflows; for example, best practice alert, visual dashboard, smart order set, customised referral. Those writing CDS advice should be aware of good practice in risk communication (for example, use 1 in 100 rather than 1% risk) and avoid interrupting clinical tasks unless there is an immediate risk of serious error. - **Right channel:** User experience needs to be considered. CDS should focus on the context of use, to ensure presented information does not disrupt workflows. Where possible and appropriate, it should be integrated within the primary EHR system. - **Right time:** CDS must be integrated and seamless within the clinical workflow. CDS should operate across different care settings. - Really must know the WHY is there a need for CDSS first before doing any work. - "### Defining the problem An organisation will need to establish existing technological, cultural, workforce and governance capabilities, to shape expectations of how CDS can be introduced and highlight where investment is required to maximise the potential of digital interventions. Technology can rapidly analyse data and present actionable insights to support clinical decision-making. However, this only works if the CDS system ‘asks’ a specific and well-defined question. Thus, understanding current workflows and local policy to define the impact of CDS systems is critical. A design-led approach can be used to define a CDS use case. This starts with gaining insights into the rate, severity and causes of the clinical problem in question from those closest to the clinical situation, the end users. This includes considering the wider change management, education and training that may be required for successful implementation, as well as exploring alternatives to CDS. Through this process there is an opportunity to reflect on current workflows and redefine ways of working if required. This needs to be an iterative process and through end-user input a solution-free problem definition should be agreed and documented to guide design of the preferred decision support solution." - CDSS should be interoperable across care settings, present a "single source of truth" to clinicians. If not integrated, it will be a barrier. - Information should be interoperable across care settings, to present a ‘single source of truth’ to clinicians. Poor integration with existing IT systems and EHRs is a key barrier to CDS adoption ([Sharma et al, 2021](https://informatics.bmj.com/content/28/1/e100253)). - Common Barriers - Alert Fatigue - "### Managing alert fatigue Many CDS systems provide clinicians with information at specific timepoints to support their decision-making. However, if clinicians are presented with inappropriate alerts or an overwhelming number of alerts they can suffer alert fatigue ([Khalifa and Zabani, 2016](https://pubmed.ncbi.nlm.nih.gov/27350464/)). If an appropriate ‘threshold’ of sensitivity and sensitivity is not set, up to 95% of alerts are inconsequential, and sometimes clinicians will distrust, disagree with or ignore them ([Ash et al, 2007](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2813668/)). The CDS system design needs to ensure that only clinically meaningful recommendations are presented. Alerts should appear at the right time and provide context-relevant information that influences clinical decision-making, in a practical, standardised format. If not managed appropriately, there is a risk the CDS system has unintended consequences, most commonly manifest as ‘over-alerting’. The context in which a CDS system is being implemented is an important consideration; they form a part of wider care pathways in a complex adaptive environment. CDS systems that trigger out of context and do not consider multiple patient factors will likely lead to inappropriate alerts, and with this users can lose confidence in the alerts and thus not engage with them even when they are relevant ([Wasylewicz et al, 2018](https://pubmed.ncbi.nlm.nih.gov/31314237/))."