Clinical Decision Support Systems: Clinical Application Case Studies

The comprehensive implementation and evaluation of Clinical Decision Support Systems (CDSSs) across diverse medical environments are best understood through specific clinical application case studies. There are many examples detailing the varied objectives, interventions, and measured impacts of CDSS deployment, ranging from optimising medication safety to improving chronic disease management in specialised and generalist settings.

I. Medication Management and Patient Safety

CDSS application in medication management is arguably the most common and robust area of study, often focusing on reducing high-risk errors such as Adverse Drug Events (ADEs) and drug-drug interactions (DDIs).

High-Risk Prescribing and DDI Management

CDSS interventions commonly target high-risk medication scenarios:

• Drug-Drug Interactions (DDIs): CDSSs are widely implemented to detect potential DDIs, particularly concerning complex therapies like Direct Oral Anticoagulants (DOACs) combined with CYP3A4/P-gp inhibitors and inducers, to prevent serious ADEs. Evaluation studies compare different commercial CDSS tools, revealing statistically significant concordance but poor agreement between them (e.g., Lexicomp and Medscape). This suggests that reliance solely on lower reliability programs may lead to incorrect guidance. Expert review further emphasises that automated systems often generate high numbers of non-clinically relevant alerts (alert fatigue), requiring refinement to incorporate patient-specific co-factors like comorbidities and laboratory results to increase specificity.

• Specialised Drug Dosing: In complex fields like transplant medicine, CDSS tools for immunosuppressant management (e.g., cyclosporine) are critical. Systems have shown feasibility and utility in aiding physicians to identify clinically relevant DDIs in kidney transplant patients. Additionally, a CDSS proved capable of identifying and correcting deficiencies in cefazolin prophylaxis in surgical patients, resulting in cost savings associated with inappropriate prescribing.

• Reducing Unnecessary Use: CDSS implemented in a tertiary care centre demonstrated effectiveness in reducing the use of high-cost medications like intravenous acetaminophen and unnecessary nasal MRSA PCR testing, aiding in cost containment and resource optimisation.

Antimicrobial Stewardship (AMS)

CDSS plays a vital role in AMS programmes across primary and acute care settings, facilitating appropriate antimicrobial selection and dosing.

• Behavioural Change: CDSS interventions have been shown to lead to immediate decreases in inappropriate prescriptions, such as azithromycin. For urinary tract infections (UTIs) in primary care, CDSS informed prescribers in real-time about the ecology and surveillance of E. coli resistance, underlining the risk of using certain antibiotics empirically.

• System Effectiveness: While CDSS for AMS shows promise in optimising antimicrobial use and enhancing clinical decision-making, evaluation must be stringent; some studies report no significant impact on antibiotic use or guideline compliance, highlighting the need for systems adapted for non-specialised prescribers.

II. Chronic Disease Management and Public Health Screening

CDSS applications in chronic care often involve supporting adherence to complex, evolving guidelines for conditions that require long-term monitoring.

• Cardiovascular Disease (CVD) and Stroke Prevention: Numerous studies focus on CVD prevention. CDSS interventions, particularly those embedded in EHRs, have shown effectiveness in improving compliance with antihypertensive treatment, particularly in primary care and resource-constrained settings. In the management of Atrial Fibrillation (AF), CDSSs prompt physicians to prescribe appropriate oral anticoagulants for stroke prevention. Notably, a cluster randomised trial in Community Health Centres (CHCs) found that a CVD CDSS, despite low overall adoption (19.8% of encounters), was associated with improved reversible CVD risk among socioeconomically vulnerable patients with high baseline risk when the tool was actually used.

• Cancer Care and Screening: CDSS tools facilitate shared decision-making (SDM) in lung cancer screening (LCS) by characterising values clarification methods and helping patients weigh complex trade-offs of benefits and harms. CDSSs are also used in managing complex conditions like breast cancer, assisting Multidisciplinary Tumour Boards (MTBs) and providing patient-specific treatment recommendations. For cervical cancer, different CDSS models implemented in EHRs were compared, showing variability in accuracy depending on the data extraction methods (Natural Language Processing vs. standardised fields).

• Diabetes and Kidney Management: CDSS aids in diabetic care by helping clinicians manage home blood pressure (HBP) records for hypertension and ensuring appropriate dosing of renally cleared drugs for patients with Chronic Kidney Disease (CKD). Studies also highlight the effectiveness of CDSS in improving process measures for Type 2 Diabetes Mellitus (T2DM) care in primary settings.

III. Acute Care, Prognosis, and Specialised Fields

In acute and specialised settings, CDSS is deployed primarily for rapid diagnosis, risk stratification, and timely adherence to life-saving protocols.

Acute Care and Critical Decisions

• VTE Prophylaxis: Computerised CDSSs (CCDSSs) have been demonstrated to increase the proportion of surgical patients who receive adequate Venous Thromboembolism (VTE) prophylaxis, correlating with a reduction in VTE events.

• Acute Kidney Injury (AKI): CDSS implementation significantly reduced mortality and increased the proportion of patients receiving AKI recognition and investigations, demonstrating strong clinical outcomes.

• Sepsis Detection: CDSSs such as POC Advisor and Modified Early Warning System (MEWS) are evaluated for identifying sepsis risk and influencing time to treatment, although the potential for alarm fatigue remains a challenge.

• Diagnosis and Prognosis: CDSSs support diagnostic reasoning in the Emergency Department (ED), often leading to better clinical management and patient outcomes. For instance, implementing evidence-based CDS in the ED was associated with a significant decrease in the use and increase in the yield of CT pulmonary angiography for evaluating acute Pulmonary Embolism (PE).

Nursing and Allied Health Applications

While often overlooked compared to physician-focused tools, CDSS for nurses and allied health professionals is critical for care efficiency and patient monitoring.

• Workload Optimisation: CDSS designed for Nurse-to-Patient Assignment (NPA) showed positive outcomes, including improved workload distribution, enhanced nursing efficiency, and reduced overtime, contributing to greater nurse satisfaction.

• Symptom Management: Machine learning-based CDSS has demonstrated effectiveness in improving nurses' decision-making accuracy and efficiency in complex scenarios, such as symptom management for patients receiving Transarterial Chemoembolisation (TACE).

• Usability and Impact: Studies targeting nurses in Intensive Care Units (ICUs) and Nursing Homes (NHs) suggest positive effects on patient care outcomes and process measures (e.g., guideline compliance). However, findings on improvement in specific outcomes like pressure ulcer incidence remain limited or non-significant, highlighting the need for better adoption and sustained interaction by nurses.

Surgical and Imaging Support

In surgical contexts, CDSS serves to standardise procedure, minimise variation, and support complex planning.

• Perioperative Care: CDSS implementation in perioperative settings is associated with improved guideline adherence, decreased medication errors, and improvements in patient safety measures, though significant impacts on postoperative mortality were not observed.

• Imaging Referral: A multicentre cluster randomised trial (MIDAS study) evaluated the implementation of CDSS for imaging referral (ESR iGuide), emphasising that success requires addressing technical issues, local workflow adaptations, and user acceptance. CDSS has also been shown to reduce unnecessary CT scans for head trauma.

IV. Evaluation and the Implementation Gap

Across all these clinical applications, a recurrent theme is the necessity for robust evaluation before implementation.

• Evidence Gap: Despite numerous successes in improving process measures (e.g., guideline adherence, appropriate orders), the evidence linking CDSS to improvements in hard clinical outcomes (e.g., mortality, morbidity) remains mixed or inconsistent, often due to a lack of rigorous, long-term randomised trials.

• Implementation Challenges: Many successful prototypes (e.g., those using text mining or machine learning) fail to achieve routine clinical use due to barriers like complexity, insufficient collaboration between developers and end-users, lack of appropriate training, and poor fit with clinical workflows. For a CDSS to be truly successful in clinical practice, it must be an interpretable and efficient tool that augments clinical judgment, requiring addressing end-user needs for model transparency and workflow integration.

Bibliography

Abell, B., Naicker, S., Rodwell, D., Donovan, T., Tariq, A., Baysari, M., ... & McPhail, S. M. (2023) 'Identifying barriers and facilitators to successful implementation of computerized clinical decision support systems in hospitals: a NASSS framework-informed scoping review', Implementation Science : IS, 18(1), p. 32.

Ali, S. I., Jung, S. W., Bilal, H. S. M., Lee, S. H., Hussain, J., Afzal, M., ... & Lee, S. (2021) 'Clinical Decision Support System Based on Hybrid Knowledge Modeling: A Case Study of Chronic Kidney Disease-Mineral and Bone Disorder Treatment', International Journal of Environmental Research and Public Health, 19(1).

Bienvenu, A. L., Ducrocq, J. M., Augé-Caumon, M. J., & Baseilhac, E. (2025) 'Clinical decision support system to guide antimicrobial selection: a narrative review from 2019 to 2023', The Journal of Hospital Infection, 162, pp. 140–152.

Bright, T. J., Wong, A., Dhurjati, R., Bristow, E., Bastian, L., Coeytaux, R. R., ... & Lobach, D. (2012) 'Effect of clinical decision-support systems: a systematic review', Annals of Internal Medicine, 157(1), pp. 29–43.

Cai, J., Li, P., Li, W., Zhu, T., & Li, X. (2024) 'Systematic review and meta-analysis of the impact of clinical decision support systems on perioperative care', International Journal of Surgery (London, England), 110(12), pp. 8057–8072.

Conway, N., Adamson, K. A., Cunningham, S. G., Emslie Smith, A., Nyberg, P., Smith, B. H., ... & Wake, D. J. (2018) 'Decision Support for Diabetes in Scotland: Implementation and Evaluation of a Clinical Decision Support System', Journal of Diabetes Science and Technology, 12(2), pp. 381–388.

Ferraris, V. A. (2023) 'The Value of Decision Support Tools for Deciding About Lung Cancer Treatments', The Annals of Thoracic Surgery, 115(2), pp. 308–309.

Gholamzadeh, M., Abtahi, H., & Safdari, R. (2023) 'The Application of Knowledge-Based Clinical Decision Support Systems to Enhance Adherence to Evidence-Based Medicine in Chronic Disease', Journal of Healthcare Engineering, 2023, 8550905.

Jiménez, D., Resano, S., Otero, R., Jurkojc, C., Portillo, A. K., Ruiz-Artacho, P., ... & Yusen, R. D. (2015) 'Effect of a clinical decision support algorithm on the use and yield of CT pulmonary angiography in the emergency department', Thorax, 70(9), pp. 909–911.

Kruse, C. S., & Ehrbar, N. (2020) 'Effects of Computerized Decision Support Systems on Practitioner Performance and Patient Outcomes: Systematic Review', JMIR Medical Informatics, 8(8), e17283.

Kwon, H., & Lee, D. (2025) 'Clinical Decision Support System for Clinical Nurses' Decision Making on Nurse-to-Patient Assignment: A Scoping Review', Studies in Health Technology and Informatics, 329, pp. 1674–1675.

Molloy, M. J., Muthu, N., Orenstein, E. W., Shelov, E., & Luo, B. T. (2024) 'Clinical Decision Support Principles for Quality Improvement and Research', Hospital Pediatrics, 14(4), e219–e224.

Sarıköse, S., & Çelik, Ş. S. (2024) 'The Effect of Clinical Decision Support Systems on Patients, Nurses, and Work Environment in ICUs: A Systematic Review', Computers, Informatics, Nursing : CIN, 42(4), pp. 298–304.

Susanto, A. P., Lyell, D., Widyantoro, B., Berkovsky, S., & Magrabi, F. (2024) 'How Well Do AI-Enabled Decision Support Systems Perform in Clinical Settings?', Studies in Health Technology and Informatics, 310, pp. 279–283.

Sutton, R. T., Pincock, D., Baumgart, D. C., Sadowski, D. C., Fedorak, R. N., & Kroeker, K. I. (2020) 'An overview of clinical decision support systems: benefits, risks, and strategies for success', NPJ Digital Medicine, 3, 17.