Grand challenges in clinical decision support

From Clinfowiki
Jump to: navigation, search

This is a perspective by Dean F. Sittig, Adam Wright, Jerome A. Osheroff, Blackford Middleton, Jonathan M. Teich, Joan S. Ash, Emily Campbell, & David W. Bates. (2008) in the Journal of Biomedical Informatics entitled “Grand challenges in clinical decision support.” [1]

Introduction

The paper is a viewpoint about common challenges in clinical decision support (CDS). According to the authors, an advanced level of clinical decision support systems are needed to achieve a high quality of care, such as interoperability and development of longitudinal electronic health records EHR for all patients. Despite the potential benefits, the implementation and use of CDS in routine practice is lacking. This article will list the top 10 “grand challenges” under three large categories that were identified by the authors and clinicians by utilizing an iterative and consensus-building process.

The Top 10 Clinical Decision Support Grand Challenges

Improve the effectiveness of CDS interventions

1. Improve the human-computer interface A poor human-computer interface is one of the common reasons why clinicians override unsolicited CDS alerts. Similarly, CDS alerts without a follow-up or recommendations further interrupts the workflow, increases frustration, and subsequently forms a barrier to adoption. Therefore, clearer information displays and CDS alerts with follow-up action will reduce challenges in CDS.

2. Summarize patient-level information It is impossible for physicians to retain and recall all the patient information during a visit. A succinct summary is needed on each patient to reduce cognitive load and improve decision making.

3. Prioritize and filter recommendations to the user Automatically prioritize recommendations according to a ‘multi-attribute utility model’ by uniting patient- and provider-specific information to consider predictable mortality or morbidity reduction, patient preferences, cost of care, efficacy of the test or therapy, patient tolerability, workflow position, services covered by insurance plans, genetic and genomic concerns, clinician’s experience, and other issues.

4. Combine recommendations for patients with co-morbidities Incorporating two or more guidelines help improve care outcome in patients with comorbidities. This challenge requires new combinatorial, logical, or semantic approaches.

5. Use freetext information to drive clinical decision support A CDS with an automated text processing ability can significantly improve the quality of care as at least 50% of the patient data is documented in an unstructured format.

Create new CDS interventions

6. Prioritize CDS content development and implementation One of the disadvantages of CDS is the high cost for implementation and content development. Prioritizing the content can greatly reduce the cost and enable widespread use of CDS.

7. Mine large clinical databases to create new CDS Mere incorporation of guidelines in the CDS content is not enough. A complete, highly scientific, rigorously Evidence-based medicine large clinical database is needed to improve CDS interventions.

Disseminate existing CDS knowledge and interventions

8. Disseminate best practices in CDS design, development, and implementation Sharing or exchange of experiences about design, development, implementation, maintenance, and evaluation can facilitate in developing high-level CDS.

9. Create an architecture for sharing executable CDS modules and services Standardization of CDS interfaces enables its utilization more effectively.

10. Create internet-accessible clinical decision support repositories Design CDS that are easy to download, maintain, modify, install, upgrade and used on any Certification Commission for Healthcare Information Technology (CCHIT)-certified EHR product.

Conclusion

In order to achieve the full benefits of CDS, the stakeholders have to overcome common challenges described in this paper. Improving the efficacy of current CDS interventions, creating new CDS interventions, and pooling of resources through sharing CDS related experiences are essential for improved health outcomes.

Comments

I chose this paper because of its unique and thorough presentation. The common challenges of CDS are well addressed and a concise solution is offered to each challenge.

Related Articles

References

  1. Sittig, et.al. (2008). Grand Challenges in Clinical Decision Support. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2660274/ .