Electronic health records (EHRs) are consuming an increasing amount of human and material resources across healthcare. This comes at a time when margins are dwindling with an increasing number of unemployed and noninsured patients, rising costs of supplies and equipment, and aging facilities in need of renovation. While many organizations are vying for meaningful use dollars, questions of EHR benefits over the staggering costs are inevitable and ongoing.
The ultimate benefit of the time, energy, and cost of an EHR is clinical intelligence (CI). CI is the electronic aggregation of accurate, relevant and timely clinical data into meaningful information and actionable knowledge in order to achieve optimal structures, processes, and outcomes.1,2 EHRs and other clinical information systems are rapidly increasing the amount of healthcare data, creating a powerful asset previously unattainable. These data will help build CI that will transform healthcare.
This article highlights opportunities for nurse leaders to ensure that CI in their organizations is being built for maximum returns.
The term CI is derived from the term business intelligence (BI), which has been around for more than 2 decades. BI is the process of translating business data into information and then into knowledge.1 The purpose of BI is to improve the timeliness and quality of inputs through technology to the decision processes of the organization.3 CI is analogous to BI with the exception of focusing on the use of clinical data.
There is a convergence of evidence supporting the necessity of CI coming from multiple sources including clinical practice, payors, legislators, and regulatory agencies. For clinicians, the rate of change in medical knowledge is staggering, and the gap between what is known and what must be known to optimally care for patients is widening.4 The passage of federal healthcare reform in March 2010 and particularly the establishment of accountable care organizations, medical homes, and value-based purchasing have further pushed the agenda of CI forward.5
Friedman’s6 fundamental theorem of biomedical informatics helps explain the concept of CI. According to Friedman, the performance of an individual, group or organization is better when working in partnership with an information resource than when working unassisted. The theorem only holds true if the information resource offers correct, reliable, and relevant information that the user does not already know. Friedman reports that the theorem can fail if the information resource is not used as intended by the end user or the quality of the information resource is lacking.
CI recognition and utilization are dependent on information resources and end user understanding. Three applications of CI are discussed below and include data management, practice-based evidence (PBE), and workforce preparation.
Data management refers to the input of timely, accurate, and relevant data with the express purpose of using it. The data that nurses enter into EHR or are downloaded from other clinical information systems, such as smart pumps and patient monitoring systems, can be used for multiple purposes. These include patient care assessment and planning, quality and safety, risk management, operational efficiencies, evaluation of staff performance, equipment and supply inventory management, billing, nursing research, and innovation.
Staff nurses play a critical role in data management and the development of CI. Their knowledge and appreciation of the use of data include understanding the significance and impact of data that are entered or downloaded into the EHR. Staff nurses are the largest contingent of healthcare workers who are involved in the building of CI through data management.
The value of CI to patient care cannot be overestimated, however; it requires timely and accurate data. To expect or accept less severely jeopardizes opportunities for early identification and prevention of patient complications, loss of efficiencies in lengths of stay, and so forth.
There are several important steps to improving data entry. Researchers have demonstrated that data documentation by staff nurses, including a reduction in missing data, can be improved through training and education.7,8 Improved usability and workflow of data in the EHR are necessary to optimize data entry.9
CI also requires access to aggregated data in a usable format. When indicated, actionable views of data should be pushed forward to clinicians at the point of service. Data analytics useful for clinical practice include descriptive analytics, which permit end users to query databases and find answers to retrospective questions such as the following: What happened? How often did it happen? And where did it happen?10 Predictive analytics promote a more prospective view and allow users to get answers to questions such as what might happen if the data continue to trend in the same manner.10 Prescriptive analytics are the ultimate goal of CI providing users with optimal solutions to actual or potential clinical issues.10
Data management begins with the design of the EHR. Good design is essential for sound data analytics, promoting superior patient outcomes and operational efficiencies. High-quality data form the basis for PBE, another key outcome of CI.
PBE refers to the use of aggregated clinical data to make healthcare decisions. PBE differs from evidence-based practice (EBP) in several important ways. Evidence-based practice based on research typically involves patient inclusion and exclusion criteria, well-defined treatment protocols, a limited number of variables and is powered for the outcome of interest and elimination of confounders in the study design or limitations.11 In contrast, PBE includes an analysis of data from all patients, all variations in treatments and interventions, and all variables and confounders captured in the EHR, representing the real world of healthcare.
There are significant gaps in the current evidence12 regarding CI. Negative research findings are less frequently published. By contrast, typically clinical trials with significant results are published and can influence a meta-analyses or syntheses of data.
PBE is an important part of CI and is dependent upon sound data management. PBE is useful in analyzing EBP as the focus moves from the more controlled research environment to the less controlled practice environment. Data collection, analysis, presentation, and use must be seamlessly integrated at the point of care. All require new directions for the nursing workforce.
Nursing Workforce Preparation
It should be clear that staff nurses play a significant role in building CI. Broad preparation in nursing informatics, including usability and workflow of EHRs, is essential. Specific education on data management and PBE should be included.
Nurse managers will take on an increasing role in monitoring and feedback ensuring that data are input in an accurate, timely, and appropriate manner. Monitoring can be automated with alerts or reports identifying progress and opportunities for improvement. Equally important will be the routine feedback to staff nurses on how they are progressing.
Chief nursing informatics officers are essential in developing and communicating the vision for CI and overseeing the design, implementation, evaluation, and optimization of EHRs and other clinical information systems, driving usability and workflow and reducing the amount of time and effort required to input relevant data. They must be part of the EHR governance structure and present when allocation of resources is being discussed to present the business case for CI.
For CNOs, incorporating CI into their strategic plan for the organization is essential. Leaders should begin operationalizing the vision by supporting the development of technology and information savvy nursing staff and promoting managerial oversight of data management. Investing in the hiring and development of well-educated nursing informaticists is a key strategy to drive success in CI use and application.
The volume of clinical data in digital formats will continue to proliferate at increasing speeds as more EHRs and other clinical information systems are deployed and improved. Failure to actively engage an agenda of increasing accuracy and timeliness of data entry into the EHR could create a landslide of poor quality data, resulting in missed opportunities to improve clinical outcomes and operational efficiencies. An understanding of PBE is similarly important in avoiding prematurely freezing the data being collected in EHRs through the use of misaligned EBPs, thereby missing better solutions. Preparation of the workforce in new areas is also essential in building CI. Although the full benefits of CI are still in the future, the journey is beginning.
1. Golfarelli M, Rizzi S, Cella I. Beyond data warehousing: what’s next in business intelligence? In: Proceedings 7th International Workshop on Data Warehousing and OLAP (DOLAP 2004). Washington, DC; 2004.
2. Donabedian A. An Inroduction to Quality Assurance in Health Care. New York, NY: Oxford Univesity Press; 2002: 46–47.
3. Negash S. Business intelligence. Commun Assoc Inform Syst. 2004; 14: 177–195.
4. Smallwood RA. The realities of evidence-based medicine. Asia Pacific Heart J. 1998; 7 (2): 121–124.
5. Hagland M. Got clinical intelligence? Applying business intelligence tools to the core of operations = revolution. Healthc Inform. 2011; 28 (3): 28–32.
6. Friedman C. A “fundamental theorem” of biomedical informatics. JAMIA. 2009; 169–170.
7. de Lusignan S, Chan T, Wood O, et al.. Quality and variability of osteoporosis data in general practice computer records: implications for disease registers. Public Health. 2005; 119: 771–780.
8. Forster M, Bailey C, Brinkhof M, et al.. Electronic medical record systems, data quality and loss to follow-up: survey of antiretroviral therapy programmes in resource-limited settings. Bull World Health Organ. 2008; 86: 939–947.
9. Chan KS, Fowles J, Wiener J. Electronic health records and the reliability and validity of quality measures: a review of the literature. Med Care Res Rev. 2010; 67 (5): 503–527.
10. Apte C. The role of machine learning in business optimization. Proceedings of the 27th International Conference on Machine Learning. 2010. Available at http://www.icml2010.org/papers/903.pdf
. Accessed May 22, 2011.
11. Horn S, Gassaway J. Practice-based evidence study design for comparative effectiveness research. Med Care. 2007; 45: S50–S57.
12. Tunis S, Stryer D, Clancy C. Practical clinical trials: increasing the value of clinical research for decision making in clinical and health policy. JAMA. 2003; 290: 1624–1632.
© 2011 Lippincott Williams & Wilkins, Inc.