Here are 10 lessons learned for achieving patient safety improvements with biomedical device integration.
By Terri Gocsik, CRNA, MS, CPHIMS
Manager, Clinical Informatics
Aspen Advisors
In 2012, the Institute of Medicine (IOM) released the report “Health IT and Patient Safety: Building Safer Systems for Better Care” recognizing that information technology (IT) systems put in place to make healthcare more efficient and effective have introduced new patient safety concerns. 1 Data entry delays and transcription errors, for example, can create safety issues. Biomedical device integration (BMDI) – sending data directly from biomedical devices into the patient electronic medical record (EMR) – can reduce these patient safety concerns by removing delays in recording, errors in interpretation and transcription, and delayed or incorrect treatments.
Patient Safety Improvement Opportunities
It is not uncommon that on a general medical-surgical unit, a nursing care technician or assistant obtains the vital sign readings and writes them on paper only to later be transcribed into the record by the nurse. This can lead to delays in data flowing into the medical record and transcription errors – both of which are serious patient safety concerns.
A research study by Gearing (2006) found that there was a 14.9 percent error rate when manually transcribing vital signs into the EMR.2 In another study by Vawdry (2007), manually entered ventilator data resulted in a 3.9 percent error rate in ventilator protocol recommendations.3
With Clinical Decision Support (CDS) and Computer Physician Order Entry (CPOE), physicians rely on the availability of accurate, near real-time data to make clinical decisions on patient care. Errors and delays put patient care decisions at risk for providers and, more importantly, patients. Additionally, effective use of CPOE and documentation of vital signs in the EMR are components of Stage 1 Meaningful Use. Stage 3 Meaningful Use proposes to capitalize upon the interoperability of clinical documentation and medical devices, though this has yet to be well-defined.
Even with the clear patient safety benefits of implementing biomedical device integration, the process can challenge the best project team. Here are some lessons learned from successful biomedical device implementations.
10 Key Success Factors for BMDI Implementations:
1. Learn your device output and usage. Conduct a thorough workflow analysis, understanding where the device data is captured, recorded and the units of measure utilized. Standardization of these processes will be essential, and the process improvement steps to get there need to be planned for.
2. Set the medical device integration scope. Conduct a device assessment and define which devices will be in the first phase and which ones will be designated for later phases. All stakeholders need to understand and agree to the project scope. For example, devices that are not up to standards for exchanging data with an EMR will need to be budgeted for replacement within an adequate time frame for testing and validation before implementation.
3. BDMI rollout should match the overall documentation workflow. Approaching the rollout of BMDI should be sequenced to follow the flow of the documentation as well as the work stream. Activating a single device across the enterprise versus a group of devices in a care area on a single application platform may influence your approach to BMDI rollout.
4. Engage the BDMI team, and set service metrics. Assess the relationship with key players in BMDI - clinicians, IT and biomedical engineering – and develop a communication plan to engage them throughout the lifecycle of the initiative. All will play key roles in the entire process including development, rollout and post go-live support. Make sure to define solid service level agreements (SLAs) with both IT and biomedical engineering. As device data has typically been discrete and self-contained in the devices, IT and biomedical engineering have not crossed paths to the degree in which they must operate collaboratively with device integration. Define roles clearly for addressing and managing issues as they arise. Patient care is dependent on the real-time data, so there cannot be gaps in the service level to the clinicians.
5. BMDI and EMR go together. Plan concurrently with your EMR implementation. Often there are upgrades or service packs that must be scheduled and sequenced in order to address technical issues encountered in integrating devices with the EMR.
6. Clear technical specifications are a must. Complete an exhaustive technical assessment with your EMR vendor and keep them involved with the BMDI effort. Integrating the Healthcare Enterprise (http://www.ihe.net/ ) is making strong progress toward the standardization and interoperability of devices and device data, but this goal is still far from being met. Use of a middleware product that allows prior device mapping, HL7 interfacing and plug and play technology will greatly enhance your ability to be device vendor agnostic and afford your project greater flexibility and scalability.
7. Data storage, filtering and purging. Know how you will handle the plethora of data that comes from the device. Your EMR will receive an inordinate amount of data from device integration. For example, Al Campanella, CIO at Virtua, a four-hospital health system in New Jersey, estimates that one average surgical case will produce two gigabytes of data. You will need to plan for storage, selected filtering (such as taking the last reading every 30 seconds) and purging of this data, which is tied to the following lesson learned.
8. Medical records implications. Engage your health information management (HIM) and legal departments early in the process. Device data will have legal medical record implications from the amount and timing of data capture to the type of validation steps the clinicians must take to confirm accuracy. The record must reflect how the data was captured and assign ownership at a very granular level. Questions will need to be answered, such as who is validating at what point of care? Should clinicians be allowed to change device generated data? If so, how often? What populates the legal medical record versus the database?
9. Testing. Incorporate adequate testing time into your project plan. Every connection and every variable must be tested and validated. Involve your clinicians in the testing whenever possible and obtain sign-off from your key stakeholders. This will prevent any unnecessary discrepancies at the time of implementation.
10. Clinician training and adoption. In the end, clinicians need to trust the system and trust that the data flowing into their record is timely and accurate. With proper training and education, you’ll help clinicians feel the data is reliable, show them how they can address artifact or erroneous data and help them understand what controls are in place for managing the electronically transmitted data as their own.
When implemented appropriately and used correctly, IT systems have significant potential to make healthcare safer and more efficient. To avoid patient safety concerns that can be introduced by IT systems, employ BMDI to remove delays in recording, errors in interpretation and transcription and delayed or incorrect treatments.
A rigorous project plan that includes a thorough assessment, engagement of key stakeholders and proper planning will build the foundation for a successful BMDI initiative. Done right, BMDI will be widely welcomed and embraced by your clinicians as one EMR project that immediately reduces the workload on the clinician while producing a safer care environment.
References
1. IOM (Institute of Medicine). (2012). Health IT and Patient Safety: Building Safer Systems for Better Care. Washington, DC: The National Academies Press.
2. Gearing P, Olney CM, Davis K, Loranzo D, Smith LB, Friedman B. Enhancing patient safety through electronic medical record documentation of vital signs. Journal of Healthcare Information Management. 2006;20(4):40-45.
3. Vawdry, DK, Gardner, RM, Evans, RS, Orme, JF, Clemmer, TP, Greenway, L, Drews, FA. (2007). Assessing data quality in manual entry of ventilator settings. Journal of American Medical Informatics Association. Retrieved from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2244881/.
CLICK HERE to view the original article from Healthcare Executive Insight. Volume 3, Number 6 – June 18, 2012
