Readers Write: Trauma Thoughts

Trauma Thoughts
By Nicole Cook, APRN

Nicole Cook, APRN, MSN is a trauma clinical nurse specialist with WakeMed of Raleigh, NC and a clinical advisor for TraumaCare.AI.

The science and technology of healthcare is relentless in its pursuit of improvement and change. Techniques and practices that were once held as best practice have been improved upon or disproven, moved to the files of antiquity to be remembered fondly or with embarrassment as we realized that the sacred cow really wasn’t quite so golden. The evolution of care for traumatically injured patients is no exception.

Trauma nurses from 20 years ago would be flabbergasted to receive a trauma patient not on a long spine board. Mixing high-dose Solu-Medrol for an acutely injured spinal cord was cutting edge best practice. Peripheral IVs were flushed with heparin every eight hours, and we certainly preferred a good bolus of crystalloids before considering blood products. Yet now, every one of these interventions is no longer best practice. Our understanding of trauma pathology and care of trauma patients has evolved and will ceaselessly continue to do so.

Given the inevitability of change, trauma nurses must anticipate future shifts to our profession and the care and assessment of these critically injured patients. No new technology is poised to revolutionize healthcare more than the integration and incorporation of artificial intelligence and machine learning (AI/ML)-driven decision-making models into patient care and assessment. Never has technological and scientific change been so ripe with questions, apprehension, and seemingly limitless opportunity. 

We would be remiss to pretend that the last several years have not harshly affected healthcare and the field of nursing. Nurses have left the bedside, turnover in most healthcare organizations is at an all-time high, and uncertainty and change are our continual companions. As the future approaches, we look to lose more staff due to retirement as our patient population only continues to age and grow. Increasing the number of nursing educators and bolstering our nursing education pipeline cannot be ignored, but the impact of that would be felt years in the future. Staff need assistance now.

As healthcare moved into the digital age and EHRs became the norm, new challenges presented themselves. A deluge of digital data is filed into the record, often automatically. This is intended to improve decision-making, but the sheer volume often ends up overwhelming staff who may miss trends and changes among the pages and tabs of data. What was intended to make healthcare easier is resulting in information overload.

Imagine this not too unusual scenario. A busy emergency department, bursting at the seams. A nurse with a heavy assignment receives a multi-trauma patient after resuscitation who is now awaiting an ICU bed. Inpatient beds are at a premium, so the patient boards in the emergency department. This nurse has been well trained, but with a 1:5 nurse to patient ratio, it’s all they can do to keep up with basic tasks for their patients.

Minute changes in labs and assessment for the trauma patient indicate an impending worsening of clinical status, which is not readily apparent on the vital sign monitor. As the nurse treads water trying to keep up with their patient assignment, the very real risk exists that the nurse and care team will miss these subtle clues, leading to adverse patient outcomes.

Now consider one small change to this scenario. As the nurse logs into the EHR, they receive a notification of the impending clinical deterioration that was calculated by a decision-making model that is integrated into the record. The nurse notifies the admitting physician, who responds to the bedside. The clinical team reviews the notification and the pertinent data, adjusts the care plan accordingly, and the patient stabilizes. The nurse is also able to use this information to advocate that the patient be moved up in the queue to receive the next available ICU bed.

One of the concerns with any integration of technology into healthcare assessment is the potential for loss of clinical expertise and critical thinking that is secondary to overreliance. This concern is not unwarranted. The more our reliance on technology grows, the more it has the potential to pull staff from the direct bedside, facing away from the patient and toward a computer screen.

Nursing is an art and a science, twisted and entangled into one inseparable form. The anticipation of impending clinical deterioration is often described as a gut feeling, or simply “I am worried about this patient.” Artificial intelligence cannot replace nurse intuition and excellent assessment skills at the bedside. But this added layer of safety could be a constant background presence that is assessing and reassessing minute changes and alterations, as bedside nurses juggle ever-increasing documentation and regulatory requirements and a thousand little tasks that keep them from having the time to comb the data and see all the details.

Patient safety is created in layers. The Swiss Cheese Model is a well-known illustration of risk management and prevention. Layers upon layers of preventative measures and mitigation strategies result in improved safety. The integration of AI/ML can be seen as a robust addition to the Swiss Cheese Model due to its potential for adaption and development. Trauma nurses have a responsibility to investigate its potential in the evolution of our specialty to anticipate potential gaps in safety, equity, and education, and to take an active role in shaping this technology to assist us in caring for some of the most complex patients in all of healthcare.