"Primum non nocere - first, do no harm"

We hear with distressing frequency the horror stories of patients who have been injured or killed as a result of medical mistakes. We think, "How could that possibly have happened?" and then realize that some of those stories even come from our own ICUs. We think if we just try harder, we will do a better job. The reality is that we all try very hard every day. We can't afford not to - the stakes (for our patients) are just too high. But we are human.

So what can we do to protect our critically ill patients, the most vulnerable ones in the hospital, and the whole healthcare system, from "us"? We hear constant talk about "improving communication," "streamlining the process," or "fixing the system," as has been done in the airline industry. In fact, much in the system can be improved to protect both our patients and our multidisciplinary team. We can standardize a good deal of our care. Some fairly simple measures, such as using a daily checklist, can help us ensure that we don't forget the little things. It is the details that count in critical care.

Of course, we cannot standardize everything. Patients are human, too, and their responses to our treatments are not always consistent. This difference is where improving healthcare diverges from the airline industry's model of improvement. Our processes and procedures are often adjusted according to each individual patient's response to treatment. Standardizing our processes so that we approach the same issues the same way every time decreases medical errors.

By examining our own procedures and routines, and making adjustments to our processes, we can improve patient outcomes. Without such examination, it is difficult, if not impossible, to improve the care we provide to our patients. Whether you use an established database, or have your own institutional system for measuring outcomes, measurement tools that provide reliable data are essential to creating a safe and effective system.

Appropriate use of technology must also enter into the improvement equation, which can help advance our system and thus provide safe, effective care to our patients. Rapid advances in noninvasive monitoring techniques have enabled us to gain better information about the physiologic status of our patients, without the risks of invasive monitoring. We must continue to expand our understanding of the roles for different types of monitoring, as well as to refine the therapeutic decisions based upon that monitoring, to apply both invasive and noninvasive monitoring techniques optimally.

The use of computerized physician order entry (CPOE) can markedly decrease errors in ordering medications for our patients, most of whom receive multiple medications. I, for one, cannot hope to remember all of the possible drug interactions that may occur in my most critically ill patients, and would welcome a computer system to help prevent adverse events.

Another area of our healthcare system that we must examine is training the next generation of healthcare professionals. I grew up in the age of "see one, do one, teach one." Clearly this "system" provided me early opportunities to learn procedures, and I quickly became quite skilled. But was this system the safest one for my patients? I sometimes shudder to think of what I thought I knew then, what I actually knew then, and what I know now! How then, are we to balance the issues of patient safety against the need for first-hand training experience?

Obviously close supervision is essential, and supervision of trainees has clearly improved since my training days, with a concomitant improvement in patient safety. But no matter how closely I supervise a trainee, it is still not my hands on the needle, tube, etc. There is a limit to how safe I can make the patient merely by my presence.

We can take another lesson from the airline industry, as well as from our anesthesia colleagues, by developing additional types of simulators and incorporating them into our training processes. Even though a simulator is not "real," and the practice situation will never carry the same emotional overlay as occurs when caring for a patient, much can be learned from practicing in situations that are calm, free from distraction, and do not result in harm to a real person.

The organization of the ICU team is another important factor we should consider in our efforts to improve patient safety; in fact it may be THE most important factor. My model ICU consists of integrated care by dedicated experts, including physicians, nurses, respiratory care practitioners, pharmacists, nutritionists, social workers, chaplains, and potentially a number of others working together toward common, articulated goals.

Every multidisciplinary team member should know what these goals are and remember that the patient and family are also part of the team. Management guidelines, ongoing education for all groups, and especially close communication between all members of the healthcare team all contribute to better coordination of care. Neither we, nor our patients, can afford "turf battles." We can and must all work together to implement current methods and develop new ones to improve patient safety. We must continue to be advocates for our patients.