Most people would agree that correctly diagnosing a patient’s medical problem is critically important. I’m a doctor – I think it’s important. Patients certainly think it’s important. And I’m fairly certain the community at large considers it pretty important as well.
But not one health-care organisation in the world is measuring the scale of diagnostic error in its patients.
Best estimates put the incidence of diagnostic error in the range of 10% to 15% but if we want to confirm exactly what’s happening in practice, we still have a very large amount of work to do. In particular, we need to figure out exactly what fraction of these errors result in harm.
Australia faces similar issues to other Western counties. Estimates put the number of diagnostic errors here at 140,000 cases a year, with 21,000 cases of serious harm and between 2000 and 4000 deaths.
These figures alone should be enough reason to take a more in-depth look at the problem. Some surveys have shown that diagnostic error is the greatest fear of patients entering the health-care system.
But if that’s not convincing enough, consider the impact this has on medical malpractice claims and, ultimately, doctors’ fees. In the United States, diagnosis-related claims account for the largest proportion of malpractice claims by far.
The situation may be more balanced in Australia but the fact remains that these errors cause appreciable suffering and harm everywhere.
In a recent address to the Hospital Alliance for Research Collaboration in Sydney, I pointed out that part of the problem is that there’s no consensus definition on what diagnostic error actually is.
There are two other major impediments. The tools we now use to monitor patient safety don’t detect diagnostic errors, and there’s a serious ownership issue in that no one seems willing to take responsibility for the problem. It seems that diagnostic errors fall into our collective blind spot.
Hospitals and health-care organisations think this is a doctor issue, doctors think it’s a problem for other doctors and educators don’t see it as their responsibility at all.
Why does diagnostic error happen?
It’s worth taking a moment to consider why these errors happen in the first place. First, we should acknowledge that diagnosis is difficult.
There are patient variables such as how a disease might manifest and there’s the complexity of the health system, with communication barriers and disjointed care.
And then there are doctor variables such as expertise, access to patient data, stress and distractions.
There are also more than 12,000 diseases according to the World Health Organization. Never heard of yellow nail syndrome or alien hand syndrome? Many doctors wouldn’t have either – they’re just two of the many new diseases that are constantly emerging.
We know that errors occur either as a result of problems with the system or from errors in cognitive reasoning. And we know that by far the main factor behind more than 80% of cognitive errors is faulty synthesis of information.
One example of faulty synthesis might be the hospital doctor who treats a patient referred from a private specialist. The specialist includes the doctor’s diagnosis in the referral notes.
Does the hospital doctor accept that diagnosis as the most likely possibility? Or does she assume nothing and look at the problem with a fresh pair of eyes?
Sadly, we hear of the former scenario too often. The recent US tragedy of 12-year-old Rory Staunton being sent home from hospital with a diagnosis of gastroenteritis only to die three days later of septic shock is a case in point. You would have read similar stories.
What we know and what we need to know
So we know that diagnosis is challenging, our health-care systems are complex and imperfect and, as human thinkers, we are fallible.
We need to know how to assess competency in clinical reasoning. And we need to understand how to measure the incidence of diagnostic error in specific settings, such as emergency departments. You can’t improve what you can’t measure.
There are possible ways of measuring diagnostic error that are worthy of being trialled.
“Trigger tools” that prompt further investigation are one such approach. An example of a trigger is an unplanned hospital admission within two weeks of a primary care visit, developed by Dr Hardeep Singh from the US Department of Veterans Affairs Medical Center.
We can also explore innovative ways to encourage doctors to report error. We will need to consider cultural factors here, but one study of facilitated physician reporting run by Dr Robert Trowbridge at Maine Medical Center established a successful desktop icon that helps doctors report diagnostic error.
We can also devote more energy to following up patients after health-care visits, and encouraging them to speak up and keep track of their health records.
What else can we do?
It is within policy makers’ grasp to solve the issue of ownership and they can make a real difference.
People who run health-care systems also have a role: to create the “space” to talk about error, address the common system flaws that contribute, provide decision support for doctors and help develop pathways for feedback.
Doctors can bring system errors to the attention of those in management and insist on feedback from both patients and peers. They can also be open to using decision support resources and seek help through collaborating with colleagues.
And patients? I repeat: Speak up! I know that can be difficult, particularly when there are perceived power imbalances in the doctor-patient relationship. This is where patient advocates may have a role.
But knowing what questions to ask and being a good historian is important. So is giving feedback about errors to your doctor, and knowing when and how to get back to him if your symptoms persist, change, or worsen. Your life may just depend on it.