One of the realities of healthcare in the US is that diagnostic safety and quality receive relatively minimal funding despite the enormous burden of harm, which is linked to misdiagnosis-related medical errors.
Earlier this year, a new study in the BMJ Quality & Safety Journal reported that every single year, the misdiagnosis of dangerous diseases results in an estimated 795,000 Americans becoming permanently disabled or dying across care settings.
Yet, as David Newman-Toker, professor of neurology at Johns Hopkins Medicine and lead author on the study, points out, the US still spends more on smallpox research each year than it does on trying to tackle vital patient safety issues around diagnosis.
“We spend $39 million on a disease that was eradicated in 1980 and kills zero people in the world,” he says. “And we spend $20 to 30 million on research around diagnostic safety and quality. This is questionable in terms of our priority setting.”
Half of the harm reported in the study is caused by diagnostic errors relating to 15 specific diseases –
- Spinal abscess
- Aortic aneurysm and dissection
- Meningitis and encephalitis
- Endocarditis
- Arterial thromboembolism
- Lung cancer
- Venous thromboembolism
- Stroke
- Melanoma
- Sepsis
- Colorectal cancer
- Pneumonia
- Breast cancer
- Prostate cancer
- Myocardial infarction.
Newman-Toker points out that eliminating just half of these errors could prevent around 200,000 cases of harm. He highlights spinal abscess, where the misdiagnosis rate is a shocking 62%.
“If you look at the data on causes of diagnostic error, the biggest bucket is always some set of failures that happened at the bedside,” he says. “Whether that’s failures of eliciting a proper history or considering and weighting the differential diagnosis or ordering the right test, it’s all bedside failures. And for something like spinal abscess, there’s a double whammy because it’s relatively rare and we haven’t done anything to try to fix the problem.”
Preventing medical errors from misdiagnosis
So what can we do better? According to Newman-Toker, there are five critical actions that can be taken.
- Invest in better differential diagnostics: Just as the biomarker troponin helped diagnose heart attacks more accurately, new assays will undoubtedly emerge which make it easier for medical professionals to pin down the precise problem when a patient is experiencing vague symptoms that could relate to any number of diseases.
“I do believe that in some cases, there’s going to be some magical new test that solves the problem,” says Newman-Toker. “Because we don’t miss things when they’re obvious. We miss them when they’re subtle.”
- Bridge critical gaps in medical training: Safety experts are increasingly finding that a lot of diagnostic errors result from gaps in expertise. Newman-Toker points to the fact that it is possible to become a board-certified physician in the US without having completed any neurology during residency. “This is despite the fact that one in five complaints across care settings is neurological of some sort,” he says. “It clearly needs to change.”
- Improve the quality of data reporting: While AI has the potential to help clinicians with decision-making in the future, Newman-Toker points out that many of the tools currently being researched are being trained on electronic health records, which are often filled with incomplete or missing information. “To realize the promise of AI, it is going to take a concerted effort to construct the gold standard datasets required to train these algorithms to be capable of the kind of decision support that everybody’s hoping for,” he says.
- Incentivize correct diagnoses: Right now, unfortunately healthcare payment models actively incentivize diagnostic errors, something which urgently needs to change. “If you get the diagnosis wrong, and order all the wrong tests, you make a lot of money,” says Newman-Toker. “So there’s no financial or other incentive, besides just being good human beings, for people to actually do better.”
- Political or regulatory pressure: If we are to reduce the enormous burden of harm quantified in his study, Newman-Toker says that diagnostic excellence has to become part of the fabric of healthcare. For this to become systemic at a national level, it will undoubtedly require both investment and pressure from people in positions of power.
“Whether it’s political authorities, healthcare purchasers, regulatory bodies, or research funding agencies, they’re going to have to step up and prioritize the importance of this issue if we’re actually going to make a difference,” he says.
