Consider the following scenario:

A patient enters the hospital and is treated in the ER by Dr. A who diagnoses the patient with problem X and the patient is then released. The patient returns with worsening symptoms and is seen by Dr. B who again diagnoses the patient with problem X. The patient returns a third time and requires emergency surgery with a diagnosis of problem Y.

In this scenario, whether Doctors A and B ever find out about their misdiagnosis through a “feedback loop” or whether there are any efforts made to investigate the cause is anyone’s guess and sometimes only becomes an issue when the patient, frustrated over the lack of acceptance of responsibility, seeks recourse. 

Errors made in a medical diagnosis can be catastrophic, and in some cases, results in a patient’s death. The human impact of misdiagnosis cannot be overemphasized. Unfortunately for patients, medical errors are not always captured in data gathering efforts. In some cases, an organization’s culture does not lend itself to admitting them, investigating them or even taking corrective action. 

The Pennsylvania Patient Safety Authority in their September 2010 Advisory published an article “Diagnostic Errors in Acute Care” examining these errors in acute care settings. 

One of the problems the Authority found is that data gathering efforts do not always capture events caused by diagnostic errors. Such deficiencies notwithstanding, the Authority embarked on an effort to look at those events that would fall under the rubric of “diagnostic errors” between 2004 and 2009. The Authority, in its own study, found five categories which accounted for the most common cases of misdiagnoses: cancer, fractures, pulmonary embolism, acute coronary syndrome and appendicitis. The objective was to understand the processes that led to the errors.

Rates of such errors have been found to vary widely by specialty. Emergency department rates, for example, have been found in the range of 0.6% and 12% compared to other fields, such as pathology, where the rates are below 5%.

Autopsy studies have shown that diagnostic error rates can be significant. In one study, a median error rate of 23.5 % was found while in another 10-12 % and in a third 14%.

And the cost is similarly staggering. For example, one insurance company study cited in the article stated that the cost of such errors during the period of January 2005 through July 2007 amounted to $127 million in contrast to a price tag of $125 million for all other categories of patient injuries considered “high severity”.

The Authority’s September article cites a number of reasons why this area of patient safety has not been given proper attention. A patient may simply be unaware that he has been misdiagnosed if there are no consequences to his health. Further, a patient may simply walk away from a practice and seek treatment elsewhere, so the provider and the health system, in general, is never informed of the error. A third reason rests with the deficiency of what is termed a “feedback loop”. Because there is limited data readily available concerning diagnostic errors, providers may not be made aware of their errors.
   
According to the Patient Safety Advisory’s study, a lot depends on the organization’s culture and whether such feedback serves to result in change, either by the individual (such as thinking beyond the most obvious diagnosis) or organizationally (such as screening patients who are returning to the ER with recurring complaint within a 48 hour period). The absence of such feedback leads to overconfidence in diagnostic conclusions that go unchecked. 

Although they found many factors at work which contribute to diagnostic errors, the Authority concluded that healthcare facilities themselves can go a long way in minimizing them, ensuring that the following are in place:

  • Implementing interventions that establish strong and effective feedback loops between and among physicians concerning diagnostic accuracy
  • Ensuring that all steps in the diagnostic testing phase occur correctly and that all results are communicated back to ordering physicians and patients
  • Acknowledging the lack of feedback mechanisms in healthcare facilities and seeking ways to give and receive collegial diagnostic feedback
  • Accepting the possibility of diagnostic error …lead[ing] to greater metacognition and recognition of diagnostic error when it does occur
  • Involving patients in the diagnostic process

Such processes are critical to improvements in patient safety, which are necessary in order to prevent future incidents of pain, suffering, injuries and deaths.