Posted by Sten Westgard, MS
Diagnostic errors are one of the "new" hot topics in the healthcare field. A new study from British Medical Journal of Quality and Safety has a chilling estimate of just how common diagnostic errors are occurring in outpatient settings.
So what's your guess? How often in the US are diagnostic errors being made in outpatient settings?
- about 1 in 5 visits
- about 1 in 10 visits
- about 1 in 20 visits
- about 1 in 50 visits
- about 1 in 100 visits
The study's conclusion, after the jump...
The frequency of diagnostic errors in outpatient care: estimations from three large observational studies involving US adult populations, Hardeep Singh, Ashley N D Meyer, Eric J Thomas, BMJ Qual Saf 2014;23:727-731.
The study concludes that the rate of outpatient diagnostic errors is 5.08%, about 1 in 20 visits, which, extended to the US patient population, means that approximately 12 million US adults are impacted by diagnostic errors every year.
For those of you keeping Sigma scores at home, that's a 3.2 Sigma process. Airlines handle your luggage better than your doctor makes your outpatient diagnosis.
Drilling down, the estimate is derived from three different studies, first and second by looking at colorectal and lung cancers, including follow-up (or lack thereof) of abnormal clinical findings, and finally by looking at unplanned hospital visits or unexpected return visits after an initial primary care visit. Within each segment of the study, the diagnosis error rates are pretty divergent:
- there were 26 errors out of 291,773 patients seen for colorectal cancer, or 5.3 Sigma
- there were 127 errors out of 587 lung cancer cases, or 2.3 Sigma
- in the patient record study, 5,126 errors were found in 81,483 patients, or 5.06%, or 3.2 Sigma
I suppose this puts the overall estimate (again that's 5.08%) into context. The usual patient visit should experience about 3.2 Sigma, but if they have some more serious condition, the diagnosis error rate will change. For those undergoing colorectal screening, the news is better. For those with lung cancer, this is bad news piling on top of bad news. Lung cancer is a deadly disease - and it appears clinicians are pretty bad at making the correct diagnosis.
One last question that's relevant for the laboratory: how many of these diagnosis errors might have been caused by test results? (the study doesn't delve into that level of detail)
We often extoll the role of the laboratory in making clinical decisions. Would we be so proud if we had to accept responsibility for about 75% of those diagnosis errors because of inaccurate test results? Or is this when we shrink back and let the clinician take all the blame?