Read more

October 14, 2024
2 min read
Save

Diagnostic errors occur in over 7% of hospitalized patients receiving general care

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Key takeaways:

  • Among the 160 cases of diagnostic errors, most may have been preventable.
  • Process failures, like failures in diagnostic testing and patient experience, were greatly tied to diagnostic errors.

Harmful diagnostic errors may occur in one in 14 hospital patients, the findings of a retrospective cohort study published in BMJ Quality & Safety suggested.

Most of the errors were preventable, while the results “offer direction for improving surveillance approaches and developing preventative interventions,” according to the researchers.

PC1024Dalal_Graphic_01_WEB
Data derived from: Dalal A, et al. BMJ Qual Saf. 2024;doi:10.1136/bmjqs-2024-017183.

Healio previously reported that diagnostic errors contribute to around 800,000 serious harmful injuries and deaths annually.

In the current analysis, Anuj K. Dalal, MD, an associate professor of medicine at Harvard University, and colleagues developed and validated a structured case review process to allow clinicians to use the electronic health record to evaluate the diagnostic process during a hospital visit, assess the likelihood of a diagnostic error and determine the impact and severity of the error.

This process was used to determine the prevalence of diagnostic errors in a cohort of 675 hospital patients randomly selected out of 9,147 who received care between July 2019 and September 2021. The analysis excluded cases that occurred from April 2020 to December 2020.

Cases characterized as being high-risk for diagnostic error included those with a transfer to the ICU 24 hours or more after admission (n = 130), death within 90 days of admission, either in the hospital or after discharge (n = 141), and those with complex clinical issues but no transfer to the ICU or death within 90 days of admission (n = 298).

Cases defined as low risk for diagnostic error met none of the criteria (n = 106).

Researchers defined diagnostic harm as minor, moderate, major and fatal, with errors that included either of the latter two defined as severe.

Diagnostic errors occurred in 160 cases, whereas harmful diagnostic errors occurred in 82 cases. The severity of harm experienced by patients was most often moderate or major.

The researchers found that, based on the weighted sample, the proportion of harmful, preventable and severely harmful diagnostic errors in patients was 7.2% (95% CI, 4.66-9.8), 6.1% (95% CI, 3.79-8.5) and 1.1% (95% CI, 0.55-1.68), respectively.

Overall, 84.7% of the diagnostic errors were preventable based on the findings.

Dalal and colleagues noted that 48% of diagnostic errors as being related to the primary diagnosis at admission or discharge, whereas 52.5% were related to a secondary diagnosis. Additionally, they characterized 62% of errors as delays.

Severely harmful diagnostic errors occurred in 55.1% of high-risk cases. Harmful diagnostic error rates appeared higher for older, White, non-Hispanic and high-risk patients.

Researchers reported highly significant associations between process failures were greatly associated with diagnostic errors. These included failures in:

  • diagnostic testing (OR = 4.24);
  • subspecialty consultation (OR = 3.11);
  • patient experience (OR = 2.93); and
  • history taking (OR = 2.5).

The most frequent diagnoses tied to diagnostic errors included heart failure, acute kidney failure, pneumonia, respiratory failure, altered mental status, abdominal pain and hypoxemia.

The researchers acknowledged several study limitations. For example, limiting the period of harm after admission to 90 days may have prevented the detection of harmful diagnostic errors with delayed impacts, they wrote.

The data drawn also came from a single center, whereas the cohort was restricted to patients with a hospital stay of less than 21 days.

Dalal and colleagues explained that certain approaches, like those including the use of AI and machine learning, could help lead to preventive interventions “to promote a culture of diagnostic safety.”

References: