Lower Diagnostic Error Rates Found Among Hospitalized Patients During Care Transitions

A lower prevalence of diagnostic errors was observed among hospitalized patients when examining errors during care transitions, suggesting that this approach may offer a more precise understanding of diagnostic challenges, according to a research letter published in Annals of Internal Medicine.¹

Although diagnostic errors are common, the optimal identification approach remains unknown. For example, previous studies used triggers, like death or intensive care unit transfer, to identify patients who may have experienced diagnostic errors.² However, using clinical triggers may introduce hindsight bias, potentially influencing the reported rates of harm associated with diagnostic errors.¹

Conversely, care transitions represent opportunities for error recognition since the oncoming clinician can review and reassess the previous clinician’s care.³ Consequently, the researchers aimed to develop a program to leverage end-of-week clinician care transitions to identify diagnostic errors and possible causes in hospitalized patients.¹

Examining care transitions in hospitalized patients revealed lower diagnostic error rates compared to traditional methods. | Image Credit: Pcess609 - stock.adobe.com

Therefore, the researchers conducted a single-center study at Northwestern University, an 897-bed teaching hospital, from April 2019 to August 2021. In this practice, clinicians work 7-day rotations, with handoffs to the oncoming physician conducted via telephone.

Those assuming care received an electronic questionnaire on day 2 of their service week, identifying 1 to 2 randomly selected patients who had been admitted 2 to 5 days before their transition. The questionnaire asked whether any diagnoses had changed, and subsequently about descriptions of the original and new diagnoses, respectively.

Cases with a diagnostic change were independently reviewed by 2 physicians using the Revised Safer Dx Instrument to determine if the changes may be attributed to a diagnostic error.⁴ The reviewers also used the National Coordinating Council for Medication Error Reporting and Prevention Index to classify potential harm.¹

Overall, the reviewers reported the number of diagnostic changes, errors, and the severity of harm. Each study team member independently reviewed the cases and provided open-ended descriptions of the causes of errors. The team members then met to discuss these assessments and generated 4 final labels for the causes of diagnostic errors: missed information from the electronic health record, missed examination finding, failure to order a test or consultation, and failure to consider a diagnosis.

Of 641 questionnaires, 621 (96.9%) were completed by 36 clinicians. Among 1011 patients, 184 (18.2%) had a diagnostic change and 59 (5.8%) were associated with a diagnostic error. The researchers found that the most common cause of error was missed information from the electronic health record (34 of 68; 50.0%).

They noted that few errors were associated with harm (10 of 59; 16.9%), and only 1 likely contributed to death. Conversely, most missed diagnoses involved infectious diseases (16 of 59; 27.1%), followed by cardiovascular (15 of 59; 25.4%) and neurologic (7 of 59; 11.8%) conditions.

The percentage of patients experiencing diagnostic errors was lower than previously reported. The researchers suggested that this discrepancy may stem from the triggers used in previous studies, which may have overestimated error rates. Consequently, they noted that using care transitions as a trigger provides a more accurate estimate of missed or delayed diagnoses.

However, the researchers acknowledged their study’s limitations, including its small single-site sample. They also noted that transition practices may vary across hospital medicine groups, limiting generalizability. Despite these limitations, Gopi J. Astik, lead investigator and hospitalist at Northwestern Medical Group, told The American Journal of Managed Care^® (AJMC^®) that their findings demonstrate the value of care transitions.

"Now that more of our care is shift- or time-based, I think there are care transitions that are just inherent to the practice of hospital medicine," she said. "This provides a good opportunity to prompt that review...rather than just carrying forward what was previously done. Care transitions have always been seen as negative and vulnerable, but I think they're actually a valuable asset to health care."

References

1. Astik GJ, Olson APJ, Steker D, et al. Utilizing care transitions for diagnostic error detection in hospital medicine patients. Ann Intern Med. doi:10.7326/ANNALS-24-00563 
2. Auerbach AD, Lee TM, Hubbard CC, et al. Diagnostic errors in hospitalized adults who died or were transferred to intensive care. JAMA Intern Med. 2024;184(2):164–173. doi:10.1001/jamainternmed.2023.7347
3. Lane KP, Chia C, Lessing JN, et al. Improving resident feedback on diagnostic reasoning after handovers: the LOOP Project. J Hosp Med. 2019;14(10):622-625. doi:10.12788/jhm.3262
4. Singh H, Khanna A, Spitzmueller C, Meyer AND. Recommendations for using the Revised Safer Dx Instrument to help measure and improve diagnostic safety. Diagnosis (Berl). 2019;6(4):315-323. doi:10.1515/dx-2019-0012