Good data requires a systems approach – scenario
The respiratory therapy service of a large academic medical center received its patient care scorecard, and the hospital’s quality assessment team was concerned by the findings. The scorecard provided performance benchmarks and improvement targets for the service. Patient satisfaction, process of care measures, and outcomes were among the performance benchmarks assessed. What troubled the quality assessment team were the poor marks for patient outcomes, specifically the patient mortality rate.
The respiratory service physicians questioned whether or not the data were correct when they reviewed the scorecard results. The results were summarized from the hospital’s R-ADT system. To better assess the data, the data steward for the service requested a listing of all respiratory patients who had been admitted and discharged from the service for the period covered by the scorecard report. The listing was provided digitally and the data steward first reviewed the dates of admission and discharge for each patient, then the admission and discharge service for each patient, and compared the admission and discharge diagnosis for each patient.
On review of the data, the data steward found a significant number of discharge diagnoses were related to a cardiac rather than a respiratory diagnosis. Due to this finding, the data steward then studied a list of all the attending physicians for each patient. This review revealed that all of the patients with a cardiac discharge diagnosis had initially presented to the emergency department with respiratory symptoms and had been assigned to the respiratory service. However, during hospitalization, cardiac problems arose and each of these patients had been transferred to the cardiac service. During the transfer process, a change in service had not been noted in the service change menu in the patient’s EHR. Consequently, for these patients, the respiratory service was being evaluated with the wrong data. The data steward also reviewed who the responsible practitioner was for the transfer process. This led to determining that two individuals were responsible for 90 percent of the documentation errors.
This case demonstrates how one documentation error can have significant consequences. In this EHR, documentation of a service transfer was not automatically invoked when there was a change in attending physician. Instead, noting a patient transfer to a different medical service required the responsible practitioner to change the service from a drop-down menu. On further evaluation, it was discovered that a lack of training in the EHR system for two practitioners was the underlying cause of failure to document correctly.
This case also demonstrates the value of a data steward who has responsibility for investigating data issues and has knowledge of the assigned domain or business unit. It shows the importance of communication and responsibility channels and reporting structure. Once the data steward identified the documentation problem, the findings were communicated through formal channels to the data steward council. The data steward council initiated problem resolution and the two responsible practitioners were provided additional training in the EHR system.
Post your responses to these two questions.
- What policy or procedure should be in place to make sure that this problem does not occur again?
- What other actions should be taken to make sure that other data have not been skewed because of this type of error?