From the 2019 HVPAA National Conference
Dr. Navkiran Dhillon (Loyola University Medical Center), Dr. Nabeela Mohideen (Loyola University Medical Center), Dr. Michelle Lundholm (Loyola University Medical Center), Dr. Tony Kurian (Loyola University Medical Center), Dr. Joshua Newman (Loyola University Medical Center), Dr. Jonathan Sachs (Loyola University Medical Center)
Hospital discharge summaries allow physicians to communicate the inpatient care plan to other providers in an organized manner in order to effectively transition patient care. Furthermore, hospital billing and coding departments utilize information in order to accurately capture the severity of illness and risk of mortality. Inaccurate or insufficiently documented primary diagnoses in a discharge summary may adversely affect patient care, inaccurately characterize quality of care given, and incur additional costs to hospitals. Numerous studies have shown that these discharge summaries are often inaccurate. Some studies have looked at the accuracy of primary diagnoses while other projects have implemented drop-down menus to improve documentation.
Our study implemented a drop-down menu with the aim to improve accuracy of primary diagnosis documentation in discharge summaries by health care providers.
A drop-down menu was created for the primary diagnoses of pneumonia, CHF, COPD, respiratory failure, UTI, atrial fibrillation, and sepsis requiring providers to pick from a menu of more specific diagnoses, i.e. for pneumonia, physicians may select “community acquired pneumonia presumed bacterial with sepsis”, “aspiration pneumonia”, etc. Patients admitted with an admitting diagnosis of pneumonia or heart failure at our institution were included in the study. Fifty consecutive patients admitted with each diagnosis were reviewed prior to implementation and after implementation of the modification described above from September to December 2018. Patients found to have an alternative primary diagnosis after admission were excluded. Primary diagnoses documented in the discharge summary were reviewed, For pneumonia, documented diagnoses were considered “accurate” if documentation specified whether the pneumonia was community acquired, healthcare acquired, or aspiration. For heart failure, diagnoses were considered “accurate” if documentation specified whether it was acute on chronic or new onset as well as whether it was systolic, diastolic, or both. The data collected was analyzed using a Fisher Exact Test for a 2×2 contingency table using statistical software.
Providers did a suboptimal job of documenting the accurate diagnosis prior to the implementation of the template (21% for heart failure and 38% overall).There appears to be a statistically significant improvement in accuracy of diagnosis for heart failure (p value = 0.0003) and overall (p value = 0.0002). There was no significant difference in accurate pneumonia diagnosis (although the raw data did show an improvement from 59% to 76%).
Both resident and attending physicians displayed an improvement in accuracy of documentation after implementation of a drop-down menu created for the primary diagnosis in discharge summaries. However, the review is limited by the small sample selection.
It is promising that utilizing an electronic prompting system to complete discharge summaries was associated with an improvement in the quality and accuracy of discharge summaries, which may improve handover of patient care, reduce adverse patient outcomes, and increase hospital reimbursement.