Algorithm Accurately Identifies Pregnant Women with Inherited Bleeding Disorders

Ming Y Lim, MBBCh

Credit: University of Utah Health

Results of a recent study demonstrated the achievability of using an algorithm to accurately identify pregnant patients with specific types of inherited bleeding disorders within an electronic health record (EHR).¹ Findings were presented at the 2023 American Society of Hematology (ASH) Annual Meeting and Exposition, held in San Diego, California.

Lead investigator Ming Y Lim, MBBCh, associate professor in the Division of Hematology and Hematologic Malignancies at the University of Utah, and colleagues, noted previous research showed using diagnoses codes alone when searching EHRs resulted in a high number of false positive identifications.

“To improve data integrity, in-depth chart reviews are often required which typically can only be done from a local institutional level,” she wrote. “This limits the potential to use national research infrastructure such as the National Patient-Centered Clinical Research Network (PCORnet) which draws its data from millions of EHRs across healthcare institutions in the United States.”

Therefore, Lim and her team aimed to test the diagnostic accuracy of an algorithm designed to use common data definitions coupled with multiple data elements in EHR to identify affected patients. Data from pregnant women with inherited blood disorders who delivered at a hospital affiliated with the University of Utah from January 2016 through December 2020 were collected to test the accuracy of the algorithm. Inherited blood disorders included hemophilia and hemophilia carriers, von Willebrand disease, and rarer bleeding disorders. Data elements from the EHR were included if they aligned to the PCORnet Common Data Model (CDM), which aided in ensuring consistent data definitions and formats across multiple sites.

The first version of the algorithm incorporated: (1) ICD-9/10 codes for inherited bleeding disorder based on discharge diagnosis, (2) medications for the management of inherited bleeding disorders, and (3) Coagulation factor test and results. Inherited bleeding disorders were confirmed if patients fulfilled the criteria for either 1 and 2, or 1 and 3.

However, a revision to the initial criteria was introduced based on the results of retrospective queries, which were confirmed using a manual chart review and the local registry. The registry houses information on demographics and laboratory data of all patients with an inherited bleeding disorder diagnosis treated at a federally funded adult hemophilia treatment center.

The original algorithm (query 1.0) identified 301 pregnant women who fulfilled criterion 1 and had ≥1 live birth or fetal death at the institution during the study period. Within this cohort, 25 patients fulfilled criteria 2 and/or criteria 3. In the other 276 cases, diagnosis was verified using the registry or through a manual chart review.

Results showed certain ICD-diagnosis codes resulted in contamination and could not be used to diagnosis patients who were carriers of bleeding disorders or those with rarer bleeding disorders. Therefore, the affected codes were removed, and a fourth criterion was added to the revised version of the algorithm. This criterion stated for cases fulfilling criteria 1 but not 2 or 3, a diagnosis was confirmed if a patient had ≥2 identical ICD diagnosis code for an inherited bleeding disorder at ≥2 separate visit types.

In the revised algorithm (query 1.1), 35 pregnant women were identified with inherited bleeding disorders, of which 32 were confirmed as true positive. The 3 incorrectly diagnosed received further laboratory testing and were ruled out. Only 1 patient experienced a false negative using the revised version. The sensitivity of the revised algorithm was 97.0% and the positive predicted value (PPV) was 91.4%.

References

1. ^{Lim MY, Sivaloganathan V, and Simonsen SE. Developing an Algorithm to Better Identify Pregnant Women with Inherited Bleeding Disorders within Electronic Health Records. Presented at: ASH Annual Meeting and Exposition. San Diego, CA. December 9-12, 2023.}