BACKGROUND
Reporting of pelvic inflammatory disease (PID) from private providers could be incomplete because of time and staff constraints, lack of knowledge of reporting requirements and of case definitions. Reporting burden can be alleviated with the use of administrative data.
GOAL
The goal of this study was to determine the validity of clinical diagnostic codes assigned in electronic medical records (EMR) for identifying PID and their use in enhancing surveillance.
STUDY DESIGN
A random sample of 296 records with a PID International Classification of Diseases, 9th Revision (ICD-9), code (614.9) were reviewed to assess for the presence of the Centers for Disease Control and Prevention (CDC) criteria for the case definition of PID. We used the records meeting the CDC clinical case definition criteria as the reference standard to determine the sensitivity, specificity, and predictive values of various data elements.
RESULTS
Used alone, the positive predictive value (PPV) of ICD-9 code 614.9 for a CDC case definition of PID was 18.1%. The PPV increased to 100% and 56% when the ICD-9 code visit was associated with a positive test for Neisseria gonorrhoeae (GC) and Chlamydia trachomatis (CT), respectively.
CONCLUSION
In this multispecialty group practice, a positive test for GC and CT coupled with ICD-9 code 614.9 could be used to enhance reporting of cases of PID.