What is new?
What this adds to what was known? International Classification of Disease Ninth Revision clinical modification (ICD-9-CM) and ICD-10 disease codes have good positive predictive value (PPV) for identifying upper gastrointestinal bleeding (UGIB) from electronic health-care records (EHRs). Use of less granular terminology [ie, International Classification of Primary Care (ICPC)] or free text for identifying UGIB results in a lower PPV. Use of more specific case definitions does not affect the magnitude of risk of UGIB associated with pre-specified drugs, but does reduce its precision. ICD-9-CM and ICD-10 disease codes can reliably identify UGIB in EHRs, whereas less granular terminology (eg, ICPC) may require additional strategies. Additional strategies need to be implemented to improve case identification when conducting pharmacoepidemiological drug safety studies in EHR databases that use disease-coding systems that have low granularity.
What is the implication?
What should change now?