OBJECTIVES: Identifying rare outcomes in administrative claims data is limited by the existence of specific International Classification of Diseases (ICD) codes and overall sample size of the database. We aimed to assess the ability to identify rare animal bite events via a large, claims database.

METHODS: The Merative MarketScan Commercial and Medicare Databases were used for this analysis. Three cohorts of patients were identified via animal bite ICD-10 diagnosis codes on claims during the study period of January 1, 2016 - January 31, 2022 (Shark: W56.41XA; Pig: W55.41XA; Raccoon: W55.51XA). Incidence proportions were calculated among the cohort of patients with ≥1 day of continuous enrollment in the database during the study period. For the descriptive analyses, patients were required to have ≥12 months of continuous enrollment prior to the bite event to assess baseline clinical characteristics, including the Deyo-Charlson Comorbidity index (DCI).

RESULTS: Approximately 60 million unique patients were identified as having enrollment in the Merative MarketScan Database during the study period. A total of 67 patients had a claim with an ICD-10 diagnosis code for shark bite, representing an incidence of approximately 1 in a million during the study period. Small sample sizes were also observed for pig bites (N=256); however, raccoon bites were more common (N=615). Shark bite victims were more likely to be male and average age was 27.5 years. The baseline period DCI scores were low for all animal bite cohorts (mean DCI<0.5), indicating low overall comorbidity burden.

CONCLUSIONS: Whether your research is about a shark, pig, raccoon bites or about a rare disease or treatment, these findings demonstrate the utility of large databases to answer research questions about rare events. The value of strong data with high patient counts coupled with highly specific ICD-10 codes can fuel research, policy and treatment insights.