Spatio-temporal patterns of tick-borne disease diagnoses in Indiana, USA (2009–2018)

Aims

Although tick-borne disease (TBD) incidence has increased in the United States (U.S.) in the past decade, new evidence suggests that notifiable diseases surveillance records may not accurately reflect the true magnitude of TBD diagnoses. Furthermore, while regional electronic health records (EHR) are readily accessible their potential use as a more stable and consistent source of TBD diagnoses data has remained largely unexplored.

Methods and Results

In this study, we used EHR from a database of more than 100 hospitals, healthcare networks, and insurance providers in Indiana, U.S., to better understand incidence, spatio-temporal and demographic distribution of TBD Diagnoses from 2009–2018. Our results revealed that in Indiana, from 2009 to 2018, there were 5173 unique TBD Diagnoses across three diagnoses categories: Lyme disease (72.5%, n = 3751), Rickettsioses (12.0%, n = 623) and Other TBD Diagnoses (15.4%, n = 799). Using EHR, the average yearly Lyme disease diagnoses was more than double the cases obtained using notifiable disease surveillance data for the same period. Patients with a TBD Diagnoses were generally older (ages 45–59) and less racially diverse (96.3% white). Rickettsiosis diagnoses were reported more among male patients (55.2%), while Lyme disease diagnoses were higher among female patients (57.1%). Temporal data illustrated higher frequencies of diagnoses from May to July. Hot spot analysis identified a Lyme disease hot spot in northwest Indiana, while hotspots of Rickettsiosis and Other TBD Diagnoses category were identified in southwest Indiana. Extrapolated to the Indiana population, chi-squared (χ²) tests of independence revealed that the observed distribution of TBD diagnoses in our data was significantly different from the expected distribution in the Indiana population-based race, gender and age groups.

Conclusions

Our study findings demonstrate that in Indiana, EHR provide a stable data source for elucidating TBD disease burden and for monitoring spatio-temporal trends in TBD diagnoses.