Do nurse advice calls follow ozone pollution? Using syndromic surveillance to evaluate short-term respiratory effects of ambient ozone concentrations in the Washington, DC metropolitan area

Abstract

TPS 683: Short-term health effects of air pollutants 2, Exhibition Hall, Ground floor, August 28, 2019, 3:00 PM - 4:30 PM Background: Syndromic surveillance, a tool to evaluate the presence and scale of disease outbreaks, has had only limited utilization in environmental health research. We adopted this framework to examine the relationship between ambient ozone concentrations and respiratory symptoms in a large HMO-based healthcare system in the Washington, DC area. We investigated the associations between changes in ozone concentrations and utilization events with particular attention to nurse calls/emails as a leading indicator. Methods: We assembled a database of data from Kaiser Permanente patients in the region for 2013-2014. Daily utilization events for respiratory-related diagnoses were aggregated by census block-group: nurse calls/emails, provider visits, emergency department/urgent care (ED/UC) visits, and hospital admissions. The database was linked to 8-hr average daily measures of ozone concentration for each block group and we modelled utilization event rates per 100,000 patients as a function of ambient ozone concentration. Results: We found that changes in 1-day lag and 3-day moving average (MA) ozone levels were associated with increases in nurse calls/emails, provider visits and ED/UC visits. The highest effects were found for nurse calls/emails, specifically, a 10 ppb change in 3-day MA ambient ozone was significantly associated with a 3.50% (95%CI: 2.08%, 4.91%) change in nurse calls/emails. The same change in ozone was significantly associated with a 1.53% (95%CI: 0.67%, 2.39%) change in provider visits and a 1.60% (95%CI: -0.09%, 3.30%) increase in ED/UC visits. We did not identify associations with hospitalizations. Conclusions: We utilized the framework of syndromic surveillance to evaluate associations between ozone and respiratory-related healthcare interactions. We found the highest effects between changes in ozone concentrations and subsequent nurse calls/emails. To our knowledge this is the first analysis to note such effects. Our findings indicate the usefulness of utilizing previously little-studied healthcare interactions data to understand respiratory effects of ozone exposure.