Short-term exposure to ozone in relation to mortality and out-of-hospital cardiac arrest : exploring sensitive subgroups by previous hospitalizations

Abstract

Background: Several epidemiological studies have associated short-term ozone (O3) exposure with mortality, as well as with out-of-hospital cardiac arrest (OHCA) and other morbidity. Knowledge on which parts of the general population that are susceptible to adverse O3-related health effects is scarce.

Aim: The aim of this thesis is to explore the role of previous disease in conferring susceptibility to O3 exposure in relation to total, cardiovascular and respiratory mortality, and OHCA.

Materials and Methods: We obtained information on age, sex, date and cause of death [classified using the International Classification of Diseases, Ninth (ICD-9) and Tenth Revision (ICD-10)] on all non-traumatic deaths that occurred in Stockholm County from 1990 to 2010. We considered all non-traumatic deaths, as well as cardiovascular and respiratory deaths separately. Data on all Emergency Medical Service-assessed OHCA were obtained from the Swedish Register for Cardiopulmonary Resuscitation. We included all OHCAs that occurred in Stockholm County from 2000 to 2014, and all OHCA occurring in 2006-2014 in Gothenburg and Malmö. All deaths and OHCA were linked to their previous hospitalizations as recorded in the National Patient Register, using personal identification numbers. Hourly values of O3 were obtained from single urban background monitoring stations in each city. We generated daily 8-h maximum levels as well as 2-h, 24-h, 2d, and 7d means. The associations between ambient O3 levels and health outcomes were estimated with time-series analyses using generalized additive and linear models, and with time-stratified case-crossover analysis.

Results: Short-term increases in O3 levels were associated with increased risks of total, cardiovascular and respiratory mortality, irrespective of previous hospitalizations. Individuals previously hospitalized for myocardial infarction demonstrated a higher O3-related risk of total and cardiovascular mortality in comparison with the general population (1.7 % vs 0.5 %; 2.1 % vs 0.8 %, per 10 μg/m3 increase in O3 during a 2d period). Individuals with previous hospitalization for chronic obstructive pulmonary disease exhibited higher risk of O3-related respiratory mortality compared to the general population (5.5 % vs 2.7 %). Furthermore, O3 exposure was associated with OHCA. A 10 μg/m3 increase of 2-h and 24-h averaged O3 was associated with an odds-ratio of 1.02 (95% CI: 1.01, 1.05) and 1.04 (95% CI: 1.01 – 1.07), respectively. We did not however observe a difference in O3-related risk of OHCA in individuals hospitalized for any of the pre-specified diagnoses of acute myocardial infarction, heart failure, diabetes, hypertension, or stroke, in comparison with the general population.

Conclusions: Our results suggest that previous hospitalizations for myocardial infarction or chronic obstructive pulmonary disease increases the susceptibility for mortality following short-term exposure to O3. In contrast, previous hospitalizations for cardiovascular diseases did not seem to modify the associations between short-term increases in O3 levels and the risk out-of-hospital cardiac arrest.