Using logic regression to characterize extreme heat exposures and their health associations: a time-series study of emergency department visits in Atlanta

Abstract

Background Short-term associations between extreme heat events and adverse health outcomes are well-established in epidemiologic studies. However, the use of different exposure definitions across studies has limited our understanding of extreme heat characteristics that are most important for specific health outcomes or subpopulations. Methods Logic regression is a statistical learning method for constructing decision trees based on Boolean combinations of binary predictors. We describe how logic regression can be utilized as a data-driven approach to identify extreme heat exposure definitions using health outcome data. We evaluated the performance of the proposed algorithm in a simulation study, as well as in a 20-year time-series analysis of extreme heat and emergency department visits for 12 outcomes in the Atlanta metropolitan area. Results For the Atlanta case study, our novel application of logic regression identified extreme heat exposure definitions that were associated with several heat-sensitive disease outcomes (e.g., fluid and electrolyte imbalance, renal diseases, ischemic stroke, and hypertension). Exposures were often characterized by extreme apparent minimum temperature or maximum temperature over multiple days. The simulation study also demonstrated that logic regression can successfully identify exposures of different lags and duration structures when statistical power is sufficient. Conclusion Logic regression is a useful tool for identifying important characteristics of extreme heat exposures for adverse health outcomes, which may help improve future heat warning systems and response plans.

Association between ambient temperature and risk of stroke morbidity and mortality: A systematic review and meta‐analysis

“…Then, a total of 62 articles were excluded after a full‐text screening because exposure was not heat/cold effect, the outcomes were not stroke, or information were not useful. Finally, 20 articles were leaved on meta‐analysis (Bai et al., 2018 ; Ban et al., 2017 ; Breitner et al., 2014 ; Chen et al., 2013, 2017 ; Fu et al., 2018 ; Goggins et al., 2012 ; Guo et al., 2017 ; Hong et al., 2003 ; Ikefuti et al., 2018 ; Jiang et al., 2021 ; Luo et al., 2018 ; Polcaro‐Pichet et al., 2019 ; Roye et al., 2019 ; Schulte et al., 2021 ; Wang & Lin, 2014 ; Wang et al., 2015 ; Xu et al., 2021 ; Yao et al., 2020 ; Zhan et al., 2022 ). The PRISMA flowchart is shown in Figure 1 .…”

Heatwave Definition and Impact on Cardiovascular Health: A Systematic Review

“…The most frequently used meteorological measure as heat indicator was air TEMP, applied in 56 of 57 cases. In nine studies, air TEMP was included in the computation of more complex indices, such as “Apparent Temperature” [ 42 , 53 , 60 , 71 , 76 , 83 ], “Perceived Temperature” [ 39 ], “Pseudo-equivalent daily temperature” [ 88 ] or “Discomfort Score” [ 56 ], by combining it with relative humidity, water vapor pressure, dew point TEMP, wind speed and/or heat radiation. The relevant calculation formulas are summarized in Table 2 , except for “Perceived Temperature” [ 39 ] as it was not provided in the relevant study.…”

“…However, in 11 studies (20.4%) no statistically significant results were found, while in 6 (11.1%) a statistically significant decrease in the risk of developing a CV acute event during heat was surprisingly reported. Of these six, one study analyzed all CV diseases together [ 83 ], while the others referred to specific diseases, i.e., myocardial infarction [ 79 ], hypertensive diseases [ 76 , 79 ], heart failure [ 45 , 63 , 79 ], and acute aortic dissection [ 75 ].…”

Use of machine learning tools to predict health risks from climate-sensitive extreme weather events: A scoping review

“…Geographies covered and study settings. The seven studies analyzed in here report data from the United States of America [16], Europe [17][18][19], China [20,21] and South Korea [22]. Data was collected at the country [18,19], county [16] or division [20,22], and city level PLOS CLIMATE PLOS CLIMATE [17,21].…”

“…The seven studies analyzed in here report data from the United States of America [16], Europe [17][18][19], China [20,21] and South Korea [22]. Data was collected at the country [18,19], county [16] or division [20,22], and city level PLOS CLIMATE PLOS CLIMATE [17,21]. There were 16 European countries included, namely Austria, Belgium, Croatia, the Czech Republic, Denmark, France, Germany, Italy, Luxembourg, Netherlands, Poland, Portugal, Slovenia, Spain, Switzerland and the United Kingdom (England and Wales only) [18,19].…”