3.2 Specific chronic diseases

When all EHE deaths were compared with all typical weather deaths, the OR [95% confidence interval] for those with schizophrenia was 3.07 [2.39, 3.94]. EHE deaths were also significantly increased among those with chronic kidney disease and ischemic heart disease with ORs of 1.36 [1.18, 1.56] and 1.18 [1.00, 1.38], respectively. The effect estimates for most chronic diseases were null (Figure 4; Table S2). However, there were several conditions for which the odds of mortality were significantly lower among EHE deaths, including angina, hospitalized transient ischemic attack, dementia, and osteoporosis. For example, the OR was 0.74 [0.63, 0.87] for dementia and 0.58 [0.47, 0.71] for angina (Figure 4; Table S2).

3.3 Burden of chronic disease

The odds of EHE mortality were higher among those with more chronic diseases (Figure 5; Table S3). The ORs were higher than 1.0 for those with 3 or more chronic diseases, and most were statistically significant. However, there was no clear trend of increasing ORs with increasing burden of chronic disease; most estimates for 3 or more chronic diseases were similar. The ORs were null for those with 1 and 2 chronic diseases.

3.4 Subgroup analysis

3.4.1 Heat-related deaths

When the 280 deaths already classified as heat-related (X30) were compared with all typical weather deaths, odds of EHE mortality were significantly increased among those with schizophrenia, chronic kidney disease, depression, and diabetes (Figure 5A; Table S2). The ORs were 3.99 [2.62, 6.08] for schizophrenia, 1.45 [1.07, 1.96] for chronic kidney disease, 1.86 [1.42, 2.44] for depression, and 1.42 [1.08, 1.86] for diabetes. The ORs were significantly less than 1.0 for those with angina and dementia. The ORs were increased for those with 4 or more chronic diseases (Figure 5B; Table S3).

3.4.2 Deaths with pending cause (R99)

When the 382 deaths with pending cause (R99) were compared with all typical weather deaths, the odds of death during the EHE were significantly increased for those with schizophrenia, substance use disorder, and chronic obstructive pulmonary disease (Figure 5A; Table S2). The ORs were 4.95 [3.46, 7.09] for schizophrenia, 1.51 [1.14, 2.01] for substance use disorder, and 1.33 [1.01, 1.75] for chronic obstructive pulmonary disease. The ORs were significantly less than 1.0 for those with dementia and angina. There was no clear effect of higher chronic disease burden in this group (Figure 5B; Table S3).

3.4.3 Non-heat-related deaths (not X30 or R99)

When the 952 non-heat-related (not X30 or R99) deaths were compared with all typical weather deaths, the odds of EHE mortality were significantly increased among those with schizophrenia, ischemic stroke, chronic kidney disease, and ischemic heart disease (Figure 5A; Table S2). The ORs were 1.66 [1.14, 2.43] for schizophrenia, 1.39 [1.07, 1.80] for ischemic stroke, 1.39 [1.17, 1.64] for chronic kidney disease, 1.25 [1.03, 1.52], and for ischemic heart disease. The ORs were significantly decreased among those with angina and osteoporosis, and generally elevated for those with 6 or more chronic diseases (Figure 5B; Table S3).