A National Study on the Impact of Wildfire Smoke on Cause-Specific Hospitalizations Among Medicare Enrollees with Alzheimer’s Disease and Related Dementias from 2006 to 2016

Abstract

Older adults may experience worse wildfire fine particulate matter (PM_2.5) smoke-related health effects due to conditions such as Alzheimer’s disease and related dementias (ADRDs). We evaluated whether wildfire PM_2.5 was associated with acute hospitalizations among older adults with ADRD, linking modeled daily wildfire PM_2.5 concentrations and circulatory, respiratory, anxiety, and depression hospitalizations from 2006 to 2016. We employed a case-crossover design and conditional logistic regression to estimate associations between lagged daily wildfire PM_2.5 and hospitalizations. Also, we stratified cause-specific models by age, sex, emergency hospitalization status, and zip code-level urbanicity and poverty. The 1,546,753 hospitalizations among Medicare enrollees with ADRD were most coded for circulatory (71.7%), followed by respiratory (43.6%), depression (2.9%), and anxiety (0.7%) endpoints. We observed null associations between wildfire PM_2.5 and circulatory, respiratory, and anxiety hospitalizations over the six days following exposure. Same-day wildfire PM_2.5 was associated with decreased depression hospitalizations (rate ratio = 0.94, 95% CI: 0.90, 0.99). We saw some effect measure modifications by emergency hospitalization status and urbanicity. There were some stratum-specific effects for age, but the results remained mostly null. Future studies should use improved methods to identify ADRD and examine recent years with higher wildfire concentrations.

1. Introduction

Climate change increases the average temperatures and dries vegetation [1,2,3], heightening wildfire risks. Thus, wildfire smoke exposure is a growing problem [1,4,5], exposing 25 million individuals in the United States (U.S.) to extreme concentrations (>100 μg/m³) in 2020 [6] and projected to expose more in the future [7]. While wildfire smoke contains various components (e.g., carbon dioxide, nitrogen oxides, volatile organic compounds) [8], PM_2.5 is a major health-harming component [9].

Like non-wildfire PM_2.5, wildfire PM_2.5 can cause adverse health outcomes by inducing oxidative stress and inflammation and by activating the autonomic nervous system [10], but wildfire PM_2.5 contains higher concentrations of organic compounds (e.g., organic carbon) with high oxidative potential [11,12,13]. When smoke lingers, photochemical aging can result in wildfire PM_2.5 causing more lung epithelial cell death [14].

The effects of wildfire smoke may be worse for older adults (≥65 years). Although older adults spend a majority of their time indoors [15,16], wildfire smoke infiltrates homes and worsens indoor air quality [17]. Older adults often have pre-existing conditions and physiological susceptibility to environmental exposures [8,9], which puts them at a higher risk for health effects from wildfire smoke exposure compared to younger adults [10,18]. Within this population, demographic characteristics such as older age (e.g., 90 years versus 65 years), poverty or wealth, or access to health care may modify the risk of health effects from wildfire smoke exposure. For example, running costly air filtration can mitigate exposure, and distance to the nearest hospital may affect healthcare-seeking behavior.

Among older Americans, Alzheimer’s disease and related dementias (ADRD) are the fifth leading cause of death [19] and a growing burden [20]. In 2014, ADRD was estimated to affect 5 million older Americans, and by 2060, as the population ages, ADRD is predicted to affect 14 million older Americans [21]. Environmental exposures may contribute to cognitive decline [22,23,24] or dementia [25] and increase health risks for older adults with ADRD [26,27,28]. Two studies have reported an association between long-term wildfire PM_2.5 exposure and incident dementia [29,30]. Previous cohort studies of Medicare beneficiaries found that long-term all-source PM_2.5 exposure was associated with ADRD hospitalization [31,32]. A recent study reported an association between hot temperatures and hospitalization with an ADRD diagnosis code in the Medicare Fee-for-Service population [33]. As the population of older adults with ADRD grows, it is crucial to understand if wildfire smoke acutely impacts this vulnerable group.

To our knowledge, no one has evaluated associations between short-term wildfire PM_2.5 and cause-specific hospitalizations among ADRD patients. Short-term wildfire PM_2.5 could increase hospitalization among older adults with ADRD, as these individuals have a reduced ability to self-manage their health [34]. Impaired cognitive ability may reduce health-protective behaviors during wildfire PM_2.5 exposure, such as closing windows, using air filters, evacuating from nearby wildfire disasters, and seeking assistance. Moreover, older adults with ADRD face a higher risk of co-morbidities that may increase hospitalization risks, including pneumonia [35,36,37], congestive heart failure [38], chronic pulmonary disease [39], and depression [40,41]. At the same time, individuals living with ADRD may have additional home health nursing or other support, potentially protecting them from wildfire smoke exposure (i.e., turning on filtration systems) [17,42,43]. Certain factors, such as older age, sex, urbanicity, and area-level poverty, may further increase the risk of hospitalizations for ADRD patients. Additionally, we anticipate that wildfire PM_2.5 could impact emergency and non-emergency hospitalizations differently, with potential increases in emergency hospitalizations and delays in non-urgent healthcare use during wildfire smoke events.

Here, we conducted a 2006–2016 nationwide study of Medicare Fee-for-Service beneficiaries with a prior ADRD hospitalization to evaluate the relationship between lagged short-term wildfire PM_2.5 exposure and hospitalizations for circulatory and respiratory diseases, anxiety, and depression. This study population is a large sub-population of vulnerable older Americans. We tested for effect modifications by individual age, sex, and zip code-level urbanicity and poverty. We also tested whether the effects differed for emergency versus non-emergency hospitalizations.

2. Materials and Methods

This study was approved by the Institutional Review Boards at WIRB (single IRB for a multi-site project), the Harvard T.H. Chan School of Public Health, and the Columbia University Mailman School of Public Health.

2.1. Study Overview

Our study identified Medicare Fee-for-Service enrollees (65+) with an ADRD hospitalization between 2000 and 2016 and all of their following hospitalizations until 2016 from the Medicare Provider Analysis and Review (MEDPAR) file. We included individuals with a hospitalization for “Alzheimer’s disease, related disorders, or senile dementia” in the first 10 diagnostic positions between 2000 and 2016. We used the International Classification of Diseases (ICD) codes 9 and 10 from the Chronic Conditions Warehouse (CCW) to determine ADRD-relevant billing codes (Supplementary Table S1) [44]. If an individual had an ADRD-related hospitalization, we included that individual’s hospitalizations thereafter until October 31, 2016. Due to wildfire seasonality for these years, we limited the study to April to October months.

2.2. Cause-Specific Hospitalization Outcomes

We used inpatient claims from the Center for Medicare and Medicaid Services. Claims included hospitalization date, patient residential zip code, age, sex, and emergency hospitalization status. We used the ICD 9 and 10 codes in the first 5 diagnostic positions to identify hospitalizations related to circulatory, respiratory, anxiety, and depression endpoints (Supplementary Table S1). We considered the emergency hospitalizations to include emergency and urgent care visits. Otherwise, we considered the hospitalizations to be non-emergency.

2.3. Wildfire PM_2.5 Exposure

We leveraged daily wildfire smoke PM_2.5 predictions across the contiguous U.S. for April to October from 2006 to 2016 [6]. Briefly, satellite imagery-based plume classification and deviations of PM_2.5 ground measures from expected pollution concentrations were used to identify the presence of wildfire smoke. A machine learning model, which fit gradient boosting trees, was used to generate wildfire PM_2.5-specific predictions across the contiguous U.S. on a 10 km × 10 km grid. Estimates were converted to zip code tabulation areas (ZCTAs) via intersection-weighted averaging for the population. We then linked ZCTAs to corresponding zip codes using the Uniform Data System (UDS) Mapper [45]. Additional information about processes to generate the wildfire PM_2.5 can be found in a prior study [6].

2.4. Covariate Data

We extracted the zip code-level meteorological variables of daily maximum temperature and daily minimum relative humidity from gridMET [46], which combines and interpolates gridded spatial and temporal meteorologic information to produce surface-level variables. We included these as potential confounders in our models.

2.5. Effect Modification

To investigate possible effect modifications, we considered age, sex, urbanicity, and zip code-level poverty. Individual-level information, i.e., age and sex, was available from hospitalization records. We classified individuals into age categories, ≤75 and >75, and sex categories, female and male. For the urbanicity assignment, we used the rural–urban commuting area (RUCA) codes, and we categorized zip codes with RUCA codes 1–3 to be urban areas and codes 4–10 to be rural areas [47,48]. For the poverty assignment, we used the ratio of household income to poverty from the 5-year American Community Survey (ACS) 2007–2011 for study years prior to 2012 and the 5-year ACS 2012–2016 for study years from 2012 onward. We dichotomized zip codes into those with <20% and ≥20% of households living below the federal poverty threshold.

2.6. Study Design and Statistical Analysis

We employed a bi-directional, time-stratified case-crossover design [49,50]. In a case-crossover design, an individual acts as their own control to assess the impact of acute, transient exposures. As such, time-invariant confounding is eliminated by the design. Exposures before an event (i.e., a case day is a day with hospitalization) are compared to exposures before a non-event (i.e., a control day is a day without hospitalization). We matched control days on weekday, month, and year to account for confounding by long-term, seasonal, and day-of-the-week trends [51]. Though this study design accounts for time-invariant confounding, we still need to control for potential confounders that vary within a month, such as temperature and relative humidity.

Among the 1,850,590 Medicare Fee-for-Service enrollees (65+) identified with an ADRD hospitalization between 2000 and 2016, 927,581 enrollees were in our final study population because they had hospitalizations where the case and control periods had wildfire PM_2.5 exposure.

We implemented the case-crossover design using conditional logistic regression models and included distributed lag nonlinear model (DLNM) terms [52,53], allowing us to estimate the exposure–response relationship nonlinearly. We chose the degrees of freedom (df) for the nonlinear terms on the exposure–response and lag–response constraints based on the lowest Akaike Information Criterion (AIC) [54] across a range of potential values (2–5 for exposure–response and 2–5 for lag–response). For our main analyses, which included all hospitalizations (i.e., emergency and non-emergency), we included DLNM terms for daily zip code-level wildfire PM_2.5 exposure, adjusting nonlinearly for daily maximum temperature and minimum relative humidity with 0–6 day lags. We ran separate models for hospitalization outcomes related to circulatory disease, respiratory disease, anxiety, and depression. Based on the lowest AIC, wildfire PM_2.5 was modeled as linear for the exposure–response and with 4–5 df for the lag–response depending on the hospitalization subtype (i.e., 4 df for circulatory, 5 df for respiratory, 5 df for anxiety, and 5 df for depression). The meteorological variables were modeled with 3 df for the exposure–response and 4–5 df for the lag–response. For these main analyses, we were primarily interested in the cumulative associations and secondarily interested in the lag-specific associations. For the effect modification analyses, we ran models stratified by participant age, sex, urbanicity of residence, and zip code-level poverty. As a secondary analysis, we ran models for emergency hospitalizations and non-emergency hospitalizations separately. For analyses evaluating effect modification and emergency hospitalization versus non-emergency hospitalization, the df values chosen based on AIC were similar to the main analyses.

As a sensitivity analysis, we redefined the first ADRD hospitalization of interest, such that an individual’s first ADRD hospitalization occurred between 2006 and 2016 instead of between 2000 and 2016, since the epidemiologic analysis spanned from 2006 to 2016. Additionally, we considered mortality to investigate possible competing risk by evaluating death instead of hospitalization as an outcome. Analyses were conducted in R, version 4.2.0 (22 April 2022). The related code is available online at https://github.com/NSAPH-Projects/wf_pm_adrd. We presented findings for every 10 μg/m³ increase because most health studies on wildfire use this increment.

3. Results

3.1. Medicare Enrollees, Wildfire PM_2.5 Exposure, and Cause-Specific Hospitalizations

Among the 1,850,590 Medicare Fee-for-Service enrollees with a prior ADRD hospitalization from 2000 to 2016 (Supplementary Table S2), 927,581 enrollees had a hospitalization where case days or control days had wildfire PM_2.5 exposure, and this group constituted our study population (Table 1). Most ADRD patients were older than 75 years (83.9%), female (62.6%), resided in urban areas (60.3%), and lived in lower-poverty areas (77.0%). Between the case and control periods, wildfire PM_2.5 exposure, daily maximum temperature, and daily minimum relative humidity were similar across the cause-specific hospitalizations (Supplementary Table S3). Broadly, daily wildfire PM_2.5 was right skewed (Supplementary Figure S1), highest in the summer months (e.g., July, August), lowest during the late fall months (e.g., October) (Supplementary Figure S2), and more prevalent in the Western U.S. (Supplementary Figure S3).

Among this cohort, there were 1,546,753 non-mutually exclusive hospitalizations for circulatory, respiratory, anxiety, or depression endpoints from 2006 to 2016 in 11,837 zip codes. Circulatory hospitalizations (71.7%) were the most common type, followed by hospitalizations for respiratory endpoints (43.6%), depression (2.9%), and anxiety (0.7%) (Supplementary Figures S4 and S5). Most hospitalizations were emergencies (90.7%).

3.2. Associations Between Wildfire PM_2.5 and Cause-Specific Hospitalizations

We display results as rate ratios (RRs) and 95% confidence intervals (CIs) per 10 μg/m³ increase in wildfire PM_2.5. We found null cumulative associations between wildfire PM_2.5 exposure across the week prior and hospitalizations for circulatory disease, respiratory disease, and anxiety among Medicare Fee-for-Service enrollees with ADRD (Figure 1). On days closer to the case day (i.e., lag 0, lag 3), we observed null associations between wildfire PM_2.5 and these hospitalizations. On lag day 6, a 10 μg/m³ increase in exposure was positively associated with circulatory, respiratory, and anxiety hospitalizations, especially at higher concentrations of exposure (Supplementary Figure S6). For depression-related outcomes, we observed that a 10 μg/m³ increase in wildfire PM_2.5 on lag 0 (same-day) was associated with a decrease (RR = 0.94, 95% CI: 0.90, 0.99) in hospitalizations (Figure 1, Supplementary Figure S6). Other wildfire PM_2.5 lags were not cumulatively or individually associated with depression hospitalizations.

3.3. Effect Modification by Individual Characteristics and Zip Code-Level Characteristics

Overall, we did not find evidence of effect modifications by age groups but observed a stratum-specific effect. For those aged ≤75 years, a daily 10 μg/m³ increase in wildfire PM_2.5 exposure was associated with a decrease in anxiety hospitalization rates from lags 0 to 5 (RR = 0.63, 95% CI: 0.46–0.87) (Figure 2), but, otherwise, associations were null in both ages groups. Similarly, we did not observe effect modifications by sex, finding null associations for both females and males across all hospitalization endpoints in stratified models (Supplementary Figure S7).

We also assessed zip code-level characteristics (i.e., urbanicity and poverty) as effect measure modifiers. We observed effect modifications by urbanicity as in urban areas compared to rural areas, and effect estimates were lower for circulatory and respiratory hospitalizations. In urban areas, a 10 μg/m³ increase in daily wildfire PM_2.5 was negatively associated with circulatory hospitalization rates cumulatively for lags 0–2 (RR = 0.98, 95% CI: 0.97–1.00) and respiratory hospitalization rates cumulatively for lags 0–4 (RR = 0.97, 95% CI: 0.95–1.00). In rural areas, we observed an association between a 10 μg/m³ increase in daily wildfire PM_2.5 and circulatory (RR = 1.02, 95% CI: 0.99–1.04) and respiratory (RR = 1.03, 95% CI: 1.00–1.06) hospitalizations cumulatively across lags 0–6 (Figure 3). For area-level poverty, we observed that effect estimates were overall similar between strata, with an exception for lag 0 in high-poverty areas for depression hospitalizations. In high-poverty areas (i.e., ≥20%), hospitalizations for depression decreased at lag 0 (RR = 0.89, 95% CI: 0.81–0.99) (Supplementary Figure S8), but, otherwise, associations were null for other endpoints in both high- and lower-poverty zip codes.

3.4. Associations Between Wildfire PM_2.5 and Emergency Hospitalization Status

In a secondary analysis, we observed differences by emergency hospitalization status for circulatory and respiratory hospitalizations. Wildfire PM_2.5 exposure was associated with slightly lower non-emergency circulatory and respiratory admissions for portions of the lag period (Supplementary Figure S9). For example, a 10 μg/m³ increase in daily wildfire PM_2.5 was associated with a decreased cumulative risk of non-emergency circulatory hospitalizations across the entire exposure period, lags 0–6 (RR = 0.96, 95% CI: 0.93–0.99), and non-emergency respiratory hospitalizations for lags 0–4 (RR = 0.95, 95% CI: 0.91–0.99). The relationship between wildfire PM_2.5 exposure and emergency hospitalizations was null for all cause-specific hospitalizations, except for a reduced rate of same-day depression-related hospitalizations, similar to the main findings.

3.5. Sensitivity Analyses

Results from sensitivity analyses did not differ from our main analyses. For a sensitivity analysis where we restricted the first ADRD hospitalization from 2000–2016 to 2006–2016, the results were null (Supplementary Figure S10). When analyzing mortality as the outcome to evaluate possible competing risks, we did not observe an association between wildfire PM_2.5 and mortality (Supplementary Figure S11).

4. Discussion

In this national U.S. 2006–2016 study, we assessed the relationship between short-term wildfire PM_2.5 exposure and circulatory-, respiratory-, anxiety-, and depression-related hospitalizations among Medicare Fee-for-Service enrollees with a prior ADRD hospitalization. Overall, we observed null associations except for depression, for which the hospitalizations decreased on the same day as exposure. Most associations remained null within individual- and zip code-level sub-groups. However, we saw reduced non-emergency hospitalizations and circulatory and respiratory hospitalizations in urban areas and increased circulatory and respiratory visits in rural areas in the 6 days following wildfire PM_2.5 exposure.

Our null findings between wildfire smoke and several cause-specific hospitalizations among ADRD patients could be because older adults with ADRD likely have more support to meet high medical needs compared to older adults without ADRD [43]. Prior work observed that wildfire smoke increased the risk of hospitalization among older adults [55,56], often finding that the risk differed by age, sex, race, income, and urbanicity [57,58]. However, studies did not examine whether older adults’ medical and social support networks buffer against impacts of wildfire exposures. Some older adults with severe cognitive decline may reside in nursing homes [42], which could reduce exposure to wildfire smoke (e.g., closing windows, turning on air filters) or decrease the chances of hospitalization if caregivers respond to early signs of a health problem. Future studies could consider how positive support of older adults with dementia may modify the impacts of environmental exposures, such as wildfire smoke, on health. Future studies accounting for individual-level variables, such as healthcare access, household income, home air filtration use, social support, and nursing home status, are crucial to advance our understanding of the relationship between outdoor wildfire PM_2.5 exposure and health among older adults.

We observed that same-day wildfire smoke exposure was associated with reduced depression-related hospitalization rates, possibly due to short-term delays in seeking healthcare among older adults with ADRD. Wildfire smoke can induce negative mental, emotional, and psychosocial impacts, including distress and sadness resulting from changes to the environment [59,60]. Wildfire smoke may be especially harmful to older adults with ADRD due to their high underlying prevalence of mental health conditions. For example, 20–30% of ADRD patients have depressive symptoms [61,62]. Unlike circulatory, respiratory, or anxiety-related health events, depression may manifest in feeling sad, withdrawn, detached, and/or helpless [63], leading to a lower likelihood of seeking immediate medical attention. Future studies should continue to examine how wildfire smoke exposure acutely impacts depression and other mental health outcomes.

We observed different associations between wildfire PM_2.5 and circulatory and respiratory hospitalizations in urban areas versus rural areas, similar to some prior studies of all-source PM_2.5 [64,65]. For example, an investigation of Medicare beneficiaries found that PM_2.5 exposure was associated with an increased risk of cardiovascular but not respiratory hospitalizations in urban U.S. counties and an increased risk of respiratory but not cardiovascular hospitalizations in rural counties [66]. Factors such as income, transportation, and physician availability may produce differential barriers to seeking and receiving health care in urban versus rural areas. Urban areas typically have more healthcare facilities, which are especially important for responding to acute health concerns [67,68]. While there are often substantial wealth disparities within urban places, urban areas still tend to have a higher average socioeconomic status, which is associated with overall better health outcomes [69]. Collectively, these factors may lead to better management of chronic conditions and more timely medical interventions in urban areas, thus reducing the need for hospitalization in response to short-term exposures. This is consistent with our observation of decreased rates of acute circulatory and respiratory hospitalizations in urban areas and increased rates in rural areas.

We observed reduced non-emergency circulatory and respiratory hospitalizations following elevated wildfire PM_2.5 exposure. Hutchinson et al. [70] used a case-crossover study to evaluate the effects of a California wildfire event from 22 October to 5 November 2007 on hospitalizations among Medi-Cal Fee-for-Service beneficiaries aged 18–65. They observed a drop in all-cause hospitalizations in the 1–10 days following elevated wildfire PM_2.5 exposure and reduced respiratory hospitalizations 11–15 days later, potentially indicating that individuals minimized healthcare utilization for non-urgent medical needs. Expanding on Hutchinson et al.’s findings that focused on an average younger and lower-income population in California, we found a similar drop in non-emergency circulatory and respiratory hospitalizations in our nationwide study on older adults with ADRD.

Our study had several strengths and limitations. First, we leveraged the nationwide hospitalization records for the entire Medicare Fee-for-Service population. By covering a wide geographic area across the contiguous U.S., our results are representative of diverse environmental and demographic contexts, including both urban and rural settings. Despite this broad coverage of hospitalization records, we did not include years after 2017 with more wildfires and much higher concentrations of wildfire PM_2.5 [71] (our data spanned from 2006 to 2016). During years with more wildfires, there would be more exposed cases, increasing the statistical power to detect an association, and a wider range of wildfire PM_2.5 concentrations, improving our ability to detect a potential nonlinear relationship between exposure and hospitalizations. Given that many severe wildfire events occurred after 2016 (e.g., the 2017 California Camp Fire, the 2025 Los Angeles Wildfires) [72,73], future studies should include later years.

Second, by identifying our cohort using ADRD-related hospitalizations, our study likely underestimated the number of older adults with ADRD, which severely limited our study’s statistical power and ability to detect an association. Relatedly, our study likely included those with more severe ADRD and underlying conditions that would precipitate a hospitalization. We identified ADRD using hospitalization records, which can lead to false negatives for ADRD compared to using electronic health records [74]. Furthermore, using diagnostic codes rather than neurology visits and imaging studies can lead to outcome misclassification [75]. Future studies should consider additional avenues to identifying older adults with ADRD, such as linking claims data to electronic health record data or acquiring additional Medicare data, such as outpatient and neurology visits. These methods would allow studies to identify earlier, less severe ADRD cases more accurately.

The use of claims data also precluded us from studying important modifying factors, such as social support, behavioral changes, or access to air filtration, which might have explained our null results. Due to data limitations, we also did not examine patient race, ethnicity, or socioeconomic status, which prior studies have identified as important effect measure modifiers [30,57,76,77]. Additional individual-level variables measuring access to resources, such as healthcare or air filtration, could also be crucial effect measure modifiers in future studies [76].

Although we used the most spatially and temporally resolved publicly available data on wildfire PM_2.5 concentrations and developed zip code exposure estimates via population weighting, exposure measurement errors remain. Future studies may be interested in developing and using more resolved wildfire PM_2.5 exposure data in combination with individual-level location information. Some additional exposure measurement errors may have accrued when beneficiaries moved during the study period. However, given the study design, this error would not differ between case and control days; any bias would, thus, be toward the null. Finally, death may be a competing risk for hospitalizations [78]. However, we conducted a sensitivity analysis using mortality as our outcome and did not observe a relationship between short-term exposure to wildfire PM_2.5 and mortality, so it is unlikely that mortality obfuscated our overall findings.

We observed null associations between wildfire PM_2.5 and nationwide Medicare hospitalizations between 2006 and 2016 among ADRD patients, except for decreased same-day depression-related hospitalizations. As older adults with ADRD may be particularly vulnerable to the health effects of climate change-related exposures, it is crucial to continue research for this population. Future work should examine the possible role of support networks, comorbidities, socioeconomic status, and accurate ADRD status identification to advance ADRD climate and health research.

Author Contributions

V.D.—conceptualization, data curation, formal analysis, writing—original draft, writing—review and editing; H.M.—conceptualization, data curation, writing—review and editing; K.T.—conceptualization, data curation, writing—review and editing; M.L.C.—conceptualization, data curation, writing—review and editing; M.-A.K.—conceptualization, data curation, writing—review and editing; J.A.C.—conceptualization, data curation, funding acquisition, supervision, writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

This study was approved by the Institutional Review Board at the Harvard T.H. Chan School of Public Health.

Data Availability Statement

Modeled wildfire PM_2.5 data are publicly available and obtainable from https://www.stanfordecholab.com/wildfire_smoke (accessed on 5 May 2023). Medicare hospitalization data are available after purchase and an application process from the Centers for Medicare and Medicaid Services (https://www.cms.gov/Research-Statistics-Data-and-Systems/CMS-Information-Technology/AccesstoDataApplication/index (accessed on 14 January 2025)). Social and meteorologic variables can be accessed from data sources in the Methods Section.

Conflicts of Interest

Authors state that there are no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Table

Figure 1. Cumulative exposure–response curves for wildfire PM2.5 and (A) circulatory hospitalizations, (B) respiratory hospitalizations, (C) anxiety hospitalizations, and (D) depression hospitalizations across lag days 0–6. The rate ratios were estimated from a conditional logistic regression model, adjusted for daily maximum temperature and minimum relative humidity, and are for a 10 μg/m3 increase in wildfire PM2.5 for each daily lag. The dark line represents point estimates across daily lags, and the grey shading represents 95% confidence intervals. The dashed line represents a rate ratio of 1. The first ADRD hospitalization for Medicare Fee-for-Service enrollees occurred from 2000 to 2016.

Figure 2. Age-specific (i.e., ≤75 years versus >75 years) cumulative exposure–response curves for wildfire PM2.5 and circulatory hospitalizations, respiratory hospitalizations, anxiety hospitalizations, and depression hospitalizations across lag days 0–6. The rate ratios were estimated from a conditional logistic regression model, adjusted for daily maximum temperature and minimum relative humidity, and are for a 10 μg/m3 increase in wildfire PM2.5 for each daily lag. The dark line represents point estimates across daily lags, and the grey shading represents 95% confidence intervals. The dashed line represents a rate ratio of 1. The first ADRD hospitalization for Medicare Fee-for-Service enrollees occurred from 2000 to 2016.

Figure 3. Urbanicity-specific (i.e., urban versus rural) cumulative exposure–response curves for wildfire PM2.5 and circulatory hospitalizations, respiratory hospitalizations, anxiety hospitalizations, and depression hospitalizations across lag days 0–6. The rate ratios were estimated from a conditional logistic regression model, adjusted for daily maximum temperature and minimum relative humidity, and are for a 10 μg/m3 increase in wildfire PM2.5 for each daily lag. The dark line represents point estimates across daily lags, and the grey shading represents 95% confidence intervals. The dashed line represents a rate ratio of 1. The first ADRD hospitalization for Medicare Fee-for-Service enrollees occurred from 2000 to 2016.

Table 1

Individual- and zip code-level characteristics of Medicare Fee-for-Service enrollees with a prior ADRD hospitalization from 2006 to 2016. This sub-population had a hospitalization where case or control periods had wildfire PM_2.5 exposure.

Individual-Level Covariates	Medicare Enrollees with a Prior ADRD Hospitalization from 2000 to 2016
	N = 927,581
Age, n (%)
≤75 years	117,721 (12.7)
>75 years	778,292 (83.9)
Unknown	31,568 (3.4)
Sex, n (%)
Female	580,716 (62.6)
Male	346,865 (37.4)
Zip code-level covariates
Poverty a, n (%)
<20%	714,494 (77.0)
≥20%	207,620 (22.4)
Missing	5467 (0.6)
Urbanicity b, n (%)
Urban	558,940 (60.3)
Rural	368,459 (39.7)
Missing	182 (0.2)

^a The ratio of income to poverty from the 5-year American Community Survey (ACS) was used to determine the percentage of households living below the federal poverty threshold. The 2007–2011 ACS was used to categorize zip codes prior to 2012, while the 2012–2016 ACS was used for zip codes from 2012 onward. ^b Zip codes located in rural–urban commuting area (RUCA) codes 1–3 were categorized as urban, and those in codes 4–10 were categorized as rural areas.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/fire8030097/s1, Supplementary Table S1; Supplementary Table S2; Supplementary Table S3; Supplementary Figure S1; Supplementary Figure S2; Supplementary Figure S3; Supplementary Figure S4; Supplementary Figure S5; Supplementary Figure S6; Supplementary Figure S7; Supplementary Figure S8; Supplementary Figure S9; Supplementary Figure S10; Supplementary Figure S11.

References

1. Abatzoglou, J.T.; Williams, A.P. Impact of anthropogenic climate change on wildfire across western US forests. Proc. Natl. Acad. Sci. USA; 2016; 113, pp. 11770-11775. [DOI: https://dx.doi.org/10.1073/pnas.1607171113] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27791053]

2. Abatzoglou, J.T.; Battisti, D.S.; Williams, A.P.; Hansen, W.D.; Harvey, B.J.; Kolden, C.A. Projected Increases in Western US Forest Fire despite Growing Fuel Constraints. Commun. Earth Environ.; 2021; 2, 227. [DOI: https://dx.doi.org/10.1038/s43247-021-00299-0]

3. Parks, S.A.; Abatzoglou, J.T. Warmer and Drier Fire Seasons Contribute to Increases in Area Burned at High Severity in Western US Forests From 1985 to 2017. Geophys. Res. Lett.; 2020; 47, e2020GL089858. [DOI: https://dx.doi.org/10.1029/2020GL089858]

4. Marlon, J.R.; Bartlein, P.J.; Gavin, D.G.; Long, C.J.; Anderson, R.S.; Briles, C.E.; Brown, K.J.; Colombaroli, D.; Hallett, D.J.; Power, M.J. et al. Long-Term Perspective on Wildfires in the Western USA. Proc. Natl. Acad. Sci. USA; 2012; 109, pp. E535-E543. [DOI: https://dx.doi.org/10.1073/pnas.1112839109] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22334650]

5. Wilmot, T.Y.; Mallia, D.V.; Hallar, A.G.; Lin, J.C. Wildfire Plumes in the Western US Are Reaching Greater Heights and Injecting More Aerosols Aloft as Wildfire Activity Intensifies. Sci. Rep.; 2022; 12, 12400. [DOI: https://dx.doi.org/10.1038/s41598-022-16607-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35859160]

6. Childs, M.L.; Li, J.; Wen, J.; Heft-Neal, S.; Driscoll, A.; Wang, S.; Gould, C.F.; Qiu, M.; Burney, J.; Burke, M. Daily Local-Level Estimates of Ambient Wildfire Smoke PM2.5 for the Contiguous US. Environ. Sci. Technol.; 2022; 56, pp. 13607-13621. [DOI: https://dx.doi.org/10.1021/acs.est.2c02934] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36134580]

7. Liu, J.C.; Mickley, L.J.; Sulprizio, M.P.; Dominici, F.; Yue, X.; Ebisu, K.; Anderson, G.B.; Khan, R.F.A.; Bravo, M.A.; Bell, M.L. Particulate Air Pollution from Wildfires in the Western US under Climate Change. Clim. Change; 2016; 138, pp. 655-666. [DOI: https://dx.doi.org/10.1007/s10584-016-1762-6]

8. Urbanski, S.P.; Hao, W.M.; Baker, S. Chapter 4 Chemical Composition of Wildland Fire Emissions. Developments in Environmental Science; Bytnerowicz, A.; Arbaugh, M.J.; Riebau, A.R.; Andersen, C. Wildland Fires and Air Pollution Elsevier: Amsterdam, The Netherlands, 2008; Volume 8, pp. 79-107.

9. Vicente, A.; Alves, C.; Calvo, A.I.; Fernandes, A.P.; Nunes, T.; Monteiro, C.; Almeida, S.M.; Pio, C. Emission Factors and Detailed Chemical Composition of Smoke Particles from the 2010 Wildfire Season. Atmos. Environ.; 2013; 71, pp. 295-303. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2013.01.062]

10. Chen, H.; Samet, J.M.; Bromberg, P.A.; Tong, H. Cardiovascular Health Impacts of Wildfire Smoke Exposure. Part. Fibre Toxicol.; 2021; 18, 2. [DOI: https://dx.doi.org/10.1186/s12989-020-00394-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33413506]

11. Black, C.; Tesfaigzi, Y.; Bassein, J.A.; Miller, L.A. Wildfire Smoke Exposure and Human Health: Significant Gaps in Research for a Growing Public Health Issue. Environ. Toxicol. Pharmacol.; 2017; 55, pp. 186-195. [DOI: https://dx.doi.org/10.1016/j.etap.2017.08.022]

12. Williams, K.M.; Franzi, L.M.; Last, J.A. Cell-Specific Oxidative Stress and Cytotoxicity after Wildfire Coarse Particulate Matter Instillation into Mouse Lung. Toxicol. Appl. Pharmacol.; 2013; 266, pp. 48-55. [DOI: https://dx.doi.org/10.1016/j.taap.2012.10.017] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23142465]

13. Verma, V.; Polidori, A.; Schauer, J.J.; Shafer, M.M.; Cassee, F.R.; Sioutas, C. Physicochemical and Toxicological Profiles of Particulate Matter in Los Angeles during the October 2007 Southern California Wildfires. Environ. Sci. Technol.; 2009; 43, pp. 954-960. [DOI: https://dx.doi.org/10.1021/es8021667] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19245042]

14. Atwi, K.; Wilson, S.N.; Mondal, A.; Edenfield, R.C.; Symosko Crow, K.M.; El Hajj, O.; Perrie, C.; Glenn, C.K.; Easley, C.A.; Handa, H. et al. Differential Response of Human Lung Epithelial Cells to Particulate Matter in Fresh and Photochemically Aged Biomass-Burning Smoke. Atmos. Environ.; 2022; 271, 118929. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2021.118929]

15. Wu, C.-Y.; Dodge, H.H.; Reynolds, C.; Barnes, L.L.; Silbert, L.C.; Lim, M.M.; Mattek, N.; Gothard, S.; Kaye, J.A.; Beattie, Z. In-Home Mobility Frequency and Stability in Older Adults Living Alone With or Without MCI: Introduction of New Metrics. Front. Digit. Health; 2021; 3, 764510. [DOI: https://dx.doi.org/10.3389/fdgth.2021.764510]

16. Petersen, J.; Austin, D.; Mattek, N.; Kaye, J. Time Out-of-Home and Cognitive, Physical, and Emotional Wellbeing of Older Adults: A Longitudinal Mixed Effects Model. PLoS ONE; 2015; 10, e0139643. [DOI: https://dx.doi.org/10.1371/journal.pone.0139643] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26437228]

17. Liang, Y.; Sengupta, D.; Campmier, M.J.; Lunderberg, D.M.; Apte, J.S.; Goldstein, A.H. Wildfire Smoke Impacts on Indoor Air Quality Assessed Using Crowdsourced Data in California. Proc. Natl. Acad. Sci. USA; 2021; 118, e2106478118. [DOI: https://dx.doi.org/10.1073/pnas.2106478118] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34465624]

18. Liu, J.C.; Pereira, G.; Uhl, S.A.; Bravo, M.A.; Bell, M.L. A Systematic Review of the Physical Health Impacts from Non-Occupational Exposure to Wildfire Smoke. Environ. Res.; 2015; 136, pp. 120-132. [DOI: https://dx.doi.org/10.1016/j.envres.2014.10.015]

19. Underlying Cause of Death, 1999–2020. Available online: https://wonder.cdc.gov/ucd-icd10.html (accessed on 25 July 2023).

20. Fang, M.; Hu, J.; Weiss, J.; Knopman, D.S.; Albert, M.; Windham, B.G.; Walker, K.A.; Sharrett, A.R.; Gottesman, R.F.; Lutsey, P.L. et al. Lifetime Risk and Projected Burden of Dementia. Nat. Med.; 2025; pp. 1-5. [DOI: https://dx.doi.org/10.1038/s41591-024-03340-9]

21. Matthews, K.A.; Xu, W.; Gaglioti, A.H.; Holt, J.B.; Croft, J.B.; Mack, D.; McGuire, L.C. Racial and Ethnic Estimates of Alzheimer’s Disease and Related Dementias in the United States (2015–2060) in Adults Aged ≥ 65 Years. Alzheimer’s Dement.; 2019; 15, pp. 17-24. [DOI: https://dx.doi.org/10.1016/j.jalz.2018.06.3063] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30243772]

22. Parrón, T.; Requena, M.; Hernández, A.F.; Alarcón, R. Association between Environmental Exposure to Pesticides and Neurodegenerative Diseases. Toxicol. Appl. Pharmacol.; 2011; 256, pp. 379-385. [DOI: https://dx.doi.org/10.1016/j.taap.2011.05.006] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21601587]

23. Scieszka, D.; Bolt, A.M.; McCormick, M.A.; Brigman, J.L.; Campen, M.J. Aging, Longevity, and the Role of Environmental Stressors: A Focus on Wildfire Smoke and Air Quality. Front. Toxicol.; 2023; 5, 1267667. [DOI: https://dx.doi.org/10.3389/ftox.2023.1267667] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37900096]

24. Cleland, S.E.; Wyatt, L.H.; Wei, L.; Paul, N.; Serre, M.L.; West, J.J.; Henderson, S.B.; Rappold, A.G. Short-Term Exposure to Wildfire Smoke and PM2.5 and Cognitive Performance in a Brain-Training Game: A Longitudinal Study of U.S. Adults. Environ. Health Perspect.; 2022; 130, 067005. [DOI: https://dx.doi.org/10.1289/EHP10498] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35700064]

25. Shi, L.; Steenland, K.; Li, H.; Liu, P.; Zhang, Y.; Lyles, R.H.; Requia, W.J.; Ilango, S.D.; Chang, H.H.; Wingo, T. et al. A National Cohort Study (2000–2018) of Long-Term Air Pollution Exposure and Incident Dementia in Older Adults in the United States. Nat. Commun.; 2021; 12, 6754. [DOI: https://dx.doi.org/10.1038/s41467-021-27049-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34799599]

26. Zhang, Y.; Ebelt, S.T.; Shi, L.; Scovronick, N.C.; D’Souza, R.R.; Steenland, K.; Chang, H.H. Short-Term Associations between Warm-Season Ambient Temperature and Emergency Department Visits for Alzheimer’s Disease and Related Dementia in Five US States. Environ. Res.; 2023; 220, 115176. [DOI: https://dx.doi.org/10.1016/j.envres.2022.115176]

27. Xu, Z.; Tong, S.; Cheng, J.; Zhang, Y.; Wang, N.; Zhang, Y.; Hayixibayi, A.; Hu, W. Heatwaves, Hospitalizations for Alzheimer’s Disease, and Postdischarge Deaths: A Population-Based Cohort Study. Environ. Res.; 2019; 178, 108714. [DOI: https://dx.doi.org/10.1016/j.envres.2019.108714] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31520832]

28. Klompmaker, J.O.; Laden, F.; Browning, M.H.E.M.; Dominici, F.; Jimenez, M.P.; Ogletree, S.S.; Rigolon, A.; Zanobetti, A.; Hart, J.E.; James, P. Associations of Greenness, Parks, and Blue Space With Neurodegenerative Disease Hospitalizations Among Older US Adults. JAMA Netw. Open; 2022; 5, e2247664. [DOI: https://dx.doi.org/10.1001/jamanetworkopen.2022.47664]

29. Zhang, B.; Weuve, J.; Langa, K.M.; D’Souza, J.; Szpiro, A.; Faul, J.; Mendes de Leon, C.; Gao, J.; Kaufman, J.D.; Sheppard, L. et al. Comparison of Particulate Air Pollution From Different Emission Sources and Incident Dementia in the US. JAMA Intern. Med.; 2023; 183, pp. 1080-1089. [DOI: https://dx.doi.org/10.1001/jamainternmed.2023.3300] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37578757]

30. Elser, H.; Frankland, T.B.; Chen, C.; Tartof, S.Y.; Mayeda, E.R.; Lee, G.S.; Northrop, A.J.; Torres, J.M.; Benmarhnia, T.; Casey, J.A. Wildfire Smoke Exposure and Incident Dementia. JAMA Neurol.; 2025; 82, pp. 40-48. [DOI: https://dx.doi.org/10.1001/jamaneurol.2024.4058]

31. Mork, D.; Braun, D.; Zanobetti, A. Time-Lagged Relationships between a Decade of Air Pollution Exposure and First Hospitalization with Alzheimer’s Disease and Related Dementias. Environ. Int.; 2023; 171, 107694. [DOI: https://dx.doi.org/10.1016/j.envint.2022.107694]

32. Lee, M.; Schwartz, J.; Wang, Y.; Dominici, F.; Zanobetti, A. Long-Term Effect of Fine Particulate Matter on Hospitalization with Dementia. Environ. Pollut.; 2019; 254, 112926. [DOI: https://dx.doi.org/10.1016/j.envpol.2019.07.094] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31404729]

33. Delaney, S.W.; Stegmuller, A.; Mork, D.; Mock, L.; Bell, M.L.; Gill, T.M.; Braun, D.; Zanobetti, A. Extreme Heat and Hospitalization Among Older Persons With Alzheimer Disease and Related Dementias. JAMA Intern. Med.; 2025; [DOI: https://dx.doi.org/10.1001/jamainternmed.2024.7719] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39899291]

34. Ibrahim, J.E.; Anderson, L.J.; MacPhail, A.; Lovell, J.J.; Davis, M.-C.; Winbolt, M. Chronic Disease Self-Management Support for Persons with Dementia, in a Clinical Setting. J. Multidiscip. Health; 2017; 10, pp. 49-58. [DOI: https://dx.doi.org/10.2147/JMDH.S121626] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28182172]

35. Foley, N.C.; Affoo, R.H.; Martin, R.E. A Systematic Review and Meta-Analysis Examining Pneumonia-Associated Mortality in Dementia. Dement. Geriatr. Cogn. Disord.; 2014; 39, pp. 52-67. [DOI: https://dx.doi.org/10.1159/000367783] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25342272]

36. Natalwala, A.; Potluri, R.; Uppal, H.; Heun, R. Reasons for Hospital Admissions in Dementia Patients in Birmingham, UK, during 2002–2007. Dement. Geriatr. Cogn. Disord.; 2008; 26, pp. 499-505. [DOI: https://dx.doi.org/10.1159/000171044] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19005254]

37. Sampson, E.L.; Blanchard, M.R.; Jones, L.; Tookman, A.; King, M. Dementia in the Acute Hospital: Prospective Cohort Study of Prevalence and Mortality. Br. J. Psychiatry; 2009; 195, pp. 61-66. [DOI: https://dx.doi.org/10.1192/bjp.bp.108.055335] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19567898]

38. Phelan, E.A.; Borson, S.; Grothaus, L.; Balch, S.; Larson, E.B. Association of Incident Dementia With Hospitalizations. JAMA; 2012; 307, pp. 165-172. [DOI: https://dx.doi.org/10.1001/jama.2011.1964]

39. Davis, M.A.; Chang, C.-H.; Simonton, S.; Bynum, J.P.W. Trends in US Medicare Decedents’ Diagnosis of Dementia From 2004 to 2017. JAMA Health Forum; 2022; 3, e220346. [DOI: https://dx.doi.org/10.1001/jamahealthforum.2022.0346]

40. Kunicki, Z.J.; Frietchen, R.; McGeary, J.E.; Jiang, L.; Duprey, M.S.; Bayer, T.; Singh, M.; Primack, J.M.; Kelso, C.M.; Wu, W.-C. et al. Prevalence of Comorbid Depression and Insomnia Among Veterans Hospitalized for Heart Failure with Alzheimer Disease and Related Disorders. Am. J. Geriatr. Psychiatry; 2023; 31, pp. 428-437. [DOI: https://dx.doi.org/10.1016/j.jagp.2023.01.026] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36863973]

41. Elser, H.; Horváth-Puhó, E.; Gradus, J.L.; Smith, M.L.; Lash, T.L.; Glymour, M.M.; Sørensen, H.T.; Henderson, V.W. Association of Early-, Middle-, and Late-Life Depression With Incident Dementia in a Danish Cohort. JAMA Neurol.; 2023; 80, pp. 949-958. [DOI: https://dx.doi.org/10.1001/jamaneurol.2023.2309]

42. Toot, S.; Swinson, T.; Devine, M.; Challis, D.; Orrell, M. Causes of Nursing Home Placement for Older People with Dementia: A Systematic Review and Meta-Analysis. Int. Psychogeriatr.; 2017; 29, pp. 195-208. [DOI: https://dx.doi.org/10.1017/S1041610216001654] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27806743]

43. ASPE Office of the Assistant Secretary for Planning and Evaluation. A Profile of Older Adults with Dementia and Their Caregivers Issue Brief. Available online: https://aspe.hhs.gov/reports/profile-older-adults-dementia-their-caregivers-issue-brief-0 (accessed on 6 September 2024).

44. Dong, S. ADRD Medicare Hospitalization Data. 2022; Available online: https://github.com/ShuxinD/ADRDdata (accessed on 7 July 2023).

45. Health Center Program GeoCare Navigator. ZIP Code to ZCTA Crosswalk. Available online: https://geocarenavigator.hrsa.gov/ (accessed on 20 January 2025).

46. Abatzoglou, J.T. Development of Gridded Surface Meteorological Data for Ecological Applications and Modelling. Int. J. Climatol.; 2013; 33, pp. 121-131. [DOI: https://dx.doi.org/10.1002/joc.3413]

47. USDA. ERS—Rural-Urban Commuting Area Codes. Available online: https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/ (accessed on 15 February 2024).

48. U.S. Department of Health & Human Services Health Resources & Services Administration. Defining Rural Population. Available online: https://www.hrsa.gov/rural-health/about-us/what-is-rural (accessed on 3 May 2024).

49. Maclure, M. The Case-Crossover Design: A Method for Studying Transient Effects on the Risk of Acute Events. Am. J. Epidemiol.; 1991; 133, pp. 144-153. [DOI: https://dx.doi.org/10.1093/oxfordjournals.aje.a115853] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/1985444]

50. Mittleman, M.A.; Mostofsky, E. Exchangeability in the Case-Crossover Design. Int. J. Epidemiol.; 2014; 43, pp. 1645-1655. [DOI: https://dx.doi.org/10.1093/ije/dyu081]

51. Mittleman, M.A. Optimal Referent Selection Strategies in Case-Crossover Studies: A Settled Issue. Epidemiology; 2005; 16, 715. [DOI: https://dx.doi.org/10.1097/01.ede.0000183170.92955.25] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/16222159]

52. Gasparrini, A.; Armstrong, B.; Kenward, M.G. Distributed Lag Non-Linear Models. Stat. Med.; 2010; 29, pp. 2224-2234. [DOI: https://dx.doi.org/10.1002/sim.3940] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20812303]

53. Gasparrini, A. Modeling Exposure–Lag–Response Associations with Distributed Lag Non-Linear Models. Stat. Med.; 2014; 33, pp. 881-899. [DOI: https://dx.doi.org/10.1002/sim.5963] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24027094]

54. Gasparrini, A. Modelling Lagged Associations in Environmental Time Series Data: A Simulation Study. Epidemiology; 2016; 27, pp. 835-842. [DOI: https://dx.doi.org/10.1097/EDE.0000000000000533] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27400064]

55. Melton, C.C.; De Fries, C.M.; Smith, R.M.; Mason, L.R. Wildfires and Older Adults: A Scoping Review of Impacts, Risks, and Interventions. Int. J. Environ. Res. Public Health; 2023; 20, 6252. [DOI: https://dx.doi.org/10.3390/ijerph20136252]

56. Reid, C.E.; Brauer, M.; Johnston, F.H.; Jerrett, M.; Balmes, J.R.; Elliott, C.T. Critical Review of Health Impacts of Wildfire Smoke Exposure. Environ. Health Perspect.; 2016; 124, pp. 1334-1343. [DOI: https://dx.doi.org/10.1289/ehp.1409277] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27082891]

57. Liu, J.C.; Wilson, A.; Mickley, L.J.; Ebisu, K.; Sulprizio, M.P.; Wang, Y.; Peng, R.D.; Yue, X.; Dominici, F.; Bell, M.L. Who Among the Elderly Is Most Vulnerable to Exposure to and Health Risks of Fine Particulate Matter From Wildfire Smoke?. Am. J. Epidemiol.; 2017; 186, pp. 730-735. [DOI: https://dx.doi.org/10.1093/aje/kwx141] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28525551]

58. Reid, C.E.; Considine, E.M.; Watson, G.L.; Telesca, D.; Pfister, G.G.; Jerrett, M. Effect Modification of the Association between Fine Particulate Air Pollution during a Wildfire Event and Respiratory Health by Area-Level Measures of Socio-Economic Status, Race/Ethnicity, and Smoking Prevalence. Environ. Res. Health; 2023; 1, 025005. [DOI: https://dx.doi.org/10.1088/2752-5309/acc4e1]

59. Eisenman, D.P.; Galway, L.P. The Mental Health and Well-Being Effects of Wildfire Smoke: A Scoping Review. BMC Public Health; 2022; 22, 2274. [DOI: https://dx.doi.org/10.1186/s12889-022-14662-z] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36471306]

60. Eisenman, D.; McCaffrey, S.; Donatello, I.; Marshal, G. An Ecosystems and Vulnerable Populations Perspective on Solastalgia and Psychological Distress After a Wildfire. EcoHealth; 2015; 12, pp. 602-610. [DOI: https://dx.doi.org/10.1007/s10393-015-1052-1]

61. Tsuno, N.; Homma, A. What Is the Association between Depression and Alzheimer’s Disease?. Expert Rev. Neurother.; 2009; 9, pp. 1667-1677. [DOI: https://dx.doi.org/10.1586/ern.09.106]

62. Zubenko, G.S.; Zubenko, W.N.; McPherson, S.; Spoor, E.; Marin, D.B.; Farlow, M.R.; Smith, G.E.; Geda, Y.E.; Cummings, J.L.; Petersen, R.C. et al. A Collaborative Study of the Emergence and Clinical Features of the Major Depressive Syndrome of Alzheimer’s Disease. Am. J. Psychiatry; 2003; 160, pp. 857-866. [DOI: https://dx.doi.org/10.1176/appi.ajp.160.5.857] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/12727688]

63. Kanter, J.W.; Busch, A.M.; Weeks, C.E.; Landes, S.J. The Nature of Clinical Depression: Symptoms, Syndromes, and Behavior Analysis. Behav. Anal.; 2008; 31, pp. 1-21. [DOI: https://dx.doi.org/10.1007/BF03392158]

64. Chen, C.; Chan, A.; Dominici, F.; Peng, R.D.; Sabath, B.; Di, Q.; Schwartz, J.; Bell, M.L. Do Temporal Trends of Associations between Short-Term Exposure to Fine Particulate Matter (PM2.5) and Risk of Hospitalizations Differ by Sub-Populations and Urbanicity—A Study of 968 U.S. Counties and the Medicare Population. Environ. Res.; 2022; 206, 112271. [DOI: https://dx.doi.org/10.1016/j.envres.2021.112271]

65. Hasegawa, K.; Tsukahara, T.; Nomiyama, T. Short-Term Associations of Low-Level Fine Particulate Matter (PM2.5) with Cardiorespiratory Hospitalizations in 139 Japanese Cities. Ecotoxicol. Environ. Saf.; 2023; 258, 114961. [DOI: https://dx.doi.org/10.1016/j.ecoenv.2023.114961] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37137261]

66. Bravo, M.A.; Ebisu, K.; Dominici, F.; Wang, Y.; Peng, R.D.; Bell, M.L. Airborne Fine Particles and Risk of Hospital Admissions for Understudied Populations: Effects by Urbanicity and Short-Term Cumulative Exposures in 708 U.S. Counties. Environ. Health Perspect.; 2017; 125, pp. 594-601. [DOI: https://dx.doi.org/10.1289/EHP257]

67. Haggerty, J.L.; Roberge, D.; Lévesque, J.-F.; Gauthier, J.; Loignon, C. An Exploration of Rural–Urban Differences in Healthcare-Seeking Trajectories: Implications for Measures of Accessibility. Health Place; 2014; 28, pp. 92-98. [DOI: https://dx.doi.org/10.1016/j.healthplace.2014.03.005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24793139]

68. Chen, X.; Orom, H.; Hay, J.L.; Waters, E.A.; Schofield, E.; Li, Y.; Kiviniemi, M.T. Differences in Rural and Urban Health Information Access and Use. J. Rural. Health; 2019; 35, pp. 405-417. [DOI: https://dx.doi.org/10.1111/jrh.12335]

69. Blumenthal, S.J.; Kagen, J. The Effects of Socioeconomic Status on Health in Rural and Urban America. JAMA; 2002; 287, 109. [DOI: https://dx.doi.org/10.1001/jama.287.1.109-JMS0102-3-1]

70. Hutchinson, J.A.; Vargo, J.; Milet, M.; French, N.H.F.; Billmire, M.; Johnson, J.; Hoshiko, S. The San Diego 2007 Wildfires and Medi-Cal Emergency Department Presentations, Inpatient Hospitalizations, and Outpatient Visits: An Observational Study of Smoke Exposure Periods and a Bidirectional Case-Crossover Analysis. PLoS Med.; 2018; 15, e1002601. [DOI: https://dx.doi.org/10.1371/journal.pmed.1002601] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29990362]

71. Burke, M.; Childs, M.L.; de la Cuesta, B.; Qiu, M.; Li, J.; Gould, C.F.; Heft-Neal, S.; Wara, M. The Contribution of Wildfire to PM2.5 Trends in the USA. Nature; 2023; 622, pp. 761-766. [DOI: https://dx.doi.org/10.1038/s41586-023-06522-6]

72. O’Neill, S.M.; Diao, M.; Raffuse, S.; Al-Hamdan, M.; Barik, M.; Jia, Y.; Reid, S.; Zou, Y.; Tong, D.; West, J.J. et al. A Multi-Analysis Approach for Estimating Regional Health Impacts from the 2017 Northern California Wildfires. J. Air Waste Manag. Assoc.; 2021; 71, pp. 791-814. [DOI: https://dx.doi.org/10.1080/10962247.2021.1891994] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33630725]

73. US EPA. 2025 California Wildfires. Available online: https://www.epa.gov/ca/2025-california-wildfires (accessed on 29 January 2025).

74. Festa, N.; Price, M.; Weiss, M.; Moura, L.M.V.R.; Benson, N.M.; Zafar, S.; Blacker, D.; Normand, S.-L.T.; Newhouse, J.P.; Hsu, J. Evaluating The Accuracy Of Medicare Risk Adjustment For Alzheimer’s Disease And Related Dementias. Health Aff.; 2022; 41, pp. 1324-1332. [DOI: https://dx.doi.org/10.1377/hlthaff.2022.00185]

75. Lee, E.; Gatz, M.; Tseng, C.; Schneider, L.S.; Pawluczyk, S.; Wu, A.H.; Deapen, D. Evaluation of Medicare Claims Data as a Tool to Identify Dementia. J. Alzheimer’s Dis.; 2019; 67, pp. 769-778. [DOI: https://dx.doi.org/10.3233/JAD-181005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30689589]

76. Do, V.; Chen, C.; Benmarhnia, T.; Casey, J.A. Spatial Heterogeneity of the Respiratory Health Impacts of Wildfire Smoke PM2.5 in California. GeoHealth; 2024; 8, e2023GH000997. [DOI: https://dx.doi.org/10.1029/2023GH000997] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38560560]

77. Chen, C.; Schwarz, L.; Rosenthal, N.; Marlier, M.E.; Benmarhnia, T. Exploring Spatial Heterogeneity in Synergistic Effects of Compound Climate Hazards: Extreme Heat and Wildfire Smoke on Cardiorespiratory Hospitalizations in California. Sci. Adv.; 2024; 10, eadj7264. [DOI: https://dx.doi.org/10.1126/sciadv.adj7264] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38306434]

78. Chen, Z.; Yang, Y.; Zhang, D.; Guo, J.; Guo, Y.; Hu, X.; Chen, Y.; Bian, J. Predicting the Risk of Alzheimer’s Disease and Related Dementia in Patients with Mild Cognitive Impairment Using a Semi-Competing Risk Approach. Informatics; 2023; 10, 46. [DOI: https://dx.doi.org/10.3390/informatics10020046]