https://orcid.org/0000-0002-7287-970X
Figures
Abstract
Objectives
To assess the association between substance use and crash injury severity among older adults and how the relationship differs by rurality/urbanicity.
Methods
We pooled 2017–2021 cross-sectional data from the United States National Emergency Medical Service (EMS) Information System. We measured injury severity (low acuity, emergent, critical, and fatal) predicted by substance use, defined as self-reported or officer-reported alcohol and/or drug use. We controlled for age, sex, race/ethnicity, road user type, anatomical injured region, roadway crash, rurality/urbanicity, time of the day, and EMS response time. We performed a partial proportional ordinal logistic regression and reported the odds of worse injury outcomes (emergent, critical, and fatal injuries) compared to low acuity injuries, and the predicted probabilities by rurality/urbanicity.
Results
Our sample consisted of 252,790 older adults (65 years and older) road users. Approximately 67%, 25%, 6%, and 1% sustained low acuity, emergent, critical, and fatal injuries, respectively. Substance use was reported in approximately 3% of the population, and this proportion did not significantly differ by rurality/urbanicity. After controlling for patient, crash, and injury characteristics, substance use was associated with 36% increased odds of worse injury severity. Compared to urban areas, the predicted probabilities of emergent, critical, and fatal injuries were higher in rural and suburban areas.
Citation: Adeyemi O, Bukur M, Berry C, DiMaggio C, Grudzen CR, Konda S, et al. (2023) Substance use and pre-hospital crash injury severity among U.S. older adults: A five-year national cross-sectional study. PLoS ONE 18(10): e0293138. https://doi.org/10.1371/journal.pone.0293138
Editor: Yanyong Guo, Southeast University, CHINA
Received: April 26, 2023; Accepted: October 6, 2023; Published: October 25, 2023
Copyright: © 2023 Adeyemi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared publicly because of data sharing agreement between NEMSIS and the states providing EMS data. Data are available from the NEMSIS upon request at https://nemsis.org/using-ems-data/request-research-data/
Funding: The authors received no specific funding for this work
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Every day in the United States (U.S.), approximately 700 older adults sustain crash injuries with varying degrees of severity [1]. As of 2020, there were over 44 million licensed older adult drivers in the U.S. ‐ a 68% increase compared to two decades ago (1). These older adult drivers are at increased crash risk due to low visual acuity, poor peripheral vision, presence of other eye diseases, hearing loss, decline in motor skills, and other environmental road conditions such as nighttime driving [2,3]. However, crash injuries involving older adults extend beyond being car occupants but include pedestrians, riders of bicycles and tricycles, and bus or truck occupants. While motor vehicular crashes account for about 55% of crash injuries among older adults [4,5], pedestrian crash rates (secondary to motor vehicle use) have also been on the rise, increasing from 40.7 to 45.0 per 100,000 population between 2009 and 2019 [6].
Further increasing the risk of crash involvement and injury among older adults is alcohol and drug use (collectively referred to as substance use) [7–10]. It is estimated that approximately 38,000 older adults receive opioid prescriptions every day, one out of every eight older adults takes alcohol daily, and a smaller proportion report daily use of marijuana and cocaine [11,12]. Across all age groups, substance use while driving is associated with 45% increased odds of adverse crash outcomes and increased risk of pre-hospital crash fatality [13,14]. Alcohol is associated with two to seven folds increased odds of crash involvement in a crash [15–17], and a 15-fold increased odds of severe injury [16]. Marijuana, opioids, narcotics, stimulants, and depressants are associated with two to six folds increased crash risks and odds of fatal crash injuries [18–23].
Although crash injuries are preventable, rapid provision of care can improve injury outcomes among older adults. The rural-urban disparity in Emergency Medical Service (EMS) response [24,25], may disproportionately predispose older adults with crash injuries to worse injury severity compared to older adults with similar injuries in urban areas [26]. Earlier studies have reported that, while fatal injuries occur more in rural areas, minor and serious injuries occur more in urban areas [14,27,28]. Additionally, substance use differs across rurality/urbanicity. While urban areas have a higher incidence of hallucinogens, cocaine, marijuana, and other illicit drug use, rural areas have a higher incidence of alcohol and opioid misuse [29].
It is unknown to what extent substance use is associated with injury severity among older adults. Additionally, it is not known how the relationship between substance use and crash injury severity among older adults differs across rural and urban areas. Identifying these regional differences may inform policies on safe driving, road infrastructural design, and targeted behavioral interventions for older adults. Assessing the risk of crash injury severity among older adults is important due to the increasing older US adult population [30,31], and older licensed drivers [1]. This study, therefore, aims to assess the relationship between substance use and crash injury severity among older adults and the rural-urban differences that further define this problem.
2. Methods
2.1. Study design
We conducted a cross-sectional analysis by pooling five years of data (2017 to 2021) from the National Emergency Medical Services (EMS) Information System (NEMSIS). The NEMSIS is the national database of all EMS cases across U.S. States and territories [32]. Between 2017 and 2021, the number of states and territories that reported their EMS statistics to NEMSIS and permitted its use for research increased from 35 to 53 and the number of 9-1-1 events captured in the NEMSIS data increased from 7,907,829 to 48,982,990 [33].
2.2. Inclusion and exclusion criteria
Between 2017 and 2021, 157,114,790 persons were managed following an EMS activation (Fig 1). We identified the older adult population (age 65 years and older) (n = 58,272,048). We further restricted the population to age 65 years and older road users that sustained motor vehicle crash injuries using the International Classification of Disease version 10 (ICD-10) codes V00 to V79 (n = 488,422). We excluded cases whose substance use status was coded as "not applicable" (n = 19,519; 4% of 488,422). Thereafter, we excluded cases whose injury status was not reported (n = 213,897; 45.6% of 468,903). These unreported cases represent patients who either canceled the 9-1-1 call, refused care, or were evaluated but no treatment or transport was required. Also, we performed a listwise deletion for cases whose missingness was less than one percent (n = 1,203; 0.5% of 255,006) and when the crash response time was greater than 60 minutes (n = 1,103; 0.4% of 255,006). We excluded cases whose EMS response time exceeded 60 minutes, consistent with an earlier study [26]. These outlier cases are typically associated with unique environmental conditions such as tornadoes [34–36]. The final analytic data, therefore, was a total of 252,790 older adult road users who sustained motor vehicle injuries.
2.3. Injury severity
Our outcome measure, injury severity status, is a four-point categorical variable: low acuity, emergent, critical, and fatal injury [37]. Patients with low acuity injuries have injuries with a low probability of worsening or developing serious complications in the absence of intervention. Patients with emergent injuries have the potential of worsening if intervention is not initiated quickly. Patients with critical injuries have life-threatening injuries at high mortality risk if intervention is not initiated immediately. Finally, those with fatal injuries either died at the crash scene or while in transit to the hospital. These categorizations were made by EMS providers at the crash scene using the Model of the Clinical Practice of Emergency Medicine [38,39].
2.4. Substance use
The main predictor variable is the presence or absence of substance use. We defined substance use using the variable eHistory.17, representing alcohol/drug use indicators [40]. The NEMSIS defines substance use in seven categories: 1) alcohol containers/paraphernalia at the scene, 2) drug paraphernalia at the scene, 3) patient admits to alcohol use, 4) patient admits to drug use, 5) positive level (of alcohol or drugs) known from law enforcement or hospital record, 6) smell of alcohol on breath, 7) none reported. We defined the presence of substance use at the time of the crash event as cases in the first to sixth categories while the absence of substance use at the time of the crash was defined with cases coded in the seventh category [40]. Hence, an individual who is positive for substance use either used alcohol, drugs, or both. In the NEMSIS documentation, an individual can be assigned multiple categories. We, therefore, defined cases of “alcohol use only” using categories 1, 3, 5, and 6 and “drug use only” using categories 2 and 4.
2.5. Patient and injury characteristics
We controlled for age, sex, race/ethnicity, road user type, anatomical injured region, rurality/urbanicity, roadway location of the crash, the time of the day of the crash, and the EMS response time. We selected these patient and injury characteristics a priori from the literature [14,24,26]. Age was measured as a three-level categorical variable (65–74 years, 75–84 years, and 85 years and older), consistent with earlier definitions of phases of aging [41–43]. Sex was measured as a binary variable. We defined race/ethnicity in four categories: non-Hispanic White, non-Hispanic Black, Hispanic, and other races. Road user type was measured in six categories using the ICD-10 codes: car occupants (V40 –V49), pedestrians (V00 –V09), two-wheel vehicle occupants (V10 –V29), three-wheel vehicle occupants (V30 –V39), occupants of buses (V70 –V79), and occupants of trucks and industrial vehicles (V50 –V69). We defined the injured anatomic region in five categories: injury to the head and neck, abdomen and genitals, chest and back, upper and lower limbs (extremities), and multiple body injuries. NEMSIS reports the geographical location as a four-point categorical variable: wilderness, rural, suburban, and urban, using the United States Department of Agriculture Urban influence codes [37]. We recoded this variable into three categories: rural/wilderness (hereafter referred to as rural), suburban, and urban.
We assessed whether whether the older adult road user had a roadway crash or not. Using the ICD-10 codes, we identified roadway crash as crash events that occurred on street, highway and other paved roadways (Y92.4). All other location, which includes places of residence, businesses, stores, recreational areas, schools, and places not otherwise specified, were classified as not a roadway crash. We defined the time of the crash injury using a proxy measure–the time the 9-1-1 call was initiated. The time of crash injury was defined in four categories: morning rush hour period, afternoon rush hour period, nighttime, and other hours. Using a recently published meta-analysis as a guide [44], the morning and afternoon rush hour periods were defined as crash injuries sustained between 6 to 9 am and 3 to 7 pm, respectively. Nighttime crashes were defined as crash injuries occurring between 12 midnight and 5 am. We defined the EMS response time as the duration from chute initiation at the base station to arrival at the crash scene. We measured the crash response time as a four-point categorical variable: less than nine minutes, nine to 17.59 minutes, 18 to 26.59 minutes, and 27 minutes or higher. The nine-minute benchmark is based on the guidelines of the National Fire Protection Agency and the Fire and EMS Department, which requires the EMS travel time to be less than 9 minutes [45,46].
2.6. Handling of missing data
We encountered missing values in the following variables: substance use (18.0%), race/ethnicity (30.5%), roadway crash (1.6%), and anatomical injured region (25.7%). We performed multiple imputations for missing data, using the multiple imputations with chained equation (MICE) after justifying that missingness was at random [47]. Additionally, NEMSIS had advised researchers not to assume that missingness in the NEMSIS data is “Not Missing at Random” [48], further stressing the need to perform some measures of missing data analysis whenever such missingness is encountered. The MICE model was strengthened using injury severity, age, sex, crash response time, road user type, and the time of the crash as predictors. We performed 100 iterations, generated 100 predicted values for all missing values, and assigned the final value using the mean of the predicted values, consistent with earlier literature on multiple imputations [49,50].
2.7. Analysis
We computed the frequency distribution of demographic, injury, crash, and substance use characteristics. We assessed differences across injury severity status and rurality/urbanicity using chi-square statistics. We performed unadjusted and adjusted partially proportional ordinal logistic regression [51] to assess the odds of worse injury outcomes (critical, emergent, and fatal injuries) and computed the predicted probabilities of substance use-associated injury severity. The decision to use a partially proportional ordinal logistic regression, as opposed to a proportionally ordinal regression was based on the violation of the parallel lines assumption evidenced by a significant Brant test [51]. Also, we performed the interaction analysis between substance use and rurality/urbanicity and we reported the predicted probabilities of each substance use-related injury severity category in rural, suburban, and urban areas. Data were analyzed using SAS 9.4 [52] and STATA version 17 [53].
2.8. Ethical concern
This research used the 2017 ‐ 2021 NEMSIS data, publicly available de-identified data [54]. Based on the guidance from the New York University Langone Health Institutional Review Board (IRB), secondary data analysis of de-identified data that is publicly available does not require IRB approval [55]. Hence, informed consent was not required for this study. Also, since the de-identified data is available publicly and obtainable upon request from NEMSIS, our secondary data analysis was adjudged as not human subject research [55]. Our study followed the STROBE guidelines for reporting cross-sectional studies (available as a supporting information file).
3. Results
A total of 252,790 older adults met our inclusion and exclusion criteria (Table 1). The majority of the population was between 65 and 74 years (62%), female (51%), non-Hispanic Whites (71%), and car occupants (76%). Thirty-six percent of the sample population sustained injuries to the chest and back. The crash injuries occurred mostly in urban areas (83%) and were mostly roadway crashes (80%), with 16% of the crash events occurring during the afternoon rush hour period. Approximately 75% of the older adults experienced an EMS response time of less than nine minutes. Substance use was identified in approximately 3% of the sample population with cases of only alcohol and only drug impairments being 2.8% and 0.4%, respectively. Furthermore, 67% of the sample population had low acuity injuries, 25%, and 6% sustained emergent and critical injuries and 1% died. Age, sex, race/ethnicity, road user type, the anatomical injured region, geographical location, roadway crash, time of the day, EMS response time, and measures of substance use were significantly associated with pre-hospital crash injury severity (p<0.001).
Between 2017 and 2021, the proportion of emergent injuries ranged from 24.8–27.2%, while the proportion of critical and fatal injuries ranged between 5.7% ‐ 8.4%, and 1.0 ‐ 1.2%, respectively (Fig 2). Among older adults with emergent injuries, the proportions in urban areas ranged between 23.9 ‐ 26.2%, while the proportions in suburban and rural areas ranged between 25.6–31.0%, and 31.2 ‐ 33.4%, respectively (p<0.001). Also, among older adults with critical injuries, the proportions ranged between 5.2 ‐ 7.9% in urban areas, and in suburban and rural areas, the proportions ranged between 8.6 ‐ 13.0% and 7.9 ‐ 9.3%, respectively (p<0.001). Furthermore, among older adults with fatal injuries, the proportions in urban areas ranged between 0.8 ‐ 1.1%, and in suburban and rural areas, the proportions ranged between 1.8 ‐ 2.2%, and 1.5 ‐ 2.1%, respectively (p<0.001). While there was a decline in emergent and critical injuries between 2017 and 2019, the proportions of these injuries gradually increased from 2019 to 2021 in rural, suburban, and urban areas (p<0.001).
There were significant differences in the age and sex of older adults who sustained crash injuries in rural, suburban, and urban areas (Table 2). The proportion of non-Hispanic Whites who sustained crash injuries in urban areas was 68% while the proportions in suburban and rural areas were 86% and 87%, respectively (p<0.001). While the proportion of pedestrians with crash injuries was highest in urban areas (urban– 10%, suburban– 8%, rural– 9%), the proportion of crashes among occupants of trucks and industrial vehicles was highest in rural areas (urban– 3%, suburban– 7%, rural– 8%) (p<0.001). The proportion of older adults who sustained multiple body injuries was 25% in urban areas, and in suburban and rural areas, the proportions were 28% and 29%, respectively (p<0.001). The proportion of roadway crashes was 81% in urban areas, and 76% and 75% in suburban and rural area, respectively (p<0.001). The proportion of older adults who experienced EMS response time of less than nine minutes was 77% in urban areas, and in suburban and rural areas, the proportions were 67% and 63%, respectively (p<0.001). Although the proportion of drug use in urban areas was significantly higher compared to suburban and rural areas (p<0.001), there were no differences in alcohol use in urban, suburban, and rural areas.
Compared to low acuity injury, the unadjusted odds of worse injury severity were higher with increasing age, among males, pedestrians, occupants of two and three-wheeled vehicles, and occupants of trucks and industrial vehicles (Table 3). Worse injury severity was also higher among those with injuries to the head and neck, those that were involved in nighttime driving, and with increasing EMS crash response time. Compared to injuries in urban areas, injuries that occurred in rural (OR: 1.65; 95% CI: 1.60 ‐ 1.69) and suburban areas (OR: 1.35; 95% CI: 1.31 ‐ 1.39) were associated with increased odds of worse injury severity. Substance use was associated with 1.7 (95% CI: 1.67 ‐ 1.83) times the odds of worse injury severity. After adjusting for potential confounders, substance use was associated with 1.36 times the adjusted odds of worse injury severity (95% CI: 1.30 ‐ 1.43). Alcohol use alone was associated with 1.33 (95% CI: 1.27 ‐ 1.40) times the adjusted odds of worse injury severity while drug use alone was associated with 1.15 (95% CI: 1.00–1.32) times the adjusted odds of worse injury severity.
The predicted probability of substance use-associated emergent injury was 32.0% (95% CI: 31.0 ‐ 33.0) and the predicted probability increased step-wisely from urban (31.2%; 95% CI: 30.1 ‐ 32.3) to suburban (33.6; 95% CI: 30.0 ‐ 37.3) and rural areas (36.9%; 95% CI: 33.4 ‐ 40.4) (p<0.001) (Fig 3). Also, the predicted probability of substance use-associated critical injury was 6.0% (95% CI: 5.6 ‐ 6.5). The predicted probability was lowest in urban areas (5.8%; 95% CI: 5.3 ‐ 6.2) and in suburban and rural areas, the values were 8.4% (95% CI: 6.7 ‐ 10.0) and 7.1% (95% CI: 5.7 ‐ 8.5), respectively (p<0.001). Furthermore, the predicted probability of substance use-associated fatal injury was 0.2% (95% CI: 0.1 ‐ 0.2). The predicted probability was 0.2% (95% CI: 0.1 ‐ 0.2) in urban areas, and in suburban and rural areas, the values were 0.5% (95% CI: 0.2 ‐ 0.7) and 0.4% (95% CI: 0.2 ‐ 0.6), respectively (p<0.001).
4. Discussion
We present one of the few studies that demonstrate an association between substance use and crash injury severity among older adult road users. Also, our study provides a statistical basis for the conceptual linkage between substance use and injury severity among older adult road users, and how this association vary by rurality/urbanicity. Additionally, the observation that the predicted probabilities of substance use-related severe injuries are higher in rural and suburban areas compared to urban areas despite no rural-urban differences in the proportions of substance use among older adult road users highlight the rural-urban disparity in crash outcomes. Furthermore, the uptrend pattern in the proportions of emergent, critical, and fatal injuries between 2019 and 2021 in rural, suburban, and urban areas requires urgent public health intervention.
Earlier studies have reported the harmful effect of alcohol and drug use among older adults some of which include increased risk of falls [56–58], cognitive impairment [59,60], increased risk of alcohol or drug dependence [61,62], and worsening health conditions [63–65]. Our report of the increased odds of worse injury severity from substance use adds to the extant literature on the harms associated with alcohol and/or drug use. The worse injury outcomes associated with substance use among older adult road users may be explained by the reduced efficiency in metabolizing alcohol and drugs, longer toxic exposure, and an attenuation of the injury response mechanism when older adults are exposed to acute and/or chronic use of alcohol or drugs [66]. Acute and chronic alcohol use is associated with orthostatic hypotension and hypertension [67–69], respectively, and drugs such as opioids and benzodiazepines are central nervous system depressants [70,71]. Alcohol and drugs may further impair age-related physiologic response to acute trauma, hence increasing injury morbidity among older adults.
Similar to earlier studies that reported no rural-urban differences in alcohol and drug consumption [72,73], we report that there are no rural-urban differences in substance use-related crash injuries among older adult road users. However, there are significant rural-urban differences in the injury severity among older adults with substance use-related crash injuries. While the predicted probabilities of low acuity injuries decreased from urban to suburban and rural areas, the predicted probabilities of emergent injuries increased step-wise from urban to suburban and rural areas. Additionally, suburban and rural areas had higher critical and fatal injury probabilities compared to urban areas. These observed variations may reflect the rural-urban differences in driving behavior and access to timely and appropriate emergency care. Earlier studies have reported increased speeding behavior among road users in rural areas [74] and non-use of seat belts among older adult drivers [75]. Rural areas also experience significantly prolonged response times [24,26], increased deaths at the crash scene [26], and increased proportions of crash fatalities [76–78]. Additionally, the impact of hospital closures in rural areas may further explain the rural-urban disparity in crash outcomes since hospital closures is directly linked with reduced availaibility of healthcare professionals and reduced access to healthcare [79–82].
Crash injuries among older adult road users are preventable and interventions aimed at reducing substance use among road users represent a strategy for reducing crash-related morbidity and mortality. Excluding crash events in 2017 and 2018, our report showed an increasing trend in emergent, critical and fatal injuries among older adult road users in rural, suburban, and urban areas. There is therefore a need for intensified and focused public health intervention in reducing older adult crash injury rates across the U.S. Preventing the interactions of several co-existing risk factors of crash occurrence is a strategy recommended by the Governors Highway Safety Administration [74]. For example, our study showed that night driving is associated with worse injury severity among older adult road users. However, intentionally increasing nighttime police presence and nighttime enforcement of speed limits on road sections and highways associated with high clusters of crash occurrence may reduce crash occurrence at night. Preventing the interaction effect of risk factors requires improved data collection and the creation of more sensitive spatial and non-spatial models. Additionally, there is a need to extend substance use educational intervention to non-conventional research settings where older adults commonly gather. Such locations include medical clinics, senior centers, retirement communities, places of worship, and parks and recreation centers. Achieving behavioral change in alcohol and illicit drug use among older adults requires identifying motivators for change, using positive messaging techniques, encouraging peer support, and exercising patience [83,84]. Furthermore, primary care providers should educate older adults on the harm of driving when given prescription drugs.
Our findings have a number of practical implications. First, there is a need for health education campaigns that target older adults and their communities. Increasing awareness about the potential risks associated with substance use, even in older age, can lead to more informed choices. Secondly, healthcare providers, especially those in primary care settings, can use this information to prioritize screening and early identification of older adults at risk of substance use-related crash injuries. Despite no significant difference in substance use among older adults living in rural, suburban, and urban areas, injury severity is worse among those living in rural and suburban areas. There is, therefore, a need for continued government funding for hospitals, healthcare facilities, and EMS systems in rural and suburban areas to allocate resources strategically to facilitate rapid response and care of crash injuries. Since older adults might have different social dynamics and support systems, community-based programs can play a significant role. These programs could include support groups, social activities, and initiatives that address both substance use and injury prevention. Furthermore, in rural and suburban areas, transportation and access to emergency medical services could be challenging. The federal, state, and local government as well as community leadership should regularly assess the community-level transportation needs and prioritize measures that will improve transportation options and safety for their older populations.
This study has its limitations. As a cross-sectional design, causal inferences cannot be made. We were unable to control for other risky driving behaviors such as the non-use of seatbelts, distracted driving, and speeding because these variables were not captured in the NEMSIS. We did not adjust for the level of certification of crash scene EMS staff since this information is not part of the publicly released data from the NEMSIS. Substance use, a heterogeneous term for alcohol, marijuana, narcotics, stimulants, depressants, benzodiazepines, and other illicit drugs is inconsistently screened across the U.S. Substance use screening differs across states with substantial under-reporting of drug use while driving [85], falsely lowering the effect size we report. The gold standard for a diagnosis of substance use remains a serological assessment [86]. However, with some cases of substance use identified by self-reported measures and the presence of paraphernalia of alcohol and drugs, the possibility of misclassification bias is likely. Misclassification of the outcome measure is less likely since injury severity classification was based on a complex matrix and applied by trained EMS staff. Despite these limitations, this study has several strengths, which include the generalizability of this study to older adults who sustain injuries across the U.S. Also, this is one of the few studies that assess the rural-urban association of substance use and crash injury severity among older adult road users. The awareness of the association between substance use and crash injury severity as well as the rural-urban differences will improve interventions aimed at reducing crash involvement of older adult road users.
5. Conclusion
Substance use is associated with worse crash injury severity among older adult road users. Despite no significant difference in rural-urban proportions of substance use among older adults, emergent injuries increase from urban to rural areas. With the increasing trends in critical and fatal crash injuries among older adults, there is an urgent need for increased community awareness of the risks associated with substance use among older adult road users. Also, reversing the uptrend pattern in substance use-related emergent, critical, and fatal crash injuries might require increased screening and identifying older adults who are at risk of substance use-related crash involvement, and increased healthcare resource allocation that will strengthen the emergency response systems, especially in rural and suburban areas.
Supporting information
S1 Checklist. STROBE statement—checklist of items that should be included in reports of observational studies.
https://doi.org/10.1371/journal.pone.0293138.s001
(DOCX)
Acknowledgments
The authors appreciate the National Highway Traffic Safety Administration’s Office of Emergency Medical Service and the Technical Assistance Center at the University of Utah for providing the data.
References
- 1.
Centers for Disease Control and Prevention. Older Adult Drivers 2021 [Available from: https://www.cdc.gov/transportationsafety/older_adult_drivers/index.html.
- 2. National Institute for Occupational Safety and Health. Older Drivers in the Workplace: How Employers and Workers Can Prevent Crashes: Centers for Disease Control and Prevention; 2016 [Available from: https://www.cdc.gov/niosh/docs/2016-116/pdfs/2016-116.pdf.
- 3. Federal Highway Administration. Nighttime Visibility: United States Department of Transportation; 2022 [Available from: https://safety.fhwa.dot.gov/roadway_dept/night_visib/general-information.cfm.
- 4.
Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Overall All Transport Nonfatal Emergency Department Visits and Rates per 100,000: 2019, United States, All Races, Both Sexes, Ages 65 to 85+, Disposition: All Cases Atlanta, GA: CDC; 2022 [Available from: https://wisqars.cdc.gov/nonfatal-reports.
- 5.
Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Overall MV-Occupant Nonfatal Emergency Department Visits and Rates per 100,000: 2019, United States, All Races, Both Sexes, Ages 65 to 85+, Disposition: All Cases Atlanta, GA: CDC; 2022 [Available from: https://wisqars.cdc.gov/nonfatal-reports.
- 6.
Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Overall Pedestrian Nonfatal Emergency Department Visits and Rates per 100,000, 2009 ‐ 2020, United States: All Races, Both Sexes, Ages 65 to 85+, Disposition: All Cases Atlanta, GA: CDC; 2022 [Available from: https://wisqars.cdc.gov/nonfatal-reports.
- 7. Alcañiz M, Santolino M, Ramon L. Drinking patterns and drunk-driving behaviour in Catalonia, Spain: A comparative study. Transportation Research Part F: Traffic Psychology and Behaviour. 2016;42:522–31.
- 8. Bondallaz P, Favrat B, Chtioui H, Fornari E, Maeder P, Giroud C. Cannabis and its effects on driving skills. Forensic Science International. 2016;268:92–102. pmid:27701009
- 9. Kumar S, Bansal YS, Singh D, Medhi B. Alcohol and Drug Use in Injured Drivers ‐ An Emergency Room Study in a Regional Tertiary Care Centre of North West India. Journal of Clinical and Diagnostic Research. 2015;9(7):Hc01-4.
- 10. Freeman DG. Drunk driving legislation and traffic fatalities: New evidence on BAC 08 laws. 2007;25(3):293–308.
- 11. Hoots BE, Xu L, Kariisa M, Wilson NO, Rudd RA, Scholl L, et al. 2018 Annual surveillance report of drug-related risks and outcomes—United States2018 04/20/2023. Available from: https://www.cdc.gov/drugoverdose/pdf/pubs/2018-cdc-drug-surveillance-report.pdf.
- 12. Mattson M, Lipari RN, Hays C, Van Horn SL. A day in the life of older adults: Substance use facts. The CBHSQ report [Internet]. 2017 04/15/2022. Available from: https://www.ncbi.nlm.nih.gov/books/NBK436750/pdf/Bookshelf_NBK436750.pdf.
- 13. DiMaggio CJ, Avraham JB, Frangos SG, Keyes K. The role of alcohol and other drugs on emergency department traumatic injury mortality in the United States. Drug and Alcohol Dependence. 2021;225:108763. pmid:34049099
- 14. Adeyemi OJ, Paul R, DiMaggio CJ, Delmelle EM, Arif AA. An assessment of the non-fatal crash risks associated with substance use during rush and non-rush hour periods in the United States. Drug and Alcohol Dependence. 2022:109386. pmid:35306398
- 15. Asefa NG, Ingale L, Shumey A, Yang H. Prevalence and factors associated with road traffic crash among taxi drivers in Mekelle town, northern Ethiopia, 2014: a cross sectional study. PLoS One. 2015;10(3):e0118675. pmid:25781940
- 16. Compton RP, Berning A. Drug and Alcohol Crash Risk. Traffic Safety Facts: Research Note [Internet]. 2015. Available from: http://www.nhtsa.gov/staticfiles/nti/pdf/812117-Drug_and_Alcohol_Crash_Risk.pdf.
- 17. Penmetsa P, Pulugurtha SS. Risk drivers pose to themselves and other drivers by violating traffic rules. Traffic Injury Prevention. 2017;18(1):63–9. pmid:27257740
- 18. Blows S, Ivers RQ, Connor J, Ameratunga S, Woodward M, Norton R. Marijuana use and car crash injury. Addiction. 2005;100(5):605–11. pmid:15847617
- 19. Preuss UW, Huestis MA, Schneider M, Hermann D, Lutz B, Hasan A, et al. Cannabis Use and Car Crashes: A Review. Frontiers in psychiatry. 2021;12. pmid:34122176
- 20. Brubacher JR, Chan H, Erdelyi S, Macdonald S, Asbridge M, Mann RE, et al. Cannabis use as a risk factor for causing motor vehicle crashes: a prospective study. Addiction. 2019;114(9):1616–26. pmid:31106494
- 21. Drummer OH, Gerostamoulos D, Di Rago M, Woodford NW, Morris C, Frederiksen T, et al. Odds of culpability associated with use of impairing drugs in injured drivers in Victoria, Australia. Accid Anal Prev. 2020;135:105389. pmid:31812899
- 22. Drummer OH, Gerostamoulos J, Batziris H, Chu M, Caplehorn J, Robertson MD, et al. The involvement of drugs in drivers of motor vehicles killed in Australian road traffic crashes. Accident Analysis & Prevention. 2004;36(2):239–48. pmid:14642878
- 23. Laumon B, Gadegbeku B, Martin JL, Biecheler MB. Cannabis intoxication and fatal road crashes in France: population based case-control study. Bmj. 2005;331(7529):1371. pmid:16321993
- 24. Byrne JP, Mann NC, Dai M, Mason SA, Karanicolas P, Rizoli S, et al. Association Between Emergency Medical Service Response Time and Motor Vehicle Crash Mortality in the United States. JAMA surgery. 2019;154(4):286–93. pmid:30725080
- 25. Adeyemi OJ, Paul R, Arif A. An assessment of the rural-urban differences in the crash response time and county-level crash fatalities in the United States. The Journal of Rural Health. 2021. pmid:34664745
- 26. Adeyemi OJ, Paul R, DiMaggio C, Delmelle E, Arif A. The association of crash response times and deaths at the crash scene: A cross-sectional analysis using the 2019 National Emergency Medical Service Information System. The Journal of Rural Health. 2022. pmid:35452139
- 27. Cabrera-Arnau C, Prieto Curiel R, Bishop SR. Uncovering the behaviour of road accidents in urban areas. Royal Society open science. 2020;7(4):191739. pmid:32431872
- 28.
Insurance Institute for Highway Safety. Fatality Facts 2019: Urban/rural comparison: Insurance Institute for Highway Safety, Highway Loss Data Institute; 2022 [Available from: https://www.iihs.org/topics/fatality-statistics/detail/urban-rural-comparison.
- 29. Rural Health Information Hub. Substance Use and Misuse in Rural Areas 2020 [Available from: https://www.ruralhealthinfo.org/topics/substance-use.
- 30.
Vespa J. The U.S. Joins Other Countries With Large Aging Populations: United States Census Bureau; 2021 [Available from: https://www.census.gov/library/stories/2018/03/graying-america.html.
- 31. Pallin DJ, Espinola JA, Camargo CA, Jr. US population aging and demand for inpatient services. J Hosp Med. 2014;9(3):193–6. pmid:24464735
- 32.
National Emergency Medical Services Information System. How NEMSIS Works 2019 [Available from: https://nemsis.org/what-is-nemsis/how-nemsis-works/.
- 33.
NEMSIS. Research Data Resources 2023 [Available from: https://nemsis.org/using-ems-data/request-research-data/research-data-resources/.
- 34. Schreiner B, Salter J. Kentucky hardest hit as storms leave dozens dead in 5 states. Chattanooga Times Free Press. 2021 December 11, 2021.
- 35.
Childs JW. Newnan, Georgia, Tornado: ’Our Hearts Are Broken’. The Weather Channel. 2021.
- 36.
National Weather Service. Tornadoes of March 3, 2019: Event Summary for Central Alabama 2019 [Available from: https://www.weather.gov/bmx/event_03032019.
- 37.
Emergency Medical Services. NEMSIS Data Dictionary2020 03/22/2021; version 3.4.0. Available from: https://nemsis.org/media/nemsis_v3/release-3.4.0/DataDictionary/PDFHTML/DEMEMS/index.html.
- 38. Beeson MS, Ankel F, Bhat R, Broder JS, Dimeo SP, Gorgas DL, et al. The 2019 Model of the Clinical Practice of Emergency Medicine. Journal of Emergency Medicine. 2020;59(1):96–120. pmid:32475725
- 39. Counselman FL, Babu K, Edens MA, Gorgas DL, Hobgood C, Marco CA, et al. The 2016 Model of the Clinical Practice of Emergency Medicine. The Journal of emergency medicine. 2017;52(6):846–9. pmid:28351510
- 40.
National Emergency Medical Services Information System. eHistory.17 - Alcohol/Drug Use Indicators: National Emergency Medical Services Information System; 2021 [Available from: https://nemsis.org/media/nemsis_v3/release-3.4.0/DataDictionary/PDFHTML/DEMEMS/sections/elements/eHistory.17.xml.
- 41. Lee TM, Vargas A, Dua S, Dafer RM. Cerebral Infarctions Following Palliative Transarterial Chemoembolization with Embozene of a Vertebral Body Metastatic Tumor. J Stroke Cerebrovasc Dis. 2017;26(12):e224–e5. pmid:28870434
- 42.
Little W, McGivern R. Aging and the Elderly. 2014 [cited 08/22/2023]. In: Introduction to Sociology ‐ 1st Canadian Edition [Internet]. Canada: BC Campus, [cited 08/22/2023]. Available from: https://opentextbc.ca/introductiontosociology/chapter/chapter13-aging-and-the-elderly/.
- 43.
Lally M, Valentine-French S. Age Categories in Late Adulthood. 2017 [cited 08/22/2023]. In: Lifespan Development–A Psychological Perspective [Internet]. [cited 08/22/2023]. Available from: https://opentextbooks.concordia.ca/lifespandevelopment/chapter/9-3-age-categories-in-late-adulthood/.
- 44. Adeyemi OJ, Arif AA, Paul R. Exploring the relationship of rush hour period and fatal and non-fatal crash injuries in the U.S.: A systematic review and meta-analysis. Accid Anal Prev. 2021;163:106462. pmid:34717204
- 45.
National Fire Protection Agency. Standard for the Organization and Deployment of FIre Suppression Operations, Emergency Medical Operations, and Special Operations to the Public by Career Fire Departments. NFPA 1710 [Internet]. 2020. Available from: https://www.nfpa.org/codes-and-standards/all-codes-and-standards/list-of-codes-and-standards/detail?code=1710.
- 46.
Fire and EMS Department. EMS Response Time 2020 [Available from: https://fems.dc.gov/page/ems-response-time.
- 47. Lee KJ, Carlin JB. Multiple imputation for missing data: fully conditional specification versus multivariate normal imputation. Am J Epidemiol. 2010;171(5):624–32. pmid:20106935
- 48.
NEMSIS Technical Assistance Center. National EMS Database NEMSIS Public Release Research Data Set: 2020 User Manual2021 05/11/2022; (v3.4.0). Available from: https://nemsis.org/wp-content/uploads/2021/05/2020-NEMSIS-RDS-340-User-Manual_v3-FINAL.pdf.
- 49. Dray S, Josse J. Principal component analysis with missing values: a comparative survey of methods. Plant Ecology. 2015;216(5):657–67.
- 50. McNeish D. Exploratory Factor Analysis With Small Samples and Missing Data. J Pers Assess. 2017;99(6):637–52. pmid:27929657
- 51. Williams R. Gologit2: A program for generalized logistic regression/partial proportional odds models for ordinal variables. STATA Journal. 2005;12:2005.
- 52.
SAS Institute Inc. SAS 9.4. 9.4 ed. Cary, NC: SAS Institute Inc; 2019.
- 53.
StataCorp. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC; 2020.
- 54.
National EMS Information System Technical Assistance Center. 2020 Public Release Research Dataset Frequently Asked Questions:2021 04/23/2023. Available from: https://nemsis.org/wp-content/uploads/2021/10/Research-dataset-FAQ-v2.pdf.
- 55.
NYU Langone Health Institutional Review Board. Human Research Protections Policies and Procedures2023 04/23/2023. Available from: https://med.nyu.edu/research/office-science-research/clinical-research/sites/default/files/nyu-som-irb-policies-and-procedures-for-human-subjects-research-protection.docx.
- 56. de Jong MR, Van der Elst M, Hartholt KA. Drug-related falls in older patients: implicated drugs, consequences, and possible prevention strategies. Ther Adv Drug Saf. 2013;4(4):147–54. pmid:25114778
- 57. Shakya I, Bergen G, Haddad YK, Kakara R, Moreland BL. Fall-related emergency department visits involving alcohol among older adults. Journal of safety research. 2020;74:125–31. pmid:32951773
- 58. Sun Y, Zhang B, Yao Q, Ma Y, Lin Y, Xu M, et al. Association between usual alcohol consumption and risk of falls in middle-aged and older Chinese adults. BMC geriatrics. 2022;22(1):750.
- 59. Moore AR, O’Keeffe ST. Drug-induced cognitive impairment in the elderly. Drugs Aging. 1999;15(1):15–28. pmid:10459729
- 60. Zhang R, Shen L, Miles T, Shen Y, Cordero J, Qi Y, et al. Association of Low to Moderate Alcohol Drinking With Cognitive Functions From Middle to Older Age Among US Adults. JAMA network open. 2020;3(6):e207922-e. pmid:32597992
- 61. Hendriks HFJ. Alcohol and Human Health: What Is the Evidence? Annu Rev Food Sci Technol. 2020;11:1–21. pmid:32209032
- 62. Kranzler HR, Soyka M. Diagnosis and Pharmacotherapy of Alcohol Use Disorder: A Review. Jama. 2018;320(8):815–24. pmid:30167705
- 63. Rehm J. The risks associated with alcohol use and alcoholism. Alcohol Res Health. 2011;34(2):135–43. pmid:22330211
- 64. Sterling SA, Palzes VA, Lu Y, Kline-Simon AH, Parthasarathy S, Ross T, et al. Associations Between Medical Conditions and Alcohol Consumption Levels in an Adult Primary Care Population. JAMA network open. 2020;3(5):e204687. pmid:32401315
- 65. Varghese J, Dakhode S. Effects of Alcohol Consumption on Various Systems of the Human Body: A Systematic Review. Cureus. 2022;14(10):e30057.
- 66. Moore AA, Whiteman EJ, Ward KT. Risks of combined alcohol/medication use in older adults. The American Journal of Geriatric Pharmacotherapy. 2007;5(1):64–74. pmid:17608249
- 67. Narkiewicz K, Cooley RL, Somers VK. Alcohol Potentiates Orthostatic Hypotension. Circulation. 2000;101(4):398–402.
- 68. Husain K, Ansari RA, Ferder L. Alcohol-induced hypertension: Mechanism and prevention. World J Cardiol. 2014;6(5):245–52. pmid:24891935
- 69. Klatsky AL, Gunderson E. Alcohol and hypertension: a review. J Am Soc Hypertens. 2008;2(5):307–17. pmid:20409912
- 70. Smink BE, Egberts ACG, Lusthof KJ, Uges DRA, de Gier JJ. The relationship between benzodiazepine use and traffic accidents: A systematic literature review. CNS Drugs. 2010;24(8):639–53. pmid:20658797
- 71. Li G, Chihuri S. Prescription opioids, alcohol and fatal motor vehicle crashes: a population-based case-control study. Injury epidemiology. 2019;6:11-. pmid:31245260
- 72. Dixon MA, Chartier KG. Alcohol Use Patterns Among Urban and Rural Residents: Demographic and Social Influences. Alcohol Res. 2016;38(1):69–77. pmid:27159813
- 73. Derefinko KJ, Bursac Z, Mejia MG, Milich R, Lynam DR. Rural and urban substance use differences: Effects of the transition to college. The American journal of drug and alcohol abuse. 2018;44(2):224–34. pmid:28726520
- 74.
Governors Highway Safety Association. America’s rural roads: Beautiful and deadly2022 03/15/2023. Available from: https://www.ghsa.org/sites/default/files/2022-09/America%E2%80%99s%20Rural%20Roads%20-%20Beautiful%20and%20Deadly%20FNL.pdf.
- 75. Adeyemi O, Paul R, Arif A. Spatial Cluster Analysis of Fatal Road Accidents From Non-Use of Seat Belts Among Older Drivers. Innov Aging. 2020;4(Supplement_1):113–4.
- 76. Adeyemi OJ, Paul R, DiMaggio C, Delmelle E, Arif A. Rush Hour-Related Road Crashes: Assessing the Social and Environmental Determinants of Fatal and Non-Fatal Road Crash Events [Ph.D.]. Ann Arbor: The University of North Carolina at Charlotte; 2021.
- 77.
Insurance Institute for Highway Safety. Fatality Facts 2018: Urban/rural comparison: Insurance Institute for Highway Safety/Highway Loss Data Institute; 2019 [Available from: https://www.iihs.org/topics/fatality-statistics/detail/urban-rural-comparison.
- 78. National Center for Statistics and Analysis. Rural/Urban Comparison of Traffic Fatalities. Traffic Safety Fact: 2017 Data [Internet]. 2019 06/12/2022. Available from: https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812741.
- 79. Bailey V. Rural Hospitals Facing Risk of Closure Due to Financial Stress. Revcycle Intelligence. 2022.
- 80. Germack HD, Kandrack R, Martsolf GR. When Rural Hospitals Close, The Physician Workforce Goes. Health affairs (Project Hope). 2019;38(12):2086–94. pmid:31794309
- 81. Kaufman BG, Thomas SR, Randolph RK, Perry JR, Thompson KW, Holmes GM, et al. The Rising Rate of Rural Hospital Closures. The Journal of rural health. 2016;32(1):35–43. pmid:26171848
- 82. Miller KEM, James HJ, Holmes GM, Van Houtven CH. The effect of rural hospital closures on emergency medical service response and transport times. Health services research. 2020;55(2):288–300. pmid:31989591
- 83. Prochaska JO, Redding CA, Evers KE. The transtheoretical model and stages of change. Health Behavior: Theory, Research, and Practice. 2015;97.
- 84. Watakakosol R, Suttiwan P, Ngamake ST, Raveepatarakul J, Wiwattanapantuwong J, Iamsupasit S, et al. Integration of the Theory of Planned Behavior and Transtheoretical Model of Change for Prediction of Intentions to Reduce or Stop Alcohol Use among Thai Adolescents. Subst Use Misuse. 2021;56(1):72–80. pmid:33106107
- 85. Berning A, Smither DD. Understanding the limitations of drug test information, reporting, and testing practices in fatal crashes: traffic safety facts: research note. United States. Department of Transportation. National Highway Traffic Safety …; 2014.
- 86.
DiMaggio C, Wheeler-Martin K, Oliver J. Alcohol-Impaired Driving in the United States: Review of Data Sources and Analyses. 2018 08/06/2021. In: Getting to Zero Alcohol-Impaired Driving Fatalities: A Comprehensive Approach to a Persistent Problem [Internet]. The National Academies Press. Available from: https://www.ncbi.nlm.nih.gov/books/NBK500064/.