Driving Under the Influence of Alcohol and Drugs in New York: Prevalence, Trends, and Population Patterns
Introduction
Driving under the influence of alcohol and drugs remains an important traffic safety concern in New York. This research note examines self-reported driving under the influence of alcohol or select drugs (DUI) among New York residents aged 16 or older, with a particular focus on driving under the influence of alcohol (DUIA) and cannabis (DUIC). The analyses address three key questions: How common is self-reported impaired driving in New York? Which population groups have higher prevalence? And among alcohol and cannabis users, how does impaired driving vary by recency of substance use and substance use disorder status?
The research note uses data from the National Survey on Drug Use and Health (NSDUH) generated through the Substance Abuse and Mental Health Services Administration (SAMHSA) Data Analysis System (DAS). NSDUH provides representative data on alcohol and other substance use, including illicit drug use and substance use disorders, for each of the 50 states. Sensitive questions are administered in a manner designed to protect respondents’ privacy and encourage accurate reporting. The analyses use two-year annual average estimates for 2021–2022, 2022-2023 and 2023–2024. All percentages and population counts are weighted survey estimates and are subject to sampling error. Estimates presented in this report met SAMHSA’s criteria for statistical precision. Statistical testing was used to assess differences between population groups and changes over time.
In 2023–2024, 5.5% of New York residents aged 16 or older reported DUI during the past year, representing an estimated 864,000 individuals. DUIC prevalence (3.2%) was comparable to DUIA prevalence (3.0%), indicating that self-reported cannabis-impaired driving was as common as alcohol-impaired driving. Overall DUI and DUIA prevalence declined significantly from 2021–2022, while DUIC prevalence did not change significantly. The analyses also identified high DUIC prevalence among past-month cannabis users (17.9%) and among past-year cannabis users with a cannabis use disorder (28.1%).
As with all self-reported data, the estimates are subject to recall error and social desirability bias. In addition, respondents may differ in how they interpret whether they were “under the influence” when driving.
DUI (Alcohol or Select Drugs)
Prevalence by Substance Type
Among New York residents aged 16 or older, 5.5% reported driving under the influence of alcohol or select drugs (DUI) during the past year, representing an estimated 864,000 individuals based on the 2023–2024 annual average (Figure 1; Table 1).
DUIC prevalence (3.2%; 518,000 individuals) was comparable to DUIA prevalence (3.0%; 482,000 individuals), indicating that self-reported cannabis-impaired driving was as common as alcohol-impaired driving. In contrast, the prevalence of driving under the influence of select drugs other than cannabis (DUID) was substantially lower at 0.7% (113,000 individuals).
Compared with the 2021–2022 annual average, significant declines in overall DUI and DUIA prevalence corresponded to approximately 243,000 fewer residents reporting DUI and 200,000 fewer reporting DUIA in 2023–2024, respectively. DUIC and DUID prevalence did not change significantly during the same period (Figure 1; Table 1).
| Substance Type |
2021–2022
|
2023–2024
|
||
|---|---|---|---|---|
| Prevalence | Estimated Number | Prevalence | Estimated Number | |
| Alcohol or Select Drugs | 7.1% | 1,107,000 | 5.5% | 864,000 |
| Alcohol | 4.3% | 682,000 | 3.0% | 482,000 |
| Cannabis | 3.7% | 592,000 | 3.2% | 518,000 |
| Select Drugs Other Than Cannabis | 0.5% | 76,000 | 0.7% | 113,000 |
| Note: Select drugs include cannabis, cocaine (including crack), heroin, hallucinogens, inhalants, or methamphetamine. | ||||
DUI in the New York Population
Prevalence by Age
According to the 2023–2024 annual average, self-reported past-year DUI prevalence was substantially higher among residents aged 21–49 (7.2%–8.9%) than among those aged 16–20 (2.4%) or 50 or older (3.5%) (Figure 2; Table 2).
Compared with the 2021–2022 annual average, DUI prevalence in 2023–2024 did not change significantly for any age group (Figure 2).
| Age Group | Prevalence | Estimated Number of Individuals |
|---|---|---|
| 16–20 | 2.4% | 27,000 |
| 21–25 | 7.9% | 102,000 |
| 26–34 | 8.9% | 217,000 |
| 35–49 | 7.2% | 261,000 |
| 50 or Older | 3.5% | 257,000 |
Prevalence by Sex
According to the 2023–2024 annual average, self-reported past-year DUI prevalence was more than twice as high among men as among women (8.1% vs. 3.0%), representing an estimated 615,000 men and 249,000 women (Figure 3; Table 3).
Compared with the 2021–2022 annual average, DUI prevalence declined significantly among women (5.2% to 3.0%), whereas prevalence among men did not change significantly (Figure 3).
| Sex | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Male | 8.1% | 615,000 |
| Female | 3.0% | 249,000 |
Prevalence by Race/Ethnicity
According to the 2023–2024 annual average, past-year DUI prevalence was highest among non-Hispanic White (6.9%) and non-Hispanic Black (6.2%) residents (Figure 4; Table 4). Because non-Hispanic White residents constitute the largest racial/ethnic group in New York, this prevalence translates to an estimated 605,000 individuals reporting DUI during the past year, compared with approximately 138,000 non-Hispanic Black residents, 74,000 Hispanic residents, and 47,000 non-Hispanic residents of other or multiple races (Table 4).
Looking at trends over time, non-Hispanic White residents were the only racial/ethnic group to experience a statistically significant decline in DUI prevalence, decreasing from 9.1% in 2021–2022 to 6.9% in 2023–2024 (Figure 4). In contrast, DUI prevalence among all other racial and ethnic groups did not change significantly over the same period.
| Race/Ethnicity | Prevalence | Estimated Number of Individuals |
|---|---|---|
| White, Not Hispanic | 6.9% | 605,000 |
| Black, Not Hispanic | 6.2% | 138,000 |
| Other or Multiple, Not Hispanic | 2.5% | 47,000 |
| Hispanic | 2.6% | 74,000 |
Prevalence by County Type1
According to the 2023–2024 annual average, past-year DUI prevalence was more than twice as high among residents of small metro or nonmetro counties (9.8%) as among residents of large metro counties (4.4%) (Figure 5; Table 5). Despite the lower prevalence in large metro counties, an estimated 548,000 residents reported past-year DUI, compared with 316,000 residents of small metro or nonmetro counties (Table 5).
Compared with the 2021–2022 annual average, DUI prevalence in 2023–2024 did not change significantly for either county type (Figure 5).
| County Type | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Large Metro | 4.4% | 548,000 |
| Small or Non Metro | 9.8% | 316,000 |
DUI Alcohol
DUIA in the New York Population
Prevalence by Age
According to the 2023–2024 annual average, self-reported past-year DUIA prevalence was significantly higher among residents aged 26–49 (4.1%–4.9%) than among those aged 16–20 (0.5%) and 50 or older (2.3%) (Figure 6; Table 6).
Compared with the 2021–2022 annual average, the most robust reduction occurred among residents aged 50 or older, with DUIA prevalence declining from 4.5% to 2.3% (Figure 6).
| Age Group | Prevalence | Estimated Number of Individuals |
|---|---|---|
| 16–20 | 0.5% | 6,000 |
| 21–25 | 2.9% | 38,000 |
| 26–34 | 4.9% | 120,000 |
| 35–49 | 4.1% | 150,000 |
| 50 or Older | 2.3% | 167,000 |
Prevalence by Sex
According to the 2023–2024 annual average, self-reported past-year DUIA prevalence was more than twice as high among men as among women (4.5% vs. 1.7%), representing an estimated 344,000 men and 138,000 women (Figure 7; Table 7).
Compared with the 2021–2022 annual average, DUIA prevalence declined significantly among women (3.0% to 1.7%), whereas prevalence among men did not change significantly (Figure 7).
| Sex | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Male | 4.5% | 344,000 |
| Female | 1.7% | 138,000 |
Prevalence by Race/Ethnicity
According to the 2023–2024 annual average, past-year DUIA prevalence was highest among non-Hispanic White residents (4.2%) (Figure 8; Table 8). Non-Hispanic White residents also accounted for the largest estimated number reporting DUIA (374,000), reflecting both their higher prevalence and larger population in New York (Table 8).
Compared with the 2021–2022 annual average, DUIA prevalence in 2023–2024 did not change significantly for any racial/ethnic group (Figure 8).
| Race/Ethnicity | Prevalence | Estimated Number of Individuals |
|---|---|---|
| White, Not Hispanic | 4.2% | 374,000 |
| Black, Not Hispanic | 2.2% | 50,000 |
| Other or Multiple, Not Hispanic | 1.0% | 19,000 |
| Hispanic | 1.4% | 40,000 |
Prevalence by County Type2
According to the 2023–2024 annual average, past-year DUIA prevalence was more than twice as high among residents of small metro or nonmetro counties (6.0%) as among residents of large metro counties (2.3%) (Figure 9; Table 9). Despite the lower prevalence in large metro counties, an estimated 288,000 residents reported DUIA, compared with 195,000 residents of small metro or nonmetro counties (Table 9).
Compared with the 2021–2022 annual average, DUIA prevalence among residents of large metro counties declined significantly from 3.5% to 2.3% in 2023–2024 (Figure 9).
| County Type | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Large Metro | 2.3% | 288,000 |
| Small or Non Metro | 6.0% | 195,000 |
DUIA Among Past-Year Alcohol Users
The preceding analyses describe how DUIA prevalence varies across demographic groups in the New York population. To better identify populations that may benefit from targeted prevention efforts, the following analyses focus specifically on past-year alcohol users and examine DUIA prevalence by recency of alcohol use and alcohol use disorder status.
Prevalence by Alcohol Use Recency
According to the 2023–2024 annual average, approximately 97% of New York residents who reported DUIA during the past year were past-month alcohol users, representing an estimated 468,000 individuals (Table 10). Among past-month alcohol users, 6.0% reported DUIA, compared with 0.6% of those who used alcohol within the past year but not the past month (Figure 10). These findings suggest that past-month alcohol users are an important audience for DUIA prevention and education efforts, although most did not report DUIA.
Compared with the 2021–2022 annual average, DUIA prevalence among past-month alcohol users declined significantly from 8.0% to 6.0% in 2023–2024 (Figure 10).
| Alcohol Use Recency | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Used Within the Past Year, Not the Past Month | 0.6% | 14,000 |
| Used Within the Past Month | 6.0% | 468,000 |
Prevalence by Alcohol Use Disorder Status
According to the 2023–2024 annual average, four in ten New York residents who reported DUIA during the past year had a past-year alcohol use disorder (AUD)3, representing an estimated 191,000 individuals (Table 11).
Among past-year alcohol users, DUIA prevalence was more than three times higher among those with a past-year AUD (11.8%) than among those without AUD (3.4%). However, approximately six in ten residents who reported DUIA did not have a past-year AUD, suggesting that prevention efforts should not be limited to those with AUD.
Compared with the 2021–2022 annual average, DUIA prevalence among individuals with a past-year AUD declined significantly from 18.9% to 11.8% in 2023–2024, while prevalence among those without AUD did not change significantly (Figure 11).
| Alcohol Use Disorder Status | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Past Year Use Disorder | 11.8% | 191,000 |
| No Past Year Use Disorder | 3.4% | 291,000 |
DUI Cannabis
DUIC in the New York Population
Prevalence by Age
According to the 2023–2024 annual average, self-reported past-year DUIC prevalence was significantly higher among residents aged 21–34 (5.9%–6.2%) than among those aged 16–20 (2.3%) and 50 or older (1.7%) (Figure 12; Table 12).
DUIC prevalence was more than four times higher than DUIA prevalence among residents aged 16–20 (2.3% vs. 0.5%) and more than twice as high among those aged 21–25 (6.2% vs. 2.9%).
Compared with the 2021–2022 annual average, DUIC prevalence did not change significantly for any age group (Figure 12).
| Age Group | Prevalence | Estimated Number of Individuals |
|---|---|---|
| 16–20 | 2.3% | 26,000 |
| 21–25 | 6.2% | 81,000 |
| 26–34 | 5.9% | 145,000 |
| 35–49 | 3.9% | 142,000 |
| 50 or Older | 1.7% | 125,000 |
Prevalence by Sex
According to the 2023–2024 annual average, self-reported past-year DUIC prevalence was more than twice as high among men as among women (4.7% vs. 1.9%), representing an estimated 364,000 men and 155,000 women (Figure 13; Table 13).
Compared with the 2021–2022 annual average, DUIC prevalence did not change significantly for men or women (Figure 13).
| Sex | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Male | 4.7% | 364,000 |
| Female | 1.9% | 155,000 |
Prevalence by Race/Ethnicity
According to the 2023–2024 annual average, past-year DUIC prevalence was 5.1% among non-Hispanic Black residents and 3.9% among non-Hispanic White residents, with both groups having significantly higher prevalence than Hispanic residents (1.1%) and non-Hispanic residents of other or multiple races (1.5%) (Figure 14; Table 14). Because non-Hispanic White residents comprise the largest racial/ethnic group in New York, they accounted for the largest estimated number reporting DUIC (342,000).
Compared with the 2021–2022 annual average, DUIC prevalence in 2023–2024 did not change significantly for any racial/ethnic group (Figure 14).
| Race/Ethnicity | Prevalence | Estimated Number of Individuals |
|---|---|---|
| White, Not Hispanic | 3.9% | 342,000 |
| Black, Not Hispanic | 5.1% | 114,000 |
| Other or Multiple, Not Hispanic | 1.5% | 29,000 |
| Hispanic | 1.1% | 34,000 |
Prevalence by County Type4
According to the 2023–2024 annual average, past-year DUIC prevalence was similar among residents of large metro counties (3.0%) and small metro or nonmetro counties (4.1%) (Figure 15; Table 15). Despite similar prevalence, substantially more residents reporting DUIC lived in large metro counties because of their larger population—an estimated 386,000 compared with 132,000 in small metro or nonmetro counties (Table 15).
Compared with the 2021–2022 annual average, DUIC prevalence in 2023–2024 did not change significantly for either county type (Figure 15).
| County Type | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Large Metro | 3.0% | 386,000 |
| Small or Non Metro | 4.1% | 132,000 |
DUIC Among Past-Year Cannabis Users
The preceding analyses describe how DUIC prevalence varies across demographic groups in the New York population. To better identify populations that may benefit from targeted prevention efforts, the following analyses focus specifically on past-year cannabis users and examine DUIC prevalence by recency of cannabis use and cannabis use disorder status.
Prevalence by Cannabis Use Recency
According to the 2023–2024 annual average, approximately nine in ten New York residents who reported DUIC during the past year were past-month cannabis users, representing an estimated 468,000 individuals (Table 16). Among past-month cannabis users, 17.9% reported DUIC, compared with 4.1% of those who used cannabis within the past year but not the past month (Figure 16). These findings suggest that past-month cannabis users represent a priority population for DUIC prevention and education efforts.
Compared with the 2021–2022 annual average, DUIC prevalence among past-month cannabis users did not change significantly. In contrast, prevalence among those who used cannabis within the past year but not the past month increased significantly from 1.3% to 4.1% in 2023–2024 (Figure 16).
| Cannabis Use Recency | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Used Within the Past Year, Not the Past Month | 4.1% | 51,000 |
| Used Within the Past Month | 17.9% | 468,000 |
Prevalence by Cannabis Use Disorder Status
According to the 2023–2024 annual average, six in ten New York residents who reported DUIC during the past year had a past-year cannabis use disorder (CUD)5, representing an estimated 322,000 individuals (Table 17).
Among past-year cannabis users, DUIC prevalence was nearly four times higher among those with a past-year CUD (28.1%) than among those without CUD (7.2%) (Table 17). These findings suggest opportunities for targeted interventions among cannabis users with CUD, alongside broader education and awareness efforts for cannabis users generally.
Compared with the 2021–2022 annual average, DUIC prevalence did not change significantly for either group in 2023–2024 (Figure 17).
| Cannabis Use Disorder Status | Prevalence | Estimated Number of Individuals |
|---|---|---|
| Past Year Use Disorder | 28.1% | 322,000 |
| No Past Year Use Disorder | 7.2% | 196,000 |
Footnotes
County type is based on the 2013 USDA Rural-Urban Continuum Codes, where Large Metro includes counties in Metropolitan Statistical Areas (MSAs) with 1 million or more residents, and Small or Non-Metro encompasses counties in MSAs with fewer than 1 million residents as well as all non-metropolitan counties. In New York State, Large Metro counties comprise New York City (Bronx, Kings, New York, Queens, Richmond), Long Island (Nassau, Suffolk), the Lower Hudson Valley (Dutchess, Orange, Putnam, Rockland, Westchester), Erie and Niagara counties (Buffalo–Niagara Falls MSA), and Monroe county (Rochester MSA).↩︎
County type is based on the 2013 USDA Rural-Urban Continuum Codes, where Large Metro includes counties in Metropolitan Statistical Areas (MSAs) with 1 million or more residents, and Small or Non-Metro encompasses counties in MSAs with fewer than 1 million residents as well as all non-metropolitan counties. In New York State, Large Metro counties comprise New York City (Bronx, Kings, New York, Queens, Richmond), Long Island (Nassau, Suffolk), the Lower Hudson Valley (Dutchess, Orange, Putnam, Rockland, Westchester), Erie and Niagara counties (Buffalo–Niagara Falls MSA), and Monroe county (Rochester MSA).↩︎
Respondents who used alcohol on 6 or more days in the past 12 months were classified as having an alcohol use disorder if they met two or more criteria from the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Relevant criteria can be found in the 2024 National Survey on Drug Use and Health: Methodological Summary and Definitions.↩︎
County type is based on the 2013 USDA Rural-Urban Continuum Codes, where Large Metro includes counties in Metropolitan Statistical Areas (MSAs) with 1 million or more residents, and Small or Non-Metro encompasses counties in MSAs with fewer than 1 million residents as well as all non-metropolitan counties. In New York State, Large Metro counties comprise New York City (Bronx, Kings, New York, Queens, Richmond), Long Island (Nassau, Suffolk), the Lower Hudson Valley (Dutchess, Orange, Putnam, Rockland, Westchester), Erie and Niagara counties (Buffalo–Niagara Falls MSA), and Monroe county (Rochester MSA).↩︎
Respondents who used cannabis on 6 or more days in the past 12 months were classified as having a cannabis use disorder if they met two or more criteria from the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Relevant criteria can be found in the 2024 National Survey on Drug Use and Health: Methodological Summary and Definitions.↩︎