- Email: [email protected]
Differences in alcohol risk profiles between college students and college-age non-students presenting for care in the emergency department
482
ABSTRACTS (ACE)
AEP Vol. 10, No. 7 October 2000: 449–483
patterns exist or their associations with sociodemographic/lifestyle factors. We examined these issues in third National Health and Nutrition Examination Survey data. METHODS: Analyses were based on 5,794 US-born, white participants. Interviews included a 60-item food frequency questionnaire (FFQ) and questions on sociodemographics and health behaviors. Dietary patterns were identified by principal components analysis on FFQ responses. Component scores representing intake level of each pattern were dichotomized at the median. Logistic regression analysis was used to estimate odds ratios (OR) and 95% confidence intervals (CI). RESULTS: We identified two primary dietary patterns—a “western” pattern characterized by processed and red meats, eggs, potatoes, and refined grains, and a “prudent” pattern characterized by cruciferous vegetables, greens, carrots, salads, and fresh fruits. The two patterns occurred across geographic regions and in both women and men. After age- and sex-adjustment, the “western” diet was associated with rural residence (OR ⫽ 1.9, 95% CI 1.7–2.2), working class status (OR ⫽ 1.4, 95% CI 1.2–1.6), and lack of high school completion (OR ⫽ 1.6, 95% CI 1.4–1.8). The “prudent” diet was associated with high school completion (OR ⫽ 2.8, 95% CI 2.5–3.3) and behaviors reflecting greater health awareness, such as daily physical activity (OR ⫽ 2.0, 95% CI 1.8–2.2), non-smoking (OR ⫽ 2.1, 95% CI 1.8–2.4), and supplement use (OR ⫽ 1.8, 95% CI 1.6–2.0). CONCLUSIONS: The two dietary patterns resemble those found in previous US studies. Findings regarding correlates agree with social histories describing the emergence of those patterns. Principal components analysis is potentially useful for identifying fundamental dietary patterns for future investigations of dietdisease associations. PII S1047-2797(00)00088-0
DIFFERENCES IN ALCOHOL RISK PROFILES BETWEEN COLLEGE STUDENTS AND COLLEGE-AGE NON-STUDENTS PRESENTING FOR CARE IN THE EMERGENCY DEPARTMENT J Helmkamp1, D Hungerford2, J Williams1, M Furbee1, B Manley1, K Horn1, 1Center for Rural Emergency Medicine, West Virginia University, Morgantown, WV; 2National Center for Injury Prevention and Control, CDC PURPOSE: Determine whether college students and nonstudents seeking emergency medical care experience different lifestyle choices, alcohol-use patterns, and risk profiles. METHODS: Patients aged 18-25 seen in a university-affiliated ED between August 1998 and June 1999 who had ⭓1 drinks within the past year were eligible. Patients scoring ⬎5 out of a possible 40 on the Alcohol Use Disorder Identification Test (AUDIT) were considered at elevated risk for alcohol problems and were interviewed further to determine alcohol-use patterns and history. RESULTS: Of 1,436 consenting patients, students were more likely to screen positive for alcohol problems than non-students (55% vs 43%; 478/870 vs 245/566). Furthermore, among screenpositive patients, students were more likely than non-students to
binge drink in the two weeks prior to screening (82% vs 65%) and to drink illegally (60% vs 38%). Students needed fewer drinks to feel the effects of alcohol (3.9 drinks vs 5.1) and were less likely to request a referral for alcohol treatment (2% vs 8%). However, students had lower baseline risk for alcohol problems than nonstudents (AUDIT ⫽ 11.2 vs 13.2). Students were younger than non-students (20.2 years vs 21.4), older at their first drunken experience (15.3 vs 14.7), and less likely to smoke (51% vs 76%) or have a family history of heavy drinking (5% vs 23%). (p ⬍ .01 for all comparisons). CONCLUSIONS: Students exhibit a higher risk than nonstudents for behaviors associated with acute alcohol-related problems. However, they may be at lower risk for chronic alcohol problems. This university-based ED seems an appropriate venue for early identification and brief, on-site intervention for students with alcohol problems. PII S1047-2797(00)00115-0
GUN AVAILABILITY, PSYCHOPATHOLOGY, AND RISK OF DEATH FROM SUICIDE ATTEMPT BY GUN ED Shenassa, SN Catlin, SL Buka, Department of Epidemiology, Harvard School of Public Health, Boston, MA PURPOSE: Suicide is the third leading cause of death among adolescents and the 8th leading cause of death overall. Guns figure prominently in this problem. Guns are the most lethal means of suicide, and in the United States about 60% of all suicides are committed by guns. We examined hospital admissions for suicide attempts (para-suicide) in the city of Chicago between 1990–1997. Our aim was to 1) quantify the risk of death from para-suicide by gun versus other methods, 2) determine whether individual with psychopathology are more likely than others to attempt suicide by gun, 3) determine proportion of variation in para-suicide by gun explained by community-level gun availability independent of individual-level determinants of para-suicide by gun. METHODS: Hospital admissions from city of Chicago during 1990–1997 were examined for cases of attempted or completed suicide. 11,584 individuals were identified with a diagnosis of parasuicide (E950–E959, classifiable to ICD-9CM). These individuals were then classified into suicide type (e.g., gun, crash, cut, poison) and were also identified as having a mental disorder or not (mental diagnoses classifiable to ICD-9CM). RESULTS: 1) Those attempting suicide by gun are about 70 times more likely to die from their injuries than those attempting suicide by other means. 2) Depressed and psychotic individuals compared with individuals with no-mental disorders are significantly more likely to attempt suicide with a gun (p ⬍ 0.001). 3) Inclusion of community-level gun availability in a hierarchical model reduced the variation in the model by 16%. CONCLUSIONS: Guns are the most lethal suicide method. Community-level availability, independent of individual-level factors, is an important determinant of suicide attempt by gun. PII S1047-2797(00)00166-6
ABSTRACTS (ACE)
AEP Vol. 10, No. 7 October 2000: 449–483
patterns exist or their associations with sociodemographic/lifestyle factors. We examined these issues in third National Health and Nutrition Examination Survey data. METHODS: Analyses were based on 5,794 US-born, white participants. Interviews included a 60-item food frequency questionnaire (FFQ) and questions on sociodemographics and health behaviors. Dietary patterns were identified by principal components analysis on FFQ responses. Component scores representing intake level of each pattern were dichotomized at the median. Logistic regression analysis was used to estimate odds ratios (OR) and 95% confidence intervals (CI). RESULTS: We identified two primary dietary patterns—a “western” pattern characterized by processed and red meats, eggs, potatoes, and refined grains, and a “prudent” pattern characterized by cruciferous vegetables, greens, carrots, salads, and fresh fruits. The two patterns occurred across geographic regions and in both women and men. After age- and sex-adjustment, the “western” diet was associated with rural residence (OR ⫽ 1.9, 95% CI 1.7–2.2), working class status (OR ⫽ 1.4, 95% CI 1.2–1.6), and lack of high school completion (OR ⫽ 1.6, 95% CI 1.4–1.8). The “prudent” diet was associated with high school completion (OR ⫽ 2.8, 95% CI 2.5–3.3) and behaviors reflecting greater health awareness, such as daily physical activity (OR ⫽ 2.0, 95% CI 1.8–2.2), non-smoking (OR ⫽ 2.1, 95% CI 1.8–2.4), and supplement use (OR ⫽ 1.8, 95% CI 1.6–2.0). CONCLUSIONS: The two dietary patterns resemble those found in previous US studies. Findings regarding correlates agree with social histories describing the emergence of those patterns. Principal components analysis is potentially useful for identifying fundamental dietary patterns for future investigations of dietdisease associations. PII S1047-2797(00)00088-0
DIFFERENCES IN ALCOHOL RISK PROFILES BETWEEN COLLEGE STUDENTS AND COLLEGE-AGE NON-STUDENTS PRESENTING FOR CARE IN THE EMERGENCY DEPARTMENT J Helmkamp1, D Hungerford2, J Williams1, M Furbee1, B Manley1, K Horn1, 1Center for Rural Emergency Medicine, West Virginia University, Morgantown, WV; 2National Center for Injury Prevention and Control, CDC PURPOSE: Determine whether college students and nonstudents seeking emergency medical care experience different lifestyle choices, alcohol-use patterns, and risk profiles. METHODS: Patients aged 18-25 seen in a university-affiliated ED between August 1998 and June 1999 who had ⭓1 drinks within the past year were eligible. Patients scoring ⬎5 out of a possible 40 on the Alcohol Use Disorder Identification Test (AUDIT) were considered at elevated risk for alcohol problems and were interviewed further to determine alcohol-use patterns and history. RESULTS: Of 1,436 consenting patients, students were more likely to screen positive for alcohol problems than non-students (55% vs 43%; 478/870 vs 245/566). Furthermore, among screenpositive patients, students were more likely than non-students to
binge drink in the two weeks prior to screening (82% vs 65%) and to drink illegally (60% vs 38%). Students needed fewer drinks to feel the effects of alcohol (3.9 drinks vs 5.1) and were less likely to request a referral for alcohol treatment (2% vs 8%). However, students had lower baseline risk for alcohol problems than nonstudents (AUDIT ⫽ 11.2 vs 13.2). Students were younger than non-students (20.2 years vs 21.4), older at their first drunken experience (15.3 vs 14.7), and less likely to smoke (51% vs 76%) or have a family history of heavy drinking (5% vs 23%). (p ⬍ .01 for all comparisons). CONCLUSIONS: Students exhibit a higher risk than nonstudents for behaviors associated with acute alcohol-related problems. However, they may be at lower risk for chronic alcohol problems. This university-based ED seems an appropriate venue for early identification and brief, on-site intervention for students with alcohol problems. PII S1047-2797(00)00115-0
GUN AVAILABILITY, PSYCHOPATHOLOGY, AND RISK OF DEATH FROM SUICIDE ATTEMPT BY GUN ED Shenassa, SN Catlin, SL Buka, Department of Epidemiology, Harvard School of Public Health, Boston, MA PURPOSE: Suicide is the third leading cause of death among adolescents and the 8th leading cause of death overall. Guns figure prominently in this problem. Guns are the most lethal means of suicide, and in the United States about 60% of all suicides are committed by guns. We examined hospital admissions for suicide attempts (para-suicide) in the city of Chicago between 1990–1997. Our aim was to 1) quantify the risk of death from para-suicide by gun versus other methods, 2) determine whether individual with psychopathology are more likely than others to attempt suicide by gun, 3) determine proportion of variation in para-suicide by gun explained by community-level gun availability independent of individual-level determinants of para-suicide by gun. METHODS: Hospital admissions from city of Chicago during 1990–1997 were examined for cases of attempted or completed suicide. 11,584 individuals were identified with a diagnosis of parasuicide (E950–E959, classifiable to ICD-9CM). These individuals were then classified into suicide type (e.g., gun, crash, cut, poison) and were also identified as having a mental disorder or not (mental diagnoses classifiable to ICD-9CM). RESULTS: 1) Those attempting suicide by gun are about 70 times more likely to die from their injuries than those attempting suicide by other means. 2) Depressed and psychotic individuals compared with individuals with no-mental disorders are significantly more likely to attempt suicide with a gun (p ⬍ 0.001). 3) Inclusion of community-level gun availability in a hierarchical model reduced the variation in the model by 16%. CONCLUSIONS: Guns are the most lethal suicide method. Community-level availability, independent of individual-level factors, is an important determinant of suicide attempt by gun. PII S1047-2797(00)00166-6