Adolescent injury in the emergency department: Opportunity for alcohol interventions?

2 5 2 A N N A L S O F E M E R G E N C Y M E D I C I N E 3 5 : 3 M A R C H 2 0 0 0

I N J U R Y P R E V E N T I O N / P E D I A T R I C S / S U R V E Y A R T I C L E

Study objective: Alcohol, the most commonly used sub-stance among adolescents, is frequently associated with injury.Little is known regarding the drinking characteristics of injuredadolescents. Such data are critical for developing emergencydepartment interventions to decrease alcohol-related injuryamong adolescents. We sought to describe the drinking charac-teristics of injured adolescents and to describe the relationshipof injury severity and mechanisms with drinking characteristics.

Methods: This study was a prospective cohort study per-formed in a university hospital (sampled May 1, 1995, to July15, 1995) and a large urban teaching hospital (sampled May 1,1996, to August 1, 1996). The participants were aged 12 to 20years, presenting within 6 hours of an injury. We performed asaliva alcohol test and self-administered questionnaire. Age,sex, E-code, injury severity score (ISS), and ED disposition wererecorded. An alcohol frequency/quantity index was calculated.Descriptive statistics and 95% confidence intervals were calcu-lated.

Results: Two hundred sixty-three patients with a mean age of17 years and a mean ISS of 2.1 (SD 3.5) were recruited. Onehundred fifty-two (50%) were males, and 33 (13%) were admit-ted. Ten (4%) patients had a positive saliva alcohol testresponse. On average, within the last year, these adolescentshad 1.7 adverse alcohol consequences. Sixty percent drank inunsupervised settings, and 36% reported drinking 5 or moredrinks in a row.

Conclusion: Alcohol use/misuse is a substantial problem amonginjured adolescents regardless of severity or mechanism ofinjury. ED physicians should consider screening/intervention orprimary prevention of alcohol problems for all injured adolescents.

[Maio RF, Shope JT, Blow FC, Copeland LA, Gregor MA,Brockmann LM, Weber JE, Metrou ME. Adolescent injury in theemergency department: opportunity for alcohol interventions?Ann Emerg Med. March 2000;35:252-257.]

Adolescent Injury in the Emergency

Department: Opportunity for Alcohol

Interventions?

From the University of MichiganInjury Research Center, Departmentof Emergency Medicine,* Universityof Michigan Transportation ResearchInstitute,‡ Ann Arbor VeteransAdministration Medical Center,§ andDepartment of Emergency Medicine,University of Michigan MedicalCenter,II Ann Arbor, MI.

Received for publication March 23, 1999. Revision received July 27, 1999. Accepted for publication October 5, 1999.

Presented in part at the ResearchSociety on Alcoholism Annual ScientificMeeting, San Francisco, CA, July 1997;the 41st Annual Conference of theAssociation for the Advancement ofAutomotive Medicine, Orlando, FL,November 1997; and the 4th WorldConference on Injury Prevention andControl, Amsterdam, The Netherlands,May 1998.

Supported by the University ofMichigan Substance Abuse ResearchCenter and the Department ofSurgery, University of MichiganMedical Center.

Address for reprints: Ronald F.Maio, DO, MS, University ofMichigan Medical Center, TCB1380/0305, 1500 E Medical CenterDrive, Ann Arbor, MI 48109-0305;734-763-9849, fax 734-763-9298; E-mail [email protected].

Copyright © 2000 by the AmericanCollege of Emergency Physicians.

0196-0644/2000/$12.00 + 047/1/104300doi:10.1067/mem.2000.104300

Ronald F. Maio, DO, MS*

Jean T. Shope, MSPH, PhD*‡

Frederic C. Blow, PhD*§

Laurel A. Copeland, MPH§

Mary Ann Gregor, MHSA*

Laurie M. Brockmann, MPH, MSW§

James E. Weber, DOII

Mary E. Metrou, MDII

A D O L E S C E N T I N J U R Y A N D A L C O H O L I N T E R V E N T I O N SMaio et al

M A R C H 2 0 0 0 3 5 : 3 A N N A L S O F E M E R G E N C Y M E D I C I N E 2 5 3

Preventing adolescent alcohol use/misuse shoulddecrease adolescent alcohol-related injury. No studies,however, have reported on the alcohol use/misuse char-acteristics of adolescent patients with injuries presentingto the ED. Such information would be critical to develop-ing strategies to prevent adolescent alcohol-relatedinjury. The current study provides data on adolescentsexperiencing a wide range of injury and their alcoholinvolvement both at the time of and during the yearbefore injury.

The purpose of this study was (1) to describe thedrinking characteristics of injured adolescents presentingto the ED, and (2) to describe the relationship of thosecharacteristics to injury mechanism and injury.

M A T E R I A L S A N D M E T H O D S

A 2-site, prospective, cohort study design was used.Subjects were patients 12 to 20 years of age presenting tothe ED within 6 hours after injury, excluding suicideattempts. Site 1 is a 950-bed, tertiary, academic medicalcenter verified as a Level I trauma center, with 50,000annual ED visits and located in a suburban setting with alarge majority population. Site 2 is a 540-bed teachinghospital verified as a Level II trauma center with 55,000annual ED visits, located in an urban setting with a largeminority patient population. The communities of the 2sites complement each other to provide an overall samplethat is more representative of the geographic area as awhole. Both sites are staffed by board-certified or board-eligible emergency medicine faculty and emergencymedicine residents. Patients were recruited between 4 PM

and 12 AM on Friday and Saturday at both sites. At site 1,patients were also recruited between 5 PM and 11 PM fromSunday through Thursday. At site 2, patients were alsorecruited on Sunday, Monday, and Wednesday between 5PM and 11 PM. At site 1, data were collected from May 1,1995, to July 15, 1995, and at site 2, data were collectedfrom May 1, 1996, to August 1, 1996. This samplingscheme was followed to maximize the number of patientsrecruited, given budgetary considerations and time con-straints.

Current and past alcohol use/misuse was assessed byself-administered questionnaire items developed, vali-dated, and previously used. The questionnaire contained134 items and required approximately 20 minutes tocomplete. Alcohol use was measured by a frequency/quantity index (AFQ) based on separate items for beer,wine, and liquor (6 items).13 Alcohol misuse was mea-sured by an index formed from 10 items quantifying the

I N T R O D U C T I O N

The major preventable health problems of adolescenceresult primarily from (1) injuries that kill or disable, and(2) lifestyle choices with long-term, adverse, health con-sequences. Injury is the most common cause of adoles-cent morbidity and mortality. More adolescents areinjured or killed as a result of motor vehicle–relatedcrashes, homicides, suicides, or drownings than any ill-ness.1 Although the mortality rate has declined duringthe past 30 years for other age groups in the United States,adolescents have experienced an increase in death ratesfrom injuries.2 Alcohol is frequently associated with fatalinjuries, including fatal injuries of adolescents. Fully 30%of adolescent traffic deaths involve alcohol.3 At least 40%of homicide victims 15 to 24 years old have a measurableblood alcohol level, and 25% have blood alcohol concen-trations (BACs) of 0.10% or more.4 Inebriated suicidevictims between the ages of 15 and 24 are more than 7times as likely to have used a firearm than sober ones.5

Alcohol is involved in as many as 40% of adolescentdrownings.6

For every injury death among adolescents (aged 13 to19 years), there are an estimated 41 hospitalizations and1,100 patients who receive treatment in an emergencydepartment.7 Unfortunately, a paucity of data existsregarding adolescent alcohol use and nonfatal injury. Twostudies on admitted adolescent trauma patients foundthat about one third of those tested were positive for alco-hol or drugs. The first study found no relationship amonginjury mechanism, injury severity, and a positive alcoholtest response, but the investigators noted that individualswith a positive BAC were more likely to have had a previ-ous injury.8 The other study noted that individuals withan intentional injury were more likely to test positive foralcohol or other drugs than those who sustained an unin-tentional injury.9 A recent study of injured adolescentspresenting to the ED reported that of 170 patients aged 10to 16 years, only 1 (0.06%) tested positive for alcohol. Incontrast, of 125 patients aged 17 to 21 years, 14 (11.2%)tested positive for alcohol. No significant relationshipwas noted between the mechanism of injury and alcoholtest status.10 A study from Milwaukee reported a 40%positive BAC rate for injured adolescents presenting tothe ED for treatment and found no significant relation-ship between mechanism of injury and a positive urinealcohol test response.11 Another study, based on self-reports from 14- to 18-year-old patients, found that alco-hol or other drugs were reported particularly often amongvictims of falls, cuts, and gun and assaultive injuries.12



Demographic and injury status characteristics werealso determined for those entering the sample frame butnot completing an SAT or a questionnaire. Alcoholuse/misuse characteristics of the study sample weredescribed and stratified by injury severity, admitted sta-tus, and age group (12 to 14, 15 to 17, and 18 to 20 years)by using 95% confidence intervals. Participants weregrouped in the following age groups: 12.00 to 14.99 yearsin the 12- to 14-year age group; 15.00 to 17.99 years inthe 15- to 17-year age group; and 18.00 to 20.99 years inthe 18- to 20-year age group. To address the extent towhich the ED patients in this study were similar to or dif-ferent from the general population, the findings on 2 vari-ables, AFQ index and reporting 5 or more drinks in a rowin the past year, were compared with those from a previ-ous school-based study.13 The school-based studyreported results from the same 613 students followedfrom 6th through 12th grade between 1985 and 1991.

This study was approved by the institutional reviewboards at both sites.

R E S U L T S

A total of 440 patients entered the sampling frame. Sixty(14%) were missed because they were not identified in atimely manner by the research assistants. Of the 380approached for participation in the study, 321 (84%) con-sented, and 59 refused. Of these, 263 (82%) completedthe questionnaire, and 58 did not. Patients who did notparticipate or did not complete questionnaires had thefollowing characteristics: mean age of 16.6 years (SD 2.3);62.4% male subjects; mean time since injury of 1.12 hours(SD 0.95); mean ISS of 2.19 (SD 3.74); intentional injuryrate of 5.7%; and an admission rate of 14.5%. Two hundredsixty-three patients had complete information. Theircharacteristics were as follows: mean age of 16.9 years (SD2.6); 57.8% male subjects; mean time since injury of 1.31hours (SD 1.21 hours); mean ISS of 2.12 (SD 3.48); inten-tional injury rate of 11.4%; and admission rate of 12.5%.

Table 1 shows alcohol use/misuse characteristics for 3 agecategories. In general, reported alcohol use/misuseincreased with age; however, even patients in the youngestage group reported alcohol use/misuse, with the exceptionof driving after drinking.

Table 2 compares alcohol use/misuse characteristics ofpatients with less serious (no single injury >AIS 1 or ISS≤3) and more serious injury (at least one injury >AIS 2 orISS >3).19 Overall, the rate of positive SAT responses waslow (3.9%). Positive SAT responses ranged from 2 mmol/L(10 mg/dL) to 47 mmol/L (215 mg/dL). Small differences

negative consequences of overindulgence, trouble withpeers, and trouble with adults from drinking alcohol.13

Additional items covered were drinking 5 or more drinksin a row,13,14 driving after drinking, riding in a car with adriver who had been drinking, drinking alcohol whiledriving in a car, and the CAGE questions.15 Saliva alcoholtests (SATs) were conducted by using the quantitativeenzyme diagnostic (QED).16 This is an enzymatic test thatcan be performed rapidly in the ED setting to produce aneasily read colored-bar result. The correlation betweensaliva and alcohol blood levels is reported to be 0.96 to0.97, and correlation of the QED with the BACs has beenreported at 0.98.16 One report has also demonstratedhigh interrater and test/retest reliability with this device.17

Adolescent trauma patients were recruited by researchassistants with the help of ED staff. Consecutive recruit-ment of eligible patients occurred, except when the vol-ume of patients in the ED precluded approaching allpotential patients. For all patients, a medical recordreview was completed, recording time of arrival, vitalsigns, and description of trauma. Consent was obtainedfrom patients 18 through 20 years of age. For patientsyounger than 18 years, assent was obtained from thepatient, and consent was obtained from a parent orguardian. The QED swab was performed immediately onall patients for whom consent was obtained. The self-administered questionnaire was completed by the patientin the ED. When the questionnaire could not be com-pleted in the ED, it was sent home with the patient, alongwith a stamped self-addressed envelope to mail the com-pleted questionnaire. If the questionnaire was notreturned within 2 weeks, the patient was contacted andasked to complete the questionnaire by means of a tele-phone interview.

Study personnel recorded medical data from patientcharts onto a standardized form, including age, sex,time of injury, and ED disposition. E-codes18 (E 800-829, 880-928.9, 960-968.9) were used to define mech-anism of injury, including whether the injury was inten-tional or unintentional, on the basis of a review of theinformation in the medical record. Anatomic injuryseverity was determined after reviewing the medicalrecord, and the Abbreviated Injury Scale (AIS)19 wasused to compute the injury severity score (ISS) for eachpatient.20 ISS was calculated by individuals who hadreceived training in injury scoring. Data were collected,coded, and entered by 2 different individuals. The dataanalyst ran comparison checks for 100% verificationand corrected discrepancies and errors by using theoriginal data sheets.



adolescents presenting to the ED. Although rates of posi-tive SAT responses were low, frequent alcohol use/misusewas reported by the study population. No age categorywas immune from this problem. Even among 12- to 14-year-old patients, substantial alcohol use/misuse wasreported. Particularly alarming from a risk-of-injurypoint of view was the finding that 29% of patients in thisyoung subgroup had ridden with a drinking driver.

This study is similar to several others that have notedlow rates of positive alcohol test responses among adoles-cents presenting to the ED after injury. One study,11 how-ever, reported almost a 40% frequency of positive BACsamong adolescent ED trauma patients in Milwaukee. It ispossible that this high value may have been due to selec-tion bias. The results from the current study agree withthose of a previous retrospective study of adolescent traumapatients admitted to the hospital,8 which concluded thatinjury mechanism and severity could not be used to accu-rately identify patients using alcohol.

Comparing a portion of the current study population bygrade with one school-based study, alcohol use/misusewas fairly similar between these 2 populations. This find-ing is in contrast to what was noted when the adult generalpopulation was compared with a group of ED patients inmotor vehicle crashes.21

There are 2 important implications of our study findings.First, they suggest that the injured adolescent presentingto the ED is not in a high-risk category, in that his or heralcohol use/misuse characteristics are similar to those ofadolescents in general. Second, they suggest that any pre-vention interventions delivered in the ED setting can be

were noted between injury severity groups, with the moreseriously injured group having higher values for all but 4measures. The mean ISS for all admitted patients was 8.5compared with 1.3 for patients who were released from theED.

Table 3 compares alcohol use/misuse characteristicsbetween nonmotor vehicle crash– and motor vehicle crash–induced injuries. Although patients with motor vehiclecrash–induced injuries had a lower percentage of positiveSAT responses, this group had higher values for all but oneof the other measures.

The AFQ indexes for the 8th, 10th, and 12th graders inthe current ED study were 0.88, 1.58, and 2.35, respectively,and those from a previous school-based study13 were 0.77,1.43, and 2.02, respectively (not substantially different).There were too few 6th-grade students in the current studyfor comparison. The percentages of 10th- and 12th-gradestudents reporting 5 or more drinks in a row in the past yearin the current ED study were 35.5% and 38.5%, respec-tively, and for the school-based study, the percentageswere 41.8% and 56.8%, respectively. The question wasnot asked of younger students in the school-based study.

D I S C U S S I O N

This study is the first to obtain self-reported alcohol use/misuse information, as well as alcohol levels from injured

Table 1.Alcohol use/misuse measures by age group.

12–14 Years 15–17 Years 18–20 Years% (95% CI) % (95% CI) % (95% CI)

Measure (n=75) (n=80) (n=108)

SAT+ 1.3 (0.1–8.2) 5.0 (1.6–13.0) 4.7 (1.8–11.1)Drank in past week 10.7 (5.1–20.5) 28.8 (19.5–40.2) 39.8 (30.6–50.0)Drank in past year 37.3 (26.6–49.3) 65.0 (53.4–75.1) 85.2 (76.8–91.0)Drank in unsupervised 33.3 (23.1–45.2) 57.5 (46.0–68.3) 79.6 (70.5–86.5)

settings in past yearReported 5+ drinks in 10.7 (5.1–20.5) 33.8 (23.8–45.3) 54.6 (44.8–64.1)

a row in past yearDrove after drinking in — 21.3 (13.2–32.2) 20.4 (13.5–29.5)

past yearDrank while riding in a 4.0 (1.0–12.0) 25.0 (16.3–36.2) 27.8 (19.8–37.4)

car in past yearRode with drinking 29.3 (19.6–41.1) 42.5 (31.7–54.0) 48.2 (38.6–58.0)

driver in past yearCAGE 9.0 (3.9–18.4) 27.7 (18.6–39.0) 33.0 (24.4–42.8)Mean AFQ index 0.61 (0.17–1.05) 1.89 (1.46–2.31) 2.96 (2.60–3.33)Mean misuse index 0.79 (0.34–1.23) 1.76 (1.33–2.19) 2.22 (1.84–2.59)

CAGE, Cut down, annoyed, guilty, eye opener.

Table 2.Alcohol use/misuse measures by injury severity.

Overall ISS ≤3 ISS >3 % % (95% CI) % (95% CI)

Measure (n=256) (n=204) (n=52)

SAT+ 3.9 3.5 (1.5–7.3) 5.8 (1.5–16.9)Drank in past week 28.9 30.4 (24.3–37.3) 23.1 (13.0–37.2)Drank in past year 65.6 64.2 (57.2–70.7) 71.2 (56.7–82.5)Drank in unsupervised settings 59.8 58.8 (51.7–65.6) 63.5 (48.9–76.0)Reported 5+ drinks 36.3 34.8 (28.4–41.8) 42.3 (29.0–56.7)Drove after drinking 14.8 15.7 (11.1–21.6) 11.5 (4.8–24.1)Drank while riding in a car 20.3 19.6 (14.5–25.9) 23.1 (13.0–37.2)Rode with drinking driver 41.0 40.2 (33.5–47.3) 44.2 (30.7–58.6)CAGE 24.1 25.0 (18.9–32.2) 20.0 (9.6–36.1)Mean AFQ index 1.98 1.94 (1.64–2.23) 2.13 (1.54–2.72)Mean misuse index 1.68 1.71 (1.43–1.99) 1.58 (1.02–2.14)

CAGE, Cut down, annoyed, guilty, eye opener.



Three limitations of this study need to be mentioned.Given that the study was conducted at 2 sites with differ-ent research staff and a large gap in time between studies,it is possible that selection bias occurred. However, thesame study protocol was implemented at both sites withthe same data collection instrument. Use of consistentdecision rules, supervisory study staff, protocols, andtraining procedures limited the potential for bias.

It is also possible that excluding patients frombetween midnight and 5 PM introduced selection bias.ED logs from the study sites indicated that very fewinjured adolescents entered the ED during this timeperiod. Severe injuries often occur between midnightand 6 AM and are more likely to have alcohol involved,whereas less severe injuries typically occur in the latermorning and early afternoon and are less likely to havealcohol involved. Including patients with more severeinjuries might have resulted in a greater number ofpatients with positive alcohol test results but probablywould not have altered our conclusions.

Selection bias may also have occurred because not allpatients who entered the sampling frame participated inthe study, were tested for alcohol, or completed a ques-tionnaire. However, we did not note substantial differ-ences in regard to age, sex, injury severity, or time frominjury between patients who did not consent or completea questionnaire and those who did. It is possible, how-ever, that nonrespondents differed from respondents inother unmeasured ways.

Our subjects were recruited from 2 hospital EDs. It ispossible that similar studies conducted in other geo-graphic areas of the country would yield different findings.Because our data were collected in the summer months, itis possible that conducting the study at other times ofthe year might alter our results. A larger sample sizewould have produced estimates of greater precision.Although this limitation prevents a precise determina-tion of differences among groups, it does not changeour conclusion that a substantial degree of alcoholuse/misuse is present among injured adolescents,whether they are seen and released from the ED or admit-ted to the hospital.

Further research is needed to answer the followingquestions: Can effective ED-based interventions to pre-vent alcohol-related adolescent injury be implemented?Do adolescents encountered in the ED represent a seg-ment of the population not effectively reached throughother venues? What is the cost-effectiveness of these inter-ventions compared with interventions delivered in othersettings? Who will pay for these interventions? What is

based, conceptually, on broad-based primary preventioninterventions that are targeted toward the general adoles-cent population, such as school-based prevention inter-ventions.22-28

Several factors suggest that the ED is an appropriatesetting for adolescent alcohol use intervention. First,many patients at high risk for substance abuse and subse-quent injury do not routinely receive medical care in set-tings other than the ED and are therefore unlikely to receiverisk prevention counseling from a primary medical careprovider.29-31 Even if an adolescent patient has a primarycare provider, it cannot be taken for granted that theprovider will address concerns regarding alcohol use.32

A recent study33 found that almost 75% of adolescentpatients did not receive appropriate prevention interven-tions from their primary care physician. Emergencyphysicians may be able to help fill this void.

Another reason to support ED-based prevention is theconcurrence in time of the injury event and interventionfor alcohol use/misuse. According to the TranstheoreticalModel,34 the chance of an adolescent being at a favorablestage for changing his or her alcohol-related behavior (eg,contemplation, preparation, or action) may be increasedby the ED experience. The immediate postinjury period,irrespective of whether the adolescent had a positive BACat the time of injury, may provide a “teachable moment” tostress interventions to prevent future injuries resultingfrom alcohol use/misuse. The concept of the teachablemoment has been used effectively by dental health practi-tioners to prevent smokeless tobacco use by adoles-cents.35,36

Table 3.Alcohol use/misuse measures by motor vehicle crash status.

Non-MVC MVC % (95% CI) % (95% CI)

Measure (n=197) (n=66)

SAT+ 6.2 (3.4–10.8) 3.1 (0.6–11.6)Drank in past week 28.9 (22.8–35.9) 25.8 (16.2–38.3)Drank in past year 64.5 (57.3–71.1) 68.2 (55.5–78.8)Drank in unsupervised settings 57.9 (50.7–64.8) 65.2 (52.4–76.2)Reported 5+ drinks 32.5 (26.1–40.0) 45.5 (33.4–58.2)Drove after drinking 13.7 (9.4–19.5) 18.2 (10.2–30.0)Drank while riding in a car 18.8 (13.7–25.1) 24.2 (14.9–36.6)Rode with drinking driver 42.6 (35.7–49.8) 36.4 (25.2–49.2)CAGE 23.2 (17.6–29.8) 27.3 (17.4–39.9)Mean AFQ index 1.87 (1.85–1.89) 2.24 (2.17–2.31)Mean misuse index 1.59 (1.30–1.87) 1.91 (1.41–2.40)

MVC, Motor vehicle crash; CAGE, Cut down, annoyed, guilty, eye opener.



21. Maio RF, Waller PF, Blow FC, et al. Alcohol abuse/dependence in motor vehicle crash vic-tims presenting to the emergency department. Acad Emerg Med. 1997;4:256-262.

22. Botvin GJ, Baker E, Dusenbury L, et al. Long-term follow-up results of a randomized drugabuse prevention trial in a white middle-class population. JAMA. 1995;273:1106-1112.

23. Ellickson PL, Bell RM. Drug prevention in junior high: a multi-site longitudinal test. Science.1990;247:1299-1305.

24. Shope JT, Copeland LA, Maharg R, et al. Effectiveness of a high school alcohol misuse pre-vention program. Alcohol Clin Exp Res. 1996;20:791-798.

25. Shope JT, Copeland LA, Kamp ME, et al. Twelfth grade follow-up of the effectiveness of amiddle school-based substance abuse prevention program. J Drug Educ. 1998;28:185-197.

26. Shope JT, Copeland LA, Marcoux BC, et al. Effectiveness of a school-based substance abuseprevention program. J Drug Educ. 1996;26:323-337.

27. Shope JT, Kloska DD, Dielman TE, et al. Longitudinal evaluation of an enhanced AlcoholMisuse Prevention Study (AMPS) curriculum for grades six-eight. J School Health. 1994;64:160-166.

28. Shope JT, Dielman TE, Butchart AT, et al. An elementary school-based alcohol misuse pre-vention program: follow-up evaluation. J Stud Alcohol. 1992;53:106-120.

29. Bindman AB, Grumback K, Keane D, et al. Consequences of queing for care at a public hos-pital emergency department. JAMA. 1991;266:1091-1096.

30. Grumbach K, Keane D, Bindman A. Primary care and public emergency department over-crowding. Am J Public Health. 1993;83:372-378.

31. Pane GA, Farner MC, Salness KA. Health care access problems of medically indigent emer-gency department walk-in patients. Ann Emerg Med. 1991;20:730-733.

32. Blum RW, Beuhring T, Wunderlich M, et al. Don’t ask, they won’t tell: the quality of adoles-cent health screening in 5 practice settings. Am J Public Health. 1996;86:1767-1772.

33. Rosen DS, Elster A, Hedberg V, et al. Clinical preventive services for adolescents: positionpaper of the Society for Adolescent Medicine. J Adolesc Health. 1997;21:203-214.

34. Prochaska JO, Johnson S, Lee P. The transtheoretical model of behavior change. In:Shumaker SA, ed. The Handbook of Health Behavior Change. 2nd ed. New York, NY: SpringerPublishing Co; 1998:59-84.

35. Hovell MF, Slymen DJ, Jones JA, et al. An adolescent tobacco-use prevention trial inorthodontic offices. Am J Public Health. 1996;86:1760-1766.

36. Stevens VJ, Severson H, Lichtenstein E, et al. Making the most of a teachable moment: asmokeless tobacco cessation intervention in the dental office. Am J Public Health. 1995;85:231-235.

the magnitude of effect from these interventions? Willemergency physicians support alcohol interventions inthe ED?

Alcohol misuse is substantial among injured adoles-cent patients presenting to the ED. Therefore screening,interventional strategies, or both should not be limited toadmitted patients. Emergency physicians must take anactive role in developing optimal ED-based strategies toprevent alcohol-related injury among adolescents.

R E F E R E N C E S1. National Center for Health Statistics. Injury-Related Visits to Hospital EmergencyDepartments: United States, 1995 (Advance Data No. 131). Washington, DC: US Department ofHealth and Human Services; 1998. DHHS Publication No. 98-1792.

2. Hoekelman RA, Pless IP. Decline in mortality among young Americans during the 20th cen-tury: prospects for reaching national mortality reduction goals for 1990. Pediatrics.1988;82:582-595.

3. National Highway Traffic Safety Administration. Fact Sheet: Young Drivers. Washington,DC: National Highway Traffic Safety Administration; 1998.

4. Berkelman RL, Callaway JL, Howard LB, et al. Fatal injuries and alcohol. Am J Prev Med.1985;116:21-28.

5. Brent DA, Perper JA, Allman CJ. Alcohol, firearms, and suicide among youth: temporaltrends in Allegheny County, Pennsylvania, 1960 to 1983. JAMA. 1987;257:3369-3372.

6. Howland J, Hingson R. Alcohol as a risk factor for drownings: a review of the literature(1950-1985). Accid Anal Prev. 1988;10:19-25.

7. Gallagher SS, Finison K, Guyer B, et al. The incidence of injuries among 87,000Massachusetts children and adolescents: results of the 1980-81 statewide childhood injuryprevention program surveillance system. Am J Public Health. 1984;74:1340-1347.

8. Maio RF, Portnoy JD, Blow FC, et al. Type, severity and repeat occurrence of injury in ado-lescent alcohol-related trauma. Alcohol Clin Exp Res. 1994;18:261-264.

9. Loiselle JM, Baker MD, Templeton JM, et al. Substance abuse in adolescent trauma. AnnEmerg Med. 1993;22:1530-1534.

10. Metropol SB, Moscati RM, Lillis KA, et al. Alcohol-related injuries among adolescents inthe emergency department. Ann Emerg Med. 1995;26:180-186.

11. Mannenbach MS, Hargarten SW, Phelan MB. Alcohol use among injured patients aged 12to 18 years. Acad Emerg Med. 1997;4:40-44.

12. Spirito A, Rasile DA, Vinnick LA, et al. Relationship between substance use and self-reported injuries among adolescents. J Adolesc Health. 1997;21:221-224.

13. Shope JT, Copeland LA, Dielman TE. The measurement of alcohol use and misuse in acohort of students followed from grade 6 through grade 12. Alcohol Clin Exp Res. 1994;18:726-733.

14. Rachal JV, Williams JR, Brehm Ml, et al. A National Study of Adolescent DrinkingBehavior, Attitudes and Correlates: Final Report. Rockville, MD: National Institute on AlcoholAbuse and Alcoholism; 1975.

15. Ewing JA. Detecting alcoholism: the CAGE questionnaire. JAMA. 1984;252:1905-1907.

16. Christopher TA, Zeccardi JA. Evaluation of the QED saliva alcohol test: a new, rapid, accu-rate device for measuring alcohol in saliva. Ann Emerg Med. 1992;21:1135-1137.

17. Jones AW. Inter- and intra-individual variations in the saliva/blood alcohol during ethanolmetabolism in man. Clin Chem. 1979;25:1394-1398.

18. International Classification of Diseases, 9th Revision, Clinical Modification. Washington,DC: US Department of Health and Human Services; 1980. DHHS Publication No. PHS 80-1260.

19. Association for the Advancement of Automotive Medicine. The Abbreviated Injury Scale(AIS)–1990 Revision. Des Plaines, IL: Association for the Advancement of AutomotiveMedicine, 1990.

20. Baker SP, O’Neill B, Haddon W Jr, et al. The injury severity score: a method for describingpatients with multiple injuries and evaluating emergency care. J Trauma. 1974;14:187-196.

Documents

Adolescent injury in the emergency department: Opportunity for alcohol interventions?