p, 0.01). Presentations were more frequent over the weekend (102, 64%) than on ... Correspondence to: Robert Tait, Department of Psychiatry and Behavioural ...
Addiction (2001) 96, 1059–1067
RESEARCH REPORT
Adolescent emergency department presentations with alcohol- or other drug-related problems in Perth, Western Australia GARY K. HULSE1, SUZANNE I. ROBERTSON2 & ROBERT J. TAIT1 1
University Department of Psychiatry and Behavioural Science, QE II Medical Centre, Nedlands & 2Adolescent Services, Princess Margaret Hospital for Children, Subiaco, Western Australia Abstract Aims. To identify the morbidity, type of substance used and the pattern of presentation by adolescents with problems related to alcohol or other drug (AOD) use. Design. A 4-week retrospective review of hospital records. Setting. Four metropolitan hospitals in Perth, Australia. Participants. There were 1064 presentations by people aged 12–19 years of which 160 (15%) were related to AOD use. The median age of the AOD cases was 17 (interquartile range 16–19) of whom 97 (61%) were male and 19 (12%) were Indigenous Australians. Findings. Alcohol was the most frequent precursor to presentation (66, 41%) followed by heroin (24, 15%) and prescription/over-the-counter drugs (24, 15%). Injury was the most common diagnosis at presentation (50, 31%), followed by overdose/drug use (47, 29%). A diagnosis of injury was signi cantly more likely following the use of alcohol than other categories of substances (c2 5 42.07, df 5 3, p , 0.001). Deliberate self-harm (DSH) occurred in more female than male cases (c2 5 7.4, df 5 1, p , 0.01). Presentations were more frequent over the weekend (102, 64%) than on weekdays, and the length of stay was signi cantly shorter for weekend cases (Mann–Whitney U 2132, p , 0.05). Conclusions. Given the small window of opportunity to provide AOD treatment to youth following hospital presentation, a number of suggestions are made. From a harm-minimization perspective the focus of interventions should be on alcohol use by male youth and DSH associated with prescription/over-the-counter drug use by female adolescents. In addition, Indigenous youth are over-represented in hospital presentations, but there is currently a lack of evaluated interventions designed for them. Introduction There is a high prevalence of illegal drug use among young people in Australia with a nationally representative sample of school students
(aged 12–17) nding that 40% had used illicit drugs on at least one occasion. Cannabis was reported as the most widely used illicit drug, with a life-time prevalence of 36%, while opiate
Correspondence to: Robert Tait, Department of Psychiatry and Behavioural Science, QEII Campus, University of Western Australia, Nedlands WA, 6009, Australia. Submitted 10th August 2000; initial review completed 24th November 2000; nal version accepted 18th December 2000. ISSN 0965–2140 print/ISSN 1360–0443 online/01/071059–09 Ó Carfax Publishing, Taylor & Francis Limited DOI: 10.1080/09652140120053110
Society for the Study of Addiction to Alcohol and Other Drugs
1060
Gary K. Hulse, Suzanne I. Robertson & Robert J. Tait
use was reported by only 3.7% (Lynskey et al., 1999). The use of alcohol is also prevalent with nearly 40% of students aged 12–19 reporting that they currently consume alcohol (Odgers, Houghton & Douglas, 1997). The population prevalence is likely to be greater than these gures indicate, as surveys based on school samples are likely to underestimate the true prevalence of illicit drug and alcohol use because those not attending school are more likely to use drugs than their age-cohort (Bauman & Phongsavan, 1999; Lynskey et al., 1999). In 1996, there were 135 deaths in Australia attributed to the use of alcohol and 56 due to other drugs in people aged 10–19 (Unwin & Codde, 1998). In addition, these substances have a further impact in terms of the morbidity related to their usage, with an Australian study (age range 11–97) nding that nearly 10% of all hospital admissions were probably drug-related (Stanton et al., 1994). In that study, the most frequently cited reasons for drug-related admissions were drug overdose (38%) de ned as deliberate or accidental self-poisoning and adverse drug reactions (30%). Notably, there were only three (0.4%) admissions that were because of recreational use or addiction to a drug. The median age in the Stanton et al. sample (1994) was 67 years, thus it is unlikely to be valid to extrapolate from these data to a teenage population, as young people are likely to have a different pattern of presentation and use a different range of substances. Data from America showed that 34% of adolescent (13–19 years) patients admitted due to trauma gave positive toxicology results for alcohol or other drug (AOD) use (Loiselle et al., 1993). These ndings were replicated in an investigation of adolescent trauma patients associated with alcohol use (36%), again veri ed by biochemical analysis (Maio et al., 1994). The effective assessment and treatment of adolescents with AOD problems requires the co-ordination of a range of specialised services (Crome, 1999) and a comprehensive knowledge of adolescent AOD presentations is required to enable the optimal deployment of these services. The objective of this study was to investigate the pattern of presentations by adolescent users of AOD with particular emphases on (a) the principle substance used, (b) the morbidity recorded at presentation, (c) the time of presentation and (d) the inter-relationship of these factors.
Methods Participants All adolescents aged 12–19 years (inclusive) presenting to the emergency department with conditions related to AOD use were eligible. Pilot data revealed that some AOD cases did not wait for a full assessment by a doctor. Therefore, we deemed patients eligible for inclusion once the triage nurse made an assessment. (The triage evaluation was the assessment by a specialist emergency department nurse when the patient arrived.)
Procedure The study was a retrospective review of presentations at four hospitals in Perth, Australia. We collected data for a 4-week period, during the school/university term-time, but including a 1day public holiday. Three of the hospitals were tertiary teaching institutions while the fourth was a specialist hospital for children. All the hospitals had emergency departments, which effectively provide emergency cover for the main metropolitan area. In three hospitals, participants were identi ed from their emergency department/ triage diagnosis. In the children’s hospital, the cases were identi ed from hospital admission information, which included the observation area in the emergency department. We inspected the hospital records for all cases where AOD was implicated by such terms as: overdose, adverse drug reactions, intoxication, drug induced psychosis and psychosis. We also reviewed the records in cases of fractures, wounds, deliberate self-harm/suicide attempt and nausea/vomiting. There was no time limit on the duration between exposure to AOD and hospital presentation. We de ned AOD to include the use of alcohol, illicit drugs, misuse of prescription drugs or over-the-counter drugs and volatile solvents. We also noted any cases where iatrogenic effects were suspected. We included all cases where AOD use was recorded in hospital records, both with and without biochemical validation. Heroin use was de ned to include its use alone and in combination with other drugs. Ethnic status was taken from hospital records (Aboriginal Australian vs. Other) which would be based on self-report. Cases were categorized as deliberate self-harm (DSH) if they were listed as “suicide attempt” at triage or diagnosis or if the hospital’s
Adolescent emergency department presentations youth self-harm worker was used in the management of the case. The data were collected as part of a larger intervention designed to link young people with substance use problems to appropriate community treatment agencies. Cases were identi ed by a health professional employed by the intervention at three of the hospitals and by the data coordinator at the fourth hospital. Methods of identi cation were discussed extensively to ensure a standard and optimal approach to case detection. The data were abstracted by the intervention’s data coordinator onto pre-designed forms. Human rights and ethics approval was obtained from each participating institution.
Conditions and substances We included any hospital presentation where the use of AOD might have been a contributory factor. Information on the type of substance used was obtained from ambulance notes, hospital transfer documents, emergency department/ triage notes, examination notes and, where available, blood tests. Substances were included on an additive basis for each presentation: all substances reported by a patient or recorded by paramedical or medical staff were listed. The review would not have detected those presenting due to the use of AOD by other people (e.g. victim of drink driving accident).
Analysis Due to the skewed distributions, we used nonparametric statistics on all analyses that involved age or length of stay (LOS) and report medians together with their interquartile ranges (IQR) (the interval between the 25th and 75th percentiles). LOS was calculated as the interval between presentation and discharge. When patients are admitted, the date but not the time of discharge is routinely recorded, so the latest time recorded on the day of discharge was used. A default value of 0900 h was used if there was no time on the discharge day. The mean frequency of presentations (see Fig. 1a, b) was calculated as the number of presentation in each time period divided by the number of sessions that made up that period over the 4-week interval. All analyses were performed using SPSS 10.0.5 and used two-tailed tests with the alpha level set at 0.05.
1061
Results We identi ed 1064 presentations in the target age range at the four hospitals, of which 160 were related to AOD use. We excluded two cases where the presentation did not appear to be related to the substance used: a passenger (marijuana use) injured in a car crash and a case of abdominal pain (alcohol use) where the notes stated that it was not related to alcohol. We included a head injury caused by a motor vehicle accident (alcohol use) where the patient was suspected to be the driver. We identi ed one patient who had three presentations (all heroinrelated) and three patients who had two presentations (one heroin-related, one heroin/ mari juana case, and one alcohol/volatile substance) over the study period with AOD problems. There were 97 (61%) male and 63 (39%) female presentations with a median age of 17 years (IQR 16–19). There was no signi cant difference in age by gender for the sample, but AOD cases at the children’s hospital were signi cantly younger (median 15) than those at the other hospitals (median 18) (Kruskal–Wallis 5 19.5, df 5 3, p , 0.001) and the hospital for children had predominantly female cases. There were 19 (12%) people recorded as Aboriginal Australians and 141 (88%) as other. One person died in hospital after admission with an illicit drug overdose. Table 1 shows the main reasons for presentation recorded on arrival at hospital. The most frequent cause was injuries 50 (31.3%). These could be subdivided into: alleged assaults 29 (58%), vehicle accidents (including skateboards) four (8%), falls three (6%), lacerations (not due to assault) ve (10%) and other causes nine (18% ). Overdose/drug use was the next most frequent cause of presentation with 47 (29.4%) cases. The table also shows that 23 (14.4%) presentations were recorded as deliberate selfharm (DSH) or suicide attempts in the hospital notes. There were signi cantly more female (22%) than male (9%) cases categorized as DSH (c2 5 7.4, df 5 1, p , 0.01). There was no signi cant relationship between day of presentation and DSH diagnosis but there were signi cantly more presentations between 08.00 and 11.59 than during other time periods (Fisher’s exact c2 5 15.25, df 5 5, p , 0.005). More of the injuries occurred in male than female cases (c2 5 11.4, df 5 1, p , 0.001) while a greater proportion of abdominal pain/nausea
1062
Gary K. Hulse, Suzanne I. Robertson & Robert J. Tait
Figure 1. The mean number of presentations for each time period for (a) weekends and (b) weekdays. Note: the time periods were chosen so that the rst two segments (8.00 to 16.30) correspond to the designated “day shift” period. Therefore, the segments are not all of equal periods.
cases were female patients (c2 5 6.4, df 5 1, p , 0.05). Among the Indigenous adolescents, injury was the most common diagnosis (5, 26%) and Indigenous people were represented in all categories of diagnosis except psychosis and drug-seeking. The most frequently identi ed drug was alcohol, which was implicated in 66 (41.3%) presentations, equating to 6.2% of all hospital presentations for this age group. In addition, in a further 17 (11%) instances alcohol was used in combination with other drugs (Table 2). Alcohol used on its own was more frequently identi ed in male than female cases (c2 5 3.87, df 5 1, p , 0.05) and also where alcohol was used in combination with other drugs (c2 5 7.9, df 5 1, p , 0.01). There were 24 (15%) cases associated with heroin (including other opiates and combi-
nations) and also prescription/over-the-counter drugs. Of the latter, three cases were recorded as adverse drug reactions (treatment for cerebral palsy, an antibiotic combination and a NSAID); the remainder was coded as overdose (n 5 17), suicide attempt (n 5 3) and abdominal pain (n 5 1). The use of prescription/over-the-counter drugs was more common in females than males (c2 5 4.25, df 5 1, p , 0.05). Alcohol was the most frequently identi ed substance used by Indigenous youth (n 5 7) but this was not signi cantly different from the frequency in nonIndigenous people. All the cases (n 5 3) involving inhalants were by Aboriginal youth. If the time of presentation is split into “day shift” (0800–1630 h) and “overnight” (1631– 0759 h), the majority of presentations (123, 77%) were at night, with the rate of presentation
Adolescent emergency department presentations
1063
Table 1. Frequency of diagnosis at presentation for male and females together with the number of each type of diagnosis recorded as deliberate self-harm (DSH) Male Admission diagnosis Injury Overdose/drug use Intoxication (alcohol) Abdominal pain/nausea Suicide attempt Psychosis Adverse reaction Drug seeking Other Total
All causes n (%) 40 (41.2)*** 23 (23.7) 13 (13.4) 5 (5.2) 3 (3.1) 2 (2.1) 1 (1) 2 (2.1) 8 (8.2) 97 (100)
Female DSH n (%) 6 (26) 3 (100)
9 (9)
Signi cant differences between males and females: *p ,
being 0.283 cases/hour at night and 0.155/hour during the “day shift”. Figure 1a and b illustrates the different patterns of presentation on weekdays and weekends (between 1631 h on Friday and 0759 h and Monday). There were more presentations (102, 64%) at the weekend than during the week, of which 86 (85%) were on the Friday, Saturday or Sunday nights. All except 16 of the weekend presentations left the hospital before 0800 h on Monday. The LOS ranged from 8 minutes to 14 days with the median LOS for weekend presentations being 2.9 hours (IQR 1.9–8.4) compared with 4.8 hours (IQR 2.4– 32.4) for weekday presentations (Mann–Whitney U 5 2132, p , 0.05). In ve instances, the default of 0900 h was used for the discharge time. The time of presentation also varies with the type of substance used. We grouped the substances into four categories: alcohol plus alcohol and drugs, heroin plus heroin combinations, prescription/over-the-counter drugs and other drugs. Figure 2a and b illustrates the pattern of presentations for different types of substance. There was a signi cant difference in the frequency of alcohol related presentations over the week (c2 5 18.07, df 5 6, p , 0.005) and by time of day (c2 5 27.27, df 5 5, p , 0.001) with a peak in presentations over the weekend period and between 2000 and 0359 h. Presentations related to prescription/over-the-counter drugs are relatively evenly spread across the week, but with a small peak on Fridays. The presentations due to heroin use are also relatively evenly spread over
All causes n (%)
DSH n (%)
10 (15.9) 24 (38.1) 7 (11.1) 11 (17.5)* 3 (4.8) 1 (1.6) 2 (3.2) 1 (1.6) 4 (6.3) 63 (100) 0.05, **p ,
Total
8 (33) 2 (18) 3 (100)
1 (25) 14 (22)**
0.01, ***p ,
All causes n (%) 50 (31.3) 47 (29.4) 20 (12.5) 16 (10) 6 (3.8) 3 (1.9) 3 (1.9) 3 (1.9) 12 (7.5) 160 (100)
DSH n (%) 14 (30) 2 (13) 6 (100)
1 (8) 23 (14)
0.001.
the week, with the highest total being for Saturdays (n 5 7). Due to the large number of cells with less than ve cases, these differences were assessed in terms of weekend vs. weekday presentations. There were signi cantly more weekday than weekend presentations proceeded by prescription drug use (c2 5 14.61, df 5 2, p , 0.001) but not for heroin use.
Discussion This review of adolescents presenting at four emergency departments with problems related to the use of AOD was conducted as part of an intervention to link young people with appropriate treatment agencies. The principal objective was to establish the pro le of adolescent presentations with AOD-related problems, particularly the morbidity diagnosed on arrival, time of presentation and the principle substance used prior to presentation. There are previous data on adolescent trauma admissions related to alcohol consumption (Maio et al., 1994) and AOD use (Loiselle et al., 1993). However, this study attempted to identify all adolescent emergency department presentations related to the full range of AOD use. It included all causes of presentation, not just those with injuries and all degrees of severity, including those people who did not progress beyond the triage station. We found that AOD-related problems accounted for 15% of emergency presentations in people aged 12–19 years. Over half of these cases
1064
Gary K. Hulse, Suzanne I. Robertson & Robert J. Tait Table 2. Frequency of each type of drug recorded at presentation for both sexes Male n (%)
Female n (%)
Total n (%)
46 (47.4)* 13 (13.4)** 14 (14.4) 4 (4.1) 1 (1.0) 2 (2.1) 1 (1.0) 10 (10.3) 4 (4.1) 2 (2.1) 97
20 (31.7) 4 (6.3) 10 (15.9) 5 (7.9) 4 (6.3) 1 (1.6) 1 (1.6) 14 (22.2)* 3 (4.8) 1 (1.6) 63
66 (41.3) 17 (10.6) 24 (15) 9 (5.6) 5 (3.1) 3 (1.9) 2 (1.3) 24 (15) 7 (4.4) 3 (1.9) 160
Substance Alcohol Alcohol & drug(s) Heroin Stimulants Ecstasy Inhalants Marijuana Prescription/over-the-counter Unknown drug(s) Drug seeking Total
Signi cant differences between males and females: *p ,
were preceded by the use of either alcohol on its own or alcohol in conjunction with other substances. Heroin and prescription/over-the counter drugs were the next most frequent substances implicated. The most frequent presentation diagnosis was injury, including related conditions such as wounds and fractures. The preponderance of cases occurred over the weekend and at night, with the majority of these leaving before 0800 h on Monday. An American study of 18–19-year-old patients found that in 5.4% of cases, the presentation was preceded by the use of alcohol and that most alcohol-related admissions occurred at night and especially over the period Thursday to Sunday (Wagner et al., 1999). These American ndings correspond very closely to our results, which showed that 6.2% of all presentations were related to alcohol use and that there was a peak in presentations over the weekend. The prevalence of AOD use we detected for all adolescent presentations was considerably lower than that reported for investigations of adolescent trauma admissions (Loiselle et al., 1993; Maio et al., 1994). This difference re ects the increased odds that (most types of) injury will require medical attention where proceeded by substance use (Spirito et al., 1997) and the use of assays to con rm substance use. In contrast to alcohol presentations, an Australian report on non-fatal heroin overdoses found that overdoses occur generally in the afternoon or at night and that they are evenly distributed throughout the week, with 29% at the weekend (Darke, Ross & Hall, 1996). All our heroin-related presentations were in the after-
0.05, **p ,
0.01.
noon or at night, with no cases between 0700 and 1330 h. However, our results showed that just over 40% of these cases were at the weekend. The Darke and co-worker study (Darke et al., 1996) was of an older sample (mean age 30.2 years) and was thus likely to include a larger proportion of dependent users, whereas our younger sample was likely to have more experimental/recreational users and hence a less even distribution of overdoses. The distribution of presentations, with most cases being at the weekend and at night, has implications for the provision of services and research. During the period of the review, we had researchers in three of the institutions attempting to link adolescents with AOD problems into external treatment agencies. We were originally only covering the period of the “day shift” not only for reasons of cost and convenience but also because of the limited availability of the external services with which we were trying to make immediate contact. We were only able to “link in” 2% of presentations with external agencies over the review period. The low frequency of presentations during the “day shift” increases the cost of recruiting or treating adolescents who use substances. Thus, there may be some justi cation in targeting alcohol support services overnight and especially at the weekends. However, the nding that heroin-related presentations are relatively evenly distributed and that prescription/over-the-counter presentations are more common on weekdays means that a targeted approach may not be appropriate with these groups. A 3-month prospective study of youth (10–24
Adolescent emergency department presentations
1065
Figure 2. (a) shows the day of presentation, (b) shows the time of presentation for the four main categories of substances. Note: the time periods were chosen so that the rst two segments (8.00 to 16.30) correspond to the designated “day shift” period. Therefore, the segments are not all of equal periods.
years) DSH at the same institutions covering all methods of DSH reported that drug overdose/ poisoning was the main method used, and that it was equally frequent for both sexes (Silburn et al., 1991). They found that overall there was a ratio of 1.6 females per male, but that this ratio varied by age, with the highest ratio being 4:1 in those aged 15 or less. Cases were spread evenly over the week, but with a disproportionate number at night. We found signi cantly more female presentations were recorded as DSH, which may re ect the age range of our study. We also found that cases were spread over the week but with signi cantly more DSH presentations during the day than at night. This difference may be be-
cause our study only included DSH related to AOD use. It is possible that there is a greater delay between the DSH action and presentation with this form of self-harm. Indigenous people represent 2.8% of the population of Western Australia, but their age structure means that nearly 40% are aged under 15 years compared with 22% of the non-Indigenous people (Australian Bureau of Statistics, 1996). Previous research has shown that the incidence of rst-time hospital admissions for problems related to illicit drug use was greater in the Indigenous than the non-Indigenous population of Western Australia in 1995 (Patterson et al., 1999). In addition, although the proportion
1066
Gary K. Hulse, Suzanne I. Robertson & Robert J. Tait
of Indigenous Australians who drink alcohol is lower than that of non-Indigenous people, the proportion that drinks at a hazardous level is greater (Australian Bureau of Statistics, 1999). Thus, the disproportion (12%) of Indigenous people in our sample is not surprising. We found a range of substances identi ed on presentation, with alcohol being the most common, but not signi cantly greater than among non-Indigenous people. However, all the presentations involving the use of volatile substances were among Indigenous youth. The most frequent diagnosis in this group was injury, but this represented less than 30% of cases. There are some limitations to the study that should be considered in the interpretation of the data. The data for the length of stay were heavily positively skewed, with most patients having a short period in an emergency department, but a few were admitted for psychiatric or surgical treatment, with the longest admission being 14 days. The discharge time was not recorded for all patients who were admitted. We calculated the LOS to the last recorded time or to the default time of 0900 h on the day of discharge. In addition, we excluded two patients who were transferred to external psychiatric facilities and for whom we had no subsequent discharge date from this section of the analysis. The study was also open to classi cation errors as categorization was primarily based on clinical assessment in the emergency department rather than biochemical assay. In addition, many of the substances reported were illegal or involved underage consumption of alcohol. Therefore, the data probably underestimate adolescent AOD usage, particularly in relation to diagnoses where AOD use would not be a vital indicator, such as injuries. The information that we collected re ects the number of presentations rather than the number of adolescents with AOD problems. To ensure con dentiality, we did not extract identi ed data from the medical records, so it was not possible to tell if a patient had presented at more than one hospital during the survey period, but we could detect multiple presentations at the same hospital. The review also only covers a 4-week period, so these ndings may not be representative of the annual gures, in particular, the use of AOD may be different during school/university vacations.
Implications The ndings of this review of adolescent hospital presentations involving AOD use have implications for the provision of support and treatment for this population. These are: (1) Attending an emergency department following a trauma or overdose may represent a unique “teachable” opportunity to encourage young people to engage in substance use treatment programmes (Gentilello et al., 1988). “Brief interventions” have been demonstrated to be an effective model in treating a range of addictive behaviours, including those in adolescents (Colby et al., 1998; Lawendowski, 1998). Brief interventions for alcohol use typically target people who engage in risky or hazardous levels of drinking rather than those who are severely dependent (Babor & Higgins-Biddle, 2000) a pattern likely to be found in young drinkers. Intervention is dif cult due to the short length of time that adolescents remain in hospital, especially at weekend presentations. However, this approach has been successfully demonstrated with youth that use alcohol (Monti et al., 1999). (2) Suf cient staff need to be rostered at these high demand times to cover both the medical treatment of presentations and the provision of brief interventions, even if this is only to motivate the adolescent to attend a treatment programme. This is particularly pertinent to the high demand at night, especially over the weekend. (3) Brief interventions” offer an opportunity to reduce costs both to the hospital and society (Fleming et al., 2000). We identi ed four people who had repeat presentations within the study period. Treatment for the use of AOD could reduce the number of repeat presentations for medical treatment. (4) These preliminary data suggest from a harm-minimization perspective; there needs to be a focus on preventing injuries associated with alcohol use, especially in male adolescents. In female youth, the prime focus should be on DSH and the use of prescription/over-the-counter drugs. Indigenous youth presented with a range of problems and substances, but the small numbers mean that rm conclusions cannot be drawn. However, to date there have been few
Adolescent emergency department presentations evaluated alcohol misuse interventions in the Indigenous community and none of these followed the brief intervention approach (Gray et al., 2000). Therefore, any intervention using this method should be evaluated closely. Future research should investigate how the services can be successfully targeted at young people and tailored to meet the requirements of different subgroups. The lack of a unique medical identi cation number may an impediment both to this research and to the optimal deployment of resources to assist, what is likely to be a mobile population.
Acknowledgements A grant from Healthway and a University of Western Australia, Faculty of Medicine scholarship supported this work. References
AUSTRALIAN BUREAU OF STATISTICS (1996) Year Book of Australia, 1996 (Canberra, ABS Catalogue no. 1301.0). AUSTRALIAN BUREAU OF STATISTICS (1999) The Health and Welfare of Australia’s Aboriginal and Torres Strait Islander Peoples (Canberra, ABS Catalogue 4704.0). BABOR , T. F. & HIGGINS-BIDDLE, J. C. (2000) Alcohol screening and brief intervention: dissemination strategies for medical practice and public health, Addiction, 95, 677–686. BAUMAN, A. & PHONGSAVAN, P. (1999) Epidemiology of substance use in adolescence: prevalence, trends and policy implications, Drug and Alcohol Dependence, 55, 187–207. COLBY , S. M., MONTI, P. M., BARNETT, N. P., ROHSENOW, D. J., WEISSMAN, K., SPIRITO, A., WOOLARD, R. H. & LEWANDER, W. J. (1998) Brief motivational interviewing in a hospital setting for adolescent smoking—a preliminary study, Journal of Consulting and Clinical Psychology, 66, 574–578. CROME, I. B. (1999) Overview—tackling troubled teenagers together, Drug and Alcohol Dependence, 55, 185–186. DARKE, S., ROSS, J. & HALL, W. (1996) Overdose among heroin users in Sydney, Australia 1. Prevalence and correlates of non-fatal overdose, Addiction, 91, 405–411. FLEMING, M. F., MUNDT, M. P., FRENCH, M. T., MANWELL, L. B., STAUFFACHER, E. A. & BARRY, K. L. (2000) Bene t–cost analysis of brief physician advice with problem drinkers in primary care settings [see comments], Medical Care, 38, 7–18. GENTILELLO, L. M., DUGGAN, P., DRUMMOND, D., TONNESEN, A., DEGNER, E. E., FISCHER, R. P. &
1067
REED, R. L. (1988) Major injury as a unique opportunity to initiate treatment in the alcoholic, American Journal of Surgery, 156, 558–561. GRAY, D., SAGGERS, S., SPUTORE, B. & BOURBON , D. (2000) What works? A review of evaluated alcohol misuse interventions among Aboriginal Australians, Addiction, 95, 11–22. LAWENDOWSKI, L. A. (1998) A motivational intervention for adolescent smokers, Preventive Medicine, 27, A39–A46. LOISELLE, J. M., BAKER, M. D., TEMPLETON, J. M., JR, SCHWARTZ, G. & D ROTT, H. (1993) Substance abuse in adolescent trauma, Annals of Emergency Medicine, 22, 1530–1534. LYNSKEY, M., WHITE, V., HILL, D., LETCHER, T. & HALL, W. (1999) Prevalence of illicit drug use among youth: results from the Australian school students’ alcohol and drugs survey, Australian and New Zealand Journal of Public Health, 23, 519–524. MAIO, R. F., PORTNOY, J., BLOW , F. C. & HILL, E. M. (1994) Injury type, injury severity, and repeat occurrence of alcohol-related trauma in adolescents, Alcoholism: Clinical and Experimental Research, 18, 261–264. MONTI, P. M., COLBY , S. M., BARNETT, N. P., SPIRITO, A., ROHSENOW, D. J., MYERS, M., WOOLARD, R. & LEWANDER, W. (1999) Brief intervention for harm reduction with alcohol-positive older adolescents in a hospital emergency department, Journal of Consulting and Clinical Psychology, 67, 989–994. ODGERS, P., HOUGHTON, S. & DOUGLAS, G. (1997) The prevalence and frequency of drug use among Western Australian metropolitan high school students, Addictive Behaviors, 22, 315–325. PATTERSON, K. M., HOLMAN, C. D., ENGLISH, D. R., HULSE, G. K. & UNWIN, E. (1999) First-time hospital admissions with illicit drug problems in indigenous and non-indigenous Western Australians: an application of record linkage to public health surveillance, Australian and New Zealand Journal of Public Health, 23, 460–463. SILBURN, S., ZUBRICK , S., HAYWARD, L. & REIDPATH, D. (1991) Attempted Suicide among Perth Youth. A report to the Hon. Keith Wilson, Minister for Health, on the initial ndings of a prospective study of youth admitted to Perth hospitals following deliberate self-harm (Perth, Health Department of Western Australia). SPIRITO, A., RASILE, D. A., VINNICK, L. A., JELALIAN, E. & ARRIGAN, M. E. (1997) Relationship between substance use and self-reported injuries among adolescents, Journal of Adolescent Health, 21, 221–224. STANTON, L. A., PETERSON, G. M., RUMBLE, R. H., COOPER , G. M. & POLACK, A. E. (1994) Drug-related admissions to an Australian hospital, Journal of Clinical Pharmacy and Therapeutics, 19, 341–347. UNWIN, E. & CODDE, J. (1998) Comparison of Deaths due to Alcohol, Tobacco and Other Drugs in Western Australia and Australia (Perth, Health Department of Western Australia). WAGNER, E. F., BROWN, S. A., MONTI, P. M., MYERS, M. G. & WALDRON, H. B. (1999) Innovations in adolescent substance abuse intervention, Alcoholism: Clinical and Experimental Research, 23, 236–249.
