Reach and effectiveness of a computer-based alcohol intervention in a swedish emergency room

International Emergency Nursing (2010) 18, 138– 146 a v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m j o u r n a l h o m e p a g e : w w w . e l s e v i e r h e a l t h . c o m / j o u r n a l s / a a e n Reach and effectiveness of a computer-basedalcohol intervention in a Swedish emergency room Anna Trinks MSc (Student) *, Karin Festin PhD,Preben Bendtsen PhD (Professor), Per Nilsen PhD (Associate Professor) Department of Medical and Health Science, Division of Community Medicine, Linko Received 28 April 2009; received in revised form 19 August 2009; accepted 25 August 2009 Objectives: This study evaluates a computerized alcohol intervention implemented in a Swedish emergency department (ED) with regard to the effectiveness of two different types of tailored brief feedback on patients’ drinking patterns and the reach of the intervention.
Methods: The study was a prospective, randomized controlled trial of ED patients. The desig-nated target population was the ED population aged 18–69 years who registered at the triageroom before receiving care. Patients who were categorized as risky drinkers and completed thecomputerized test were randomized to either a long or a short feedback. The feedback was tai-lored on the basis of the individual patient’s responses to questions on their drinking patterns.
Results: The computerized intervention reached 41% of the target population. Those who com-pleted the computerized test and received the feedback were younger than those who did notreceive the intervention. Among those who could be followed up, the feedback was effective inreducing the patient’s weekly alcohol consumption and the number of heavy episodic drinkingoccasions. The long feedback was slightly more effective than the short feedback, but the dif-ferences were not statistically significant.
ª 2009 Elsevier Ltd. All rights reserved.
Alcohol consumption in Sweden has increased substantiallyover the last decade, reaching its highest levels in 100 years. Between 1996 and 2004, consumption increased Corresponding author. Tel.: +46 13227132.
from 8.8 L of 100% alcohol per year per person (over 15 years) to 10.4 L. Since then, consumption has stabilized 1755-599X/$ - see front matter ª 2009 Elsevier Ltd. All rights reserved.
Reach and effectiveness of a computer-based alcohol intervention in a Swedish emergency room ). The alcohol consumption trends in Sweden have led Study participants were eligible if they were aged to concern as to whether the number of alcohol-related 18–69 years. Patients were excluded for the following emergency department (ED) presentations might be on the increase. Patients presenting to EDs have higher weeklyconsumption and more frequently engage in heavy episodic  Structural reasons: arrived by ambulance or received drinking (HED) than the general population immediate care without presenting to the triage room.
 Health reasons: the triage nurses considered the patients setting thus offers an important opportunity to intervene to be too ill, injured, intoxicated or fragile to do the with patients to reduce future alcohol intake  Feasibility reasons: the triage nurses perceived ‘‘logisti- scribed as ‘‘an ideal setting in which to identify and initiate cal’’ problems such as many patients presenting within interventions for alcohol abuse’’ and a short time period or the ED waiting room being being ‘‘ideally situated for alcohol screening, interventions, Despite the ED’s potential as an important venue for The remaining patients comprised the designated target alcohol interventions, emergency care patients are rarely population: patients who should be given a card by the ED triage nurses with an instruction to do the computerized This can be partially explained by the operational difficul- The computerized intervention concept was designed to ties of delivering alcohol interventions in the ED environ- require minimal input by the researchers. The amount of ment, which tends to be busy and overcrowded. Lack of time the researchers devoted to maintaining the interven- time, fear of negative patient response due to the perceived tion did not exceed 2 h/week over the 1-year study period.
sensitivity of the subject, negative attitudes to and lack of The implementation of the concept is described in detail in interest in alcohol prevention, and insufficient knowledge about alcohol-related issues have been cited as key obsta-cles for implementation of alcohol interventions in ED set- Patients who were registered at the ED triage room were Although empirical support is emerging for computerized given an instruction card by a triage nurse and requested health behaviour interventions, very few studies have eval- to answer alcohol-related questions on a touch-screen com- uated the effectiveness of computer-based alcohol inter- puter that was positioned in the adjacent ED waiting room.
ventions delivered in ED settings. We have only been able Participation was voluntary and the patient could exit the computer programme at any point. Patients who completed ). This paucity of research suggests that more the programme received a one-page printout, containing studies are needed to explore the use of computerized either a long or a short feedback on their drinking patterns interventions implemented in ED settings. Effectiveness is and motivation to change drinking behaviour. The patient obviously a critical aspect, but it is also important to con- picked up the printout by the computer and it was not avail- sider the reach of computer-based interventions in terms able to any of the staff. No further person-to-person feed- of the proportion and representativeness of patients who are willing to participate in such interventions. The public Using a randomization algorithm within the computer health impact is a function of effectiveness and reach programme, patients were allocated to one of two types of feedback. The ‘‘long feedback’’ group received tailored a knowledge gap by evaluating a computerized alcohol advice and information concerning the weekly alcohol in- intervention implemented in a Swedish ED with regard to take level, frequency of heavy episodic drinking (HED), its reach and the effectiveness of two different types of tai- and motivation to change current drinking patterns. The lored feedback on patients’ drinking patterns.
printout also included a graphic illustration of a traffic light,indicating the patient’s weekly alcohol consumption and frequency of HED, represented as ‘‘hazardous level’’, an‘‘elevated risk’’ or ‘‘no risk’’. The advice, information, and traffic lights were tailored based on the patient’s an-swers. The ‘‘short feedback’’ group received only the gra- The study was a prospective, randomized controlled trial of phic illustration showing the risk levels regarding their ED patients, conducted over a 1-year period at the ED facil- weekly alcohol consumption and frequency of HED. The ity of the Motala County Hospital, beginning in March 2007.
decision to use two different types of feedback was based The population of Motala is 42,000. Motala County Hospital on the ED staff’s wishes to provide feedback to all patients is a public hospital with a total catchment area of approxi- who did the computerized test. A similar methodology was mately 80,000 people Swedish health care is also used in one of the previous computer-based ED studies publicly funded, i.e. residents are insured by the state and health care services are funded through a taxation scheme At the end of the computer programme, the patients of the county councils. Approval and permission to conduct were asked if they would be willing to respond to a fol- this study were given by an ethical committee.
low-up postal questionnaire. The follow-up questionnaire was mailed 6 months after the ED visit. Two reminders were HED occasions per month, we applied the same principle of using the highest amount in an interval.
Risk drinking was defined as having a weekly consumption of 10 or more drinks for women (P120 g) and 15 or more(P180 g) for men (i.e. hazardous weekly consumption) Several different sources were used to obtain data for this and/or engaged in HED (as defined above) once a month study. ED logs provided data on the total number of pa- or more often. This composite risk drinking definition has tient’s aged 18–69 years presenting to the ED during the been promoted by the National Public Health Institute and study period. Baseline data were collected through the computer, which stored participant-provided information ). Those who drank at above these levels of weekly con- in a database. This made it possible to access socio-demo- sumption and/or frequency of HED are referred to as risky graphic and alcohol consumption information about the par- ticipants and to examine the number of tests initialized andcompleted. Follow-up data were collected by means of a postal questionnaire that was mailed to the patients6 months after their ED visit. Questions on the patient’s alcohol consumption since the ED visit were included inthe questionnaire.
(1) Participation was examined in terms of the propor- Individual interviews were conducted with five triage tions of patients who initialized and completed the nurses and one group interview was conducted with a fur- computerized test, were willing to be followed up ther five triage nurses to obtain information on the different 6 months later, and responded to the follow-up types of patients presenting to the ED who did not initialize the computer test, i.e. estimates of how many patients (2) Representativeness was examined by comparing age were excluded due to structural, health or feasibility rea- and sex for patients who initialized but did not com- sons. After the study was over, the number of ‘‘instruction plete the computerized test with those who com- cards’’ remaining was counted to obtain a measure of how many patients were not given a card by the triage nurses.
investigated by comparing socio-demographic andalcohol data for those who were unwilling to be fol- lowed up with a questionnaire (non-participants),those who did not respond to the follow-up question- The following data were recorded on the computer: (a) naire (non-responders), and those who responded to socio-demographic data on sex, age, education, and occu- the follow-up questionnaire (responders).
pation; (b) data on three drinking variables: frequency of (3) Effectiveness was measured by comparing the long drinking, typical quantity of drinking, and frequency of and short feedback conditions in terms of absolute HED. The data on drinking variables were also recorded in and relative changes, from baseline to follow-up, in weekly alcohol consumption (in grams) and number Frequency of drinking was measured as follows: every of HED occasions per month, and the proportion of day; almost every day; 3–4 times per week; 1–2 times patients who changed from risk drinking to non-risk per week; 2–3 times per month; about once per month; less drinking levels, according to the previously stated often than monthly; had not been drinking during the past year. Typical quantity of drinking was measured as follows:1 standard glass; 2–3 standard glasses; 4–5 standardglasses; 6–7 standard glasses; 8–9 standard glasses; 10 standard glasses or more. One standard drink equals 12 gof pure alcohol.
Pearson’s v2-test and Fisher’s exact test, when appropriate, Responses regarding frequency of drinking and typical were used to analyse the differences in distribution regard- quantity were combined to calculate the weekly consump- ing socio-demographic characteristics ), type tion for each patient, according to a method suggested by of feedback (and proportion of risky drinkers Differences in average weekly consumption were per week (counted as two times per week) and a typical tested with one-way ANOVA (for all three categories in quantity of 4–5 standard glasses (counted as drinking five ) and t-test (all tables). Differences concerning fre- standard glasses) has a weekly consumption of 10 standard quency of HED occasions per month between the two types of feedback were tested with non-parametric tests, HED was defined as consuming four drinks or more on one the Kruskal–Wallis test (for all three categories in occasion for women and five drinks or more on one occasion and the Mann–Whitney test (all three tables). In for men. This standard is widely applied in the international absolute change in consumption within each feedback con- dition was tested with the paired t-test (average weekly in- Frequency of HED was categorized as follows: take) and with the non-parametric Wilcoxon signed-rank never; less than monthly; about once per month; 2–3 times test (number of HED occasions per month). A p-value per month; 1–2 times per week; 3–4 times per week; al- <0.05 was considered statistically significant. SPSS 15.0 most every day or every day. To estimate the number of was used for the statistical calculations.
Socio-demographic and drinking characteristics of the three categories.
Frequency of HED, no. of HED occasions/month (p = 0.009)Median (range) a Non-participants, risky drinkers who completed the computerized questionnaire but answered that they not were willing to respond to the follow-up questionnaire.
b Non-responders, risky drinkers who completed the computerized test and were willing to be followed up, but chose not to respond to the questionnaire.
c Responders, risky drinkers who completed the computerized test, wanted to be followed up and answered the follow-up questionnaire.
patients aged 18–69 years were registered at the ED be- Baseline socio-demographic and drinking charac- fore receiving care during the 1-year study period. Of teristics according to type of feedback.
these, 3016 patients were not given a card with an instruc- tion to use the computer due to structural, health or fea- sibility reasons. The remaining 3848 patients comprisedthe target population for the study, i.e. patients who should be given a card with an instruction to do the com- Of the target population, 1508 patients (39% of the tar- get population) were lost due to oversight by the nurses to give the patients the card with the request to do the com- puterized test. An additional 278 patients (7% of the target population) were given the card but chose not to partici- pate. The test was initialized by 2062 patients and was com- pleted by 1570 patients, constituting 41% of the target The 1570 patients who completed the computerized test were randomized to receive either the long or short feed- back. Of those who completed the test, 560 patients (36%) were categorized as risky drinkers. However, 415 (74%) of those who completed the test did not want to be followed up. The remaining 145 patients (26%) were followed up with a questionnaire. Ninety-three patients replied to the ques- tionnaire, yielding a response rate of 64% for the follow- A number of patients were excluded from the study. One patient who completed the test died within 6 months andwas thus excluded. We could not obtain the addresses of four patients (they were included in the non-participants group as we had baseline data for them). One patient sub- mitted a follow-up questionnaire that had been filled in by Frequency of HED, no. of HED occasions per month another person (the patient was treated as a non-responder since we had relevant baseline data). One patient was ex- cluded because we lacked information about which type of feedback the patient had received. We also excluded20 patients because they reported a weekly consumptionthat was three times higher than the cut-off for risky describes the participation and reasons for attrition In total, 492 patients initialized the computerized test but did not complete it. This group of patients was to a greater Effectiveness of the two types of feedback.
Average weekly consumption (g) at baseline, mean (median) Average weekly consumption (g) at follow-up, mean (median) Absolute change in average weekly consumption, g (p-value) Relative change in average weekly consumption, % Number of HED occasions per month at baseline, mean (median) Number of HED occasions per month at follow-up, mean (median) Absolute change in no. of HED occasions per month (p-value) Relative change in no. of HED occasions per month, % a Test for change in average weekly intake within the ‘long feedback’ group.
b Test for change in average weekly intake within the ‘short feedback’ group.
c Test for change in number of HED occasions per month within the ‘long feedback’ group.
d Test for change in number of HED occasions per month within the ‘short feedback’ group.
Reach and effectiveness of a computer-based alcohol intervention in a Swedish emergency room Patients who received the long feedback after Patients who received the short feedback extent older, 44 versus 41 years (p = 0.004), than those who completed the test (n = 1570). Both groups were comprisedof 55% women.
compares characteristics for the patients who re- presents key data on the three categories who ceived the long and short feedback. There were no statisti- completed the test: non-participants, non-responders, and cally significant differences between the two feedback responders. There were statistically significant differences conditions concerning socio-demographic and drinking char- in age between non-participants and non-responders (p = 0.014); the non-participants were older. However, presents the results concerning the effectiveness there were no significant differences in sex, education or of the two types of feedback, i.e. changes from baseline to occupation between the three categories.
the 6-month follow-up. No significant differences were ob- There were no significant differences between the three served between the long and short feedback pertaining to groups regarding weekly alcohol consumption (p = 0.123), weekly consumption at baseline (p = 0.330) or follow-up but the number of HED occasions per month differed signif- (p = 0.100). Patients who received the long feedback de- icantly between the groups (p = 0.009). Responders had a creased their weekly consumption by 26 g between baseline lower number of HED occasions per month than non- and follow-up (34% reduction), which was statistically sig- responders (p = 0.014) and non-participants (p = 0.005).
nificant (p = 0.029). Patients who received the short feed- back also improved, reducing their weekly consumption by Three-quarters of the patients who initialized the test 24 g between baseline and follow-up (26%), which was also completed it and received the tailored feedback. Those statistically significant (p = 0.039).
who completed the test were younger than those who did There were no statistically significant differences be- not complete the test. Of those patients who completed tween the long and short feedback regarding the frequency the test, the responders (those who responded to the fol- of HED at either baseline or follow-up. Patients who re- low-up questionnaire) had somewhat less detrimental drink- ceived long feedback reduced their frequency of HED by ing patterns than non-participants (those who did not want 1.2 occasions per month (40% reduction) between baseline to be followed up) and non-responders (those who did not and follow-up, which was significant (p < 0.001). Patients respond to the follow-up questionnaire), with responders who received the short feedback reduced their HED fre- having a significantly lower number of HED occasions per quency by 1.0 occasions per month (33% reduction), which month. Our findings are in line with previous brief alcohol was not statistically significant (p = 0.115).
The proportions of risky drinkers at baseline who chan- ged to non-risky drinkers at follow-up did not differ signifi- have found that patients who do not want to be followed up after an intervention tend to drink more than those feedback. Of the patients who received the long feedback, 56% of the risky drinkers at baseline became non-risky drink- The computer-based intervention was effective in reduc- ers at follow-up. The corresponding figure for patients who ing the weekly alcohol consumption and number of HED received the short feedback was 39%. In total, 48% of the occasions per month for patients in both feedback condi- target population became non-risky drinkers at follow-up.
tions who could be followed up. The long feedback wasslightly more effective than the short feedback, but the dif-ference was not statistically significant. The decrease in weekly alcohol consumption and number of HED occasionsper month from baseline to 6-month follow-up ranged be- This study set out to evaluate a computerized alcohol inter- tween 26% and 40% for the two types of feedback. These vention implemented in a Swedish ED with regard to reach beneficial effects are largely consistent with the large body in terms of the proportion and representativeness of pa- of literature on brief face-to-face alcohol interventions.
tients who participated in the intervention, and the effec- There is a solid evidence base that supports the effective- tiveness of two different types of tailored computer- ness of brief alcohol interventions at reducing hazardous generated feedback on patients’ drinking patterns. The and harmful alcohol consumption in non-dependent, non- intervention was integrated into ordinary ED practice and treatment-seeking patients in a variety of settings, includ- required little maintenance, thus adhering to requirements on realistic models for alcohol interventions described in Only two previous computer-based ED studies have been German university hospital ED. They found that patients The reach of the intervention was acceptable. The test who received the computer-based intervention reduced was completed by two-fifths (41%) of the target population, their weekly alcohol consumption by 36% from baseline to i.e. patients who should have been given a card with an 6-month follow-up; weekly consumption in the control instruction to do the computerized test. Unquestionably, this proportion is smaller than in most conventional alcohol set in an American university hospital ED, combined com- intervention studies, even though the loss of patients in puter-based feedback with face-to-face advice under four such studies tends to vary a great deal and is not always re- conditions: generic feedback complemented with brief ad- ported in sufficient detail to allow for comparisons among vice, generic feedback without the advice, computer-gen- erated feedback that was tailored according to the ticipation rates can be achieved by applying more research- patient’s responses complemented with brief advice, and er-supported procedures, e.g. with researchers on hand in tailored feedback without the advice. The feedback con- the ED to encourage patients to initialize the test and even sisted of a booklet printed by the computer for each partic- help them complete the test. However, the present concept ipant. The tailored and generic booklets were identical in was implemented with the intention of being sustained with length, content, and graphics, but the generic version in- minimal researcher input in order to study its effectiveness cluded standard text and graphics rather than content tai- and reach under realistic conditions.
lored to the individual responses. Patients in the tailored Relatively few patients who received a card with a re- feedback and brief advice condition achieved a 48% de- quest to do the computerized test declined to participate.
crease in weekly alcohol consumption from baseline to Our ‘‘refusal rate’’ of 7% of the target population compares 12-month follow-up; for the other conditions, reductions favourably with many ED intervention studies (e.g. of between 26% and 36% were achieved. The corresponding reductions in HED occasions per month ranged from 20% to due to patient refusal typically ranges from about 15% to Our findings suggest that the computerized intervention 35%. The fact that nearly nine out of ten patients who re- could have considerable public health impact, if sustained.
ceived the ‘‘instruction card’’ chose to initialize the test The reach and effectiveness of the intervention must be could possibly be seen as an indicator of patient interest considered very promising, but further research is needed before firm conclusions can be drawn concerning the viabil- Reach and effectiveness of a computer-based alcohol intervention in a Swedish emergency room ity of computer-based interventions implemented in ED set- tings. Research on computer-assisted health behaviourinterventions in general has indicated that computer Adair, J.G., 1984. The Hawthorne effect: a reconsideration of the solutions may have several advantages over conventional methodological artefact. Journal of Applied Psychology 69, face-to-face counselling. For instance, the use of computers has been found to decrease the effect of social desirability ´asson, S., Allebeck, P., 2005. Alkohol och ha and increase the amount of information disclosed ¨versikt om alkoholens positiva och negative effekter Findings also suggest that patients prefer to reveal health. A knowledge review of positive and negative healthconsequences of alcohol].
information of a personal and potentially embarrassing nat- Babor, T.F., Kadden, R.M., 2005. Screening and interventions for alcohol and drug problems in medical settings: what works? Journal of Trauma 59, 80–87 (Discussion 94–100).
interventions can also achieve improved consistency of Barnett, N.P., Monti, P.M., Cherpitel, C., Bendtsen, P., Borges, C., interventions and closer matching of intervention to patient Colby, S.M., Nordqvist, C., Johansson, K., 2003. Identification characteristics and recommended guidelines ( and brief treatment of alcohol problems with medical patients: an international perspective. Alcoholism: Clinical and Experi- This study has a number of obvious shortcomings and lim- itations that must be considered when interpreting the re- Beich, A., Gannik, D., Saelan, H., Thorsen, T., 2007. Screening and sults. We used self-reports of alcohol use. Although self- brief intervention targeting risky drinkers in Danish generalpractice – a pragmatic controlled trial. Alcohol & Alcoholism 42, reports are widely assumed to have reasonable validity Blow, F.C., Barry, K.L., Walton, M.A., Maio, R.F., Chermack, S.T., influence these self-reports, such as the measuring instru- Bingham, C.R., Ignacio, R.V., Strecher, V.J., 2006. The efficacy ment, the types of questions, and how they are adminis- of two brief intervention strategies among injured, at-risk drinkers in the emergency department: impact of tailored The questions on frequency and typical quantity of drinking messaging and brief advice. Journal of Studies on Alcohol 67, and frequency of HED were based on the Alcohol Use Disor- ders Identification Test (AUDIT-C). This instrument has been Bradley, K.A., Bush, K.R., McDonell, M.B., Malone, T., Fihn, S.D., validated and is considered to be a good measurement for 1998. Screening for problem drinking: comparison of CAGE and AUDIT. Journal of General and Internal Medicine 13, 379–388.
Charalambous, M.P., 2002. Alcohol and the accident and emergency ). It is very difficult to determine the extent to which department: a current review. Alcohol & Alcoholism 37, 307– underreporting may have occurred and whether a study set in Sweden produces greater bias than studies conducted Crawford, M.J., Patton, R., Touquet, R., Drummond, C., Byford, S., Barrett, B., Reece, B., Brown, A., Henry, J.A., 2004. Screening Regression-to-the-mean effects are often offered as an and referral for brief intervention of alcohol-misusing patients in explanation for favourable intervention effects, i.e. ex- an emergency department: a pragmatic randomised controlled treme scores on any measure at one point in time will, for purely statistical reasons, probably have less extreme Cherpitel, C.J., 1999. Emergency room and primary care services utilization and associated alcohol and drug use in the United ). Another reason could be Hawthorne effects, i.e.
States general population. Alcohol & Alcoholism 34, 581–589.
Cherpitel, C.J., 2007. Alcohol and injuries: a review of international unintended reactivity to study conditions by study partici- emergency room studies since 1995. Drug and Alcohol Review pants, with a resulting effect on perceived socially desirable behaviour (Thus, awareness that one’s drink- Cryer, H.G., 2005. Barriers to interventions for alcohol problems in ing is being monitored may induce motivational effects trauma centers. Journal of Trauma 59 (Suppl.), S104–S111 and actual behaviour change. However, it would seem plau- sible that Hawthorne effects are smaller when using com- Daeppen, J.B., 2003. Screening and brief alcohol interventions in trauma centers. Swiss Medical Weekly 133, 495–500.
However, regression-to-mean and Hawthorne effects can- Dawson, D., Grant, B., Stinson, F., Zhou, Y., 2005. Effectiveness of not be ruled out in our study. There are also a number of the derived alcohol use disorders identification test (AUDIT-C) in other possible explanations for the positive results seen in screening for alcohol use disorders and risk drinking in the USgeneral population. Alcoholism: Clinical and Experimental our study. However, the mechanisms of change involved in brief alcohol interventions are not fully understood and Del Boca, F.K., Darkes, J., 2003. The validity of self-reports of there is no cohesive theory to explain why these interven- alcohol consumption: state of the science and challenges for research. Addiction 98 (Suppl. 2), 1–12.
more research to explore process factors that may influence Edwards, A.G., Rollnick, S., 1997. Outcome studies of brief alcohol intervention outcomes, including investigations into how intervention in general practice: the problem of lost subjects.
screening and assessment reactivity may promote behaviour Fleming, M.F., Graham, A.W., 2001. Screening and brief interven- tions for alcohol use disorders in managed care settings. RecentDevelopments in Alcoholism 15, 393–416.
Forsberg, L., Ekman, S., Halldin, J., Ronnberg, S., 2000. Brief interventions for risk consumption of alcohol at an emergency The authors have no conflicts of interest to declare.
surgical ward. Addictive Behaviors 25, 471–475.
Glasgow, R.E., Klesges, L.M., Dzewaltowski, D.A., Estabrooks, P.A., Nilsen, P., Kaner, E., Babor, T.F., 2008. Brief intervention, three Vogt, T.M., 2006. Evaluating the impact of health promotion decades on – an overview of research findings and strategies for programs: using the RE-AIM framework to form summary more widespread implementation. Nordic Studies on Alcohol and measures for decision making involving complex issues. Health Nilsen, P., Festin, K., Guldbrandsson, K., Carlfjord, S., Holmqvist, Gmel, G., Rehm, J., 2003. Harmful alcohol use. Alcohol Research & M., Bendtsen, P., 2009. Implementation of a computerized alcohol advice concept in routine emergency care. International Graham, D.M., Maio, R.F., Blow, F.C., Hill, E.M., 2000. Emergency physician attitude concerning intervention for alcohol abuse/ Noell, J., Glasgow, R.E., 1999. Interactive technology applications dependence delivered in the emergency department: a brief for behavioral counseling. Issues and opportunities for health report. Journal of Addictive Diseases 19, 45–53.
care settings. American Journal of Preventive Medicine 17, 269– Hadida, A., Kapur, N., Mackway-Jones, M., Guthrie, E., Creed, F., 2001. Comparing two different methods of identifying alcohol Nordqvist, C., Holmqvist, M., Nilsen, P., Bendtsen, P., Lindqvist, K., related problems in the emergency department: a real chance to 2006. Usual drinking patterns and non-fatal injury among intervene? Emergency Medicine 18, 112–115.
patients seeking emergency care. Public Health 120, 1064–1073.
Heather, N., Rollnick, S., Bell, A., Richmond, R., 1996. Effects of Peters, J., Brooker, C., McCabe, C., Short, N., 1998. Problems encountered with opportunistic screening for alcohol-related on general hospital wards. Drug and Alcohol Review 15, 29– problems in patients attending an accident and emergency department. Addiction 93, 589–594.
Hungerford, D.W., 2005. Interventions in trauma centers for Reinert, D.F., Allen, J.P., 2007. The alcohol use disorders identi- substance use disorders: new insight on an old malady. Journal fication test: an update of research findings. Alcoholism: Clinical and Experimental Research 31, 185–199.
Johnston, B.D., Rivara, F.P., Droesch, R.A.M., Dunn, C., Copass, Rhodes, K.V., Lauderdale, D.S., Stocking, C.B., Howes, D.S., M.K., 2002. Behaviour change counseling in the emergency Roizen, M.F., Levinson, W., 2001. Better health while you wait: department to reduce injury risk: a randomized, controlled trial.
a controlled trial of a computer-based intervention for screening and health promotion in the emergency department. Annals of Littlejohn, C., 2006. Does socio-economic status influence the acceptability of, attendance for, and outcome of, screening and Richmond, R., Heather, N., Wodak, A., Kehoe, L., Webster, I., brief interventions for alcohol misuse: a review. Alcohol & 1995. Controlled evaluation of a general practice-based brief intervention for excessive drinking. Addiction 90, 119–132.
Locke, S.E., Kowaloff, H.B., Hoff, R.G., Safran, C., Popovsky, M.A., Robinson, R., West, R., 1992. A comparison of computer and Cotton, D.J., Finkelstein, D.M., Page, P.L., Slack, W.V., 1992.
questionnaire methods of history-taking in a genito-urinary Computer-based interview for screening blood donors for risk of clinic. Psychology and Health 6, 77–84.
HIV transmission. JAMA 268, 1301–1305.
Rodriguez, R.M., Kreider, W.J., Baraff, L.J., 1995. Need and desire Maio, R.F., 1995. Alcohol and injury in the emergency department: for preventive care measures in emergency department opportunities for intervention. Annals of Emergency Medicine patients. Annals of Emergency Medicine 26, 615–620.
Maio, R.F., Shope, J.T., Blow, F.C., Copeland, L.A., Gregor, M.A., ¨, R., Sillanaukee, P., 1995. Effectiveness of the Brockmann, L.M., Weber, J.E., Metrou, M.E., 2000. Adolescent alcohol use disorder identification test in occupational health injury in the emergency department: opportunity for alcohol screenings. Alcoholism: Clinical and Experimental Research 19, interventions? Annals of Emergency Medicine 35, 252–257.
Midanik, L., 1982. The validity of self-reported alcohol consumption Tate, D.F., Wing, R.R., Winett, R.A., 2001. Using Internet technol- and alcohol problems: a literature review. British Journal of ogy to deliver a behavioral weight loss program. JAMA 285, Neumann, T., Neuner, B., Gentilello, L.M., Weiss-Gerlach, E., Thomas, R., Cahill, J., Santilli, L., 1997. Using an interactive Mentz, H., Rettig, J.S., Schroder, T., Wauer, H., Muller, C., computer game to increase skill and self-efficacy regarding safer Schutz, M., Mann, K., Siebert, G., Dettling, M., Muller, J.M., sex negotiation: field test results. Health Education & Behavior Kox, W.J., Spies, C.D., 2004. Gender differences in the perfor- mance of a computerized version of the alcohol use disorders Tourangeau, R., Smith, T.W., 1996. Asking sensitive questions. The identification test in subcritically injured patients who are impact of data collection mode, question format, and question admitted to the emergency department. Alcoholism: Clinical context. Public Opinion Quarterly 60, 275–304.
and Experimental Research 28 (4), 1693–1701.
Wallace, P., 2001. Patients with alcohol problems – simple Neumann, T., Neuner, B., Weiss-Gerlach, E., Tonnesen, H., Gen- questions is the key to effective identification and management.
tilello, L.M., Wernecke, K.D., Schmidt, K., Schroder, T., Wauer, British Journal of General Practice 51, 172–173.
H., Heinz, A., Mann, K., Muller, J.M., Haas, N., Kox, W.J., Spies, Wallace, P., Cutler, S., Haines, A., 1988. Randomised controlled C.D., 2006. The effect of computerized tailored brief advice on trial of general practitioner intervention in patients with at-risk drinking in subcritically injured trauma patients. Journal excessive alcohol consumption. BMJ 297, 663–668.

Source: https://www.alexit.se/Trinks_et_al_010.pdf