THE EFFECTIVENESS OF MOTIVATIONAL INTERVIEWING DELIVERED BY YOUTH WORKERS IN REDUCING DRINKING, CIGARETTE AND CANNABIS SMOKING AMONG YOUNG PEOPLE: QUASI-EXPERIMENTAL PILOT STUDY

EMILY GRAY, JIM McCAMBRIDGE* and JOHN STRANG

National Addiction Centre, Institute of Psychiatry (King's College London), 4 Windsor Walk, Camberwell, London SE5 8AF, UK

* Author to whom correspondence should be addressed at: Fax: +44 (0)207 7018454; E-mail: J.McCambridge{at}iop.kcl.ac.uk

(First received 10 May 2005; first review notified 8 July 2005; in final revised form 27 July 2005; accepted 28 July 2005)


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Aim: To test whether a single session of Motivational Interviewing (MI) focussing on drinking alcohol, and cigarette and cannabis smoking, would successfully lead to reductions in use or problems. Methods: Naturalistic quasi-experimental study, in 162 young people (mean age 17 years) who were daily cigarette smokers, weekly drinkers or weekly cannabis smokers, comparing 59 receiving MI with 103 non-intervention assessment-only controls. MI was delivered in a single session by youth workers or by the first author. Assessment was made of changes in self-reported cigarette, alcohol, cannabis use and related indicators of risk and problems between recruitment and after 3 months by self-completion questionnaire. Results: 87% of subjects (141 of 162) were followed up. The most substantial evidence of benefit was achieved in relation to alcohol consumption, with those receiving MI drinking on average two days per month less than controls after 3 months. Weaker evidences of impact on cigarette smoking, and no evidence of impact on cannabis use, were obtained. Conclusions: Evidence of effectiveness for the delivery of MI by youth workers in routine conditions has been identified. However, the extent of benefit is much more modest than previously identified in efficacy studies.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Motivational interviewing (MI) is a directive client-centred counselling style, which aims to encourage reflection on the risks associated with behaviours such as drug use, in the context of personal values and goals (Miller and Rollnick, 2002Go). It is highly influenced by the work of Rogers (1967)Go, which postulates that when the critical conditions of empathy, positive regard, and genuineness are combined in therapy, a client can naturally change in a healthy positive way. The advantages and disadvantages of someone's drug-related behaviour, as viewed by the persons themselves, are discussed with a view to resolving ambivalence and facilitating change (Miller and Rollnick, 2002Go).

MI was originally developed in the treatment context with problem drinkers, and although there are many applications and adaptations targeting a wide range of behaviours, the strongest evidence of effectiveness remains in reducing alcohol consumption and associated problems (Dunn et al., 2001Go; Burke et al., 2003Go, 2005Go; Hettema et al., 2005Go). MI has the potential also to be offered to people for the purposes of prevention, either targeting existing harms or seeking to avoid future harms or both. Among young people, impressive evidence of successful long-term impact on both alcohol consumption and problems has been obtained by Baer et al. (2001)Go in an individual study. Systematic reviews of more traditional prevention interventions are otherwise not encouraging (Foxcroft et al., 2002Go).

Young people not infrequently use more than one legal or illegal drug, and MI has been adapted to simultaneously target multiple drugs (McCambridge and Strang, 2003Go). This application was tested in a randomized trial to establish efficacy in London Further Education (FE) colleges among young people who were already involved in regular illegal drug use in a comparison with education-as-usual assessment-only (McCambridge and Strang, 2004Go). At 3-month follow-up, a number of benefits were robustly attributed to MI, including reduced cigarette smoking, alcohol consumption, cannabis use, and improvements in other indicators of risk. After 12 months these effects had dramatically reduced, but not disappeared altogether (McCambridge and Strang, 2005Go; Miller, 2005Go).

Effectiveness studies are intended to improve generalizability compared with efficacy studies, and there is a concern that brief intervention effects detected in efficacy studies may be diminished when tested in routine conditions (Kaner et al., 2003Go). The current naturalistic study (reported below) aims to replicate the short-term findings of our previous study (McCambridge and Strang, 2004Go), while extending the method to embrace intervention delivery by college-based youth work practitioners. The literature on training practitioners in MI has developed relatively recently, since the first pilot study by Miller and Mount (2001)Go. The present study was designed to pilot training and other methods for a full effectiveness trial, including daily cigarette smokers and weekly drinkers in addition to cannabis users, in order to utilize more fully the opportunities for MI delivery that occur during routine practice.


    METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Study design
A quasi-experimental study design was used, in which MI was compared with assessment-only, as in the previous study (McCambridge and Strang, 2004Go). Without randomization, attempts to attribute change to intervention must deal with the possibility that intervention and control groups may differ in known or unknown ways at baseline, and subsequently have differing non-intervention histories, which may influence outcomes (Cook and Campbell, 1979Go). Randomization was not considered possible in the circumstances of this study as it would interfere with ongoing relationships with students, and also be highly likely to suffer contamination. The intention was for college practitioners themselves to recruit students naturalistically in their usual place of work, and to deliver interventions within routine conditions. This study was undertaken without dedicated funding. No payment or other allowance of time was set aside for practitioner involvement in the study beyond attendance at a training workshop, and it was agreed that the college practitioners themselves should seek to minimize interference with existing ongoing commitments. Three separate London colleges were selected to act as sites for the recruitment of study controls. FE colleges in UK are non-traditional educational settings, serving largely those of post-compulsory schooling age (16–18 years old) distinct from universities. Ethical approval was obtained.

Practitioner training
Seven practitioners attended training, though two subsequently failed to recruit participants and dropped out of the study. Of the five participating practitioners, four were youth workers, two each from urban and rural areas, along with the lead author (E.G.) who was a psychology graduate working on research studies of MI with young people. For E.G., two additional London FE colleges were identified for interventions delivery. Training of practitioners took place over 4 days, 1 month prior to the start of the data collection period, facilitated by the second author (J.M.). The training was split into two 2-day blocks, held 2 weeks apart, to allow trainees opportunities to practise newly acquired skills between the training blocks.

Training delivery itself was based upon MI principles, with an emphasis on developing skills through practice and supportive feedback. MI techniques such as reflective listening, asking open-ended questions, providing summaries and affirmations were practised and developed as methods for minimizing resistance and eliciting change talk. The previously developed intervention model (McCambridge and Strang, 2003Go) was offered as one possible way to do MI. After training, further practice development was encouraged, through peer supervision amongst practitioners, and by creating opportunities for the examination of audio-recordings of interventions delivered. Practitioners were requested to submit a recording of an intervention session of their choice for feedback to be given. Technical difficulties precluded feedback, as inaudible recordings resulted.

Study procedures
Practitioners recruited intervention participants either through making contacts with students in the natural course of their work, or through purposively approaching students in common areas in college. Control participants were also recruited in this latter way by the first author. Following initial interest from a student, the practitioners checked verbally whether the student met inclusion criteria for the study (daily cigarette smoking, or weekly drinking, or cannabis use) and distributed information and consent materials, containing appropriate assurances of confidentiality.

Recruitment and baseline data collection took place over a period of 3 months. Questionnaires were completed by the students themselves, with help available from practitioners. Where possible, practitioners checked participants' baseline questionnaires for omissions. For those students in the intervention group, the intervention generally took place immediately after the completion of the baseline questionnaire. Following the intervention, practitioners and students rated the intervention separately on feedback sheets.

The effect of this single session of MI was the core object of study, though it was recognized that further discussions with individual study participants were likely to take place. Follow-up data collection took place 3 months after study recruitment. To ensure a good balanced rate of follow-up, participants in the assessment-only control group were paid £5 on successful collection of the follow-up questionnaire (intervention participants received no remuneration). The follow-up questionnaires were also self-completed, and were administered in the same way as the baseline questionnaires, except that practitioners did not check the completed questionnaires of intervention participants, as it was believed that this might compromise the reliability of reporting. Subsequent to self-completion of the follow-up questionnaires, the participants themselves placed these in sealed envelopes, having been given assurances that college-based practitioners would not have access to individual questionnaire results. Where face-to-face contact with participants was not achieved, follow-up questionnaires were distributed and returned by post [5% of those successfully followed up (n = 7)]. Where students failed to return questionnaires within 2 weeks, questionnaires were administered by telephone by E.G. [a further 5% of those successfully followed up (n = 7)].

Outcomes
Outcome data were collected in brief outline for cigarette smoking, alcohol consumption, and cannabis use to minimize the possibility of reactivity to assessment (Bien et al., 1993Go; McCambridge and Strang, 2005Go). For these three drugs, the proportions of students who currently used each drug were analysed, as were consumption rates. The following frequency measures were used: number of cigarettes smoked in the last week; number of days in the last month on which alcohol was consumed; units of alcohol drunk in the previous week; number of days in the last month on which cannabis was smoked; and number of times in the last week when cannabis was smoked. The proportions of individuals in each group who had reduced their consumption, or discontinued their use altogether, for at least 1 week during the study period, were also considered. In addition, as in our earlier study, perceived personal importance and existence of interpersonal problems related to each drug were studied (McCambridge and Strang, 2004Go), along with the interval between waking up in the morning and smoking the first cigarette of the day as a proxy for nicotine dependence, and the frequency of getting drunk among drinkers.

Data analysis
Data were analysed using STATA Version 8. Initial analyses were carried out to establish whether the control and intervention groups were equivalent at baseline on all variables measured. For continuous variables, independent samples t-tests were used, and for categorical variables, {chi}2-test and Fisher's exact test were used as appropriate. Upon discovery of significant baseline differences on several variables (see Results section), further analyses were carried out controlling for the effects of these differences. For this purpose, linear regression was used with continuous outcomes and logistic regression was used with binary outcomes. All baseline variables that were non-equivalent between groups were included in the models with stepwise procedures used to remove variables above a P > 0.1 threshold. These variables were included regardless of their relationship to the particular outcome under study, to guard against confounding by other unknown and unmeasured confounders. Outcomes were evaluated among all participants who provided follow-up data. No corrections for multiple statistical tests were made in the light of the nature of the study.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Sample characteristics and non-equivalence between groups
A total of 162 students were recruited for the study; 59 received MI and 103 were recruited as non-intervention control participants. A satisfactory rate of follow-up was achieved—87% (n = 141). There was no statistically significant difference in attrition rates between groups [Control Group 90% (n = 93) and MI Group 81% (n = 48) retained, respectively, {chi}2 = 2.65; d.f. = 1; P > 0.1], or in mean follow-up study interval (Control Group 95 days compared with MI Group 98 days). Male students were more likely than female students not to be retained at follow-up, with 20% lost to follow-up across both groups (n = 17; {chi}2 = 8.18; d.f. = 1; P = 0.004). Otherwise there were no differences between those lost to follow-up and those remaining in the study. One case was dropped from the analysis for inconsistent information.

Participants were 53% (n = 74) female and 47% (n = 66) male. Most lived with their parents or other family, with only 12% (n = 15) not doing so. Of the participants, 29% (n = 41) reported having ever been in trouble with the police. The drug use characteristics of the achieved sample are presented in Table 1, with it being possible for study participants to surpass more than one inclusion threshold. Other drug use was relatively rare, with only 8 of the MI Group and 9 of the Control Group reporting ever having used any other drugs at study entry.


View this table:
[in this window]
[in a new window]
 
Table 1. Sample composition according to inclusion criteria

 
There were statistically significant baseline differences between the intervention and control groups on some demographic and drug use variables. These differences are presented in Table 2 for the 140 participants for whom usable outcome data are available. These variables were thus all included as covariates in regression models, with the exception of the lifetime alcohol consumption variable owing to collinearity with the current drinking variable.


View this table:
[in this window]
[in a new window]
 
Table 2. Baseline differences between study groups

 
Process data
The first author delivered 26 interventions, for which 21 participants provided follow-up data, while the youth workers delivered the other 33 interventions. Outcome data were available for 16/18 interventions delivered by the two urban practitioners and 11/15 interventions delivered by the two rural practitioners. In the 48 interventions for which there were follow-up data, drinking alcohol was discussed during 85% (n = 41) of intervention sessions, slightly more often than cigarette (79%, n = 38) and cannabis (65%, n = 31) smoking. However, cigarette smoking was identified as the principal drug use behaviour discussed in 52% (n = 25) of all sessions, compared with 25% (n = 12) for drinking alcohol and 15% (n = 7) for cannabis use.

There were no differences in which drugs were discussed between the youth workers (n = 27) and the first author (n = 21). The youth workers had held previous discussions with 74% (n = 20) of those with whom they used MI, though in only 4 of these cases was drug use ever discussed. Further youth worker contacts post-MI session were reported with a mean frequency of 2.29 occasions during the study period. Interventions delivered by the youth workers were on average much briefer than those delivered by the first author (mean duration 20 min compared with 35 min, t = 4.17; P < 0.001). Fewer sessions delivered by the youth workers were audio-recorded—5/27 compared with 11/21 ({chi}2 = 6.1; d.f. = 1; P = 0.014). On simple measures briefly recorded at the end of the intervention session, there was only one statistically significant difference between the youth workers and the first author—participants reported higher ratings of feeling understood by the latter (6.86 compared with 6.26 on a scale of 1–7; t = 2.74, P = 0.01). There were no differences detected in any outcomes between youth workers and the first author.

Cigarette smoking
The prevalence of current cigarette smoking remained stable in both groups during the study period [Control Group 69% (n = 63) at baseline, 70% (n = 64) at follow-up; MI Group 79% (n = 38) at baseline, 73% (n = 35) at follow-up]. The mean number of cigarettes smoked in the previous week by participants in the Control Group decreased from 34.6 to 27.3, and while the MI Group changed little (34.7–33.0), this difference was not statistically significant in the regression model (P > 0.1). There were also no differences in interpersonal problems relating to smoking, the perceived importance of smoking, or in time to first cigarette.

Among all cigarette smokers at study entry there was a significant between-group difference in reported attempts to cut down or stop smoking for a week or more during the study period [73% (n = 27) of the MI Group compared with 45% (n = 28) of the Control Group]. Participants in the MI Group were 3.93 times (OR 1.52–10.18; P = 0.005) more likely to have attempted to change their cigarette smoking, after taking account of the various baseline differences between the two groups in the regression model. This effect was also statistically significant when considering only those who smoked weekly or more frequently at baseline [OR = 4.92 (1.68–14.37); P = 0.004]. Attempts to reduce cigarette smoking were much more common in both groups than actual quit attempts maintained for one week [Control Group 32% (n = 28) compared with 6% (n = 5); MI Group 47% (n = 21) compared with 15% (n = 7)].

Drinking alcohol
The prevalence of current alcohol consumption also remained stable between baseline and follow-up [Control Group 62% (n = 57) at baseline, 60% at follow-up (n = 55); MI Group 88% (n = 42) at baseline, 83% (n = 40) at follow-up]. Among those who were currently drinking at study entry, there was no change in the number of units consumed in the week prior to follow-up data collection [Control Group 8.5 and 7.9 units for baseline and follow-up, respectively, MI Group 9.6 and 10.0 units]. There was a small increase in the number of drinking days in the past month in the Control Group (from 5.2 to 5.8 days), and a larger decrease in the MI Group (from 6.3 to 4.6 days). After controlling for other potential confounders, the MI group drank on average 1.97 fewer days (unstandardized B = 0.30–3.65, P = 0.021) in the month prior to follow-up compared with the Control Group.

A greater proportion of the Control Group reported having stopped or cut down their drinking for 1 week or longer [56% (n = 31) of the Control Group, compared with 36% (n = 15) of the MI Group]. However, when controlling for other potentially confounding variables, this was not statistically significant (P > 0.1). There were also no between-group differences in reported interpersonal problems relating to alcohol, perceived importance of alcohol, or frequency of getting drunk.

Cannabis smoking
There was no change in the prevalence of current cannabis smoking over the 3-month study period [Control Group 35% (n = 32) and 33% (n = 30), respectively; MI Group 52% (n = 25) and 43% (n = 20)]. The mean number of times that cannabis was smoked in the week prior to data collection among baseline users rose slightly in both groups (Control Group 4.0–5.0; MI Group 3.9–4.3). The mean number of days in the past month on which these participants smoked cannabis was only slightly changed in both groups (Control Group 9.0–9.6 days; MI Group 12.3–11.6 days, respectively).

In terms of attempts to cut down or stop smoking cannabis during the 3-month study period, there were no differences between the groups. Approximately half of the current smokers in each group at baseline, including those who smoked less frequently than every week had attempted to cut down or stop their cannabis smoking during the study period (n = 15/32 for the Control Group and n = 12/25 for the MI Group). No differences were detected in reported interpersonal problems relating to cannabis, or perceived importance of cannabis.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Evidence of benefit has been identified for MI delivered by youth workers in FE colleges in this study, although the benefit is uneven across the relevant behaviours studied, with some areas showing clear change, while other areas show little or no change. Students who received MI were almost four times more likely than were controls to have attempted to change their cigarette smoking, either by quitting or cutting down for 1 week or more during the study period, The absence of any difference between the two groups in actual cigarette smoking at the 3-month follow-up interval points towards these attempts not being successfully maintained. There are no data available on the timing of these attempts. Reported attempts to reduce smoking may be particularly vulnerable to information bias among MI participants. On the other hand, it seems entirely plausible that many making attempts to reduce their smoking will not succeed (Prochaska et al., 1992Go), and that a single session of MI is simply not designed to deliver support to assist further change attempts.

Those receiving MI reduced the number of days per month on which they drank alcohol. Students in the Control Group had drunk alcohol on nearly two more days in the last month, than the MI Group at 3-month follow-up. This monthly frequency of drinking effect has been obtained among a population drinking at low levels, which may explain the lack of an effect in the past week consumption outcome. This is a more substantial effect than the cigarette smoking finding, given that it remained present at the end of the 3-month study period. Although the capacity to detect differences in outcomes between youth workers and the first author is limited, it is nonetheless encouraging that it would seem possible to train youth workers to deliver MI to the benefit of the students in FE colleges.

This assumes that these self-reported data are indeed reliable, and the potential for bias may be most acute in the MI condition where the follow-up questionnaire was frequently completed in the presence of the person who delivered the intervention. This arrangement was made to permit assistance with the questionnaire itself, and to prevent additional attrition, but may have adversely impacted upon the reliability of self-report, amongst those who were not convinced of the assurances given on access to individual data. Consideration of demand characteristics indicates that this may have been more likely to occur among those with ongoing relationships with those college-based youth workers who delivered MI. The lack of differences in outcomes for this group suggests that this possibility was successfully constrained by following identified best practice (Harrison, 1995Go), though this cannot be securely known without dedicated study.

Compared with earlier findings (McCambridge and Strang, 2004Go), these results are, however, less impressive and warrant careful scrutiny of differences between the two studies. The inclusion criteria differed, with participants in the previous study being cannabis users, whereas most participants in the present study were not. Study procedural differences that may have contributed to discrepancy between the sets of findings are differences in recruitment methods (peer recruitment was employed in the previous study, but not here) and in follow-up questionnaire administration (self-completion here compared with researcher standardized interview previously). Also, all participants were paid in the previous study, whereas in the present study only control participants were given a £5 payment for their time and to encourage continued study participation.

The quasi-experimental approach of the present study did not succeed in creating equivalence between study groups. Clearly, recruiting controls from different colleges is problematic in this regard, and it will be necessary to stratify allocation by college in future studies. The MI group, although younger, were generally more involved with drug use, and it may be that statistical control of known potential confounders has not been sufficient to eliminate other unknown differences in receptivity to intervention.

It is possible that differences in study findings are also attributable to the differing study conditions (efficacy vs effectiveness) or to variation in the content and quality of MI. Practitioners in the present study reported difficulties in applying MI in practice and the learning process was not straightforward. No formal analyses were undertaken to monitor the content of the interventions, or to assess the practitioners' level of competence. Recent randomized trial evidence identifies that workshop learning is not sufficient for the acquisition of skilful MI practice, and that detailed individual supervision is needed (Miller et al., 2004Go). Other recent studies point towards a consensus on how the effects of MI training should be studied, with audio-recorded evaluation of practice post-workshop using the MI Skills Code needed (see for example Baer et al., 2004Go). In this study, practitioners were encouraged, but not required, to use the previous structured approach to MI delivery (McCambridge and Strang, 2003Go). Attention to outcomes for individual practitioners, as well as for different professional groups, and detailed examination of actual practice will be needed in a future full trial.

There was no evidence of benefit here, with either the use of cannabis or in reported problems with any drugs, and in both cases there were low baseline levels. For all drugs, there appear to be quite substantial fluctuations in use according to attempts to change—regardless of whether students were in the intervention or control group. This may be indicative of instability of drug use patterns amongst this age group, or may reflect information bias, or alternatively be explained by the common observation that the experience of taking part in a research study, even as a non-intervention control participant, invites reactivity to assessment (McCambridge and Strang, 2005Go). We had been mindful of this possibility and sought to minimize the assessment burden, but it may owe more to the nature of the activity of completing an assessment than the actual volume of data (although a dose–response relationship may nonetheless be expected). Further studies of MI will need to pay careful attention to separation of assessment and intervention effects. According to the effects detected here, alcohol consumption may present a much more straightforward target for MI intervention by youth workers than cigarette smoking. Larger individual studies with appropriate statistical power will be needed to ascertain whether the effectiveness of youth worker intervention does indeed vary across drugs, in the way suggested by the meta-analytic data in largely adult treatment studies (Hettema et al., 2005Go).


    ACKNOWLEDGEMENTS
 
This study could not have taken place but for the active support of the practitioners, the colleges involved, and the study participants themselves. No specific funding was secured for this study. The second author is supported by a Wellcome Trust Health Services Research Fellowship.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Baer, J. S., Kivlahan, D. R., Blume, A. W. et al. (2001) Brief intervention for heavy drinking college students: Four-year follow-up and natural history. American Journal of Public Health 91, 1310–1316.[Abstract/Free Full Text]

Baer, J. S., Rosengren, D. B., Dunn, C. W. et al. (2004) An evaluation of workshop training in motivational interviewing for addiction and mental health clinicians. Drug and Alcohol Dependence 73, 99–106.[CrossRef][ISI][Medline]

Bien, T. H., Miller, W. R. and Tonigan, S. J. (1993) Brief interventions for alcohol problems: a review. Addiction 88, 315–336.[ISI][Medline]

Burke, B. L., Arkowitz, H. and Menchola, M. (2003) The efficacy of Motivational Interviewing: a meta-analysis of controlled clinical trials. Journal of Consulting and Clinical Psychology 71,843–861.[CrossRef][ISI][Medline]

Burke, B., Dunn, C., Atkins, D. C. et al. (2005) The emerging evidence base for motivational interviewing: a meta-analytic and qualitative inquiry. Journal of Cognitive Psychotherapy, in press.

Cook, T. D. and Campbell, D. T. (1979) Quasi-Experimentation: Design & Analysis Issues for Field Settings. Rand McNally, Chicago.

Dunn, C., Deroo, L. and Rivara, F. P. (2001) The use of brief interventions adapted from motivational interviewing across behavioural domains: a systematic review. Addiction 96, 1725–1742.[CrossRef][ISI][Medline]

Foxcroft, D. R., Ireland, D., Lister-Sharp, D. J. et al. (2002) Primary prevention for alcohol misuse in young people. The Cochrane Database of Systematic Reviews 2002, Issue 3. Art. No.: CD003024. Doi: 10.1002/14651858.CD003024.

Harrison, L. (1995) The validity of self-reported data on drug use. Journal of Drug Issues 25, 91–111.[ISI]

Hettema, J., Steele, J. and Miller, W. R. (2005) Motivational Interviewing. Annual Review of Clinical Psychology 1.

Kaner, E., Campbell, C., Pienaar, E. D. et al. (2003) Brief interventions for excessive drinkers in primary care health settings. The Cochrane Database of Systematic Reviews 2003, Issue 2. Art. No.: CD004148. Doi: 10.1002/14651858.CD004148.

McCambridge, J. and Strang, J. (2003) Development of a structured generic drug intervention model: a brief application of motivational interviewing with young people. Drug and Alcohol Review 22, 391–399.[CrossRef][ISI][Medline]

McCambridge, J. and Strang, J. (2004) The efficacy of single session motivational interviewing in reducing drug consumption and perceptions of drug-related risk and harm among young people: results from a multi-site cluster randomised trial. Addiction 99, 39–52.[CrossRef][ISI][Medline]

McCambridge, J. and Strang, J. (2005) Deterioration over time in effect of motivational interviewing in reducing drug consumption and related risk among young people. Addiction 100, 470–478.[CrossRef][ISI][Medline]

Miller, W. R. (2005) Motivational interviewing and the amazing shrinking treatment effect. Addiction 100, 421.[CrossRef][ISI][Medline]

Miller, W. R. and Mount, K. A. (2001) A small study of training in motivational interviewing: does one workshop change clinician and client behavior? Behavioural and Cognitive Psychotherapy 29, 457–471.

Miller, W. R. and Rollnick, S. (2002) Motivational Interviewing: Preparing People for Change. Guilford Press, New York.

Miller, W. R., Yahne, C. E., Moyers, T. B. et al. (2004) A randomized trial of methods to help clinicians learn motivational interviewing. Journal of Consulting and Clinical Psychology 72, 1050–1062.[CrossRef][ISI][Medline]

Prochasksa, J. O., DiClemente, C. C. and Norcross, J. C. (1992) In search of how people change: Applications to addictive behaviors. American Psychologist 47, 1102–1114.[CrossRef][ISI][Medline]

Rogers, C. R. (1967) On becoming a person: a therapist's view of psychotherapy. Constable, London.





This Article
Abstract
Full Text (PDF)
All Versions of this Article:
40/6/535    most recent
agh199v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by GRAY, E.
Articles by STRANG, J.
PubMed
PubMed Citation
Articles by GRAY, E.
Articles by STRANG, J.