Journal Club

1. Journal Club Alcohol and Health: Current Evidence May–June 2005 www.alcoholandhealth.org

2. Featured Article Brief physician and nurse practitioner-delivered counseling for high-risk drinking. Results at 12-month follow-up. Reiff-Hekking S, et al. J Gen Intern Med. 2005;20:7–13. www.alcoholandhealth.org

3. Study Objective To examine whether… 5–10 minutes of counseling during a regularly scheduled primary care visit …can decrease weekly alcohol consumption at 12 months www.alcoholandhealth.org

4. Study Design • Researchers randomly assigned 3 primary care internal medicine practices to deliver… • an alcohol brief intervention during regular primary care visits (19 trained clinicians) or • usual care (27 clinicians); a fourth practice was assigned to the usual care arm to ensure adequate patient recruitment. • The intervention clinicians received 2.5 hours of training. Further, their practices offered chart prompts, an intervention algorithm reminder, and patient educational materials. www.alcoholandhealth.org

5. Study Design (cont.) • Researchers identified 530 adults drinking risky amounts1; 445 patients provided data at 12-month follow-up and were include in analyses. • Analyses were adjusted for potential confounders (e.g., age sex, baseline consumption). 1Defined in this study as >12 standard drinks per week or binge drinking on 1 or more occasions in the previous month for men; >9 standard drinks per week or binge drinking on 1 or more occasions in the previous month for women www.alcoholandhealth.org

6. Assessing Validity of an Article about Therapy • Are the results valid? • What are the results? • How can I apply the results to patient care? www.alcoholandhealth.org

7. Are the Results Valid? • Were patients randomized? • Was randomization concealed? • Were patients analyzed in the groups to which they were randomized? • Were patients in the treatment and control groups similar with respect to known prognostic variables? www.alcoholandhealth.org

8. Are the Results Valid? (cont.) • Were patients aware of group allocation? • Were clinicians aware of group allocation? • Were outcome assessors aware of group allocation? • Was follow-up complete? www.alcoholandhealth.org

9. Were patients randomized? • The patients, as well as the physicians, were not randomized. • 3 physician practices were randomized. • 2 adjacent practices were combined into 1. • Another was added and assigned to usual care. • As a result, 2 practices delivered brief intervention and 2 practices delivered usual care. www.alcoholandhealth.org

10. Was randomization concealed? • Researchers used a random number generator to assign practices to intervention or usual care. • After that point, randomization could no longer be concealed. www.alcoholandhealth.org

11. Were patients analyzed in the groups to which they were randomized? • Patients were analyzed in the groups to which the practices they visited were randomized. www.alcoholandhealth.org

12. Were the patients in the treatment and control groups similar? • Yes, they were similar on… • age, • sex, • educational level, • race, • family history of alcohol abuse, • smoking, and • alcohol consumption. www.alcoholandhealth.org

13. Were patients aware of group allocation? • Practices and physicians knew whether or not they were to deliver a brief intervention. • Patients reported whether or not they were counseled. It is not known if they knew the status of the practice they visited. www.alcoholandhealth.org

14. Were clinicians aware of group allocation? • Yes, clinicians knew because they were either trained or not. • Also, their offices were either provided with a support system or not. www.alcoholandhealth.org

15. Were outcome assessors aware of group allocation? • It is unknown if research staff who assessed drinking outcomes were aware of group allocation. • 65% of patients in the intervention group were interviewed to determine the content of delivered interventions. Clearly, research staff were aware of group allocation for those patients. www.alcoholandhealth.org

16. Was follow-up complete? • 84% of enrolled patients completed follow-up. www.alcoholandhealth.org

17. What Are the Results? • How large was the treatment effect? • How precise was the estimate of the treatment effect? www.alcoholandhealth.org

18. How large was the treatment effect? Outcomes at 12 Months 1>=5 drinks on 1 occasion for men; >=4 drinks on 1 occasion for women 2Limits that do not exceed binge amounts or excessive weekly amounts *Analyses were adjusted for age, sex, and consumption at baseline. They also accounted for the nesting of patients within physicians and practice sites. www.alcoholandhealth.org

19. How precise was the estimate of the treatment effect? • The difference in consumption between groups was statistically significant. • Difference in mean drinks per week: -2.6 (95% CI, -4.5 to -0.3) • Difference in mean number of binge episodes: -0.4 (95% CI, -1.3 to -0.5) www.alcoholandhealth.org

20. How precise was the estimate of the treatment effect? (cont.) • The difference in safe drinking between groups was borderline significant. • Adjusted odds ratio 1.58* (95% CI, 0.99-2.52; P=0.06) *Analyses were adjusted for age, sex, consumption at baseline, and physician sex, and accounted for the nesting of patients within physicians and practice sites. www.alcoholandhealth.org

21. How Can I Apply the Results to Patient Care? • Were the study patients similar to the patients in my practice? • Were all clinically important outcomes considered? • Are the likely treatment benefits worth the potential harm and costs? www.alcoholandhealth.org

22. Were the study patients similar to those in my practice? • Patients were primarily white. • About one-third were smokers. www.alcoholandhealth.org

23. Were all clinically important outcomes considered? • Potentially important outcomes not considered include… • alcohol-related consequences, • healthcare utilization, and • health-related quality of life. www.alcoholandhealth.org

24. Are the likely treatment benefits worth the potential harm and costs? • The study reported no harms of brief intervention. • The costs involved in the brief intervention include costs of screening all primary care patients, the training of physicians, time during initial and follow-up office visits, and office support systems. • The benefits are modest improvements in alcohol consumption, which may be associated with fewer alcohol-related consequences. • The balance of cost and benefit is not clear. www.alcoholandhealth.org

25. Summary • The intervention in this study, which included physician training, office systems support, screening, and (when indicated) a 5–10 minute brief intervention during a regular primary care visit,… • produced modest, yet lasting, reductions in alcohol intake among patients drinking risky amounts. • Limitations to this study include: • the small number of randomized practices (therefore, confounding may affect patient-level analyses despite the use of sophisticated statistical techniques), • lack of blinding of outcome assessors, and www.alcoholandhealth.org

26. Summary (cont.) • Limitations (cont.) • costs of physician training, which will limit translation of these results into practice; however, having the physician conduct the intervention during a regular office visit enhances applicability. • Additional research is needed to… • determine whether brief counseling over multiple visits in the context of a long-term patient-physician relationship can further reduce risky drinking and related consequences, and • clarify the costs and benefits of such interventions. www.alcoholandhealth.org