Current issues of ACP Journal Club are published in Annals of Internal Medicine


Intensive, multiple intervention improved smoking cessation rates in men

ACP J Club. 1991 Sept-Oct;115:50. doi:10.7326/ACPJC-1991-115-2-050

Source Citation

Weissfeld JL, Holloway JJ. Treatment for cigarette smoking in a Department of Veterans Affairs outpatient clinic. Arch Intern Med. 1991 May;151:973-7.



To evaluate the effectiveness of smoking-cessation interventions in encouraging men attending Veterans Affairs clinics to stop smoking.


Randomized controlled study of 8 to 12 months' duration with control, low- and high-intensity treatment groups.


Stop-smoking clinic adjacent to outpatient clinics at a Veterans Affairs medical center.


467 of 1382 smokers approached in 2 general medicine clinics, and another 226 smokers from other sources, registered in a stop-smoking clinic. 533 of the 693 registered were eligible and enrolled. Men with military veteran status who smoked ≥ 20 cigarettes/wk and were receiving medical care at Veterans Affairs outpatient clinics were included. Those planning to move or currently in a stop-smoking program, an alcohol- or drug-treatment program, or receiving psychiatric care were excluded. Follow-up was 94%.


Before randomization, participants filled out questionnaires on general health and demographics and on smoking history, attitudes, and perceptions. Control patients received only literature. Intervention group patients had counseling (20 to 30 min), received information on quitting, were given encouragement, and established a mutually agreed-on quit date. One half of this group (high-intensity) received telephone follow-up, further counseling, literature, nicotine gum prescriptions, and other assistance. All participants were interviewed and asked for a urine sample 6 months after intervention.

Main outcome measure

Self-reported and urine-confirmed quit rates.

Main results

Analysis was done according to intention to treat. For all 5 levels of progressively more restrictive definitions of smoking cessation, the intervention groups had higher quit levels. For self-reporting of no cigarettes (least severe definition), the control group had 8% quitters and the low- and high-intervention groups had 14% and 23% (P < 0.001), respectively. For urine-sample confirmed quitters (most severe definition), the rates were 1%, 6%, and 6% (P < 0.05), respectively.


6-month quit rates were low. At all levels of reporting, the control group (physician advice only) had lower quit levels. The 2 intervention groups (counseling vs counseling plus follow-up) did not differ, especially for confirmed nicotine abstinence.

Source of funding: Department of Veterans Affairs.

Address for article reprint: Dr. J.L. Weissfeld, Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, 130 DeSoto Street, Pittsburgh, PA 15261, USA.


This well-designed trial shows the modest efficacy of clinician efforts to induce smoking cessation among patients who still continue to smoke. The investigators' careful attention to design features makes it possible to examine the results from the perspective of several different outcome measures, to search for maldistribution to treatment groups of patients more or less predisposed to quitting, to characterize recruitment and cessation rates on a (clinical) population base, and to describe the extent to which the intended interventions were actually received. Problematic features of the trial include vigorous screening and trial recruitment efforts (making it unclear to what extent trial results can be applied to other populations of outpatients); a design decision to measure smoking status 6 months after last interventional contact (meaning that time from randomization to smoking status outcome assessment varied systematically by intention-to-treat group); and lack of information about how continued use of nicotine gum affected the more restrictive measures of smoking cessation involving urinalysis for products of nicotine. The net effect of all 3 residual methodologic concerns does not diminish the credibility of the observed difference between the control and active intervention groups but may pose a problem for the assertion of no outcome differences between the low- and high-intensity active treatment groups.

Readers who may consider themselves to be competent clinicians fully prepared to assist patients quit smoking would do well to first outline their own notion of the critical content for an interventional session and next read the brief, apt description of the low-intensity intervention in this article (page 974). The differences will be instructive.

Thomas S. Inui, MD, ScM
University of WashingtonSeattle, Washington, USA