Evidence for Genetic Influence on Tobacco Use in Humans 157schools, for example, do not provide comprehensive tobacco
cessation training to medical students (Ferry, Grissino, &
Runfola, 1999). Receiving specialized training, however,
has been shown to lead to increased delivery of smoking ces-
sation counseling among health care professionals (Lancaster,
Silagy, & Fowler, 2000; Sinclair et al., 1998). According to
the •Clinical Practice Guideline for Treating Tobacco Use and
Dependence,Ž which summarizes the results of more
than 6,000 published articles, “ve key components of tobacco
cessation counseling are: (a) ask patients whether they use
tobacco, (b) advise tobacco users to quit, (c) assess patients•
interest in quitting, (d) assist patients with quitting, and
(e) arrange follow-up care (Fiore et al., 2000).
Although physicians are aware of the health conse-
quences of using tobacco (Wechsler, Levine, Idelson, Schor,
& Coakley, 1996), smoking status is assessed in only about
one-half to two-thirds of patient clinic visits, and cessation
assistance is provided in only about one-“fth of smokers•
visits (Goldstein et al., 1997; Thorndike, Rigotti, Stafford, &
Singer, 1998).
Group cessation programs offer an alternative to individ-
ual counseling. In a meta-analysis of 13 studies comparing
group programs to self-help programs (Stead & Lancaster,
2000), group program participants were signi“cantly more
likely to have quit for six or more months (odds ratio 2.1).
Group therapy exhibited similar ef“cacy as similar -intensity
individual counseling. A principal drawback of group pro-
grams is their limited reach, because participation rates tend
to be low (Stead & Lancaster, 2000). Smokers must be moti-
vated not only to attempt to stop, but also to commit the time
and effort required to attend group meetings.
Although data indicate that quit rates are enhanced with
more intensive group programs or counseling, smokers tend
to prefer less intense, briefer forms of self-help counseling
(Fiore, Smith, Jorenby, & Baker, 1994; Hughes, 1993).
Mandatory counseling, such as that required by many health
insurers if a patient is to receive cessation medications at no
cost or at a reduced price (co-pay), may act as a barrier to pa-
tients• quitting (Fiore et al., 1994). Thus, from a health policy
standpoint, it will be important to weigh the costs/bene“ts
of offering a brief, less intense, and less effective treatment
to more potential quitters versus the costs/bene“ts of of fering
a more intense, more effective treatment to fewer potential
quitters.
Future Directions
The market for smoking cessation aids is relatively small
but growing. Based on available sales data from drug manu-
facturers, the current worldwide market for pharmaceutical
aids for cessation is estimated to be $670 million ($505 million
nonprescription products + $165 million, bupropion SR).
Sales of nonprescription NRT products appear to be enjoy-
ing continued growth while that of bupropion SR appears to
be leveling off or even declining in some market segments
(Through the Loop Consulting. Corporate Focus: Health care,
January, 2000 see http://www.throughtheloop.com/focus; also
GlaxoSmithKline Financial Report, 2000; GlaxoSmithKline
Annual Report, 2000 see http://corp.gsk.com/“nancial/
reports/ar/report/op_“nrev_prosp/op_“nrev_prosp.html). It is
clear that improved methods of promoting cessation are
needed; this might include new medications (Centers for Dis-
ease Control and Prevention, 2000), new indications for ex-
isting medications, combination therapy, new or improved
behavioral approaches, and/or increased knowledge for effec-
tive methods of matching medications and behavioral ap-
proaches to individual patients. In addition, research is
needed to examine the safety and ef“cacy of different med-
ications for use in special populations, such as adolescents,
pregnant women, patients with depression, and smokeless to-
bacco users (Fiore et al., 2000).
While risk factors for relapse following treatment with
pharmacological and nonpharmacological cessation ap-
proaches have been identi“ed, relatively little work has been
done to identify which smokers should receive which treat-
ments. One reason for this is that the typical analytic ap-
proach used to identify risk factors (e.g., multiple, logistic, or
Cox regression) lends itself to identi“cation of risk factors
and not necessarily the nature of smokers most or least at
risk. For example, it is incorrect to assume that because all
women, who have been shown to be at higher risk for relapse
than men by conventional statistical methods, require the
same treatment approach. In fact, it can be shown with
appropriate analytic tools, that some women do very well in
response to treatment. Previous research has identi“ed sub-
groups of smokers with wide variation in responsiveness to
both pharmacological (Swan, Jack, Niaura, Borrelli, &
Spring, 1999; Swan, Jack, & Ward, 1997) and nonpharmaco-
logical (Swan, Ward, Carmelli, & Jack, 1993) treatments.
One of the keys to the future of matching treatments to indi-
vidual smokers will be consistent attention to and analysis of
individual differences in treatment responsiveness.EVIDENCE FOR GENETIC INFLUENCE
ON TOBACCO USE IN HUMANSIn humans, as more is learned about the genetic basis of to-
bacco dependence, the chronic use of tobacco is increasingly
appreciated as a complex genetic trait and is likely in”uenced