J Canc Educ (2016) 31:338–347

DOI 10.1007/s13187-015-0819-y

The Relationship Between Young Adult Smokers’ Beliefs

About Nicotine Addiction and Smoking-Related
Affect and Cognitions
Erika A. Waters 1 & Eva Janssen 2 & Annette R. Kaufman 3 & Laurel M. Peterson 4 &
Nicole L. Muscanell 5 & Rosanna E. Guadagno 6 & Michelle L. Stock 7

Published online: 24 April 2015

# Springer Science+Business Media New York (outside the USA) 2015

Abstract Risk beliefs and self-efficacy play important roles suggests that it is important for health messages to counter
in explaining smoking-related outcomes and are important to the possible negative effects of messages that strongly empha-
target in tobacco control interventions. However, information size the addictiveness of nicotine, possibly by emphasizing the
is lacking about the underlying beliefs that drive these con- importance of quitting earlier rather than later. The findings
structs. The present study investigated the interrelationships also add to the evidence base that feelings of risk are powerful
among young adult smokers’ beliefs about the nature of nic- predictors of behavioral intentions. Furthermore, our results
otine addiction and smoking-related affect and cognitions suggest that in some circumstances, feelings of risk predict
(i.e., feelings of risk, worry about experiencing the harms of quit intentions beyond that predicted by worry and self-effi-
smoking, self-efficacy of quitting, and intentions to quit). cacy. Gaining additional understanding of the tobacco-related
Smokers (n=333) were recruited from two large universities. beliefs that can increase feelings of risk and incorporating
Results showed that quit intentions were associated with feel- those beliefs into educational campaigns may improve the
ings of risk, but not with worry or self-efficacy. Furthermore, quality of such campaigns and reduce tobacco use.
higher feelings of risk were associated with lower beliefs that
addiction is an inevitable consequence of smoking and with Keywords Gene-environment interaction . Tobacco use .
lower beliefs that the harms of smoking are delayed. This Risk perception . Health beliefs

* Erika A. Waters Introduction

waterse@wudosis.wustl.edu
Tobacco use is one of the most important preventable health
1
Division of Public Health Sciences, Washington University School of
risks in the world, accounting for over 500,000 deaths in the
Medicine, 660 S. Euclid Ave, Campus, Box 8100, St. United States alone each year [1]. Cigarette smoking causes a
Louis, MO 63110, USA wide range of diseases, including multiple forms of cancer,
2
Maastricht University School for Public Health and Primary Care coronary heart disease, stroke, and obstructive pulmonary dis-
(Caphri), Department of Health Promotion, Faculty of Health, ease. It is therefore not surprising that many public health
Medicine and Life Sciences, Maastricht, The Netherlands agencies identify reducing tobacco use as a top priority.
3
Division of Cancer Control and Population Sciences, National Recommended strategies to reduce tobacco use include
Cancer Institute, Rockville, MD, USA attempting to motivate smokers to quit by educating them about
4
Department of Psychology, Bryn Mawr College, Bryn Mawr, PA, the long-term health problems caused by tobacco use and provid-
USA ing them with information and resources about how to quit [2].
5
Knowledge Media Research Center, Tübingen, Germany Empirical research supports targeting health risk beliefs for
6
Emerging Media and Communication; Department of Psychology, tobacco control purposes. In one study, smokers who per-
University of Texas at Dallas, Richardson, TX, USA ceived themselves at high risk of experiencing a negative
7
Department of Psychology, George Washington University, health outcome expressed greater interest in quitting [3].
Washington, DC, USA Another study demonstrated that smokers who were worried
J Canc Educ (2016) 31:338–347 339

about their health were more likely to attempt to quit [4]. physiological and psychological factors [17], including be-
There is also extensive literature supporting the premise that liefs that are associated with continued smoking behavior.
smokers who have high confidence in their ability to quit For example, they underestimate the addictiveness of nico-
smoking (i.e., self-efficacy) are more likely to succeed [5]. tine and overestimate the ease of quitting [17–19].
These constructs—perceived risk, worry about negative Moreover, they display unrealistic optimism about their
health outcomes, and self-efficacy of changing one’s behav- ability to resist nicotine addiction and their ability to quit
ior—are integral components of many health behavior theo- smoking easily [20]. That is, they believe that they are less
ries, including protection motivation theory [6] and the ex- likely to become addicted to nicotine and that will be able to
tended parallel process model [7]. quit more easily than the average smoker their age. In ad-
However, many health behavior theories conceptualize dition, many young adults believe that either the harms of
perceived risk as cognitive in nature [6–8]. For example, cigarette smoking are delayed [21] or that there are no
participants might be asked to indicate how likely they health risks associated with smoking for “only a few years”
think they are to experience an outcome. More recent [22]. However, no amount of smoking is safe [1]. Even
theories suggest that cognitively based perceptions of risk smoking a few cigarettes per day—so-called “light”
are missing an important component—affect [9, 10]. The smoking—increases the risk of a variety of immediate
affective aspect is manifested in feelings about a risk rath- health problems including lower respiratory tract infections
er than thoughts about a risk. Affective risk perceptions and impaired healing [23], as well as delayed health prob-
are correlated with, yet distinct from, cognitive risk per- lems such as coronary heart disease [1]. Considered togeth-
ceptions [11]. They are distinguished by the assertion that er with evidence about the power of nicotine addiction [1],
affective feelings are integral and vital components of risk this research suggests that many (though not all) young
perceptions themselves [9, 10]. Furthermore, although af- smokers may not make plans to quit smoking until they
fective risk perceptions are correlated with other affective are already addicted to nicotine. The challenge for public
constructs such as worry about experiencing a negative health organizations is to identify ways to encourage all
health outcome, they are not identical constructs [11]. smokers, including young adults, to quit smoking immedi-
Measures of affective risk perceptions ask people to indi- ately instead of “in a few years.”
cate how vulnerable they feel, or the extent to which they
feel they are going to experience the outcome.
There is extensive empirical support for distinguishing Objective and Research Questions
between cognitive and affective risk perceptions. For a
variety of health behaviors (including but not limited to Typically, health behavior constructs such as perceived risk,
cigarette smoking), feelings of risk have been found to be worry, self-efficacy, and intentions are examined in relation to
more predictive of health behavior and/or behavioral in- one another and/or to an outcome of interest. However, to
tentions than cognitively oriented risk perceptions improve the effectiveness of behavioral interventions, a better
[12–15], worry about experiencing the health problem understanding of how these health behavior constructs are
[14, 15], and self-efficacy of changing behavior [13]. related to specific tobacco-related beliefs is needed.
Therefore, it is likely important for smoking cessation Additional knowledge is also needed about feelings of risk
interventions to consider people’s feelings about their and their relationship with cognitions related to tobacco use,
risk of experiencing negative health outcomes from particularly among young adult smokers. Understanding these
smoking. This idea is consistent with many current inter- issues might lead to the identification of potential content
ventions, which attempt to increase cognitive perceptions areas for future smoking prevention or cessation interventions
of risk, self-efficacy of changing behavior, and negative targeted to young adults. The present study addresses these
emotions such as fear of developing lung cancer [2]. gaps in the literature by examining the following research
However, these interventions do not focus on feelings question and hypotheses:
of risk per se.
Unfortunately, several factors act counter to the efforts of & RQ1: How are beliefs about the nature of nicotine addic-
public health agencies, which results in millions of continu- tion, the ease of quitting smoking, and perceptions of de-
ing smokers. One reason for the resistance of cigarette layed harm related to cognitive and affective constructs
smoking to tobacco control efforts is that the vast majority associated with quitting smoking, such as feelings of risk,
of smokers begin smoking when they are teenagers and worry, self-efficacy of quitting, and quit intentions?
young adults [1]. However, very little is known about feel- & H1: Feelings of risk will have a stronger relationship with
ings of risk among youth and young adults [16]. quit intentions than worry.
Furthermore, youth and young adults are uniquely vulner- & H2: Feelings of risk will have a stronger relationship with
able to becoming addicted to nicotine due to a variety of quit intentions than self-efficacy.
340 J Canc Educ (2016) 31:338–347

Methods Analysis Plan

This paper describes the secondary analysis of data collected Preliminary Analyses Participants who reported smoking
for the purpose of testing the effect of genetic risk information even a puff on some or every day of the 30 days prior to the
on smoking-related beliefs [16]. That study reported that experiment were considered smokers. Smokers were further
informing college-aged smokers that there was a genetic basis classified into those who smoked less than one cigarette per
for nicotine addiction did not affect beliefs about nicotine day vs. those who smoked at least one per day. Pearson and
addiction, quitting, or the immediacy of harm. Nor did it affect Spearman correlations were used to identify potential con-
other social-cognitive variables related to quitting smoking founding variables. Details about the exploratory factor anal-
(i.e., self-efficacy, feelings of risk, worry about the harms of ysis that led to the scales and subscales described in Table 1
smoking, and intentions to quit). can be found in [16].

Main Analyses The research question and hypotheses were

Participants examined using a series of hierarchical linear regressions.
Beliefs about addiction and quitting and beliefs about delayed
Participants were recruited from undergraduate psychology harm were analyzed separately. The dependent variables were
classes at the George Washington University (GWU) and the feelings of risk, worry about becoming ill due to smoking,
University of Alabama (UA) during the Fall 2010 and Spring self-efficacy of quitting, and intentions to quit smoking.
2011 semesters. A total of 746 students completed the survey, Each analysis was identical such that only the dependent var-
of whom 278 reported not having smoked even a puff of a iable differed across the regressions. Thus, step 1 of each
cigarette in the last 30 days (i.e., never or former smokers). analysis contained the covariates, which were uniform across
This study focuses only on the smokers in the sample. Of the analyses and included experimental condition (dummy cod-
392 smokers who began the survey, 333 (85.0 %) were eligi- ed), school, gender, race, and number of cigarettes smoked
ble for analysis. Six participants were excluded because they daily (<1 vs. 1+). Step 2 added items assessing either addic-
had participated in the study previously, 43 because they an- tion and quitting beliefs, or perceptions of delayed harm.
swered the comprehension verification item inaccurately, and Worry was added in step 3, followed by feelings of risk (step
17 did not check a box indicating they consented to partici- 4) and self-efficacy of quitting (step 5).
pate. Seven participants were excluded for more than one
reason.
Results

Procedure and Measures Sample Characteristics

Participants completed the study using the Internet-based As noted previously [16], most of the 333 individuals who
survey administration platform SurveyMonkey. After completed the study were aged 18–22 (97 %; n=322; M=
consenting, participants provided information about their 19.3, SD =2.0), were women (66.4 %; n=221), and were
demographics and past smoking behavior. Demographics Caucasian (85 %; n=283). On average, participants smoked
included age, race, gender, and school attended. Smoking few cigarettes daily (M=2.7, SD=4.3), with more than half
behavior was assessed with the following items: “Within smoking less than one cigarette per day (58.6 %; n=195).
the last 30 days did you smoke…? [Not at all, not even a Over half of participants were considered “established”
puff/Some days/Every day]”; “Have you smoked at least smokers; that is, they had smoked at least one cigarette on at
100 cigarettes in your entire life? [Yes/No]”; “On average, least one of the last 30 days and had smoked at least 100
how many cigarettes do you now smoke per day?”. cigarettes during their lifetimes (53 %; n = 177).
Participants were then randomly assigned to read one of Approximately one-third (34.5 %, n=115) of those who had
three news articles described in [16]. Next, they completed smoked 100 cigarettes in their lifetime had made at least one
survey items assessing comprehension of the information unsuccessful quit attempt. The main analysis includes only the
and beliefs about genetic determinism. Beliefs about nico- 306 participants who provided complete data on all the vari-
tine addiction and quitting, feelings of risk, worry about ables of interest.
experiencing harm from smoking, self-efficacy of quitting,
and intentions to quit were assessed. See Table 1 for word- Main Analysis
ing of items. We debriefed participants and provided them
with smoking cessation resources. The complete survey Step Demographic and smoking characteristics. Table 2
can be obtained from the corresponding author. 1 contains detailed information about the relationships
J Canc Educ (2016) 31:338–347 341

Table 1 Survey items by construct (in italics) and subscale

Addiction and quittinga

Amount to addiction (α=.82) “People can become addicted to cigarettes even if they only smoke one or two cigarettes a day.”
“People can become addicted to cigarettes even if they only smoke one or two cigarettes a week.”
“People can become addicted to cigarettes even if they only smoke occasionally with friends.”
Can avoid addiction (α=.62) “If I have not yet become addicted to nicotine then I never will.”
“If signs of nicotine addiction haven’t appeared by the time someone smokes a TOTAL of 5
packs of cigarettes, they will never become addicted.”
“If people who smoke have good genes, they won’t get addicted to nicotine.”
“There’s not much risk in smoking during college because people have plenty of time to quit.”
Judgment (α=.64) “People who say they are addicted to nicotine just don’t have the willpower to quit smoking.”
“People who are addicted to nicotine are emotionally weak.”
“People don’t become addicted to nicotine unless they enjoy smoking.”
“People who are addicted to nicotine are not in control of their lives.”
Determinism1 (never addicted-reverse scoredb) “Some people will never become addicted to nicotine, no matter how much or how long they smoke.”
Determinism2 (unavoidable addictionb) “Addiction to nicotine is an unavoidable consequence of smoking cigarettes.”
Agentic1 (complete controlb) “People who smoke have complete control over their nicotine addiction.”
Agentic2 (really want to stopb) “People can stop smoking if they really want to.”
Agentic3 (willpowerb) “Willpower is the best way to quit smoking.”
Delayed harma (α=.76) “There is usually no risk to this person’s health at all for the first few years.”
“Although smoking may eventually harm this person’s health, there is really no harm from
smoking the very next cigarette.”
“There is usually no risk of nicotine addiction for the first few years.”
Quitting self-efficacy acdefg
(α=.86) “Whether or not I quit smoking cigarettes completely in the next month would be entirely up to me.”
“How much personal control would you feel you have over quitting smoking completely in the next
month?”
“Quitting cigarettes in the next month would be…”
“How confident would you be that you could quit smoking completely in the next month if you really
wanted to?”
“How certain would you be that you could quit smoking completely in the next month if you
really wanted to?”
“How difficult would it be for you to quit smoking cigarettes in the next month?”
Worryh (α=.87) “How much do you worry about…getting lung cancer in the future?”
“How much do you worry about…having a heart attack in the future?”
“How much do you worry about…becoming addicted to nicotine within in the next year?”
Feelings of riska (α=.81) “I feel that I will develop lung cancer in the future if I continue to smoke.”
“I feel that I will have a heart attack in the future if I continue to smoke.”
“I feel that I will get addicted to nicotine in the next year if I continue to smoke.”
Quit intentionsa “I intend to quit smoking within the next year.”

Response options all included 5-point Likert-type scales. The anchors changed based on the question
a
Strongly disagree–strongly agree
b
These items were intended to be components of two scales titled Deterministic Addiction (never addicted, unavoidable addiction) and Agentic Quitting
(complete control, really want to stop, willpower), but their internal consistency was limited
c
No control–total control
d
Not at all possible–completely possible
e
Not at all confident–completely confident
f
Not at all certain–completely certain
g
Not at all difficult–extremely difficult
h
Not at all worried–extremely worried

between demographic and smoking characteristics and associated with higher feelings of risk and higher
the outcomes of interest. In brief, being a woman was worry. Being non-white was also associated with
Table 2 Addiction and quitting beliefs (n=306)a
342

Intentions Feelings of risk Worry Self-efficacy

Variables entered β p R2 (F change, β p R2 (F change, β p R2 (F change, β p R2 (F change,

sig. F change) sig. F change) sig. F change) sig. F change)

Step 1 0.05 (F=2.47, p=0.024) 0.09 (F=4.65, p=0.000) 0.22 (F=14.31, p=0.000) 0.09 (F=4.78, p=0.000)
Demographic and smoking characteristicsb
School −0.06 0.36 −0.02 0.69 0.06 0.26 −0.02 0.78
Experimental condition
A vs. C 0.04 0.53 −0.10 0.11 0.00 0.96 0.03 0.62
B vs. C 0.06 0.41 0.01 0.90 0.03 0.57 0.00 0.94
Gender 0.09 0.12 0.22 0.00 0.22 0.00 0.03 0.65
Race −0.03 0.62 −0.03 0.60 −0.18 0.001 0.03 0.61
Number of cigarettes −0.15 0.01 0.17 0.003 0.40 0.00 −0.29 0.000
Step 2 0.08 (F=1.31, p=0.24) 0.15 (F=2.91, p=0.004) 0.24 (F=0.56, p=0.81) 0.20 (F=4.89, p=0.000)
Addiction beliefs
Covariates were included in each step of the analysis but are not shown here for clarity of presentation.
Amount to addiction 0.06 0.35 0.08 0.15 0.07 0.22 0.06 0.28
Can avoid addiction −0.06 0.33 0.01 0.86 0.00 0.98 −0.07 0.28
Judgment 0.05 0.44 −0.05 0.43 0.01 0.90 0.04 0.44
Determinism1 (never addicted) −0.05 0.46 −0.13 0.03 0.04 0.44 0.11 0.07
Determinism2 (unavoidable) −0.05 0.38 −0.16 0.01 −0.07 0.22 0.06 0.33
Agentic1 (control addiction) −0.02 0.76 −0.02 0.77 0.02 0.68 0.18 0.00
Agentic2 (quit if really want to) 0.13 0.05 0.05 0.42 −0.04 0.45 0.16 0.01
Agentic3 (willpower) 0.02 0.71 0.02 0.77 0.02 0.75 0.04 0.53
Step 3 0.09 (F=3.69, p=0.06)
Worry
Amount to addiction 0.05 0.42
Can avoid addiction −0.06 0.32
Judgment 0.05 0.44
Determinism1 (never addicted) −0.05 0.41
Determinism2 (unavoidable) −0.05 0.45
Agentic1 (control addiction) −0.02 0.72
Agentic2 (quit if really want to) 0.13 0.04
Agentic3 (willpower) 0.02 0.74
Worry 0.12 0.06
Step 4 0.11 (F=6.15, p=0.014)
Feelings of risk
Amount to addiction 0.04 0.51
Can avoid addiction −0.07 0.31
J Canc Educ (2016) 31:338–347
Table 2 (continued)

Intentions Feelings of risk Worry Self-efficacy

Variables entered β p R2 (F change, β p R2 (F change, β p R2 (F change, β p R2 (F change,

sig. F change) sig. F change) sig. F change) sig. F change)

Judgment 0.06 0.36

Determinism1 (never addicted) −0.03 0.68
Determinism2 (unavoidable)
J Canc Educ (2016) 31:338–347

−0.02 0.69
Agentic1 (control addiction) −0.02 0.78
Agentic2 (quit if really want to) 0.12 0.06
Agentic3 (willpower) 0.02 0.75
Worry 0.04 0.56
Feelings of risk 0.17 0.01
Step 5 0.12 (F=2.81, p=0.10)
Self-efficacy
Amount to addiction 0.03 0.61
Can avoid addiction −0.06 0.36
Judgment 0.05 0.40
Determinism1 (never addicted) −0.04 0.56
Determinism2 (unavoidable) −0.03 0.66
Agentic1 (control addiction) −0.04 0.56
Agentic2 (quit if really want to) 0.10 0.12
Agentic3 (willpower) 0.02 0.81
Worry 0.06 0.42
Feelings of risk 0.18 0.01
Self-efficacy 0.11 0.10
a
Includes only the 306 participants who provided complete data
b
Experimental conditions: A=There is a genetic link to nicotine addiction; B=There is not a genetic link; C=Attention control. Gender: 1=Men; 2=Women. Race: 1=Non-white; 2=White. Number of
cigarettes: 1=less than 1 per day; 2=1 or more per day
343
344 J Canc Educ (2016) 31:338–347

higher worry. Smoking more than one cigarette per day Discussion
was associated with higher feelings of risk, higher
worry, lower quitting self-efficacy, and lower inten- Strong prior correlational evidence for the importance of cog-
tions to quit. These significant demographic and nitive and affective risk perceptions and self-efficacy in
smoking characteristic covariates were controlled for explaining smoking-related outcomes indicates that that they
in the remaining steps. are important to target in risk communications [3–5, 14].
Step Correlates of feelings of risk, worry, self-efficacy, and However, information about the underlying beliefs driving
2 intentions. There were mixed results for the these constructs is scant, despite their possible role in guiding
relationships between addiction beliefs and future smoking prevention and cessation interventions. The
perceptions of delayed harm and the cognitive and results of the present study supported our hypotheses that
affective variables of interest. None of the feelings of risk would be more strongly associated with quit
addiction or quitting beliefs (Table 2) or perceived intentions than worry about becoming ill due to smoking and
delayed harm measures (Table 3) were associated self-efficacy of quitting. This suggests that interventions that
with worry about experiencing the harms of increase feelings of risk may motivate quit attempts and re-
smoking. However, smokers had lower feelings of duce tobacco use. This finding adds to the growing body of
risk if they endorsed beliefs that addiction was evidence suggesting that feelings of risk are stronger predic-
inevitable or of they believed that the harms of tors of health behavior and/or behavioral intentions than some
cigarette smoking would be delayed. None of the social cognitive and affective constructs [12–15]. It is also
other addiction belief variables were associated consistent with several theoretical perspectives emphasizing
with feelings of risk. Self-efficacy was also asso- the important role of feelings in influencing perceptions of risk
ciated with few addiction beliefs. Two of the three and health decision making [9, 10]. The reasons why feelings
items assessing agentic beliefs about quitting pre- would be more influential or motivating than cognitive risk
dicted higher quitting self-efficacy, but none of the perceptions should be investigated in future research.
remaining addiction beliefs or perceptions of de- Our results indicate that the belief that addiction is an in-
layed harm items were associated with self-effi- evitable consequence of cigarette smoking is associated with
cacy. In this step, intention to quit was associated lower feelings of risk. It could be that smokers, who may feel
only with smoking less than one cigarette per day; personally threatened by addiction, disregard the health threat
neither addiction beliefs nor perceptions of de- to reduce the threatening feeling. It is therefore important that
layed harm were related to quit intentions. public health messages counter the possible negative effects of
Steps Cognitive and affective correlates of intentions to messages that strongly emphasize the addictiveness of nico-
3–5 quit. The roles of worry, feelings of risk, and self- tine by providing information about smoking cessation strat-
efficacy in predicting intentions to quit were ex- egies [6, 7]. Moreover, believing that the harms of smoking
amined in steps 3–5. Adding worry to the model are delayed was associated with lower feelings of risk. This
that included addiction beliefs did not account for suggests that smokers may believe they will quit before they
a statistically significant amount of increased var- experience harm. Hence, it is important for tobacco control
iance in intentions (Table 2, step 3). Furthermore, messages to emphasize the urgency of quitting earlier rather
worry was not associated with quit intentions. than later.
However, feelings of risk did account for a statis- The limited associations found between the remaining ad-
tically significant amount of increased variance diction and quitting beliefs variables and feelings of risk and
(Table 2, step 4). Higher feelings of risk were intentions to quit indicate that they may not be optimal targets
associated with higher intentions to quit. Self-ef- for intervention. It also suggests that risk feelings may origi-
ficacy, on the other hand, was not associated with nate from some source other than beliefs about addiction,
quit intentions after feelings of risk had been possibly more visceral or experiential in nature (e.g., the ex-
accounted for (Table 2, step 5), nor did it account tent to which people are able to imagine themselves getting
for increased variance in intentions. The same lung cancer) [24, 25]. Future research efforts and public health
pattern of findings can be seen for the analyses practice would benefit from investigating these other sources
examining perceptions of delayed harm. Worry of addiction and quitting beliefs.
was not related to intentions to quit (Table 3, step
3), but higher feelings of risk were associated with Limitations and Future Directions
higher quit intentions (Table 3, step 4). However,
self-efficacy was only marginally related to inten- As with all cross-sectional surveys, the directionality of the
tions once feelings of risk were accounted for relation between feelings of risk and beliefs about the nature
(Table 3, step 5). of addiction cannot be determined. Longitudinal and
Table 3 Delayed harm beliefs (n=306)a

Intentions Feelings of risk Worry Self-efficacy

J Canc Educ (2016) 31:338–347

Variables entered β p R2 (F change, sig. F change) β p R2 (F change, sig. F change) β p R2 (F change, sig. F change) β p R2 (F change, sig. F change)

Step 1 The findings for step 1 in this table are identical to those described in Table 2, step 1.
Demographic and smoking
characteristics
Step 2 0.06 (F=2.40, p=0.12) 0.10 (F=4.81, p=0.03) 0.22 (F=0.35, p=0.56) 0.1 (F=3.48, p=.06)
Delayed harm
Covariates were included in each step of the analysis but are not shown here for clarity of presentation.
Delayed harm −0.09 0.12 −0.12 0.03 −0.03 0.56 −0.10 0.06
Step 3 0.07 (F=3.52, p=0.06)
Worry
Delayed harm −0.08 0.14
Worry 0.12 0.06
Step 4 0.09 (F=8.05, p=0.005)
Feelings of risk
Delayed harm −0.06 0.25
Worry 0.03 0.67
Feelings of risk 0.19 0.00
Step 5 0.1 (F=3.90, p=.05)
Self-efficacy
Delayed harm −0.05 0.39
Worry 0.05 0.52
Feelings of risk 0.21 0.00
Self-Efficacy 0.12 0.05
a
Includes only the 306 participants who provided complete data
b
Experimental conditions: A=There is a genetic link to nicotine addiction; B=There is not a genetic link; C=Attention control. Gender: 1=Men; 2=Women. Race: 1=Non-white; 2=White. Number of
cigarettes: 1=less than 1 per day; 2=1 or more per day
345
346 J Canc Educ (2016) 31:338–347

experimental research should explore this issue further. References

Additional research is also needed to determine the condi-
tions under which high feelings of risk lead to actual quit 1. US Department of Health and Human Services (2014) The health
attempts or smoking cessation among smokers. The role of consequences of smoking—50 years of progress. A report of the
surgeon general. Atlanta, GA: U.S. Department of Health and
feelings of risk in preventing initiation among young non-
Human Services, Centers for Disease Control and Prevention,
smokers should also be examined. Another consideration National Center for Chronic Disease Prevention and Health
is that cognitive measures of perceived risk (e.g., “How Promotion, Office on Smoking and Health
likely do you think you are to develop lung cancer”) were 2. Centers for Disease Control and Prevention (2014) Best Practices
for Comprehensive Tobacco Control Programs. Atlanta, GA
not included in the survey. Thus, it is not possible to make
3. Lipkus IM, Eissenberg T, Schwartz-Bloom RD, Prokhorov AV,
direct comparisons of the relative power of worry and cog- Levy J (2011) Affecting perceptions of harm and addiction among
nitive and affective perceived risk on intentions to quit. In college waterpipe tobacco smokers. Nicotine Tob Res 13(7):599–
addition, it is possible that smokers hold beliefs regarding 610
addiction and quitting that were not assessed in this study. 4. McCaul KD, Hockemeyer JR, Johnson RJ, Zetocha K, Quinlan K,
Glasgow RE (2006) Motivation to quit using cigarettes: a review.
Consequently, it is not possible to say that such beliefs Addict Behav 31(1):42–56
overall do not affect quitting intentions. It is also important 5. Gwaltney CJ, Metrik J, Kahler CW, Shiffman S (2009) Self-
to note that our sample was comprised primarily of efficacy and smoking cessation: a meta-analysis. Psychol Addict
smokers who were white, young, and smoked less than Behav 23(1):56–66
6. Rogers RW, Prentice-Dunn S (1997) Protection motivation theory.
one cigarette per day. This may limit the generalizability In: Gochman DS (ed) Handbook of Health Behavior Research I:
of the findings to other populations. Future research should personal and social determinants. Plenum Press, New York, pp
examine these research questions in more established 113–32
smokers and in more socioeconomically diverse popula- 7. Witte K (1992) Putting the fear back into fear appeals: the extended
parallel process model. Commun Monogr 59:329–49
tions. Future work should also systematically test educa- 8. Rosenstock IM (1974) Historical origins of the belief model. In:
tional strategies for enhancing feelings of risk. One possi- Becker MH (ed) The Health Belief Model and Personal Health
ble approach would be to convey information about the Behaviors. p. 1-8
harms of smoking via personal narratives, which are vivid, 9. Loewenstein GF, Hsee CK, Weber EU, Welch N (2001) Risk as
feelings. Psychol Bull 127(2):267–86
memorable, and increase feelings of risk [24, 25].
10. Finucane ML, Alhakami A, Slovic P, Johnson SM (2000) The
affect heuristic in judgments of risks and benefits. J Behav Decis
Mak 13(1):1–17
Implications 11. Janssen E, van Osch L, Lechner L, Candel M, de Vries H (2012)
Thinking versus feeling: differentiating between cognitive and af-
fective components of perceived cancer risk. Psychol Health 27(7):
This study adds to the evidence base that feelings of risk 767–83
are powerful predictors of behavioral intentions. It also 12. Dillard AJ, Ferrer RA, Ubel PA, Fagerlin A (2011) Risk perception
demonstrates that, at least in some circumstances, feelings measures’ associations with behavior intentions, affect, and cogni-
of risk related to cigarette smoking predict quit intentions tion following colon cancer screening messages. Health Psychol
13. Janssen E, van Osch L, de Vries H, Lechner L (2013) Examining
above and beyond worry and self-efficacy, two constructs direct and indirect pathways to health behaviour: the influence of
that have been shown to be important predictors of quit cognitive and affective probability beliefs. Psychol Health 28(5):
intentions and behavior. Gaining additional understanding 546–60
of what tobacco-related beliefs can increase feelings of risk 14. Janssen E, Waters EA, van Osch L, Lechner L, de Vries H (2012)
The importance of affectively laden beliefs about health risks: the
and incorporating those beliefs into tobacco control cam- case of tobacco use and sun protection. J Behav Med
paigns may help increase the effectiveness of such cam- 15. Weinstein ND, Kwitel A, McCaul KD, Magnan RE, Gerrard M,
paigns and reduce tobacco use. Gibbons FX (2007) Risk perceptions: assessment and relationship
to influenza vaccination. Health Psychol 26(2):146–51
16. Waters EA, Kincaid C, Kaufman AR, Stock MM, Peterson LM,
Muscanell NL et al (2014) Concerns about unintended negative
consequences of informing the public about multifactorial risks
may be premature for young adult smokers. Br J Health Psychol
19(4):720–36
Acknowledgments The results presented in this paper were pre- 17. US Department of Health and Human Services (2012) Preventing
sented as a poster at the 34th Annual Meeting of the Society of Behavioral tobacco use among youth and young adults: a report of the surgeon
Medicine in 2013. This research was supported by a Mentored Research general: U.S. Department of Health and Human Services, Centers
Scholar Grant awarded to EW by the American Cancer Society (ACS), for Disease Control and Prevention, National Center for Chronic
MRSG-11-214-01-CBBP, and the Barnes Jewish Hospital Foundation Disease Prevention and Health Promotion, Office on Smoking and
(BJHF). The funding agreement ensured the authors’ independence in Health
designing the study, interpreting the data, writing, and publishing the 18. Weinstein ND, Slovic P, Gibson G (2004) Accuracy and optimism in
report. The content is solely the responsibility of the authors and does smokers’ beliefs about quitting. Nicotine Tob Res 6(Supplement 3):
not necessarily represent the official view of the ACS or BJHF. S375–S80
J Canc Educ (2016) 31:338–347 347

19. Slovic P (ed) (2001) Smoking: risk, perception, and policy. Sage 23. Schane RE, Ling PM, Glantz SA (2010) Health effects of
Publications, Inc., Thousand Oaks, CA light and intermittent smoking: a review. Circulation
20. Arnett JJ (2000) Optimistic bias in adolescent and adult smokers 121(13):1518–22
and nonsmokers. Addict Behav 25(4):625–32 24. Janssen E, van Osch L, de Vries H, Lechner L (2013) The influence
21. Weinstein ND, Marcus SE, Moser RP (2005) Smokers’ unrealistic of narrative risk communication on feelings of cancer risk. Br J
optimism about their risk. Tob Control 14(1):55–9 Health Psychol 18(2):407–19
22. Slovic P (2000) What does it mean to know a cumulative risk? 25. Allen JA, Duke JC, Davis KC, Kim AE, Nonnemaker JM, Farrelly
Adolescents’ perceptions of short-term and long-term conse- MC (2014) Using mass media campaigns to reduce youth tobacco
quences of smoking. J Behav Decis Mak 13:259–66 use: a review. Am J Health Promot
Journal of Cancer Education is a copyright of Springer, 2016. All Rights Reserved.