Factors in Nonadherence to Quitline Services: Smoker Characteristics Explain Little

Article (PDF Available)inHealth Education & Behavior 39(5):596-602 · January 2012with11 Reads
DOI: 10.1177/1090198111425186 · Source: PubMed
Abstract
Background: Quitlines offer evidence-based, multisession coaching support for smoking cessation in the 50 U.S. states, Canada, and several other countries. Smokers who enroll in quitline services have, ipso facto, shown readiness to attempt to quit, but noncompletion of coaching services appears widespread and has not been widely investigated. The current study explored the magnitude and correlates of quitline service abandonment. Method: A state's quitline intake, coaching, and nicotine patch/gum utilization data were obtained for smokers who enrolled during the period July 2007 to June 2008 (n = 20,882). Analyses examined demographic, socioeconomic status, nicotine dependence-related, and nicotine replacement therapy--utilization factors associated with completion of only one coaching session (of five offered). Results: Almost half of enrollees (47.8%) completed only one session. All significant predictors together explained less than 4% of variance; not being sent nicotine replacement therapy was most strongly correlated with completion of only one session. A framework is proposed for directing research toward reducing quitline service nonadherence. Conclusions: Premature user abandonment of coaching calls is widespread within a quitline. Further research should determine the extent of the problem in national quitline systems, increase knowledge of mediators of nonadherence, and develop strategies for increasing coaching completion.
Health Education & Behavior
39(5) 596 –602
© 2012 Society for Public
Health Education
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1090198111425186
http://heb.sagepub.com
Telephone quitlines (QLs) provide tobacco cessation treat-
ment that is effective compared with minimal or no counseling
or self-help (odds ratio = 1.6, 95% confidence interval = 1.4-1.8),
and QL–medication combination therapy is more effective
than medication alone (odds ratio = 1.3, 95% confidence
interval = 1.1-1.6; Fiore et al., 2008). Optimal QL service tim-
ing and duration have not been established, but experimental
studies have found better outcomes with delivery of more
than one counseling or coaching session (Stead, Perera, &
Lancaster, 2007; Zhu et al., 1996). Similarly, quit rates are
higher among callers who self-select multiple sessions ver-
sus one session (25.6% vs. 16.1%, p < .001; Zhu, Melcer,
et al., 2000; Zhu, Tedeschi, et al., 2000).
Although national guidelines in the United States recom-
mend at least four sessions with 90 minutes total contact time
(Fiore et al., 2008), multisession QL coaching services are
rarely completed. In a trial of three versus five sessions, with
and without additional “booster” calls, only 1 in 10 study
participants completed all assigned sessions (Rabius, Pike,
Hunter, Wiatrek, & McAlister, 2007). The average number
of completed calls is often half or less of the available ses-
sions (An et al., 2006). In one California study, 21% of call-
ers (who also received nicotine replacement therapy [NRT])
discontinued after one counseling call (Zhu, Melcer, et al.,
2000; Zhu, Tedeschi, et al., 2000), and in another, approxi-
mately one third did not complete the first scheduled coun-
seling session (Zhu et al., 1996).
Studies of treatment nonadherence may be guided by a
World Health Organization (WHO, 2003) framework, which
was based on a critical review across chronic conditions.
The framework identifies five interrelated domains of adher-
ence factors: patient, condition, therapy, health system/health
care team, and social/economic context. Although not
explicitly guided by this framework, most of the consider-
able research on nonadherence has focused on patient factors
across a variety of medical treatments and contexts. Despite
this emphasis on patient factors, nonadherent patient sub-
populations remain hard to predict (Bosworth, Weinberger,
& Oddone, 2006). Demographics typically do not strongly
predict adherence, according to a meta-analysis across
diseases and conditions including diabetes, cardiovascular
425186HEBXXX10.1177/1090198111425186Bu
rns et al.Health Education & Behavior XX(X)
1Colorado School of Public Health
2University of Colorado Cancer Center
Corresponding Author:
Emily K. Burns, Colorado School of Public Health, University of Colorado
Denver, 13001 E. 17th Place MS F542, Aurora, CO 80045, USA
Email: emily.burns@ucdenver.edu
Factors in Nonadherence to
Quitline Services: Smoker
Characteristics Explain Little
Emily K. Burns, MD1, Arnold H. Levinson, PhD1,2,
and Elizabeth A. Deaton, MA2
Abstract
Background. Quitlines offer evidence-based, multisession coaching support for smoking cessation in the 50 U.S. states, Canada,
and several other countries. Smokers who enroll in quitline services have, ipso facto, shown readiness to attempt to quit, but
noncompletion of coaching services appears widespread and has not been widely investigated. The current study explored
the magnitude and correlates of quitline service abandonment. Method. A state’s quitline intake, coaching, and nicotine patch/
gum utilization data were obtained for smokers who enrolled during the period July 2007 to June 2008 (n = 20,882). Analyses
examined demographic, socioeconomic status, nicotine dependence-related, and nicotine replacement therapy—utilization
factors associated with completion of only one coaching session (of five offered). Results. Almost half of enrollees (47.8%)
completed only one session. All significant predictors together explained less than 4% of variance; not being sent nicotine
replacement therapy was most strongly correlated with completion of only one session. A framework is proposed for
directing research toward reducing quitline service nonadherence. Conclusions. Premature user abandonment of coaching
calls is widespread within a quitline. Further research should determine the extent of the problem in national quitline
systems, increase knowledge of mediators of nonadherence, and develop strategies for increasing coaching completion.
Keywords
quitline, smoking cessation, treatment adherence
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
Burns et al. 597
disease, cancer, and infectious disease (DiMatteo, 2004).
There are exceptions, though. In studies of antihypertensive
medication or medications in general among older patients,
nonadherence was more common among Blacks than among
Whites (Bosworth, Dudley, et al., 2006; Gerber, Cho,
Arozullah, & Lee, 2010).
Cognitive and psychosocial factors may also play a role
in treatment nonadherence. A systematic overview of
research on adherence to chronic disease treatment found
nonadherence related to issues with managing the treatment
regimen (such as forgetting), cognitive motivation, symp-
toms, and social support (Dunbar-Jacob et al., 2000).
The few studies of adherence to smoking cessation treat-
ment suggest that demographic, smoking, and quitting char-
acteristics may predict adherence. A randomized controlled
study of prescription smoking cessation medications identi-
fied younger age, not achieving abstinence for at least 2
weeks, and more prequit cigarettes per day (CPD) as predic-
tors of nonadherence (defined as not taking at least one dose
of the medication for at least 80% of days in the 12-week
treatment; Hays, Leischow, Lawrence, & Lee, 2010). A
population-based study of NRT use in Colorado found that
certain demographic subpopulations, including adults
younger than 45 years, non-Latino non-Whites, and people
with low socioeconomic status (SES; <200% poverty level)
were more likely to discontinue treatment for reasons other
than quitting smoking (Burns & Levinson, 2008). Other fac-
tors associated with nonadherence may include higher ciga-
rette consumption and nicotine dependence, failure to stop or
reduce smoking during treatment, and depression or other
psychiatric comorbidities (WHO, 2003). Finally, treatment
intensity may play a role; one intervention study found that
smokers who received “community care” consisting of a
self-help and referral had lower rates of initiating and con-
tinuing cessation medications compared with a group
receiving face-to-face counseling in an autonomy- and com-
petence-supportive manner (Williams et al., 2006).
To our knowledge, no study has investigated reasons for
QL service nonadherence. The current study used existing
data to explore available demographic, smoking, and quit-
ting characteristics as potential correlates of nonadherence to
QL coaching calls. Based on published studies, we expected
that lower SES, higher nicotine dependence and being
younger would be associated with higher rates of nonadher-
ence; the magnitude of the association as well as associations
of other covariates with QL nonadherence were largely
unknown based on little published evidence.
Method
Institutional review board approval was obtained for second-
ary analysis of one state’s existing data maintained by a
multi-state QL provider (National Jewish Health, Denver,
Colorado). The study period was July 2007-June 2008. The
study population included callers who enrolled in the QL
program during the study period, smoked cigarettes, and
were 18 years or older. Participants with more than one
enrollment during the period were limited to the most recent
enrollment.
The free coaching program included five sessions, and
enrollment was a prerequisite to receipt of free NRT. The
first, smoker-initiated QL call began with intake data collec-
tion (demographics, contact information), then usually tran-
sitioned directly into the first coaching session which
included collection of additional smoking and cessation his-
tory variables. A total of approximately 30 questions were
asked throughout this initial session, requiring at least 30
minutes to complete. Subsequent sessions were proactively
initiated by a coach at scheduled times based on participant
preference. Coaches made up to three call attempts to con-
duct a scheduled session; if no contact was made, the partici-
pant was disenrolled but could reactivate enrollment by
calling the QL. Free NRT was offered to all eligible enroll-
ees during the first call and mailed to the participant’s home
address; physician approval to receive NRT was required for
enrollees with heart conditions, uncontrolled hypertension,
or pregnant smokers. The amount of NRT was determined
by smoking level: 4-week supplies for 20 CPD, up to
8-week supplies (in two separate mailings) for >20 CPD.
Self-help materials were mailed to all enrollees.
Coaches were college graduates who were trained in a
3-week curriculum on addiction, tobacco cessation tech-
niques, motivational interviewing, customer service, and
working with different cultural and disparate populations.
Coaching content was based on Clinical Practice Guidelines
(Fiore et al., 2000), which support smokers to develop con-
fidence and coping skills for quitting. The initial coaching
session included setting a quit date, planning the cessation
process, and ordering NRT (if appropriate). Subsequent ses-
sion protocols were tailored to the enrollee’s stage of change
and status in the process of quitting. Prepare sessions com-
pleted and extended the quit plan, verified receipt and proper
use of NRT (if appropriate), and discussed coping strategies.
Support sessions included updating quit status and NRT use,
reinforcing NRT benefits, and addressing problem-solving
skills. Support/relapse sessions addressed barriers to long-
term abstinence, lessons from past quit attempts, and motiva-
tion for quitting again after relapse.
For the current study, descriptive and regression analyses
were conducted in 2009 using SAS 9.2 (SAS Institute, Inc.,
Cary, NC). Descriptive analysis compared proportions of
available characteristics with number of completed coaching
sessions (Table 1). For logistic regression analysis, the out-
come of interest was number of completed coaching ses-
sions, coded dichotomously as one vs. more than one session,
because almost half of participants completed only the initial
session. Potential predictors of completing only one session
included available demographic characteristics (age, sex, race/
ethnicity, sexual orientation, children in home, rurality); SES
(education, insurance status); smoking characteristics (CPD,
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
598 Health Education & Behavior 39(5)
Table 1. Percentage of Study Population Who Completed One to Five Calls Before Discontinuing Quitline Program by Demographics,
Smoker and Cessation History, and Nicotine Replacement Therapy Use
Percentage of Colorado Quitline Enrollees Who Complete No. of Total Calls
Sample Size (N)1 Call 2 Calls 3 Calls 4 Calls 5 Calls
Total 20,882 47.8 23.4 13.6 8.1 7.1
Gendera
Men 8,798 45.8 23.4 14.1 8.5 8.2
Women 12,084 49.2 23.4 13.2 7.8 6.4
Race/ethnicitya
White, non-Latino 14,260 47.3 23.9 13.8 8.1 6.8
Latino 4,314 46.7 22.7 13.6 8.4 8.6
Black 1,303 55.3 21.4 11.4 6.8 5.1
Other 864 49.0 22.6 13.1 8.1 7.3
Agea (years)
18-24 3,101 56.3 23.9 10.8 5.6 3.5
25-44 9,751 50.3 23.5 13.1 7.3 5.9
45-64 7,155 43.1 23.3 15.1 9.4 9.2
65+875 27.7 22.9 17.4 15.8 16.3
Gay, lesbian, or bisexual
Heterosexual 19,480 47.8 23.4 13.6 8.1 7.1
Gay, lesbian, or bisexual 915 49.4 23.0 12.6 7.9 7.2
Educationa
<High school 3,226 45.9 23.3 14.2 8.8 7.7
High school/GED 7,455 49.2 23.7 12.3 8.1 6.7
Some college/tech 5,955 47.9 23.2 14.2 7.8 6.9
College and/or graduate school 3,866 46.9 23.3 14.4 7.9 7.5
Insurance statusa
No insurance 7,945 49.6 23.6 13.1 7.5 6.2
Private insurance 8,531 47.0 22.8 14.4 8.3 7.6
Medicaid 2,678 48.3 25.4 11.8 7.7 6.8
Medicare 884 36.5 21.7 18.0 12.9 10.9
Children in homea
Ye s 9,167 50.5 23.3 13.1 7.2 5.9
No 11,506 45.6 23.5 13.9 8.8 8.1
Rural status
Ye s 3,457 46.3 24.6 13.9 8.8 6.5
No 17,406 48.1 23.2 13.6 8.0 7.2
Cigarettes per daya
20 15,459 49.4 23.7 13.4 7.3 6.3
>20 5,423 43.1 22.8 14.3 10.4 9.5
Time to smoke
Within 30 minutes 16,041 48.3 23.3 13.4 8.0 7.0
>30 minutes 4,542 46.2 24.1 13.9 8.3 7.5
Live with tobacco usera
Ye s 9,153 49.8 23.6 13.3 7.5 5.8
No 11,488 46.2 23.3 13.7 8.6 8.2
Number of lifetime quit attemptsa
0 1,883 53.2 23.0 12.6 5.7 5.4
1 3,122 48.9 22.1 13.1 8.1 7.8
2+15,547 46.9 23.8 13.7 8.4 7.1
Quit attempt in past year
Ye s 9,661 47.5 23.7 13.3 8.5 7.0
No 10,763 48.0 23.1 13.8 7.9 7.2
(continued)
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
Burns et al. 599
Percentage of Colorado Quitline Enrollees Who Complete No. of Total Calls
Sample Size (N)1 Call 2 Calls 3 Calls 4 Calls 5 Calls
Fax referral
Ye s 1,907 47.4 23.1 13.4 7.6 8.6
No 18,975 47.8 23.5 13.6 8.2 7.0
Motivation to quit
Motivated or very motivated 16,523 47.8 23.2 13.5 8.2 7.3
Somewhat or not very
motivated
1,931 49.9 23.6 13.1 7.6 5.9
Nicotine replacement therapy
senta
Ye s 18,043 44.9 24.3 14.5 8.7 7.6
No 2,839 65.8 18.1 8.0 4.4 3.8
a. Covariates for which chi-square p < .05.
Table 1. (continued)
time to first morning cigarette, living with a smoker, motiva-
tion to quit); cessation history (number of previous quit
attempts, quit attempts in the past year); origin of first QL
contact (self-initiated or fax referred by a health provider);
and whether NRT was mailed. Based on preliminary results
of NRT mailing analysis, a further post hoc analysis exam-
ined medical conditions (heart conditions, pregnancy) that
required physician approval to receive NRT. All contrasts
and tests used two-sided p < .05 as the significance
criterion.
A multivariate regression model was built by examining
each potential predictor (covariates with significant chi-
square associations from Table 1) separately and adding uni-
variately significant predictors in a forward stepwise manner.
All predictors with significance (p < .05) were retained in the
final full multivariate model.
Results
Study participants were more often female (57.9%), aged 25
to 44 years (46.7%) or 45 to 64 (34.3%), White (68.8%), and
nonrural residents (83.4%); the majority smoked 20 CPD
(74.0%), smoked within 30 minutes of waking (77.9%), had
multiple previous quit attempts (75.7%), and were very
motivated or motivated to quit (89.5%). Almost half lived
with another smoker (44.3%) and/or had children at home
(44.3%). Nearly half (47.8%) completed only the first
coaching session. Most were sent NRT (86.4%).
Characteristics significantly associated with number of
sessions included sex, race/ethnicity, age, education level,
insurance status, CPD, children in the home, living with a
tobacco user, number of lifetime quit attempts, and being
sent NRT (Table 1). Completion of only one call occurred
among more than half of the following groups: Black
(55.3%), callers aged 18-24 years (56.3%), callers with chil-
dren in the home (50.5%), those who had never tried to quit
(53.2%), and those who were not sent NRT (65.8%).
Logistic regression factors that univariately predicted
completion of only one session included not being sent NRT;
being 18 to 24 years old, female, or Black; having no more
than high school education; having no insurance, no previ-
ous quit attempts, or any children in the home; smoking 20
CPD, and living with a tobacco user (Table 2). None
accounted alone for more than 2% of variance. The final
multivariate regression model retained all of the initial sig-
nificant univariate predictors with little change in odds ratio
estimates for any predictors compared with univariate val-
ues; no interactions were present. The model had good fit
(Hosmer–Lemeshow goodness of fit p = .13) and explained
4.7% of variance.
Among enrollees sent NRT, 1.2% reported being preg-
nant (vs. 12.1% if not sent NRT, p < .0001), and 13.4%
reported high blood pressure or heart disease (vs. 24.7% if
not sent NRT, p < .0001).
Discussion
This large study of a state’s QL users found that almost half
of enrollees completed only the first of five available coach-
ing sessions. Being aged 18 to 24 years, Black, or smoking
20 or fewer cigarettes per day weakly predicted coaching
discontinuation; not being sent NRT was the strongest pre-
dictor. Other available demographic, socioeconomic, smok-
ing, and cessation history characteristics did not strongly
predict discontinuation; together all the available predictors
explained little of the overall variance of call completion.
Despite strong efficacy evidence and wide availability,
smoking cessation treatments are underutilized. The current
study adds to evidence that those who do initially adopt one
of these treatments may still receive only a fraction of the
potential benefit because of nonadherence, a problem that
is starting to be documented (Tinkelman, Wilson, Willett,
& Sweeney, 2007; Williams et al., 2006; Zbikowski,
Hapgood, Smucker Barnwell, & McAfee, 2008), but remains
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
600 Health Education & Behavior 39(5)
Table 2. Multivariate Logistic Regression Analysis of Completing One Call Versus Two to Five Calls by Covariates, Among Adults
Colorado Quitline Enrolleesa
Independent Variable Univariate OR 95% CI Multivariate OR 95% CI Multivariate p Value
Gender
Women vs. men 1.1 1.1-1.2 1.1 1.1-1.2 .0002
Age (years)
18-24 1.3 1.2-1.4 1.3 1.2-1.4 <.0001
25-44 Ref Ref Ref Ref
45-64 0.8 0.7-0.8 0.8 0.7-0.8
65+0.4 0.3-0.5 0.4 0.3-0.5
Race/ethnicity
White, non-Latino Ref Ref Ref Ref <.0001
Latino 1.0 0.9-1.1 1.0 0.9-1.0
Black 1.4 1.2-1.5 1.3 1.2-1.5
Other 1.1 0.9-1.2 1.0 0.9-1.2
Education level
<High school 0.9 0.8-1.0 0.9 0.8-0.9 .0148
High school/GED Ref Ref Ref Ref
Some college/tech 0.9 0.9-1.0 0.9 0.9-1.0
College and/or graduate school 0.9 0.8-1.0 1.0 0.9-1.0
Insurance status
No insurance 1.1 1.0-1.2 1.1 1.0-1.2 .0051
Private insurance Ref Ref Ref Ref
Medicaid 1.1 1.0-1.2 1.0 0.9-1.1
Medicare 0.7 0.6-0.8 0.9 0.8-1.1
Children in the home vs. not 1.2 1.2-1.3 1.1 1.0-1.2 .0066
Cigarettes per day
20 vs. >20 1.3 1.2-1.4 1.2 1.2-1.3 <.0001
Live with tobacco user vs. not 1.2 1.1-1.2 1.1 1.0-1.2 .0005
Lifetime quit attempts
0 1.2 1.1-1.4 1.2 1.0-1.3 <.0001
1 Ref Ref Ref Ref
2+0.9 0.9-1.0 0.9 0.8-1.0
Being sent no NRT vs. at least one NRT shipment 2.4 2.2-2.6 2.6 2.4-2.9 <.0001
Note. OR = odds ratio; CI = confidence interval; NRT = nicotine replacement therapy.
a. Full model R2 = .047.
underinvestigated in smoking cessation research. Study of
treatment adherence in general as well as the results of the
current study cast doubts on the WHO patient domain
(demographics, personal history, etc.) and socioeconomic
domain as fruitful areas of inquiry. Although socioeconomic
factors in particular were expected to be associated with non-
adherence, perhaps there is a selection effect based on the
study population having already called the QL. In other
words, socioeconomic status may be trumped by the charac-
teristics of the population that led them to initiate contact
with the QL in the first place.
Our current findings lead us to speculate that the condi-
tion domain, when defined as smoking dependence vari-
ability, may also be less informative given the small
magnitude of the association between cigarettes per day
and adherence. In addition, the finding that lighter smok-
ers were actually more likely to discontinue after one
coaching call was contrary to our proposed hypothesis.
Other studies show mixed results on nicotine dependence
and cessation success (Biener, Hamilton, Siegel, &
Sullivan, 2010; Hyland et al., 2004; Hymowitz et al., 1997;
Levy, Romano, & Mumford, 2005). Because of the lack of
association of nonadherence with nicotine dependence,
individually focused domains may need to be expanded to
include loss of motivation as well as perceived completion
(Table 3).
The WHO domains of therapy and health system/care
team, that is, coaching protocols, telephone–computer infra-
structure and coaches, remain promising areas of research
related to nonadherence to QL coaching that were not able to
be explored in this study of secondary data. Across these
domains, we propose that the most relevant categories of
inquiry include protocol failure to accommodate enrollee
needs/preferences and system failure, defined as deviance
from service protocols. Table 3 summarizes these four fac-
tors related to nonadherence to QL services; further research
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
Burns et al. 601
in these domains will delineate the relationships between the
domains, the prevalence and impact of each domain, and will
lead to interventions that will improve adherence to QL
coaching.
The positive findings of this study suggest initial places
for improvement in nonadherence. Coaching discontinuation
was more than twice as prevalent among enrollees who were
not mailed NRT. Slightly more than one fourth of this non-
NRT group comprised pregnant women or smokers with
heart conditions requiring physician approval. This finding
may be an example of protocol failure; further research
should determine the extent to which the physician consent
process poses an unnecessarily high treatment barrier in light
of safety evidence for NRT among most smokers with cardio-
vascular disease (Fiore et al., 2008; Ford & Zlabek, 2005).
Additionally, alternate protocols may need to be developed
for those who do not want or qualify for NRT. Another proto-
col failure related to NRT may be NRT distribution proto-
cols; perhaps the reason that lighter smokers more often
discontinued coaching calls was because they did not have
the possibility of receiving a subsequent supply of NRT like
the heavier smokers did. NRT distribution protocols may be
designed to maximize adherence to coaching calls while
simultaneously delivering NRT in cost-effective doses.
Young adults and Blacks had approximately 30% greater
odds of discontinuing coaching after one call. Enrollees older
than 44 years exhibited the exact opposite trend, and those
older than 64 years were half as likely as young adults to dis-
continue coaching after the first call. This finding supports
speculation that system failure might play a lesser role in QL
nonadherence, since such failures are unlikely to be age or
race related. Protocol failure might be a more helpful domain,
suggesting a potential need for alteration or flexibility in
coaching call timing, frequency, content, or delivery for young
adults and/or Black enrollees.
Limitations
The factors examined in this secondary data analysis
were based on responses to multiple choice or short-answer
questions without further probing. Coaching nonadherence
is probably influenced by factors that were not asked or are
difficult to measure, as described in detail throughout the
framework and discussion. Nevertheless, the results pro-
vide an important starting point for research and suggest
action steps even at this early stage.
Implications for Practice
Nonadherence to QL coaching is widespread, may occur
after just one session, and affects most subpopulations of
smokers. Campaigns to get smokers to call a QL may need
components specifically designed to promote adherence and
completion of coaching programs. For enrollees who need
medical approval for NRT, QLs should examine their proto-
cols and physician referral systems to ensure that they are
accessible and efficient. In states that link coaching with
NRT, protocols may need to be changed for enrollees who
do not want NRT. Finally, young adult, Black, and lighter
smokers who enroll in the QL may be more adherent to
coaching protocols designed specifically for these groups.
Conclusions
Nonadherence to QL coaching calls is significant across
almost all subpopulations of QL enrollees; further research
should explore if cessation outcomes are improved by
increasing adherence to QL coaching which is nationally
available. Furthermore studies should expand understanding
of mediators of nonadherence to coaching calls with the goal
to design interventions for the first call or shortly thereafter
to improve adherence to coaching calls.
Acknowledgments
The authors would like to thank National Jewish Health, Denver,
Colorado for provision of data.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Table 3. Factors in (Non)Adherence to Quitline Services
Category of Quitline Nonadherence Examples
System failure Nondelivery of promised medication
Delay or dropped call before connecting enrollee with a coach
Failure to initiate outreach call at the scheduled time
Lack of adherence to coaching protocols
Protocol failure Reliance on enrollee’s ability to pay for cell-phone minutes
Lack of protocol tailored to enrollee’s needs
Enrollee perceived completion Enrollee plans to maintain abstinence without further coaching
Enrollee loss of motivation Enrollee loses motivation to make or maintain quit attempt
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
602 Health Education & Behavior 39(5)
Funding
The authors disclosed receipt of the following financial support for
the research, authorship, and/or publication of this article:
Emily K. Burns was supported by a Pfizer Scholar Grant in
Public Health. The study was also supported by the Colorado
Tobacco Education, Prevention and Cessation Grant Program,
grant FLA08-00233. Neither funding entity was involved in any
stages of this manuscript production.
References
An, L. C., Schillo, B. A., Kavanaugh, A. M., Lachter, R. B.,
Luxenberg, M. G., Wendling, A. H., & Joseph, A. M. (2006).
Increased reach and effectiveness of a statewide tobacco
quitline after the addition of access to free nicotine replace-
ment therapy. Tobacco Control, 15, 286-293.
Biener, L., Hamilton, W. L., Siegel, M., & Sullivan, E. M. (2010).
Individual, social-normative, and policy predictors of smoking
cessation: A multilevel longitudinal analysis. American Jour-
nal of Public Health, 100, 547-554.
Bosworth, H. B., Dudley, T., Olsen, M. K., Voils, C. I., Powers, B.,
Goldstein, M. K., & Oddone, E. Z. (2006). Racial differences in
blood pressure control: Potential explanatory factors. American
Journal of Medicine, 119(1), 70.e9-70.e15.
Bosworth, H. B., Weinberger, M., & Oddone, E. Z. (2006). Intro-
duction. In H. B. Bosworth, E. Z. Oddone, & M. Weinberger
(Eds.), Patient treatment adherence: Concepts, interventions,
and measurement (pp. 3-12). Mahwah, NJ: Lawrence Erlbaum.
Burns, E. K., & Levinson, A. H. (2008). Discontinuation of nico-
tine replacement therapy among smoking-cessation attempters.
American Journal of Preventive Medicine, 34, 212-215.
DiMatteo, M. R. (2004). Variations in patients’ adherence to medi-
cal recommendations: A quantitative review of 50 years of
research. Medical Care, 42, 200-209.
Dunbar-Jacob, J., Erlen, J. A., Schlenk, E. A., Ryan, C. M.,
Sereika, S. M., & Doswell, W. M. (2000). Adherence in chronic
disease. In J. Goeppinger & J. J. Fitzpatrick (Eds.), Annual
review of nursing research: Focus on chronic illness (Vol. 18,
pp. 48-90). New York, NY: Springer.
Fiore, M., Bailey, W., Cohen, S., Dorfman, S., Goldstein, M., Gritz,
E., . . . Wewers, M. E. (2000). Treating tobacco use and depen-
dence. Clinical practice guideline. Rockville, MD: U.S. Depart-
ment of Health and Human Services, Public Health Service.
Fiore, M., Jaen, C., Baker, T., Bailey, W., Benowitz, N., Curry, S.,
. . . Wewers, M. E. (2008). Treating tobacco use and depen-
dence: 2008 update. Clinical practice guideline. Rockville,
MD: U.S. Department of Health and Human Services, Public
Health Service.
Ford, C. L., & Zlabek, J. A. (2005). Nicotine replacement therapy and
cardiovascular disease. Mayo Clinic Proceedings, 80, 652-656.
Gerber, B. S., Cho, Y. I., Arozullah, A. M., & Lee, S. Y. (2010).
Racial differences in medication adherence: A cross-sectional
study of Medicare enrollees. American Journal of Geriatric
Pharmacotherapy, 8, 136-145.
Hays, J. T., Leischow, S. J., Lawrence, D., & Lee, T. C. (2010).
Adherence to treatment for tobacco dependence: Association
with smoking abstinence and predictors of adherence. Nicotine
& Tobacco Research, 12, 574-581.
Hyland, A., Li, Q., Bauer, J. E., Giovino, G. A., Steger, C., &
Cummings, K. M. (2004). Predictors of cessation in a cohort of
current and former smokers followed over 13 years. Nicotine &
Tobacco Research, 6(Suppl. 3), S363-S369.
Hymowitz, N., Cummings, K. M., Hyland, A., Lynn, W. R.,
Pechacek, T. F., & Hartwell, T. D. (1997). Predictors of smok-
ing cessation in a cohort of adult smokers followed for five
years. Tobacco Control, 6(Suppl. 2), S57-S62.
Levy, D. T., Romano, E., & Mumford, E. (2005). The relation-
ship of smoking cessation to sociodemographic characteristics,
smoking intensity, and tobacco control policies. Nicotine &
Tobacco Research, 7, 387-396.
Rabius, V., Pike, K. J., Hunter, J., Wiatrek, D., & McAlister, A. L.
(2007). Effects of frequency and duration in telephone counselling
for smoking cessation. Tobacco Control, 16(Suppl. 1), i71-i74.
Stead, L. F., Perera, R., & Lancaster, T. (2007). A systematic
review of interventions for smokers who contact quitlines.
Tobacco Control, 16(Suppl. 1), i3-i8.
Tinkelman, D., Wilson, S. M., Willett, J., & Sweeney, C. T. (2007).
Offering free NRT through a tobacco quitline: Impact on utili-
sation and quit rates. Tobacco Control, 16(Suppl. 1), i42-i46.
Williams, G. C., McGregor, H. A., Sharp, D., Levesque, C.,
Kouides, R. W., Ryan, R. M., & Deci, E. L. (2006). Testing a
self-determination theory intervention for motivating tobacco
cessation: Supporting autonomy and competence in a clinical
trial. Health Psychology, 25, 91-101.
World Health Organization. (2003). Adherence to long-term thera-
pies: Evidence for action. Geneva, Switzerland: Author.
Zbikowski, S. M., Hapgood, J., Smucker Barnwell, S., & McAfee,
T. (2008). Phone and web-based tobacco cessation treatment:
Real-world utilization patterns and outcomes for 11,000 tobacco
users. Journal of Medical Internet Research, 10(5), e41.
Zhu, S. H., Melcer, T., Sun, J., Rosbrook, B., & Pierce, J. P. (2000).
Smoking cessation with and without assistance: A population-
based analysis. American Journal of Preventive Medicine, 18,
305-311.
Zhu, S. H., Stretch, V., Balabanis, M., Rosbrook, B., Sadler, G.,&
Pierce, J. P. (1996). Telephone counseling for smokingcessa-
tion: Effects of single-session and multiple-session interventions.
Journal of Consulting and Clinical Psychology, 64, 202-211.
Zhu, S. H., Tedeschi, G., Anderson, C. M., Rosbrook, B.,
Byrd, M., Johnson, C. E., & Gutiérrez-Terrell, E. (2000).
Telephone counseling as adjuvant treatment for nicotine
replacement therapy in a “real-world” setting. Preventive Medi-
cine, 31, 357-363.
at UNIV OF CO HEALTH SCIENCE CTR on October 4, 2012heb.sagepub.comDownloaded from
  • [Show abstract] [Hide abstract] ABSTRACT: Background Internet and telephone treatments for smoking cessation can reach large numbers of smokers. There is little research on their costs and the impact of adherence on costs and effects. Objective To conduct an economic evaluation of The iQUITT Study, a randomised trial comparing Basic Internet, Enhanced Internet and Enhanced Internet plus telephone counselling (‘Phone’) at 3, 6, 12 and 18 months. Methods We used a payer perspective to evaluate the average and incremental cost per quitter of the three interventions using intention-to-treat analysis of 30-day single-point prevalence and multiple-point prevalence (MPP) abstinence rates. We also examined results based on adherence. Costs included commercial charges for each intervention. Discounting was not included given the short time horizon. Results Basic Internet had the lowest cost per quitter at all time points. In the analysis of incremental costs per additional quitter, Enhanced Internet+Phone was the most cost-effective using both single and MPP abstinence metrics. As adherence increased, the cost per quitter dropped across all arms. Costs per quitter were lowest among participants who used the ‘optimal’ level of each intervention, with an average cost per quitter at 3 months of US$7 for Basic Internet, US$164 for Enhanced Internet and US$346 for Enhanced Internet+Phone. Conclusions ‘Optimal’ adherence to internet and combined internet and telephone interventions yields the highest number of quitters at the lowest cost. Cost-effective means of ensuring adherence to such evidence-based programmes could maximise their population-level impact on smoking prevalence.
    Full-text · Article · Sep 2012
  • [Show abstract] [Hide abstract] ABSTRACT: The use and effectiveness of tobacco quitlines by weight is still unknown. This study aims to determine if baseline weight is associated with treatment engagement, cessation, or weight gain following quitline treatment. Quitline participants (n = 595) were surveyed at baseline, 3 and 6 months. Baseline weight was not associated with treatment engagement. In unadjusted analyses, overweight smokers reported higher quit rates and were more likely to gain weight after quitting than obese or normal weight smokers. At 3 months, 40 % of overweight vs. 25 % of normal weight or obese smokers quit smoking (p = 0.01); 42 % of overweight, 32 % of normal weight, and 33 % of obese quitters gained weight (p = 0.05). After adjusting for covariates, weight was not significantly related to cessation (approaching significance at 6 months, p = 0.06) or weight gain. In the first quitline study of this kind, we found no consistent patterns of association between baseline weight and treatment engagement, cessation, or weight gain.
    Article · Sep 2013
  • [Show abstract] [Hide abstract] ABSTRACT: Objective To study whether demographic and smoking-related characteristics are associated with participation (reach) in a smoking cessation trial and subsequent use (uptake) of two specific smoking interventions (internet-based program and proactive telephone counseling). Methods We used data from a four-arm randomized smoking cessation trial (2011). Participants (n = 1,809) were recruited among 9,924 smokers who previously participated in two health surveys in Denmark (2007-2008 and 2010). Interventions were: 1) an internet-based smoking cessation program 2) proactive telephone counseling, 3) reactive telephone counseling, 4) self-help booklet. Results Reach (defined as the proportion accepting to participate in the trial of those invited) was highest among persons aged 40-59, women, heavy smokers and persons with long education. Among trial participants, uptake (defined as any use of the specific intervention at one-month follow-up) was 69% for the internet-based program, 74% and 9 % for proactive and reactive telephone counseling, and 84% for the self-help booklet. Young age was associated with uptake of the internet-based program and short education was associated with using proactive telephone counseling. Conclusions Internet-based interventions and proactive telephone counseling appeal to different age and educational groups. Further, offering similar intervention content by a proactive and a reactive approach can be associated with different intervention uptake.
    Full-text · Article · May 2014
  • [Show abstract] [Hide abstract] ABSTRACT: Most telephone quitlines provide free nicotine replacement therapy (NRT). An 8-week course is recommended, but few users complete it. Information is needed to help quitlines distribute NRT cost-effectively. Randomised two-group trial. Colorado QuitLine callers who smoked 16-20 cigarettes per day at enrolment and who were eligible for and agreed to receive free NRT. Provision of 4-week versus 8-week NRT supply; the 8-week supply was shipped in halves and required participants to request the second half (split-shipment protocol). Enrolment occurred during March 2010-February 2011, follow-up concluded in November 2011, and analysis was performed in 2012. Point abstinence (7 and 30 day) and prolonged abstinence (6 month) from tobacco use. Overall, 1495 study participants were enrolled and 57.7% completed follow-up. Abstinence rates did not differ significantly between study conditions: 13.8% versus 12.4% in 4-week versus 8-week arms, respectively, (30-day point abstinence, non-respondents treated as smokers). NRT duration was similar in both groups, due in part to purchase of additional patches in the 4-week group. About one-third of the 8-week group requested the full 8-week supply and had higher abstinence rates. Cost per quit was lower in the 4-week (compared to 8-week) group. A randomised trial did not find worse cessation outcomes among quitline users who received half the minimum recommended course of NRT, but offering the full recommended course using a split-shipment protocol may be reasonably cost-effective and supportive of NRT adherers. NCT01889771. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.
    Full-text · Article · Nov 2014