Nurse Education Today 48 (2017) 48–54

Contents lists available at ScienceDirect

Nurse Education Today

journal homepage: www.elsevier.com/nedt

Tobacco prevention and reduction with nursing students:

A non-randomized controlled feasibility study
Anneke Bühler, PhD a,⁎, Katrin Schulze a, Christa Rustler b, Sabine Scheifhacken b,
Ines Schweizer c, Mathias Bonse-Rohmann, Prof PhD d
a
IFT Institut für Therapieforschung, München, Parzivalstr. 25, D-80804, München, Germany
b
Deutsches Netzwerk rauchfreie Krankenhäuser (DNrfK), Pettenkoferstrasse 16-18, D-10247, Berlin, Germany
c
Hochschule Esslingen, Flandernstr. 101, D-73732 Esslingen a.N., Germany
d
Hochschule Hannover, Blumhardtstraße 2, D-30625, Hannover, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Background: Prevalence of tobacco use among nurses and nursing students is disproportionally high in Germany.
Received 12 May 2016 However, from a public health perspective they are considered to be an important group for delivering smoking
Received in revised form 29 August 2016 cessation interventions. As delivery of tobacco-related treatment depends on own smoking status, smoking pre-
Accepted 19 September 2016 vention and cessation among the nursing professions is indicative for improving nurse and public health.
Objective: To evaluate the feasibility and effects of a comprehensive tobacco prevention and reduction program
Keywords:
on psychosocial and environmental factors related to smoking behavior of nursing students.
Nursing students
Methods: Between 2014 and 2015, a non-randomized, controlled feasibility study was conducted in 12 schools of
Tobacco use
Smoking nursing with 397 nursing students in Germany. Students in the intervention group received a program (ASTRA)
Prevention consisting of an introductory session, steering committee workshop, stress prevention lessons, evidence-based
Cessation smoking cessation intervention, and action project. Six months after baseline assessment, change in smoking-
Feasibility studies related protective and risk factors was determined. Secondary endpoints included smoking behavior.
Results: The program was implemented in total in 5 of 7 intervention schools. About one third of smoking nursing
students participated in a cessation intervention. The program seems to do better than a minimal intervention
booklet in four primary outcomes: perceived descriptive, subjective, and injunctive norms towards smoking
and nursing as well as perceived social support. As anticipated, there was no change in smoking behavior.
Conclusions: The applied approach is feasible and able to improve important smoking-related norm perceptions
of student nurses and perception of social support. However, additional context measures to influence the set-
tings of nursing education currently rather supporting smoking seem to be necessary in order to promote
smoking cessation among nursing students and to scale up implementation of the program.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction From a public health perspective, nurses are an extraordinary target

To stop or not to start smoking is a significant if not the most effec-
tive health-related decision a person can take. Smokers are estimated
to die ten to 11 years earlier than non-smokers (Doll et al., 2004;Pirie
et al., 2012). Quitting smoking at any age is beneficial. The earlier this
is accomplished the more substantial are reductions in overall mortality
and morbidity at any given age compared to continuous smokers (Doll
et al., 2004; Pirie et al., 2012).
⁎ Corresponding author.
E-mail addresses: buehler@ift.de (A. Bühler), schulze@ift.de (K. Schulze),
christa.rustler@rauchfrei-plus.de (C. Rustler), sabine.scheifhacken@rauchfrei-plus.de
(S. Scheifhacken), Ines.Schweizer@hs-esslingen.de (I. Schweizer),
mathias.bonse-rohmann@hs-hannover.de (M. Bonse-Rohmann).
group when it comes to tobacco-related intervention. They play an “in-
strumental role” in tobacco reduction among the general population
(Schultz, 2003, p.571) as nurses' advice is effective in supporting pa-
tients to quit (Rice et al., 2013). At the same time, prevalence of smoking
is high among the nursing population itself. In Germany, rates range
from 31% in health care nurses to 42% in elderly care nurses
(Statistisches Bundesamt, 2014). Nurses mostly initiate smoking before
they start their nursing education (Kolleck, 2004). It is estimated that
about every second nursing student uses tobacco at the beginning of
their vocational career (Hirsch et al., 2010; Vitzthum et al., 2013).
Given the fact that own smoking behavior impedes cessation support
by nursing staff (Vitzthum et al., 2013; Sarna et al., 2014) any interven-
tion which leads to a lower proportion of smoking nurses will improve
nurse and public health.

http://dx.doi.org/10.1016/j.nedt.2016.09.008
0260-6917/© 2016 Elsevier Ltd. All rights reserved.
 A. Bühler et al. / Nurse Education Today 48 (2017) 48–54
Tobacco control in Germany compares unfavorably to many European
countries (Schaller and Pötschke-Langer, 2012). Only one tobacco pre-
vention and cessation measure with nursing students has been initiated
and evaluated (Rapp et al., 2006). By learning brief cessation counseling
in school, nursing students were expected to quit smoking. The applied
approach improved student nurses' medical knowledge and promoted
their competence in giving advice to smokers, but had no effect on their
own smoking behavior (Rapp et al., 2006). Reviving the Federal Drug
Comissioner's statement about the importance of health care profes-
sionals in tobacco control (Die Drogenbeauftragte, 2008), the Federal
Health Ministry supported the development of an intervention to prevent
and reduce tobacco consumption among nursing students. This manu-
script describes the intervention ASTRA and reports results from the fea-
sibility study.
2. Background
In order to effectively change psychosocial and environmental fac-
tors related to smoking behavior in the short-term and prevent and re-
duce nursing students' smoking behavior in the long-term, the design of
the intervention is based on research and best-practice from several
areas.
2.1. Environmental Measures
Currently, nursing education in Germany is perceived as a context
characterized by smoking-friendly norms and regulations (Kolleck,
2004). Social-cognitive models of behavior emphasize the importance
of perceived descriptive, subjective and injunctive norms as well as self-
efficacy to be able to behave in a certain way (Bandura, 1989; Ajzen,
1991). If smoking is perceived as being common and accepted in a
given setting, then initiation and maintenance of tobacco use is more
probable (Riou França et al., 2009; Freedman et al., 2012). In contrast,
perceived incompatibility of professional self-concept as nurse and
smoking indicates cognitive dissonance towards smoking which is asso-
ciated with less smoking behavior (Pericas et al., 2009). Finally, in a
smoking-restricted context students may be likelier to believe they can
resist tobacco use and therefore smoke less. Resistance self-efficacy pre-
dicts smoking cessation in young adults (Cengelli et al., 2012).
An audit tool developed by the ENSH-Global Network for Tobacco
Free Health Care Services has been adapted for the ASTRA workshop
to guide the way to smoke-free nursing education (ENSH-Global,
2011). Introduction of the ENSH-process has been shown to produce a
substantial increase in smoke-free policy at hospitals and service vol-
ume of smoking cessation (Huang et al., 2015).
According to substance abuse prevention research, increased stu-
dent participation, improved relationships and positive school ethos is
associated with reduced drug use (Fletcher et al., 2008). Thus the
ASTRA action component gives nursing students the opportunity to ac-
tively improve their school social and physical environment and make it
a healthier place.
2.2. Psychosocial Measures
Turning to the behavioral level, ASTRA takes into account evidence
which points to the importance of stress coping skills and break man-
agement for successful smoking cessation among nurses (Sarna et al.,
2005; Perdikaris et al., 2010). Activating social support seems to be cru-
cial. Reeve et al. (2013) report that nursing students experience high
levels of anxiety, worry and depression in response to stress and cope
through utilizing social support from fellow nursing students rather
than from faculty staff. In their review on stress prevention measures
Galbraith and Brown (2011) conclude that the most successful profes-
sional interventions with student nurses are theory-based and incorpo-
rate skills to enable cognitive reappraisal of maladaptive cognitions, as
well as relaxation. In line with these conclusions, the theoretical basis
49
of ASTRA's stress prevention component is Lazarus and Folkman's
(1984) work on stress, reappraisal and coping.
The most effective smoking cessation strategies with young adults
were found to be cognitive-behavioral and social-cognitive approaches
as well as motivation enhancement (Villanti et al., 2010). In order to
overcome well known recruitment difficulties (Bühler and Thrul,
2012), ASTRA offers three formats based on these approaches: group
format, quit line, and web-based cessation.
This background guided development of the intervention as well as
its evaluation in the feasibility study. The aim of the evaluation was to
explore change processes initiated by ASTRA. Evaluation outcomes
need to be realistic (EMCDDA, 2011). Considering the current context
of nursing education which rather supports tobacco use, primary out-
comes were psychosocial and environmental factors related to smoking
behavior rather than smoking behavior itself.
3. Method
3.1. Intervention Model, Content and Delivery Method
Fig. 1 depicts the intervention model of the ASTRA program. ASTRA
aims at changing intermediate, proximal and distal predictors of smoking
behavior through five environmental or psychosocial components.
The ASTRA program implemented in nursing schools during the first
year of school consists of an introductory session, a steering committee
workshop, four stress prevention lessons including after care text mes-
sages, an offer of three evidence-based smoking cessation interventions,
and an action project (www.astra-programm.de). The introductory ses-
sion (90 min) informs all stakeholders about the program, gives general
and nursing-specific information about smoking and aims at promoting
motivation for a smoke-free nursing education. In the ASTRA workshop
(120 min) school administration, teachers, practice mentors reflect on
the school's tobacco and health policy along the ten ASTRA tool dimen-
sions (adapted from ENSH-Global, 2011). Five goals are priorized and
three goals are set to be achieved during the next three months. In the
stress prevention lessons (adapted from Müller, 2013) students analyze
their stressful situations and social support resources. Three coping
strategies, self-rewarding techniques, and assertive behavior are trained
in simulated nursing and smoking situations. In addition, a buddy sup-
port system is introduced and participants are sent supporting text
messages during their following practice semester. Students are offered
evidence-based cessation interventions in terms of a group format
(Wenig et al., 2013), helpline (Lindinger et al., 2012) and online-
program (Zeidler and Kleiber, 2016). Students not attending the
smoking cessation group participate in the action project. Here, the
topic health promotion is addressed and school-specific environmental
health or smoke-free initiatives are developed and implemented by the
students (e.g. new lounge area for break time). The program compo-
nents were delivered by program developers during school lessons.
3.2. Design
A three wave non-randomized controlled trial with waiting list con-
trol condition was conducted in the three project regions in Germany
(around Munich, Berlin and Hamburg, and Esslingen). A convenience
sample of 12 nursing schools with N = 584 first year students agreed
to participate in the study when asked because of their proximity to
study centers and/or because of established cooperation. In order to
reach a heterogeneous sample, health care and elderly care nursing
schools, small and large schools as well as urban and rural schools
were approached. Schools chose to whether be part of the intervention
group (IG) or control group (CG) with the option of receiving the inter-
vention after final data assessment of first study phase. All first year stu-
dents were eligible for study participation.
Finally, four treatment conditions were compared pre- to post-test
(T0-T2) in this feasibility study. CG students took part in the assessment
50 A. Bühler et al. / Nurse Education Today 48 (2017) 48–54
ASTRA Smoking-related
Introductory Perceived social,
session subjective, injunctive norms
Workshop Perceived social support
Stress prevention Stress coping self-efficacy
Smoke-free break self-
Cessation
efficacy
Smoking policy
Action project
endorsement
Fig. 1. Intervention model of the ASTRA program.
and were provided with a booklet about smoking and smoking cessa-
tion. IG 1 group schools planned to implement the complete ASTRA pro-
gram but only about half of the components were realized. This was due
to (1) school difficulty to find a time slot for the cessation group and
parallel action project and (2) lack of interest of smokers to participate
in the cessation group despite repeated recruitment efforts in class,
through letter or personal approach. IG2 group schools planned and im-
plemented ASTRA in total during the first intervention phase. IG3
schools implemented all components of the ASTRA program during in-
tervention phase 2. For the second phase ASTRA components had been
revised based on formative evaluation results in phase 1: (a) practice
mentors, clinic/ward administration, doctors are explicitly invited to
the introductory session, (b) in the ASTRA tool more simple language
is used and its design is more clearly arranged, (c) the sequence of the
stress prevention sessions is changed and examples focus more on nurs-
ing than on smoking situations (d) a maximum of one stress session per
day, and (e) the smoking cessation course is scheduled during school
hours.
3.3. Procedure
In schools, research staff collected informed consent from students.
Data assessment was conducted in the class room and lasted for
30 min. Staff was responsible for the distribution, help in completion
and collection of the questionnaire. To permit matching T0 and T2
data while assuring anonymity, students generated an individual code.
Finally, all questionnaires were placed in an envelope which was sealed
in front of the class. All study procedures were approved by the ethics
commission of the German Psychological Society.
Project staff implemented the ASTRA program between September
2014 and May 2015 and documented which components were realized
in a log book.
3.4. Measures
Following the intervention model (Fig. 1), the analyses relied on the
following outcome measures. Descriptive norm perception was
assessed by asking students how high they estimate smoking preva-
lence among their fellow nursing students, peers and graduated nurses
(3 items, α = 0.55). Perceived subjective norm was measured using
four items (e.g. “Smoking is accepted at nursing school”, α = 0.79).
Three items assessed perceived injunctive norm, e.g. “Smoking during
work or in a health care center is incompatible with my self-
conception as a nursing student”, α = 0.81). Confidence in receiving so-
cial support throughout nursing education was estimated based on an-
swers to three items (e.g. “How confident are you to receive sufficient
factors
Smoking behavior
Motivation to
quit or reduce
Resistance
self-efficacy
support through your fellow students during your education?”, α =
0.72). Self-efficacy to cope with stress was assessed by single item:
“How confident are you to cope with future work stress?” Similarly, stu-
dents were asked to rate their self-efficacy concerning smoke-free work
breaks in four items (e.g. “How confident are you to spend a smoke-free
break at nursing school?”, α = 0.76). Resistance self-efficacy towards
cigarette use in different situations was measured using seven items
(e.g. “How confident are you to resist smoking a cigarette if you need
a short time-out at work?”, α = 0.95). Finally, students indicated their
attitude towards a strict smoking policy at work by answering two
items (e.g. “Smoking should be prohibited on all clinic/ward premises”,
r = 0.48). All mentioned items had Likert-type answering categories
(scale 1 to 5).
Although not a primary outcome, smoking behavior was assessed
via number of smoking days during the last 30 days and mean number
of cigarettes smoked on a typical smoking day. Students reporting cur-
rent cigarette use were considered smokers. A quantity-frequency
index was calculated ((days of cigarette use × mean number of ciga-
rettes per smoking day)/30) indicating average cigarette use per day
(Kraus et al., 2013). Motivation to change was measured asking “How
motivated are you to change your smoking behavior, i.e. stop or reduce
smoking?” Students ticked one of the following categories: not motivat-
ed (0), rather not motivated (1), rather motivated (2), very motivated
(3).
3.5. Sample Size, Baseline Equivalence and Attrition Analysis
Reaching 68% of eligible first year students, the baseline sample con-
sists of n = 397 students with n = 55 in IG1, n = 79 in IG2, n = 31 in IG3,
and n = 232 in CG (Fig. 2). On average, first year students were
22.2 years old (SD = 6.4, Min = 16, Max = 54). The majority of students
was female (75.6%, n = 300) and currently smoking (52.9%, n = 210).
Among smokers, mean cigarette use per day was M = 7.7 (SD = 6.4)
and motivation to change was at a medium level (M = 1.3 (SD =
0.88)). Baseline equivalence of the four treatment conditions was tested
(Table 1). With regard to socio-demographics and smoking-related var-
iables, the only significant difference is age with IG2 and IG3 members
being older than CG members (Table 1). School level scores in a school
smoking policy checklist (Bühler and Piontek, 2015) did not differ signif-
icantly between treatment conditions (data not shown).
It was possible to collect and merge post-test data of n = 227 stu-
dents from 10 schools (57% retention rate). Dropout rates varied signif-
icantly between groups (CHI2(3) = 23.0, p b 0.0001) being higher in CG
and IG1 than in IG2 and IG3 (Table 1). Reasons for dropout were exclu-
sion of two CG schools from analysis because of low enrollment at T0
(b30% of students), failure of matching T0-T2 data, absence on
 A. Bühler et al. / Nurse Education Today 48 (2017) 48–54 51

Eligible (n=584, 12 schools)

Excluded (n=187) (32%)

Enrollment Declined to participate (n=12)
Not present (n=175)

Non-randomized (n= 397)

Allocation /T0
Allocated to intervention (n=165) Allocated to control (n=232)
IG1 (n=55, 2 schools) CG (n=232, 5 schools)
IG2 (n=79, 3 schools)
IG3 (n=31, 2 schools)

Post-Test (T2)
Lost to follow-up (n=38) due to discontinuation
of training (n=8), not present (n=22), no further
interest (n=3) or no matching (n=5)
Lost to follow-up (n=122) due to
discontinuation of training (n=10), not present
(n=58), 2 schools with low enrollment at T0
(n=18), no further interest (n=8) or no matching
(n=29)

Analysis
Analysed (n=117, 70%)) Analysed (n=110, 47%)
IG1 (n=25, 2 schools, 45%) CG (n=110, 3 schools)
IG2 (n=68, 3 schools, 86%)
IG3 (n=24, 2 schools, 77%)

Fig. 2. Flow chart.

assessment day, discontinuation of training, and no further interest to
participate in the study. Differential attrition from IGs and CG was sig-
nificant only in respect of age with IG3 members dropping out being
much older than CG members (Table 1).
3.6. Analyses
To test whether hypothesized change processes were initiated in the
intervention groups, linear and logistic regressions were computed
predicting T2-scores by treatment group controlled for corresponding
T0-score and age. Reference group was CG. The STATA statistical pack-
age was utilized. In case of significant findings significance level (p),
and effect size (η2) is reported. According to Cohen et al. (2003) an η2
of 0.02 is considered a small effect and an η2 of 0.13 is considered a me-
dium effect.
4. Results
4.1. Implementation
The introductory session and stress prevention were implemented
at all seven IG schools. The workshop took place at six IG schools. The
smoking cessation group and the action project were realized at five
schools. Thus, the complete ASTRA program was feasible at five of

Table 1
Baseline equivalence (T0) and attrition analysis.

CG (n = 232) IG1 (n = 55) IG2 (n = 79) IG3 (n = 31) Test

Subsample retention drop-out retention drop-out retention drop-out retention drop-out Contrasts baseline Dropout
% (n) 47.4 (110) 52.6 (122) 45.4 (25) 54.6 (30) 86.1 (68) 13.9 (11) 77.4 (24) 22.6 (7) (p) (p)

Women
in % (n) 77.1 (84) 75.4 (92) 72.0 (18) 70.0 (21) 80.9 (55) 90.9 (10) 58.3 (14) 85.7 (6) 0.1561 0.5111
Age IG1 vs. CG 0.5692 0.9353
Mean (SD) 21.3 (5.2) 22.0 (4.8) 20.4 (3.6) 21.2 (6.1) 23.9 (8.5) 26.5 (7.3) 22.0 (7.6) 30.0 (14.8) IG2 vs. CG 0.0222 0.3823
[Range] [16–42] [16–48] [17–29] [17–47] [17–49] [17–41] [17–53] [16–54] IG3 vs. CG 0.0012 0.0093
Smokers
in % (n) 57.3 (63) 48.4 (59) 40.0 (10) 46.7 (14) 57.4 (39) 54.6 (6) 62.5 (15) 57.1 (4) 0.3721 0.9391
QFI4 IG1 vs. CG 0.5802 0.1673
Mean (SD) 7.1 (6.0) 6.9 (6.0) 4.2 (4.1) 8.0 (7.4) 8.7 (6.2) 8.3 (5.5) 12.2 (9.1) 9.6 (3.9) IG2 vs. CG 0.6072 .9623
[Range] [0.04–25] [0.03–20] [0.1–11] [0−20] [0.08–20] [2–16] [0.58–35] [6–15] IG3 vs. CG 0.4112 .5223
Motivation to IG1 vs. CG 0.1212 0.5223
change
Mean (SD) 1.4 (0.9) 1.4 (1.0) 1.2 (0.7) 1.0 (0.7) 1.4 (0.8) 1.0 (0.6) 1.1 (0.8) 1.3 (1.0) IG2 vs. CG 0.2202 0.3323
[Range] [0–3] [0–3] [0–2] [0–2] [0–3] [0–2] [0–3] [0–2] IG3 vs. CG 0.7252 0.7883
1
CHI2-Test;
2
Anova Effect Treatment group;
3
Anova Interaction Effect Group × Drop-out;
4
QFI = Quantity-Frequency-Index.
52 A. Bühler et al. / Nurse Education Today 48 (2017) 48–54

seven nursing schools (71%). For each component, ASTRA trainers re-
ported to have taught 80 to 100% of component content. However, the
sequence and timeline of implementation had to be adapted in some
cases.
The smoking cessation group format was utilized by n = 45 first year
nursing students (n = 32 in IG2 and n = 13 in IG3) as documented by
trainers. In the retention sample n = 27 students reported to have par-
ticipated (n = 22 in IG2 and n = 5 in IG3). Relative to n = 64 T0-
smokers of the retention sample this would be a recruitment rate of
42% (27/64). In relation to n = 88 smokers of the baseline sample, the
recruitment rate would be 31% (27/88).
4.3. Smoking-Related Behavior
Although descriptive smoking rates developed quite differently in
the four conditions, e.g. from 40% to 48% in IG1 and 62.5% to 54.2% in
IG3, there are no significant differences between CG and all three IGs.
Students of CG and IG1 who smoked at the beginning of the study re-
ported more cigarettes per day at T2 whereas no increase was observed
in IG2 and IG3. However, these descriptive results did not reach statisti-
cal significance.
5. Discussion

In this controlled feasibility study we found that a comprehensive

4.2. Changes in Primary Outcomes
On descriptive level, changes in primary outcomes were small. How-
ever, students in the four treatment groups differed significantly in
some risk and protective factors of smoking behavior (Table 2). Com-
pared to the descriptive norm perception of control students, members
of IG1 and IG2 perceived less tobacco use among their fellow students
after the intervention phase (b = 0.16 (CI: 0.01–0.31), p b 0.035,
η2 = 0.01 and b = 0.19 (CI: 0.08–0.30), p b 0.001, η2 = 0.04). In addi-
tion, after the intervention IG2 students estimated acceptance and pos-
sibility of smoking to be less pronounced in their nursing school and
clinic/ward than CG students, indicating a stronger decrease in subjec-
tive norms (b = − 0.39 (CI: .-0.57- -0.21), p b 0.0001, η2 = 0.07).
With regard to injunctive norms, control students endorsed the opinion
of the incompatibility of smoking behavior and being a health profes-
sional to a lesser extent at T2 whereas IG2 students even increased
their opinion of incompatibility (b = 0.49 (CI: 0.13–0.74), p = 0.003,
η2 = 0.02). Finally, perceived social support decreased substantially
among control students and less among members of IG2 and IG3
(b = 0.33 (CI: 0.08–0.57), p = 0.008, η2 = 0.03 and b = 0.35 (CI:
0.001–0.70), p = 0.043, η2 = 0.02). There was no differential change
in self-efficacy variables and attitude towards strict smoking policy.
There were no gender differences in effects as well as no differential
intervention effects depending on initial smoking status of the partici-
pant (data not shown).
tobacco prevention and reduction program (ASTRA) can be implement-
ed in nursing schools. The ASTRA program comprising environmental
and psychosocial elements did better than a minimal intervention
booklet in four primary outcomes: perceived descriptive, subjective,
and injunctive norms towards smoking and nursing as well as perceived
social support. However, the effect sizes were small and ASTRA couldn't
be realized in total by two of seven schools. As anticipated, no effects on
smoking behavior were observed.
5.1. Outcomes in Relation to the Goals and Rationale of the ASTRA Program
We preliminarily conclude that stated goals were reached for the
time being. It was possible to successfully implement a program and
hereby initiate change processes. These are supposed to ultimately
lead to the reduction of smoking prevalence among nursing students
in the long-term. Despite a general denormalising process of smoking
behavior in Germany, momentarily there is still an accepting if not
supporting climate of cigarette use in nursing education. Nine in ten
nursing students in our sample perceived that smoking was accepted
at nursing school or clinic/ward (Bühler et al., 2016). In addition, work-
ing conditions for nurses have deteriorated during the last years (DBfK,
2015), assumably making smoking cessation even more difficult
(Eriksen, 2005). Finally, the ethical question about nurse's autonomy
and personal choice in health behaviors and professionally expected
role modelling (Blake, 2013) hasn't yet been answered explicitly by

Table 2
Primary outcome measures and smoking-related behavior at baseline (T0) and post-test (T2).

CG (n = 110) IG1 (n = 25) IG2 (n = 68) IG3 (n = 24) Intervention effect3

T0 T2 T0 T2 T0 T2 T0 T2 IG1 IG2 IG3

Descriptive 2.0 (0.4) 2.0 (0.4) 1.7 (0.5) 1.7 (0.4) 2.0 (0.3) 1.8 (0.4) 2.1 (0.3) 2.0 (0.3) 0.17 (0.01–0.32) 0.19 (0.08–0.30) -0.05 (−0.21–0.10)
Norm1 p = 0.035 p = 0.001 p = 0.543
Subjective 4.2 (0.7) 4.0 (0.5) 4.1 (0.9) 4.1 (0.5) 4.2 (0.6) 3.6 (0.7) 4.0 (0.7) 3.8 (0.6) 0.12 (−0.13–0.37) -0.40 (−0.58-−0.22) -0.18 (−0.44–0.07)
Norm1 p = 0.365 p b 0.0001 p = 0.161
Injunctive 2.8 (1.3) 2.5 (1.2) 2.5 (1.4) 2.5 (1.3) 2.5 (1.2) 2.8 (1.2) 2.2 (1.1) 1.9 (1.1) 0.02 (−0.40–0.45) 0.46 (0.16–0.76) -0.31 (−0.75–0.12)
Norm1 p = 0.920 p = 0.003 p = 0.160
Social support1 3.9 (0.7) 3.4 (0.9) 3.9 (0.5) 3.5 (0.8) 3.9 (0.8) 3.8 (0.8) 3.9 (0.5) 3.8 (0.7) 0.02 (−0.32–0.37) 0.33 (0.09–0.57) 0.36 (0.01–0.71)
p = 0.897 p = 0.008 p = 0.043
Stress coping 3.9 (0.8) 3.9 (1.0) 4.0 (0.7) 4.2 (0.8) 3.8 (0.9) 3.9 (0.8) 3.9 (0.7) 4.0 (0.8) 0.17 (−0.20–0.53) -0.05 (−0.31–0.21) -0.02 (−0.34–0.39)
self-efficacy1 p = 0.366 p = 0.714 p = 0.900
Smokefree 3.5 (1.1) 3.6 (1.0) 3.9 (0.9) 4.2 (0.9) 3.3 (1.1) 3.4 (1.1) 3.1 (0.9) 3.3 (0.8) 0.29 (−0.04–0.63) -0.06 (−0.29–0.17) -0.13 (−0.46–0.20)
break p = 0.083 p = 0.620 p = 0.440
self-efficacy1
Resistance 3.7 (1.3) 3.8 (1.2) 3.6 (1.3) 4.0 (1.2) 3.5 (1.4) 3.6 (1.4) 3.6 (1.3) 3.6 (1.4)
0.26 (−0.10–0.62) -0.03 (−0.29–0.22) 0.00(−0.39–0.38)
Self-efficacy1 p = 0.165 p = 0.798 p = 0.981
Strict smoking 3.0 (1.3) 2.8 (1.2) 2.6 (1.1) 2.5 (1.3) 2.8 (1.2) 2.7 (1.3) 2.6 (1.2) 2.6 (1.0) -0.06 (−0.48–0.35) -0.02 (−0.31–0.27) 0.00 (−0.43–0.41)
policy1 p = 0.759 p = 0.893 p = 0.966
Smoker2 57.1 (63) 55.4 (61) 40.0 (10) 48.0 (12) 57.4 (39) 55.9 (38) 62.5 (15) 54.2 (13) 2.5 (0.53–11.7) 1.1 (0.33–3.34) 0.49 (0.11–2.22)
p = 0.243 p = 0.933 p = 0.360
QFI1 7.3 (6.0) 7.7 (6.3) 4.2 (4.1) 5.8 (5.5) 8.7 (6.2) 8.6 (6.3) 12.2 (9.1) 11.7 (7.8) 0.25 (−2.2–2.7) -1.2 (−2.7–0.29) 0.22 (−1.8–2.3)
p = 0.842 p = 0.114 p = 0.824
Motivation 1.4 (0.9) 1.3 (0.9) 1.2 (0.7) 1.3 (0.8) 1.4 (0.8) 1.1 (0.8) 1.1 (0.8) 0.7 (0.8) -0.14 (−0.68–0.41) -0.114 (−0.46–0.24) -0.43 (−0.89–0.02)
to change1 p = 0.614 p = 0.524 p = 0.060
1
unadjusted M (SD),
2
% (n),
3
Regression coefficient/Odds Ratio of predictor Treatment group controlled for T0 and Age (95%-Confidence Interval) with reference group CG.
 A. Bühler et al. / Nurse Education Today 48 (2017) 48–54
the German nursing community. The new generation of nursing stu-
dents finds itself in a situation without clear professional guidance
with regard to health-related decisions. Thus, rather modest goals
were set in terms of influencing empirically sound risk and protective
factors of smoking behavior rather than smoking behavior itself which
seems appropriate in order to remain realistic (EMCDDA, 2011) and to
avoid frustration among stakeholders.
Nursing students who completed all original ASTRA components
were more likely to perceive less smokers around them (descriptive
norm) and smoking to be less accepted at school and clinic or ward
(subjective norm) after the intervention. In addition, they were more
convinced that smoking and nursing are incompatible (injunctive
norm). To be able to influence social norms is a relevant outcome as per-
ception of norms was shown to be a significant factor in uptake and ces-
sation of smoking among young adults (Freedman et al., 2012; Pericas
et al., 2009; Riou França et al., 2009). Furthermore, ASTRA students
were more confident they would receive social support during the
time at nursing school. Perceived social support has been discussed as
a protective factor in coping with stress among nursing students
(Reeve et al., 2013).
It is noteworthy that the factors in which some change was observed
were proximal to the intervention and rather distal to smoking behavior
(see Fig. 1). The question remains why motivation to quit or reduce
smoking didn't change as self-efficacy in coping with stress, in resisting
smoking in work-related situations, and in spending a smoke-free break
had not changed either. These were prominent issues in the ASTRA be-
havioral training. A ceiling effect might be the explanation with regard
to the self-efficacy beliefs. Especially non-smoking students started
out with the very firm expectation that they would be able to resist
smoking and spend a smoke-free break so that an additional benefit
might not have been observable in the short-term.
In contrast, motivation to quit or reduce smoking remained low
among smokers even though a substantial proportion of smokers (con-
servative estimate: 30% of smokers) joined the cessation group inter-
vention. Possibly, the context characterized by positive smoking
norms and negative quitting norms may counteract motivation to quit
(Dohnke et al., 2011). This would call for a norm shaping campaign
within the nursing professions in general including addressing the eth-
ical issue of nurse's personal freedom of choice and professional role
modelling of health behaviors. Specifically, the ASTRA program would
need to expand and provide more intensive clinic- or ward-oriented
components in addition to interventions at nursing schools ensuring
both settings of nursing education are targeted. Due to the lessons
learned with regard to the importance of context factors, in the current
second ASTRA study we focus on norm building within the nursing pro-
fessions in order to facilitate implementation and improve effectiveness
of the ASTRA program.
5.2. Strengths and Limitations
The major strengths of this study are its focus on a neglected group
of smokers who in addition are assigned a pivotal role in tobacco-
related public health; Its attempt to implement a comprehensive pro-
gram in a smoking-supporting context; Its explicit intervention
model; Its reliance on evidence-based components where possible; Its
longitudinal and controlled design; As well as use of modern statistical
techniques, and its realistic goal-setting with regard to the initiation of
empirically sound mediating processes of smoking behavior.
This study suffers from obvious limitations. The non-randomized de-
sign with self-selection into intervention groups and post-hoc grouping
of intervention schools endanger internal validity and prohibit firm ef-
fectiveness conclusions. It might well be that schools who decided to
implement ASTRA in full or in part differed systematically from the con-
trol schools in other relevant characteristics than tested for in baseline
equivalence analysis. These eventual differences might have influenced
the favorable development of their students rather than participation in
53
ASTRA. High drop-out restricts external validity of the results. Two con-
trol schools were excluded from analysis because study enrollment was
so low and thus the risk of selectivity bias very high. If at all, generaliz-
ability of findings would be appropriate to rather younger than older
nursing students. However, given the scarceness of international and
national approaches, and the novelty of an approach combining envi-
ronmental and psychosocial elements, the intervention first required a
non-randomized feasibility study. To encounter bias, eventual baseline
differences were controlled for in statistical analyses. Also, schools
were grouped depending on program implementation dosage and con-
tent to approach intervention similarity within comparison groups.
Nevertheless, it has to be stressed that a randomized, controlled clinical
trial with a standardized implementation of the program is needed to
test the efficacy or effectiveness of the ASTRA program.
6. Conclusions
The ASTRA program appears to be an intervention that can success-
fully be implemented in the heterogeneous landscape of German nurs-
ing schools. Although this study has its limitations, the positive
outcomes for nursing students in a challenging context, the scientifically
sound intervention model, and the incorporated evidence-based ele-
ments justify further development of the program and facilitation of
broad implementation. Program effects on smoking behavior via psy-
chosocial and environmental factors should be evaluated in a next
study with longer follow-up periods.
Funding
This study was funded by the German Federal Ministry of Health
(ZMVI5-2515DSM202) which had no role in the study design; the col-
lection, analysis or interpretation of the data; the writing of the paper,
or the decision to submit this paper for publication.
Contribution
All authors made substantial contributions to the conception and de-
sign of the study, or acquisition of data, or analysis and interpretation of
data, drafting the article or revising it, and approved the submitted
version.
Acknowledgments
We thank the nursing schools, teachers and nursing students who
participated in this research. We thank Jens Kalke and Hermann
Schlömer for help in recruiting schools and Tessa-Virginia Hannemann
for proof-reading the manuscript.
References
Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50,
179–211.
Bandura, A., 1989. Human agency in social cognitive theory. Am. Psychol. 44, 1175–1184.
Blake, H., 2013. Should nurses be role models for health? https://www.nursingtimes.net/
holly-blake-should-nurses-be-role-models-for-health/5052877.article.
Bühler, A., Piontek, D., 2015. Entwicklung einer Checkliste zur Erfassung der schulischen
Tabakpolitik. Das Gesundheitswesen S1, S112–S113.
Bühler, A., Thrul, J., 2012. Rauchstopp bei jungen Rauchenden. Grundlagen und interven-
tion. Sucht 58, 297–316.
Bühler, A., Schulze, K., Rustler, C., Scheifhacken, S., Schweizer, I., Bonse-Rohmann, M.,
2016. Geschlechterbezogene Analyse des Tabakkonsums bei Auszubildenden in
Pflegeberufen. Sucht 62, 73–81.
Cengelli, S., O'Loughlin, J., Lauzon, B., Cornuz, J., 2012. A systematic review of longitudinal
population-based studies on the predictors of smoking cessation in adolescent and
young adult smokers. Tob. Control 21, 355–362.
Cohen, J., Cohen, P., West, S., Aiken, L., 2003. Applied Multiple Regression/Correlation
Analysis for the Behavioral Sciences. Third edition. New York, Routledge.
Die Drogenbeauftragte der Bundesregierung, 2008. Frauen und Rauchen: Neue Wege in
der Prävention. http://www.fact-antitabak.de/pdf/Handlungsempfehlungen%20-%
20deutsch.pdf.
54 A. Bühler et al. / Nurse Education Today 48 (2017) 48–54
Deutscher Berufsverband für Pflegekräfte (DBfK), 2015. Zahlen - Daten - Fakten Pflege.
https://www.dbfk.de/media/docs/download/Allgemein/Zahlen-Daten-Fakten-
Pflege-2015-03.pdf.
Dohnke, B., Weiss-Gerlach, E., Spies, C., 2011. Social influences on the motivation to quit
smoking: main and moderating effects of social norms. Addict. Behav. 36, 286–293.
Doll, R., Peto, R., Boreham, J., Sutherland, I., 2004. Mortality in relation to smoking: 50
years' observations on male British doctors. BMJ 328, 1519–1527.
ENSH-Global Network for Tobacco Free Health Care Services. The ENSH-global self-audit
tool. http://www.ensh.org/docs/11-2016_ENSH-Global%20Self-Audit%20Tool.pdf.
Eriksen, W., 2005. Work factors and smoking cessation in nurses' aides: a prospective co-
hort study. BMC Public Health 5, 142.
European Monitoring Centre for Drugs and Drug Addiction (EMCDDA), 2011M. European
drug prevention quality standards. A manual for prevention professionals. EMCDDA
Manuals 7. EMCDDA, Lisbon.
Fletcher, A., Bonell, C., Hargreaves, J., 2008. School effects on young people's drug use: a
systematic review of intervention and observational studies. J. Adolesc. Health 42,
209–220.
Freedman, K., Nelson, N., Feldman, L., 2012. Smoking initiation among young adults in the
United States and Canada, 1998–2010: a systematic review. Prev. Chronic Dis. 9,
1–14.
Galbraith, N., Brown, K., 2011. Assessing intervention effectiveness for reducing stress in
student nurses: quantitative systematic review. J. Adv. Nurs. 67, 709–721.
Hirsch, K., Voigt, K., Gerlach, K., Kugler, J., Bergmann, A., 2010. Tabak -, Alkohol- und
Drogenkonsum sowie Impfverhalten von Gesundheits- und KrankenpflegeschülerInnen
in Sachsen-Anhalt. HB Science 1, 127–132.
Huang, C., Lin, M., Huang, G., Lee, H., Tsai, C., Hsieh, W., 2015. Implementation and out-
come analysis of tobacco-free hospitals in Taiwan. Poster, 12.06., Oslo.
Kolleck, B., 2004. Rauchen in der pflegerischen Ausbildung. Pflege 17, 98–104.
Kraus, L., Piontek, D., Pabst, A., de Matos, E.G., 2013. Studiendesign und Methodik des
Epidemiologischen Suchtsurveys 2012 [study design and methodology of the 2012
epidemiological survey of substance abuse]. Sucht 59, 309–320.
Lazarus, R.S., Folkman, S., 1984. Stress, Appraisal and Coping. Springer, New York.
Lindinger, P., Strunk, M., Nübling, M., Lang, P., 2012. Arbeitsweise und Wirksamkeit einer
Telefonberatung für Tabakentwöhnung. Sucht 58, 33–43.
Müller, H., 2013. Der erfolgreiche Umgang mit täglichen Belastungen: Das IFT-Programm
zur Stressbewältigung. Kursleitermanual. IFT-Gesundheitsförderung, München.
Perdikaris, P., Eleni, K., Elpida, G., Vasiliki, M., 2010. The relationship between workplace,
job stress and nurses' tobacco use: a review of the literature. Int. J. Environ. Res. Public
Health 7, 2362–2375. http://dx.doi.org/10.3390/ijerph7052362.
Pericas, J., González, S., Bennasar, M., De Pedro, J., Aguiló, A., Bauzá, L., 2009. Cognitive dis-
sonance towards the smoking habit among nursing and physiotherapy students at
the University of Balearic Island in Spain. Int. Council Nurses 56, 95–101.
Pirie, K., Peto, R., Reeves, G., Green, J., Beral, V., 2012. The 21st century hazards of smoking
and benefits of stopping: a prospective study of one million women in the UK. Lancet
381, 133–141.
Rapp, K., Büchele, G., Jähnke, A., Weiland, S., 2006. A cluster-randomized trial on smoking
cessation in German student nurses. Prev. Med. 42, 443–448.
Reeve, K., Shumaker, C., Yearwood, E., Crowell, N., Riley, J., 2013. Perceived stress and so-
cial support in undergraduate nursing students' educational experiences. Nurse Educ.
Today 33, 419–424.
Rice, V., Hartmann-Boyce, J., Stead, L.F., 2013. Nursing interventions for smoking cessa-
tion. Cochrane Db Syst Rev, Issue 8, Art. No.: CD001188.
Riou França, L., Dautzenberg, B., Falissard, B., Reynaud, M., 2009. Are social norms associ-
ated with smoking in French university students? A survey report on smoking corre-
lates. Subst. Abuse Treat. Prev. Policy 4, 1–14.
Sarna, L., Bialous, S.A., Wewers, M.E., Froelicher, E.S., Danao, L., 2005. Nurses, smoking and
the workplace. Res. Nurs. Health 28, 79–90.
Sarna, L., Bialous, S., Kraliková, E., Kmetova, A., Felbrová, V., Kulovaná, S., Malá, K.,
Roubicková, E., Wells, M., Brook, J., 2014. Impact of a smoking cessation educational
program on nurses' interventions. J. Nurs. Scholarsh. 46, 314–321.
Schaller, K., Pötschke-Langer, M., 2012. Tabakkontrolle in Deutschland und weltweit.
Pneumologe http://dx.doi.org/10.1007/s10405-011-0560-0.
Schultz, A., 2003. Nursing and tobacco reduction: a review of the literature. Int. J. Nurs.
Stud. 40, 571–586.
Statistisches Bundesamt, 2014. Mikrozensus 2013 – Fragen zur Gesundheit –
Rauchgewohnheiten der Bevölkerung. Statistisches Bundesamt, Wiesbaden.
Villanti, A.C., Mckay, H.S., Abrams, D.B., Holtgrave, D.R., Bowie, J.V., 2010. Smoking-cessa-
tion interventions for US young adults a systematic review. Am. J. Prev. Med. 39,
564–574.
Vitzthum, K., Koch, F., Groneberg, D., Kusma, B., Mache, S., Marx, P., Hartmann, T., Pankow,
W., 2013. Smoking behaviour and attitudes among German nursing students. Nurse
Educ. Pract. 13 (5), 407–412.
Wenig, J., Kröger, C., Erfurt, L., Nowak, D., 2013. Smoking cessation in groups – who ben-
efits in the long term? Health Educ. Res. 28, 869–878.
Zeidler, W., Kleiber, D., 2016. SQUIN – ein neuartiges, onlinebasiertes Gruppentraining zur
Tabakentwöhnung. Rep. Psychol. 1, 6–17.