Frequency of nurses’ smoking cessation interventions: report from a national survey

Authors

  • Linda Sarna,

    1. Authors:Linda Sarna, RN, DNSc, FAAN, Professor, School of Nursing, University of California, Los Angeles, CA, USA; Stella A Bialous, RN, MScN, Dr.PH, FAAN, President, Tobacco Policy International, San Francisco, CA, USA; Marjorie Wells, PhD, RN, Project Director, School of Nursing, University of California, Los Angeles, CA, USA; Jenny Kotlerman, MS, Principal Statistician, School of Medicine, University of California, Los Angeles, CA, USA; Mary E Wewers, RN, MPH, PhD, FAAN, Professor and Associate Dean for Research, College of Public Health, The Ohio State University, Columbus, OH, USA; Erika S Froelicher, RN, MA, MPH, PhD, FAAN, Professor, School of Nursing and Medicine, University of California, San Francisco, CA, USA
    Search for more papers by this author
  • Stella A Bialous,

    1. Authors:Linda Sarna, RN, DNSc, FAAN, Professor, School of Nursing, University of California, Los Angeles, CA, USA; Stella A Bialous, RN, MScN, Dr.PH, FAAN, President, Tobacco Policy International, San Francisco, CA, USA; Marjorie Wells, PhD, RN, Project Director, School of Nursing, University of California, Los Angeles, CA, USA; Jenny Kotlerman, MS, Principal Statistician, School of Medicine, University of California, Los Angeles, CA, USA; Mary E Wewers, RN, MPH, PhD, FAAN, Professor and Associate Dean for Research, College of Public Health, The Ohio State University, Columbus, OH, USA; Erika S Froelicher, RN, MA, MPH, PhD, FAAN, Professor, School of Nursing and Medicine, University of California, San Francisco, CA, USA
    Search for more papers by this author
  • Marjorie Wells,

    1. Authors:Linda Sarna, RN, DNSc, FAAN, Professor, School of Nursing, University of California, Los Angeles, CA, USA; Stella A Bialous, RN, MScN, Dr.PH, FAAN, President, Tobacco Policy International, San Francisco, CA, USA; Marjorie Wells, PhD, RN, Project Director, School of Nursing, University of California, Los Angeles, CA, USA; Jenny Kotlerman, MS, Principal Statistician, School of Medicine, University of California, Los Angeles, CA, USA; Mary E Wewers, RN, MPH, PhD, FAAN, Professor and Associate Dean for Research, College of Public Health, The Ohio State University, Columbus, OH, USA; Erika S Froelicher, RN, MA, MPH, PhD, FAAN, Professor, School of Nursing and Medicine, University of California, San Francisco, CA, USA
    Search for more papers by this author
  • Jenny Kotlerman,

    1. Authors:Linda Sarna, RN, DNSc, FAAN, Professor, School of Nursing, University of California, Los Angeles, CA, USA; Stella A Bialous, RN, MScN, Dr.PH, FAAN, President, Tobacco Policy International, San Francisco, CA, USA; Marjorie Wells, PhD, RN, Project Director, School of Nursing, University of California, Los Angeles, CA, USA; Jenny Kotlerman, MS, Principal Statistician, School of Medicine, University of California, Los Angeles, CA, USA; Mary E Wewers, RN, MPH, PhD, FAAN, Professor and Associate Dean for Research, College of Public Health, The Ohio State University, Columbus, OH, USA; Erika S Froelicher, RN, MA, MPH, PhD, FAAN, Professor, School of Nursing and Medicine, University of California, San Francisco, CA, USA
    Search for more papers by this author
  • Mary E Wewers,

    1. Authors:Linda Sarna, RN, DNSc, FAAN, Professor, School of Nursing, University of California, Los Angeles, CA, USA; Stella A Bialous, RN, MScN, Dr.PH, FAAN, President, Tobacco Policy International, San Francisco, CA, USA; Marjorie Wells, PhD, RN, Project Director, School of Nursing, University of California, Los Angeles, CA, USA; Jenny Kotlerman, MS, Principal Statistician, School of Medicine, University of California, Los Angeles, CA, USA; Mary E Wewers, RN, MPH, PhD, FAAN, Professor and Associate Dean for Research, College of Public Health, The Ohio State University, Columbus, OH, USA; Erika S Froelicher, RN, MA, MPH, PhD, FAAN, Professor, School of Nursing and Medicine, University of California, San Francisco, CA, USA
    Search for more papers by this author
  • Erika S Froelicher

    1. Authors:Linda Sarna, RN, DNSc, FAAN, Professor, School of Nursing, University of California, Los Angeles, CA, USA; Stella A Bialous, RN, MScN, Dr.PH, FAAN, President, Tobacco Policy International, San Francisco, CA, USA; Marjorie Wells, PhD, RN, Project Director, School of Nursing, University of California, Los Angeles, CA, USA; Jenny Kotlerman, MS, Principal Statistician, School of Medicine, University of California, Los Angeles, CA, USA; Mary E Wewers, RN, MPH, PhD, FAAN, Professor and Associate Dean for Research, College of Public Health, The Ohio State University, Columbus, OH, USA; Erika S Froelicher, RN, MA, MPH, PhD, FAAN, Professor, School of Nursing and Medicine, University of California, San Francisco, CA, USA
    Search for more papers by this author

Linda Sarna, School of Nursing, University of California, 700 Tiverton Ave, Los Angeles, CA 90095-6916, USA. Telephone: +1 310 825 8690.
E-mail:lsarna@sonnet.ucla.edu

Abstract

Aims and objectives.  To describe the frequency of nurses’ delivery of tobacco cessation interventions (‘Five A’s’: Ask, Advise, Assess, Assist, Arrange) and to determine the relationship of interventions to nurses’ awareness of the Tobacco Free Nurses initiative.

Background.  Tobacco cessation interventions can be effectively provided by nurses. The delivery of smoking cessation interventions by healthcare providers is mandated by several organisations in the USA and around the world. Lack of education and resources about tobacco cessation may contribute to the minimal level of interventions. The Tobacco Free Nurses initiative was developed to provide nurses with easy access to web-based resources about tobacco control.

Design.  Cross-sectional survey of nurses (n = 3482) working in 35 Magnet-designated hospitals in the USA (21% response rate).

Method.  A valid and reliable questionnaire used in previous studies to assess the frequency of the nurse’s delivery of smoking cessation interventions (‘Five A’s’) was adapted for use on the web.

Results.  The majority of nurses asked (73%) and assisted (73%) with cessation. However, only 24% recommended pharmacotherapy. Only 22% referred to community resources and only 10% recommended use of the quitline. Nurses familiar with TFN (15%) were significantly more likely to report delivery of all aspects of interventions, including assisting with cessation (OR = 1·55, 95% CI 1·27, 1·90) and recommending medications (OR = 1·81, 95% CI 1·45, 2·24).

Conclusions.  Nurses’ delivery of comprehensive smoking cessation interventions was suboptimal. Awareness of Tobacco Free Nurses was associated with increased interventions.

Relevance to clinical practice.  Further efforts are needed to ensure that nurses incorporate evidence-based interventions into clinical practice to help smokers quit. These findings support the value of Tobacco Free Nurses in providing nurses with information to support patients’ quit attempts.

Introduction and background

The Tobacco Free Nurses (TFN) initiative was launched in 2003 in the USA. It was designed to enhance and support nursing efforts in tobacco control by reducing smoking among nurses and to increase awareness of the nurses’ role in tobacco control through collaborations with nursing organisations (Sarna et al. 2006a). Additionally, the initiative focussed on increasing availability to nurses of smoking cessation materials for patients via the use of the Internet (http://www.tobaccofreenurses.org). Advertisements in major nursing journals in the USA, national presentations and collaborations with nursing organisations were used to publicise the availability of these resources. However, it is unclear if these efforts were successful in making staff nurses aware of the TFN resources or if this information influenced clinical practice by increasing nursing involvement in tobacco control (increasing frequency of nurses’ delivery of cessation interventions to their patients).

The effort to enhance the availability of tobacco control resources for nurses was in response to the prevalence of tobacco use in the U.S. (one of five adults, 20·8%) (Centers for Disease Control and Prevention (CDC) 2007a) and the resulting morbidity and mortality (U.S. Department of Health and Human Services (USDHHS) 2004). Registered nurses (RNs) comprise the largest group of healthcare professionals in the USA (US Department of Health and Human Services (USDHHS) 2006) (i.e. 2·5 million employed RNs). Adequately prepared nurses may thus be especially useful in broadening efforts to provide tobacco cessation interventions to all smokers, including low-income, underserved populations where tobacco use is most prevalent (Orleans et al. 2006, CDC 2007a,b).

Similarly, there is a worldwide appeal to improve nurses’ preparation and involvement in tobacco control, particularly in cessation interventions. The World Health Organization (WHO) has urged all health professionals, including nurses, to become more involved in tobacco control efforts to address the worldwide epidemic of tobacco-related diseases (WHO 2005). WHO specifically addressed the need for better education in the area of tobacco cessation, as healthcare providers need to be prepared to provide tobacco dependence treatment interventions. The work of TFN was highlighted as a model of a national initiative by the World Health Organization Tobacco Free Initiative (2005).

Evidence-based tobacco cessation strategies are the most cost-effective interventions for preventing morbidity and mortality (Fiore et al. 2000, 2002) but are underused by healthcare providers (Cokkinides et al. 2005). Treatment of tobacco dependence has been identified as a priority condition for action by healthcare systems (Institute of Medicine 2003), and performance on the delivery of smoking cessation interventions is one of the quality measures assessed by The Joint Commission on Accreditation of Health Care Organizations (Williams et al. 2005a,b).

The USA Public Health Service’s publication, Treating Tobacco Dependence a Clinical Practice Guideline, provides evidence-based strategies for assisting patients with tobacco cessation by all healthcare providers (Fiore et al. 2000). Although 42·5% of the 45 million current smokers (CDC 2006) annually attempt to stop smoking for at least 24 hours, few will be able to sustain long term abstinence without support and assistance (Fiore et al. 2002). Use of behavioural counselling and pharmacotherapy to aid in quit efforts are implemented using the ‘Five A’s’ (ask about tobacco use, advise to quit, assess readiness to quit, assist with quit efforts including recommendations of pharmacotherapy and arrange for follow-up). Some have suggested shortening this intervention to ‘ask, advise, refer’, with the referral to a smoking cessation telephone quitline for busy clinicians, or for clinicians who are not skilled in the delivery of smoking cessation interventions (Zhu et al. 2002, Schroeder 2005). In 2004, a national quitline network was established to allow all smokers in every state in the USA access to a telephone quitting smoking helpline. Quitlines have been shown to increase quit rates by 30–50% (Schroeder 2005), however it is unknown if nurses routinely refer patients to these resources.

The National Action Plan for Tobacco Cessation calls for the involvement of all clinicians, to enhance the delivery of tobacco dependence interventions (Fiore et al. 2004). To incorporate tobacco cessation interventions into clinical practice, healthcare providers need training, resources, and an expectation that such interventions are an integral component of quality care (CDC 2007b). The involvement of multiple healthcare professionals improves quit rates (Fiore et al. 2000), and the evidence that nurses are effective in providing cessation interventions is mounting (Sivarajan Froelicher et al. 2004, Persson & Hjalmarson 2006, Wewers et al. 2006, Rice & Stead 2008). However, several studies have suggested that, despite the evidence, interventions for smoking cessation are infrequently delivered by nurses and other healthcare professionals (Denny et al. 2003, Braun et al. 2004, Katz et al. 2004, Puffer & Rashidian 2004, Segaar et al. 2007). It is unknown if specially tailored resources for nurses, such as TFN, would enhance nurses’ delivery of cessation interventions to patients.

These issues are not unique to the USA (Percival et al. 2003). There only have been a few initiatives worldwide to enhance nurses’ readiness to deliver cessation interventions. Even fewer were systematically integrated into clinical practice after the end of projects, or were limited to a special patient population (e.g. prenatal care, diabetics, cardiac, etc.). Nonetheless, examples from the Netherlands (Segaar et al. 2007), Germany (Rapp et al. 2006), Australia (Cooke et al. 1998), and Denmark (Sejr & Osler 2002) demonstrate that providing additional training, along with system changes, to both nursing students and practicing nurses can have a positive impact on clinical practice by increasing nurses’ delivery of cessation interventions.

Barriers to smoking cessation interventions by nurses’ engagement include lack of training and skills, presence of smoking among nursing professionals, lack of administrative support, and other factors (e.g. the perception that patients do not want to quit smoking) (Sarna et al. 2001). Many nurses are unprepared to intervene with patients because tobacco cessation content is inadequate in most schools of nursing curricula both in the USA (Wewers et al. 2004) and in other countries (Sarna et al. 2006b, Chan et al. 2007). Results from the Global Health Professionals Survey of third year health professional students in 31 countries (including 209 nursing schools in 18 countries), concluded that there is a need to help students quit smoking and to implement programmes to train all health professionals in effective cessation interventions. In 16 of the countries, less than half of students reported receiving smoking cessation training (range 57–7·4%). Puska et al. (2005) found that nurses who enrolled in a smoking cessation programme for themselves also increased their cessation interventions with patients. Other factors, such as changing expectations and implementation of tracking systems for tobacco use and delivery of cessation, also influence interventions in healthcare organisations (Orleans et al. 2006). Another barrier to cessation is the lack of appropriate patient and provider cessation resources (Blumenthal 2007).

The National Quality Forum (NQF) established a Nursing Care Performance Measures project in 2003 to examine ‘nurse-sensitive’ indicators of quality nursing performance. Smoking cessation interventions for patients who smoked within the year prior to hospitalisation and who have a diagnosis of acute myocardial infarction, heart failure or pneumonia were endorsed as one of NQF’s 15 nursing-sensitive outcomes (National Quality Forum [NQF] 2006). This endorsement recognises the contributions of nurses to health outcomes in the area of tobacco cessation.

In summary, the tobacco epidemic is a major cause of preventable death; effective interventions are available. Nurses encountering patients who smoke can provide support for patients in their quit attempts, but few do. The availability of the TFN resources may assist nurses in these efforts.

Aim and objectives

The purposes of this study were to describe the frequency of nurses’ delivery of tobacco cessation interventions to adult patients and to explore the relationship of the frequency of interventions to awareness of TFN. The specific objectives were: (1) to describe of the frequency of nurses’ self-reported delivery of cessation interventions with patients based on the ‘Five A’s’ (ask, advise, assess, assist [including recommending pharmacotherapy], arrange [including referral to a telephone quit line and/or referral to community resources]), (2) to determine the awareness of TFN and (3) to examine the relationship of the nurses’ awareness of TFN, demographics (age, sex, smoking status, state of residence), professional characteristics (level of education, years as a nurse) and work-setting (type of unit, full-time status and shift) to delivery of the ‘Five A’s’. It was hypothesised that nurses who were aware of TFN would be more likely to engage in self-reported smoking cessation interventions with patients than nurses who were not aware.

Method

Design

A descriptive cross-sectional web-based survey was used to assess frequency of smoking cessation interventions and differences in frequency by awareness of TFN. Inclusion criteria for participation in the survey included being a nurse (Registered Nurse, RN or Licensed Practical Nurse, LPN) and providing care for adult patients. To conduct this national survey, we obtained a discrete list of Magnet-designated facilities identified through the American Nurses’ Credentialing Center (ANCC) website. Magnet recognition indicates excellence in nursing care within an institution (McClure & Hinshaw 2002). Surveying nurses in these settings would allow us to describe the practice of nurses in institutions with the highest standards of patient care.

Instruments

The Helping Smokers Quit (HSQ) 30-item, web-based questionnaire was used to describe frequency of smoking cessation interventions delivered by nurses and to assess awareness of TFN. This instrument was based upon a previous survey developed by the investigators (Sarna et al. 2000a, 2003) and used in a study to evaluate oncology nurses as to delivery of tobacco control interventions, as to attitudes and as to behaviours. In this study, eight items assessed tobacco cessation interventions according to the ‘Five A’s’. Response options for the frequency of these interventions included: ‘always’, ‘usually’, ‘sometimes’, ‘rarely’ and ‘never’. The reliability of this scale in the oncology nurse population was acceptable (ά = 0·92) (Sarna et al. 2000a). The psychometrics of the revised questions prepared for the web-based survey were re-evaluated prior to the national survey (Kappa = 0·70). The survey took approximately 5–7 minutes to complete.

The questionnaire included questions about demographic, professional and work characteristics of nurse respondents and new questions regarding awareness of TFN (yes/no) and its resources (website and printed materials). Demographic information included: age, sex, race/ethnicity, smoking status (never, former, current), state of residence. Professional characteristics included: level of nursing education, years of nursing practice, primary position (staff nurse, head nurse/supervisor, nurse practitioner, clinical nurse specialist, educator/case manager, other). Work-related characteristics included: full/part-time position, usual shift (day, evening, night) and unit where the nurse usually worked.

Procedure

Recruitment

An email invitation was sent to Chief Nursing Officers (CNOs) at all listed Magnet-designated facilities that met inclusion criteria:

  • • Three follow-up emails were sent over a four-week period to encourage participation. If the CNO expressed an interest in the study, further information was provided about the study, including assistance with obtaining approval from the institutional review board at the facility, if necessary. Strategies for distributing the survey to the staff within their facilities, most commonly via the facility’s local area network (LAN), were discussed.

Survey procedures

The HSQ survey was administered using SurveyMonkey.com© (2008), a professional programme where the subscriber can create a survey using multiple types of question formats and where participants can easily access and complete the survey online:

  • • The survey web-link was disseminated by CNO’s to their nursing staff. In Magnet-designated organisations with multiple facilities, the survey link was sent out via their organisation’s LAN, thus we were able to identify responses only from the organisation, not the individual hospital facilities.
  • • Regular email updates to participant organisations were used to provide feedback as to number of survey responses.
  • • After nurses accessed the survey website, an introductory letter described the survey process and the nurses’ rights as research subjects. Voluntary completion and submission of the survey constituted informed consent for participation in the project. No direct queries were sent by the investigators to the nurses in the healthcare setting.
  • • The website link ensured that the survey could be administered wherever the nurse had access to a computer connected to the Internet (work, home, laptop, etc.) and at any convenient time.
  • • Nurses had four weeks to complete the survey.
  • • Nurses who participated were eligible to enter in a lottery to win $100 in cash.
  • • Additionally, a description of the organisation (i.e. the number of beds, number of RNs and LPNs employed by the facility) was obtained from the CNOs and hospital databases (e.g. Hospital-data.com 2008).

Data management

An automated email notification and list management tool allowed us to track the data as they were collected. All data were downloaded to a secure location at a TFN computer at the University of California Los Angeles, School of Nursing, for data management and statistical analysis. The study was approved for exemption by the Institutional Review Board at the University of California, Los Angeles.

Data analysis

Descriptive statistics were used to characterise both the nurse respondents and the healthcare facilities. Data from responses to the survey questionnaires were examined to determine if participants met the inclusion criteria and if they completed responses to the questions describing frequency of delivery of tobacco cessation interventions and awareness of TFN. To increase the power of the analysis, some response options were combined: education (associate degree and diploma, and masters and doctoral), advanced practice nurse category (nurse practitioner, clinical nurse specialist, educator and case manager), unit (medical–surgical and cardiovascular; home health and outpatient), race/ethnicity (white or non-white).

Because the prevalence of smoking in a given region might have influenced the overall frequency of hospitalised smokers as well as the frequency of delivery of smoking cessation, we categorised nurses as residing in states with ‘high’ prevalence (above the national smoking prevalence median of 20·2%) or ‘low’ prevalence (below the national smoking prevalence median) prevalence states in 2006 (CDC 2007c), as listed below:

  • • High prevalence state category: Arkansas, Florida, Illinois, Indiana, Maine, North Carolina, Ohio, Pennsylvania, South Carolina and Wisconsin.
  • • Low prevalence state category: Colorado, Connecticut, Georgia, Maryland, New Jersey, New York, Texas, Utah and Virginia, plus the District of Columbia.

Chi-square analysis was used to examine the difference in frequency of the delivery of tobacco cessation interventions by awareness of TFN. We examined each component of the evidence-based cessation interventions separately. The relationships of unadjusted frequencies of the univariate demographic, professional, work-site and state smoking prevalence characteristics with the frequency of delivery of each of the ‘Five A’s’ were examined for statistical significance. We collapsed responses to the frequency of intervention for each of the ‘Five A’s’ (always/usually or sometimes/rarely/never). Multiple logistic regression analyses were conducted to estimate whether sets of characteristics [demographic (smoking status, high/low smoking prevalence state), professional (education, position), work-site (unit) and awareness of TFN] with p-values < 0·10 in the univariate analyses were associated with more frequent cessation intervention (always/usually intervene) for each of the ‘Five A’s’ compared with less frequent interventions (sometimes/rarely/never).

Results

Sample characteristics

At the time of the survey (September 2006), 181 health-care organisations representing 225 individual facilities in 42 states were listed on the ANCC’s Magnet Recognition Programme web pages. Ten were paediatric facilities and thus did not meet eligibility criteria, resulting in 167 eligible organisations (some included more than one hospital facility). CNOs from 35 organisations agreed and participated in the survey (21% of 171). These facilities were from 19 states, plus the District of Columbia. Twenty-three facilities declined or were unable to participate with the most common reason being concerns about staff overload, most often due to frequent requests for participation in surveys, and 101 facilities did not respond. Although CNOs at an additional 12 organisations initially agreed to participate, they were unable to complete the process of obtaining institutional review approvals in a timely manner.

The initial sample included 4489 nurses. The response rate, based upon the number of nurses working at the individual facilities reported by the CNO’s varied for each institution (range, 0·1–33·7%) with 26% of facilities having a response rate of ≥15%. The median response rate was 9·3%. After removing respondents who did not meet the inclusion criteria or who had incomplete data, 3482 nurses were maintained in the final sample for the analysis (Table 1). The respondents were similar in age, gender and ethnicity to the USA RN population (46·8 years of age, 5·8% male, 81·8% white), but were more highly educated (51·7% diploma/associate degree, 34·2% baccalaureate, 13·0% masters/doctoral education) (USDHHS 2006). RNs working in hospital settings are younger (43·2 years). A few nurse respondents (2·3%) were LPNs. The majority of the respondents (54·74%) worked in healthcare facilities in low smoking prevalence states.

Table 1.   Demographic, professional and work-setting characteristics of the sample (n = 3482)
  1. Percentages based upon number of respondents: *smoking rate > 20·2%, smoking rate ≤ 20·2%.

Demographics
Age [years] Mean (SD)42·77 (10·81)
Sex [n (%)]
 Female 3197 (92·91)
 Male 244 (7·09)
Race/ethnicity [n (%)]
 White 3020 (88·28)
 Non-white 401 (11·72)
Education [n (%)]
 Licensed practice nurse 77 (2·26)
 Associate/Diploma 1368 (40·09)
 Baccalaureate 1582 (46·37)
 Masters/Doctoral 385 (11·28)
Characteristics [n (%)]
Unit
 Intensive care 563 (17·06)
 Medical/surgical/cardiovascular 1520 (46·06)
 Obstetrics/gynaecology 379 (11·48)
 Psychiatric 100 (3·03)
 Emergency room 298 (9·03)
 Home health care/outpatient 440 (13·33)
Shift
 Day 2321 (72·49)
 Evening 209 (6·53)
 Night 672 (20·99)
Position
 Staff nurse 2589 (76·55)
 Head nurse/supervisor 410 (12·12)
 Advanced practice nurse 383 (11·32)
Smoking status
 Never 2246 (66·81)
 Former (ever who quit) 804 (23·.91)
 Current smoker 312 (9·28)
State smoking prevalence
 High prevalence state*1576 (45·26)
 Low prevalence state1906 (54·74)

The median bed size of the facilities where the respondents worked was 567 (range, 82–1136), and the length of time that a hospital had Magnet-designation ranged from 1–12 years (median, three years). The geographic distribution of states and number of hospitals (percent of the survey sample) was: five states in the Northeast (n = 11 facilities, 15·02%), four states in the Midwest (n = six facilities, 11·28%), eight states in the South (n = 15 facilities, 61·21%), and three states in the West (n = 3 facilities, 12·49%). Seven organisations (20·0%) contributed 151–440 respondents; 16 (45·71%) organisations had 51–150 respondents. Twelve (34·29%) had 1–50 respondents.

Frequency of cessation interventions and awareness of TFN

Overall, the majority (73%) of nurses frequently (‘always’ or ‘usually’) asked about tobacco use, advised (62%) about the risks of tobacco use and benefits of quitting and assessed (62%) motivation to quit tobacco use (Fig. 1). A minority (37%) ‘always’ or ‘usually’ assisted (37%) with cessation efforts, arranged (19%) or recommended (24%) cessation medications, referred (22%) to resources or recommended (10%) use of the quitline. Fifteen per cent of the participants were aware of TFN. Nurses who were aware of TFN reported significantly more frequent smoking cessation intervention activities as compared to those who were not aware of TFN.

Figure 1.

 Differences in delivery of cessation intervention by awareness of TFN. Unadjusted chi-square: *p ≤ 0·01, **p ≤ 0·0001.

Factors associated with frequency of cessation interventions

Multiple logistic regressions on factors associated with the frequency of the nurses’ delivery of each of the ‘Five A’s’ (Table 2) demonstrated that familiarity with TFN increased the odds of delivering interventions to patients in almost all aspects. The frequency of each intervention differed by demographic, professional characteristics and workplace setting. Current smokers were less likely to arrange for follow-up, but there were no other differences by smoking status. Nurses working in hospitals in states with high tobacco prevalence also were significantly more likely to deliver cessation interventions on a more frequent basis. Advanced practice nurses (nurse practitioners and clinical nurse specialists) were significantly more likely to ‘advise’ patients to quit, arrange for follow-up, recommend medications and twice as likely to ‘arrange’ referral of smokers to other resources, including the quitline.

Table 2.   Predictors of likelihood of always/usually delivering tobacco cessation interventions (n = 2861)
 AskAdviseAssessAssistRecommend medicationsArrangeRefer to QuitlineRefer to community resource
  1. Values are given as OR (95% Cl). Bold letters indicate significantly significant factors (those having confidence intervals that do not include 1·0).
    OR, odds ratio; CI, confidence interval; TFN, Tobacco Free Nurses; Med/Surg, medical/surgical; OB/GYN, obstetrics/gynecology; psych, psychiatric; ER, emergency room; OutPt, outpatient.
    *p < 0·05, **p < 0·01.

Not Familiar with TFN1·00 1·00 1·00 1·00 1·00 1·00 1·00 1·00
Familiar with TFN1·16, (0·92, 1·46)1·27* (1·03, 1·57)1·34** (1·08, 1·65)1·55** (1·27, 1·90)1·81** (1·45, 2·24)1·90** (1·52, 2·38)2·20** (1·66, 2·92)1·71**(1·38, 2·14)
Smoking status
Never smoker1·00 1·00 1·00 1·00 1·00 1·00 1·00 1·00
Former smoker0·91 (0·75, 1·09)0·95 (0·80, 1·13)0·98 (0·83, 1·16)1·03 (0·86, 1·22)1·20 (0·98, 1·45)0·87 (0·70, 1·07)0·80 (0·59, 1·08)0·94 (0·76, 1·15
Current smoker1·02 (0·77, 1·36)0·88 (0·69,1·13)0·94 (0·73, 1·21)0·98 (0·76, 1·27)1·10 (0·83, 1·47)0·70* (0·50, 0·97)0·67 (0·42, 1·09)0·85 (0·62, 1·16
Education
MS/PhD 1·43** (1·05, 1·96)1·35* (1·03, 1·79)1·30 (0·99, 1·70)1·10 (0·84, 1·44)1·28 (0·95, 1·72)1·02 (0·74, 1·41)0·81 (0·52, 1·24)1·04 (0·77, 1·41)
BS1·00 1·00 1·00 1·00 1·00 1·00 1·00 1·00
Associate/Diploma 1·00 (0·84, 1·19)1·01 (0·87, 1·18)1·09 (0·94, 1·28)1·08 (0·92, 1·27)1·15 (0·96, 1·38)1·17 (0·97, 1·42)0·82 (0·63, 1·07)0·99 (0·82, 1·19)
LPN0·56* (0·33, 0·94)0·87 (0·53, 1·.43)0·87 (0·53, 1·43)1·40 (0·85, 2·33)0·96 (0·52, 1·76)1·57 (0·88, 2·81)1·22 (0·56, 2·65)0·78 (0·41, 1·49)
Position
Staff nurse1·00 1·00 1·00 1·00 1·00 1·00 1·00 1·00
Advanced practice1·03 (0·77, 1·37)1·38* (1·06, 1·81)1·10 (0·85, 1·43)1·07 (0·83, 1·39)1·40* (1·06, 1·85)1·65** (1·23, 2·22)2·11** (1·45, 3·07)1·90** (1·43, 2·51)
Unit
Med/Surg 1·00 1·00 1·00 1·00 1·00 1·00 1·00 1·00
Intensive care1·18 (0·95, 1·48)1·39** (1·13, 1·72)1·23 (0·99, 1·51)1·14 (0·93, 1·39)1·38** (1·11, 1·72)1·37** (1·09, 1·72)1·14 (0·84, 1·56)1·48** (1·18, 1·86)
OB/GYN 1·34* (1·03, 1·74)0·81 (0·64, 1·02)0·81 (0·64, 1·02)0·51** (0·39, 0·65)0·51** (0·38, 0·70)0·51** (0·36, 0·70)0·53** (0·34, 0·84)0·68* (0·50, 0·92)
Psych2·65** (1·46, 4·83)1·36 (0·87, 2·13)1·47 (0·92, 2·34)1·29 (0·85, 1·96)2·47** (1·62, 3·77)0·80 (0·48, 1·35)0·63 (0·30, 1·35)0·74 (0·44, 1·26)
ER3·13** (2·16, 4·55)1·24 (0·95, 1·62)0·45** (0·35, 0·58)0·35** (0·25, 0·47)0·45** (0·31, 0·64)0·32** (0·21, 0·50)0·42** (0·24, 0·72)0·48** (0·33, 0·69)
OutPt0·93 (0·74, 1·18)1·11 (0·88, 1·38)0·80* (0·64, 0·99)0·73** (0·58, 0·91)0·65** (0·49, 0·85)0·70* (0·53, 0·94)0·68 (0·46, 1·02)1·16 (0·90, 1·50)
Low prevalence state1·001·00 1·00 1·00 1·00 1·00 1·00 1·00
High prevalence state1·15 (0·98, 1·35)1·26** (1·09, 1·46)1·31** (1·13, 1·52)1·40** (1·20, 1·62)1·13 (0·96, 1·34)1·51** (1·26, 1·81)1·94** (1·51, 2·49)1·70** (1·43, 2·03)

Compared with nurses working in medical–surgical units, nurses in intensive care units were more likely to intervene. Nurses working in obstetrical/gynecological units (OB/GYN) and emergency rooms were more likely to ‘ask’ patients about their tobacco use, but less likely to offer cessation interventions. Nurses working in outpatient facilities were less likely to offer assistance with cessation.

Discussion

The frequency of smoking cessation interventions among this sample was similar to that reported by nurses in earlier national and international surveys (Sarna et al. 2000a, 2003, Rice & Stead 2008). The frequency of cessation interventions by nurses overall was suboptimal, even in healthcare organisations that had received a national recognition for excellence in patient care. Nurses who were aware of TFN were significantly more likely to deliver evidence-based smoking cessation interventions on a more frequent basis.

Providing advice to smokers to quit more than doubles the quit rate (Bao et al. 2006). Our findings demonstrate that the majority of nurses consistently asked and advised patients to quit. These findings are similar to a national survey of 3000 physicians conducted on behalf of the American Legacy Foundation where 84% asked about smoking status, 86% advised patients to stop smoking and 63% assessed a patient’s willingness to quit (Association of American Medical Colleges (AAMC) 2007).

Consistent with our findings, fewer provided the support to actively assist patients in quitting. Although the majority of physicians (68%) in that survey discussed pharmacotherapy, only 31% recommended nicotine replacement, fewer (25%) prescribed other medications to aid quit attempts. In general, our findings support reports from several countries including China (Chan et al. 2007), Canada (Schultz et al. 2006), North Ireland (Slater et al. 2006), Germany (Thyrian et al. 2006), Netherlands (Bakker et al. 2005), UK (Condliffe et al. 2005, Hall et al. 2005) and Korea (Kim et al. 2004), demonstrating that nurses’ smoking status negatively impacts their intervention with patients, that nurses are generally ill-prepared to provide tobacco dependence interventions, and that when they address tobacco use it tends to be limited to asking about smoking status without additional intervention or follow-up. However, unlike the previously cited studies, our study is the first to report findings across a broad geographic area and cross-specialty focus. International experiences also demonstrate that when nurses receive training or resources, such as those housed on the TFN website, they tend to increase their frequency in providing cessation interventions (Cooke et al. 1998, Sejr & Osler 2002, Puska et al. 2005, Rapp et al. 2006).

Smoking interventions that begin during hospitalisation, regardless of diagnosis, increase quit rates, especially when at least one month of follow-up is included in the intervention (Rigotti et al. 2007). The findings from this survey indicate that arranging for follow-up has not yet been incorporated into clinical nursing practice.

Our results indicate that the majority (81%) of nurses did not refer patients to the one free resource associated with improving quit rates – a telephone quitline. If lack of knowledge and skills, limited time and other job pressures are barriers to providing comprehensive cessation services, the referral of patients to the quitline can allow them to receive evidence-based cessation advice and support (Zhu et al. 2002, Schroeder 2005). Similar to findings of the physician survey described above (AAMC 2007), where only 13% of doctors referred patients for cessation treatment and 7% to a quitline, few nurses in this study (10% always or usually) referred patients to this resource.

The delivery of smoking cessation interventions could be viewed as a marker of quality of care. As Magnet-designation denotes quality nursing care, it has implications for overall excellence in care which might indicate increased delivery of smoking cessation interventions. However, we did not evaluate if this was the case. Further study is needed to assess how frequency of delivery of tobacco cessation interventions by nurses from such facilities compares to those settings without Magnet designation.

The frequency of delivery of cessation interventions varied by the nurses’ personal and professional characteristics and by worksite factors. As has been reported, smoking status negatively influences the delivery of smoking cessation intervention (Sarna et al. 2001, Heath et al. 2004). In our study, nurses who self-reported as current smokers were less likely to arrange for follow-up services compared to never smokers, but there were no other statistically significant differences. The findings that frequency of interventions was significantly greater among nurses in states with a high smoking prevalence might reflect increased state tobacco control efforts, as was seen in an earlier physician survey (AAMC 2007) or may be due to caring for a higher number of patients who smoke.

Professional characteristics, including advanced practice roles (e.g. nurse practitioners), were associated with increased delivery of smoking cessation interventions. Nurses who were least educated (i.e. LPNs) were the least likely to ask and advise patients to quit. As has been reported by others, lower levels of education is linked with lower levels of delivery of smoking cessation interventions (Sarna et al. 2000b, Slater et al. 2006).

Frequency of interventions varied by the clinical setting. Although nurses working in emergency rooms were more likely to assess smoking status compared with nurses in medical surgical units, cessation interventions were considerably less frequent. An emergency visit for a tobacco-induced event provides an opportunity for a ‘teachable moment’ in the delivery of cessation advice to quit. Even in this acute situation, cessation interventions could include, at a minimum, referral to community resources, including the telephone quitline, as supported by the Emergency Medicine Organisations, including the Emergency Nurses Association (Bernstein et al. 2006).

Patients with mental health illnesses are more than twice as likely to smoke. There has been a reluctance to assist psychiatric patients with cessation in the fear that it might interfere with treatment of their mental illness (Lasser et al. 2000). However, our findings reveal that nurses in psychiatric units were more than twice as likely to assess smoking status and recommend medications compared to nurses in medical–surgical settings. Of concern was the low incidence of intervention among nurses in OB/GYN settings; however, this could be due to lower smoking prevalence, perceived or actual, among these patients.

Further research is needed to determine if the TFN website and materials are more relevant to supporting tobacco cessation interventions than other widely available websites (e.g. Agency for Healthcare Research and Quality) in supporting nurses in clinical care. Additionally, as we were unable to determine if differences in nursing interventions were due to healthcare practices in the individual organisations, further research is needed to determine if system level factors (e.g. requiring documentation of treatment strategies for smoking cessation) influenced the frequency of nursing interventions. We did not examine differences in nurses’ delivery of interventions by acuity of patient population. This is not differentiated in the Joint Commission evaluations, but it deserves further study.

To increase cessation attempts, healthcare providers need to be aware of resources that can equip them with skills and knowledge, as well as access to cessation services. Quitting smoking in the USA is becoming the norm, and the majority of people who have smoked in the USA (50·2%) have quit (CDC 2007a). The sustained efforts that are needed to assist smokers to quit can be augmented if nurses are actively involved in these efforts.

Limitations

There are several limitations to this study, including the low response rate from CNOs (20·5%) and from the nurses in each facility who responded to the web-survey (median 9·3%). Although web-based research is a growing field in nursing research, low response rates and selection bias have been identified as challenges (Im & Chee 2004, Ahern 2005, Im et al. 2006). In this study, we were not able, proactively to email individual nurses or request participation of individual nurses.

We needed to rely on the dissemination of a web-link by the CNO. Although some research studies were able to produce higher response rates after proactively emailing participants multiple times (Saul et al. 2007) and expanding the time frame for data collection, others have found even lower response rates after multiple recruitment strategies and pro-actively emailing nurses. For example, Im et al. (2006) only obtained a 0·72% response rate after four months of data collection and active recruitment. Im and Chee’s (2004) assessment of response rates from three different studies showed that response rates varied from 2–10%. The sample in this study is the largest ever to describe smoking cessation interventions by nurses. Despite the large sample, some subsets were too small to allow for the testing for statistically significant differences.

The low response rate from the CNOs and the nurses could have biased our findings, making generalisability difficult. Those CNOs who did respond and provided us with a reason for lack of participation, let us know that they experienced numerous requests for participation in [local and national] surveys, resulting in survey ‘fatigue’. Most previous studies of cessation interventions by nurse are cross-sectional and did not collect data regarding characteristics of the healthcare facility.

These data are not meant to be representative of all nurses working in Magnet facilities, nor are they an evaluation of smoking cessation efforts in Magnet organisations overall. The majority of respondents were nurses who worked in hospital-based facilities. The Joint Commission indicator mandates that smoking cessation interventions are delivered to patients with selected diagnoses, despite their acuity, during a hospital stay. It is recognised that cessation interventions are more commonly delivered in primary care outpatient settings (Fiore et al. 2000). We did not determine differences in interventions delivered by nurses in acute and primary care settings.

As this is a cross-sectional survey, we were only able to determine the association of TFN awareness with self-reported cessation interventions. We were unable to determine if nurses who were more actively involved in cessation prior to the initiation of TFN were more likely to be aware of and use this resource. Additionally, we are unable to determine differences by facility, such as nurse/patient ratios, which have been attributed to the quality of patient outcomes and directly associated with cessation interventions (Aiken et al. 2002). We are unable to determine if lack of participation in the survey indicated less positive involvement in tobacco control and less frequent interventions.

A media campaign promoted our partnership with NursesQuitnet® as a resource to support nurses with their own smoking cessation efforts. It is possible that nurses, regardless of smoking status, may not have viewed TFN as a resource for all nurses for assistance with patient-focussed interventions. Finally, these data were based upon self-reports from a web-based survey. We are unable to assess the relationship of these reports with evidence from actual clinical practice.

Relevance to clinical practice

The findings of this study support a strong positive relationship between awareness of a nurse-focussed tobacco cessation resource, TFN, with increased frequency of smoking cessation interventions. In particular, nurses who were aware of TFN were significantly more likely to refer patients to a telephone quitline, a free but underused resource. As expected, advanced practice nurses were more likely to engage in cessation interventions, but all levels of healthcare providers, regardless of preparation, can refer to the toll free telephone quitline. The differences in the frequency of cessation interventions by nursing unit suggests that more attention may be needed in changing expectations for delivery of interventions based upon the type of patient. Further efforts to promote the resources of TFN to nurses within healthcare settings might increase awareness of expectations to help smokers quit by using evidence-based approaches.

Acknowledgements

This research was supported by The Robert Wood Johnson Foundation grant no. 441056. The authors would like to thank Lisa Chang for her assistance with this project.

Contributions

Study design: LS, SA, MEW; data collection and analysis: LS, SA, MW, JK and manuscript preparation: LS, SA, MW, MEW, JK.

Ancillary