Are smoking cessation interventions prior to elective surgery, effective in achieving short and long term smoking cessation?

ABSTRACT: 178
Objectives: Smoking cessation intervention programmes prior to elective surgery are reviewed in relation to achieving short and long term abstinence. Research studies examining smoking cessation interventions and its overall effectiveness were critiqued and evaluated using a systematic literature review.
Methods: Specific medical databases were utilised, followed by a methodical search strategy to identify key articles. Through evaluation of the studies, reliability and validity will be established. Evidence based practice will be explored throughout the report and will demonstrate how to source and appraise relevant research, whilst indicating how research findings can be implemented within todays nursing practice.
Findings: It has been identified that smoking cessation intervention programmes are successful in short and long term abstinence with evidence identifying that post-operative complications are higher in patients that smoke.
Implications: The right time to introduce smoking cessation pre-operatively is debatable although the majority of research argues that even the shortest time will be of benefit to the patients overall health and recovery time. Patients attitudes and willpower towards smoking cessation have a considerable impact upon how effective the intervention is.
INTRODUCTION: 110
Modern medical practice is founded on applying scientific principles to the art of medicine (Lauerman: 2008). The National Institute for Nursing Research (2013) defines nursing research as the underpinning knowledge that enables nurses to promote and improve health and provides the scientific foundation for clinical practice whilst preventing disease and disability. Bowling (2014) suggests health care research ranges from descriptive experiences of illness, the individuals’ perception to evaluations on health services in relation to its appropriateness, effectiveness and cost. The Nursing and Midwifery Council (2015) states that health care professionals must implement evidence based practice to ensure the delivery of safe and effective care and practice to all. A questioning and critical approach to practice will enhance practice and help to develop ways to accomplish this.

Background: (600) 426 so far
Globally, over 230 million adults undergo surgery each year with millions experiencing major respiratory and cardiovascular complications, alongside impaired wound healing (Weiser et al: 2008). The National Health Service (NHS), (2015) claims smoking is one of the biggest causes of death and illness in the UK. Each year 100,000 people die from smoking, with many more deaths caused by smoking-related illnesses. Smoking increases risk of developing more than 50 serious health conditions. Some are potentially fatal and others can cause irreversible long-term damage to health. The U.S Department of Health and Human Services, (2010) agrees that cigarette smoking harms nearly every organ of the body, causes many diseases, and reduces the health of smokers in general and suggest that quitting smoking lowers the risk of smoking-related diseases adding years to their life.
Cigarette smoking has been implicated as a risk factor for post-operative complications across a spectrum of surgical specialities (Glassman et al: 2000, and Brooks-Brunn: 1997). Previous research by Peters et al, (2004) suggests the advisability of performing surgery on patients that smoke has become questionable based on evidence of greater surgical risks and healthcare costs for patients who smoke. Smokers have a higher risk of developing post-operative complications such as impaired wound and or bone healing to life threatening pulmonary and cardiovascular complications (Selber et al:2006, Finan et al:2005 and Sadr Azodi et al:2006). Moller et al (2002) suggest that increased surgical risks could be reduced if patients were encouraged to stop smoking prior to surgery.
Research suggests that most smokers accept that smoking is harmful and understand the benefits of stopping but continue to do so regardless (Taylor et al:2005). Interventions by health care professionals can encourage and help them to quit as previous research by Hughes et al (2004) implies that some will give up when they have a good reason to. National Institute for Health and Care Excellence (NICE) (2006) recommend that healthcare pofessionals give the following advice; ask them if they are interested in quitting, referrals to relevant cessation services, offer smoking aids, offer advice and support and update monitoring systems to match patients smoking status.
Although there are studies suggesting that stopping smoking shortly before surgery increases complications, the weight of evidence does not favour this especially with evidence showing that cardiac and surgical risk decrease reduces soon after smoking cessation (Warner: 2005). According to Theadom et al, (2006) smoking cessation eight weeks pre-operatively is associated with reduced complications risks. However, studies implementing smoking cessation at different periods need to be examined to provide clear guidance on its benefits.
SEARCH STRATEGY: 283
An effective literature search helps to locate and identify current and evidence based information (Parahoo: 2006). The acronym PICO was used to search terms examining the context around the issue highlighted and interventions addressing the problem (appendix 1). National Collaborating Centre for Methods and Tools (2015) suggests the PICO tool assists in articulating the search enabling to formulate a question for qualitative research. Cheng, (2004) states PICO improves the specificity and conceptual clarity of issues with a focused way of identifying precise results (Booth et Al:2000).
The Cumulative Index to Nursing Allied Heath (CINAHL) database was searched using the keywords (appendix 2) with an inclusion and exclusion criteria adhered to (appendix 3). Due to multitudes of results found within CINAHL, MEDLINE was not explored. If results were limited, a MEDLINE search would provide depth as National Library of Medicine, (2015) states MEDLINE, incorporates medical, nursing and social care literature globally. The British Nursing Index (BNI) could be explored to reduce the risk of omitting British based studies. Library support was accessed to gain knowledge upon robust search strategies. Perspectives from an Advanced Laparoscopic Nurse who admittedly saw poorer wound healing in patients that smoke was sought. The Boolean operator tool was used allowing for expansion, narrowing or combining of keywords. Alliant Libraries, (2015) argue it focuses searches and minimises the risk of irrelevant documents (appendix 4). Abstracts of the articles were read to ensure that the focus of those studies were relative to the proposed question; however, the transferability of research could be argued due to the varying surgical interventions. Four papers were selected from the search results in relation to the proposed question (appendix 5) Are smoking cessation interventions prior to elective surgery, effective in achieving short and long term smoking cessation?

Article 1 Critique 619
The first study, written by Wolfenden et al (2005) assesses how effective smoking cessation interventions are in increasing abstinence rates pre and post-operatively. A quantitative randomised control test (RCT) was used which is the most scientifically rigorous method of hypothesis testing available (Akobeng: 2005) and are regarded as gold standard in evaluating the effectiveness of interventions (McGovern, 2001). Some argue that whilst RCT’s are a good experimental design, the potential shortcomings and implementation of RCT’s are often overlooked (Grossman and Mackenzie, 2005).
Parahoo, (2006) suggests powerfulness can be determined by the sample size whilst Gerrish and Lacey (2010) highlight that smaller studiers need support to strengthen its generalizability. Inclusion exclusion criteria clearly identified who was eligible for the study (see appendix 6). However; the study could have been translated into other languages to include more participants. This could provide strength to the transferability of the study however; funding may have been the issue. Sample size must represent the population it was drawn from but genders were not specified its adequacy was questionable. The population within Newcastle, Australia is 288,733 with 22,000 (14%) smokers. With 210 of the participants being smokers, over-representation is argued. Upon further research, the National Health Performance Authority (2013) suggests the area contains just 6% of the smoking population across Australia, resulting in a misrepresented sample size.
Parahoo (2006) suggests ethical implications within all stages of research that must be addressed to be deemed trustworthy. Written consent was obtained whilst approval and monitoring was sought by the University of Newcastle Human Research Ethics Committee (HREC) (2015) and the Hunter Area Research Ethics Committee. The HREC operates in accordance with the Australian Governments National Health and Medical Research Council guidelines on Ethical Conduct in Human Research 2007. This appears the equivalent to National Institute for health and Care Excellence (NICE) guidelines adhered to within the UK so assumingly all ethical issues had been appropriately addressed.
Participants completed a computerised assessment declaring themselves as current smokers which allocated a random number function to receive either the intervention or usual carein a ratio of 3:2 retrospectively. This allocation enabled the majority to receive the benefits of the intervention. Groups were divided into categories of; non-dependant (10 cigarettes) and usual care and were given the appropriate intervention (see appendix 7). Data was collected during a 3 month telephone survey. Individual reading and computer literate skills were not assessed therefore results may be varying due to personal. Outcomes were assessed after three months via a follow up survey conducted by a blinded assistant. A blinded assistant ensures unbiased results however; patients and clinic staff were not blinded to the groups’ allocations suggesting patients allocated to receive usual care may have obtained interventional components. Limitations include self-declaration due to some potentially reporting incorrectly. Biochemically validating smokers’ abstinence is recognised as gold standard.
Data was analysed using SAS version 8.2 statistical software with all tests being 2-tailed. Differences between the groups were assessed using chi-squared tests and independent sample t-tests. The primary outcomes were assessed by comparing abstinence rates between groups using chi-squared tests odds ratios and 95% CI as part of the intention to treat analysis.
Results were clearly illustrated within a flow chart which evidently points out that some participants were lost to follow up now questioning its validity and reliability. Results indicate that the intervention group were significantly more likely to be abstinent pre-operatively and at the three month follow up. For all smokers combined, a greater proportion receiving interventions reported abstinence and improved stage of change compared to those receiving usual care. In conclusion, the findings suggest that the provision of smoking cessation has the potential to improve smoking behaviour.

Article 2 critique: 601
The second paper, authored by Andrews et al, (2005) identified its primary aim as assessing the effectiveness of a letter, from a consultant surgeon in causing smokers to stop smoking pre-operatively. Stolberg et al (2004) state RCT’s are the most powerful experimental design in clinical trials as discussed previously. Block randomisations of each patient were numbered sequentially. The corresponding numbers were placed in a bag with the first drawn out assigned to the intervention status. The second number was assigned to control and so on alternatively. This block size of two randomisation works by randomising participants within blocks ensuring equal numbers are assigned to each intervention. Efird (2011) suggests allocation continues by randomly selecting and assigning the next block according to the specified sequence. Disadvantages suggest that allocation becomes predictable resulting in selection bias. Efird (2011) suggests random block sizes and blinded investigators would be necessary to reduce selection bias.
The sample size consisted of 51 participants within each group with one from the control group withdrawing. The participant could have withdrew due to receiving the control method and may see it is no benefit. Sample size is important as errors decreasing with larger samples. Smaller samples may be over-representative of small subgroups within the target population (Burns and Grove: 1997) as discussed previously. Russell (2005) suggests inclusion and exclusion criteria should be clearly defined in order to avoid bias, however this was not discussed.
The trial took place in Sandwell General between January and April 2004 and permission was obtained the research governance committee. Approval has been assumed and suggests that ethical principles are applied and the rights of the individual adhered to (Burns and Grove, 1999). Smokers were asked to take part following a brief explanation and upon agreement, they were introduced to the researchers who explained in more depth and were asked to sign a consent form. Bowrey and Thompson (2014) state that written, informed consent must be obtained for participation in most clinical trials.
Self-reported abstinence results were collected via questionnaire administered to the patients on the day of their surgery. Coughlan et al (2007) claims questionnaires are most commonly used in data collection and could be considered anonymous if identification was not stated. Parahoo (2006) suggests confidentiality breaches are avoided and are less intrusive than an interview or an experiment. Participants quit status was not confirmed by testing exhaled carbon monoxide, therefore the intervention group may have felt obliged to state abstinence more than the control group.
Declaration of what statistical tests were involved with analysis the data were not provided. Coughan et al (2007) suggests inferential statistical tests limit some threats to the validity and reliability of the study. In qualitative studies, identification of significance could be shown as p values (