Family Practice Vol. 19, No. 2, 154-160
© Oxford University Press 2002
Original Paper |
A new method for describing smokers' consulting behaviours which indicate their motivation to stop smoking: an exploration of validity and reliability
a Division of General Practice, School of Community Health Sciences, University of Nottingham, Queen's Medical Centre, Nottingham NG7 2UH, UK. E-mail: tim.coleman{at}nottingham.ac.uk
b Department of General Practice and Primary Health Care, University of Leicester, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW, UK.
Coleman T, Stevenson K and Wilson A. A new method for describing smokers' consulting behaviours which indicate their motivation to stop smoking: an exploration of validity and reliability. Family Practice 2002; 19: 154–160.
Received 18 December 2000; Revised 14 August 2001; Accepted 1 November 2001.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Smokers vary in their readiness to try stopping smoking, but there currently are no objective tools for identifying smokers' consulting behaviours which indicate their level of motivation to try stopping smoking.
Objective. The aim of this study was to investigate the construct validity and inter-observer reliability of the Smokers' Motivation Code (SMC).
Methods. General practice consultations between 29 different Leicestershire GPs and their patients were video-recorded. In 47 consultations, regular or occasional smokers discussed smoking with their GPs and their consulting behaviour was coded using the SMC. The reliability of three different observers' codings was investigated. Construct validity was also investigated by comparing smokers' consulting behaviours coded using the SMC with measures of motivation to stop smoking recorded on pre-consultation questionnaires.
Results. Two pairs of observers achieved good reliability when using the SMC to code smokers' consulting behaviours during a subset of 11 video-recorded consultations. For readiness behaviours (indicating motivation to stop smoking), kappas were 0.82 and 0.65, and for resistance behaviours (indicating little motivation to stop smoking), kappas were 0.74 and 1.0. For the 37 consultations attended by regular smokers, complete pre-consultation questionnaires were obtained. Smokers displaying readiness behaviours were significantly more likely than others to report having tried to stop in the past year, thinking about or trying to stop and to agree that their health would improve if they stopped smoking. Smokers displaying resistant behaviours were significantly less likely to report thinking about stopping/trying to stop smoking.
Conclusion. We have provided some evidence to support the construct validity and inter-observer reliability of the SMC and have identified some consulting behaviours which might indicate smokers' motivation to stop smoking. Further work is needed to determine whether smokers' consulting behaviour can be used to predict future quit attempts.
Keywords. Consulting behaviour, reliability, Smokers' Motivation Code, validity.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Smoking remains a massive public health problem in the UK.1 Brief anti-smoking advice from GPs has been shown to have a small but beneficial effect on their patients' smoking behaviour.2 GPs' advice works by triggering advised smokers to make more quit attempts rather than increasing the likelihood that smokers will be more successful in stopping at any one quit attempt.3 Smokers vary in their motivation to try to stop smoking, and those who have greatest motivation are most likely to attempt this.4 Consequently, these motivated smokers are most likely to respond to GPs' anti-smoking advice. Questionnaires can identify if heavy drinkers are more ready to cut their alcohol intake5,6 and have also been used to predict smokers' readiness to stop7,8 but, unfortunately, there currently is no way of identifying smokers who are motivated to stop as they consult with health professionals. Objective methods of assessing smokers' motivation to stop as they consult could help health professionals to tailor their anti-smoking interventions to smokers' levels of motivation. This paper, therefore, investigates the construct validity and inter-observer reliability of the Smokers' Motivation Code (SMC), which is a new method for describing smokers' consulting behaviours, hypothesized to indicate both higher and lower motivation to try stopping smoking.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Process of data collection
For this study, a smoker is defined as anyone who has smoked at least one cigarette during the previous year. This definition includes those who smoke occasionally and also those who have stopped very recently, and it allowed us to study the consulting behaviour of smokers who had stopped smoking recently or currently were taking action against their habit. Forty-two Leicestershire GPs participated in the study and these were a random sample of all GPs in Leicestershire, stratified by their attitudes towards discussing smoking with patients.9,10 Recruitment of GPs and their characteristics are described elsewhere.9–11 Each GP agreed to have one surgery session video-recorded and all patients attending data collection surgeries were asked to complete a pre-consultation questionnaire (details of contents below). After completing the questionnaire, patients' consent to video-recording of their consultations was sought and, where this was given, video-recording occurred.11
Contents of pre-consultation questionnaire
This recorded patients' age and sex and identified all patients who had smoked a cigarette in the previous year. Additionally, those who reported smoking cigarettes on at least most days (regular smokers) were asked about their strength of nicotine addiction, their past quitting behaviour, future intentions to quit (or not), strength of desire (or lack of it) to stop smoking, attitudes toward smoking and their confidence (self-efficacy) in their ability to stop smoking. These qualities have all been associated with successful smoking cessation in prospective studies,12 and further details are given below.
Nicotine addiction
Lennox13 summarized the evidence that heavier smoking is associated with either difficulty in stopping smoking or making lower numbers of attempts to quit, though many studies recruited motivated volunteers who may differ from general practice patients. Two UK12,14 community-based studies and one Australian15 study have found heavier smoking associated with difficulty maintaining complete abstinence from smoking.
Past quitting behaviour
Higher numbers of past quit attempts are associated with increased future smoking cessation activity, and smokers reporting previous quit attempts are more likely to make quit attempts in the future.4,16–19 Also, longer periods of abstinence from smoking in previous quit attempts are associated with increased chances of stopping in subsequent attempts.14
Intentions about quitting
Smokers who state they intend to stop are more likely to do so.3,16,20 The transtheoretical model of behaviour change4,21 uses smokers' intentions to quit (or not) to categorize smokers into different ‘stages of change'.22 Using this model, smokers' intentions (to quit) have been demonstrated as being associated with making future quit attempts.4
Motivation/desire to stop
Smokers who desire or want to stop smoking are more likely to achieve this.14–16
Confidence in ability to stop (self-efficacy)
This attribute is associated with successful smoking cessation.13,16,20 The evidence that smokers who have greater self-efficacy are more active in trying to quit smoking has been summarized by others.23
Attitudes towards smoking
There is less consensus about the relationship between smokers' attitudes towards smoking and their subsequent smoking behaviour, as little work has been conducted on this topic. Developing negative attitudes to smoking may pre-empt quit attempts,16 and smokers who have never made a quit attempt are less likely than others to acknowledge the health risks that smoking poses.12 We could find no prospective studies that explore the relationship between attitudes towards smoking and subsequent smoking behaviour.
Coding of smokers' consulting behaviour
TC viewed all consultations between GPs and self-reported smokers, and if either a GP or patient mentioned smoking during a consultation, this was used for further analysis. This approach identified 47 consultations between 47 occasional and regular smokers and 29 different GPs. For each consultation, TC transcribed verbatim the conversation about smoking which took place. TC and KS then independently watched each consultation whilst reading the appropriate transcript and coded smokers' consulting behaviour using the SMC.11,24 The SMC is an interaction analysis system, developed by TC and KS for this project—a process which is described in detail elsewhere.11,24 The SMC divides smokers' consulting behaviours into eight categories, four which are intended to indicate higher motivation or ‘readiness' to stop smoking and four which are intended to indicate lower motivation or ‘resistance' towards stopping smoking. The SMC is used with transcriptions of consultations to categorize smokers' consulting behaviours into one of these eight categories. Details of the behaviours described by the SMC are given in Boxes 1 and 2
. During the development of the SMC, TC and KS demonstrated that they could use the SMC to code smokers' consulting behaviours reliably, with each of them coding all 47 consultations (kappa = 0.71 for readiness behaviours and kappa = 0.73 for resistance behaviours11,24). To investigate whether or not clinicians unfamiliar with the SMC could use it reliably, a third researcher and GP (AW), who had not been involved in SMC development, coded a subsample of 11 (23%) of the 47 consultations. AW was given one 2 hours training session in using the SMC before beginning coding.
Box 1 Smokers' resistant consulting behaviours
|
Box 2 Smokers' readiness consulting behaviours
|
Analysis of data—reliability
Cohen's kappa25 was used to assess inter-observer agreement between the two pairs of observers TC/AW and KS/AW. For each pair, two levels of agreement were tested:
- Whether or not observers agreed that any resistant behaviour was present.
- Whether or not observers agreed that any readiness behaviour was present.
Analysis of data—construct validity
Next, to assess the construct validity of the SMC, we investigated relationships between smokers' consulting behaviour as coded by the SMC and relevant pre-consultation questionnaire variables. For this analysis, we used the codings of the GP who helped develop the SMC (TC). Smokers' observed consulting behaviours (as coded by TC) were, therefore, compared with their attitudes towards smoking and their past, present and intended (future) smoking behaviours as recorded on the pre-consultation questionnaire. Table 1 demonstrates the nature of the relationships that we expected. We hypothesized that consulting behaviour indicating readiness to stop would be associated with more motivation to stop (as recorded on the pre-consultation questionnaire, and vice versa). Pre-consultation questionnaires were dichotomized at appropriate points (illustrated by Table 1) and simple chi-squared tests were used in this analysis.
|
Additionally, we explored the relationship between smokers' age and heaviness of smoking and the presence or absence of readiness and resistance behaviour. Due to the small numbers in each group, the two questionnaire items dealing with nicotine dependence were dichotomized.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Reliability
Tables 2–5
show the level of agreement between the pairs of observers (KS/AW and TC/AW), and the range of kappa values (0.65–1.0) suggests that agreement is good or very good.25 Confidence intervals on these values are wide, however, ranging from poor to excellent in each case, so we must interpret these findings with caution.
|
|
|
|
Construct validity
Characteristics of sample. Of the 47 discussions about smoking observed within video-recorded consultations, smokers' behaviours in five (11%) could not be coded as containing readiness or resistant behaviours. In the remaining 42 consultations, smokers exhibited readi-ness behaviour in 28 (67%) and resistant behaviour in 25 (60%)—readiness and resistant behaviours could be exhibited by the same smoker and 11 (23%) smokers did so. Of the 47 smokers, all completed the pre-consultation questionnaire and 37 (79%) smoked on at least most days (regular smokers), with the remainder being occasional smokers, smoking on less than most days or less frequently. As only regular smokers completed the pre-consultation questionnaire, only data from these 37 smokers could be used to test the construct validity of the SMC. The mean age of these 37 regular smokers was 42.3 years (SD = 16.7) with 32 (86%) smoking 20 or less cigarettes daily and 17 (46%) smoking their first cigarette within 30 min of waking. Four of the 37 regular smokers who discussed smoking with their GP were not coded as displaying either readiness or resistant behaviours, and of the remaining 33 smokers, 20 (61%) exhibited some resistance and 22 (67%) some readiness behaviour.
Relationships between smokers' behaviours and pre-consultation questionnaire variables..
There was no significant difference in the mean ages of smokers displaying readiness behaviours and those who did not [mean age of those displaying readiness, 45.2 years versus 37.9 years for those not (95% CI for difference between means, –3.9 to 18.5)]. Similarly, those displaying resistant behaviours had similar mean ages to those that did not [mean age of those displaying resistance, 41.0 years versus 44.0 years for those not (95% CI for difference between means, –14.4 to 8.3)]. There were also no significant differences between the variables which indicated heaviness of smoking and the presence or absence of either readiness or resistant behaviours. Table 6 summarizes the relationships between pre-consultation questionnaire variables and smokers' observed readiness behaviours. This demonstrates that where readiness behaviour was observed, smokers indicated more motivation to stop on pre-consultation questionnaires. In particular, smokers showing readiness behaviour were significantly more likely to have tried to give up in the past year, to be thinking about giving up and to believe that their health would improve if they stopped smoking. Table 7 summarizes the relationships between smokers' observed resistant behaviour and their pre-consultation variables. Smokers displaying resistant behaviour are significantly less likely to be thinking about or trying to stop smoking. Of the smaller, non-significant differences in the reported attitudes of smokers displaying and not displaying resistant behaviour, three are in the expected direction but three are not.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
To our knowledge, this is the first attempt to investigate whether or not smokers' behaviour during consultations with their GP can indicate their motivation to stop smoking. We have demonstrated that it is possible to describe smokers' consulting behaviour reliably using the SMC, though a larger study would be needed to determine how accurate our estimates of this reliability are. Although this is a small, exploratory study, the results are encouraging. Our findings suggest that where smokers display readiness behaviour, they report greater motivation to try stopping smoking. Of the seven constructs explored by pre-consultation questionnaire items, five differences were in an expected direction, three of which were statistically significant, with one remaining so after correction for multiple significance testing using a conservative Bonferroni correction.25 The results for resistant behaviour are slightly less encouraging. Smokers who were coded as displaying no resistant behaviours were more likely to indicate greater motivation to stop on four pre-consultation questionnaire items, with one of these differences reaching statistical significance. The remaining three differences were in an unexpected direction. In summary, of the 14 comparisons made in construct validity assessment, only three were in an unexpected direction. There is, therefore, some evidence to support the notion that descriptors are valid.
Our work complements that which has used questionnaires to predict the readiness of smokers7,8 and heavy drinkers5,6 to change their unhealthy behaviours. Study findings suggest that smokers who report different levels of motivation to stop smoking on pre-consultation questionnaires also exhibit different behaviours during consultations with their GP when discussing smoking. Although exploratory, this work could be a first step towards helping clinicians identify which smokers are most likely to respond positively to their stop smoking advice by making and perhaps being successful at a quit attempt. Being able to do this as patients consult would remove the need for clinicians to use questionnaires to assess smokers' motivation to stop smoking.
Study findings could be biased if smokers altered their consulting behaviour as a consequence of completing the pre-consultation questionnaire. This seems unlikely though, as patients generally have other concerns during their consultations26 and also there was no possible secondary gain for them from altering their consulting behaviour. The validity of study findings could also be compromised by either doctors or patients altering their consulting behaviour as a consequence of video-recording. No studies have investigated whether patients' awareness of video-recording influences their consulting behaviour, but where their views have been sought immediately after recorded consultations, recording appears to have been of little consequence to them.27 Also, in our study, most conversations about smoking were initiated by GPs, suggesting that data collection procedures did not trigger many patients towards discussing smoking. Additionally, GPs' awareness of being video-recorded has been found to have no influence on their consulting behaviours, including the amount of time devoted to preventive medicine.28 It is plausible, therefore, that our video-recorded consultations displayed the ‘normal' behaviours of GPs and patients.
It is worth noting that we have not categorized smokers as ‘motivated' or ‘not motivated' to try stopping smoking. Indeed, 23% of smokers demonstrated both readiness and resistance within a single consultation, perhaps reflecting the real ambivalence that many smokers feel about smoking.4 Further work is needed to refine our method of coding smokers' behaviours for use as a research tool. In particular, it would be useful to explore whether coding without transcriptions is possible and to investigate further the SMC's reliability. It is most important, however, to assess whether the SMC has predictive validity.29 This means we need to assess whether smokers' observed behaviours in their consultations are related to their future smoking behaviour. If predictive validity were demonstrated, our work could be used to develop an evidence-based approach towards the issue of smoking with individual patients in their consultations. It could become possible to tailor anti-smoking advice effectively to smokers' levels of motivation.
|
Acknowledgments |
|---|
We are indebted to Mrs Margaret Whatley for secretarial help. This study was funded by the Scientific Foundation Board of the Royal College of General Practitioners and Trent NHS Executive.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Health Education Authority. The Smoking Epidemic: A Prescription for Change. London: HEA, 1993.
2
Ashenden R, Silagy C, Weller D. A systematic review of the effectiveness of promoting lifestyle change in general practice. Fam Pract 1997; 14:160–176.
3 Russell MA, Wilson C, Taylor C, Baker CD. Effect of general practitioners' advice against smoking. Br Med J 1979; 2:231–235.
4 DiClemente CC, Prochaska JO, Fairhurst SK, Velicer WF, Velasquez MM, Rossi JS. The process of smoking cessation: an analysis of precontemplation, contemplation, and preparation stages of change. J Consult Clin Psychol 1991; 59:295–304.[Web of Science][Medline]
5 Heather N, Rollnick S, Bell A. Predictive validity of the Readiness to Change Questionnaire. Addiction 1993; 88:1667–1677.[Web of Science][Medline]
6 Rollnick S, Heather N, Gold R, Hall W. Development of a short ‘readiness to change' questionnaire for use in brief, opportunistic interventions among excessive drinkers. Br J Addiction 1992; 87:743–754.[Web of Science][Medline]
7 Richmond RL, Bell AP, Rollnick S, Heather BB. Screening for smokers in four Sydney teaching hospitals. J Cardiovasc Risk 1996; 3:199–203.[Medline]
8 Cordoba R, Martin C, Casas R et al. Value of brief questionnaires in predicting smoking cessation in primary care. Atencion Primaria 2000; 25:32–36.
9
Coleman T, Williams M, Wilson A. Sampling for qualitative research using quantitative methods. 1. Measuring GPs' attitudes towards discussing smoking with patients. Fam Pract 1996; 13:526–530.
10
Coleman T. Sampling for qualitative research using quantitative methods. 2. Characteristics of GPs who agree to video-taping of consultations. Fam Pract 1996; 13:531–535.
11 Coleman, T. Anti-smoking advice in general practice consultations: a description of factors influencing provision of advice and the development of a method for describing smokers' responses. MD thesis, University of Leicester, 1998.
12 Lennox AS, Taylor RJ. Factors associated with outcome in unaided smoking cessation, and a comparison of those who have never tried to stop with those who have. Br J Gen Pract 1994; 44:245–250.[Web of Science][Medline]
13 Lennox AS. Determinants of outcome in smoking cessation. Br J Gen Pract 1992; 42:247–252.[Web of Science][Medline]
14 Jackson PH, Stapleton JA, Russell MA, Merriman RJ. Predictors of outcome in a general practitioner intervention against smoking. Prev Med 1986; 15:244–253.[Web of Science][Medline]
15 Richmond RL, Kehoe LA, Webster IW. Multivariate models for predicting abstention following intervention to stop smoking by general practitioners. Addiction 1993; 88:1127–1135.[Web of Science][Medline]
16 Marsh A, Matheson J. Smoking Attitudes and Behaviour. London: HMSO, 1983.
17 Marlatt GA, Curry S, Gordon JR. A longitudinal analysis of unaided smoking cessation. J Consult Clin Psychol 1988; 56:715–720.[Web of Science][Medline]
18 Curry S, Thompson B, Sexton M, Omenn GS. Psychosocial predictors of outcome in a worksite smoking cessation program. Am J Prev Med 1989; 5:2–7.[Web of Science][Medline]
19 Russell MA, Stapleton JA, Feyerabend C et al. Targeting heavy smokers in general practice: randomised controlled trial of transdermal nicotine patches. Br Med J 1993; 306:1308–1312.
20 Sanders D, Peveler R, Mant D, Fowler G. Predictors of successful smoking cessation following advice from nurses in general practice. Addiction 1993; 88:1699–1705.[Web of Science][Medline]
21 Prochaska JO, DiClemente CC. Stages and processes of self-change of smoking: toward an integrative model of change. J Consult Clin Psychol 1983; 51:390–395.[Web of Science][Medline]
22 Prochaska JO, Velicer WF, DiClemente CC, Fava J. Measuring processes of change: applications to the cessation of smoking. J Consult Clin Psychol 1988; 56:520–528.[Web of Science][Medline]
23 Strecher VJ, McEvoy B, Becker MH, Rosenstock IM. The role of self-efficacy in achieving health behaviour change. Health Educ Q 1986; 13:73–91.[Web of Science][Medline]
24 Coleman T, Stevenson K, Wilson A. Using content analysis of video recorded consultations to identify smokers ‘readiness' and ‘resistance' to stopping smoking. Patient Educ Counsel 2000; 41:305–311.[Web of Science][Medline]
25 Altman DG. Practical Statistics for Medical Research. London: Chapman and Hall, 1991.
26 Cromarty I. What do patients think about during their consultations? A qualitative study. Br J Gen Pract 1996; 46:525–528.[Web of Science][Medline]
27 Martin E and Martin PM. The reactions of patients to a video camera in the consulting room. J R Coll Gen Practitioners 1984; 34:607–610.[Web of Science][Medline]
28 Pringle M, Stewart-Evans C. Does awareness of being video recorded affect doctors' consultation behaviour? Br J Gen Pract 1990; 40:455–458.[Web of Science][Medline]
29 Striener DC, Norman GR. Health Measurement Scales: A Practical Guide to Their Development and Use. London: Oxford University Press, 1989.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||