|Home | About | Journals | Submit | Contact Us | Français|
To examine the relationship between physicians’ tobacco-related medical training and physicians’ confidence in their tobacco-related skills and smoking-related interventions with parents of child patients.
The survey was mailed to 800 family physicians and 800 pediatricians across Canada, with a corrected response rate of 65% (N = 900).
Physicians’ self-reported tobacco-related education, knowledge, and skills, as well as smoking-related interventions with parents of child patients. Cochran-Mantel-Haenszel χ2 tests were used to examine relationships between variables, controlling for tobacco-control involvement and physician specialty. Data analysis was conducted in 2008.
Physicians reporting tobacco-related medical education were more likely to report being “very confident” in advising parents about the effects of smoking and the use of a variety of cessation strategies (P < .05). Furthermore, physicians with tobacco-related training were more likely to help parents of child patients quit smoking whether or not the children had respiratory problems (P < .05). Physicians with continuing medical education in this area were more likely to report confidence in their tobacco-related skills and to practise more smoking-related interventions than physicians with other forms of training.
There is a strong relationship between medical education and physicians’ confidence and practices in protecting children from secondhand smoke. Physicians with continuing medical education training are more confident in their tobacco-related skills and are more likely to practise smoking-related interventions than physicians with other tobacco-related training.
Vérifier la relation entre la formation des médecins sur le tabagisme et leur confiance en leur habileté d’intervenir à ce sujet auprès des parents de jeunes enfants.
L ’enquête a été adressée à 800 médecins de famille et à 800 pédiatres du Canada, le taux de réponse corrigé étant de 65 % (N = 900).
Déclarations des médecins concernant leur formation, connaissances et habiletés au sujet du tabagisme, de même que leurs interventions à ce sujet auprès de parents de jeunes enfants. On a utilisé les tests de χ2 de Cochran-Mantel-Haenszel pour examiner les relations entre les variables, en tenant compte de l’engagement envers la réduction du tabagisme et de la spécialité du médecin. L’analyse des données a été faite en 2008.
Les médecins qui disaient avoir reçu une formation sur le tabagisme étaient plus susceptibles de se dire « très confiants » pour conseiller les parents sur les effets du tabac et sur l’utilisation de plusieurs stratégies pour cesser de fumer (P < ,05). En outre, ceux qui avaient eu une formation sur le tabac étaient plus susceptibles d’aider les parents de jeunes patients à arrêter de fumer, que les enfants souffrent ou non de problèmes respiratoires (P < ,05). Par rapport à ceux qui avaient eu un autre type de formation, les médecins qui profitaient d’une formation médicale continue sur ce sujet étaient plus susceptibles de se dire confiants en leur habileté dans le domaine du tabac et de faire des interventions à ce sujet.
Il existe une forte relation entre la formation médicale et la confiance et les pratiques des médecins pour protéger les enfants de la fumée secondaire. Les médecins qui reçoivent une formation médicale continue sont plus susceptibles de se dire confiants en leurs habiletés en matière de tabagisme et d’intervenir dans ce domaine que ceux qui ont reçu un autre type de formation.
In 2006, the US Surgeon General’s report on the health consequences of exposure to second-hand smoke (SHS) concluded that there is no “risk-free level” of exposure to SHS, and that exposure leads to increased risk of sudden infant death syndrome, respiratory infection, ear problems, and severe asthma among children.1 It is recognized that minimizing children’s exposure to SHS should be a priority for physicians who treat children.2–6
Physician advice to quit smoking is an effective component of a smoking cessation strategy.7 Several studies have examined the proportion of patients receiving advice to quit smoking5–12 but few have examined the level of advice offered to parents when children are the patients.13 In one US population-based survey, about half of parents who had children seen by physicians in the past year reported that they were asked about household smoking; of parents who smoked, less than half recalled being advised of health consequences of SHS exposure.14 In earlier surveys of pediatricians in Maine and Vermont, most pediatricians reported advising parents to quit smoking and spent on average 4 to 5 minutes with parents. However, negative parental expectations, lack of time, lack of skills or confidence, and perceptions of professional norms were noted as barriers to providing this advice.15–17
Several studies demonstrate that physician education on tobacco counseling is associated with increased comfort and practice in advising patients who smoke, and, in one randomized controlled trial with pediatric residents, with advising parents to quit.18–21 Also, tobacco cessation training might increase the success rate of helping patients to quit smoking.22 However, the tobacco-related content in medical curricula remains limited,23–25 as is research into its relationship with assisting parents of child patients to stop smoking.
This study examines the association between tobacco-related medical education, during and after training, and physicians’ confidence in their tobacco-related skills, and the advice and assistance offered to parents of pediatric patients to quit smoking.
In 2003, random samples of 800 family physicians and 800 pediatricians across Canada were obtained with the help of the College of Family Physicians of Canada and the Canadian Paediatric Society (CPS). A sample size of 1600 was chosen to provide 95% confidence intervals with at least 0.05 half-width precision around proportions estimates stratified by specialty and assuming a 50% response rate.
The survey employed methods recommended by Dillman for self-administered mailed surveys, including an introductory letter, a reminder postcard, and 2 follow-up mailings for nonrespondents.26
Practising members of the College of Family Physicians of Canada and the CPS whose primary practices were located in Canada were considered eligible to be surveyed. For the purposes of these analyses, those who reported no patient care (n = 26, 2.8%) were excluded. Furthermore, the CPS permits nonpediatrician members; therefore, to ensure participants were family physicians or pediatricians, respondents were excluded if they did not indicate a specialty or if they indicated a specialty other than pediatrics or family medicine (n = 4, 0.4%).
The questionnaire was drafted based on a literature review, with input from family physicians and pediatricians. It was pretested with practising physicians from both specialties. The goal of the questionnaire was to assess the tobacco-related role perceptions and practices of primary care physicians in addressing parental smoking when a child is the patient. We also assessed tobacco-related education, knowledge, and attitudes. The study was approved by the Centre for Addiction and Mental Health Research Ethics Board.
Completed questionnaires were received from 437 family physicians and 493 pediatricians (corrected response rate27 was 65% after ineligible respondents [eg, retired, moved out of Canada, no medical licence] were removed from the denominator, n = 169).
Approximately 47% of respondents were family physicians. About 54% were women. On average they had been practising for 19.7 years (range < 1 year to 65 years). Only 2.7% of respondents were current smokers of cigarettes, and 4.4% had used tobacco in any form within the last 6 months. Characteristics of respondents’ primary practices are presented in Table 1.
Participants were asked 7 questions about their confidence in their knowledge and skills to advise patients and parents of child patients on tobacco-related issues (Box 1). To examine the relationship between education and elevated confidence, responses to each item were dichotomized as “very confident” versus “somewhat” or “not very confident.”
How confident are you in your knowledge and skills to …
*Responses to these questions were combined to reflect the recommendation of any pharmacotherapy (ie, nicotine replacement or bupropion).
Participants were asked 6 questions about their practices with the parents of child patients with and without respiratory disease (Box 2). Responses were dichotomized as “all or most parents” versus “some” or “few or no parents” to examine the relationship between education and a high-degree of tobacco-related practice. Respondents indicating that the question was “not applicable” were excluded from analysis.
In the past year, approximately how many smoking parents of child patients with (without*) respiratory problems did you …
*Each practice question was asked twice: once for child patients with respiratory disease, and once for child patients without respiratory disease.
†Responses to these questions were combined to reflect the recommendation of any pharmacotherapy (ie, nicotine replacement or bupropion).
Participants were asked if, during medical training, they had participated in any full courses, lectures (≥ 1 hour), seminars or workshops, or clinical placements that dealt specifically with nicotine, tobacco, or smoking cessation. A second question asked respondents if they had taken any continuing medical education (CME) programs that dealt with nicotine, tobacco, or smoking cessation. Respondents were categorized into 1 of 3 categories: those who had taken tobacco-related CME, those who had not taken tobacco-related CME but had some tobacco-related education during their medical training, and those who reported no prior tobacco-related education or training.
The goal of this analysis was to examine tobacco-related education as it relates to practice. However, physicians involved in tobacco-related public education or advocating policy might be more likely to seek tobacco-related training and be more confident in their practices.18 Respondents were asked if they had participated in tobacco-related activities (ie, advocating smoke-free spaces, tobacco-control policies, sales, advertising, etc; educating groups about tobacco and smoking cessation; or other). Respondents who had participated in any of these activities (21.9%) were considered to be involved in tobacco control and were accounted for in the analysis.
Family physicians and pediatricians differ in their medical training and patient populations. These differences might affect their confidence and skills in addressing tobacco cessation with parents of child patients. Therefore, specialty was also controlled for in the analysis.
All data analyses were conducted using SAS, version 9.1. Cochran-Mantel-Haenszel χ2 tests were used to compare confidence in practice and tobacco-related practice dependent variables with previous tobacco-related training, controlling for both tobacco-control involvement and physician specialty. For tobacco-related practices, separate analyses were conducted for pediatric patients with and without respiratory problems. Breslow-Day tests of homogeneity, with a cutoff of P = .10, were used to confirm the assumption of homogeneity of odds ratios.
Overall, most physicians reported high levels of confidence in discussing the effects of smoking and the effects of SHS on child health with patients and parents of child patients (Table 2). However, confidence in discussing cessation aids or strategies was lower. Compared with those with no training, physicians reporting tobacco-related education during medical training were more likely to report high levels of confidence in advising parents on the effects of smoking and SHS on child health, the use of pharmacotherapy, and following up on quit progress (P < .05). Moreover, the odds of reporting high levels of confidence were 1.5 to 2.5 times greater for those with tobacco-related CME compared with those with tobacco-related education during their medical training (P < .05).
Almost half of physicians reported asking the smoking status of all or most mothers (n = 469, 52%) and fathers (n = 440, 49%) of child patients when the smoking status was unknown. Overall, about three-quarters of physicians reported discussing the effects of SHS with all or most parents of child patients with respiratory problems when they knew the parents smoked; and more than four-fifths reported advising parents to cut down on or quit smoking (Table 3). However, less than one-third of physicians reported suggesting pharmacotherapy or providing specific assistance to all or most of these parents. When child patients had no respiratory problems, less than half of physicians reported discussing the effects of SHS or advising all or most parents who smoked to cut down or quit; less than 20% reported advising all or most parents about pharmacotherapy or giving assistance with quitting (Table 4).
When child patients had respiratory problems, physicians with any tobacco-related education during medical training were more likely to discuss the effects of SHS with parents compared with physicians without tobacco education; CME did not increase the likelihood of discussing SHS. Tobacco-related education during training was not related to other interventions. Physicians with tobacco-related CME were more likely to intervene in all ways, except discussing the effect of SHS, compared with physicians with other training (P < .05) (Table 3).
When the child patients did not have respiratory problems, physicians with tobacco-related education during their medical education were more likely to assist most parents with quitting (Table 4) compared with physicians reporting no tobacco-related education. However, the odds of physicians with CME advising, following up with, and discussing SHS with all or most parents were at least 1.7 times greater compared with physicians with tobacco-related education only during their medical training (Table 4). This association was consistent regardless of the respiratory status of the child patients (Tables 3 and and44).
We found that most physicians who treat children advise parents about the effects of SHS when children have respiratory problems. Physicians reporting any tobacco-related education are significantly and consistently more likely to have increased confidence in skills in promoting tobacco cessation and to intervene with the parents of child patients (P <.05). Nonetheless, there is room for improvement. Although those with previous tobacco-related education are more likely to advise and assist parents with smoking cessation, the proportion of physicians performing these interventions remains at or below 50%. Burnett and Young13 also found that levels of advice and assistance for parents are low, unless the child patients had illnesses exacerbated by SHS.
The finding that those with tobacco-related CME are more likely to discuss the effects of SHS with parents when children do not have respiratory problems compared with physicians with only education during medical training underscores the importance of continuing education during medical training and supports the call for increased tobacco-related education at all levels for all primary care physicians who care for pediatric patients.25,28
That tobacco-related CME is related to both confidence and practice might be a function of the temporal proximity of CME and practice, or could indicate that physicians who are interested in tobacco issues are more likely to take related CME and to be involved in cessation practices with parents of child patients. Nevertheless, the association between CME and increased confidence and intervention remained after controlling for tobacco-control involvement.
Data from self-completed questionnaires might be subject to response bias; however, the validity of the data are maximized when response rates are high. We achieved a 65% corrected response rate, which is relatively high for surveys of physicians.29 Responses to practice questions might have been subject to “social desirability bias.” We attempted to minimize this effect by asking 3-point Likert questions. Respondents not wishing to identify little or no practice with parents would be able to indicate “some parents.” Although we asked about tobacco-related education during medical training before practice, the self-report and cross-sectional nature of this survey precludes causal inference, and those practising cessation techniques might be more likely to recall this training.
Our sampling did not limit respondents to pediatricians and family practitioners practising with a general patient population. The practice behaviour of physicians is complex, and this analysis does not allow us to conclude that education is the only factor related to practice. Despite these limitations, the strength of the associations between CME and confidence and practice suggests that tobacco-related training and, in particular, CME is beneficial.
The findings provide evidence that tobacco-related education is important in promoting physicians’ confidence in tobacco-related skills and in increasing interventions with parents who smoke when children are the patients. Furthermore, they suggest that ensuring tobacco-related content in the curricula of medical schools will be beneficial to the health of children and their parents. Continuing education after completion of training is particularly important to enhance and maintain physicians’ tobacco-related practice.
This work was supported by grants XG 02 to 012 from the SickKids Foundation. We thank the Institute for Social Research at York University in Toronto, Ont, for data collection. Dr Michèle Tremblay provided advice, support, and other contributions to the study. The College of Family Physicians of Canada and the Canadian Paediatric Society provided us with the samples of physicians.
Mr Victor and Dr Brewster were responsible for developing the research question, analyzing the data, and presenting results. All authors were involved in the research design, the development of the questionnaire, and the writing of this manuscript. Mr Victor had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
This article has been peer reviewed.