Physician counseling of pregnant women about active and secondhand smoking in Argentina
Eliseo J Pérez-Stable, Department of Medicine, Division of General Internal Medicine, Medical Effectiveness Research Center for Diverse Populations, UCSF Box 0856, 3333 California Street, San Francisco, CA 415-502-4088, USA. E-mail: email@example.com
Objective. Describe physicians' practices of smoking cessation and secondhand smoke (SHS) exposure counseling during prenatal visits. Design. Cross-sectional survey. Setting. A total of 13 public and private hospitals from three cities in Argentina. Population. A total of 300 obstetrician/gynecologists. Methods. Self-administered survey included knowledge and attitudes about tobacco use during pregnancy, frequency, type and duration of smoking cessation counseling, barriers to counseling, communication skills, level of understanding and personal smoking history. Main outcome measures. Composite outcomes of four items, each representative of counseling on smoking cessation and SHS exposure. Results. A total of 235 (78.3%) questionnaires were completed; 54.5% men, mean age 45, 35% current smokers. Only 22% had received training in smoking cessation counseling and 48.5% reported insufficient knowledge to provide smoking cessation advice. Although 88.9% always or almost always advised women to stop smoking, 75% believed it was acceptable for pregnant women to smoke up to 6 cigarettes per day. The risk of SHS exposure was ‘always or almost always discussed’ by only 34.5% of physicians. Multivariate logistic regression showed that lack of training was associated with less counseling about smoking cessation (OR 0.18; 95% CI 0.04–0.82) and SHS exposure (OR 0.27; 95% CI 0.12–0.59). Current compared to never smokers had lower odds of smoking cessation counseling (OR 0.39; 95% CI 0.05–0.82). Current smokers were less likely than former smokers to counsel about SHS (OR 0.25; 95% CI 0.11–0.62). Conclusions. Smoking cessation counseling during pregnancy in Argentina occurs infrequently, interventions are needed to assist physicians motivate and counsel women to quit smoking and avoid SHS exposure. Physicians taking care of pregnant women also need to quit smoking.
In 2006, according to a report from the Argentinian Ministry of Health and Environment, 28.6% of women 18–64 years of age had smoked cigarettes in the previous 12 months (1). The highest prevalence was among women who were 18–49 years old (2) and smoking rates increased with education among women, peaking among those with partial high school (29.6%) and decreasing slightly thereafter (3). In a recent survey of 800 pregnant women from public hospitals in Argentina, 44% had been or were regular smokers and 11% continued smoking during pregnancy (4). Furthermore, secondhand smoke (SHS) exposure is prevalent among pregnant women in Argentina with 49% of women who had never smoked reportedly living with a smoker, 67% of those who had quit and 78% of those who continued smoking (4). There is compelling evidence that maternal tobacco consumption has harmful effects on the fetus (5–9). Maternal tobacco consumption has postnatal consequences such as neonatal death (10), respiratory infections, asthma (11) and sudden infant death syndrome (12). Exposure to SHS during pregnancy is also associated with low birthweight among other adverse effects (13). Thus, assisting pregnant women to quit smoking and avoid SHS during pregnancy is a public health priority.
Approximately 20–40% of women quit smoking during pregnancy and most of them quit after their first prenatal care visit (14–17). Furthermore, physician mediated interventions are effective for reducing tobacco consumption during pregnancy (18–20). Brief physician counseling has been shown to increase the smoking cessation rates of patients in primary care settings, is cost effective when a minimum of 5 minutes are used (21) and is widely recommended (22, 23). However, it is estimated that fewer than half of the physicians, from different countries, recommend smoking cessation to their pregnant patients (16, 24). There is insufficient evidence of effectiveness and some potential risk to recommend use of pharmacological treatments in smoking cessation in pregnant women (22).
In Latin America, the role of physician counseling in smoking cessation has not been emphasized despite the recent development of national guidelines (23). As smoking rates increase among women, the prenatal visit provides an opportunity for physician counseling intervention to promote smoking cessation and decrease SHS exposure. Little is known about physicians counseling about smoking cessation and SHS exposure in their pregnant patients in Argentina. The objective of this study was to estimate the prevalence and describe the type of smoking cessation and SHS exposure counseling provided by physicians during prenatal visits in Argentina in order to identify policy needs that would target obstetricians.
Material and methods
A cross-sectional self-administered survey was distributed to all physicians taking care of pregnant women in 13 urban hospitals. Physicians were sampled from 5 hospitals in Salta, 5 in Jujuy and 3 in Buenos Aires. Salta and Jujuy provinces are located in the northwest of Argentina and constitute the principal tobacco-growing regions. All physicians caring for pregnant women in the 13 hospitals were identified from administrative lists. The questionnaire and the consent form were delivered to the physician's outpatient office and were completed by the participants between 1 March and 31 August 2005.
The questionnaire included items translated from surveys used in the US (Tobacco Use in California Survey) (25), the CDC Global Health Professional Student Survey (26), and in Guatemala (27), and items developed by the authors. Items in English were translated to Spanish, back translated to English and reviewed by three Argentinean investigators. Pre- test of the instrument was conducted with physicians of the Hospital de Clinicas, University of Buenos Aires.
The survey consisted of 41 questions and took about 15 minutes to complete. There were questions regarding demographic background, medical training received, characteristics of the medical workplace (public or private hospital), personal history of tobacco use and intentions to quit. Questions about knowledge and attitudes about tobacco use, frequency, type and duration of smoking cessation counseling provided to pregnant women, and SHS exposure counseling provided to their patients were also included. Finally, barriers to tobacco counseling, communication skills in tobacco cessation and training received in tobacco counseling were queried.
Data were analyzed using SAS and descriptive statistics reported means and standard deviations. Multivariate logistic regression models were constructed to identify physician factors associated with providing smoking cessation counseling to pregnant patients. Consistent provision of smoking cessation counseling was defined by physician responses of ‘always’ or ‘almost always’ engaging in each of the following four activities with their pregnant patients. First, advice to reduce the number of cigarettes smoked per day; second, explain the risks of active smoking; third, inform about the benefits of smoking cessation; and fourth, explain the risks to the fetus of continued smoking. Similar models were constructed to identify physician factors associated with counseling about SHS exposure. Consistent provision of SHS exposure counseling was defined by physicians indicating ‘always’ or ‘almost always’ performing each of the following four activities with their pregnant patients. First, asking about other persons who smoke at home; second, asking about who smokes around them at work; third, explaining the meaning of SHS; and fourth, explaining the risks of SHS to self and others. Modeled explanatory factors included age in 10-year increments, gender (woman as referent), self-identified practice as obstetrician, gynecologist and obstetrician, or neither (referent), location of hospital (Buenos Aires as referent), physician smoking behavior (never smoker as referent), work in private or public setting (private as referent), average number of pregnant patients seen per day (> 10 as referent) and any training in smoking cessation.
Of 300 questionnaires distributed, we received 235 completed for a 78.3% response rate. Respondents were 55% men, had a median age of 45 years (SD 12.7) and 18% self-defined as obstetricians only, and 20% as gynecologists only. Among respondents, 14% worked exclusively in public hospitals, 24.7% in private hospitals only and 60.9% in both, and on average they saw 10 patients/day in the ambulatory setting (mean 10.3; SD 6.27) (Table 1). Of these physicians, 62 (26.3%) were former smokers, 83 (35.3%) were current smokers and over half of current smokers, 47 (56.6%), wanted to stop.
Table 1. Demographics and training of 235 obstetrician/gynecologists from Buenos Aires, Salta and Jujuy, Argentina, 2006.
| 25–39||79 (33.6)|
| 40–49||75 (31.9)|
| ≥ 50||81 (34.4)|
|Working in a tobacco growing province||171 (72.7)|
|Specialty self-identified as|
| Gynecologist only||47 (20)|
| Obstetrician only||43 (18.3)|
| Obstetrician and gynecologist||113 (48.1)|
| No specific specialty||29 (12.3)|
| Residents||30 (12.8)|
|Type of hospital or workplace|
| Only public sector||33 (14)|
| Only private setting or practice||58 (24.7)|
| Works in both||143 (60.9)|
| Current smokers Former smokers Never smokers||83 (35.3) 62 (26.3) 86 (36.6)|
| Never received any training on counseling patients on tobacco issues||183 (77.8)|
|Patients seen in an average working dayb|
| 1–10 11–20 ≥ 21||149 (64.7) 59 (25.6) 22 (9.5)|
About half of physicians (48.5%) did not consider themselves to have sufficient knowledge of how to advise their patients on smoking cessation or SHS exposure. Only 51 respondents (22%) had received previous training in smoking cessation counseling, but almost half of these respondents (n = 24) mentioned that ‘attending a conference’ was the only training received.
The type of smoking cessation counseling provided by the physicians is shown in Table 2. Most physicians (89.3%) asked their pregnant patients about smoking and 88.9% recommended complete smoking abstinence during pregnancy. However, over half (56.8%) never encouraged their patients to use nicotine replacement therapy, and almost all (93.1%) never advised patients about the use of bupropion for smoking cessation. The risk of SHS exposure was always discussed by only 16.7% of physicians. However, 152 (64.6%) respondents would allow pregnant women to smoke up to six cigarettes per day if they were unable to completely stop smoking. Physician smoking status was significantly associated with SHS exposure counseling (p = 0.008).
Table 2. Type of cigarette smoking cessation and SHS exposure counseling provided by the 235 obstetrician/gynecologists in Buenos Aires, Salta and Jujuy, Argentina, 2006.
|Smoking cessation counseling|
| Ask patients about cigarette smoking?||210 (89.3)||20 (8.5)||5 (2.1)||None||None|
| Recommend that your patients stop smoking completely?||209 (88.9)||12 (5.1)||2 (0.8)||9 (3.8)||3 (1.2)|
| Suggest that the patient reduce the number of cigarettes?a||198 (85.3)||28 (11.9)||6 (2.6)||None||None|
| Explain the risks to their health of continued smoking?a||171 (73.1)||35 (15)||23 (9.8)||4 (1.7)||1 (0.4)|
| Inform the patient of the benefits of quitting smoking?a||155 (66.5)||49 (21)||17 (7.3)||6 (0.6)||6 (2.6)|
| Explain the risk of smoking to the fetus?a||184 (78.6)||27 (1.5)||15 (6.4)||6 (2.6)||2 (0.9)|
| Prescribe bupropion for smoking cessation?||None||2 (0.9)||2 (0.9)||12 (5.2)||216 (93.1)|
| Prescribe nicotine replacement therapy for smoking cessation?||14 (6)||13 (5.6)||36 (15.4)||38 (16.2)||133 (56.8)|
| Ask about SHS at home?b||63 (27)||61 (26.2)||49 (21)||13 (5.6)||47 (20.2)|
| Ask about SHS at work?b||57 (24.6)||39 (16.8)||50 (21.6)||20 (8.6)||66 (28.4)|
| Explain to the patient what it means to be a ‘passive smoker’?b||75 (32.2)||55 (23.6)||40 (17.2)||26 (11.2)||37 (15.9)|
| Explain the risks of SHS?b||92 (39.3)||47 (20.1)||34 (14.5)||22 (9.4)||39 (6.7)|
| Explain the risks of sudden neonatal death?b||84 (35.9)||53 (22.6)||42 (17.9)||23 (9.8)||32 (3.7)|
Table 3 shows results of a multivariate logistic regression model to define factors associated with cessation counseling about active smoking and counseling about SHS exposure. The only factor associated with counseling about both outcomes was any training in smoking cessation. Current smokers compared to never smokers were less likely to counsel about smoking cessation and current smokers compared to former smokers were less likely to counsel about SHS exposure. Former smokers tended to be more likely to counsel about SHS exposure compared to never smokers, but this did not reach significance (p = 0.07). Clinicians working in a public setting were less likely to provide smoking cessation counseling and men were more likely to counsel their patients about SHS exposure.
Table 3. Factors associated with physician counseling about risks of smoking and SHS exposure in pregnant women, Argentina, 2006.
|Age in 10 year increments||1.01 (0.68–1.50)||0.87 (0.64–1.20)|
|Men (women)||0.76 (0.34–1.71||3.49c (1.67–7.27)|
|Jujuy (Buenos Aires)||1.22 (0.43–3.45)||1.71 (0.70–4.15)|
|Salta (Buenos Aires)||1.06 (0.36–3.09)||2.21 (0.86–5.65)|
|Neither specialty (Ob/Gyn)||2.55 (0.49–13.29)||0.88 (0.33–2.35)|
|Gynecologists only (Ob/Gyn)||4.44 (0.78–25.4)||0.56 (0.23–1.35)|
|Public setting only (private only)||0.20c (0.05–0.81)||0.92 (0.28–3.06)|
|Both public and private (private only)||0.34 (0.11–1.09)||1.57 (0.71–3.48)|
|Current smoker (never)||0.39c (0.16–0.95)||0.55 (0.25–1.20)|
|Former smoker (never)||0.95 (0.30–3.04)||2.17 (0.93–5.04)|
|Current smoker (former)||0.41 (0.14–1.19)||0.25c (0.11–0.61)|
|Training: none||0.18c (0.04–0.82)||0.27c (0.12–0.59)|
|Patients per day: > 10||1.11 (0.49–2.54)||0.93 (0.47–1.84)|
To our knowledge, this is the first Argentinian study on the frequency with which physicians caring for pregnant women address tobacco use. According to the respondents of this study, Argentinian obstetricians and gynecologists asked about tobacco consumption in almost every medical visit with a pregnant patient. Obstetricians answered that they strongly encouraged smoking cessation to all their patients and recommended non-pharmacologic treatment for all. However, physicians infrequently discussed the harm of SHS. These observations are not unusual, as other studies have shown that most obstetricians and gynecologists reported identification of smoking status, but that the provision of smoking treatment was not consistent (28, 29). Any training in smoking cessation was significantly associated with counseling about both active smoking and SHS exposure even though the amount of training was quite limited for half of the physicians.
Although participants appeared to be well informed about the consequences of tobacco, most of them allowed their patients to smoke up to six cigarettes a day if they were unable to quit. This could reflect a lack of confidence in their own counseling skills since a number of studies have found that while physicians may advise pregnant women to stop smoking, the vast majority of them do not adopt a smoking cessation proactive attitude because they were particularly pessimistic that they could influence women's smoking behavior (28, 30–32). This could also be due to a lack of knowledge of the major adverse effects of tobacco smoke on pregnancy, and thus their previous answers may reflect a ‘respondent’ bias because participants tend to respond what they think they are expected to. Finally, poorly informed physicians may not be willing to disagree with their patients about quitting tobacco completely instead of the ‘harm reduction’ of fewer cigarettes per day in order to preserve the patient–physician relation. Since most of participants admitted to a lack of expertise in tobacco cessation counseling skills, they should probably give greater emphasis to the importance of advising and supporting pregnant women to stop smoking completely. These topics would potentially be addressed in a formal training program.
Other than smoking cessation training few factors were associated with smoking cessation or SHS exposure counseling. Since 35% of respondents were current smokers, they may have been inhibited to ask patients to ‘do what I say, not what I do’, and so were less likely to counsel patients about cessation or SHS exposure. Physicians in a private setting may spend more time with patients and thus have more opportunities to counsel about smoking cessation. Finally, male physicians were significantly more likely to counsel about SHS exposure and this may reflect a heightened parental concern.
As expected, few physicians recommended nicotine replacement therapy and almost none prescribed bupropion during pregnancy given that guidelines do not recommend their use (22). The role of pharmacological therapy to assist pregnant women quit smoking is in need of additional research as the limited evidence from randomized trials is inconclusive in terms of additional risk or benefit. It is likely that nicotine replacement therapy in pregnancy is less harmful to the fetus than a moderate amount of maternal smoking since it results in a slower increase in serum nicotine levels, it does not increase serum carbon monoxide levels or other tobacco-related combustion products and will limit the duration of nicotine exposure for the fetus if it leads to maternal smoking abstinence.
The limitation of this study is that the conclusions are based exclusively on the physician's self-reported answers. We did not ask patients about their experience with the obstetricians and gynecologists and we did not review the clinical records for checking the accuracy of the answers. However, most physicians would tend to over-report a behavior that is expected or considered recommended by experts and thus bias the responses to a higher rate of counseling than what really occurs. Additional research is needed to confirm these results, but given the context, the low rates of smoking cessation and SHS exposure counseling are striking.
In conclusion the obstetricians interviewed in this study are not taking full advantage of the unique opportunity pregnancy affords for smoking cessation intervention and are not using recommended methods to assist pregnant women to stop smoking. Furthermore, over one-third of respondents to this survey were current smokers and a first step in helping their patients quit smoking may be to quit themselves. Interventions to train physicians on tobacco counseling of pregnant smokers are urgently needed and may well increase counseling rates and decreased exposure to tobacco among pregnant women. Tobacco counseling needs to become widely integrated into routine antenatal care in Argentina and elsewhere.
This research was funded by grant No. TW05935 from the Tobacco Research Network Program, Fogarty International Center, National Institute of Drug Abuse, National Institutes of Health, USA.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.