Asthma is not enough : Continuation of smoking among parents with an asthmatic child

1Section of Allergy and Clinical Immunology, Department of Pediatrics and Child Health; 2Department of Community Health Sciences, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba; 3Department of Sociology, University of Victoria, Victoria, British Columbia; 4Manitoba Institute of Child Health, University of Manitoba, Winnipeg, Manitoba Correspondence: Dr Allan B Becker, Section of Allergy and Clinical Immunology, Department of Pediatrics and Child Health, Room AE101, 820 Sherbrook Street, Winnipeg, Manitoba R3A 1R9. Telephone 204-787-2537, fax 204-787-5040, e-mail becker@cc.umanitoba.ca JJ Liem, AL Kozyrskyj, CM Benoit, AB Becker. Asthma is not enough: Continuation of smoking among parents with an asthmatic child. Can Respir J 2007;14(6):349-353.

One would expect that parents, on learning that their child is asthmatic and given the knowledge that cigarette smoking exacerbates the illness, would give up smoking or at the very least smoke outside the home (9).Yet, existing research has not addressed this issue.Part of the difficulty is that the prevalence of smoking in the general population has decreased significantly over the past decade, but the reason for this decrease may have little or nothing to do with a specific child's diagnosis of asthma.Interventions to change parental smoking behaviour have largely been unsuccessful (9)(10)(11)(12)(13)(14)(15)(16).In the present study, we asked whether parents would stop smoking once they knew their child had asthma.

METHODS
The present study was approved by the Health Information Policy Committee of Manitoba Health and by the Health ©2007 Pulsus Group Inc.All rights reserved

ORIGINAL ARTICLE
Research Ethics Board of the University of Manitoba, Winnipeg, Manitoba.
A mail-out survey to the parents of children born in the province of Manitoba in 1995 was sent on our behalf by Manitoba Health in November 2002 and again in January 2003.The survey focused on the health of the child and the child's family, including atopic (ie, does your child have asthma, hay fever or food allergies?)and nonatopic conditions (ie, does your child have diabetes, inflammatory bowel disease or arthritis?).As well, there were questions (current and retrospective) about the child's home environment: Does anyone living in this home smoke (cigarettes or tobacco)?Do they smoke inside the home?In 1995, did anyone living in your home smoke (cigarettes or tobacco)?Did they smoke inside the home in 1995?
A sample of the survey is appended to the end of the manuscript.
Methods to increase the response rate of the survey, such as the use of a short, colourful survey and two mail-outs, were used.
No incentives were provided for the return of the surveys.Returned surveys were stratified by postal code into rural and urban locations.Urban families were stratified by postal code into five income quintiles.Low and high incomes were defined by the bottom and top quintiles, respectively.
Smoking behaviour change was defined as having quit smoking if parents reported indoor or outdoor smoking in 1995 but not in 2002/2003, smoking outdoors if parents reported indoor smoking in 1995 and outdoor smoking in 2002/2003, and any change if parents quit smoking or smoked outdoors in 2002/2003.The likelihood of smoking behaviour change was related to the diagnosis of asthma in the child born in 1995, adjusted for family history of asthma or allergy (hay fever and food allergy), household income and urban or rural location.
Data analysis was performed with SAS software (SAS Institute, USA), and χ 2 tests were used to determine significant differences in rates of smoking.Multivariate logistic regression analysis was performed to determine the likelihood of smoking change.Statistical significance was set at P<0.05.Results are reported as OR with 95% CI.

RESULTS
A total of 16,384 children were born in the province of Manitoba in 1995.Of these, 13,980 children lived in the province in 2002.The surveys were sent to all of these families, except for those living in a First Nations community ('reserves', which will be studied separately) (n=12,556).The first and second mail-outs resulted in 2114 and 1725 returned surveys, respectively.A total of 3580 surveys (28.5%) were returned.
The demographics of survey participants are shown in Table 1.The urban to rural ratio reflects the population distribution of Manitoba.
The second mailing of the survey increased the response rate by 12%, but the final response rate was limited by the 17% response rate to the first mailing and the 809 noncurrent mailing addresses.The fact that the survey participants were representative of the asthmatic population was reassuring.The overall prevalence of asthma was 12.3 per 100 children.Asthma was reported in 7.9% of children living in northern rural areas, 10.3% in southern rural areas and 13.9% in urban areas.
A comparison of respondents to nonrespondents is shown in Table 2.
A total of 1151 surveys (32.2%) indicated that a smoker was in the home in 1995 when their child was born.A similar proportion (approximately 33%) of homes that had a smoker present was reported for participants living in both urban (32.3%) and rural (31.9%) locales.In 2002/2003, the prevalence of smoking in the home had decreased from 32.2% (n=1151) to 23.4% (n=839).This decline as shown in Figure 1 was similar for participants residing in both urban and rural locations (23.6% and 23.2%, respectively).
Just over 12% (n=439) of households reported that their child had asthma in 2002/2003.Among this group, 40.0%(n=171) had someone who smoked in 1995.Eight years later, the number of homes with smokers decreased to 29.8%.The prevalence rate was significantly greater than homes without an asthmatic child, where the smoking rates dropped from 31.2% in 1995 to 22.5% in 2002/2003.Homes with an asthmatic child were more likely to have a smoker present in 1995 (P=0.001) and in 2002/03 (P<0.001)(Figure 2).
When compared with homes without an asthmatic child, parents of asthmatic children were not more likely to quit smoking or smoke outside.This was independent of family history of asthma, urban or rural location, and household  income (Table 3).As well, parental smoking behaviour (quit smoking or smoke outside) did not change if there was a positive family history of asthma, if they lived in a rural versus an urban location, or if they were from a low-versus high-income household (Table 4).

DISCUSSION
Over the past decade, the prevalence of smoking has decreased.In Canada, there has been an 8.8% decrease in smokers aged 15 years or older from 1994/1995 to 2001 (most recent data available) (17).The results from our study indicate a similar downward trend -dropping from 32.2% in 1995 to 23.4% in 2002/2003.The aim of our study was to determine whether a specific health factor, ie, a diagnosis of asthma in a child, could play a role in changing parental smoking behaviour.Our data suggested that this was not the case; rather, the parents participating in our study with a child diagnosed with asthma were not any more likely to alter their smoking behaviour than counterparts without an asthmatic child.
The present study suggests that ETS exposure is a risk factor for the development and persistence of childhood asthma (Figure 2).It appears that many parents continue to smoke, in spite of having children with asthma.Results from the Childhood Asthma Management Program (CAMP) study (11) revealed that of the children who reported cigarette smoking causing asthma symptoms, 26% had at least one parent who continued to smoke cigarettes.Given a similar proportion of parents who smoked in our study, this suggests, even if smoking causes obvious symptoms in the child, that this relationship does not induce parents to stop.Weiss et al (11) concluded that although environmental modification of asthmatic homes may occur, many children remain exposed to agents involving tobacco smoke that are known to trigger childhood asthma.It appears that the parents' altruism toward the wellbeing of their children may not be sufficient to give them the ability to quit smoking.
Many social factors are likely to play a role in the overall decrease in smoking prevalence.Public pressure resulting in changes to governmental health policy appears to have a major impact (ie, increasing taxes for cigarettes and banning smoking in public settings).What, if anything, can be done to help the parents reduce or eliminate exposure of their asthmatic child to tobacco smoke?Most interventions have, to date, been ineffective in changing parental smoking behaviour even when parents are well aware that their asthmatic child's health may improve (9)(10)(11)(12)(13)(14)(15)(16)18).For example, Wakefield et al (10) attempted to provide feedback to low-income parents by using urinary cotinine levels and advice on home smoking restrictions to help decrease exposure.Unfortunately, the intervention proved to be unsuccessful (10).
In fact, some studies (18,19) suggest that counselling parents in smoking cessation may be detrimental.Irvine et al (18) found that a brief intervention to advise smoking parents about the risks of passive smoking to their asthmatic child may have made some parents less inclined to quit smoking.Their conclusion was that if a clinician believes that a child's health is being affected by parental smoking, the parent's smoking needs to be addressed as a separate issue from the child's health.Continuing to 'badger' a family to quit smoking may even turn smoking parents away from using preventive health services due to the stigma that parents may feel because they are unable to stop the habit (19).In one of the few positive studies, Hovell et al (20) showed a significant decrease in second-hand smoke exposure among mothers of preschool asthmatic children after providing seven counselling sessions over a three-month period.The authors concluded that this intensive form of counselling service should be made available to all.However, the cost burden of this intervention can be enormous and likely unacceptable to most publicly funded health care systems.This study demonstrates the positive effect of social support for these mothers.They are likely isolated with their child for long periods of time each day, and social supports may be critical to help alter negative behaviours.
We hypothesized that there would be greater resistance to smoking cessation among parents who participated in the survey from a lower socioeconomic status or from rural locales; however, we found no differences when compared with participants from high-income and urban locales, respectively.What this seems to indicate is that a minority of parents, regardless of social class and geographical location, are highly addicted to cigarette smoking.
For a general health and population survey, our response rate of 28.5% is greater than a recent general population health survey of comparable size, in which a response rate of only 20% was achieved (21).This is often considered a low response rate but only when compared with specific population surveys or problem-specific surveys (22,23).Our survey demographic population reflects the overall population in the province of Manitoba (63% urban and 37% rural, according to Statistics Canada 2001 census) (24), although we had slightly more surveys returned from southern rural regions compared with northern rural regions.The male to female response rate coincided with the sex distribution of children born in the province in 1995, as expected.There was a slight over-representation of returned surveys from those in the higher income quintiles; however, there was good representation from each of the income groups.Finally, we were reassured by how representative our survey was of the asthmatic population.The overall prevalence of asthma was 12.3 per 100 children.Asthma was reported in 7.9% of children living in northern rural areas, 10.3% in southern rural areas and 13.9% in urban areas.This geographic distribution was virtually identical to the prevalence of health care prescriptions for asthma as defined by the Manitoba Health databases for the full 1995 birth cohort (7.9% in northern rural areas, 9.6% in southern rural areas and 13.8% in urban areas) (25).Thus, the survey results were a good reflection of the entire 1995 Manitoba birth cohort.
There may have been an under-reporting of smoking behaviour in our survey because parents may have associated the question of smoking with the question regarding asthma.However, the survey was sent out as a general 'child health and environment' questionnaire, and questions regarding the health of their child -ie, does your child have asthma?-were separated from the questions regarding participant's smoking behaviour.The fact that a greater proportion of surveys indicated that smokers were present in the homes and that their child had asthma compared with homes without children with asthma suggests that participants likely did not associate the two questions.Furthermore, the prevalence of smokers in the survey were similar to Statistics Canada data of trends of smoking across the country (17).
The prevalence of smoking in the province of Manitoba, like elsewhere in Canada, has significantly decreased in the past decade or so.Public pressure and changes to health care policy are likely among the major factors behind this trend.Nevertheless, there is still a substantial minority of Canadian adults who appear to be unable to quit smoking, many of whom are parents with children living at home.

CONCLUSIONS
Smoking is a major preventable cause of illness and death, even for those who are passively exposed.Advising parents that their asthmatic child's condition is worsened by ETS is not enough to change behaviour.Innovative social policies need to be developed to help addicted individuals overcome their habit and deal with the underlying causes that promote smoking in the first place.Only then can the health burden for parents and their children be reduced if not completely eliminated.
Parents of asthmatic children continue to smoke Can Respir J Vol 14 No 6 September 2007 351

Figure 1 )Figure 2 )
Figure 1) Smoking prevalence in the province of Manitoba in 1995 and 2002/2003.The black bar represents total smoking prevalence.The white and grey bars represent the smoking prevalence in urban and rural locales, respectively.There were no statistically significant differences between locations

TABLE 2
*Current asthma diagnoses were determined by linking nonrespondent surveys to Manitoba Health Services Insurance Plan database records, and by determining which children had asthma as defined by the International Classification of Diseases ninth revision (ICD-9) code of 493 in-hospital or medical claims, or a prescription of an asthma medication (inhaled corticosteroids, inhaled or oral bronchodilators, cromolyn glycate and leukotriene receptor antagonist) in 2002; † Maternal asthma in 1995 was determined by linking nonrespondent surveys to 1995 Manitoba Health Services Insurance Plan database records, and by determining which mothers had asthma as defined by the ICD-9 code of 493 in-hospital or medical claims, or a prescription of an asthma medication (inhaled corticosteroid, inhaled or oral bronchodilators, cromolyn glycate and leukotriene receptor antagonists) in 1995

TABLE 3
Adjusted for family history of asthma, rural or urban location, and income *