- Email: [email protected]
Anti-tobacco socialization in homes of African-American and white parents, and smoking and nonsmoking parents
JOURNAL OF ADOLESCENT HEALTH 1999;24:329–339
ORIGINAL ARTICLE
Anti-tobacco Socialization in Homes of AfricanAmerican and White Parents, and Smoking and Nonsmoking Parents PAMELA I. CLARK, Ph.D., ANNEMARIE SCARISBRICK-HAUSER, Ph.D., SHIVA P. GAUTAM, Ph.D. AND SARAH J. WIRK, B.S.
Purpose: To examine parental perceptions and behaviors with regard to teen smoking, comparing AfricanAmerican and white parents, and those who did and did not smoke. Methods: Focus groups consisting of African-American and white parents who smoked provided initial in-depth information. A computer-assisted telephone survey of a biracial sample of 311 parents of children ages 8 to 17 years provided more generalizable information regarding parental beliefs and behaviors. Results: Nearly 50% of households either allowed teen smoking, had no ground rules, or had set restrictive rules but never communicated them to the children. Compared to white parents, African-American parents felt more empowered to affect their children’s behaviors and were more likely to actively participate in anti-tobacco socialization within the home (all p values < 0.01). Among the African-American parents, 98% reported 18 years or older to be an appropriate age for teens to make up their own minds about using tobacco, whereas 26% of white parents thought 16 years to be an appropriate age (p < 0.001). Parents who smoked reported more frequent rule-making than those who did not smoke (p 5 0.02), but were more likely to believe that childhood tobacco use is inevitable (p 5 0.01). Conclusions: Many parents are not engaged in antitobacco socialization in the home. Differences in the degree of parental participation may contribute to the From Northeastern Ohio Universities College of Medicine, Rootstown, and the Institute for Policy Studies, University of Akron, Akron, Ohio. Address reprint requests to: Pamela I. Clark, Ph.D., Associate Professor of Epidemiology, Division of Community Health Sciences, Northeastern Ohio Universities College of Medicine, 4209 State Rte. 44, Rootstown, OH 44272. Manuscript accepted March 12, 1998.
variance in smoking prevalence between African-American and white children. © Society for Adolescent Medicine, 1999 KEY WORDS: Adolescents Tobacco Smoking Parenting Socialization Race/ethnicity
That adult intermediaries knowingly or inadvertently assist children in obtaining and using tobacco products is evident in that 3,000 children become regular smokers each day (1). Parents could, however, be the most effective and the most motivated adults to stand between tobacco and their children. They could act as protective gatekeepers to the family; providing anti-tobacco socialization within the home (2), and politically influencing schools, merchants, and government outside the home. Some understanding of the relationship between parents and childhood tobacco use is emerging. Childhood tobacco use has been found to increase with parental smoking status, lower parental educational attainment and social class, and living in single-parent homes (3–9). By portraying attitudes favorable to smoking and otherwise increasing normative expectations about tobacco use, parents may increase the likelihood a child will smoke (6,10,11). The influence of parenting styles also has been implicated, with teen smoking reported to be related
© Society for Adolescent Medicine, 1999 Published by Elsevier Science Inc., 655 Avenue of the Americas, New York, NY 10010
1054-139X/99/$–see front matter PII S1054-139X(98)00117-7
330
CLARK ET AL.
directly to measures of nonsupportive/neglectful parenting (12) and harsh/inconsistent parenting (13), and inversely related to nurturant/involved parenting (13). Family involvement also has been found to be a necessary component of school and community programs intended to decrease childhood use and intention to use tobacco (14,15). Although some parental characteristics (e.g., social class, marital status, educational attainment, and even parenting styles) may not be amenable to public health interventions, with enough practical information, it may be possible to mobilize parents in a campaign to decrease the tobacco industry’s access to children. In order to do so, it is necessary to have a better understanding of the perceptions that parents hold regarding their legitimate roles in protecting their children and the barriers that they perceive in fulfilling protective roles. This study combined qualitative and quantitative methods to investigate parental anti-tobacco socialization beliefs and behaviors, and to compare the beliefs and behaviors among parents in households with any adult smoker and those with no adult smokers. Because African-American smokers generally begin smoking at a later age than do white smokers (16 –20) and tobacco use rates of AfricanAmerican teens are significantly lower than those of white teens (21–24), indicating the possibility of significant group differences in anti-tobacco socialization within households, we compared the beliefs and behaviors of African-American parents with those of white parents.
Methods Moderated focus groups were used to provide indepth information about adult smokers’ knowledge, attitudes, and practices. Twelve focus groups were completed prior to commencing a computer-assisted telephone survey of parents of children, ages 8 to 17 years. Information derived from the focus groups guided the content of the telephone survey.
The Focus Groups Current, regular smokers from northeast Ohio (excluding the city of Akron) were recruited into the focus groups in March and April 1997. Recruitment was done by posting notices in public places and by word of mouth. Members of the groups were paid $35.00 for their participation. Group members were at least 18 years of age and were parents of children
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
aged 8 to 17 years. Several were grandparents with primary responsibilities for the care of grandchildren in the household. Groups were composed of either self-identified African-American smokers or selfidentified white smokers. To assure that subjects in the focus groups and the survey respondents were mutually exclusive, candidates for the focus groups had telephone exchanges not included in the sampling frame used for the survey. Procedures. The focus group sessions were audiotaped for subsequent transcription. Each lasted between 1.5 and 2 hours. The discussions were structured around a set of items in focus group discussion guidelines, including a set of open-ended questions to cover the maximum range of relevant topics, fostering group interaction that explored the participants’ knowledge, beliefs, and feelings. We used low moderator involvement. The number of groups provided enough repetition to confirm dominant themes. The analytic approach was that of Knodel (25). Briefly, audiotapes were transcribed and then converted into transcripts. An initial set of codes were developed, with topics and subtopics corresponding to each item in the discussion guidelines. Additional codes were created for topics that arose and were of interest, and codes were consolidated as necessary. Finally, detailed codes to use for analysis of specific topics (e.g., opinions about the inevitability of teens’ smoking) were developed. An overview grid was constructed, providing a descriptive summary of the content of the discussions. One axis had the topic headings and the other the focus group identifiers. The cells contained brief descriptions of the content by group and topic, and formed the basis for content analysis. Reliability of the data was improved by comparing statements within and across groups. Reliability of the analysis was improved by involving two members of the investigative team in the interpretations. The content was interpreted independently by each, then compared. Disagreements were discussed and resolved. At least two analysts were present during the focus groups, providing field notes to enhance further the accuracy of the data. Statements made during the discussions that needed further clarification were incorporated into the content of subsequent groups. In order to generate some information not contaminated by the group interactions and to provide some data common to both the focus groups and the
May 1999
survey respondents, the discussions were preceded by a brief self-administered questionnaire.
The Telephone Survey Parents of child(ren) in each of 311 households in Akron, Ohio, previously identified by random-digitdialing techniques for another study were interviewed using a computer-assisted telephone interviewing system during June and July 1997. The survey consisted of a set of mostly close-ended questions, and was 8 to 10 minutes in length. The survey items were planned to elicit household and family policies and behaviors, as well as the knowledge, opinions, and behaviors of individual respondents. The survey was administered only in English. Procedures. Each called number was recontacted up to three times, spread over the field period including day, evening, and weekend calls. Initial refusals were recontacted for conversion. Appointments were made whenever an eligible household was identified and an immediate interview was not possible. Appointments were made for respondent convenience including day and weekend hours. Those appointments also were pursued over the field period, with several attempts made to complete the appointed interview. After screening for the presence of age-eligible child(ren) in the household, the interviewer requested to speak to a parent or guardian. To increase the likelihood of an adequate sampling of adults who smoked, if one or both of the parents smoked, the interviewer asked to speak to a/the parent who smoked. Because parental attitudes and policies may be very different for younger children compared to older children, each respondent was asked for the ages of all children in the household. The child whose age was closest to 12 years was selected as the referent child. If two children had ages equidistant from 12 years, the older child (if ,18 years old) was chosen. A coin toss was used to choose between children of the same age. The respondent was asked to think of that child when responding to questionnaire items, and the request to focus on that child was reenforced several times during each interview. The study protocol was reviewed and approved by the Institutional Review Board for the Protection of Human Subjects of Northeastern Ohio Universities College of Medicine. Written informed consent was obtained from subjects in the focus groups, and
ANTI-TOBACCO SOCIALIZATION BY PARENTS
331
verbal consent obtained from adults in the survey sample. Statistical Analysis. The outcome variables were categorical and were first explored through frequency tables, appropriately stratified. Because the extreme categories in the variables with ordered choices (e.g., “strongly agree,” “agree,” “disagree,” “strongly disagree”) were used infrequently, most were collapsed to dichotomous outcomes (“strongly agree/agree” or “disagree/strongly disagree”) without loss of significant information. The independent variables were dichotomous: smoking status of the household’s adults and selfidentified race. Separate analyses were performed comparing respondents who did and did not smoke, and households with any adult smoker (defined as a smoker older than age 18 years) versus households without any adult smokers. Because both results were virtually the same, the data are presented comparing households in which there were and were not adult smokers. Households also were classified according to the self-reported ethnicity of the respondent. The associations of household smoking status and race with the categorical outcome measures were assessed by Chi-square analysis. Because the age of the child referent may contribute significantly to parents’ beliefs and behaviors, the distributions of children’s ages among African-American respondents versus white respondents, and respondents who did and did not smoke, were explored. Because the age distributions were not different, none of the results were adjusted for age of the child referent.
Results A total of 70 parents who smoked participated in the 12 focus groups. There were five African-American groups and seven white groups. The characteristics of the participants are shown in Table 1. The response rate for the telephone survey was 86% (311 completed of 362 attempted). Reasons for noncompletion were: no age-eligible children in the household (n 5 14), refused (n 5 23), interrupted interview (n 5 3), or telephone no longer in service (n 5 11). The relatively high response rate may have been attributable to the nature of the telephone list, that is, parents who had been randomly sampled but who had given permission to interview their children for an earlier study. Characteristics of the parent respondents, the households, and the child referents are shown in Table 2.
CLARK ET AL.
332
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
Table 1. Characteristics of the Focus Group Participants (n 5 70) Number female (%) Race/ethnicity (%) African-American parents White parents Education level (%) Did not graduate high school High school graduate Some college or technical school College graduate Household income ($) (%) ,10,000 10,000 –29,999 30,000 – 49,999 50,000 or more Mean age (years) (SD) African-American parents White parents Mean age of first smoking (SD) African-American parents White parents Mean number of children (SD) African-American parents White parents
32 (45.7) 32 (45.7) 38 (54.3) 4 (3.7) 15 (21.4) 45 (64.3) 6 (8.6) 5 (7.2) 26 (37.1) 25 (35.7) 14 (20.0) 45.9 (9.1) 43.1 (6.1) 19.9 (4.1) 14.2 (4.0) 2.3 (0.8) 2.1 (0.9)
SD 5 standard deviation.
Household Ground Rules Regarding Tobacco Use by Children The focus group discussions revealed notable differences in the way the African-American and the white families approached tobacco use ground rules. The African-American groups were adamant in their insistence that they had a duty to set the rules, that their children would not use tobacco because they had set and conveyed the rules, and that there would be certain and effective consequences if the rules were broken (“They’re my kids, it’s my job.” “Who else cares enough?” “That’s what parents do.” “It is tough, but it’s my responsibility. I didn’t take this [parenting] on because I thought it was easy.” “They know who’s in charge.” “If we don’t tell them, how are they supposed to know?”). The white groups displayed much less assurance about the usefulness of setting ground rules and about the ability of parents to have any effect on teen tobacco use (“What can we do? We can’t follow them around everywhere. If we set the rules, they’re different as soon as they step outside . . .” “If we make rules, they’ll only break them.”). The quantitative survey revealed a similar pattern. The survey respondents were asked to select from five possible household ground rule options (no tobacco use by anyone in the household, use restricted to adults only, children allowed to use to-
Table 2. Characteristics of the Survey Sample (n 5 311) Characteristics of the Respondents [number (%)] Female Self-identified ‘‘race’’ African-American White Other Education level Did not graduate high school High school graduate Some college or technical school College graduate Smoking status Never smoker Former smoker Current smoker Characteristics of the Households Two parents Smoking status Other parent smokes Neither parent smokes One parent smokes Both parents smoke Any parent smokes Annual household income ($) ,10,000 10,000 –29,999 30,000 – 49,999 50,000 or more Refused item Characteristics of the Child Referents Age (years) 8 –10 11–13 14 –15 16 –17 Gender female Respondent knowledge of child’s smoking status Certain child does not smoke Doesn’t know if child smokes Has reason to suspect child smokes No reason to suspect child smokes Certain child smokes
247 (79.4) 51 (16.4) 256 (82.3) 4 (1.3) 3 (1.0) 125 (40.2) 81 (26.0) 102 (32.8) 158 (50.8) 70 (22.5) 83 (26.7) 225 (82.0) 66 (25.9) 195 (62.7) 83 (26.7) 33 (10.6) 126 (40.5) 4 (1.3) 76 (24.4) 92 (29.6) 130 (41.8) 9 (2.9)
19 (6.1) 98 (31.5) 105 (33.8) 89 (28.6) 165 (53.4) 216 (69.5) 45 (14.5) 15 (4.8) 30 (9.7) 50 (16.0)
bacco in the house, children allowed to use tobacco only outside the house, or no ground rules). If the parents reported that they had set ground rules, they were asked if they or any other responsible adult in the household had ever had a specific conversation with the child(ren) about the rules. Most of the parents (86.8%) reported that they had set ground rules in the household. Of those who had set ground rules, most reported that either smoking was not permitted for any household members (55.2%) or was limited to adults only (34.1%); 29 respondents (10.7%) reported that children were allowed to use tobacco either inside or outside of the home. Of those parents who reported having any
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
333
Table 3A. Household Ground Rules Regarding Use of Tobacco, by Race/Ethnicity African-American Parent
White Parent
OR (95% CI)
P Value
Have the parents set ground rules regarding tobacco use by children? Yes 51 (100) 216 (84.4) 19.27 0.003 No 0 40 (1.17, 318.6) If parents have set rules do the rules restrict tobacco use to either adults only or to no one in the household? Yes 47 (92.2) 191 (88.4) 4.00 0.44 No 4 25 (0.51, 4.61) If parents have set rules, have they told the children what the rules are? Yes 43 (84.3) 142 (65.7) 2.80 0.01 No 8 74 (1.28, 6.12)
ground rules, 69.6% had explained the rules to the child(ren), 27.8% felt that the children knew what the rules were without their having to have a specific conversation about them, and 2.6% felt that their children were too young to require setting rules. In households with rules restricting any childhood use of tobacco, 68.5% had explained the restrictions to their children. Nearly half of the households (46.9%) had rules restricting childhood tobacco use, but had never explicitly articulated them to the child(ren) what the rules were, allowed children to use tobacco, or had not formulated any rules. African-American households were more likely than those of white families to have set ground rules and were more likely to have had a discussion with the children about the rules (Table 3A). Among the parents who had set rules, the African-American parents reported restriction of tobacco use to adults only as frequently as did the white parents. Respondents in households with any adult smoker differed significantly from households with no adult smokers only in that parents in smoking households were more likely to report having set any rules (Table
3B), and, as would be expected, that parents in households with smoking adult(s) were more likely to allow tobacco use only by adults. There was no difference in the two groups regarding whether or not they had told the child(ren) about the ground rules. Perceptions of Parental Efficacy in the Prevention of Childhood Tobacco Use The perception of the power that parents have to prevent tobacco use was also markedly different in the African-American groups compared to the white parents. The white groups discussed their powerlessness to have any effect (“They are all going to try it sometime.” “It’s part of being a kid.” “It’s all peer pressure, what can we do about it?” “If we say don’t do it, they’ll only want it more” “What can ya do? They’re teenagers. They do what their friends want. It’s peer pressure.”). The causative relationship between peer pressure and childhood smoking was mentioned frequently and thought of by the white groups to be a relatively immutable phenomenon. When asked what they had or would do if they
Table 3B. Household Ground Rules Regarding Use of Tobacco, by Parental Smoking Status Any Adult Smoker in Household
No Adult Smoker in Household
OR (95% CI)
P Value
Have the parents set ground rules regarding tobacco use by children? Yes 116 (92.1) 154 (83.2) 0.43 0.02 No 10 31 (0.20, 0.90) If parents have set rules, do the rules restrict tobacco use to either adults only or to no one in the household? Yes 103 (88.8) 138 (89.6) 1.09 0.83 No 13 16 (0.50, 2.33) If parents have set rules, have they told the children what the rules are? Yes 81 (69.8) 107 (69.5) 1.02 0.87 No 35 47 (0.60, 1.72)
334
CLARK ET AL.
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
Table 4A. Perceptions of Parental Efficacy in Reducing Teen Tobacco Use, by Race/ Ethnicity African-American Parent
White Parent
OR (95% CI)
P Value
All kids will try tobacco, it’s a part of growing up Agreea 17 (33.3) 137 (53.5) 0.4 Disagreeb 34 119 (0.2, 0.8) 0.01 Punishing children for trying tobacco is not likely to keep them from trying it again Agree 11 (21.6) 193 (72.4) 0.09 Disagree 40 63 (0.05, 0.20) 0.0001 If parents forbid teens to use tobacco, they will only want it more Agree 6 (11.8) 127 (49.6) 0.14 Disagree 45 129 (0.06, 0.29) 0.0001 Schools can be more effective than parents in teaching children about the dangers of tobacco Agree 4 (8.7) 80 (32.0) 0.27 Disagree 42 156 (0.08, 0.55) 0.004 ‘‘Don’t know’’ responses not included. a Combines ‘‘strongly agree’’ and ‘‘agree.’’ b Combines ‘‘disagree’’ and ‘‘strongly disagree.’’
found their kids using tobacco, the reactions were mixed among the members of the white groups. There were some who reported that there was nothing they could do that would make any difference (“If I beat her to a pulp she’d keep it up to spite me.” “What are you going to do? They are teenagers.”). Others felt that talking, grounding, withdrawal of privileges, or other interventions might make a difference. Some alluded to corporal punishment, assured that it would be an effective deterrent. Several of the men mentioned that they would do what their fathers had done, such as make them smoke an entire pack quickly to make them sick, or make them eat a pack of cigarettes. When asked if their fathers’ interventions had any effect on them, they conceded that it had only made them more cautious about getting caught again. The African-American groups were in agreement about their ability to influence their children’s use of tobacco. When asked what they would do if they caught their children using tobacco, they agreed that either it wouldn’t happen (“She knows not while she’s under my roof”) or that consequences would be certain and effective (“He knows what would happen if I caught him. He’d never do it again.”). Many statements were apparent hyperbole (“. . . grounded for life . . .” “He wouldn’t get over it anytime soon.”). Specific interventions were not elicited, except from two African-American women who had at one time suspected that their children might be using. Each had let the children know how disappointed they would be if the children ever used tobacco, and both felt that such communication was adequate to keep the teens from using it again.
Many of the white parents who smoked believed that it was not a behavior worthy of getting into a conflict with their teens (“Nobody ever got killed getting behind the wheel of a car after smoking a cigarette.” “At least it’s not drugs.” “We have to pick our battles. That’s the least of our concerns.”). In contrast, members of the African-American groups expressed the importance of the issue and the need to involve themselves actively in their children’s decisions regarding tobacco use (“It will kill them eventually.” “It’s addicting, kids don’t have the sense to know that.” “We need to protect them from it.”). Only two individual members of the AfricanAmerican groups expressed a laissez-faire attitude regarding children and tobacco. The telephone survey elicited similar differences (Table 4A). The white parents were more than twice as likely as the African-American parents to either agree or strongly agree with the statement, “All kids will try tobacco, it’s a part of growing up.” The white parents were seven times more likely to agree with the statement, “If parents forbid teens to use tobacco, they will only want it more,” eight times more likely to agree with, “If parents make a big deal out of tobacco, children will be more likely to try it,” and 11 times more likely to agree with, “Punishing children for trying tobacco is not likely to keep them from trying it again.” Parents in households with a least one adult smoker did not differ significantly from those with no adult smokers on any of these items, except on the inevitability of children trying tobacco. Smokers were 1.8 times as likely as nonsmokers to agree or strongly agree with that statement (Table 4B).
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
335
Table 4B. Perceptions of Parental Efficacy in Reducing Teen Tobacco Use, by Parental Smoking Status Any-Adult Smoker in Household
No Adult Smoker in Household
OR (95% CI)
P Value
All kids will try tobacco, it’s a part of growing up Agreea 74 (58.7) 81 (43.8) 1.8 Disagreeb 52 104 (1.2, 2.9) 0.01 Punishing children for tryping tobacco is not likely to keep then from trying it again Agree 92 (73.0) 113 (61.0) 1.72 Disagree 34 72 (1.06, 2.82) 0.03 If parents forbid teens to use tobacco, they will only want it more Agree 55 (43.7) 80 (43.2) 1.02 Disagree 71 105 (0.64, 1.61) 0.94 If parents ‘‘make a big deal’’ out of tobacco, children will be more likely to try it Agree 51 (40.5) 68 (36.8) 1.17 Disagree 75 117 (0.72, 1.91) 0.51 Schools can be more effective than parents in teaching children about the dangers of tobacco Agree 43 (34.7) 82 (23.9) 0.87 Disagree 81 134 (0.55, 138) 0.55 ‘‘Don’t know’’ responses not included. a Combines ‘‘strongly agree’’ and ‘‘agree.’’ b Combines ‘‘disagree’’ and ‘‘strongly disagree.’’
Perceptions of the Role of the Schools Many from the groups of white parents expressed the belief that they did not have to worry about their children using tobacco because of the influence of the schools (“They learn in school not to want it.” “He comes from school like filled with the stuff [about tobacco] and gives me hell about my smoking. I don’t need to worry about it.”). They acknowledged that tobacco control among teens is an appropriate role for the schools, releasing them from much of the burden. There were no gender differences in white parents’ beliefs. The African-American women showed near unanimity in their belief that the schools should only reinforce their own teachings (“It’s my job. The schools can help, but I’m the one . . .”). The AfricanAmerican men were more certain of the efficacy of the schools in reducing teen tobacco use, but confirmed that it was the role of the parents to do the primary teaching. One item in the survey dealt with the relative roles of parents and schools in teaching children about the dangers of smoking. White respondents were almost five times as likely to agree that the schools can be more effective than can parents (Table 4A). There was no significant difference in the responses from parents who did and did not smoke (Table 4B). Problems of Parents Who Smoke in Talking to Their Children About Tobacco The white parents in focus groups frequently mentioned their difficulties in telling their children that
they could not smoke (“What am I going to do, like, do as I say, not as I do.” “I feel like a hypocrite if I say anything.”). It was not unusual for the mothers to express guilt about the role that their smoking had in the inevitability of their children’s smoking. This guilt response was apparent from only one father. Such comments were not heard from the African-American group members. They felt that their own smoking made it more difficult, but did not prevent them from instructing their children not to smoke. In the survey, both African-American and white parents agreed that parents who smoked would have more difficulty talking to their children about tobacco and that children whose parents smoke would be more likely to begin smoking themselves. In contrast, respondents from households in which no parent smoked were 2.7 times more likely to agree with the statement, “Children who see their parents smoking are likely to try smoking,” and 5.7 times more likely to agree that “Parents who smoke have difficulty telling their children not to smoke” (Table 5).
Ownership of, Use of, and Policies Regarding Tobacco Promotional Items This issue was not discussed in the focus groups, but several items addressed it in the survey (Tables 6A and 6B). As expected, respondents who smoked were more likely than those who did not to own tobacco promotional items, be willing to use them themselves, and to not restrict their children from
336
CLARK ET AL.
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
Table 5. Perceptions of the Influence of Parental Smoking on Teen Tobacco Use Any Adult Smoker in Household
No Adult Smoker in Household
Children who see their parents smoking are likely to try smoking Agreea 93 (73.9) 160 (88.4) Disagreeb 33 21 Parents who smoke have difficulty telling their children not to smoke Agree 72 (57.1) 159 (88.4) Disagree 54 21
OR (95% CI)
P Value
0.37 (0.21, 0.67)
0.001
0.17 (0.10, 0.30)
0.0001
‘‘Don’t know’’ responses not included. a Combines ‘‘strongly agree’’ and ‘‘agree.’’ b Combines ‘‘disagree’’ and ‘‘strongly disagree.’’
using them. Parents who smoked were also more likely to allow their children to use candy cigarettes (Table 6B). The African-American respondents did not differ significantly from the white respondents on any of these items, except that they were significantly less likely to allow the use of candy cigarettes (Table 6A). Availability and Monitoring of Cigarettes in the Homes of Adults Who Smoked The preliminary questionnaire administered to the 70 focus group participants revealed that most (68.2%) of the participants bought their cigarettes by the pack, yet 23.8% typically had more than one pack open at any time, and 50.8% often stored their open packs in unattended locations. The mean number of
separate rooms or places in which unopened packs were stored was two. Both the African-American and the white group members did virtually no monitoring of their tobacco supplies and did not attempt to limit children’s access to them. Frequently cited reasons were that they had never thought about it as a potential problem, or it was too inconvenient to lock up a product that they used so frequently. Those who had discovered their own tobacco in the possession of their children invariably were surprised, even those who had related earlier in the discussion that their own initial experimentation with tobacco had involved either cigarettes pilfered from family members or cigarettes that their friends had surreptitiously obtained from their parents’ or other adults’ tobacco supplies.
Table 6A. Ownership of, Use of, and Policies Regarding Tobacco Promotional Items African-American Parent
White Parent
OR (95% CI)
P Value
Do you own any product that promotes a tobacco brand or was distributed by a tobacco company? Yes 5 (10.2) 55 (21.7) 0.43 0.08 No 42 198 (0.14, 1.20) Do you think you would ever use a tobacco industry promotional item, such as a tee shirt? Yes 14 (27.5) 32 (17.3) 1.81 0.11 No 37 153 (0.82, 3.14) Which of the following statements best describes your policies about tobacco promotional items and your child? No Problema 14 (27.5) 74 (29.7) 0.89 0.75 Disapproveb 37 175 (0.43, 1.83) Which of the following statements best describes your policies about your child’s use of [candy cigarettes]? No Problema 13 (25.5) 124 (48.4) 0.36 0.003 Disapproveb 38 132 (0.17, 0.75) a Includes the responses: ‘‘I have no problem with my child using these items and would help obtain them for my child’’; ‘‘I have no problem with my child using these items, but would not help obtain them for my child’’; and ‘‘I don’t have any opinion about these items.’’ b Includes the responses: ‘‘I disapprove of my child using these items, but would not forbid it’’; and ‘‘I don’t allow my child to use these items.’’
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
337
Table 6B. Ownership of, Use of, and Policies Regarding Tobacco Promotional Items, by Parental Smoking Status Any Adult Smoker in Household
No Adult Smoker in Household
OR (95% CI)
P Value
Do you own any product that promotes a tobacco brand or was distributed by a tobacco company? Yes 46 (37.1) 15 (8.2) 6.74 No 76 167 (3.40, 13.51) ,0.001 Do you think you would ever use a tobacco industry promotional item, such as a tee shirt? Yes 61 (48.4) 32 (17.3) 4.49 No 65 153 (2.59, 7.79) ,0.001 Which of the following statements best describes your policies about tobacco promotional items and your child? No Problema 47 (37.6) 41 (22.9) 2.03 0.005 Disapproveb 78 138 (1.19, 3.46) Which of the following statements best describes your policies about your child’s use of [candy cigarettes]? No Problema 67 (53.2) 70 (37.8) 1.87 0.007 Disapproveb 59 115 (1.15, 3.03) a Includes the responses: ‘‘I have no problem with my child using these items and would help obtain them for my child’’; ‘‘I have no problem with my child using these items, but would not help obtain them for my child’’; and ‘‘I don’t have any opinion about these items.’’ b Includes the responses: ‘‘I disapprove of my child using these items, but would not forbid it’’; and ‘‘I don’t allow my child to use these items.’’
Of the 126 survey respondents who smoked, 35.7% reported usually buying cigarettes by the carton, 18.3% sometimes bought by the carton, and 19.8% typically had more than one pack open at any time. Only 13 stored their extra supplies in a locked place. When the survey respondents were asked “How likely would you be to notice that some of your cigarettes were missing?” 80 said that it was “very likely,” 14 “somewhat likely,” 15 “unlikely,” and 9 reported “it would be impossible to tell.”
Even so, the age suggested by African-American respondents was significantly higher than that of the white respondents (p , 0.001). Only one AfricanAmerican respondent named an age ,18 years, and 10 named an age .18 years; 67 of the white parents responded with an age ,18 years, and 39 named an age .18 years (Table 7). The suggested age was not significantly different for parents who did and did not smoke (p 5 0.32).
Discussion Age at Which Children or Teens Should Make Up Their Own Minds About Tobacco Use The last question asked of each individual member of the focus groups was, “At what age should children or teens be able to make up their own minds about whether or not to use tobacco?” The white parents, with few exceptions, stated that 16 years was the appropriate age. This choice was often accompanied by an observation such as, “If they can drive a car at 16 years, they can decide whether they want to smoke.” In contrast, the African-American parents answered either age 18 years or “When they are out of my house.” None of the African-American parents stated any age younger than 18 years. The same question was asked of the survey respondents. In the absence of group discussions preceding the question, the responses were not as uniform as those from the focus group participants.
Compared to the African-American parents, the white parents reported lower levels of involvement in anti-tobacco socialization within the home. They were less likely to report setting household ground rules regarding teen tobacco use and communicating the rules to the children. The African-American parents were less likely than the white parents to believe
Table 7. Age of Responsibility in Choosing to Use Tobacco African-American Parent
White Parent
Chi-Square
P-Value
At what age should children or teens be allowed to make up their own minds about whether or not to use tobacco? Age ,18 1 (2.0) 67 (26.2) 14.47 ,0.001 Age 18 40 (78.4) 150 (58.6) Age .18 10 (19.6) 39 (15.2)
338
CLARK ET AL.
that tobacco use is an inevitable part of growing up, and that parental interventions would have either no effect on teen use or make the teens “want it more.” They were also more likely than the white parents to believe that schools played a role in anti-tobacco socialization that was subordinate to their own. There are several possible explanations for the relatively low rates of African-American teen smoking. Whereas the delay in the initiation of tobacco use among African-Americans may contribute to the difference, it is unlikely that it accounts for the full magnitude of the discrepancy. In addition, the delay in youth smoking onset may in itself be a salutatory effect of the African-American beliefs and behaviors reported here. A causative relationship between active participation in anti-tobacco socialization and the relatively low rates of teen tobacco use in African-American homes is not certain, but evidence gathered from teens supports the supposition. For instance, in a National Cancer Institute study, utilizing focus group methods, parental advice to not smoke, regardless of parental smoking status, was implicated in the lower rates of smoking among African-American youth.(23) McDermott (26) reported that, regardless of parents’ own drug use behavior, adolescents who perceive their parents as holding permissive views about teen drug use were significantly more likely to use drugs than those who perceived their parents as having nonpermissive views. In a more recent report specific to tobacco use, Jackson and Henriksen (2) concluded that third- and fifth-grade children who experience anti-smoking socialization in the home report substantially lower onset rates, even when one or both parents currently smoke. Specifically, children exposed to one or two parents who smoked were at significantly greater risk for intending to smoke, perceiving easy access to cigarettes, and for having almost tried smoking, but those who believed that their parents would know if they were smoking, and those who expected negative consequences for smoking, reported lower smoking rates than those who did not hold such beliefs. Flay et al. (10) found that seventhgrade adolescents were more likely to copy smoking behaviors of their peers rather than of those of their parents, but that perceived parental disapproval of adolescent smoking modified that relationship. Hunter et al. (27) reported that parental approval contributed to continued smoking after initial trial. That tobacco use by parents is a risk factor for teen smoking seemed to be generally accepted among the adults in this study. This knowledge may have had unintended negative consequences. Feelings of guilt
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
and powerlessness apparently discouraged the parents who smoked from openly talking about the issue, thereby disassociating themselves from their children’s decision-making regarding tobacco use. It is not known whether the number of parents who successfully quit when pressured by their children outnumbers those who have been silenced because of their continued use. The knowledge that parental tobacco use is a risk factor for children’s use may have worked against the parents who did not smoke, as well. They may have felt that they did not have to be concerned about their children initiating tobacco use and, therefore, had not bothered with this element of anti-tobacco socialization in the home. The ubiquitous pro-tobacco messages outside the home would therefore have had fewer anti-tobacco counter messages in those households in which there were no adult smokers (28). Educational efforts may be required for many parents to understand and support the proposed United States Food and Drug Administration rules outlawing tobacco promotional items.(29) In addition, 44% had no problem with their children using candy cigarettes. It may be that the parents did not associate the use of these items with promoting tobacco use as a social norm, or recognize the problem of candy cigarettes contributing to the identification of cigarettes as a treat. Protective practices regarding tobacco supplies were not found to be the norm in this study. The ready availability of unmonitored parental cigarette supplies suggests that the behavior of parents may be significant in any supply-side strategies to reduce tobacco use by youth. Whereas most such strategies have focused on reducing sales of tobacco to children, a campaign to convince parents (and the extended family) to handle adult tobacco supplies more responsibly also is important in reducing childhood tobacco use. Limitations Whereas the households likely to have age-eligible children initially were selected by random-digitdialing techniques and household screening, the respondents in the telephone survey reported here had all previously allowed their children to be interviewed regarding tobacco use. Whereas we can only speculate what effect this potential selection bias has on the results, the presence of some bias in the whole group estimates is evident from the unusually low refusal rate. It is implausible, however, that the effect would have operated differently among smokers and nonsmokers, or among African-American par-
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
339
ents versus white parents, producing bias in the between-groups comparisons reported here. The samples for both the telephone survey and the focus groups were from a single region (northeast Ohio). It is not known if there are significant differences in attitudes and behaviors of parents in different regions of the country. A more broadly representative sample, including other ethnic groups, would be an appropriate next step.
10. Flay BR, Hu FB, Siddiqui O, et al. Differential influence of parental smoking and friends’ smoking on adolescent initiation and escalation of smoking. J Health Soc Behav 1994;35:248 – 65.
Conclusions
14. Biglan A, Ary D, Yudelson H, et al. Experimental evaluation of a modular approach to mobilizing antitobacco influences of peers and parents. Am J Community Psychol 1996;24:311–39.
These data support the hypothesis that AfricanAmerican parents feel more empowered to delay or prevent their children’s tobacco use than do white parents. The challenge facing tobacco control professionals will be to find ways to help empower all parents to become actively involved in safeguarding their children from this major health threat. In order to do so, parents must first be provided with more knowledge about the immediate and long-term health consequences of tobacco use in childhood, about the realities of (and ubiquitous nature of) tobacco promotion coming from outside the home, and about the powerful role they could play in protecting their children. This study was supported through an intramural grant program, the President’s Pioneer Award for Research.
References 1. Pierce JP, Fiore MC, Novotny TE, et al. Trends in cigarette smoking in the United States. JAMA 1989;261:61–5. 2. Jackson C, Henriksen L. Do as I say: Parent smoking, antismoking socialization, and smoking onset among children. Addict Behav 1997;22:107– 4. 3. Bauman KE, Koch GG, Lentz GM. Parent characteristics, perceived health risk, and smokeless tobacco use among white adolescent males. NCI Monogr 1989;8:43– 8. 4. Green G, MacIntyre S, West P, et al. Like parent like child? Associations between drinking and smoking behaviour of parents and their children. Br J Addict 1991;86:745–58. 5. Chassin L, Presson CC, Montello D, et al. Changes in peer and parent influence during adolescence: Longitudinal versus cross-sectional perspectives on smoking initiation. Dev Psychol 1986;22:327–34. 6. Eiser JR, Morgan M, Gammage P. Adolescent smoking: Attitudes, norms and parental influence. Br J Soc Pyschol 1989;28: 193–202. 7. Bauman KE, Foshee VA, Linzer MA, et al. Effect of parental smoking classification on the association between parental and adolescent smoking. Addict Behav 1990;15:413–22. 8. Chassin L, Presson CC, Sherman SJ, et al. Parent educational attainment and adolescent cigarette smoking. J Subst Abuse 1992;4:219 –34. 9. Covey LS, Tam D. Depressive mood, the single-parent home, and adolescent cigarette smoking. Am J Public Health 1990; 80:1330 –3.
11. Bauman KE, Botvin GJ, Botvin EM, et al. Normative expectations and the behavior of significant others: An integration of traditions in research on adolescents’ cigarette smoking. Psychol Rep 1992;71:568 –70. 12. Glendinning A, Shucksmith J, Hendry L. Family life and smoking in adolescence. Soc Sci Med 1997;44:93–101. 13. Melby JN, Conger RD, Conger KJ, et al. Effects of parental behavior on tobacco use by young male adolescents. J Marriage Fam 1993;55:439 –54.
15. Foxcroft DR, Lowe G. Adolescent drinking, smoking and other substance use involvement: Links with perceived family life. J Adolesc 1995;18:159 –77. 16. Nelson DE, Giovino GA, Shopland DR, et al. Trends in cigarette smoking among U.S. adolescents, 1974 –1991. Am J Public Health 1995;85:34 – 40. 17. Centers for Disease Control. Differences in the age of smoking initiation between blacks and whites—United States. MMWR 1991;40:754 –7. 18. Escobedo LG, Anda RF, Smith PF. Sociodemographic characteristics of cigarette smoking initiation in the United States. JAMA 1990;264:1550 –5. 19. U.S. Department of Health and Human Services. Report of the Surgeon General: Reducing the health consequences of smoking. Atlanta, GA: Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1989. 20. U.S. Department of Health and Human Services. Tobacco use in 1986. Methods & Basic Tabulations from Adult Use of Tobacco Survey. Atlanta, GA: Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1986. 21. Wang MQ, Cowdery JE, Trucks J, et al. Southern adolescents: A survey of attitudes and beliefs about smoking. S Med J 1994;87:1097–102. 22. Johnston LD, O’Malley PM, Bachman JG. National survey results on drug use from Monitoring the Future Study, 1975–1994. Vol. 1: Secondary school students. NIH Publication No. 95-4026. Rockville, MD: National Institute on Drug Abuse, 1995. 23. McIntosh H. Black teens not smoking in great numbers. J Natl Cancer Inst 1995;87:564 –5. 24. Gleenlund KJ, Johnson C, Wattigney W, et al. Trends in cigarette smoking among children in a southern community, 1976 –1994: The Bogalusa Heart Study. Ann Epidemiol 1996;6:476– 82. 25. Knodel J. The design and analysis of focus group studies. A practical approach. In: Morgan DL, ed. Successful Focus Groups. Newbury Park: Sage Publications, 1993:35–50. 26. McDermott D. The relationship of parental drug use and parents’ attitude concerning adolescent drug use to adolescent drug use. Adolescence 1984;19:89 –97. 27. Hunter SM, Baugh JG, Webber LS, et al. Social learning effects on trial and adoption of cigarette smoking in children: The Bogalusa Heart Study. Prev Med 1982;11:29 – 42. 28. Institute of Medicine, Committee on Preventing Nicotine Addiction in Youth and Children. Lynch BS, Bonnie RJ, eds. Growing Up Tobacco Free. Washington, DC: National Academy Press, 1994. 29. Federal Register, August 11, 1995.
ORIGINAL ARTICLE
Anti-tobacco Socialization in Homes of AfricanAmerican and White Parents, and Smoking and Nonsmoking Parents PAMELA I. CLARK, Ph.D., ANNEMARIE SCARISBRICK-HAUSER, Ph.D., SHIVA P. GAUTAM, Ph.D. AND SARAH J. WIRK, B.S.
Purpose: To examine parental perceptions and behaviors with regard to teen smoking, comparing AfricanAmerican and white parents, and those who did and did not smoke. Methods: Focus groups consisting of African-American and white parents who smoked provided initial in-depth information. A computer-assisted telephone survey of a biracial sample of 311 parents of children ages 8 to 17 years provided more generalizable information regarding parental beliefs and behaviors. Results: Nearly 50% of households either allowed teen smoking, had no ground rules, or had set restrictive rules but never communicated them to the children. Compared to white parents, African-American parents felt more empowered to affect their children’s behaviors and were more likely to actively participate in anti-tobacco socialization within the home (all p values < 0.01). Among the African-American parents, 98% reported 18 years or older to be an appropriate age for teens to make up their own minds about using tobacco, whereas 26% of white parents thought 16 years to be an appropriate age (p < 0.001). Parents who smoked reported more frequent rule-making than those who did not smoke (p 5 0.02), but were more likely to believe that childhood tobacco use is inevitable (p 5 0.01). Conclusions: Many parents are not engaged in antitobacco socialization in the home. Differences in the degree of parental participation may contribute to the From Northeastern Ohio Universities College of Medicine, Rootstown, and the Institute for Policy Studies, University of Akron, Akron, Ohio. Address reprint requests to: Pamela I. Clark, Ph.D., Associate Professor of Epidemiology, Division of Community Health Sciences, Northeastern Ohio Universities College of Medicine, 4209 State Rte. 44, Rootstown, OH 44272. Manuscript accepted March 12, 1998.
variance in smoking prevalence between African-American and white children. © Society for Adolescent Medicine, 1999 KEY WORDS: Adolescents Tobacco Smoking Parenting Socialization Race/ethnicity
That adult intermediaries knowingly or inadvertently assist children in obtaining and using tobacco products is evident in that 3,000 children become regular smokers each day (1). Parents could, however, be the most effective and the most motivated adults to stand between tobacco and their children. They could act as protective gatekeepers to the family; providing anti-tobacco socialization within the home (2), and politically influencing schools, merchants, and government outside the home. Some understanding of the relationship between parents and childhood tobacco use is emerging. Childhood tobacco use has been found to increase with parental smoking status, lower parental educational attainment and social class, and living in single-parent homes (3–9). By portraying attitudes favorable to smoking and otherwise increasing normative expectations about tobacco use, parents may increase the likelihood a child will smoke (6,10,11). The influence of parenting styles also has been implicated, with teen smoking reported to be related
© Society for Adolescent Medicine, 1999 Published by Elsevier Science Inc., 655 Avenue of the Americas, New York, NY 10010
1054-139X/99/$–see front matter PII S1054-139X(98)00117-7
330
CLARK ET AL.
directly to measures of nonsupportive/neglectful parenting (12) and harsh/inconsistent parenting (13), and inversely related to nurturant/involved parenting (13). Family involvement also has been found to be a necessary component of school and community programs intended to decrease childhood use and intention to use tobacco (14,15). Although some parental characteristics (e.g., social class, marital status, educational attainment, and even parenting styles) may not be amenable to public health interventions, with enough practical information, it may be possible to mobilize parents in a campaign to decrease the tobacco industry’s access to children. In order to do so, it is necessary to have a better understanding of the perceptions that parents hold regarding their legitimate roles in protecting their children and the barriers that they perceive in fulfilling protective roles. This study combined qualitative and quantitative methods to investigate parental anti-tobacco socialization beliefs and behaviors, and to compare the beliefs and behaviors among parents in households with any adult smoker and those with no adult smokers. Because African-American smokers generally begin smoking at a later age than do white smokers (16 –20) and tobacco use rates of AfricanAmerican teens are significantly lower than those of white teens (21–24), indicating the possibility of significant group differences in anti-tobacco socialization within households, we compared the beliefs and behaviors of African-American parents with those of white parents.
Methods Moderated focus groups were used to provide indepth information about adult smokers’ knowledge, attitudes, and practices. Twelve focus groups were completed prior to commencing a computer-assisted telephone survey of parents of children, ages 8 to 17 years. Information derived from the focus groups guided the content of the telephone survey.
The Focus Groups Current, regular smokers from northeast Ohio (excluding the city of Akron) were recruited into the focus groups in March and April 1997. Recruitment was done by posting notices in public places and by word of mouth. Members of the groups were paid $35.00 for their participation. Group members were at least 18 years of age and were parents of children
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
aged 8 to 17 years. Several were grandparents with primary responsibilities for the care of grandchildren in the household. Groups were composed of either self-identified African-American smokers or selfidentified white smokers. To assure that subjects in the focus groups and the survey respondents were mutually exclusive, candidates for the focus groups had telephone exchanges not included in the sampling frame used for the survey. Procedures. The focus group sessions were audiotaped for subsequent transcription. Each lasted between 1.5 and 2 hours. The discussions were structured around a set of items in focus group discussion guidelines, including a set of open-ended questions to cover the maximum range of relevant topics, fostering group interaction that explored the participants’ knowledge, beliefs, and feelings. We used low moderator involvement. The number of groups provided enough repetition to confirm dominant themes. The analytic approach was that of Knodel (25). Briefly, audiotapes were transcribed and then converted into transcripts. An initial set of codes were developed, with topics and subtopics corresponding to each item in the discussion guidelines. Additional codes were created for topics that arose and were of interest, and codes were consolidated as necessary. Finally, detailed codes to use for analysis of specific topics (e.g., opinions about the inevitability of teens’ smoking) were developed. An overview grid was constructed, providing a descriptive summary of the content of the discussions. One axis had the topic headings and the other the focus group identifiers. The cells contained brief descriptions of the content by group and topic, and formed the basis for content analysis. Reliability of the data was improved by comparing statements within and across groups. Reliability of the analysis was improved by involving two members of the investigative team in the interpretations. The content was interpreted independently by each, then compared. Disagreements were discussed and resolved. At least two analysts were present during the focus groups, providing field notes to enhance further the accuracy of the data. Statements made during the discussions that needed further clarification were incorporated into the content of subsequent groups. In order to generate some information not contaminated by the group interactions and to provide some data common to both the focus groups and the
May 1999
survey respondents, the discussions were preceded by a brief self-administered questionnaire.
The Telephone Survey Parents of child(ren) in each of 311 households in Akron, Ohio, previously identified by random-digitdialing techniques for another study were interviewed using a computer-assisted telephone interviewing system during June and July 1997. The survey consisted of a set of mostly close-ended questions, and was 8 to 10 minutes in length. The survey items were planned to elicit household and family policies and behaviors, as well as the knowledge, opinions, and behaviors of individual respondents. The survey was administered only in English. Procedures. Each called number was recontacted up to three times, spread over the field period including day, evening, and weekend calls. Initial refusals were recontacted for conversion. Appointments were made whenever an eligible household was identified and an immediate interview was not possible. Appointments were made for respondent convenience including day and weekend hours. Those appointments also were pursued over the field period, with several attempts made to complete the appointed interview. After screening for the presence of age-eligible child(ren) in the household, the interviewer requested to speak to a parent or guardian. To increase the likelihood of an adequate sampling of adults who smoked, if one or both of the parents smoked, the interviewer asked to speak to a/the parent who smoked. Because parental attitudes and policies may be very different for younger children compared to older children, each respondent was asked for the ages of all children in the household. The child whose age was closest to 12 years was selected as the referent child. If two children had ages equidistant from 12 years, the older child (if ,18 years old) was chosen. A coin toss was used to choose between children of the same age. The respondent was asked to think of that child when responding to questionnaire items, and the request to focus on that child was reenforced several times during each interview. The study protocol was reviewed and approved by the Institutional Review Board for the Protection of Human Subjects of Northeastern Ohio Universities College of Medicine. Written informed consent was obtained from subjects in the focus groups, and
ANTI-TOBACCO SOCIALIZATION BY PARENTS
331
verbal consent obtained from adults in the survey sample. Statistical Analysis. The outcome variables were categorical and were first explored through frequency tables, appropriately stratified. Because the extreme categories in the variables with ordered choices (e.g., “strongly agree,” “agree,” “disagree,” “strongly disagree”) were used infrequently, most were collapsed to dichotomous outcomes (“strongly agree/agree” or “disagree/strongly disagree”) without loss of significant information. The independent variables were dichotomous: smoking status of the household’s adults and selfidentified race. Separate analyses were performed comparing respondents who did and did not smoke, and households with any adult smoker (defined as a smoker older than age 18 years) versus households without any adult smokers. Because both results were virtually the same, the data are presented comparing households in which there were and were not adult smokers. Households also were classified according to the self-reported ethnicity of the respondent. The associations of household smoking status and race with the categorical outcome measures were assessed by Chi-square analysis. Because the age of the child referent may contribute significantly to parents’ beliefs and behaviors, the distributions of children’s ages among African-American respondents versus white respondents, and respondents who did and did not smoke, were explored. Because the age distributions were not different, none of the results were adjusted for age of the child referent.
Results A total of 70 parents who smoked participated in the 12 focus groups. There were five African-American groups and seven white groups. The characteristics of the participants are shown in Table 1. The response rate for the telephone survey was 86% (311 completed of 362 attempted). Reasons for noncompletion were: no age-eligible children in the household (n 5 14), refused (n 5 23), interrupted interview (n 5 3), or telephone no longer in service (n 5 11). The relatively high response rate may have been attributable to the nature of the telephone list, that is, parents who had been randomly sampled but who had given permission to interview their children for an earlier study. Characteristics of the parent respondents, the households, and the child referents are shown in Table 2.
CLARK ET AL.
332
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
Table 1. Characteristics of the Focus Group Participants (n 5 70) Number female (%) Race/ethnicity (%) African-American parents White parents Education level (%) Did not graduate high school High school graduate Some college or technical school College graduate Household income ($) (%) ,10,000 10,000 –29,999 30,000 – 49,999 50,000 or more Mean age (years) (SD) African-American parents White parents Mean age of first smoking (SD) African-American parents White parents Mean number of children (SD) African-American parents White parents
32 (45.7) 32 (45.7) 38 (54.3) 4 (3.7) 15 (21.4) 45 (64.3) 6 (8.6) 5 (7.2) 26 (37.1) 25 (35.7) 14 (20.0) 45.9 (9.1) 43.1 (6.1) 19.9 (4.1) 14.2 (4.0) 2.3 (0.8) 2.1 (0.9)
SD 5 standard deviation.
Household Ground Rules Regarding Tobacco Use by Children The focus group discussions revealed notable differences in the way the African-American and the white families approached tobacco use ground rules. The African-American groups were adamant in their insistence that they had a duty to set the rules, that their children would not use tobacco because they had set and conveyed the rules, and that there would be certain and effective consequences if the rules were broken (“They’re my kids, it’s my job.” “Who else cares enough?” “That’s what parents do.” “It is tough, but it’s my responsibility. I didn’t take this [parenting] on because I thought it was easy.” “They know who’s in charge.” “If we don’t tell them, how are they supposed to know?”). The white groups displayed much less assurance about the usefulness of setting ground rules and about the ability of parents to have any effect on teen tobacco use (“What can we do? We can’t follow them around everywhere. If we set the rules, they’re different as soon as they step outside . . .” “If we make rules, they’ll only break them.”). The quantitative survey revealed a similar pattern. The survey respondents were asked to select from five possible household ground rule options (no tobacco use by anyone in the household, use restricted to adults only, children allowed to use to-
Table 2. Characteristics of the Survey Sample (n 5 311) Characteristics of the Respondents [number (%)] Female Self-identified ‘‘race’’ African-American White Other Education level Did not graduate high school High school graduate Some college or technical school College graduate Smoking status Never smoker Former smoker Current smoker Characteristics of the Households Two parents Smoking status Other parent smokes Neither parent smokes One parent smokes Both parents smoke Any parent smokes Annual household income ($) ,10,000 10,000 –29,999 30,000 – 49,999 50,000 or more Refused item Characteristics of the Child Referents Age (years) 8 –10 11–13 14 –15 16 –17 Gender female Respondent knowledge of child’s smoking status Certain child does not smoke Doesn’t know if child smokes Has reason to suspect child smokes No reason to suspect child smokes Certain child smokes
247 (79.4) 51 (16.4) 256 (82.3) 4 (1.3) 3 (1.0) 125 (40.2) 81 (26.0) 102 (32.8) 158 (50.8) 70 (22.5) 83 (26.7) 225 (82.0) 66 (25.9) 195 (62.7) 83 (26.7) 33 (10.6) 126 (40.5) 4 (1.3) 76 (24.4) 92 (29.6) 130 (41.8) 9 (2.9)
19 (6.1) 98 (31.5) 105 (33.8) 89 (28.6) 165 (53.4) 216 (69.5) 45 (14.5) 15 (4.8) 30 (9.7) 50 (16.0)
bacco in the house, children allowed to use tobacco only outside the house, or no ground rules). If the parents reported that they had set ground rules, they were asked if they or any other responsible adult in the household had ever had a specific conversation with the child(ren) about the rules. Most of the parents (86.8%) reported that they had set ground rules in the household. Of those who had set ground rules, most reported that either smoking was not permitted for any household members (55.2%) or was limited to adults only (34.1%); 29 respondents (10.7%) reported that children were allowed to use tobacco either inside or outside of the home. Of those parents who reported having any
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
333
Table 3A. Household Ground Rules Regarding Use of Tobacco, by Race/Ethnicity African-American Parent
White Parent
OR (95% CI)
P Value
Have the parents set ground rules regarding tobacco use by children? Yes 51 (100) 216 (84.4) 19.27 0.003 No 0 40 (1.17, 318.6) If parents have set rules do the rules restrict tobacco use to either adults only or to no one in the household? Yes 47 (92.2) 191 (88.4) 4.00 0.44 No 4 25 (0.51, 4.61) If parents have set rules, have they told the children what the rules are? Yes 43 (84.3) 142 (65.7) 2.80 0.01 No 8 74 (1.28, 6.12)
ground rules, 69.6% had explained the rules to the child(ren), 27.8% felt that the children knew what the rules were without their having to have a specific conversation about them, and 2.6% felt that their children were too young to require setting rules. In households with rules restricting any childhood use of tobacco, 68.5% had explained the restrictions to their children. Nearly half of the households (46.9%) had rules restricting childhood tobacco use, but had never explicitly articulated them to the child(ren) what the rules were, allowed children to use tobacco, or had not formulated any rules. African-American households were more likely than those of white families to have set ground rules and were more likely to have had a discussion with the children about the rules (Table 3A). Among the parents who had set rules, the African-American parents reported restriction of tobacco use to adults only as frequently as did the white parents. Respondents in households with any adult smoker differed significantly from households with no adult smokers only in that parents in smoking households were more likely to report having set any rules (Table
3B), and, as would be expected, that parents in households with smoking adult(s) were more likely to allow tobacco use only by adults. There was no difference in the two groups regarding whether or not they had told the child(ren) about the ground rules. Perceptions of Parental Efficacy in the Prevention of Childhood Tobacco Use The perception of the power that parents have to prevent tobacco use was also markedly different in the African-American groups compared to the white parents. The white groups discussed their powerlessness to have any effect (“They are all going to try it sometime.” “It’s part of being a kid.” “It’s all peer pressure, what can we do about it?” “If we say don’t do it, they’ll only want it more” “What can ya do? They’re teenagers. They do what their friends want. It’s peer pressure.”). The causative relationship between peer pressure and childhood smoking was mentioned frequently and thought of by the white groups to be a relatively immutable phenomenon. When asked what they had or would do if they
Table 3B. Household Ground Rules Regarding Use of Tobacco, by Parental Smoking Status Any Adult Smoker in Household
No Adult Smoker in Household
OR (95% CI)
P Value
Have the parents set ground rules regarding tobacco use by children? Yes 116 (92.1) 154 (83.2) 0.43 0.02 No 10 31 (0.20, 0.90) If parents have set rules, do the rules restrict tobacco use to either adults only or to no one in the household? Yes 103 (88.8) 138 (89.6) 1.09 0.83 No 13 16 (0.50, 2.33) If parents have set rules, have they told the children what the rules are? Yes 81 (69.8) 107 (69.5) 1.02 0.87 No 35 47 (0.60, 1.72)
334
CLARK ET AL.
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
Table 4A. Perceptions of Parental Efficacy in Reducing Teen Tobacco Use, by Race/ Ethnicity African-American Parent
White Parent
OR (95% CI)
P Value
All kids will try tobacco, it’s a part of growing up Agreea 17 (33.3) 137 (53.5) 0.4 Disagreeb 34 119 (0.2, 0.8) 0.01 Punishing children for trying tobacco is not likely to keep them from trying it again Agree 11 (21.6) 193 (72.4) 0.09 Disagree 40 63 (0.05, 0.20) 0.0001 If parents forbid teens to use tobacco, they will only want it more Agree 6 (11.8) 127 (49.6) 0.14 Disagree 45 129 (0.06, 0.29) 0.0001 Schools can be more effective than parents in teaching children about the dangers of tobacco Agree 4 (8.7) 80 (32.0) 0.27 Disagree 42 156 (0.08, 0.55) 0.004 ‘‘Don’t know’’ responses not included. a Combines ‘‘strongly agree’’ and ‘‘agree.’’ b Combines ‘‘disagree’’ and ‘‘strongly disagree.’’
found their kids using tobacco, the reactions were mixed among the members of the white groups. There were some who reported that there was nothing they could do that would make any difference (“If I beat her to a pulp she’d keep it up to spite me.” “What are you going to do? They are teenagers.”). Others felt that talking, grounding, withdrawal of privileges, or other interventions might make a difference. Some alluded to corporal punishment, assured that it would be an effective deterrent. Several of the men mentioned that they would do what their fathers had done, such as make them smoke an entire pack quickly to make them sick, or make them eat a pack of cigarettes. When asked if their fathers’ interventions had any effect on them, they conceded that it had only made them more cautious about getting caught again. The African-American groups were in agreement about their ability to influence their children’s use of tobacco. When asked what they would do if they caught their children using tobacco, they agreed that either it wouldn’t happen (“She knows not while she’s under my roof”) or that consequences would be certain and effective (“He knows what would happen if I caught him. He’d never do it again.”). Many statements were apparent hyperbole (“. . . grounded for life . . .” “He wouldn’t get over it anytime soon.”). Specific interventions were not elicited, except from two African-American women who had at one time suspected that their children might be using. Each had let the children know how disappointed they would be if the children ever used tobacco, and both felt that such communication was adequate to keep the teens from using it again.
Many of the white parents who smoked believed that it was not a behavior worthy of getting into a conflict with their teens (“Nobody ever got killed getting behind the wheel of a car after smoking a cigarette.” “At least it’s not drugs.” “We have to pick our battles. That’s the least of our concerns.”). In contrast, members of the African-American groups expressed the importance of the issue and the need to involve themselves actively in their children’s decisions regarding tobacco use (“It will kill them eventually.” “It’s addicting, kids don’t have the sense to know that.” “We need to protect them from it.”). Only two individual members of the AfricanAmerican groups expressed a laissez-faire attitude regarding children and tobacco. The telephone survey elicited similar differences (Table 4A). The white parents were more than twice as likely as the African-American parents to either agree or strongly agree with the statement, “All kids will try tobacco, it’s a part of growing up.” The white parents were seven times more likely to agree with the statement, “If parents forbid teens to use tobacco, they will only want it more,” eight times more likely to agree with, “If parents make a big deal out of tobacco, children will be more likely to try it,” and 11 times more likely to agree with, “Punishing children for trying tobacco is not likely to keep them from trying it again.” Parents in households with a least one adult smoker did not differ significantly from those with no adult smokers on any of these items, except on the inevitability of children trying tobacco. Smokers were 1.8 times as likely as nonsmokers to agree or strongly agree with that statement (Table 4B).
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
335
Table 4B. Perceptions of Parental Efficacy in Reducing Teen Tobacco Use, by Parental Smoking Status Any-Adult Smoker in Household
No Adult Smoker in Household
OR (95% CI)
P Value
All kids will try tobacco, it’s a part of growing up Agreea 74 (58.7) 81 (43.8) 1.8 Disagreeb 52 104 (1.2, 2.9) 0.01 Punishing children for tryping tobacco is not likely to keep then from trying it again Agree 92 (73.0) 113 (61.0) 1.72 Disagree 34 72 (1.06, 2.82) 0.03 If parents forbid teens to use tobacco, they will only want it more Agree 55 (43.7) 80 (43.2) 1.02 Disagree 71 105 (0.64, 1.61) 0.94 If parents ‘‘make a big deal’’ out of tobacco, children will be more likely to try it Agree 51 (40.5) 68 (36.8) 1.17 Disagree 75 117 (0.72, 1.91) 0.51 Schools can be more effective than parents in teaching children about the dangers of tobacco Agree 43 (34.7) 82 (23.9) 0.87 Disagree 81 134 (0.55, 138) 0.55 ‘‘Don’t know’’ responses not included. a Combines ‘‘strongly agree’’ and ‘‘agree.’’ b Combines ‘‘disagree’’ and ‘‘strongly disagree.’’
Perceptions of the Role of the Schools Many from the groups of white parents expressed the belief that they did not have to worry about their children using tobacco because of the influence of the schools (“They learn in school not to want it.” “He comes from school like filled with the stuff [about tobacco] and gives me hell about my smoking. I don’t need to worry about it.”). They acknowledged that tobacco control among teens is an appropriate role for the schools, releasing them from much of the burden. There were no gender differences in white parents’ beliefs. The African-American women showed near unanimity in their belief that the schools should only reinforce their own teachings (“It’s my job. The schools can help, but I’m the one . . .”). The AfricanAmerican men were more certain of the efficacy of the schools in reducing teen tobacco use, but confirmed that it was the role of the parents to do the primary teaching. One item in the survey dealt with the relative roles of parents and schools in teaching children about the dangers of smoking. White respondents were almost five times as likely to agree that the schools can be more effective than can parents (Table 4A). There was no significant difference in the responses from parents who did and did not smoke (Table 4B). Problems of Parents Who Smoke in Talking to Their Children About Tobacco The white parents in focus groups frequently mentioned their difficulties in telling their children that
they could not smoke (“What am I going to do, like, do as I say, not as I do.” “I feel like a hypocrite if I say anything.”). It was not unusual for the mothers to express guilt about the role that their smoking had in the inevitability of their children’s smoking. This guilt response was apparent from only one father. Such comments were not heard from the African-American group members. They felt that their own smoking made it more difficult, but did not prevent them from instructing their children not to smoke. In the survey, both African-American and white parents agreed that parents who smoked would have more difficulty talking to their children about tobacco and that children whose parents smoke would be more likely to begin smoking themselves. In contrast, respondents from households in which no parent smoked were 2.7 times more likely to agree with the statement, “Children who see their parents smoking are likely to try smoking,” and 5.7 times more likely to agree that “Parents who smoke have difficulty telling their children not to smoke” (Table 5).
Ownership of, Use of, and Policies Regarding Tobacco Promotional Items This issue was not discussed in the focus groups, but several items addressed it in the survey (Tables 6A and 6B). As expected, respondents who smoked were more likely than those who did not to own tobacco promotional items, be willing to use them themselves, and to not restrict their children from
336
CLARK ET AL.
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
Table 5. Perceptions of the Influence of Parental Smoking on Teen Tobacco Use Any Adult Smoker in Household
No Adult Smoker in Household
Children who see their parents smoking are likely to try smoking Agreea 93 (73.9) 160 (88.4) Disagreeb 33 21 Parents who smoke have difficulty telling their children not to smoke Agree 72 (57.1) 159 (88.4) Disagree 54 21
OR (95% CI)
P Value
0.37 (0.21, 0.67)
0.001
0.17 (0.10, 0.30)
0.0001
‘‘Don’t know’’ responses not included. a Combines ‘‘strongly agree’’ and ‘‘agree.’’ b Combines ‘‘disagree’’ and ‘‘strongly disagree.’’
using them. Parents who smoked were also more likely to allow their children to use candy cigarettes (Table 6B). The African-American respondents did not differ significantly from the white respondents on any of these items, except that they were significantly less likely to allow the use of candy cigarettes (Table 6A). Availability and Monitoring of Cigarettes in the Homes of Adults Who Smoked The preliminary questionnaire administered to the 70 focus group participants revealed that most (68.2%) of the participants bought their cigarettes by the pack, yet 23.8% typically had more than one pack open at any time, and 50.8% often stored their open packs in unattended locations. The mean number of
separate rooms or places in which unopened packs were stored was two. Both the African-American and the white group members did virtually no monitoring of their tobacco supplies and did not attempt to limit children’s access to them. Frequently cited reasons were that they had never thought about it as a potential problem, or it was too inconvenient to lock up a product that they used so frequently. Those who had discovered their own tobacco in the possession of their children invariably were surprised, even those who had related earlier in the discussion that their own initial experimentation with tobacco had involved either cigarettes pilfered from family members or cigarettes that their friends had surreptitiously obtained from their parents’ or other adults’ tobacco supplies.
Table 6A. Ownership of, Use of, and Policies Regarding Tobacco Promotional Items African-American Parent
White Parent
OR (95% CI)
P Value
Do you own any product that promotes a tobacco brand or was distributed by a tobacco company? Yes 5 (10.2) 55 (21.7) 0.43 0.08 No 42 198 (0.14, 1.20) Do you think you would ever use a tobacco industry promotional item, such as a tee shirt? Yes 14 (27.5) 32 (17.3) 1.81 0.11 No 37 153 (0.82, 3.14) Which of the following statements best describes your policies about tobacco promotional items and your child? No Problema 14 (27.5) 74 (29.7) 0.89 0.75 Disapproveb 37 175 (0.43, 1.83) Which of the following statements best describes your policies about your child’s use of [candy cigarettes]? No Problema 13 (25.5) 124 (48.4) 0.36 0.003 Disapproveb 38 132 (0.17, 0.75) a Includes the responses: ‘‘I have no problem with my child using these items and would help obtain them for my child’’; ‘‘I have no problem with my child using these items, but would not help obtain them for my child’’; and ‘‘I don’t have any opinion about these items.’’ b Includes the responses: ‘‘I disapprove of my child using these items, but would not forbid it’’; and ‘‘I don’t allow my child to use these items.’’
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
337
Table 6B. Ownership of, Use of, and Policies Regarding Tobacco Promotional Items, by Parental Smoking Status Any Adult Smoker in Household
No Adult Smoker in Household
OR (95% CI)
P Value
Do you own any product that promotes a tobacco brand or was distributed by a tobacco company? Yes 46 (37.1) 15 (8.2) 6.74 No 76 167 (3.40, 13.51) ,0.001 Do you think you would ever use a tobacco industry promotional item, such as a tee shirt? Yes 61 (48.4) 32 (17.3) 4.49 No 65 153 (2.59, 7.79) ,0.001 Which of the following statements best describes your policies about tobacco promotional items and your child? No Problema 47 (37.6) 41 (22.9) 2.03 0.005 Disapproveb 78 138 (1.19, 3.46) Which of the following statements best describes your policies about your child’s use of [candy cigarettes]? No Problema 67 (53.2) 70 (37.8) 1.87 0.007 Disapproveb 59 115 (1.15, 3.03) a Includes the responses: ‘‘I have no problem with my child using these items and would help obtain them for my child’’; ‘‘I have no problem with my child using these items, but would not help obtain them for my child’’; and ‘‘I don’t have any opinion about these items.’’ b Includes the responses: ‘‘I disapprove of my child using these items, but would not forbid it’’; and ‘‘I don’t allow my child to use these items.’’
Of the 126 survey respondents who smoked, 35.7% reported usually buying cigarettes by the carton, 18.3% sometimes bought by the carton, and 19.8% typically had more than one pack open at any time. Only 13 stored their extra supplies in a locked place. When the survey respondents were asked “How likely would you be to notice that some of your cigarettes were missing?” 80 said that it was “very likely,” 14 “somewhat likely,” 15 “unlikely,” and 9 reported “it would be impossible to tell.”
Even so, the age suggested by African-American respondents was significantly higher than that of the white respondents (p , 0.001). Only one AfricanAmerican respondent named an age ,18 years, and 10 named an age .18 years; 67 of the white parents responded with an age ,18 years, and 39 named an age .18 years (Table 7). The suggested age was not significantly different for parents who did and did not smoke (p 5 0.32).
Discussion Age at Which Children or Teens Should Make Up Their Own Minds About Tobacco Use The last question asked of each individual member of the focus groups was, “At what age should children or teens be able to make up their own minds about whether or not to use tobacco?” The white parents, with few exceptions, stated that 16 years was the appropriate age. This choice was often accompanied by an observation such as, “If they can drive a car at 16 years, they can decide whether they want to smoke.” In contrast, the African-American parents answered either age 18 years or “When they are out of my house.” None of the African-American parents stated any age younger than 18 years. The same question was asked of the survey respondents. In the absence of group discussions preceding the question, the responses were not as uniform as those from the focus group participants.
Compared to the African-American parents, the white parents reported lower levels of involvement in anti-tobacco socialization within the home. They were less likely to report setting household ground rules regarding teen tobacco use and communicating the rules to the children. The African-American parents were less likely than the white parents to believe
Table 7. Age of Responsibility in Choosing to Use Tobacco African-American Parent
White Parent
Chi-Square
P-Value
At what age should children or teens be allowed to make up their own minds about whether or not to use tobacco? Age ,18 1 (2.0) 67 (26.2) 14.47 ,0.001 Age 18 40 (78.4) 150 (58.6) Age .18 10 (19.6) 39 (15.2)
338
CLARK ET AL.
that tobacco use is an inevitable part of growing up, and that parental interventions would have either no effect on teen use or make the teens “want it more.” They were also more likely than the white parents to believe that schools played a role in anti-tobacco socialization that was subordinate to their own. There are several possible explanations for the relatively low rates of African-American teen smoking. Whereas the delay in the initiation of tobacco use among African-Americans may contribute to the difference, it is unlikely that it accounts for the full magnitude of the discrepancy. In addition, the delay in youth smoking onset may in itself be a salutatory effect of the African-American beliefs and behaviors reported here. A causative relationship between active participation in anti-tobacco socialization and the relatively low rates of teen tobacco use in African-American homes is not certain, but evidence gathered from teens supports the supposition. For instance, in a National Cancer Institute study, utilizing focus group methods, parental advice to not smoke, regardless of parental smoking status, was implicated in the lower rates of smoking among African-American youth.(23) McDermott (26) reported that, regardless of parents’ own drug use behavior, adolescents who perceive their parents as holding permissive views about teen drug use were significantly more likely to use drugs than those who perceived their parents as having nonpermissive views. In a more recent report specific to tobacco use, Jackson and Henriksen (2) concluded that third- and fifth-grade children who experience anti-smoking socialization in the home report substantially lower onset rates, even when one or both parents currently smoke. Specifically, children exposed to one or two parents who smoked were at significantly greater risk for intending to smoke, perceiving easy access to cigarettes, and for having almost tried smoking, but those who believed that their parents would know if they were smoking, and those who expected negative consequences for smoking, reported lower smoking rates than those who did not hold such beliefs. Flay et al. (10) found that seventhgrade adolescents were more likely to copy smoking behaviors of their peers rather than of those of their parents, but that perceived parental disapproval of adolescent smoking modified that relationship. Hunter et al. (27) reported that parental approval contributed to continued smoking after initial trial. That tobacco use by parents is a risk factor for teen smoking seemed to be generally accepted among the adults in this study. This knowledge may have had unintended negative consequences. Feelings of guilt
JOURNAL OF ADOLESCENT HEALTH Vol. 24, No. 5
and powerlessness apparently discouraged the parents who smoked from openly talking about the issue, thereby disassociating themselves from their children’s decision-making regarding tobacco use. It is not known whether the number of parents who successfully quit when pressured by their children outnumbers those who have been silenced because of their continued use. The knowledge that parental tobacco use is a risk factor for children’s use may have worked against the parents who did not smoke, as well. They may have felt that they did not have to be concerned about their children initiating tobacco use and, therefore, had not bothered with this element of anti-tobacco socialization in the home. The ubiquitous pro-tobacco messages outside the home would therefore have had fewer anti-tobacco counter messages in those households in which there were no adult smokers (28). Educational efforts may be required for many parents to understand and support the proposed United States Food and Drug Administration rules outlawing tobacco promotional items.(29) In addition, 44% had no problem with their children using candy cigarettes. It may be that the parents did not associate the use of these items with promoting tobacco use as a social norm, or recognize the problem of candy cigarettes contributing to the identification of cigarettes as a treat. Protective practices regarding tobacco supplies were not found to be the norm in this study. The ready availability of unmonitored parental cigarette supplies suggests that the behavior of parents may be significant in any supply-side strategies to reduce tobacco use by youth. Whereas most such strategies have focused on reducing sales of tobacco to children, a campaign to convince parents (and the extended family) to handle adult tobacco supplies more responsibly also is important in reducing childhood tobacco use. Limitations Whereas the households likely to have age-eligible children initially were selected by random-digitdialing techniques and household screening, the respondents in the telephone survey reported here had all previously allowed their children to be interviewed regarding tobacco use. Whereas we can only speculate what effect this potential selection bias has on the results, the presence of some bias in the whole group estimates is evident from the unusually low refusal rate. It is implausible, however, that the effect would have operated differently among smokers and nonsmokers, or among African-American par-
May 1999
ANTI-TOBACCO SOCIALIZATION BY PARENTS
339
ents versus white parents, producing bias in the between-groups comparisons reported here. The samples for both the telephone survey and the focus groups were from a single region (northeast Ohio). It is not known if there are significant differences in attitudes and behaviors of parents in different regions of the country. A more broadly representative sample, including other ethnic groups, would be an appropriate next step.
10. Flay BR, Hu FB, Siddiqui O, et al. Differential influence of parental smoking and friends’ smoking on adolescent initiation and escalation of smoking. J Health Soc Behav 1994;35:248 – 65.
Conclusions
14. Biglan A, Ary D, Yudelson H, et al. Experimental evaluation of a modular approach to mobilizing antitobacco influences of peers and parents. Am J Community Psychol 1996;24:311–39.
These data support the hypothesis that AfricanAmerican parents feel more empowered to delay or prevent their children’s tobacco use than do white parents. The challenge facing tobacco control professionals will be to find ways to help empower all parents to become actively involved in safeguarding their children from this major health threat. In order to do so, parents must first be provided with more knowledge about the immediate and long-term health consequences of tobacco use in childhood, about the realities of (and ubiquitous nature of) tobacco promotion coming from outside the home, and about the powerful role they could play in protecting their children. This study was supported through an intramural grant program, the President’s Pioneer Award for Research.
References 1. Pierce JP, Fiore MC, Novotny TE, et al. Trends in cigarette smoking in the United States. JAMA 1989;261:61–5. 2. Jackson C, Henriksen L. Do as I say: Parent smoking, antismoking socialization, and smoking onset among children. Addict Behav 1997;22:107– 4. 3. Bauman KE, Koch GG, Lentz GM. Parent characteristics, perceived health risk, and smokeless tobacco use among white adolescent males. NCI Monogr 1989;8:43– 8. 4. Green G, MacIntyre S, West P, et al. Like parent like child? Associations between drinking and smoking behaviour of parents and their children. Br J Addict 1991;86:745–58. 5. Chassin L, Presson CC, Montello D, et al. Changes in peer and parent influence during adolescence: Longitudinal versus cross-sectional perspectives on smoking initiation. Dev Psychol 1986;22:327–34. 6. Eiser JR, Morgan M, Gammage P. Adolescent smoking: Attitudes, norms and parental influence. Br J Soc Pyschol 1989;28: 193–202. 7. Bauman KE, Foshee VA, Linzer MA, et al. Effect of parental smoking classification on the association between parental and adolescent smoking. Addict Behav 1990;15:413–22. 8. Chassin L, Presson CC, Sherman SJ, et al. Parent educational attainment and adolescent cigarette smoking. J Subst Abuse 1992;4:219 –34. 9. Covey LS, Tam D. Depressive mood, the single-parent home, and adolescent cigarette smoking. Am J Public Health 1990; 80:1330 –3.
11. Bauman KE, Botvin GJ, Botvin EM, et al. Normative expectations and the behavior of significant others: An integration of traditions in research on adolescents’ cigarette smoking. Psychol Rep 1992;71:568 –70. 12. Glendinning A, Shucksmith J, Hendry L. Family life and smoking in adolescence. Soc Sci Med 1997;44:93–101. 13. Melby JN, Conger RD, Conger KJ, et al. Effects of parental behavior on tobacco use by young male adolescents. J Marriage Fam 1993;55:439 –54.
15. Foxcroft DR, Lowe G. Adolescent drinking, smoking and other substance use involvement: Links with perceived family life. J Adolesc 1995;18:159 –77. 16. Nelson DE, Giovino GA, Shopland DR, et al. Trends in cigarette smoking among U.S. adolescents, 1974 –1991. Am J Public Health 1995;85:34 – 40. 17. Centers for Disease Control. Differences in the age of smoking initiation between blacks and whites—United States. MMWR 1991;40:754 –7. 18. Escobedo LG, Anda RF, Smith PF. Sociodemographic characteristics of cigarette smoking initiation in the United States. JAMA 1990;264:1550 –5. 19. U.S. Department of Health and Human Services. Report of the Surgeon General: Reducing the health consequences of smoking. Atlanta, GA: Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1989. 20. U.S. Department of Health and Human Services. Tobacco use in 1986. Methods & Basic Tabulations from Adult Use of Tobacco Survey. Atlanta, GA: Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1986. 21. Wang MQ, Cowdery JE, Trucks J, et al. Southern adolescents: A survey of attitudes and beliefs about smoking. S Med J 1994;87:1097–102. 22. Johnston LD, O’Malley PM, Bachman JG. National survey results on drug use from Monitoring the Future Study, 1975–1994. Vol. 1: Secondary school students. NIH Publication No. 95-4026. Rockville, MD: National Institute on Drug Abuse, 1995. 23. McIntosh H. Black teens not smoking in great numbers. J Natl Cancer Inst 1995;87:564 –5. 24. Gleenlund KJ, Johnson C, Wattigney W, et al. Trends in cigarette smoking among children in a southern community, 1976 –1994: The Bogalusa Heart Study. Ann Epidemiol 1996;6:476– 82. 25. Knodel J. The design and analysis of focus group studies. A practical approach. In: Morgan DL, ed. Successful Focus Groups. Newbury Park: Sage Publications, 1993:35–50. 26. McDermott D. The relationship of parental drug use and parents’ attitude concerning adolescent drug use to adolescent drug use. Adolescence 1984;19:89 –97. 27. Hunter SM, Baugh JG, Webber LS, et al. Social learning effects on trial and adoption of cigarette smoking in children: The Bogalusa Heart Study. Prev Med 1982;11:29 – 42. 28. Institute of Medicine, Committee on Preventing Nicotine Addiction in Youth and Children. Lynch BS, Bonnie RJ, eds. Growing Up Tobacco Free. Washington, DC: National Academy Press, 1994. 29. Federal Register, August 11, 1995.