Abstract
We conducted a nationwide survey to evaluate the effect of implementing a smoke-free policy in municipalities that forbid teachers to smoke on school premises. Questionnaires were mailed to 3207 municipalities throughout Japan. After we adjusted for population size and the standardized mortality ratio for male lung cancer, we found that assigning a high priority to tobacco control in municipal health promotion activities was significantly associated with implementation of school tobacco-control policies (odds ratio = 1.50, 95% confidence interval=1.24, 1.81).
More than 80% of male smokers start smoking before age 20.1,2 Therefore, any adolescent smoking prevention program needs to include implementation of a school smoking policy and programs about social influences on smoking.3,4 Poulsen et al.5 found that adolescent smoking behavior was influenced by teachers’ smoking behavior during school hours. To our knowledge, few studies have evaluated the effect of municipalities’ public health policies on the implementation of a complete smoke-free policy that prohibits anyone, including teachers, from smoking on school premises. We used data from a nationwide survey in Japan to report on the prevalence of complete smoke-free school policies in relation to the priority given to municipal tobacco-control activities.
METHODS
The questionnaires were mailed to the health promotion sections of 3207 municipalities throughout Japan in July 2003. They included the following 3 items:
the respondent’s profession;
whether a complete smoke-free policy on school premises was implemented in the elementary and junior high schools of the municipality (all schools, some schools, or no schools);
the priority of school tobacco-control policies within the municipality’s health promotion activities (high, intermediate, or low).
Data Analysis
Categorical variables were tested with the χ2 test and the χ2 test for linear trend. The t test or the Mann–Whitney test was used to compare continuous variables.
The population size (2000 census data) and the life expectancy for men, as well as the standardized mortality ratio (SMR) for male lung cancer in 1999 (estimated by the Ministry of Health, Labour, and Welfare), were examined as potential confounders. These variables were divided into quintiles for analysis.
Bivariate and multivariate logistic regression models were constructed to estimate the odds ratios with a 95% confidence interval. In the models, the implementation status (of a smoke-free policy on school premises) as a dependent variable was dichotomized to either “implemented” or “not implemented” by merging the responses “yes at all schools” and “yes at some schools” into “implemented.” The priority levels also were dichotomized to a positive category (“high”) or a negative category (“intermediate” or “low”). SMR quintiles were divided into 3 categories—(1) first, (2) second to fourth, and (3) fifth—because of a significant nonlinear association with the implementation status.
RESULTS
Of the 3207 municipalities, 2570 (80.1%) responded. No statistically significant differences were observed in the life expectancy and the SMR for lung cancer for men between the municipalities that responded and those that did not, except for population size (median values: responding = 11 483, nonresponding = 8140; P <.001). Public health nurses accounted for 80.5% of the respondents.
Valid answers for the questions on implementation of a smoke-free policy and the priority of tobacco-control activities were available from 2246 municipalities (87.4% of all municipalities responding). Three hundred twenty-two (14.3%) municipalities implemented a complete smoke-free policy in all elementary and junior high schools, 408 (18.2%) did so in some of the schools, and 1516 (67.5%) had not implemented any complete smoke-free policies. The proportions of these responses did not differ significantly by whether the respondent was a public heath nurse.
Table 1 ▶ shows the prevalence of implementation of a smoke-free policy in schools in relation to other factors. A smoke-free policy was less likely to be implemented in municipalities that assigned a low priority to tobacco-control activities (P <.001). The school smoke-free policy was more likely to be implemented in municipalities with a large population size (P <.001) and in the first and fifth quintile of the SMR for male lung cancer (P <.005). No significant relation between life expectancy and implementation was observed.
TABLE 1—
Association of the Implementation of a Complete Smoke-Free Policy for School Premises With the Priority of Tobacco-Control Policy, Population Size, and Vital Statistics in Japanese Municipalities
| Implementation Status of a Complete Smoke-Free Policy for School Premises | ||||||
| Implemented in All Schools | Implemented in Some Schools | No Implementation | Totala | |||
| n (%) | n (%) | n (%) | n (%) | P for χ2 Test | P for Trend | |
| Priority given to tobacco-control policy in health promotion activities | ||||||
| High | 218 (16.4) | 271 (20.4) | 839 (63.2) | 1328 (100) | <.001 | <.001 |
| Intermediate | 93 (11.0) | 129 (15.3) | 620 (73.6) | 842 (100) | ||
| Low | 5 (16.1) | 3 (9.7) | 23 (74.2) | 31 (100) | ||
| Population size | ||||||
| <5000 | 71 (15.5) | 47 (10.3) | 340 (74.2) | 458 (100) | <.001 | <.001 |
| 5000–9999 | 83 (14.8) | 75 (13.4) | 403 (71.8) | 561 (100) | ||
| 10 000–29 999 | 86 (12.9) | 126 (18.9) | 455 (68.1) | 668 (100) | ||
| ≥30 000 | 82 (14.6) | 160 (28.6) | 318 (56.8) | 560 (100) | ||
| Life expectancy for men, y b | ||||||
| <76.8 | 64 (13.3) | 80 (16.6) | 338 (70.1) | 482 (100) | NS | NS |
| 76.8–77.2 | 69 (16.4) | 68 (16.2) | 283 (67.4) | 420 (100) | ||
| 77.3–77.7 | 71 (14.9) | 79 (16.6) | 325 (68.4) | 475 (100) | ||
| 77.8–78.0 | 60 (14.9) | 70 (17.4) | 273 (67.7) | 403 (100) | ||
| ≥78.1 | 58 (12.4) | 111 (23.8) | 297 (63.7) | 466 (100) | ||
| SMR for male lung cancer b | ||||||
| <76 | 64 (15.2) | 59 (14.0) | 299 (70.9) | 422 (100) | <.005 | NS |
| 76–91 | 65 (14.2) | 99 (21.7) | 293 (64.1) | 457 (100) | ||
| 92–104 | 59 (13.1) | 100 (22.2) | 292 (64.7) | 451 (100) | ||
| 104.1–120 | 66 (13.9) | 92 (19.4) | 317 (66.7) | 475 (100) | ||
| >120 | 68 (15.4) | 58 (13.2) | 315 (71.4) | 441 (100) | ||
Note. NS = not statistically significant; SMR = standardized mortality ratio.
aSome numbers do not equal 2570 because of missing values.
bEach interval is defined by quintile.
As indicated in Table 2 ▶, after we adjusted for the population size and the SMR for lung cancer in men, a high priority given to tobacco-control policy in municipal health promotion activities was significantly associated with the implementation of a complete smoke-free school policy (odds ratio = 1.50; 95% confidence interval = 1.24, 1.81).
TABLE 2—
Logistic Regression Predicting the Implementation of a Complete Smoke-Free Policy for School Premises, by the Priority of Tobacco-Control Policy in Municipalities, Population Size, and SMR for Male Lung Cancer
| Bivariate Model | Multivariate Model | |
| OR (95% CI) | Adjusted OR (95% CI) | |
| Priority given to tobacco-control policy in health promotion activities | ||
| Low or intermediate | Reference | Reference |
| High | 1.63 (1.35, 1.97*) | 1.50 (1.24, 1.81*) |
| Population size | ||
| <5000 | Reference | Reference |
| 5000–9999 | 1.13 (0.86, 1.49) | 1.13 (0.85, 1.51) |
| 10 000–29 999 | 1.34 (1.03, 1.75*) | 1.26 (0.96, 1.66) |
| ≥& 30 000 | 2.19 (1.68, 2.87*) | 1.94 (1.46, 2.59*) |
| SMR for male lung cancer | ||
| <76 | 0.77 (0.61, 0.98**) | 0.97 (0.75, 1.24) |
| 76–121 | Reference | Reference |
| >121 | 0.75 (0.59, 0.95**) | 0.86 (0.67, 1.10) |
Note. SMR = standardized mortality ratio; OR = odds ratio; CI = confidence interval.
*P <.001; **P <.05.
DISCUSSION
A complete smoke-free school policy was significantly more likely to be implemented in municipalities in which tobacco control had a high priority among health promotion activities. This finding appears to be compatible with the results of the Massachusetts survey of local restaurant smoking regulations6 and may provide a clue to the problem of the limited efficacy of school-based smoking programs that do not include an enforced tobacco-control policy.7,8 Our results could add weight to the concept of enforcing a stronger public health policy for tobacco control at the local level.
The first limitation of this study was its cross-sectional design. Second, data were gathered through questionnaires that may have been biased by responders’ attitudes and their social and cultural environments. Consequently, the prevalence of implementation could have been overestimated. Finally, these results do not address other current issues in tobacco control in schools.9 Further studies are needed to elucidate the association between tobacco control in schools and municipalities’ public health priorities.
Acknowledgments
The study was supported by a grant from the Ministry of Health, Labour, and Welfare of Japan (grant H15-ganyobou-023).
The authors thank Yoshihiko Miura for his statistical advice and Michiko Kawashima for her computational assistance and database management.
Human Participant Protection No protocol approval was needed for this study.
Peer Reviewed
Contributors K. Kayaba wrote the brief with H. Yanagawa. K. Kayaba conducted the analyses with C. Wakabayashi, N. Kunisawa, and H. Shinmura. H. Yanagawa originated the study and supervised all aspects of its implementation.
References
- 1.Sarraf-Zadegan N, Boshtam M, Shahrokhi S, et al. Tobacco use among Iranian men, women and adolescents. Eur J Public Health. 2004;14:76–78. [DOI] [PubMed] [Google Scholar]
- 2.Osaki Y, Minowa M, Suzuki K, Wada K. Adolescent smoking behavior in Japan, 1996. Nihon Arukoru Yakubutsu Igakkai Zasshi. 2003;38:483–491. [PubMed] [Google Scholar]
- 3.Nakamura M, Kitayama T, Nishioka N, Inoue M. Smoke-free schools. Jpn J Sch Health. 2004;45: 502–504. [Google Scholar]
- 4.Griesbach D, Inchley J, Currie C. More than words? The status and impact of smoking policies in Scottish schools. Health Promot Int. 2002;17:31–41. [DOI] [PubMed] [Google Scholar]
- 5.Poulsen LH, Osler M, Roberts C, Due P, Damsgaard MT, Holstein BE. Exposure to teachers smoking and adolescent smoking behaviour: analysis of cross sectional data from Denmark. Tob Control. 2002;11: 246–251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Skeer M, George S, Hamilton WL, Cheng DM, Siegel M. Town-level characteristics and smoking policy adoption in Massachusetts: are local restaurant smoking regulations fostering disparities in health protection? Am J Public Health. 2004;94:286–292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wakefield MA, Chaloupka FJ, Kaufman NJ, Orleans CT, Barker DC, Ruel EE. Effect of restrictions on smoking at home, at school, and in public places on teenage smoking: cross sectional study. BMJ. 2000; 321:333–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rooney BL, Murray DM. A meta-analysis of smoking prevention programs after adjustment for errors in the unit of analysis. Health Educ Q. 1996;23: 48–64. [DOI] [PubMed] [Google Scholar]
- 9.Laugesen M, Scollo M, Shiffman S, et al. World’s best practice in tobacco control. Tob Control. 2000;9: 228–236. [DOI] [PMC free article] [PubMed] [Google Scholar]