Printer Friendly

Cohort study of smoke-free homes in economically disadvantaged communities in the Dominican Republic/Estudio de cohortes de hogares libres de humo de tabaco en comunidades economicamente desfavorecidas de la Republica Dominicana.

Rates of smoking and associated secondhand smoke (SHS) exposure are increasing globally, with the greatest increases occurring in low- and middle-income countries (LMICs) (1). Further reducing SHS exposure is a basic tenet of tobacco control to reduce the prevalence of smoking, establish regulations restricting tobacco use in workplaces and public areas, and promote smoke-free homes (2-6). SHS exposure affects morbidity and mortality among nonsmoking adults (development of lung cancer and cardiovascular problems) and children (increased risk of Sudden Infant Death Syndrome (SIDS), acute and chronic respiratory problems, and ear problems) (3-7). The U.S. Surgeon General's report cites the home as the primary site of major SHS exposure (7-9). Home smoking bans are directly associated with better health status (9-11) and represent another mechanism to reduce the differential exposure and increased vulnerability experienced by disadvantaged populations (12).

While regulations can limit or eliminate SHS exposure outside the home, regulating exposure inside homes is the household's purview. U.S. tobacco control efforts promoting smoke-free homes have led to increases in household bans. The Tobacco Use Supplement to the Current Population Survey compared data on household smoke-free rules between 1992 and 2003 (4). This continuous monthly household survey conducted by the U.S. Census Bureau for the Bureau of Labor Statistics of the U.S. civilian, noninstitutionalized population queries respondents (> 14 years old) about smoking inside the home. By 2003, 72.2% of households surveyed (n = 127 332) across the 50 states and the District of Columbia reported being smoke-free (not allowing smoking anywhere inside the home)--an increase of 67.1% from the 1992-1993 rate of 43.2% (4). In addition, the 2008 Behavioral Risk Factor Surveillance Survey, a state-based, random-digit-dialed telephone survey conducted among a similarly defined group of adults older than 17 years, reported household smoke-free bans across 11 participating states. Nonsmokers were more likely to report a complete household ban (median: 87.7%) compared to smokers (45.0%) (13).

This trend is not unique (14); the International Tobacco Control (ITC) Four Country Survey found similar trends in other English-speaking countries (15). Among the more than 9 000 randomly selected smokers, household bans increased between 2002 and 2003 to 19.0%, 31.5%, and 43.1% in Canada, United Kingdom, and Australia respectively.

The presence of children and nonsmoking adults increased the likelihood of a home being designated as smoke-free. Subsequent studies reconfirmed these findings (16-20).

The Dominican Republic (DR) is historically a significant tobacco-growing LMIC and remains the only Latin American or Caribbean country that has not signed onto the Framework Convention on Tobacco Control (2). This lack of national action on tobacco control is consistent with the country's limited national and local tobacco control regulations despite increasing tobacco-related morbidity and mortality rates (21). Although DR regulations limiting SHS exposure in public or work areas were enacted in 2000, awareness of any restrictions was virtually absent (21, 22). No awareness campaigns about the deleterious effects of tobacco use on the smoker or of SHS on the nonsmoker have been conducted. In 2003, during the initial (qualitative) phase of the project described in this study, community members from six economically disadvantaged communities described places where smoking was not allowed (e.g., local businesses (manager decision); selected public places (e.g., churches); and public transit) out of "respeto" (respect) for others. These actions represented community self-regulation rather than adherence to governmental regulations (21). Few community members mentioned passive smoking (SHS) or household restrictions during the 2003 interviews. Individuals choosing to smoke outside did so to hide their smoking, not to reduce others' SHS exposure. This lack of knowledge is consistent with elevated risk exposure (12).

Despite the above history, there are no published studies of household smoking bans in the DR or Caribbean. This report provides the first analysis of household smoking-ban prevalence over time and associated factors among economically disadvantaged DR communities.


These analyses used data drawn from a larger study of tobacco cessation in the DR known as Proyecto Doble T (Project Double T, with the "Ts" standing for technology and tobacco). The project began with formative research about tobacco attitudes and practices in six economically disadvantaged DR communities, followed by household and smoker surveillance surveys (21-23). Post-survey, the U.S.-DR research team deployed a multifaceted intervention (posters, fairs, community charlas (talks)) in the three intervention communities to raise awareness about the health effects of smoking and to provide cessation resources (e.g., training local intervention specialists and both professional and para-professional health care workers) (23, 24). The intervention was replicated in control communities following a one-year comparison period. The cessation intervention was not specifically designed to promote household smoking bans.

Household surveillance data were collected systematically in six communities (two small urban, two peri-urban, [6] and two rural; within each pair, one was tobacco-growing and one was not, and one was a control and the other an intervention site). Data collection took place in 2004 (baseline survey) and was repeated in 2006 post-intervention (first follow-up survey) and again in 2007 after the control communities received the intervention (second follow-up survey). Initial analyses of household smoking bans (which included data from all three surveys) and subsequent analyses (which were limited to the follow-up surveys) were conducted using SAS/ STAT software, version 9.1.3 (SAS System for Windows, SAS Institute, Cary, NC, USA).

As described in Ossip-Klein et al. (23), each surveillance survey included 170-175 households per community, for a total of 1 052 homes in the baseline survey; 1 040 in the first follow-up survey; and 1 048 in the second follow-up survey. All three surveys used the same procedures (randomly selected households approached by trained data collectors hired by the DR-based team). In the baseline survey, 207-241 households per community were approached, for a completion rate of 73%-85%. The first follow-up survey covered 201-250 households per community (71%-87%) and the second follow-up survey covered 183-222 households per community (78%-96%).

The households surveyed at baseline were not systematically included or excluded from the follow-up surveys. As each survey represented less than 10% of available households, inclusion in subsequent surveys was unlikely. Interviews were conducted with the first available adult member (> 18 years old) who could respond to questions regarding basic household demographics, tobacco use, and health conditions of all adult household members.


The survey item to assess smoke-free home practices was adapted from the ITC policy survey (24). Modifications were implemented based on pretesting to ensure that the survey item would be readily understood by the respondents. The question used to assess household smoking restrictions (English translation) was "Which of the following best describes smoking in your household?" (possible responses: "Smoking is allowed in your home"; "Smoking is never allowed in your home"; "Smoking is allowed in your home for some people"; "Smoking is allowed in your home just in some places"; and "Other"). Additional survey items were drawn from multiple standard questionnaires and translated into Spanish using a back-translation method (25).


Descriptive analyses of sample characteristics included testing for trend. Subsequent bivariate analyses of changes in household smoking bans (chi-square and Student's i-test) and multivariate analyses (logistic regressions) were undertaken to identify factors associated with household bans. Model fit was assessed using the Pearson method. To protect respondent anonymity the analyses used only de-identified household data. Verbal consent and survey administration procedures, developed and approved by institutional review boards and independent ethics committees in the United States and the DR, were used (26).


The characteristics of the households responding to the three surveys varied (Table 1). More than 30% of all households surveyed had at least one member who smoked. The decreasing population over time was statistically significant (38.6% at baseline versus 30.6% in the second follow-up survey; P < 0.0001). In all three surveys more than 20% of households had a member with a respiratory illness (23.3% at baseline, 27.8% in the first follow-up survey, and 23.8% in the second followup survey; differences not statistically significant). The increase over time in households that had a member with a cardiovascular problem was statistically significant (32.7% at baseline, 40.5% in the first follow-up survey, and 40.9% in the second follow-up survey; P < 0.0001). More than 80% of all households surveyed were Catholic, with a statistically significant decline over the three periods (from 89.2% to 85.6% to 83.4%; P < 0.001). No other statistically significant differences were found.

For analytical purposes, and to focus on the target behavior (establishment of a household smoking ban), responses to the question "Which of the following best describes smoking in your household?" were dichotomized into "No smoking allowed' and "Any smoking allowed." Aggregating households across all communities, those reporting a total smoking ban increased from 23.9% at baseline to 35.9% (first follow-up survey) and 45.3% (second follow-up survey). As depicted in Table 2, these differences were statistically significant (P < 0.0001; test for trend).

Comparisons by community type (urban, peri-urban, and rural) at baseline revealed smoking bans among 21.5%27.5% of households, with statistically significant increases in household smoking-ban uptake in all communities by the second follow-up survey (P < 0.001). This increase was lowest among rural communities (36.7%) but at least doubled in peri-urban and urban communities (45.9% and 53.6% respectively). Between-community differences were not statistically significant (data not shown), so subsequent analyses used data aggregated across community type.

Statistically significant increases (P < 0.0001) were found for both intervention and control community households (data not shown). As expected, a difference-of-differences analysis comparing intervention and control conditions between baseline and the second follow-up survey did not show statistically significant results because the tobacco cessation intervention did not focus on passive smoking or smoke-free homes. Therefore, subsequent analyses used data aggregated across intervention conditions.

Additional chi-square analyses compared the presence of household smoking bans based on being in a tobacco-growing community or not. Initially, only 19.6% of households in a tobacco-growing community had bans, compared to 28.2% in non-tobacco-growing communities. By the second follow-up survey (2007), the increase in smoking bans in both types of communities was statistically significant, with larger increases occurring in non-tobacco-growing communities (59.8% versus 30.8% in tobacco-growing communities).

The difference in smoking ban prevalence between households with at least one smoker and households with no smokers (irrespective of community type) was statistically significant (Table 2). While statistically significant increases in household smoking bans were found in both groups (P < 0.0001), including high-risk (smoker) homes, the actual difference in the proportion of households with smoking bans was striking. For example, by 2007 (the second follow-up survey), 57.7% of households without a smoker had smoking bans, compared to only 16.7% of households with a smoker (P < 0.0001). A difference-of-differences analysis confirmed the absence of any intervention effect on the smoker households or the nonsmoker households (data not shown). Subsequent analyses were undertaken separately for households with and without a smoker.

The final analyses focused on factors associated with households allowing any smoking (under any circumstance) versus households with complete smoking bans. Five household variables and one community variable were examined based on prior research (16-18) and the initial (qualitative) findings of the project (19). These variables included: religion ("Catholic" versus "not Catholic," with the latter group including those who did not answer the question and those who provided the following responses: "Adventist," "Evangelist," "voodoo," "don't know," and "none"); presence of at least one household member < 5 years old or > 64 years old; household member with a respiratory condition (aggregate variable included asthma, cough, and pulmonary or respiratory problem); and member with a cardiovascular problem (heart disease and/or hypertension). The one community characteristic was: "tobacco-growing."

Table 3 depicts the distribution of these characteristics stratified by households with and without a smoker for the baseline survey and the second follow-up survey, comparing those with and without household smoking bans. Among households without a smoker, those that allowed smoking were more likely to be in tobacco-growing communities (48.4%) relative to those with a ban at baseline (39.5%; P < 0.05). This statistically significant difference increased in the second follow-up survey, which showed that nearly two-thirds (61.8%) of households without a smoker that allowed smoking were in tobacco-growing communities, where only 29.3% of these households had bans (P < 0.001). In the second follow-up survey, households without a smoker but allowing smoking were also more likely than those banning smoking to have a member with a cardiovascular problem (44.9% versus 32.2%; P < 0.001) and were more likely to be Catholic (87.3% versus 77.5%; P < 0.001).

For households with a smoker the only variable with a statistically significant association with not having a smoking ban at baseline was having a household member with a respiratory condition (27.5% of households without a ban had a household member with a respiratory condition, compared to only 8.0% among households with bans; P = 0.032) (Table 3). No statistically significant differences were found for this variable in the second follow-up survey.

All six variables were entered into full logistic regression models to identify factors associated with allowing household smoking. For households without a smoker, the models for the baseline and second follow-up surveys demonstrated good fit (based on the Pearson chi-square test) and had statistically significant associations (Table 4). At baseline, being in a tobacco-growing community was the only variable with a statistically significant association with allowing household smoking (odds ratio (OR): 1.46; 95% confidence interval (CI): 1.04, 2.07).

The OR increased in the second follow-up survey (2007), when being in a tobacco-growing community made it more than three times more likely that a household would allow smoking (OR: 3.75; 95% CI: 2.71, 5.20). In that survey, being a Catholic household and having someone in the household with a cardiovascular problem were both associated with increased likelihood that smoking would be allowed (OR: 1.87; 95% CI: 1.20, 2.90 and OR: 1.77; 95% CI: 1.24, 2.53 respectively).

Among households with a smoker, no statistically significant associations and poor model fit were found for both the baseline and second follow-up survey (data not shown). This may be due to the relatively small number of households in that subgroup that banned smoking.


The prevalence of household smoking bans among six economically disadvantaged communities in the DR increased from 23.9% at baseline to 45.3% two years later. While promising, there remains considerable room for improvement across all communities, as more than half of the participating households continued to allow smoking with some or no restriction, well below rates in industrialized countries (5, 6).

The observed increase in smoking bans occurred in the absence of any national or local tobacco control activities outside of the study project. Because the project did not focus on household smoking bans or passive smoke exposure, the uptake of bans was, as expected, non-differential between intervention and control communities. Given the absence of prior community interventions related to passive smoke, asking the question about having a household ban (along with multiple other questions regarding tobacco and health risks) in the baseline surveillance survey may have itself raised awareness about the option of banning household smoking. This phenomenon has been described by others (27-28). Increases also differed by type of community and by whether any household member smoked. Prevalence of household bans increased over time in households with and without a smoker but was dramatically higher among households without a smoker.

In households with a smoker, despite small but statistically significant increases in smoking bans, no factor had any consistent association (predictive or protective) with allowing household smoking. This finding, consistent with previous studies (29-32), reaffirms that having a smoker in the house may mitigate the establishment of these bans. In contrast, Binns et al. (18) found that a ban was more likely if the smoker lived with a nonsmoker or with a child under age 5.

In examining only households without a smoker, being located in a tobacco-growing community was strongly associated with allowing smoking at baseline. While being in a tobacco-growing area likely detracts from tobacco control efforts (15), this study demonstrates its specific mitigating influence on home smoking bans.

While not significant at baseline, several findings have implications for future interventions at the community and national levels, particularly in relation to the general population's awareness of the broad impact of smoking on both healthy and unhealthy adults. Having a household member with a cardiovascular problem was strongly associated with allowing household smoking, even when controlling for the presence of an older household member. While it may seem counterintuitive to allow smoking in a nonsmoking household, especially in the presence of someone with a cardiovascular problem, this phenomenon may be explained by the fact that individuals with a cardiovascular problem may be former smokers, so the household may simply be continuing a practice of allowing smoking that began prior to the household member's diagnosis. It may also be attributable to community members' general lack of awareness of the effect of smoking on cardiovascular health (20, 22), or to household members having friends who smoke. Several authors noted that external factors (e.g., prevalence of friends or visitors who smoke or have cravings) might decrease the establishment of complete home smoking bans (19-20, 29-32). Finally, increased cardiovascular disease prevalence may have been a consequence of SHS exposure in homes with no ban (19). Understanding the far-reaching health effects of SHS exposure is an important missed intervention opportunity that could lead to greater prevalence of household smoking-ban implementation.

Contrary to previous studies (11, 17, 19-20), in this study, having children or someone with a respiratory problem in the home did not influence house hold smoking bans. This result, which was similar to Ji et al.'s finding among urbanized Chinese households (30), underscores the need for awareness-raising on the harms of SHS to increase the prevalence of home-smoking bans and reduce SHS exposure among children and nonsmoking adults.

The relationship between religion and household smoking bans has not been previously identified. Being Catholic was associated with allowing household smoking even among households without a smoker. Based on the qualitative work conducted for this project, this may be attributable to the strong anti-smoking messages promulgated by the other prevalent religions in the DR (Adventist and Evangelical). Religion and religious beliefs do not feature prominently in findings from other research on home smoking bans (17-20, 29-32).

Study limitations and strengths

The major limitation of this study is reliance on self-report by the household member interviewed. Respondents may have provided socially acceptable responses. Given the prevalence of smoking, lack of tobacco control initiatives, and low awareness of smoking's health effects, it is not clear which responses would be considered socially acceptable in these communities. In addition, while it is likely that some of the findings are applicable to other disadvantaged DR or Caribbean communities, caution is warranted to not overgeneralize.

The major strength of this exploratory analysis is its surveillance in understudied disadvantaged communities in a historically tobacco-growing country with limited tobacco control initiatives. The focus on disadvantaged communities within an LMIC provides an opportunity to address health disparities within vulnerable populations who bear the greatest brunt of the tobacco epidemic (12). The finding that smoke-free homes increased in these communities demonstrates that this type of change is feasible and initial results can be achieved without significant effort. The use of local data collectors, rigorous sampling and surveillance techniques, and local capacity building for tobacco control provided for adequate numbers of participating households for analysis (23).

The study results also documented key differences in household smoking bans between households with and without a smoker, and across community types, and demonstrated the potential ripple effect of a smoking cessation project on home smoking restrictions.


In communities with no prior tobacco control initiatives, the prevalence of households banning smoking increased to nearly 50% overall, possibly due to the surveillance process itself. While households with and without a member who smoked were more likely to report smoking bans over the course of the study, a more dramatic increase was found among those without smokers. This is encouraging for an LMIC such as the DR that devotes few national resources to tobacco control. For communities in an LMIC (and by extension LMICs) in the early stages of tobacco control initiatives, small measures may have large effects. Including a question about household bans on health surveys or providing information at health fairs or through other media may lead to significant increases in the prevalence of smoking bans, particularly among nonsmoking households. Given the newly discovered relationship between smoking bans and religion, collaboration with or outreach through churches should also be considered.

This study also highlights the far-reaching effect of being in a tobacco-growing community, underscoring the challenge of implementing tobacco control interventions such as household smoking bans within that environment. Consistent with the role of social determinants on health among disadvantaged populations (12), across all households, lack of knowledge about the harmful health effects of smoking appeared to influence whether household bans were established. Programs specifically designed to encourage discussion of the detrimental health effects of SHS on children and adult nonsmokers (healthy and unhealthy) may be essential to increase further the prevalence of household smoking bans. These may need to be specifically tailored for tobacco-growing communities.

Establishing smoking bans in households with smokers represents a separate challenge. Where tobacco control initiatives are few, interventions specifically designed to target homes with smokers may be warranted.

Given these promising findings, the DR is encouraged to implement policies and programs to promote the establishment of household smoking bans. While different strategies may be needed for homes with smokers, increasing bans among households without a smoker may lead to changes in community norms, which in turn may influence ban establishment in households of smokers.

Acknowledgments. The authors thank the office staff at the University of Rochester and at the headquarters of Proyecto Doble T in the Dominican Republic; the site coordinators and data collectors in the participating communities; and the survey respondents who gave their time to this project.

Funding. This investigator-initiated work was supported by the National Institutes of Health (NIH) Fogarty International Center (grant no. TWO5945) and National Cancer Institute (NCI) (grant no. R01 CA132950). The funder played no other role in this work.

Conflicts of interest. None.


(1.) World Health Organization. WHO report on the global tobacco epidemic, 2011: warning about the dangers of tobacco. Geneva: WHO; 2011. Available from: tobacco/global_report/2011/en/index.html Accessed on 20 December 2013.

(2.) World Health Organization. WHO Framework Convention on Tobacco Control. Geneva: WHO; 2005.

(3.) Centers for Disease Control and Prevention (US). Best practices for comprehensive tobacco control programs--2007. Atlanta: U.S. Department of Health and Human Services, CDC, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2007.

(4.) Centers for Disease Control and Prevention (US). Vital signs: nonsmokers' exposure to secondhand smoke--United States, 1999-2008. MMWR Morb Mortal Wkly Rep. 2010;59(35):1141-6.

(5.) Fong GT, Cummings KM, Shopland DR; ITC Collaboration. Building the evidence base for effective tobacco control policies: the International Tobacco Control Policy Evaluation Project (the ITC Project). Tob Control. 2006;15 Suppl 3:iii1-2.

(6.) Wipfli H, Avila-Tang E, Navas-Acien A, Kim S, Onicescu G, Yuan J, et al. Secondhand smoke exposure among women and children: evidence from 31 countries. Am J Public Health. 2008;98(4):672-9.

(7.) Centers for Disease Control and Prevention (US). The health consequences of involuntary exposure to tobacco smoke: a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, CDC, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2006.

(8.) Lando HA, Hipple BJ, Muramoto M, Klein JD, Prokhorov AV, Ossip DJ, et al. Tobacco control and children: an international perspective. Pediatr Allergy Immunol Pulmonol. 2010;23(2):99-103.

(9.) Hill SC, Liang L. Smoking in the home and children's health. Tob Control. 2008;17(1):32-7.

(10.) Wamboldt FS, Balkissoon RC, Rankin AE, Szefler SJ, Hammond SK, Glasgow RE, et al. Correlates of household smoking bans in low-income families of children with and without asthma. Fam Process. 2008;47(1):81-94.

(11.) Behm I, Kabir Z, Connolly GN, Alpert HR. Increasing prevalence of smoke-free homes and decreasing rates of sudden infant death syndrome in the United States: an ecological association study. Tob Control. 2012;21(1):6-11.

(12.) Blas E, Kurup AS, editors. Equity, social determinants and public health programmes. Geneva: World Health Organization; 2010.

(13.) Centers for Disease Control and Prevention (US). State-specific secondhand smoke exposure and current cigarette smoking among adults--United States, 2008. MMWR Morb Mortal Wkly Rep. 2009;58(44):1232-5.

(14.) Sebrie EM, Schoj V, Glantz SA. Smokefree environments in Latin America: on the road to real change? Prev Control. 2008;3(1):21-35.

(15.) Borland R, Yong HH, Cummings KM, Hyland A, Anderson S, Fong GT. Determinants and consequences of smoke-free homes: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15 Suppl 3:iii42-50.

(16.) King BA, Hyland AJ, Borland R, McNeill A, Cummings KM. Socioeconomic variation in the prevalence, introduction, retention, and removal of smoke-free policies among smokers: findings from the International Tobacco Control (ITC) Four Country Survey. Int J Environ Res Public Health. 2011;8(2):411-34.

(17.) Hawkins SS, Berkman L. Parental home smoking policies: the protective effect of having a young child in the household. Prev Med. 2011;53(1-2):61-3.

(18.) Binns HJ, O'Neil J, Benuck I, Ariza AJ; Pediatric Practice Research Group. Influences on parents' decisions for home and automobile smoking bans in households with smokers. Patient Educ Couns. 2009;74(2):272-6.

(19.) Shelley D, Fahs MC, Yemeni R, Qu J, Burton D. Correlates of household smoking bans among Chinese Americans. Nicotine Tob Res. 2006;8(1):103-12.

(20.) Yousey YK. Household characteristics, smoking bans, and passive smoke exposure in young children. J Pediatr Health Care. 2006;20(2):98-105.

(21.) Dozier AM, Ossip-Klein DJ, Diaz S, Chin NP, Sierra E, Quinones Z, et al. Tobacco use in the Dominican Republic: understanding the culture first. Tob Control. 2006;15 Suppl 1: i30-6.

(22.) Dozier AM, Ossip DJ, Diaz S, Sierra-Torres E, Quinones de Monegro Z, Armstrong L, et al. Health care workers in the Dominican Republic: self-perceived role in smoking cessation. Eval Health Prof. 2009;32(2):144-64.

(23.) Ossip-Klein DJ, Fisher S, Diaz S, Quinones Z, Sierra E, Dozier A, et al. Tobacco use in six economically disadvantaged communities in the Dominican Republic. Nicotine Tob Res. 2008;10(5):851-60.

(24.) Thompson ME, Fong GT, Hammond D, Boudreau C, Driezen P, Hyland A, et al. Methods of the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15 Suppl 3:iii12-8.

(25.) Brislin RW. Back-translation for cross-cultural research. J Cross Cult Psychol. 1970;1(3):185-216.

(26.) McIntosh S, Sierra E, Dozier A, Diaz S, Quinones Z, Primack A, et al. Ethical review issues in collaborative research between us and low-middle income country partners: a case example. Bioethics. 2008;22(8):414-22.

(27.) Morwitz VG. The effect of survey measurement on respondent behaviour. Appl Stoch Models Bus Ind. 2005;21(4-5):451-5.

(28.) Gine X, Karlan D, Zinman J. The risk of asking: measurement effects from a baseline survey in an insurance takeup experiment. Cambridge: Harvard University Center for International Development; 2007. Available from: neudc07/docs/neudc07_s4_p12_gine.pdf Accessed on 12 February 2012.

(29.) McMillen RC, Winickoff JP, Klein JD, Weitzman M. US adult attitudes and practices regarding smoking restrictions and child exposure to environmental tobacco smoke: changes in the social climate from 2000-2001. Pediatrics. 2003;112(1 Pt 1):e55-60.

(30.) Ji M, Ding D, Hovell MF, Xia X, Zheng P, Fu H. Home smoking bans in an urbanizing community in China. Am J Prev Med. 2009;37(2):132-6.

(31.) Pizacani BA, Martin DP, Stark MJ, Koepsell TD, Thompson B, Diehr P. Longitudinal study of household smoking ban adoption among households with at least one smoker: associated factors, barriers, and smoker support. Nicotine Tob Res. 2008;10(3):533-40.

(32.) Escoffery C, Kegler MC, Butler S. Formative research on creating smoke-free homes in rural communities. Health Educ Res. 2009;24(1):76-86.

Manuscript received on 12 September 2013. Revised version accepted for publication on 2 January 2014.

Ann M. Dozier, [1] Sergio Diaz, [2] Joseph Guido, [3] Zahira Quinones de Monegro, [4] Scott McIntosh, [1] Susan G. Fisher, [5] and Deborah J. Ossip [1]

[1] Social and Behavioral Sciences, Department of Public Health Sciences, University of Rochester, Rochester, New York, United States of America. Send correspondence to: Ann M. Dozier, Ann_dozier@urmc.

[2] Centro de Atencion Primaria Juan XXIII, Santiago, Dominican Republic.

[3] Department of Biostatistics and Computational Biology, University of Rochester, Rochester, New York, United States of America.

[4] Pontificia Universidad Catolica Madre y Maestra, Santiago, Dominican Republic.

[5] Epidemiology, Department of Public Health Sciences, University of Rochester, Rochester, New York, United States of America.

[6] Urban structure but in a remote location.
TABLE 1. Characteristics of households (a) responding to
baseline/follow-up surveillance surveys conducted before/
after tobacco cessation intervention in six economically
disadvantaged communities, Dominican Republic, 2004-2007 (b)

Characteristic                      Baseline        First
                                    survey (pre-    follow-up
                                    intervention)   survey (post-
                                    (n = 1 052)     intervention)
                                                    (n = 1 050)

                                     No.      %      No.      %

Has at least one member who:
  Smokes                             406    38.6     346    33.0
  Has respiratory condition (c)      245    23.3     292    27.8
  Has cardiovascular problem (d)     344    32.7     425    40.5
  Is < 5 years old                   259    24.6     302    28.8
  Is > 64 years old                  288    27.4     284    27.1
Household religion is Catholic       938    89.2     899    85.6
In tobacco-growing community         525    49.9     523    49.8

Characteristic                      Second                 P
                                    follow-up       (test for trend)
                                    survey (post-
                                    (n = 1 048)

                                     No.      %

Has at least one member who:
  Smokes                             321    30.6         0.0001
  Has respiratory condition (c)      249    23.8         0.4005
  Has cardiovascular problem (d)     429    40.9        < 0.0001
  Is < 5 years old                   274    26.2         0.2174
  Is > 64 years old                  286    27.3         0.4822
Household religion is Catholic       874    83.4         0.0002
In tobacco-growing community         525    50.1         0.9479

(a) Mean size: baseline survey, 3.7 individuals (standard
deviation (SD): 2.04); first follow-up survey, 3.8 (SD: 2.11);
second follow-up survey, 3.6 (SD: 1.93) (P = 0.3277).

(b) Baseline survey conducted in 2004; follow-up surveys conducted
in 2006 and 2007.

(c) Asthma, cough, or pulmonary or respiratory problem.

(d) Heart disease or hypertension.

TABLE 2. Household smoking-ban prevalence (%) (a) by
household characteristic (type of community,
with/without smoker) based on surveys conducted before/after
tobacco cessation intervention in six economically
disadvantaged communities, Dominican Republic, 2004-2007

                           % households with total
                           smoking bans (No. of
                           households per variable)

Characteristic              Baseline    First follow-up
                             survey         survey

Community type
  Rural                    21.5 (349)     33.2 (350)
  Peri-urban               22.8 (342)     32.0 (347)
  Urban                    27.5 (346)     42.6 (343)
  Non-tobacco-growing      28.2 (521)     43.0 (519)
  Tobacco-growing          19.6 (516)     28.8 (521)
With no smokers            35.2 (634)     48.2 (695)
With at least one smoker   6.2 (403)      11.0 (345)

                           % households with           P
                           total smoking bans   (test for trend)
                           (No. of households
                             per variable)

Characteristic              Second follow-up

Community type
  Rural                        36.7 (343)           < 0.001
  Peri-urban                   45.9 (349)           < 0.0001
  Urban                        53.6 (334)           < 0.0001
  Non-tobacco-growing          59.8 (513)           < 0.0001
  Tobacco-growing              30.8 (513)           < 0.0001
With no smokers                57.7 (711)           < 0.0001
With at least one smoker       16.7 (312)           < 0.0001

(a) Baseline survey (2004), 23.9% (n = 1 037); first follow-up
survey (2006), 35.9% (n = 1 040); second follow-up survey
(2007), 45.3% (n = 1 026)(P < 0.0001).

(b) Subsample totals reflect missing values for this survey item.

TABLE 3. Distribution of household characteristics associated with
allowing smoking (being Catholic; having a member < 5 or > 64 years
old, or with a respiratory or cardiovascular problem; being in a
tobacco-growing community) in households with/without a smoker based
on surveys conducted before/after tobacco cessation intervention In
six economically disadvantaged communities, Dominican Republic,
2004-2007 (a, b)

                                  Households with a smoker

                                  Baseline survey
                                  (n= 403)

                                  Smoking      Smoking ban
                                  Allowed      (n= 25)
                                  (n= 378)

Characteristic                    No.    %     No.     %       P

Has member with                   104   27.5    2     8.0    0.032
  respiratory condition
Has member with                   139   36.8    8    32.0    0.631
  cardiovascular condition
Has member < 5 years old          86    22.8    5    20.0    0.750
Has member > 64 years old         131   34.7   11    44.0    0.344
Household religion Is Catholic    357   94.4   25    100.0   0.226
In tobacco-growing community      216   57.1   13    52.0    0.615

                                  Households with a smoker

                                  Follow-up survey
                                  (n = 301)

                                  Smoking      Smoking ban
                                  allowed      (n = 52)
                                  (n= 257)

Characteristic                    No.    %     No.    %       P

Has member with                   81    31.2   19    36.5   0.448
  respiratory condition
Has member with                   123   47.3   29    55.8   0.265
  cardiovascular condition
Has member < 5 years old          58    22.3   12    23.1   0.903
Has member > 64 years old         93    35.8   15    28.9   0.338
Household religion Is Catholic    233   90.7   43    82.7   0.090
In tobacco-growing community      169   65.0   38    73.1   0.261

                                  Household without a smoker

                                  Baseline survey
                                  (n= 634)

                                  Smoking      Smoking ban
                                  allowed      (n= 223)
                                  (n= 411)

Characteristic                    No.    %     No.    %       P

Has member with                   88    21.4   46    20.6   0.818
  respiratory condition
Has member with                   130   31.6   62    27.8   0.317
  cardiovascular condition
Has member < 5 years old          102   24.9   63    28.3   0.356
Has member > 64 years old         99    24.1   45    20.2   0.262
Household religion Is Catholic    360   87.6   183   82.4   0.076
In tobacco-growing community      199   48.4   88    39.5   0.031

                                  Household without a smoker

                                  Follow-up survey
                                  (n= 711)

                                  Smoking      Smoking ban
                                  allowed      (n= 410)
                                  (n= 301)

Characteristic                    No.    %     No.    %       P

Has member with                   66    21.9   71    17.3   0.123
  respiratory condition
Has member with                   135   44.9   132   32.2   <0.001
  cardiovascular condition
Has member < 5 years old          84    27.9   113   27.6   0.919
Has member > 64 years old         75    24.9   97    23.7   0.699
Household religion Is Catholic    262   87.3   317   77.5   <0.001
In tobacco-growing community      186   61.8   129   29.3   <0.001

(a) Baseline survey conducted In 2004; second follow-up survey
conducted In 2007.

(b) Missing values represent < 5% of total.

TABLE 4. Odds ratios for allowing smoking in households without
a smoker by associated characteristic based on surveys conducted
before/after tobacco cessation intervention in six economically
disadvantaged communities, Dominican Republic, 2004-2007 (a,b)

Characteristic (c)      Odds ratio (CI (d))

                        Baseline survey      Follow-up survey
                        (n= 632)             (n = 709)

Household member
  Has cardiovascular    1.14 (0.77, 1.67)    1.77 (e) (1.24, 2.53)
  Has respiratory       0.90 (0.59, 1.38)    0.88 (0.58, 1.34)
    condition (f)
  < 5 years old         0.90 (0.62, 1.31)    1.04 (0.72, 1.50)
  > 64 years old        1.22 (0.80, 1.85)    0.89 (0.59, 1.33)
  Catholic              1.50 (0.94, 2.37)    1.87 (e) (1.20, 2.90)
  Tobacco-growing (g)   1.46 (e) (1.04, 2.07)3.75 (e) (2.71,5.20)

(a) Baseline survey conducted in 2004; second follow-up survey
conducted in 2006.

(b) Missing values represent < 5% of total.

(c) Reference: absence of listed characteristic.

(d) CI: 95% confidence interval.

(e) Statistically significant.

(f) Asthma, cough, or pulmonary or respiratory problem.

(g) Each pair of study communities included one that was
COPYRIGHT 2014 Pan American Health Organization
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Original research/Investigacion original
Author:Dozier, Ann M.; Diaz, Sergio; Guido, Joseph; de Monegro, Zahira Quinones; McIntosh, Scott; Fisher, S
Publication:Revista Panamericana de Salud Publica
Date:Jan 1, 2014
Previous Article:Prevalence of root caries among adults and the elderly in southeast Brazil/Prevalencia da carie radicular na populacao adulta e idosa da regiao...
Next Article:Contribution of extended school hours to the nutritional and physical activity profile of schoolchildren/Contribuicao do turno escolar estendido para...

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters