Smoke-Free Outdoor Seating Policy: 1-Year Changes in Compliance of Bars and Restaurants in Philadelphia

Abstract

Purpose:

To evaluate changes in compliance with a smoke-free outdoor seating policy before and after passage of a local regulation in 2015, which reinterpreted Philadelphia’s Clean Indoor Air Worker Protection Law to include outdoor seating areas of food or beverage establishments.

Design:

Natural experiment.

Setting:

Philadelphia, Pennsylvania.

Sample:

Food or beverage establishments (N = 108). Establishments were comprised of sit-down restaurants, cafes, quick-service restaurants, and bars.

Measures:

Presence of outdoor smoking and smoking-related litter on a given day were measured as binary variables. A geographic information system–based survey developed for this study was used to collect observational data.

Analysis:

Logistic regressions were used to determine the change in odds of observing outdoor smoking and smoking-related litter on a given day from baseline (preregulation) to follow-up (postregulation).

Results:

Compliance with smoke-free outdoor seating increased from 84.5% to 95.4% after passage and implementation of the regulation. Results showed a significant 75% decrease (odds ratio [OR]: = 0.25, 95% confidence interval [CI]: 0.08-0.67) in odds of outdoor smoking and a slight decrease in smoking-related litter (OR: 0.81, 95% CI: 0.39-1.65) at follow-up in establishments overall. However, at baseline, bars had higher odds of outdoor smoking (OR: 2.68, 95% CI: 0.57-12.72) and smoking-related litter (OR: 4.09, 95% CI:, 1.87-9.49) compared to sit-down restaurants.

Conclusion:

Results suggest there is high compliance with low-cost, low-burden, smoke-free outdoor seating policy and that enforcement is best targeted toward bars.

Keywords

tobacco control smoke-free policy outdoor spaces smoking ban secondhand smoke health policy

Purpose

While smoke-free policies are widely regarded as effective public health interventions to reduce the harms of indoor secondhand smoke (SHS) exposure, smoke-free policies in outdoor spaces have historically garnered less support.¹ However, a growing body of evidence suggests that SHS exposure in outdoor settings can be comparable to exposure indoors² and that smoke-free outdoor policies can likely protect against SHS exposure indoors and out.^2,3 The current study aimed to evaluate changes in compliance in outdoor seating areas of food or beverage establishments before and after a 2015 local health regulation interpreted Philadelphia’s 2007 Clean Indoor Air Worker Protection Law (CIAWPL) to apply to all outdoor service areas of said establishments.⁴

Methods

Study Design

A site surveyor conducted observations at each establishment using a geographic information system–based survey app developed by the Department of Public Health. Observations were conducted from May to August of both 2015 (baseline) and 2016 (follow-up), during weekday business hours typically from 4:00 to 8:00 pm. Between observation periods, the regulation was passed effective immediately, creating a quasi-natural experiment before and after implementation. Follow-up observations were collected 7 to 10 months postimplementation. This study did not constitute human participants research as determined by the Department of Public Health institutional review board.

Policy Change

Although the 2007 CIAWPL included clear language prohibiting smoking indoors, the language prohibiting smoking in certain outdoor spaces was more complex, particularly with respect to applicability to establishments with CIAWPL exemptions. The complex language resulted in decreased enforcement of and compliance with outdoor smoking prohibitions. In October 2015, the Philadelphia Board of Health promulgated a regulation clarifying that the CIAWPL’s smoke-free outdoor food service mandate applied to all outdoor food and beverage service areas, even where establishments held CIAWPL exemptions applicable to indoor space.⁴ This facilitated uniform implementation of the existing CIAWPL enforcement mechanism, which consisted of complaint-based ticketing of smoking in outdoor food or beverage establishment service areas. Establishment owners received letters in advance of implementation, and health department inspectors issued warnings rather than tickets to establishment owners for the first 3 months after implementation if smoking occurred on site.

Sample

At baseline, 130 food or beverage establishments were randomly selected for observation within parameters set to reflect the geospatial distribution of 440 total establishments registered with the city of Philadelphia as having outdoor seating. The surveillance team determined that the sample size comprised a reasonable number of observations to be completed in a 3- to 4-month summer period, while ensuring geospatial distribution of sampled establishments remained aligned with that of registered establishments. Of the 130 selected samples, 129 observations were completed. At follow-up, 108 (83.7%) of establishments were reassessed and included in the final analysis (Figure 1). Incomplete observations were excluded from regression analysis if the establishment was permanently closed, the outdoor seating area was permanently removed, there was no record of postregulation visit, or a site was missing data on the smoking variable at baseline or follow-up.

Figure 1.

Enrollment flow of establishments included in the current study from baseline in 2015, 2 to 5 months prior to passage and implementation of smoke-free outdoor seating regulation, to follow-up in 2016, 7 to 10 months postregulation.

Measures

Establishment characteristics (eg, establishment type, number of tables in outdoor area) were recorded at each visit. A secondary classification for establishment type was optional—which allowed analysts to distinguish drinking establishments that primarily sell alcohol (ie, bars) from other establishments that primarily sell food, but might also sell alcohol (ie, sit-down, cafe/quick-service restaurants). The presence of observed smoking (including combustible tobacco or e-cigarette use) and smoking-related litter (defined as the presence of cigarette butts or cartons, including in ashtrays and trash receptacles) on a given day were recorded as binary variables.

Data Analysis

Logistic regression analyses were conducted to examine (1) changes in odds of observed outdoor smoking and (2) smoking-related litter from baseline to follow-up. Covariates included establishment type (with “sit-down restaurant” assigned as the reference category) and the number of tables in outdoor seating area to control for size. All analyses were conducted in R version 1.0.136.⁵

Results

Of 129 food or beverage establishments measured at baseline, nearly 46% (n = 59) were sit-down restaurants, 33% (n = 42) were drinking establishments (ie, bars), and 22% (n = 28) were cafes or quick-service establishments. The distribution of establishment types was similar at follow-up (Table 1).

Table 1.

Descriptive Characteristics and Unadjusted Prevalence of Observed Smoking and Smoking-Related Litter of Establishments With Outdoors Seating Areas 2 to 5 Months Prior to Passage and Implementation of a Smoke-Free Outdoor Seating Regulation (Baseline) and 7 to 10 Months Postimplementation (Follow-Up).

	Baseline	Follow-Up
Characteristic/Outcome	Count (% of Total)	Count (%)
Number of establishments surveyed	N = 129	N = 108
Establishment type
Sit-down	59 (45.7)	51 (47.2)
Drinking establishments	42 (32.6)	34 (31.5)
Cafe/quick service	28 (21.7)	23 (21.3)
All establishment types
Smoking	20 (15.5)	5 (4.6)
Smoking-related litter	22 (17.0)	16 (14.8)
Sit-down establishments	N = 59	N = 51
Smoking	3 (5.1)	1 (2.0)
Smoking-related litter	7 (11.9)	3 (5.9)
Drinking establishments	N = 42	N = 34
Smoking	12 (28.6)	4 (11.8)
Smoking-related litter	12 (28.6)	11 (32.4)

Observed Smoking

Results showed a statistically significant 75% decrease in odds of observed smoking (odds ratio [OR]: 0.25; 95% confidence interval [CI]: 0.08-0.67) at follow-up compared to baseline. At baseline, drinking establishments had significantly higher odds of observed smoking compared to sit-down restaurants (OR: 2.68, 95% CI, 0.57-12.72; Table 2). In all establishment types, the unadjusted prevalence of observed smoking was 15.5% (n = 20) at baseline and 4.6% (n = 5) at follow-up; in drinking establishments, the prevalence was 28.6% (n = 12) at baseline and 11.8% (n = 4) at follow-up (Table 1).

Table 2.

Odds Ratios of Observed Smoking and Smoking-Related Litter in Outdoor Seating Areas of Surveyed Establishments (N = 108).

Observed smoking in outdoor seating area	OR	Lower CI	Upper CI
Intercept	0.06	0.01	0.22
Year 2016	0.25	0.08	0.67
Sit-down	Ref	Ref	Ref
Drinking establishments	2.68	0.57	12.72
Smoking-related litter in outdoor seating area
Intercept	0.18	0.07	0.48
Year 2016	0.81	0.39	1.65
Sit-down	Ref	Ref	Ref
Drinking establishments	4.09	1.87	9.49

Abbreviations: CI, confidence interval; OR, odds ratio; Ref, reference group.

Smoking-Related Litter

Changes in patterns in smoking-related litter of all establishments were not statistically significant, but suggested a decreasing trend from 2015 to 2016 (OR: 0.81, 95% CI: 0.39-1.65). At baseline, drinking establishments had significantly higher odds of observed smoking-related litter (OR: 4.09, 95% CI: 1.87-9.49) compared to sit-down restaurants (Table 2).

Discussion

Summary

Odds of observed smoking in outdoor seating areas of food or beverage establishments declined by 75% after 1 year, a meaningful improvement from the previous year. Overall compliance with the smoke-free outdoor seating policy was high (95.4%), consistent with self-reported decline in SHS exposure after passage of smoke-free policies in outdoor seating areas of bars and restaurants in Canadian provinces⁶ and particulate matter analyses demonstrating bars and restaurants tend to comply with smoke-free policies.⁷ The slight decline demonstrated in observed smoking-related litter after regulation implementation is consistent with a comprehensive smoke-free outdoor policy analysis conducted in Vancouver,⁸ but differs from an evaluation of New York City’s smoke-free outdoor parks and beaches policy that found large declines.⁹

Notably, odds of observed smoking at baseline were over twice as high at drinking establishments compared to sit-down restaurants, and 4 times as high for observed smoking-related litter. This is unsurprising, given the tendency for alcohol and tobacco use to co-occur¹ and the tendency for those outlets to colocate.¹⁰ Nevertheless, the prevalence of observed smoking in outdoor seating areas of bars also declined following the passage of the regulation (29%-12%). These findings suggest that type of establishment is important to consider for enforcement and that effort may be best targeted toward drinking establishments.

Limitations

The lack of control group precludes any definitive statements about the effectiveness of the regulation in reducing SHS exposure in outdoor seating areas. Cigarette smoking has been steadily declining,¹¹ which could be an alternative explanation for the observed decline. However, the decline in unadjusted prevalence of observed smoking over 1 year was relatively large (15.5%-4.6%), and the magnitude of the decline remained meaningful when adjusting for size and establishment type. Additionally, a single site visit may not be sufficient to reliably estimate compliance as factors such as time of day, day of week, and variable staff attitudes toward smoke-free outdoor seating within an establishment may also influence customer compliance.

Significance

Given the greater presence of enclosed outdoor spaces in urban settings, which are associated with greater SHS exposure,² the high level of compliance with Philadelphia’s smoke-free outdoor seating policy is reassuring. To facilitate compliance, smoke-free policies may benefit from including feasible mechanisms for enforcement that do not contribute to health inequities,⁸ and focused attention to drinking establishments. Although some controversy remains regarding smoke-free outdoor spaces, public support for such policies is growing. The implementation of smoke-free policies may increase public support of smoke-free spaces and perceived social unacceptability of smoking¹; therefore, it is important to continue to implement, strengthen, and evaluate compliance with smoke-free outdoor policies.

SO WHAT?

What is Already Known About the Topic?

Smoke-free policies are effective in reducing harmful SHS exposure when fully implemented.

What Does this Article Add?

An evaluation of Philadelphia’s smoke-free outdoor seating policy demonstrated increased compliance after a regulation clarifying the policy was passed in 2015; however, lower compliance was observed at bars than sit-down restaurants.

What are the Implications for Health Promotion Practice or Research?

For other municipalities considering smoke-free outdoor policies, the findings suggest that low-cost, low-burden smoke-free space policies may be a promising approach to limiting SHS exposure in restaurants and that enforcement may be best targeted toward bars.

Footnotes

Acknowledgments

The study authors would like to thank Reem Hamoda and Naomi Hollo, who assisted with data collection, Amory Hillengas who designed the data collection tool, and Dr. Cheryl Bettigole for her feedback on the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Pilar Ocampo, MPH

References

Thomson

Wilson

Collins

Edwards

. Attitudes to smoke-free outdoor regulations in the USA and Canada: a review of 89 surveys. Tob Control. 2016;25(5):506–516. doi:10.1136/tobaccocontrol-2015-052426.

Sureda

Fernández

López

Nebot

. Secondhand tobacco smoke exposure in open and semi-open settings: a systematic review. Environ Health Perspect. 2013;121(7):766–773. doi:10.1289/ehp.1205806.

Fernández

Martínez-Sánchez

, et al.

Second-hand smoke exposure in indoor and outdoor areas of cafés and restaurants: need for extending smoking regulation outdoors?

Environ Res. 2016;148:421–428. doi:10.1016/j.envres.2016.04.024.

City of Philadelphia Department of Public Health Board of Health. Regulations Relating to the Clean Indoor Air Worker Protection Law. Vol 10-602; 2015. http://regulations.phila-records.com/pdfs/Department%20of%20Public%20Health%20Regs.%2011-16-15.pdf, Accessed May 28, 2019.

R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2016. https://www.r-project.org/. Accessed February 15, 2017.

Azagba

. Effect of smoke-free patio policy of restaurants and bars on exposure to second-hand smoke. Prev Med. 2015;76:74–78. doi:10.1016/j.ypmed.2015.04.012.

Wilson

Edwards

Parry

. A persisting secondhand smoke hazard in urban public places: results from fine particulate (PM2.5) air sampling. N Z Med J. 2011;124(1330):34.

Pederson

Okoli

Hemsing

, et al. Smoking on the margins: a comprehensive analysis of a municipal outdoor smoke-free policy. BMC Public Health. 2016;16(1):1–13.

Johns

Coady

Chan

Farley

Kansagra

. Evaluating New York City’s smoke-free parks and beaches law: a critical multilist approach to assessing behavioral impact. Am J Community Psychol. 2013;51(1-2):254–263. doi:10.1007/s10464-012-9519-5.

10.

Shortt

Tisch

Pearce

, et al. A cross-sectional analysis of the relationship between tobacco and alcohol outlet density and neighbourhood deprivation. BMC Public Health. 2015;15:1014. doi:10.1186/s12889-015-2321-1.

11.

Syamlal

. Tobacco use among working adults — United States, 2014–2016. MMWR Morb Mortal Wkly Rep. 2017;66(42):1130–1135. doi:10.15585/mmwr.mm6642a2.