By: Rithy Techavoan Yean, in partial completion of the California Consortium for Public Health Informatics and Technology (CCPHIT) certificate program during undergraduate studies at California State University, Long Beach (CSULB).
Supervised by: Toni Espinoza-Ferrel, Lecturer at the College of Human Health and Services at CSULB
Why is second-smoking bad?
Second-hand smoke is smoke that comes from the burning of cigarettes from someone breathing out smoke when they are smoking. Being in the vicinity can make you sick as cigarettes contain harmful chemicals such as PAHs (polycyclic aromatic hydrocarbons) which the Center of Disease Control and Prevention states can cause cancer, heart disease, birth defects, and lung problems in adults (CDC, 2024). So it is a human carcinogen, meaning it is cancer causing and therefore is detrimental to our community's health.
1 in 10 American workers breathe second-hand smoke regularly while working in restaurants, casinos, and hotels. Moreover, outdoor workplaces and private homes are usually not covered by smoke-free policies (CDC, 2024).
In recent years, Orange County has made great strides in their policies to prohibit smoking which has helped reduce second-hand smoke exposure for residents. However, not every city in Orange County has implemented smoke-free policies. Some cities do not have any smoke-free policies while some do. The difference between smoke ordinance implementation is a cause for concern for residents in cities that do not have any smoke-free policies as they risk second-hand smoke exposure. Moreover, tobacco is considered the number one preventable cause of death in California as local action is a great deterrence in controlling tobacco usage and improving public health outcomes.
The following project aims to analyze the current smoke ordinance progress of cities within Orange County and highlight any policy gaps to encourage lawmakers to advocate for policy change. According to the American Lung Association, California "used to have some of the strongest laws in the nation to protect people from harmful second hand smoke exposure" but these state laws mostly focus on workplaces and indoor areas. There were only a few laws in placed that actually prohibits secondary smoking in outdoor areas and hence it is pertinent to see how many cities
Methodology:
Smoke ordinance data of Orange County is taken directly from the 2024 State of Tobacco Control Report which is a report published by the American Lung Association that evaluates the states and federal government's actions to eliminate tobacco use. The report assigns grades ranging from A to F to states based on their implementation of smoke policies. The California report assigns grades to every city and every county by focusing on five key areas: Smokefree Outdoor Air, Smokefree Housing, Reducing Tobacco Sales, Flavored Tobacco Restrictions, and an Overall Tobacco Control Grade. The Overall Tobacco Control Grade is calculated based on the grades the city or county receives in the four aforementioned areas.
The American Lung Association awards points to each city based on each area. The scale is A = 4, B = 3, C = 2, D = 1, and F =0 and the overall grade is the cumulative of all the points. Our focus is mostly on the Smokefree Outdoor Air policies which explores smokefree policies in the following categories: (1) Dining Areas, (2) Entryways, (3) Public Events, (4) Recreation Areas, (5) Service Areas, (6) Sidewalks in Commercial Areas, and (7) Worksites.
The Smokefree Housing area focuses on: (1) Nonsmoking Apartments, (2) Nonsmoking Condominiums, and (3) Nonsmoking Common Areas. The scale is 0-4 with 4 meaning complete 100% restriction of smoking in the Apartments, Condominiums, and Common Areas.
A brief summary of each category in each category is provided below:
Dining Areas: Restrictions on smoking in outdoor seating areas at restaurants and bars.
Entryways: Restrictions on smoking in places within a certain distance of doors, windows, and other openings into all enclosed areas where smoking is prohibited.
Public Events: Smoking restrictions at events open to the public such as fairs, farmer's marketplace, parades, concerts, and other similar events that take place on public property.
Service Areas: Smoking restrictions in outdoor locations where people stand or wait for services including ATM lines, public transit, stops, taxi stands, and ticket lines.
Recreation Areas: Smoking restrictions at parks, beaches, trails, and other similar recreation areas.
Sidewalks in Commercial Areas: Smoking restrictions on sidewalks and other pedestrian walkways in commercial areas, such as downtown areas, and outdoor shopping centers.
Worksites: Smoking restrictions in outdoor places of employment, such as construction sites.
Nonsmoking Apartments: Prohibiting smoking within the units of multi-unit apartment buildings.
Nonsmoking Condominiums: Prohibiting smoking within the units of multi-unit condominium buildings.
Nonsmoking Common Areas: Smoking restrictions in both indoor common areas, which includes hallways, stairwells, laundry rooms, and recreation rooms, and outdoor common areas, which include swimming pools, play areas, outdoor eating areas, and courtyards of multi-housing units.
For Dining Areas, Entryways, Public Events, Service Areas, and Recreation Areas, a score of 4 implies that the city by law has 100% smokefree regulation in that category. For Sidewalks in Commercial Areas and Worksites, the criteria is a score of 1 or 0 with the score of 1 implying partial restriction on smoking. Our map will mostly focus on the 100% smokefree policies thereby focusing only on the complete prohibition of smoking in Dining Areas, Entryways, Public Events, Service Areas, Recreation Areas, Nonsmoking Apartments, Nonsmoking Condominiums, and Nonsmoking Common Areas.
Extraction, Transform, and Load Process (ETL):
The ETL process is a data integration method where raw data is taken from a source or various sources and it is transformed into a standardized and cleaned format suitable for analysis which is then loaded into a data warehouse. This process is necessary to retrieve, prepare, and utilize data for interpretable and meaningful insights. Ideally, ETL process is automated but due to my limited experience with some of the computer science applications, the data from the 2024 State of Tobacco Control Report is manually entered into an excel sheet as scraping the PDF could risk a potential IP address ban as the website itself does not allow web scraping as it can infringe on the website's term of service and privacy. While manually entering the data is subject to human error, the excel sheet and the report is checked thoroughly to ensure that each data point was written down accurately. A similar process is usually observed in clinical studies and so manually entering the data is not an uncommon occurrence but rather a practiced method that is validated rigorously.
After the data is taken into the excel sheet, the smokefree categories will then be transformed into binary variables 0-1. Since the project is focused on 100% smokefree policies, any score that is not a 4 would imply that the city did not fulfill its smokefree obligation in that area. Hence, that city will then be assigned a score of 0 if the area does not have 100% smokefree laws and a score of 1 that means the area in that city has 100% smokefree laws. Afterwards this data will be loaded into a Google Spreadsheet and then linked to an integrated development environment (IDE) such as RStudios and JupyterNotebook for preliminary analysis and visualization.
Preliminary Analysis:
Raw Data
The raw data is read into JupyterNotebook where preliminary analysis was conducted to uncover trends in smoke ordinance in orange county. The code that showcases the exploratory data analysis of the dataset can be found in the reference section at the bottom of the website.
Key Findings
No city in Orange County has 100% smoking restriction in Apartments, Condominiums, and Common Areas.
Placentia, Dana Point, Irvine, Laguna Beach, and La Palma have partial restrictions on smoking for sidewalks of commercial areas.
Buena Park, Dana Park, Irvine, Laguna Beach, and La Palma have partial restrictions on smoking in worksites.
Laguna Beach has the highest Overall Tobacco Control Grade of 22 followed by Irvine and Dana Point with grades of 20 each respectively.
Laguna Beach has the most smokefree policies with 5 smokefree policies followed by Buena Park, Dana Point, Irvine, Laguna Woods, and Laguna Hills all with 4 smokefree policies.
Yorba Linda, Orange, Westminster, and Rancho Santa Margarita have the lowest grade of 0.
Orange County Unincorporated which includes areas like Ladera Ranch, Coto de Caza, Emerald Bay, and more have no 100% smoke-free policies.
Transformed Data
As mentioned in the methodology section, the smoke-free grade for each category is transformed into a binary categorical variable of 0-1 where 0 equates to smoke-free grades that are not equal to 4 and 1 equates to smoke-free grades that are equal to 4. Every value of 1 means that the city has 100% smoke-free policy in that category. The total number of 100% smoke-free policies are aggregated on the last column of the dataset. The transformed dataset is stored onto a Google Spreadsheet as shown below:
The Map - Visualizing the Data
Once the data has been read, a visualization of the smoke ordinance in the form of a map is created. Why did I choose to create an interactive map? Were there alternative visualizations like a pie chart or bar graphs? Interactive map in the Geographic Integrated Systems (GIS) field helps to reveal hidden patterns and relationships, address challenges found in specific locations, and allow organizations and relevant stakeholders to optimize resources to help those locations. By using a map, we can see which regions are lacking in smoke ordinance and determine the next steps that policy makers can do to tackle that.
To create this map, the following libraries were used:
Leaflet: This is a Javascript library that allows users to create interactive maps. Javascript works alongside HTML and CSS to allow websites to have interactivity with HTML defining the structure and content of the website and CSS handling the styling of the website.
Tigris: To create a map of orange county, there is a need to gain access to U.S Census Bureau which defines how big each city is in the county and how big the county is a whole. This measurement of the area of each city is accurately measured by the U.S Census Bureau. The Tigris library contains shapefiles of cities and counties across the United States of America. Shapefiles are common geospatial vector data format that stores the location, shape, and attribute information of a certain geographic information. Without shapefiles, there would be no coordinate boundaries for the regions in our map and the Tigris library provides convenient access to these shapefiles without needing to find them on the U.S Census Bureau itself.
USA Boundaries: This library contains a basemap of the United States of America which provides a background for our map to sit on.
Simple Functions (sf): The sf library provides a standardized way to working, encoding, and manipulating spatial vector data in R. This helps with calculating the centroids of each area or any other metrics of the spatial data. This is a very useful library for working with both 2D and 3D coordinate data.
Dplyr: This library allows users to manipulate data and dataframes in R. Several useful functions include filter(), select(), mutate(), and more. The dplyr package comes with the Tidyverse package which is normally downloaded in R to allow users to have access to a variety of data science related libraries.
GoogleSheets4: This library allows RStudios to interact with Google Sheets using the Sheets Application Program Interface (API). This allows R to access the data within the Google Sheet that the user authorizes access to. In this case, the data was contained on Google Sheets to allow potential revisions to data points when new information for subsequent fiscal years become available.
Features of the Map
User-Friendliness: The importance of user-friendliness is to ensure that the targeted consumer of a technology, software, or solution is able to use it with minimal difficulties where the user interface is convenient, easy, and intuitive. The map was designed to keep this in mind, particular to help any non tech-savvy individuals to use and understand the map without requiring further research.
Icons: The map itself is very easy to use as users can simply click on the blue icon on each smoke-free city to be directed to smoke guidelines of that city. This enables to read up on the policies of the cities that they are interested in learning about. The icons themselves are placed on the center of each region and are reasonably sized so that users cannot miss them. Moreover, there are clear instructions that says "Click on the blue marker and the link" which is very direct and helps the user know what top do as well.
Zoom: The zooming feature can be achieved using the scroll wheel on the mouse or trackpad which allows users to focus on certain parts of the map up close if they wish to. Zooming out provides a broader perspective that provides a wider view of the map which really highlights the disproportionate smoke-free policies of each city in comparison to Orange County in its totality.
Legend: There is a legend to the left side of the map which states that any region colored in green equals to smoke-free and any region colored in gray equals to non-smoke free.
Clear Title: The title of the map is "Orange County Interactive Smoke Ordinance Map of 2024" is clear and straightforward. By specifying the year, it grounds the map to 2024 and so any subsequent years would require a new report with new data to supplement the map. The title is descriptive enough to let the user know what they are looking at.
Clear Region Label: Each city region is defined by its territory that is calculated by that city's shapefile. The borders of each city is differentiated by a thick orange line. When a user hovers over a region, there is a small pop-up that says what that region is for clarity. The orange is an aesthetic chosen because of the name Orange County.
Orange City Unincorporated Widget: The widget that allows users to turn off the Orange City Unincorporated areas off is a quality of life for those only interested in the main cities.
The map is saved as local HTML file which can then embedded on websites if required. This is achieved by hosting the map on GitHub pages which enables the Javascript to work. Now the interactive map can be used on this Google Sites or any other site given that it is linked correctly. For a detailed breakdown of how this map was made, I have provided a documentation of every block of code that was used. This documentation can serve as a way for you to reproduce this map or pursue a similar project but to create a different map from the 2024 State of Tobacco Control report. Given that many counties were surveyed by the report, there are boundless opportunities to create smoke ordinance map of counties like LA, Alameda, or San Diego.
You can read the documentation to get started yourself!
Conclusion
Results:
By pure introspective look, we can see that a majority of the map is mostly grey. Visually it is clear that Orange County has around 50 % - 60 % smoke-free coverage which suggests many of the cities still need to implement smoke ordinance to ensure the public safety of their residents.
Empirically, only 20 of the cities have at least 1 smokefree policy. 4 cities (Buena Park, Dana Point, Irvine, Laguna Hills, and Laguna Woods) have 4 smoke-free policies and 1 city (Laguna Beach) has 5 smoke-free policies.
20 out of 35 cities (including unincorporated areas) have at least 1 smoke-free policy which means the entire county has only 57 % smoke-free coverage.
No cities have 100 % smoke free worksites, sidewalks in commercial areas, condominiums, apartments, and common areas.
Takeaway for Policy-Makers:
Unincorporated areas are areas that do not have a central municipal government and so it is supervised by the Orange County Board of Supervisors which oversees the community's governance, land use, and public works. The nature of these unincorporated areas is that they are less densely populated and therefore there is less exposure to second-hand smoke overall. Hence, these areas could have laxed smoke restrictions. Ideally, there should be smoke ordinance policies in place to help regulate smoking regardless to avoid residents in these areas dealing with second-hand smoke exposure.
More densely populated areas generally have robust smoking restrictions as the risk for smoking is higher for the collective public health. Hence, why there are more health advocacy groups that are more active and have more resources to push for comprehensive tobacco control policies. Irvine is a potential example of this pattern. However, it should be noted that Anaheim, Huntington Beach, and Fullerton while have high populations do not have effective tobacco control policies.
A surprising takeaway is that no city in California is prevented from passing stricter tobacco laws than the state. California law provides minimum standards for ordinance in recreation areas, housing, and outdoor public spaces yet the government still encourage each city to adopt better measures to prioritize their community's collective health. This is due to a court ruling in 1998 City of San Jose v. Department of Health Services which affirms that local governments have the innate right to enact stronger smoke laws. However, this is not the case and it is apparent that smoke ordinance is disproportionate. Perhaps there are underlying factors that contribute to less smoke ordinance such as population, income level, level of education, poverty rate, crime rate, just to name a few. Further research and data will be needed to supplement these claims.
Future Work
Within our framework, we denote a city green if they have at least 1 smoke-free policy. A more robust framework would be to use different hues of green to designate cities that have less smoke-free policies. That way, it can show percent coverage of smoke-free per city so that it is more indicative of the smoke ordinance progress. Having only one color can show that there is more smoke ordinance when it reality some of these cities only have 1 smoke-free policy at the moment.
Given more manpower and time, it is possible to map the entire smoke ordinance of every county in California. This could be a good visualization to see the progress of smoke prohibition in the whole state which could leverage better sway for policy-makers. Moreover, it would be interesting to compare different counties and see the progress of each county. Perhaps we can uncover patterns or contributors that make certain cities or counties more smoke-free than others.
One hypothesis is that high education and income can lead to more comprehensive tobacco control though the extent of these factors have not really been looked at. Perhaps if data is available, this map can be further refined by also showing the literacy rate and the income levels and showcase whether or not those factors help lead to stronger smoking restrictions.
What is clear is that second-hand smoke is a serious public health concern for our communities, but second-hand smoke effects can be greatly mitigated with the correct preventative measures such as proper smoke ordinance. It is imperative that local governments and health officials implement smoke restrictions to protect the collective health of their communities.
Acknowledgements
Thank you to my supervisor, Toni Espinoza-Ferrel, for her constructive feedback and valuable support throughout this internship. The internship complemented by academic endeavors by enabling me to apply the knowledge and skills accumulated in the California Consortium for Public Health Informatics and Technology program towards a real-world situation. This experience greatly enhanced my data analysis and data visualization skills in R and Python while also teaching me newer programming skills using HTML and Javascript.
Thank you to Dr. Brian Kwan and Dr. Lisa Sparks for their relentless efforts to place me in this internship given my background as an International Student. They worked tirelessly behind the scenes for this opportunity to happen.
Thank you to Gora Datta for promoting the importance of CCPHIT throughout California.
Thank you to Dr. Kamiar Alaei for awarding me with the 2025 Outstanding Student Citation Award in recognition of my completed work in this internship.
References
Americans for Nonsmokers’ Rights. Smokefree Parks. Americans for Nonsmokers’ Rights, 2 Oct. 2017. PDF file. https://no-smoke.org/wp-content/uploads/pdf/SmokefreeParks.pdf
American Lung Association. State of Tobacco Control 2024: California Local Grades. 2024, The American Lung Association. https://www.lung.org/getmedia/f7dafd89-9007-43c6-9ba3-52a422213fee/2024-SOTC-California-Combined-v3.pdf
Centers for Disease Control and Prevention. “About Secondhand Smoke.” Smoking & Tobacco Use. U.S. Department of Health and Human Services, 15 May 2024, https://www.cdc.gov/tobacco/secondhand-smoke/index.html
City of San Jose v. Department of Health Services. California Court of Appeal, Sixth District, No. H016744. 18 Aug. 1998. Justia, https://law.justia.com/cases/california/court-of-appeal/4th/66/35.html