Cigarette smoking increases the risk of developing lung cancer and other diseases, but how effective are public health strategies?
Professor Theodore Holford at the Yale School of Public Health, USA, is part of the CISNET Lung Cancer Working Group. His work uses population health models, such as the Smoking History Generator, to show age-period-cohort effects on smoking rates across US states.
Read more in Research Features
Read the original research: doi.org/10.1016/j.amepre.2022.08.018
Image Source: Adobe Stock Images / Lainen
Transcript:
Hello and welcome to Research Pod! Thank you for listening and joining us today.
In this episode, we look at the work of Professor Theodore Holford from the Yale School of Public Health, USA, as a part of the CISNET Lung Cancer Working Group. Holford and his team aim to improve our understanding about smoking behaviours across US states by discovering the changes in smoking patterns over time.
Public health strategies are designed to encourage people to stay healthy, reduce the risk of disease, and improve overall health and wellbeing. But how can we predict whether these strategies will be helpful?
A population model can be useful to look at how factors that affect health can help guide policy changes. Population models look at individual-level responses, such as use of hospital services, and scale them up to predict what this might look like at a wider population level – such as demand on health services. Understanding which areas impact public health most can ensure policies are directed for maximum benefit. One such programme is The Cancer Intervention and Surveillance Modelling Network, or CISNET, in the USA, which has developed cancer models.
The World Health Organization reports that lung cancer is the leading cause of cancer-related deaths worldwide. Smoking is the leading cause of lung cancer, accounting for around 85% of all cases and causing approximately 1.8 million deaths each year. Despite tobacco control measures, smoking remains a significant public health problem. Globally, the total economic cost of smoking-attributable disease and mortality is estimated at US$1436 billion, equivalent to 1.8% of the world’s total GDP.
Professor Theodore Holford, at the Yale School of Public Health, belongs to the Lung Cancer Working Group of CISNET. He explains how the Smoking History Generator is one type of model used to learn more about the links between cigarette smoking and the chances of having lung cancer. Other models consider ways to implement lung cancer screening and new methods for treating lung cancer.
The Smoking History Generator allows people to look up patterns of smoking histories of different aged populations across different states. This may include initiation of smoking, how many packs/cigarettes are smoked per day, or for how long they have smoked. This can then be cross-referenced with the dates of policy changes to explore their impact. For example, did changing access to cigarettes have an impact on how many people started smoking, or have smoking rates in women decreased since the launch of a concerted public health campaign?
The CISNET Lung Cancer Working Group has already examined how published reports, such as the Surgeon General’s Report on smoking and health in 1964, impact smoking behaviour and mortality rates attributed to lung cancer. Holford’s previous work has assessed the impact of the report on deaths from all causes. His staggering findings estimated that tobacco control was associated with the avoidance of 8 million premature deaths, as well as living 19–20 years longer between 1964 and 2014. While many public health efforts have focussed on reducing cigarette smoking, more work is needed to reduce the burden of ill health and mortality associated with smoking.
The Working Group have evaluated the impact of other changes in tobacco control and usage, concluding that almost 800,000 US lung cancer deaths were avoided between 1975-2000 following changes in smoking behaviours through the 1950s.
While the US federal government sometimes enforces policies affecting cigarette availability, state-level policies vary greatly. A smoking history generator that accounts for the huge differences in smoking behaviours across different states can help to quantify the effects of health policies.
A recent study published by Holford and colleagues uses an extension of the Smoking History Generator to estimate the impacts of policies affecting cigarette smoking, such as access to cigarettes following a change in the legal age for purchasing cigarettes.
They used an age-period-cohort modelling framework to reconstruct the smoking habits of the populations of the different states. Data from the previous health surveys was mapped across each state, including smoking initiation, cessation, and intensity data. The researchers looked at this data for different ages and genders across states and reported state-specific trends in smoking habits.
Their findings highlighted significant differences between states and smoking policies. One example is California, where more aggressive tobacco control has resulted in fewer people starting to smoke cigarettes, and more people stopping cigarette use. Other states with less strict policies, such as Kentucky, had higher cigarette smoking rates.
Public health policies are not the only factors affecting smoking behaviours; other influencing features include differences in demographics and rurality between states, as well as social and cultural differences between states.
There is huge variation across states in cigarette smoking behaviours, with some states showing improvements with more recent cohorts. States which have successfully changed smoking rates and decreased lung cancer rates may provide good examples of policy changes which could be adopted by other states. While 28 states do have smoke-free air laws in place, 38% of the US population across the remaining states have no protection from smoking. The smoking behaviour profiles also help understand and predict health consequences expected to result from each state’s smoking history. This variation highlights the importance of state-level estimates of smoking behaviour, such as those provided by Holford and colleagues.
Primary prevention, such as tobacco control measures and reducing exposure to environmental risk factors, such as second-hand cigarette smoke, can reduce the incidence of lung cancer and save millions of lives.
ResearchPod was privileged to ask the research team about the inspiration behind the research conducted by CISNET, to which Holford answered:
‘Cigarette smoking has had a catastrophic public health effect worldwide due to its association with lung cancer and other smoking-related diseases. It’s important to develop strategies to control it. Modelling is an effective way to study the potential impact of a control strategy, which motivated me to work in this field. The age-period-cohort model for disease incidence and mortality has been an area of research that I’ve worked in for many years, and it offers a useful statistical tool for the study of cigarette smoking. Many health behaviours, like cigarette smoking, are taken up at a particular age, and the effect develops over a lifetime. These patterns are often influenced by others in an individual’s birth cohort; hence, we observe cohort effects. Similarly, other factors affect everyone at a particular time, such as a law banning smoking in restaurants, an effect that might be associated with a certain time period. Hence, this approach has offered a powerful way to study these associations.’
Holford went on to outline the history of the most successful tobacco control policy in the US:
‘The turning point for cigarette smoking in the US came with the publication of the report by the Surgeon General on the effects of smoking on health in 1964, as well as the epidemiological research being conducted around that time. This knowledge inspired some to change their smoking behaviour, but it also led to efforts to control tobacco use. States like California have enacted policies designed to reduce exposure to cigarette smoke. Effective approaches for this have included taxes on cigarette purchases, clean air laws that ban smoking in certain areas, limitations on ages when cigarettes can be legally purchased, and the promotion of programmes that help smokers to quit. In many tobacco-growing states like Kentucky, essentially no efforts have been made to control the use of cigarettes. By studying the trends among the states, one can quantify the effects that have been realised. This can also help health departments in the different states to learn approaches that can be most beneficial to their citizens.’
And finally, Holford concludes that the Smoking History Generator can also be used as an exciting, useful tool for effectively:
‘Modelling the health effects of cigarette smoking and it has been extensively used in analysing lung cancer mortality and all-cause mortality. In addition, the approach has been modified in analyses by health departments in other countries. There are other diseases affected by cigarette smoking, and future modelling efforts can also make use of the Smoking History Generator.’
That’s all for this episode – thanks for listening, and stay subscribed to Research Pod for more of the latest science.
See you again soon.
Leave a Reply