Data used for this work
The Alcohol and Smoking Toolkit Study, a monthly cross-sectional survey of adults has been running in England since 2006 and was extended to Scotland and Wales in 2020 as part of SPECTRUM, surveying over 2,000 people each month. The survey gathers data on smoking and alcohol consumption. Mental health was first included in the survey in 2016/17 but only for people who smoked or had recently stopped. UK Prevention Research Partnership funding for SPECTRUM enabled inclusion of mental health measures for all participants from April 2020 to June 2023. Participants stated whether since the age of 16, they had ever been diagnosed with a mental health condition (choosing all that applied from a list) and provided their level of past-month psychological distress, measured using the Kessler Psychological Distress Scale (K6) which can be categorised into none/minimal distress, moderate and severe distress. This addition to the rich data of the Alcohol and Smoking Toolkit Study allowed us to answer a wide range of research questions.
Increased distress over time and financial hardship
Combining all the data from 2020 to 2023, around 30% of respondents reported ever having had a diagnosis of a mental health condition, most commonly depression, anxiety (both about 20%) or post-traumatic stress disorder (5%). Looking at psychological distress, 23% fell into the moderate and 6% in the severe category.
Comparing 2016 and 2017 with 2020, we identified an increase in moderate and severe psychological distress among people who smoked or who had recently quit, suggesting widening inequalities in mental health. Additional data from April 2020 (a highly stressful time at the start of the COVID-19 pandemic) to December 2022 showed that severe distress in adults increased by 46% over that time to 8%. Worryingly, the increase was much more pronounced in some groups than others, in less advantaged social grades, among those aged 18 to 25 and among those smoking, one in five adults reported severe distress by the end of 2022.
We also identified an association between financial hardship and psychological distress. In 2023, around 1 in 8 adults found it hard to manage financially, and 1 in 15 reported rent or mortgage arrears. Adults facing financial difficulties were considerably more likely to report moderate or severe psychological distress, irrespective of other sociodemographic factors. Financial hardship was also much more common among people who smoked.
Smoking prevalence
We know that smoking prevalence is significantly higher among individuals with mental health conditions, and smoke-free goals (<5% prevalence) are likely to be reached around 15 years later in this group than in those without a mental health condition. However, very little research has been able to look at specific diagnoses. Including over 64,000 adults who responded from 2020 to 2023, we found that as expected, among those reporting any mental health diagnosis, currently smoking or having recently stopped was much more prevalent than among those with no diagnosis (28% compared to 15%). There was large variation between specific conditions, with over 50% past-year smoking among adults reporting a diagnosis of drug use or dependence, personality disorder or psychosis/schizophrenia. Having multiple mental health conditions was associated with high smoking rates, with 37% of those with two or more mental health conditions currently smoking and more likely to be smoking daily and more heavily. These estimates highlight the need for increased efforts to reduce smoking among people with mental health conditions.
Smoking cessation
People reporting almost any of the mental health diagnoses were at least as likely to attempt to stop smoking as those without a diagnosis (pooled data from 2020 to 2023. But success rates varied considerably across different mental health conditions and were mostly below those for people without a diagnosis; for example, among people reporting psychosis/schizophrenia, 13% of cessation attempts were successful, half the success rate among people without a diagnosis.
If people are making cessation attempts but are less likely to be successful, we need to make sure that each attempt is as likely to be successful as possible. We explored the effectiveness of different smoking cessation aids, and whether this is moderated by mental health. In line with previous evidence, using vaping products or the prescription medicine varenicline led to higher success rates whereas there was no clear evidence for a benefit for most other options. Encouragingly, the effectiveness of cessation aids was not significantly moderated by mental health conditions, supporting the use of the most effective options irrespective of mental health status
Continued nicotine use after smoking cessation
There is evidence that mental health symptoms improve after smoking cessation. However, we wanted to know how long this may take and if it differs between people who continue to use nicotine and those who do not, which may also provide information about the role of nicotine versus smoke toxicants in mental health.
People who had stopped smoking for at least 1 year had levels of psychological distress similar to people who had never smoked and much lower than people who smoked or had more recently stopped, so improvements may need some time, similar to improvements in physical health. Interestingly, continued use of nicotine products after quitting smoking, such as vaping, heated tobacco products, or nicotine replacement therapy, was linked to higher levels of distress than not using nicotine, although people with poorer mental health may be more likely to continue use rather than the continued use causing the poorer mental health.
Other questions
The inclusion of mental health measures meant that research not primarily about mental health could look at differential impacts. One paper for example estimated the effect of changes in regulation (in this case a ban on disposable vapes) and found this would disproportionately affect disadvantaged groups including those with mental health conditions.
Alcohol use and smoking
Often, alcohol use and smoking are researched separately, and the data we had enabled us to look at them together. The survey included a screener (the AUDIT questionnaire) to assess whether people drank no alcohol at all, drank alcohol in a way that presented low risk or in a way that presented increasing or higher risk.
Overall, 5% of adults smoked and drank at increasing or higher risk levels. Nearly half of this group had experienced psychological distress in the last month and 45% had ever been diagnosed with a mental health condition. Compared with adults who only smoked or drank at this level, the group doing both was more likely to have experienced distress and a higher proportion had ever had a diagnosis. There was also a close link between the two behaviours, people who smoked were much more likely to also drink at a riskier level and vice versa.
One way to help reduce smoking or drinking is to offer brief interventions at every opportunity, for example in primary care. Among people who smoked and had visited their GP in the last year, around one in three had received a brief intervention for smoking while for those who drank at a risker level, this was closer to one in twenty. People who reported a mental health diagnosis were more likely to have received an intervention for smoking or alcohol.
Conclusions
The SPECTRUM Work package 7 research highlights the interplay between mental health and unhealthy commodity use and the need to address these inequalities. For smoking, people with mental health problems are at least as likely to attempt to stop, support in the form of vapes or varenicline is effective for all people who smoke and mental health is better among people who have stopped smoking for at least a year. Accessible and effective support for people with mental health conditions who smoke or drink too much alcohol needs to be ensured.