Public stigma towards people with mental illness comprises the beliefs, attitudes and behaviour towards people perceived as having, or with a diagnosis of, a mental illness. We previously found evidence that public stigma in England worsened over the periods 1994–2003 Reference Mehta, Kassam, Leese, Butler and Thornicroft1 and 2019–2023. Reference Ronaldson and Henderson2 There was a significant reduction in public stigma between these periods, particularly in 2008 and 2019. Reference Henderson, Potts and Robinson3 The Time to Change stigma reduction programme Reference Potts and Henderson4 ran from 2009 to 2021, including a social marketing campaign and work with several target groups, and may have contributed to this improvement. Reference Evans-Lacko, Corker, Williams, Henderson and Thornicroft5
Other stigma measures from this period are available for comparison. Surveys of discrimination experienced by people using mental health services, undertaken between 2008 and 2014, showed a significant reduction overall, and in numerous areas of social life and employment. Reference Corker, Hamilton, Robinson, Cotney, Pinfold and Rose6 General population surveys measuring desire for social distance from people, described in vignettes with symptoms of depression and schizophrenia, also showed improvements between 2007 and 2021. Reference Ronaldson and Henderson2 This was maintained between 2021 and 2023, Reference Ronaldson and Henderson2 in contrast to the deterioration in public stigma measured with scales comprising items about people with mental illness in general.
This divergence suggests that interpersonal intentions towards known individuals should be distinguished from attitudes towards people with mental illness in general. To continue tracking general population stigma, we aimed to compare both aspects of stigma between 2023 and 2024.
Method
Data from the Attitude to Mental Illness (AMI) survey are available from 2008. The latest survey ran from 16 September to 10 November 2024. A quota-sampling frame was used to ensure a nationally representative sample of adults (16 years or older) living in England; respondents were not resampled in later surveys, and approximately 1700 respondents took part each time. Information about the data source, sampling methods and study measures are published elsewhere Reference Ronaldson and Henderson2 and in the Supplementary Methods. The King’s College London Psychiatry, Nursing and Midwifery Research Ethics Subcommittee exempted analysis of these data as secondary analysis of anonymised data.
Measures of stigma-related knowledge (Mental Health Knowledge Schedule (MAKS Reference Evans-Lacko, Little, Meltzer, Rose, Rhydderch and Henderson7 )), attitudes to mental illness (Community Attitudes toward the Mentally Ill (CAMI Reference Taylor and Dear8 )) and desire for social distance (Reported and Intended Behaviour Scale (RIBS-IB Reference Evans-Lacko, Rose, Little, Flach, Rhydderch and Henderson9 )) were included, as they have been since 2008 (CAMI) and 2009 (MAKS, RIBS-IB). For the second time, we included vignettes from the British Social Attitudes Surveys 10 depicting a common mental health problem (‘Stephen’: depression) and a less common problem (‘Andy’: schizophrenia). Participants were asked social distancing questions assessing their willingness to engage with the vignette character across six domains: living next door, socialising, forming a friendship, working as colleagues, accepting them as a family member by marriage and having them provide childcare.
In the current study, we compared 2024 outcomes with those from 2023 by estimating adjusted mean differences. These differences were then converted into changes in the proportion of individuals scoring above the 2023 mean, which served as the reference threshold. We used survey-weighted, multiple regression models to evaluate patterns of change in MAKS, CAMI (including CAMI subscales: ‘Prejudice and Exclusion’, ‘Tolerance and Support for Community Care’) and RIBS-IB scores. All models used the standardised scores of the measures as dependent variables, meaning that outputs were interpreted in standard deviation units. Logistic regression models were used to assess change in vignette responses (willing versus unwilling) between 2023 and 2024. Models were adjusted for sociodemographic variables: age, gender, ethnicity, socioeconomic position and government office region. Reference Ronaldson and Henderson2
Analyses were performed with Stata 18.0 (Stata Corp, College Station, Texas, USA).
Results
Detailed characteristics for the samples up to 2023 were provided previously. Reference Ronaldson and Henderson2 The 2024 sample comprised 1563 participants with the distribution of gender and age group remaining stable, whereas changes since the introduction of remote data collection in 2019 (i.e. more participants with professional/managerial occupations, greater familiarity with mental health problems) were sustained.
Survey-weighted descriptive statistics were calculated for all outcome measures for 2023 and 2024 (Supplementary Table 1). Figure 1 depicts change over time in total CAMI (plus subscales), MAKS and RIBS-IB scores. Previously we reported decreases in all scale scores between 2019 and 2023. Reference Ronaldson and Henderson2 Between 2023 and 2024, the proportion of respondents reaching 2023-level MAKS scores reduced significantly, by 3.5% (p = 0.028). The proportion of respondents reaching 2023-level total CAMI scores decreased by 7.0% (p < 0.001). Analysis of the CAMI subscales showed a 5.7% reduction in the proportion of respondents achieving 2023-level scores on the Prejudice and Exclusion subscale (p < 0.001), and a 6.9% reduction on the Tolerance and Support for Community Care subscale (p < 0.001). The proportion of respondents reaching 2023-level RIBS-IB scores declined by 1.9% but was not statistically significant (p = 0.202). Regression coefficients are presented in Supplementary Table 1.
Top, marginal estimates of stigma-related attitudes (CAMI), knowledge (MAKS) and desire for social distance (RIBS-IB) by year (confidence intervals). Bottom, marginal estimates for subscales of the CAMI scale. RIBS-IB, Reported and Intended Behaviour Scale.
Across vignettes, between 2023 and 2024 the proportion of people who were fairly or very unwilling to interact with ‘Andy’ (schizophrenia) increased across most scenarios, with the exception of working as colleagues and providing childcare. However, fully adjusted logistic regression models showed that none of the changes reached statistical significance. Fully adjusted models also revealed no statistically significant changes in unwillingness to interact with ‘Stephen’ (depression), although increases were observed in the proportion of people unwilling to interact across the following: socialising, forming a friendship, working as colleagues, accepting them as a family member by marriage and having them provide childcare. Descriptive statistics and results from regression models are provided in Supplementary Table 2.
Discussion
Further deterioration in public stigma occurred over the period 2023–2024 regarding people with mental illness/mental health problems in general, although desire for social distancing from named individuals, conveying familiarity, has not shown significant change. Negative stereotypes, leading to stigmatisation in the context of unequal power relationships between people with mental ill health and others, Reference Link, Yang, Phelan and Collins11 may be easier to apply to groups of unknown people rather than towards an individual described as if they were known. However, it should not be assumed there will be no future increase in interpersonal stigma, just as it decreased previously in line with public stigma Reference Henderson, Potts and Robinson3 and experiences of discrimination. Reference Corker, Hamilton, Robinson, Cotney, Pinfold and Rose6 Despite study limitations (see Supplementary Material), there are grounds for concern that people with mental ill health will, or may already, experience greater active discrimination and/or avoidance by others.
Several factors may be contributing to increasing public stigma. Rising levels of mental ill health and help-seeking among young people have been interpreted by some public figures as evidence that they are ‘work-shy’ and misusing the welfare system. Reference Weaver12 From a systems perspective, Reference Skivington, Matthews, Simpson, Craig, Baird and Blazeby13 greater help-seeking and rising levels of mental ill health are disruptive, leading to either greater investment in services or a backlash, reflecting economic and labour market circumstances and/or a failure to address structural stigma. The use of vignettes of people with common mental disorder that vary by age and employment status could be used to explore the intersectionality of these characteristics. Homicides by people with psychosis that receive intense media coverage and criticism of the mental health services responsible for the care of the perpetrator are likely to influence public stigma. Reference Corrigan, Powell and Michaels14,Reference Cummins15 However, given long delays in hospitalisation in parts of England, community members’ anxiety about unwell people in their neighbourhood is increasingly likely. We suggest that this should not be dismissed as public stigma: for example, instead of a process that starts with labelling, fear may then be followed by recognition that someone is unwell and thence a general desire for social distancing from people with such illnesses. However, within socially deprived communities, enacting social distancing may not be possible and there may be relatively little power differential. We therefore recommend qualitative exploration of the experience and views of people in such communities.
Supplementary material
The supplementary material is available online at https://doi.org/10.1192/bjo.2026.11050
Data availability
The data that support the findings of this study are available from the corresponding author, A.R., upon reasonable request.
Acknowledgements
We thank Kerry Goddard and Anita Fernandes at Mind for collaboration on the evaluation, and acknowledge the valuable contributions of George Hoare and Alex Viccars during their previous roles at the organisation.
Author contributions
A.R. conducted the data analysis and created the figure. A.R. and C.H. drafted and reviewed the manuscript.
Funding
The funding for the research following Time to Change was provided by Mind, as part of Exilarch’s Foundation funding of The Big Mental Report. The Time to Change evaluation was funded by the UK Government Department of Health and Social Care, Comic Relief and Big Lottery Fund. C.H. was supported by these grants during phases 1–3 of Time to Change. The funding source had no involvement in the study design, data or report writing.
Declaration of interest
C.H. has received consulting fees from Lundbeck and educational speaker fees from Janssen. A.R. declares no conflict of interest. Neither author participated in the planning or execution of Time to Change.
Open access
eLetters
No eLetters have been published for this article.