The New Childhood: Mental Health in a Connected Age

6.1 Dataset overview and definitions

The analyses in this report are based on anonymised service data from Shout, following data compliance and protection measures from MHI. The dataset contains records of conversations between texters and trained volunteers from March 2018 to December 2025 and was de-identified by MHI. Across this time over one million texters have been supported by Shout across almost three and a half million conversations with volunteers.

Two key units of analysis are used throughout the report:

Texters: Texters are defined by the distinct mobile number they use to contact the Shout service. While Shout is anonymous, each number is assigned a unique, de-identified ID to allow repeat contacts to be recognised without identifying individuals. This allows the number of texters to be separated for analyses from the total number of conversations.
Conversations: Conversations are defined as a continuous exchange of text messages between a service user and a Shout volunteer within a single time-bound interaction that begins when the texter contacts the service.

Texter-level analyses describe who is contacting the service where information is known. For example, when looking at the proportion of texters aged 13 and under in a given year, the denominator is the number of texters who provided age information in that year.

Conversation-level analyses describe how the service is used. For example, the proportion of conversations involving suicide-related discussions is based on the number of conversations where issue labels were recorded.

Because texters can contact the service multiple times, a relatively small proportion of texters may account for a larger share of conversations if they return frequently for support.

Unless otherwise stated, conversation-level analyses refer to the number of unique conversations, while texter-level analyses refer to the number of unique texters contacting the service.

6.2 Data availability and completeness

After each conversation, texters are invited to complete a post-conversation survey. This survey includes questions about age and several demographic characteristics (Box 4). Because completion of the survey is voluntary, the availability of demographic information varies across variables (Figure 11).

Age information is available for an estimated 32 per cent of conversations, meaning that all analyses for this report have been computed over the subset of conversations and texters where age information was provided unless otherwise stated (see Table 1).

Demographic variables including autism status, LGBTQ+ identity, and if the texter receive(d) free school meals, have been analysed at the texter level, and limited to texters who provided a response to the relevant survey question.

Time of day and issue labels have a complete dataset coverage and are available across all conversations.

Because data availability varies, we have provided a table (Table 1) within this appendix to disclose which subset of the dataset was used for each analysis.

6.3 Growth in Shout service use

Since its launch in March 2018, the number of conversations supported by Shout has increased significantly as awareness of the service has grown.

Within this growth, there has been an observable and anecdotal increase in young texters, with texters aged 13 and under growing from an estimated 4 per cent in 2018 to 9 per cent in 2025. Despite being less than 10 per cent of Shout’s texters, the proportion of conversations with those aged 13 and under sits between 11 per cent and 15 per cent, having peaked in 2022 at just under 100 conversations per day from this age group (Figure 13). This disproportionate share between texters and conversations indicates that children 13 and under have generated between 16 per cent and 34 per cent more conversations per texter than the Shout average, demonstrating both a high level of engagement in Shout and the need for support for this age group relative to their share of users.

6.4 Regional distribution of texter aged 13 and under

Regional analyses were conducted at the texter level with less than 19 per cent of conversations providing regional information.

Among texters who provided both age and regional information, an estimated one in 10 were aged 13 or under, matching the service wide estimations of 9 per cent (Figure 4).

Analysis found that use of Shout among this age group is distributed across all regions of the UK, with no single region accounting for a dominant share of texters aged 13 and under (Figure 14).

6.5 Support network availability

During a 60-month period from January 2018 to December 2020 and January 2022 to June 2024, the post-conversation survey included a question asking texters whether they had anyone else they could talk to about what they were experiencing.

Across this period, just over two million conversations occurred within the survey window, with responses provided across approximately one third of those conversations.

Among texters aged 13 and under who responded to the question an estimated 40 per cent reported that they had no one else to talk to. Similar patterns are observed across other age groups, with roughly one third to nearly half of respondents indicating that they lacked an alternative support network.

6.6 Conversation topic prevalence

Additional insight into why texters contact Shout can be obtained through issue labels assigned by volunteers for each conversation.

Shout volunteers can label conversations using four broad issue categories based on what was discussed:

Suicide
Self-harm
Bullying
Other issues
With an additional label of "No Issue Identified"

Conversations with the label “No Issue Identified” account for approximately 46 per cent of all conversations across Shout.

Analyses into issue and topic distribution use the subset of data that includes the labels of “Suicide”, “Self-harm”, “Bullying”, and “Other Issues”.

As the number and frequency of texters contacting Shout increases, priority is given to texters who contact Shout with severe or urgent need, such as those in crisis or who ask for support regarding suicidal ideation. However, the proportion of conversations related to suicidality across this young age group has remained consistently higher than the service-wide average and has continued to rise, even as the proportion of conversations on this topic has declined across all Shout conversations since 2023.

In 2025, of all conversations where an issue was identified, 35 per cent of conversations mentioned suicide-related topics. When looking specifically at conversations with texters aged 13 and under, this increases to 38 per cent. This indicates that children aged 13 and under make up a larger share of suicidality-related conversations than would be expected based on overall service demand. This suggests an acute need for support on this topic.

6.7 Analytical bases used in this report

To ensure transparency in the interpretation of findings, the table below summarises the analytic denominators used across the report. Because demographic information is collected through an optional post-conversation survey, different analyses rely on different subsets of the dataset depending on which variables were available.

Where analyses use filtered subsets of the dataset, we have noted this explicitly in the table below