Horizon CDT Research Highlights

Research Highlights

Exploring Loneliness and Social Isolation Elicitation Through Proxy Data Streams

  Gregor Milligan (2021 cohort)

Loneliness is defined as the subjective perception of a discrepancy between the desired and the true social relationships in terms of companionship, connectivity, or intimacy (Hawkley and Cacioppo, 2010), social isolation is a more objective measurement of one’s social network and interactions (Shiovitz-Ezra and Ayalon, 2012).
There has been a recent increase in acknowledging the significance of loneliness across the UK. A significant instigator of this was ’The Loneliness Report’ by The Jo Cox Commission on Loneliness, showing that over 9 Million people in the UK often or always feel lonely; this report also led to the appointment of a UK minister for loneliness in 2018 (ONS 2018)

This PhD will incorporate insights from several data sources to investigate the factors that impact propensity towards, elicitation, and intervention of loneliness. This PhD will apply methodologies from different disciplines, providing a comprehensive overview of the psychological theories of loneliness, loneliness measurements and
data science methods to evaluate the seldom used data generated on digital mental health interventions to understand how digital interventions are used to tackle loneliness.

This PhD aims to provide insight into the nature of loneliness, specifically in the 18-25 year old age range. Evaluating the factors that co-occur with loneliness such as deprivation, and how loneliness is elicited in digital mental health interventions. These insights will enable a further understanding of loneliness in this age group and inform practitioners within digital mental health practice through a series of recommendations for clinicians and practitioners to tackle loneliness. Through a two-pronged approach, the first part of this research evaluates how loneliness co-occurs with socio-demographic characteristics such as deprivation to understand the factors that may increase an individual’s propensity towards social isolation. Then explore loneliness in the context of digital mental health interventions to establish the occurrence of loneliness among young people, its relationship to other mental health conditions and how loneliness is being discussed on these platforms.

Research Questions and Objectives:

1) How do young adults experience loneliness across the UK?

Objectives to answer question 1:

  • Understand how loneliness co-occurs with deprivation indicators and mental health conditions
  • Explore the rates of loneliness in a representative sample of young adults in the UK and compare this to rates of loneliness of those who engage with digital mental health interventions

2) How can topic modelling approaches be applied to digital mental health text data to understand the elicitation of loneliness?

Objectives to answer question 2:

  • Explore how topic modelling approaches can be used to identify the elicitation of loneliness.
  • To Understand how the identified elicitations of loneliness can be used to assist practitioners in the provision of digital mental health platforms.
  • Define the ethical challenges, considerations and concerns regarding the accessing of public data from digital mental health services


Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness Matters: A Theoretical and Empirical Review of Consequences and Mechanisms. Annals of Behavioral Medicine, 40 (2), 218–227. https://doi.org/10.1007/s12160-010-9210-8

Shiovitz-Ezra, S., & Ayalon, L. (2012). Use of Direct Versus Indirect Approaches to Measure Loneliness in Later Life. Research on Aging, 34 (5), 572–591. https://doi.org/10.1177/0164027511423258

Office for National Statistics (2018, October). A connected society: A strategy for tackling
loneliness (Policy Paper). Department for Digital, Culture, Media and Sport. https://assets.publishing.service.gov.uk/media/5fb66cf98fa8f54aafb3c333/6.4882_DCMS_Loneliness_Strategy_web_Update_V2.pdf


Milligan, G., Bernard, A., Dowthwaite, L., Perez Vallejos, E. and Goulding, J. (2023) “Generating a Single Session Outcome Measure from Digital Mental Health Platform Footprints Using Natural Language Processing”, International Journal of Population Data Science, 8(3). doi: 10.23889/ijpds.v8i3.2269.