Independent report

Evidence gaps table summaries (updated March 2023).

Updated 17 March 2023

Gaps still evident since the 2018 review of the literature

The relationship between loneliness and local communities

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
An exploration of how loneliness risk factors at a local community level. Related research questions: • What community level risk factors might influence loneliness (e.g. poor social cohesion, limited community assets like libraries and youth centres)? • What is the relationship between loneliness and these risk factors?	Longitudinal survey data from population studies that includes neighbourhood and community variables, as well as the loneliness question. Data sets large enough to facilitate subgroup analysis.	Partial - Some data are available (see next column). The harmonised ONS recommended loneliness measure needs to be included in more longitudinal surveys.	Loneliness - Understanding Society, the UK Household Longitudinal Study. This study includes loneliness data from 1995-2001 and then again from 2017. Community risk factors - A number of community and neighbourhood variables are included in Understanding Society too.

The relationship between loneliness and key outcomes

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
Longitudinal examination of the relationships between loneliness and key outcomes. Related research question: What is the relationship, over time, between loneliness and different outcomes, specifically: • health • education • employment • wellbeing	Longitudinal survey data from population studies that includes neighbourhood and community variables, as well as the loneliness question. Data sets large enough to facilitate subgroup analysis.	Partial - Some data are available (see next column). The harmonised ONS recommended loneliness measure needs to be included in more longitudinal surveys.	Loneliness - Understanding Society, the UK Household Longitudinal Study. This study includes loneliness data from 1995-2001 and then again from 2017. Key outcomes - Good coverage of key outcomes of interest in Understanding Society.

The prevalence of loneliness amongst younger children

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
Secondary analysis to explore the rates of loneliness amongst young people, particularly younger children. Related research question: • How prevalent is loneliness amongst children under 16 years of age?	Longitudinal loneliness survey data that enables trends analyses for youth younger than 16 years.	Partial - Some data available, but relatively unexploited.	Youth loneliness data - Active Lives Children and Adolescent Survey (2019 onwards, ages 11-16 years) and the People and Nature Survey (2020 onwards, child [8-15 years] and adult survey [16+ years]).

Loneliness over the life course

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
Exploration of loneliness over time using longitudinal surveys to understand what makes people lonely at different points in life and in different circumstances. Related research questions: • Over the life course what factors predict loneliness at different points? • What stops people being lonely at different points? • How do the number and strength of connections someone has predict loneliness over time?	Longitudinal loneliness survey data over the life course.	Partial - There is some survey data, but the harmonised ONS recommended loneliness measure needs to be included in more longitudinal surveys.	Longitudinal loneliness data - The Millennium Cohort Study data includes loneliness data at age 14, and again more recently. Understanding Society data will also be useful.
There is a gap in understanding the experience of loneliness in midlife (40-60 year olds). Related research questions: • How do experiences of loneliness vary by age, including midlife which has been less reported on?	Cross-sectional and longitudinal loneliness survey data.	Partial - There is a range of data available in different population surveys.	Possible population surveys include: •Longitudinal - Understanding Society Cross-sectional - Community Life Survey

The life course approach to loneliness does not apply just to the collection of survey data, but to other types of data collection (e.g., experimental) too. It is important to consider broadly.

Measurement

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
This is a limited understanding of how the ONS recommended measure is understood by different ages. Related research question: • How far is the ONS recommended measure for loneliness (comprising the UCLA 3-item scale and the 1 item direct measure) understood in the same way by people at different ages?	Data from ONS Loneliness 3-item scale and the 1 item direct measure for different age groups and other sub-groups such as ethnicity and gender to explore invariance for other individual differences.	Partial - Available from different population surveys.	ONS recommended loneliness measure. The ONS recommended measure has been included in the Community Life Survey since 2018. Data could also be collected from other studies that have included the ONS 3 item scale (e.g., National Travel Survey); There does not appear to be any youth surveys with the 3-item measure included. The ONS 1-item direct measure is available in a large number of population surveys commissioned by DCMS, including those for youth.
An exploration of different aspects of loneliness, including duration and intensity of loneliness are needed. Related research question: • How do experiences of loneliness vary, for example, in terms of duration and intensity?	Data on loneliness duration and intensity needs to be collected. Qualitative enquiry into the experience of loneliness is needed.	No, new data is needed - Data is not available in current datasets, with the exception of the BBC Loneliness Experiment, where it has previously been explored.	The BBC Loneliness Experiment is a good example.
Clarification of the distinction between transient and chronic loneliness and how they are understood. Related research question: • How are the impacts of transient and chronic loneliness different?	New data in the form of a systematic review, conceptual review and qualitative research.	No, new data is needed.	Understanding Society could be used to explore how long people felt lonely for, and how this impacts wellbeing/health outcomes (as covered in row 1). However, this is one interpretation of chronic loneliness, more research (in the form of a review) is needed.
There is a limited understanding in how far it is possible or appropriate for others in a person’s life to adequately report on loneliness. Related research question: • How far is it possible/appropriate for someone else (e.g. someone’s teacher, parent, co-workers) to report on loneliness on their behalf?	Quantitative and qualitative data that explores the links between self and other loneliness ratings.	No, new data is needed.

Stigma

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
Clarification of why being stigmatised increases loneliness is needed. Related research question: • What are the mechanisms that link stigma and loneliness?	Qualitative and quantitative insights on the stigma experienced by minority groups, and those who experience loneliness.	No, new data is needed.
There is a need to expand the loneliness evidence base for groups at risk of marginalisation. Some characteristics, for example visual impairment, are not collected in survey data. Other characteristics of interest are collected, like ethnicity, but sample sizes are sometimes small and lack statistical power. Related research question: • How do those with protected characteristics and most at risk of exclusion from society experience loneliness? Areas of interest include: • Ethnicity • Gender • Different disabilities • Sexual orientation orientation • Class	The ONS loneliness measure (cross-sectional or longitudinal) surveys with sufficiently large samples to facilitate analysis at subgroup level with statistical power.	Partial - there is data available in different population surveys, and data from different survey years could be combined to boost sample sizes. However, further survey data would be beneficial.
Questioning and improving our understanding of social skills and social skills-based interventions is needed. Related research question: • How do social skills and social-skills based interventions reduce loneliness?	A review of the social skills interventions including a description of social skills focused on, and how they have reduced loneliness.	No, new data needed - A review of social skills interventions, including qualitative research with programme participants/staff to explore reductions in loneliness.
Examination of loneliness resulting from stigmatisation of a family member or friend is needed. Related research question: • What feelings of loneliness are experienced by family members or friends of those who are stigmatised by others, including wider society?	Data from qualitative research or quantitative surveys.	No, new data is needed.
Improvement in our understanding of structural factors and loneliness. Related research question: • What structural factors cause loneliness for marginalised groups (for example through laws)?	Explore structural factors that impact loneliness through qualitative research.	No, new data is needed.
Policy makers and researchers should consider how to ensure interventions do not increase the stigma associated with loneliness. Related research questions: • How do you ensure that interventions designed to combat loneliness do not increase stigma associated with loneliness? • What are the consequences of stigma - how far does it make people more lonely?	Recommendation to researchers and policy-makers; Exploration through qualitative research.	No, new data is needed.

Cultural background

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
There is a limited understanding about the relationship between cultural background and loneliness. Related research question(s): • What aspects of cultural background need to be understood in relation to loneliness? • Do we need to broaden our understanding of ‘cultural background’? • What is the relationship between cultural background and loneliness?	Qualitative research, with people from different cultural backgrounds, exploring how different aspects of cultural backgrounds relate to loneliness. A review of the literature to determine the aspects of culture that might link to loneliness. Loneliness data from people from different cultures living within England needs to be collected.	No, new data is needed - Specific aspects of cultural background, beyond ethnicity, are not collected as part of population surveys.
An improved understanding of how different cultural backgrounds impact loneliness and the effects is needed. Related research questions: • What are the intersections between cultural backgrounds and other demographic groups? • How do cultural backgrounds influence loneliness in relation to other things, like gender, age and community interaction? • E.g. How do different cultural attitudes and views of gender shape experiences of loneliness amongst those from different genders?	Loneliness data from people from diverse cultural backgrounds living in England needs to be collected and consequences need to be explored.	No, new data is needed - Specific aspects of cultural background are not collected as part of population surveys.

Mental health

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
Exploration of how those with mental health challenges experience loneliness is needed, particularly in relation to groups at risk of marginalisation. Related research questions: • How do those with diagnosed mental health problems experience loneliness? • How does this vary by different types and severity of mental health challenges? • What are the experiences of loneliness amongst vulnerable groups with mental health problems? For example: • children in the care system • asylum seekers	Loneliness survey data needed from specific groups, including those with existing mental health problems. Qualitative exploration of experiences of loneliness amongst those with mental health problems.	Partial - Mental health data are routinely collected (see next column). But new data is also needed from specific groups identified, such as young carers.	Understanding Society Longitudinal Study. Includes data from 1995-2001 on loneliness and then again from 2017. LSYPE2 is a longitudinal study that includes loneliness questions included in later waves (8 and 9).
Research suggests there is a bidirectional relationship between loneliness and anxiety or depression. However examination of whether there is a bidirectional relationship between loneliness and other mental health diagnoses is required. Related research question: • Is there a bidirectional relationship between loneliness and the following: schizophrenia, bipolar, eating disorders, autism, and traits suggestive of a diagnosis of ‘personality disorder’ (complex emotional needs)?	Longitudinal survey data on loneliness and mental health diagnoses.	Partial - Mental health data routinely collected in Understanding Society Longitudinal Study	Understanding Society Longitudinal Study. Includes data from 1995-2001 on loneliness and then again from 2017.
There is a lack of adequately powered trials of a broad range of psychological and social interventions.	Interventions specifically focused on those with mental ill-health that are adequately powered to detect any effects.	No, new data is needed - New data that includes development and robust evaluation of interventions specifically focused on those with mental ill-health who are experiencing loneliness.

Place

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
An exploration of the mechanisms that underlie the relationship between loneliness and place is needed. Related research questions: • What is the relationship between place and loneliness - how does place impact loneliness? • How does engagement with local community groups (e.g. local arts groups) influence loneliness? • How can social impacts of infrastructure such as libraries and leisure be maximised to tackle loneliness?	Quantitative data on transport, green spaces, etc. needs to be collected. To explore the mechanisms, data needs to be longitudinal.	Partial - cross-sectional data available on green-spaces, transport, etc. To explore mechanisms needs longitudinal data. Useful data is available in People and Nature Survey.	People and Nature Survey, which is longitudinal and includes many pieces of information relevant to place. Taking Part Survey includes engagement with museums, galleries, libraries; National Travel Survey includes transportation data.

Workplace

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
An exploration of individual and organisational determinants of loneliness in the workplace is needed. Related research question: • What determines loneliness in the workplace, at an individual and organisational level?	Loneliness data needed from people in the workplace; exploration, likely using qualitative research. to explore organisational factors that affect loneliness.	No, new data is needed - Data on specific aspects of organisations are not available in population surveys.	Understanding Society Longitudinal Study. Includes data from 1995-2001 on loneliness and then again from 2017. Includes information on the transition from education to employment. The workplace wellbeing questionnaire including benchmarkable relationship measures could be used.

A study from What Works Wellbeing using longitudinal data from the UK Understanding Society Survey (2009-2014), finds that young people with lower levels of life satisfaction are less likely to go on to find a job. The strength of the effect is determined by personality, such that the negative impact of low life satisfaction is only significant for introverts and not extraverts.

Economics

Evidence gap overview	Data needs	New data needed or data already available?	If data is available, where can we find those?
An exploration of the ability of quality-of-life instruments (Quality Adjusted Life Years - QALY) and wellbeing instruments (Wellbeing Year - WELLBY) to assess the impact of loneliness in the lives of individuals.	Need to adapt existing quality of life instruments, incorporating QALYs and WELLBYs when appropriate.	No, new data is needed.
An exploration and Identification of all key relevant economic aspects of loneliness that have an impact on people’s lives.	Review of existing evidence on how loneliness impacts people’s lives, with exploration across the life course. The health sector could look at the economic cost of loneliness on the NHS.	No, new data is needed - could loneliness data be linked with health and education data?

Recommendations

1. Encourage robust and high-quality evaluations of loneliness interventions in the UK.
   
   • A measurement bank already exists and can be used to inform researchers

2. Looking beyond the immediate impact of interventions — consider the long term.

3. Exploration of the effectiveness of interventions designed for different subgroups.

4. Economic analysis needs to be embedded into robust longitudinal trials looking at the effectiveness of interventions over time.

Longitudinal quantitative data

There is an overarching need for more longitudinal loneliness data, particularly the ONS recommended loneliness measure. However, it is important to keep in mind that other loneliness questions might be possible to compare with the ONS loneliness measure. This table outlines survey data referenced in the evidence gaps tables and gives advice on the comparability.

Survey	Loneliness question(s)[footnote 1]	Comparability with ONS recommended loneliness measure
Understanding Society	Understanding Society asks the complete indirect measure in the harmonised question set, however their direct measure uses different response options.	Near comparable
Millenium Cohort Study	The Millennium Cohort Study in 2017 asked whether participants felt lonely within the past two weeks.	Not comparable
LYPSE2 (?)	Specific questions unknown	Comparability unknown
Community Life Survey	Harmonised direct measure	Near comparable
Taking Part Survey	Harmonised direct measure	Near comparable

1. Refers to current survey waves ↩