Loneliness among people with 'Complex Emotional Needs' (CEN): A cross-sectional UK study

The objective is to use the Adult Psychiatric Morbidity Survey (APMS) 2014 dataset for research purposes. Specifically, a secondary analysis of the data will be conducted by University College London (UCL) researchers to investigate the relationship between "complex emotional needs" (CEN, or "personality disorder" traits) and loneliness and suicidality outcomes, and the relationship of loneliness to discrimination, self-harm, and suicidal ideation in people with CEN.

Background:

Considering the prevalence and impact of loneliness on both physical and mental health outcomes, recovery, and quality of life, loneliness has been identified as an intervention target with the potential to alleviate symptoms and improve outcomes for people with mental health conditions such as depression, ‘complex emotional needs’ (CEN), and psychosis. The term ‘complex emotional needs’ (CEN) is used to describe people who are diagnosable with a “personality disorder”, who are a neglected group of great research and clinical interest. This is particularly important as the quality of care and available interventions for people with CEN have been criticized as lacking by service users and clinicians.

The research evidence has demonstrated the centrality of loneliness in the day-to-day lives of people with CEN. A quantitative study describing the intensity of loneliness in the lives of people with ‘CEN’ found that poor social functioning and objective social isolation did not account for the severity of loneliness experienced by people with CEN. Not only do these findings illustrate the need to further investigate determinants of loneliness and the effects of loneliness among people with CEN, but they also highlight the probable role of negative and discriminatory societal experiences in exacerbating a sense of loneliness. Recent sociological and psychological theories and studies exploring the relationship between suicidal thoughts and self-harm and loneliness report that a lack of belonging is one of the prominent risk factors associated with suicide and self-harm. Collectively, the evidence points to the need to explore the role of self-harm, suicidal ideation as well as discriminatory experiences and their consequent effects on loneliness outcomes among people with CEN. Although there is a relationship between loneliness and CEN, there is very little known about the relationship between individual CEN traits/symptoms and loneliness outcomes, and whether specific symptoms of CEN are more relevant in relation to loneliness. Given the extent to which interventions focus on symptomatic reductions, investigating whether self-harm and suicidal ideation exacerbate levels of loneliness could provide a pathway for future interventions to target the reduction of loneliness.

Aims and objectives:

In this study, UCL aim to conduct a quantitative study to investigate the relationship between CEN and loneliness and suicidality outcomes, and the relationship of loneliness to discrimination, self-harm, and suicidal ideation in people with CEN. Moreover, UCL will also assess the effects of relevant demographic factors in the relationship between discrimination and loneliness outcomes. UCL will use the APMS 2014 database to explore loneliness in this group and therefore the objectives are to:

- investigate the relationship between the number of traits endorsed that are suggestive of complex emotional needs (CEN), and the presence of loneliness controlling for stressful life events, including serious injuries/assaults, deaths, financial/job problems, prison time, bullying, violence, sexual abuse, homelessness, being institutionalized or in foster care, and other sociodemographic variables.

- Investigate the relationship between the number of traits endorsed that are suggestive of CEN and the presence of suicidality and self-harm, controlling for stressful life events including serious injuries/assaults, deaths, financial/job problems, prison time, bullying, violence, sexual abuse, homelessness, being institutionalized or in foster care and other sociodemographic variables.

- Assess whether loneliness modifies the relationship between the number of CEN traits and suicidality/self-harm by testing interaction effects.

- investigate the association between individual CEN traits and loneliness, controlling for covariates.

- assess the possibility that experiences of discrimination modify the relationship between specific traits suggestive of CEN and loneliness by testing interaction effects.

- investigate the interaction effect between discrimination and sexual orientation, and loneliness (outcome), among people who meet the cut-off for a diagnosis of CEN.

UCL is requesting the Adult Psychiatric Morbidity Survey 2014 (APMS 2014). UK data service (UKDS) (https://ukdataservice.ac.uk/), holding APMS 2014 on the behalf of NHS England and are responsible for disseminating under the direction of NHS England, would provide the whole data set to UCL as there is no facility to select individual variables. Upon signing the data sharing agreement, UCL would then download the pseudonymised APMS data through UKDS for the period specified within the DSA. All APMS data are pseudonymised. In line with the procedures and standards, when the DSA expires, all local copies of the APMS 2014 dataset will be erased and destroyed. The PI will be dedicating a period of one year to work with the dataset, carry out secondary analysis, and write up results.

The data will be held in the UCL Data Safe Haven using UCL-approved computers. The Data Safe Haven is a highly confidential technical solution for transferring and storing data. It meets the requirements for the NHS England governance toolkit and ISO 27001 Information Security Standards.

The aim of this secondary analysis of cross-sectional data from the 2014 Adult Psychiatric Morbidity Survey (APMS) is to investigate the relationship between loneliness and “complex emotional needs” (or “personality disorder”) and suicidality outcomes in a representative sample in England. The APMS 2014, the most recent survey adult psychiatric morbidity survey, provides relevant data on “personality disorder”, loneliness, discrimination, and suicidality for a nationally representative sample of residents in private households in England, aged 16 and over. Therefore UCL requires the full database of this large nationally representative sample to achieve a sufficient sample size to obtain meaningful results for this research. The data obtained will be anonymous and stored and processed only on UCL’s Data Safe Haven. The data will only be accessed by UCL employees and registered UCL Ph.D. students and registered UCL MSc students, as required or necessary. All students sign up to the UCL Academic manual and will be working under appropriate supervision on behalf of the data controller/processor within this agreement , they will have access to the data and only for the purposes described in this agreement.

The lawful basis for processing personal data under the UK GDPR is:

Article 6(1)(e) - processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller

The lawful basis for processing special category data under the UK GDPR is:

Article 9(2)(j) - processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject.

Given the adverse effects of loneliness, with severe levels of loneliness predicting suicidal and self-injurious behaviours in people with CEN, and the prevalence of CEN in the general population, it is of public interest to conduct research on loneliness among people with CEN traits. This research would potentially build the groundwork for intervention development. Specifically, investigating whether self-harm and suicidal ideation exacerbate levels of loneliness could provide a pathway for future interventions to both target the reduction of loneliness, therefore increasing personal recovery outcomes, as well as reducing self-harm and suicidal ideation.

UCL ensures that the processing of data abides by means of ‘appropriate technical and organisational measures which includes pseudonymising data/measures.

This research is undertaken as a part of a Ph.D. research and the Ph.D. student conducting this research is sponsored by UCL and the Economic and Social Research Council (ESRC), however, the ESRC will not be involved in the processing of data by the APMS 2014.

There is no control group in this study as the APMS consists of participants (aged over 16 years) who completed the APMS survey.

Once the agreement was active, the flow of data was from the UK data service (UKDS) which allowed access for UCL to download APMS data. There were no other flows of data. Data will only be stored, processed, and held in accordance with UCL’s data protection policy. It will be accessed, held, and stored in the UCL Safe Haven, within the Division of Psychiatry, by the research team. The research team consists of a UCL Ph.D. student, a MSc student (registered solely at UCL), and senior researchers and clinicians who are employees of UCL. The UCL Data Safe Haven has its own set of accepted and standard procedures that is described on the following website: https://www.ucl.ac.uk/isd/services/file-storage-sharing/data-safe-haven-dsh. Data will be stored within the Data Safe Haven, which is characterized by: dual-factor authentication, a firewall that has a default deny policy, data enter via a managed file transfer mechanism and only the asset owner has permission by default to draw down any data. All those analysing data within the team are required to complete the Data Security Training as provided by Health Education England.

All APMS participant data will be included in the data analysis and research study to achieve meaningful results. UCL will securely destroy all local copies of the dataset once the DSA expires and will inform the Data Access Request Service (DARS) as required by standard procedures. This 2014 version of the APMS dataset available via DARS has been redacted based on Disclosure Procedure advice to minimize the likelihood of participants being identified.

Methodology

Sample

UCL is conducting a secondary analysis of cross-sectional data from the 2014 Adult Psychiatric Morbidity Survey (APMS), a survey commissioned by NHS England that is carried out by the National Centre for Social Research (NatCen). The APMS provides data on mental health and treatment access for a nationally representative sample of residents in private households in England, aged 16 and over. The APMS uses a stratified probability sampling design that consisted of two stages, the initial stage is an interview with the whole sample and the second stage involves clinically trained interviewers conducting face-to-face interviews with a subset of participants.

Measures

The following clinical measures and sociodemographic characteristics collected in the APMS 2014 survey will be used:

Exposure variable:

- Personality disorder traits: personality disorder traits were measured using the Standard Assessment of Personality- Abbreviated Scale (SAPAS). SAPAS is an eight-item screen for identifying a possible diagnosis of “personality disorder”. The eight items are the following: 1) difficulty making and keeping friends, 2) identifying self as a loner, 3) difficulty trusting others, 4) tendency to lose temper, 5) tendency of being impulsive, 6) tendency to worry, 7) tendency to depend on others, and 8) tendency to be a perfectionist. The SAPAS has been validated for use in the psychiatric population and general population. SAPAS scores correlate with nurses’ ratings of externalizing behaviours, and future functioning and response to antidepressant treatment. A cut-off point of 4 on the SAPAS indicates a high probability of a diagnosis of a “personality disorder”.

Outcomes

- Loneliness: Loneliness is assessed using one item from the eight-item Social Functioning Questionnaire, a valid and robust measure to assess social functioning. Participants were asked to what degree they agreed with the statement: “‘I feel lonely and isolated from other people’’, over the past two weeks. Participants indicated their agreement on a four-point Likert scale, ranging from ‘not at all’ (scored 0) to ‘almost all the time (scored 3)’, generating scores ranging from 0 to 3. To transform these scores and avoid including zero values, one will be added to each integer.

- Self-harm, suicidal ideation, and suicidal attempt:

- Self-harm: Questions on self-harm were asked in both the self-completion section and the face-to-face interview. Information about self-harm was obtained through questions: 1) Have you deliberately harmed yourself in any way but not with the intention of killing yourself? A variable will be created on self-harm in the past year, derived from either the self-completion section or the face-to-face section.

- Suicidal ideation: In the face-to-face interview questions on suicidal ideation was included and participants were asked: 1) Have you ever thought of taking your life, even if you would not really do it? An affirmative response would be followed by the following questions on when this last occurred. A variable will be created on suicidal thoughts in the past year. This approach has been used by other studies.

- Suicide attempts: Questions on suicide attempts were asked in both the self-completion section and the face-to-face interview. Data on suicide attempts was obtained by the question: Have you ever made an attempt to take your life, by taking an overdose of tablets or in some other way? A follow-up question on when this took place was asked. A variable will be created combining reports of suicidal attempts in the past year, in either section (i.e. the face-to-face or the self-completion section)

Potential modifiers of the association between the exposure and outcome:

- Discrimination

o The APMS includes binary measures on experiences of discrimination, based on a computer-assisted self-interview. Participants were asked a series of questions on whether people have been treated unfairly in the past year on the basis of belonging to a particular group such as their mental health condition, ethnicity, sexuality, sex, age, religious beliefs, and physical health problems or disabilities. UCL will be focusing on discrimination based on sexuality and mental health condition.

Covariates

The following sociodemographic characteristics collected and recorded in the APMS 2014 survey are covariates that might be associated with both the exposure and the outcome and therefore confound any association. Sociodemographic variables such as gender, age, Registrar General's Social Class, and marital status are all measured and recorded and included in a set of covariates and UCL will agree on these as a team.

- Gender

- Age

- Socioeconomic status

- Marital status

- Social participation

- Stressful life events

- Childhood emotional, physical, and sexual violence or abuse

Analysis plan

UCL will conduct all analyses using STATA (a statistical software) and present descriptive statistics, sociodemographic and clinical characteristics, by exposure group, and will divide the exposure group into three categories for the descriptive statistics, those scoring 0, those scoring 1-3, and those scoring a 4 or more (meeting the criteria for a probable diagnosis).

The relationship between the number of traits suggestive of CEN (continuous measure) and loneliness (binary measure) will be investigated using logistic regression models. UCL will then conduct adjusted models for the confounders described, adding blocks sequentially. Similarly, logistic regression models will be run to produce odds ratios for outcome variables, suicidality and self-harm, UCL will then assess if loneliness modifies this relationship between CEN and suicidality and self-harm by testing the interaction between the number of CEN traits endorsed and loneliness.

To assess whether experiences of discrimination modify the relationship between specific traits suggestive of CEN and loneliness, UCL will be testing for interaction effects between each trait of CEN and the modifier. UCL will carry out a separate statistical analysis to investigate the interaction effect between discrimination and sexual orientation, and loneliness outcomes.

UCL will apply appropriate survey weightings to the data, using the relevant ‘survey’ (svy) commands in Stata 15, which allow for the use of clustered data modified by probability weights and provide robust estimates of variance. This weighting will take account of the complex survey design and of non-response to ensure that estimates are representative of the household population in England.

No linkage to other data will be conducted. No linkages are requested at this time. There will be no attempt to re-identify any participants

The results of this quantitative study will be published in a relevant peer-reviewed journal specialising in psychiatry or personality disorders such as BMC Psychiatry or Journal of Personality Disorder. A jargon-free and accessible blog describing this study in layperson's terms will also be published on the Mental Elf website as a blog for easy access and on other relevant websites such as the Loneliness and Social Isolation Mental Health Network website. This study will also be submitted for a Ph.D. Thesis and presented in research meetings and therefore will be uploaded and freely available in https://discovery.ucl.ac.uk/ upon completion of the Ph.D. Outputs, included in the published papers, will include aggregated level data with small numbers supressed. The date the researcher aims to publish the paper is October 2025. The blog and other mediums of dissemination will be published after the journal accept the manuscript (i.e. December 2025). The results of this research study will be presented at relevant academic conferences with an interest in the link between complex emotional needs and loneliness and suicidality. To increase public awareness the results may also be posted on Twitter. The study results will also be shared in the Loneliness and Social Isolation Research Network Newsletter and other mental health and loneliness-related newsletters.

In order to protect patient confidentiality in publications resulting from analysis of APMS 2014 data users must:

1. Guarantee that any outputs made available to anyone other than those with whom this agreement is made, will meet required standards, including the guarantee, methods, and standards contained in the Code of Practice for Official Statistics (https://www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html) and the Office for National Statistics (ONS) Statistical Disclosure Control (https://analysisfunction.civilservice.gov.uk/statistics/methodology-2/statistical-disclosure-control) from tables produced from surveys.

2. Apply methods and standards specified in the Microdata handling and Security Guide to Good Practice (https://ukdataservice.ac.uk/app/uploads/cd171-researchdatahandling.pdf) for disclosure control for statistical outputs.

The results will be disseminated to mental health charities with a focus on targeting loneliness and supporting people with ‘complex emotional needs’ such as befriending networks, and charities such as The British and Irish Group for the Study of Personality Disorder (BIGSPD). This piece of research forms one component of a Ph.D. thesis focused on using mixed methods to explore loneliness among people with complex emotional needs. Therefore, the researcher will contextualize and elucidate quantitative findings within qualitative experiences; therefore, combining subjectivity and complexity of reality with standardised and representative findings gathered through the APMS survey. As well as this study informing policy and potential practice, the results of all these studies will be disseminated in an accessible blog in laypeople's terms, and sensitively, to inform the public (in December 2025). The Ph.D. student conducting this research is funded by the Economic and Social Research Council (ESRC) and UCL. This project is a part of a Ph.D. research study.

Improving the quality of care and broadening the range of interventions offered, to include socially focused intervention targets, such as loneliness, has been prioritised and emphasized by people with lived experience of ‘complex emotional needs’ (or a “personality disorder”), professionals, and policymakers. Given the adverse effects of loneliness and the call for a more holistic treatment approach for people with complex emotional needs, as opposed to a narrow focus on self-harm and symptomatic reduction, loneliness is a potentially promising intervention target. To develop an intervention that targets loneliness, an understanding of the relationship between loneliness and ‘complex emotional needs’ and suicidality, as well as investigating the role of discrimination would be useful and necessary. Loneliness, self-harm, and suicide have been a public health crisis and priority, and ‘complex emotional needs’ traits are common in the general population. Therefore, developing evidence-based interventions targeting loneliness among people with ‘complex emotional needs’ would be more beneficial if rooted in a detailed understanding of the associations of loneliness among people with ‘complex emotional needs’. However, pending this aim of developing an intervention, findings from this study can suggest that asking service users with ‘complex emotional needs’ about social connections and loneliness can be beneficial in formulating a treatment plan. If this study reveals a link between loneliness and self-harm/suicidality then this could provide a pathway for future interventions to target the reduction of loneliness, which would consequently contribute to reductions in symptoms of self-harm and suicidality.

The identification of the specific symptoms that could be playing a particular role in exacerbating loneliness could be helpful in developing an effective intervention, with a focus on those specific symptoms exacerbating loneliness. This would give a target in intervention development.

Given the extent to which interventions for people with complex emotional needs focus on symptomatic reductions of self-harm and suicidality, investigating whether self-harm and suicidal ideation exacerbate levels of loneliness could provide a pathway for future interventions to target the reduction of loneliness. This would increase personal recovery outcomes (i.e. loneliness), as well as reduce self-harm and suicidal ideation.

The study will also make a case for the role of loneliness and the way in which it could maintain symptoms (self-harm), which can promote future studies exploring interventions for people with complex emotional needs to include outcome measures such as loneliness.

Assessing the possibility that experiences of discrimination modify the relationship between specific traits suggestive of CEN and loneliness would both be a targetable focus of intervention and would promote clinical discussion on the way loneliness intersects with sexuality, gender, and ethnicity.

The results of this study could also contribute to a better understanding of the complex interplay between loneliness and mental health. The associated/moderators who will be assessed, could shape future questionnaire design for a loneliness scale specific to people with complex emotional needs. This is an issue raised by people with lived experience recently.

As of November 2023, there are no yielded benefits to date because the study and analysis of the data is not complete.