Cyber-victimisation of adults with long-term conditions in the UK: A cross-sectional study

Background: Individuals living with chronic conditions and disabilities experience harassment and cyber-victimisation which impose distressing consequences. This is mostly documented among children and adolescents. However, the scope of such experiences is not well-documented among adults with long-term conditions, and the potential impact was not examined from a public health perspective in this context. Objective: This study aims to examine the scope of cyber-victimisation among adults living with long term conditions in the UK and the perceived impact on the self-management of chronic conditions. Methods: This paper reports the findings of the quantitative phase of a mixed-method study in the UK. An online survey was developed and disseminated, the recruitment was online via 55 victim support groups, health support organisations, and social media accounts of non-governmental organisations and activists. Results: Quantitative data from 152 participants showed that almost one in every two adults with chronic conditions was cyber-victimised (45.39%). Most victims (76.81%) had a self-reported disability, and the relationship between cyber-victimisation and disability was statistically significant. The most common means to contact the victims included Facebook (63.24%), followed by personal email or text messaging, each accounting for 27(39.71%). Nine participants (13.24%) were victimised in online health forums. Furthermore, 61.11% of victims reported that experiencing cyber-victimisation had affected their health condition self-management plan. The highest impact was on lifestyle changes such as exercise, diet, avoiding triggers, and avoiding excessive smoking or alcohol drinking. This was followed by changes to medications and follow up with healthcare professionals. The majority of victims (69%) perceived a worsened self-efficacy on the Self-Efficacy for Managing Chronic Diseases Scales.


Table of Contents
2) If accepted for publication in a JMIR journal, would you like the PDF to be visible to the public?Yes, please make my accepted manuscript PDF available to anyone at any time (Recommended).
Yes, but please make my accepted manuscript PDF available only to logged-in users; I understand that the title and abstract will remain v Yes, but only make the title and abstract visible (see Important note, above).I understand that if I later pay to participate in <a href="http

Introduction
Millions of people worldwide live with chronic health conditions, and the prevalence of such conditions is projected to increase [1].The term 'chronic' is derived from the Greek word 'khronos', which means 'time', and the dictionary definition for a chronic condition is an illness that persists for a long time or with a recurring nature [2].In medicine, chronicity covers a group of diseases characterised by recurrence and slow progression.The medical definition of chronicity includes communicable conditions resulting from infectious agents, such as tuberculosis.In public health, and through the lens of international health organisations, 'chronic disease' typically refers to noncommunicable diseases, which are characterised by a duration of a year or longer, with slow progression and required management that includes medical follow-up and lifestyle changes with or without pharmacological treatment [3].Examples include cardiovascular diseases, diabetes, cancers, and chronic respiratory diseases including asthma and chronic obstructive pulmonary disease [1,4].These represent the leading causes of morbidity and mortality worldwide [1].The public health definition of chronic disease is the one adopted in this research.
Chronic conditions and disabilities overlap in terms of definition and day-to-day experiences.Hence, a chronic disease can result in disability, and vice versa [5].For example, 25% of people with chronic conditions have disabilities, and 80-90% of people with disabilities have chronic conditions [6].The Equality Act 2010 in the UK defines disability as a 'physical or mental impairment and the impairment has a substantial and long-term adverse effect on [an individual's] ability to carry out normal day-to-day activities'.A total of 14.6 million people in the UK had a disability in the year 2020-21, which represents 22% of the total population [7].It is important to distinguish that not all impairments are chronic conditions, and not all chronic conditions are disabling; however, they significantly overlap.The major points in this research are the chronicity factor, which indicates that a person is living with a condition, and the self-management aspect, which reflects the day-to-day changes to lifestyle or medications to manage the condition.To reflect this, from this point onward, the conditions covered in this paper are referred to as 'long-term conditions' or 'chronic conditions'.Disability will be specifically highlighted in questions specific to disability.
Living with a long-term condition is physically and mentally demanding to manage on a daily basis.This is further complicated by being treated differently in society.The 'offline' targeting of people with long-term conditions is a documented phenomenon among young individuals [8] and has also been reported as hate incidents against disabled adults [9].The increase of online communication has further reshaped this phenomenon to include 'online targeting', or cyber-victimisation.
A systematic review examined the experiences and impact of cyber-victimisation of people with long-term conditions and disabilities [10].The narrative synthesis of reported results covered a total of 3,070 people with chronic conditions from ten included studies.The sample sizes ranged from 42 to 823 participants, and the age range was 6-71 years.The reported prevalence range of cybervictimisation was 2%-41.7%[10].The risk of being targeted was consistent for people with longterm conditions, which was described as being 'different'.Such differences might include visible physical differences, invisible neurodiversity or differences in lifestyle management of the health condition, such as using an inhaler or insulin pump in front of peers [11][12][13].However, researchers from different disciplines and countries used varied terminologies to address such online incidents.
The terminology related to negative online experiences of people with long-term conditions included cyberbullying, cyberstalking, cyber-harassment, cyber-hate and cyber-victimisation.Cyberbullying is a term used to describe online abuse that involves a power imbalance between the victim and offender; it was the most commonly used term in previous studies [11].Due to its emphasis on perceived differences in power, cyberbullying is a term used with young victims, such as in schools or in workplaces where the victim has less authority than the perpetrator [10].Cyberstalking was another term used [14], which is characterised by fixation and persistence.Such persistence can also be seen in cases of cyber-hate and disability hate crimes in which victims had experienced repetitive harassment from similar groups with a fixation on the impairment [15].Cyber-victimisation and cyber-harassment were used as generic terms to describe the experience of intimidation or abuse using online communication [12,14].Accordingly, due to such differences among researchers and to facilitate communication internationally, the umbrella term 'cyber-victimisation' was adopted in this research.
The reported scope and impact of cyber-victimisation lacks examination of the phenomenon in older age groups.Moreover, limited studies have focused on health consequences.In a cross-sectional study in Sweden [12], a sample of 8,544 individuals was examined, of which 762 individuals were disabled, aged 12, 15 and 17 years.The impact on the victims was mainly subjective health complaints [12].Another public health study in Sweden [11] looked at 413 participants aged 13-15 years.The reported impact of online experiences included poor health, mental health consequences and self-harm.Both studies [11,12] provided insight into the impact of cyber-victimisation on health; however, the target population was not adults.
In the United Kingdom (UK), individuals with long-term conditions comprise 30% of the population, 64% of outpatient appointments and 70% of inpatients [16,17].
No previous research has examined the online experiences of people with long-term conditions in the UK [10].A relatively recent petition was raised to the House of Commons in the UK with concerns over the cyber-victimisation of people with disabilities.This was followed by investigations, and the governmental report acknowledged the concerns over the cyber-abuse of people with long-term conditions and disabilities.It recommended further legislative and non-legislative acts to prevent such experiences and their long-term impact on health [18].The research reported in this article was used to inform this governmental report to identify the impact of cyber-victimisation on people with long-term conditions.This study aimed to examine the scope and impact of cyber-victimisation of people with long-term conditions in the UK.

Ethical approval
Ethical approval was granted by the University Research Ethics Committee at the University of Bedfordshire, UK (IHRREC C557).Ethical considerations were an ongoing process due to the sensitivity of the topic, which also included developing a risk assessment for participants and researchers.The risk assessment included categorising the potential risks arising during the study from low to high, their likelihood, and what was planned to mitigate the risk, such as signposting to support channels, additional discussions with the ethics committee, or a need for disclosure to protect the participants from immediate harm.

Target population
The target population in this survey included individuals aged 18 or over, from any gender, any ethnic background, with a self-reported chronic condition or impairment of a minimum duration of three months, residing in the UK during the research period, and with internet access.The participants were identified as having a long-term condition if they responded 'yes' to the question 'Do you have a long-standing medical condition/illness or disability that requires monitoring, lifestyle changes, and/or taking medications?By long-standing, we mean anything that has affected you over a period of at least 3 months or that is likely to affect you over a period of at least 12 months'.To ensure that only eligible participants could complete the survey, a pre-screen at the beginning of the survey confirmed the eligibility criteria.Any missing criterion was designed to lead to a "thank you" note and to ending the survey.

Survey design
The survey questions were developed based on a literature review and discussions with experts in cyber-harassment and were further refined after the piloting stage.The final survey was put online on Qualtrics website using an institutional account.This platform provided sufficient accessibility options for this research.The process of designing the questionnaire online included several tests to check the layout, question designs, and navigation between sections.A further check was conducted to ensure that the results reports reflected the main statistical output expected from each question.When the survey became fully functional, it was used for the piloting stage.

Piloting
The development of tools included a pilot study conducted by the researchers over four weeks, after gaining ethical approval and before commencing the main data collection campaign.The aim of this stage was to test the functionality, clarity, and usability of the online questionnaire and to obtain input from respondents on the wording or other areas of concern.The respondents were approached on the university's campus and via direct contact with healthcare professionals.The researchers explained to the respondents that the study was a pilot test and invited them to fill out the questionnaire using a "think aloud" approach.The researcher asked the respondents to think loudly while completing the survey to get their real-time feedback on survey questions or use, which helped minimise memorisation issues [1,2].After completing the questionnaire, a short interview was conducted with a pre-designed set of questions derived from the literature [3][4][5][6].The set of questions covered the following points: 1) thoughts on time to complete the questionnaire, 2) issues regarding the clarity of instruction, 3) overall layout, 4) confusing questions, 5) objectionable questions, and 6) additional comments to improve the survey.
The number of respondents was 10, representing various demographics in age, gender, ethnicity, and occupation.Four of them reported living with a long-term condition, and two of them went through the experience of cyber-victimisation and provided answers and feedback based on their lived experiences.Respondents who did not have a long-term condition were given the chance to make several attempts at the questionnaire and provide different answers to give feedback on the clarity of questions and layout.The mean time spent filling out the questionnaire was approximately 15 minutes if all sections were answered.The piloting stage influenced the recruitment stage by adding pre-screen questions.This resulted in moving one question to the pre-screen to include only participants with long-term conditions.There were minor issues in skip logic that required technical support from the Qualtrics team.This stage also included changes to the wording and options in six questions (religion, health condition, level of fear and distress, clarification of online harassment, and options of contact by the harassers in two questions).The question on the self-management efficacy scale was understood by the respondents, and the results were in line with the expected statistics from it.

Survey sections
The survey was open to all visitors to the webpage and did not require registration to the website.The survey page started with a pre-screen to confirm three main criteria related to age, living in the UK, and having a long-term condition.This was followed by a briefing consent form.To fill out the questionnaire, participants had to confirm by ticking boxes that they understood the information given, the anonymity, the right to withdraw, and contact details for further information or to complain.The survey was voluntary, and the participants could skip questions, as highlighted in the consent form, to avoid eliciting distress.Additionally, most questions included 'not applicable' or 'rather no say' as answer choices.The participants were also provided with a back button to check or change answers if needed.A survey logic was implemented to show the participants the options selected in their previous answers, or automatically skip questions not relevant to them.The questions followed this logic without the randomisation of the question.The survey included validation questions to prompt giving a response without forcing it.The survey had six major sections, each of which included a number of questions.To ensure accessibility, short questions were grouped into one page, and long questions that included matrix buttons or scales were placed on separate pages.The first section focused on demographic information, such as gender, ethnicity, employment, and county of residence.The main outcomes anticipated from this section were the sample description and victims' characteristics.The second section focused on the long-term condition and self-management plan.The participants had to tick their conditions and duration, and were given additional space to add any condition.The plan was to further group the written conditions during analysis according to the nearest medical diagnosis in the 10th version of the International Classification of Diseases for 2015 [7].Participants with comorbidities were asked about the health condition that affected them most.The third section was about cyber-victimisation experience; it started with two questions to identify victims.The first question provided the definition of cyber-victimisation in this study and asked participants if they had experienced this.Cyber-victimisation in this research was defined as 'unwanted repeated contact via the internet such as email, chartroom, online forum, social network, mobile phone message, or other electronic means that was used to harass, insult, embarrass, or spread lies about the victim'.The second question was a direct question about whether they considered themselves victims of online harassment.Fear associated with distress was also included in the third section of the survey because it has been documented that the psychological effects of victimisation have more impact on health [8,9].The fourth section explored the participants' coping, self-management during or after the cybervictimisation experience, and the perceived motivation of harassment [10][11][12].The impact of cybervictimisation on the self-management plan was examined in multiple questions using impact statements, a Likert scale and a self-efficacy scale.
The fifth section was designed to examine the actions taken by the victims and the support received in response to the experience of cyber-victimisation.The last page invited participants to volunteer for the second qualitative phase, which is reported elsewhere (Manuscript 40227).

Using a standardised scale
Self-efficacy is a core concept in the self-management of chronic conditions; it represents patients' own beliefs in how capable they are of taking control of managing their health conditions [13].
Hence, the Stanford standardised efficacy scale was used to examine the perceived impact on the self-management of health [14,15].It is formed of six questions to be answered with a score from 0-10, with the average of the six scores representing the self-efficacy of the participant [14].
The researchers aimed to examine the difference in self-efficacy in the self-management of chronic conditions before and after/during the experience of cyber-victimisation.The participants were asked to respond to the set of questions twice, one considering their self-management before cybervictimisation, and the second considering the current self-management plan.A negative change before and after victimisation could indicate perceived disruption to the self-management plan [16].The limitations of using the scale are discussed in the limitations section.

Recruitment
Online recruitment was through victim support groups, patient-support groups, and social media accounts of organisations and activists in the fields of cyber abuse or disability campaigners.Search engines were used to look for victim and health support groups.The keywords used included: patient, support, chronic, health forum, disability, hate crime, online support, and specific health conditions' names.The inclusion criteria for gatekeepers included: a) established patient and victim support groups/organisation, b) based in the UK or with a significant audience from the UK, c) having terms and policies in their websites aligning with ethics to protect participants [17], d) having direct contact with patients/victims, and e) provided contact details.Further snowballing was followed to reach relevant organisations, charities, journalists, academics, and activists in the field.
The lead researcher contacted 'gatekeepers' via email.When no response was received within 1-2 weeks, an email reminder was sent.In cases where a telephone number was provided, further contact via phone was made.Gatekeepers were provided with information related to the rationale of the study, expected benefits to participants in the short and long term, inclusion criteria, the survey link, study poster, and contact details.Gatekeepers who agreed to collaborate in this research and help in recruitment sent the survey link to potential participants via their mailing lists, social media accounts, and monthly updates.
The recruitment process uncovered challenges in reaching the target population due to the sensitivity of the topic, especially since a considerable number of victims were still experiencing harassment.Four overarching themes influenced the recruitment process: social identity in online support groups, the influencing role of online gatekeepers, the contradictory role of social media, and promoting inclusivity.The challenges and lessons learned from online recruitment in this sensitive topic were theorised using social identity theory and published elsewhere [18].

Data collection process
The average time to finish the survey was 15 minutes; it was longer for participants who completed sections relevant to cybervictimisation.This was consistent between the pilot and main studies.There were daily checks of responses by the researchers to screen IP addresses, filter bots, and remove duplicate responses or false victimisation cases.A separate screening form was developed by the research team in cases of suspicion of false victimisation.False victimisation refers to responses that raise suspicions over being factitious or associated with delusional disorders.The screening tool was used once in this study, and the suspicious response was excluded from the analysis.The data was anonymised with no means to be traced to the participants' identity, and was stored in accordance with the Data Protection Act 1998.Anonymised data were stored in a password-protected device, and the data were shared only for analysis with the research team.The dataset was not put in an open repository due to the sensitivity of the topic and as another level of reassurance for participants.

Analysis
The survey data were collected over 18 months, from September 2015 to the end of March 2017.Incomplete responses were recorded 48 hours after the participants' last activity.A total of 424 individuals accessed the survey online; 310 of them were eligible based on the pre-screening, with 222 people consenting to participate and 152 participants completing more than 50% of the survey.This is the final number included in the analysis.The first step in the analysis was to use univariate statistics for descriptive statistics [19].The participants reported various chronic conditions and/or disabilities.The demographic data were presented, followed by information on the long-term condition.To ensure consistency and accuracy in categorising and reporting these conditions, each response was categorised in accordance with the International Statistical Classification of Diseases and Related Health Problems 10th Revision [7,20].Due to variations in terminology used by the participants, each condition entry was checked manually and cross-checked individually with the ICD-10 classification.
The prevalence of cyber-victimisation in the sample was calculated, and descriptive statistics of the victimisation experience were represented.Fear/distress was presented on a Likert scale, and also grouped into a binary outcome as fear vs. no fear [8].The number of respondents in this section was variable to allow for skipping questions with which they were not comfortable.Hence, the frequency reflects the number of respondents to a specific question.The impact of cyber-victimisation was analysed using descriptive statistics and the calculation of the self-management efficacy scale.For each participant, the scale was calculated before and after/during victimisation as described above.The third step in analysing the survey data was to make cross-tabulations between the independent variables.Cross-tabulation was used to identify different factors in relation to the scope and impact of cyber-victimisation.Statistical significance tests were performed using Stata 12.The main independent variables were gender, ethnicity, age, disability status, and the impact of cybervictimisation.The statistical significance was measured using the chi-square test to examine the observed versus the expected number of 2 × 2 tables, with a P value of significance if p < .05.The Fischer exact test was used when the number in any cell was less than five [19].To examine victims' characteristics, cross-tabulations were made to highlight the main characteristics of disabled victims, and to compare them with the whole sample.

The diversity of reported long-term conditions
The participants (n= 152) had a wide range of diverse health conditions, with most having multiple comorbidities.Hence, 340 health conditions and comorbidities were collectively reported.Chronic lower respiratory diseases were reported by 53 (34.87%) participants.The second category was endocrine and metabolic diseases reported by 46 (30.26%) participants, and included conditions such as diabetes mellitus, thyroid diseases, and Wilson's disease.Mental and behavioural disorders were reported by 46 (30.26%) participants in the sample.Among these, 4 (2.63%) participants were living with autism spectrum disorder and 3 (1.97%)participants reported Asperger's syndrome.Diseases of the skin-eczema and psoriasis-affected 40 (26.32%)participants.A wide spectrum of nervous system diseases, such as epilepsy, was reported by 38 (25.00%) participants.Diseases of the musculoskeletal system, such as rheumatoid arthritis, and fibromyalgia, were reported by 36 (23.68%) respondents.This category also includes a range of connective tissue disorders such as hypermobility syndrome, gout, and scoliosis.Diseases of the digestive system, such as non-infective inflammatory bowel diseases, were reported by 24 (15.79%)respondents.Other less common but no less debilitating conditions were reported, such as genitourinary conditions (15; 9.87%), circulatory system disorders (13; 8.55), congenital malformations or chromosomal abnormalities (10; 5.58%), neoplasms (9; 5.92%), hearing impairments (4; 2.63%), visual impairments (3, 1.97%), and injuries (3; 1.97%).

The experience of living with a long-term condition
The participants (n=152) were asked about the condition that affected them most.

Cyber-victimisation experience
Cyber-victimisation was found to be prevalent in this sample, as 69 (45.39%) participants were victimised online.The term 'victim' will be used from this point onward to refer to this group for clarity purposes.Due to ethical considerations, responding to questions related to cyber-victimisation was voluntary; hence, the number of respondents in this section varies.Among the victims (n = 68), the majority ( It was reported by 68 victims that 20 (29.41%) harassers were strangers, 14 (20.59%) were identified as acquaintances, and 9 (13.24%) were ex-partners; however, 10 (14.71%) victims were unsure about the identity of their harassers.Additionally, 16 (23.53%)victims specified other categories, such as neighbours, ex-partner's partners, or fellow members of online support groups.
When the victims (n=53) were asked whether they considered having this chronic condition or impairment to be related to the experience of being harassed online, 22 (41.51%)responded 'yes'.These participants were provided with a space to explain their answers, their answers included experiences of disability discrimination, harassers pretending to have the same health condition to get closer to them, or the longer time spent online due to the impairment.This finding was also examined in the qualitative phase of the study (Manuscript 40227).
To find commonalities and differences between the whole sample, all victims, and disabled victims, the characteristics of each of these groups were cross-tabulated and summarised in Table 2.The table shows minimal demographic differences between the sample, participants who experienced victimisation and disabled participants who experienced victimisation.The impact of cyber-victimisation Among 54 victims, most respondents (33; 61.11%) reported that cyber-victimisation had resulted in an impact on their self-management of chronic conditions.Among these, 32 participants provided more details, they were given their personalised management plan as they shared individually, and were asked to tick the parts of the health management plan that were affected.The majority of changes were under the lifestyle category, such as avoiding triggers that exacerbate illness (19; 59.3%) and healthy eating (12; 37.50%).They also included changes to medications, follow-up with general practitioners (GP), and self-monitoring.A detailed breakdown of the affected aspects of the self-management plan is shown in Table 3.The impact of cyber-victimisation on the self-management plan was further examined by asking the victims to endorse impact statements that apply to them, which were ranked on a 5-point Likert scale ranging from always to never.A total of 32 victims responded to this question, and their responses reflected multi-level effects on health management and provided potential explanations for the changes stated in Table 3.A detailed breakdown of the impact statements and their endorsements is reported in Table 4.To identify the conditions that were more commonly victimised, these were cross-tabulated with cyber-victimisation.Due to the low number, a statistical significance test was not performed but highlighting these conditions is important for future research.These were mainly people with asthma, diabetes, depression, chronic obstructive pulmonary disease (COPD), anxiety, MS, ME, fibromyalgia, EDS, heart disease, thyroid disease, and IBD.The results reported above were further cross-checked to identify the impact of cyber-victimisation on each chronic condition reported in the sample, and this impact was shared with the UK government to guide future mitigating actions [22,23].Table 5 summarises the impact reported based on the chronic conditions.

Category
Reported impact An additional step to measure the impact of cyber-victimisation included using the Stanford selfefficacy for managing chronic disease 6-item scale.The score was calculated for each victim (n = 55) before and after the cyber-victimisation experience; it was negative in 38 (69.09%) responses, positive in 7 (12.73%)cases, and zero in 10 (18.18%) cases.Hence, a negative difference in scale indicates a perceived change in self-efficacy before and after the cyber-victimisation experience and potentially reflects a negative impact of cyber-victimisation on the self-management of chronic conditions.

Diabetes
The relationship between gender and being cyber-victimised was not statistically significant with a pvalue of .61using the chi-square test.The Fischer exact test was used to examine the relationship between gender and the perceived impact on self-management; however, the result was 1.0, which was not statistically significant.
There was a statistically significant relationship between being a disabled person and cybervictimisation with a p-value of .23.However, there was no difference in the perceived impact of cyber-victimisation between disabled victims and non-disabled victims.The p-value using the chisquare test was 0.19, which was not significant at p < .05.
Sexual orientation and employment status in relation to cyber-victimisation were not statically significant.Reporting fear and distress was statistically significant with regard to the impact of cyber-victimisation, as shown in Table 6.
Table 6 The relationship between fear/distress and perceived cyber-victimisation impact on selfmanagement is statistically significant (n=54 The chi-square statistic is 18.82.The p-value is .00.This result is significant at p < .05.
The impact of the duration of cyber-victimisation was also examined.The chi-square test was not statistically significant, with a p-value of .20.However, when the categories were narrowed to one year or less compared to more than one year, there was a significant relationship between the duration of cyber-victimisation and its perceived impact.The chi-square statistic was 4.77.The pvalue is .029,which is significant at p < .05.

Support
The participants sought formal and informal support to cope with the cyber-victimisation experience.Informal support was commoner; among 52 respondents, a total of 37 (71.15%)victims received support from their families.When asked about how helpful it was, family support received variable ratings as: very good (14/37, 37.84%), good (10/37, 27.03%) and poor (11/37, 29.73%).Most victims also received support from their friends (40; 76.92%), which they rated as primarily very good (17/40, 42.50%).Formal support was less common and the number of respondents varied.It included approaching victim support groups (20/50, 40%), which were generally rated poor (11/20, 55.00%).Healthcare professionals were also approached (22/52, 42.31%) and this was mainly rated as very good (10/22, 45.45%).The police was contacted by victims (20; 38.46%) and was mainly rated poor (13/20, 65.00%).The support sought by the victims and the perceived effectiveness of the support are detailed in Table 7.

Summary of findings
This cross-sectional study represents the quantitative phase of a mixed-method research to examine the scope of cyber-victimisation experiences among people with long-term conditions and disabilities in the UK, and how it affected their self-managed health plan.Around one in every two people with long-term conditions in this study experienced cyber-victimisation.The sample was diverse in demographics, such as age and ethnic groups, with the majority of participants being female.The participants reported a range of chronic conditions and impairments that were grouped using ICD-10 classifications.The majority of changes to the self-management plan were under the lifestyle category, in addition to changes to medications, follow-up, and self-monitoring.The participants perceived lower self-efficacy, which potentially affected their self-management.The most common means of contacting the victims was Facebook, and most harassers were strangers.Statistical tests were significant between cyber-victimisation and disability, fear/distress, and the perceived impact of cyber-victimisation on health, long duration of abuse (more than a year), and the perceived impact of cyber-victimisation.Support was sought from formal and informal support channels, with the former generally rated as poor.

Comparison to prior work
It is challenging to compare the scope of cyber-victimisation among people with long-term conditions with the literature.This is mainly because the prevalence of cyber-victimisation depended on the definition and criteria adopted by the researchers to describe a negative online experience, which varied [8,24].This remains an issue.A recent review [25], highlighted the challenges of prevalence inconsistencies in the cyber-victimisation literature due to issues in definitions and methodological variations, in addition to contextual factors, including culture and geographical settings.Among people with chronic conditions, cyber-victimisation was reported to be as high as 41.7% [26]; however, this was in a younger age group and in a different context than this study.It is important to acknowledge cyber-victimisation as a global health issue, and further work is needed to tackle inconsistencies in definitions to have a clearer understanding and facilitate conversations between researchers internationally.
The majority of the participants in this study were females, with no statistically significant difference between the genders.In the current literature, studies that examined the cyber-victimisation phenomenon and its impact on different groups were inconsistent; in some cases, cyber-victimisation was associated with the male gender [27], and in other cases, it was associated with the female gender [28,29].Notably, most papers that focused specifically on victimising people with disabilities were male-dominated [30][31][32], and some studies showed increased cyber-victimisation towards disabled girls [33].This could be influenced by several factors, such as the young age group in previous studies or focusing on specific disabilities that are commoner among males, such as attention deficit hyperactivity disorder (ADHD).Hence, the current study added to the literature by reporting the experiences of people with long-term conditions with input from women.Further research is needed to examine whether this result reflects attitudes towards participation, higher cyber-victimisation among women, or whether cultural factors have influenced the results, for example, if men are seen as masculinity figures who should not disclose similar experiences.
The participants in this study were all adults aged 18 or over.This is an important addition to the literature.Previous studies on cyber-victimisation have focused on young age groups [24], and how cyber-victimisation affects older populations remains under-research [25].A review of behalf of the Department for Digital, Culture, Media & Sport examined the evidence on the harms of online experiences on adults, and acknowledged the scarcity of evidence in examining disability hate against adults [34].
Most of the victims in this study were disabled, and there was a statistically significant relationship between cyber-victimisation and being disabled.This is in line with previous research on cyberharassment and disability [35]), and also research examining cyberbullying among younger age groups [36,37].Additionally, almost half of the victims considered victimisation related to their conditions or impairment.One explanation could be the targeting of people with physical impairments by harassers.This is in line with the role of disability discrimination and hate in the literature [35,38].It is alarming to see disability discrimination taken to an online context and can potentially lead to cyber incidents or crimes.This study focused on people with long-term conditions, and this significant association that builds on existing literature makes disability and cyber-victimisation a research area to be examined by multidisciplinary teams.
The characteristics of the whole sample and those of the victims with long-term conditions were comparable.The age of the victims was slightly higher in those with disabilities.This finding is unlike the literature that focused on cyberbullying among children [30,32], showing how the victimisation continues throughout the life course.Employment status and professionals were less among victims, and less among victims with disabilities.This could be due to restricted physical activity in some physical or invisible impairments [39].However, this could also reflect accessibility issues, marginalisation, and stereotyping of disabled people [38].Despite the slight differences, the sample, victims, and disabled victims had comparable characteristics, suggesting an alarming risk of being victimised across all groups.
Most of the victims in this study experienced fear and distress, which is consistent with previous studies [8].The relationship between fear and cyber-victimisation impact was statistically significant.This perceived impact was also significant in cases with longer durations, which extends the literature and could be used for awareness-raising and health promotion to prevent long-term health consequences.Fear and eliciting distress were factors used in previous studies to examine the impact of cyber-harassment [40], and eliciting distress was also included in defining cyberstalking [8,41]).Fear can also be viewed as a precursor to harm, which can be physical or mental.Although fear is reported here as an impact because it might influence how the individual manages the chronic condition and results in health consequences, it can also be viewed as a factor to build on for future interventions.For example, fear of safety was one of the factors that facilitated the reporting of cyber-hate cases to the police [42].
The diversity of reported conditions in this research ensured covering different impairments, scoping the impact on each condition, and directing future research.In the literature, only a few of the conditions reported here were reported collectively, and none were specifically reported in relation to victimisation [24].Asthma was the most frequently reported condition in this study.The impact of victimisation on managing asthma was studied previously among young patients [37,43]; however, it was not examined at a later age.Diabetes was highly prevalent in the sample, which could reflect its prevalence in the general population and documented victimisation [37].Patients with thyroid diseases were also victimised; however, this has not been studied before and requires further research.These findings do not exclude people with other conditions; rather, they warn of the increase in cyber-victimisation and the need for research to examine the specific impact on health conditions.Anxiety and depression were also reported in the sample and were exacerbated by cybervictimisation, which is concerning, considering the distress caused by the experience itself [8].Individuals with autism spectrum disorders and Asperger's syndrome were included.However, the impact and victimisation of people with these conditions were less than expected compared to previous studies [31,32,44].This comparison, however, is not conclusive due to the low number of these participants.This could be influenced by the recruitment process and thus requires further research.Such findings reflect the wide range of conditions included; they might also suggest differences in impact compared to younger victims or could be a result of methodological differences.
Invisible conditions, such as multiple sclerosis and myalgic encephalomyelitis were highly reported.The victimisation of people with invisible disabilities has been documented [45] and was further confirmed by this study.Patients with epilepsy also shared the impact of cyber-victimisation on their self-management.Previous studies showed that people with epilepsy were victimised offline [46] or online at a young age [37], confirming that people with conditions documented to be victimised offline, but not studied online or among adults, could be at risk of cyber-victimisation.Diseases of the musculoskeletal system and connective tissue disorders were reported by the victims, and they require further research concerning cyber-victimisation.EDS is a rare condition in epidemiology [47].Nonetheless, it was a considerable concern to the participants.The representation of invisible and less common conditions could be linked to the participants' identity and attitudes towards participation [18].
In total, 61.11% of victims reported that experiencing cyber-victimisation affected their selfmanagement plan.Previous research has not specified changes in managing health after victimisation [10,11,48].After cyber-victimisation, the reported impact on self-management was mainly in avoiding triggers, healthy eating, and avoiding exercise.The importance of this lies in the specific aspects of each condition.Lifestyle changes are broad, and the trigger is different in each management plan [49].Additionally, healthy eating and exercise are essential aspects of selfmanagement, for example, in diabetes, musculoskeletal conditions, and depression.Moreover, triggers of neurological, mental health, and heart conditions can have an immediate effect [50].Regular medications were also affected.Missing medications, for example, in heart diseases and diabetes, can trigger life-threatening situations [51].This indicates the need to raise awareness to prevent such serious complications.In this study, 69.09% of victims perceived a worsened self-efficacy scale for the self-management of health conditions following cyber-victimisation [14].It is acknowledged that such results do not quantify the impact of cyber-victimisation, and the participants already experienced fear.However, the results reflected the victims' perceptions of how this experience affected their coping.Thus, it could be used as a rough estimate to demonstrate the health disruption caused by cyber-victimisation.
By examining the population at risk of cyber-victimisation, the diversity of the included condition, and the multi-level impact of self-management, it can be argued that cyber-victimisation is a threat to public health.This is in line with previous work that acknowledged that cyber-victimisation results in unexpected health consequences and, in turn, health-associated costs to individuals and systems [25].
Identifying cyber-victimisation as a global health issue is an essential step in an increasingly connected world with massive online communication.During the COVID-19 pandemic, online experiences changed, cyber-victimisation risks increased [52], and more hate crimes were reported in the UK [53][54][55].In public health emergencies, and without proper action, people with long-term conditions might face long-and short-term health consequences.

Strengths and limitations
This study has contributed to the body of literature by focusing on adults as an age group and addressing a diverse range of health conditions and impairments.The researchers aimed to give every person living with a chronic condition in the UK the opportunity to participate.However, equal chances for participants in this study were influenced by the recruitment strategy, because gatekeepers were approached in recruitment.The researchers recognise the influence of the recruitment process on the results and do not claim the generalisability of the findings.However, the findings provided an idea of the frequency and inter-relationship between having a chronic condition and, cyber-victimisation experience, and its impact on self-management.Additionally, the recruitment was inclusive of participants facing physical barriers and people who were determined to share their voices, for example, disability rights advocates.The sample in this study is not large; however, the study was specifically designed to examine cyber-victimisation without treating chronic conditions as a homogenous group.Previous studies utilised existing datasets that are not specifically designed for this topic, and chronic conditions were mostly reported in large samples as a homogenous group [24].Hence, the study design and specific conditions will guide future research.Using an online approach to reach participants was an inclusive option, given the range of health conditions and the sensitivity of the topic.However, this approach was also challenging.The challenges that faced this study during the recruitment stage were published elsewhere [18].Lack of internet access and socioeconomic status are also limiting factors to consider [56], as well as social desirability bias in self-reporting [57].This was managed by designing the survey in a way that more than one question was assigned to address one issue; for example, two questions covered cybervictimisation experience, and four questions covered the impact on self-management.Additionally, we encouraged the participants to elaborate on their experiences in the qualitative phase of the study.The self-efficacy scale used was a validated scale.However, the participants were asked about their self-efficacy before and after cyber-victimisation at a single point in time during data collection.Hence, the scores are not conclusive, and they might be influenced by recall bias, or exaggerated in cases of ongoing harassment or mental health impact.This question was used to examine perceived impact, in combination with other questions on the impact of cyber-victimisation, rather than a standalone score.

Conclusion and future directions
This study pioneered research on cyber-victimisation of people with long-term conditions in the UK and identified the need to build proper support that is context-specific and condition-specific.
Reaching context-specific work could be refined in future research, and a health condition-specific work can be achievable by using these findings to identify possible conditions that were targeted and their potential impact, which could help tailor specific prevention interventions and support by experts in the field.All conditions reported in this study require attention and further investigation due to their potential impact upon victims.It is also essential to tackle inconsistencies in definitions and recognise cyber-victimisation as a global health issue that requires international conversations and consistent language to grasp the scope of the issue and potential interventions.Further research is also needed to examine how public health emergencies in the age of online communication, such as the COVID-19 pandemic, have influenced the online experiences and health outcomes of people with long-term conditions and disabilities.Victimisation of people with chronic health conditions, especially those with disabilities, will continue if we do not take a holistic approach to tackling this pressing issue.

Table 1 .
60; 88.24%) reported experiencing fear and distress as a reaction to abusive communication, ranging from extreme fear and distress (22; 32.35%) to moderate fear (24; 35.29%) and slight fear (14; 20.59%).The duration of the victimisation was more than a year in 25 (36.76%)cases, and between three months and one year in 15 (22.06%) cases.The harassment was ongoing in 17 (25.00%)cases,and12 (17.65%)victimswerenot sure whether the harassment had ended.The most common means of contacting the victims (n=68) included Facebook, as reported by 43 (63.24%)victims,followed by personal email or text messaging, each accounting for 27 (39.71%) of victims as detailed in Table1.Phone calls were reported by 26 (38.24%) victims.Other means of contact included websites, such as eBay, chatrooms, spam subscriptions, or hacking into friends' accounts.Some participants (9; 13.24%) were victimised in online health forums.Most victims, 67 (98.53%), were contacted once or more per day by their harassers.The means used to contact the victim with frequency and duration (n = 68)

Table 2 .
Comparison between the main characteristics of all participants, victims, and disabled victims.

Table 3 .
Victims' responses to what specific aspects of the self-management of chronic conditions

Table 4 The endorsements by victims on impact statements that applies to them on a 5-point Likert scale (n = 32).
Exercise/physical activity (reported by multiple participants in this category).