The Impact of a Befriending Service on Health-related Quality of Life in Older Adults: An Interventional N-of-1 Pilot Study.

The Impact of a Befriending Service on Health-related Quality of Life in Older Adults: An Interventional N-of-1 Pilot Study. | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Impact of a Befriending Service on Health-related Quality of Life in Older Adults: An Interventional N-of-1 Pilot Study. Joanna McHugh Power, Eimile Holton, Brian Lawlor, Frank Kee, Thomas Scharf, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4807512/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Purpose: Befriending interventions are unlikely to reduce loneliness, but they may provide social support which buffers the negative impact of loneliness on health outcomes of older adults. An interventional N-of-1 design was used to assess the impact of a befriending intervention on health-related quality of life (HR-QoL) among older adults, and whether such intervention attenuated the impact of loneliness on HR-QoL. Methods: Participants were n = 33 new users of the service, aged 60+. Outcomes were measured at 13 timepoints across 26 weeks, and data were analysed using generalised additive modelling (GAM) with a subset of data analysed using visual analysis. Results: Results indicate that the befriending service may reduce decline of HR-QoL (i.e., health declined in the baseline phase over time: edf = 3.893, F = 3.0, p=0.002, while in the treatment phase, health remained more stable: edf = 5.98, F = 2.98, p=.008). The befriending intervention also suppressed the association between loneliness and HR-QoL. Conclusion: We supported our hypothesis, that befriending interventions may moderate the impact of loneliness on HR-QoL. Interventional N-of-1 designs however carry considerable recruitment and participant burden, which should be considered prior to onset. This research provides an insight into practical difficulties when evaluating existing community-based services, particularly in relation to adhering to best practice design guidelines. Psychology intensive longitudinal N-of-1 quality of life loneliness Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Loneliness is a painful insufficiency in the quality or quantity of social relationships [ 1 ]. The link between loneliness and health among older adults is well established [ 2 ]. This link is concerning given the high levels of loneliness observed among older populations [ 3 ]. As interventions on loneliness typically yield small effects [ 4 ] research prioritising effective intervention and evaluation is warranted. One intervention often offered to alleviate loneliness is befriending. Befriending interventions are designed under the assumption that providing companionship and social support alleviates loneliness, as per major theoretical approaches [5; 6]. Befriending interventions do not reliably reduce loneliness [7; 8] but they may improve older adult health [9; 10]. Befriending interventions may indirectly affect health by buffering the negative impact of loneliness. We invoke the stress-buffering model as comparison [ 11 ] –in which social support “buffers” the individual from the negative impact of stress on health. Social support provided in a befriending intervention may buffer the individual from the impact of loneliness on health, since loneliness is a stressor [12; 13]. The aim of this study was to test whether a befriending intervention improved health outcomes of older adults, either directly and/or by reducing the impact of loneliness on health. We focused on health-related quality of life (HR-QoL). Although limited evidence exists for a therapeutic effect of befriending interventions on HR-QoL [ 10 ] loneliness is associated with HR-QoL [ 14 ]. When evaluating a community service, feasible research design is an important consideration. Control groups are inappropriate when individuals are seeking a community service. N-of-1 designs (single case designs, SCDs) offer practical, ethical, person-centred alternatives to between-group designs [15; 16], and are well-established in health literature [17; 18]. Interventional N-of-1s include baseline observations and systematic manipulations of an independent variable, and sample sizes tend to be small (> 10), with use of visual analysis common [ 19 ]. We hypothesised that a befriending intervention would a) improve HR-QoL among older adults who experienced loneliness, and b) buffer the negative impact of loneliness on HR-QoL. Methods Design We followed TIDieR and RoBiNT guidelines [20; 21], and What Works Clearinghouse (WWC) documentation [ 22 ] in reporting our interventional N-of-1 AB design, with outcomes measured during baseline (A-phase) and intervention (B-phase). This approach is suitable for community settings where flexibility is required [ 23 ] and there is an ethical issue involved in using a control group or wait-listing service seekers (i.e. in the context of a befriending service, which has known benefits). The sample were new seekers of an existing community-based befriending service which matches older adults to volunteers based on availability, resulting in varying baseline lengths. The service provider agreed to allow two timepoints to go by before participants were matched with a befriender (thus marking the beginning of the intervention, which would then run weekly for the duration of the data collection and beyond, at the preference of the participant). Because matching occurred based on availability, baseline length varied across participants with no systematic manipulation of intervention start-points (AB designs are therefore considered pre-experimental; [ 24 ]. The study was approved by the local faculty Research Ethics Committee. All participants provided informed written consent. Intervention The evaluated intervention is the (removed for review) volunteer-led befriending service in Ireland, which has been run by (removed for review), a non-governmental organisation, since 2009. The stated aim of the service is to alleviate the negative impact that loneliness has on health [ 25 ]. Older adults can self-refer or be referred by others (typically healthcare professionals). A volunteer is paired with a service user to provide weekly social contact until either party discontinues it, or until the service user requests a new match. There is no guidance on the content of the befriending visit – service user and volunteer pairs are free to engage how they wish, and volunteers are advised to visit for one hour. Volunteers receive regular phone calls from (removed for review) to catch issues such as disengagement, issues in the partnership, or concerns about the service user’s welfare. Participants entered the study at least two weeks before their intervention. During the intervention phase, volunteer visits took place on a weekly basis. Since this is a voluntary, community-based intervention, no regular checks of procedural fidelity or adherence were completed. Recruitment Participants were new users of the befriending service who responded to the screening question “do you feel lonely” with either “all the time” or “at least some of the time”. Users were informed that the study was an evaluation of the effect of the befriending service on HR-QoL , and that their decision to engage had no bearing on their service receipt. Interested users were contacted by the research team, who established eligibility, sent information by post, and followed up after one week. Eligibility criteria were: aged 60+, ability to provide informed consent, and living within the greater Dublin area. Exclusion criteria were: self-reported receipt of a diagnosis by a medical doctor of: intellectual disability, psychotic disorder, dementia or serious memory impairment. Participants were made aware that the terms of the study meant a delay of two weeks in receiving a befriender match (although this was not, in practice, different to the length of time normally taken). In total, 246 individuals were referred to the research team as potential participants between September 2018 and June 2021, of a possible 2,485 individuals who approached or were referred for befriending. Recruitment challenges included service staff buy-in, research staff illness (which resulted in participants not being assessed prior to being matched to a befriender, thus yielding many participants who were “matched too soon” in Fig. 1 ), and the COVID-19 pandemic, which prevented recruitment for a period of three months to allow service staff to focus on existing service users. Of these 246 individuals, 76 declined (see Fig. 1 ), 23 were ineligible (20 could not provide informed consent, two were in an ineligible location, one was too young). 147 participants gave consent and were registered to the study. Of these, 21 dropped out, two died, 10 discontinued the befriending intervention, and 61 were matched to a befriender before two weeks had passed. 86 participants in total were registered. 53 participants completed the study, but because 20 participants were not matched to a befriender during the study, full datasets with baseline and intervention data suitable for inferential analyses were only obtained from n = 33 participants ( > = 2 baseline data-points). Visual analysis was also conducted from a subset ( n = 14) who met WWC design standards (see Supplementary Materials ([22; 26]. For the inferential analysis, an a priori sample size requirement of 85 participants was calculated (information removed for review) based on 13 observations per participant, once fortnightly for 26 weeks (power in N-of-1s is a function of number of observations and of participants; [ 18 ]. While the dataset of 33 complete cases fell short of this, the data for visual analysis (n = 14) exceeded the typical 2–3 participants in an N-of-1 [ 16 ]. Given the complexity of the study design, the ultimate analyses are a compromise between the feasibility of the research and the standards outlined [ 27 ] and are best considered as a pilot study. Sample Characteristics The mean age of participants was 75 (48.7% male; See Supplementary Materials). 83% of the sample lived alone (compared with 27% of those aged 65 + in the general population, [ 28 ]). We compared HR-QoL outcomes to Irish norms (from participants aged 65 and over; Irish EQ-5D-5L survey 2015–2016; www.ucd.ie/issda ). 30% of our sample had no problems with mobility relative to 53% in the normative sample. For self-care, 61% of our sample (compared to 85% of the normative sample) had no issues. For engaging in usual activities, 38% had no issues with engaging in usual activities, relative to 68% of the normative sample. 31% of our sample had no issues with pain, relative to 45% of the normative sample; 31% had no issues with anxiety or depression in our sample compared to 82% of the normative sample. Overall, despite our sample having a lower age threshold than the normative data sample, moderate and severe issues in HR-QoL domains were more frequent. A logistic regression was conducted to predict attrition. This analysis showed that those with a lower level of education (OR = 0.85, CI = − .76, − .95), and those who were separated (OR = .17, CI = .04, .72), single (OR = .27, CI = .09, .79), or widowed (OR = .31, CI = .14, .71) relative to divorced were less likely to be lost to attrition. *FIGURE 1 HERE* Procedure and Setting The first data collection point was performed at a home visit, but all data collection switched to phone after the beginning of the COVID-19 pandemic (March 2020). Data collection was scheduled for all participants every fortnight for 26 weeks (yielding max. 13 timepoints). Written consent was returned by post. Data collection was conducted by trained research assistants, who used a script. If participants could not be reached, two more attempts were made to contact them, before the datapoint was marked as “missed”. Measures HR-QoL The EQ5D-5L was used to evaluate HR-QoL [ 29 ]. This index is responsive to changes in health over time, correlates strongly with global health measures and measures of functioning, and has excellent test-retest reliability [ 30 ]. Utility scores can be calculated for five domains of health (see Supplementary Materials). Age norms for Ireland were used to convert responses into utility scores [ 31 ]. A utility score of 1 indicates full health with a score of 0 meaning a state as bad as being dead, with negative values possible. A minimally important difference of .074 was identified [ 32 ]. The utility scores were converted to a disutility index in order to reduce problematic skewness and negative values [ 33 ]. Loneliness was measured using the three-item UCLA Loneliness scale [ 34 ] which is well validated in older populations. Baseline covariates age, gender, years in education, living status (living alone or with others) were also recorded. Analytic Plan Generalised additive modelling (GAM) was used to evaluate trends in outcomes over time for each participant, controlling for covariates. GAM is a semi-parametric regression [ 35 ] where the data inform the researcher about the trend, as opposed to other single-case analytic methods that assume linear or no trend [ 18 ]. The “wiggliness” of the trend is measured using estimated degrees of freedom or “edf”; while values of 1 indicate a linear trend, higher values indicate non-linearity. Visual analysis was also conducted (see Supplementary Materials) following guidelines [ 36 ]. We used the package “mgcv” in R Studio for conducting the analyses [ 37 ]. The hypotheses (and implied models) below were preregistered ((information removed for review)). Data are not available for further analysis due to constraints in the ethical approval of this study. Results Hypothesis 1 The befriending service has a therapeutic (positive) impact on HR-QoL. We followed the steps outlined [ 38 ] in applying GAM to single-case design data from the 33 participants with sufficient data (using a robust maximum likelihood estimator given the small sample size). A Gaussian distribution with a logarithmic link was chosen. First, a model was created with Time (1–13), Phase (0 = baseline, 1 = treatment), and Case (individual ID) as predictors (M0). Then a smoothing term on Time was added (M1), which marginally improved model fit (AIC = -104 to -101). Smoothing terms allow one to define a trend as non-linear, but without interpolation, yielding a line that fits the data well. A third model (M2) was run without the case predictor, which disimproved model fit (AIC = 219) so the case predictor was retained in M3, which also specified that Time terms varied by Phase, in effect evaluating an interaction between Time and Phase which could identify whether the befriending intervention exerted a therapeutic effect on disutility scores. M3 had an improved fit (AIC = -126; adjusted R 2 = .724). Disutility scores increased albeit nonlinearly (i.e. health got worse) in the baseline phase over time (edf = 3.893, F = 3., p = 0.002; see Fig. 2 ). In the treatment phase, disutility scores decreased (i.e. health got better), and were highly nonlinear (edf = 5.98, F = 2.98, p = .008). Result patterns indicated that while health declined in the baseline phase, the befriending intervention could mitigate decline. When covariates were added, fit improved again (AIC = -128, adjusted R 2 = .73). Of the covariates, only living status had an impact (F = 3.6, p = .02) such that those living alone had poorer HR-QoL than those living with spouses. *FIGURE 2 HERE* Visual analyses [ 27 ] were conducted (See Supplementary Materials). Based on these analyses, we concluded that the intervention had therapeutic effects on disutility scores, since for 5/14 participants, the minimally important difference of 0.074 was met (although for other participants there was decline). For 6/14 participants, an effect took place within three timepoints, indicating that the effect of befriending on HR-QoL occurs quickly. Hypothesis 2 The befriending service mitigates (buffers) the negative impact of loneliness on disutility scores. Model M4 modelled the differential association between loneliness and disutility scores over time, to gauge whether the befriending intervention reduced the magnitude of this association. We used a tensor product smoother specifying an interaction between (smoothed) time and (smoothed) loneliness (the “ti” function in mgcv), specified separately for each intervention phase (Baseline, Treatment). Covariates and main effects were included. Model fit improved (AIC = -161; Adjusted R 2 = .77). There were significant interactions in the baseline (edf = 9.7, F = 2.88, p < .001) and treatment phase (edf = 9.48, F = 5.50, p < .001) - in both phases, there was an association between loneliness and disutility scores over time. Being male was associated with lower disutility scores (i.e., better health; F = 8.48, p = .003), as was living alone (F = 9.29, p < .001). To visualise these effects, we used contour plots (in the “vis.gam” function of mgcv) such that the ‘Time by Loneliness’ interaction on disutility score outcomes was plotted separately for each phase (see Fig. 3 ). Hereafter low disutility scores are described as good health, and high disutility scores as poor health. Red areas indicate better health, while the yellow areas indicate worse health. In both contour plots, health was worst at higher initial levels of loneliness. Health declined over time in the baseline phase, and for participants with high levels of loneliness, health was poor and stayed poor throughout the study. However, in the treatment phase, health was stable for longer (getting worse at the end of the study, but this happened for all participants). As such, the intervention reduced the impact that loneliness had on health over time. For participants who had lower levels of loneliness at the study onset, health worsened over time in the baseline condition, but did not get worse over time in the treatment condition until the end – that is, for people with low and moderate levels of loneliness, the intervention prevented decline in HR-QoL over time. *FIGURE 3 HERE* Discussion We explored whether a befriending intervention could improve HR-QoL in a sample of older adults and reduce the impact of loneliness on HR-QoL. While HR-QoL generally deteriorated over six months, as could be expected for a group of older adults with worse than usual functioning at baseline, it deteriorated less after the onset of the befriending intervention. Furthermore, while loneliness was associated with worsening HR-QoL over time, this association was buffered by the befriending intervention. Thus our findings are compatible with prior literature on the buffering effect [ 39 ], and with literature suggesting that loneliness itself is a stressor [ 13 ]. Although high in ecological validity, conducting an evaluation of a community-based intervention was challenging. We hope that sharing these insights may assist future researchers who are planning similar evaluations. To begin, we experienced considerable recruitment issues. When participants were recruited, our design requirements were sometimes at odds with the usual procedures of the service providers. A considerable proportion of participants were matched with a befriender too soon, which meant that they were not eligible for study registration. Because of problematic missingness and issues with matching, only 33/86 participants registered to the study had sufficient data for inferential analysis, meaning that our results are best interpreted as a pilot study. That said, our sample size is large for an interventional N-of-1. Future researchers should consider the feasibility of an interventional or experimental design and ensure that community partners or service providers are clear on the requirements for that design (see [24; 40]). When planning the study, our intention was to use an experimental multiple baseline design (MBD) with baseline (A) and intervention (B) phases, but the MBD would require planned sequential introduction of the intervention across participants (see [ 16 ]). In reality, matching occurred based on availability of befrienders, so ultimately we could not establish the experimental control required for an MBD. This rendered our design a pre-experimental AB design. While this is not uncommon in the N-of-1 literature [ 24 ], further research is required to determine if the effects of befriending observed can be replicated in more controlled design contexts. Public health guidelines also presented challenges for evaluation, since visits were paused during COVID-19 lockdowns, and contact was maintained by phone for most participants and their volunteers. The methodological constraints we experienced present an interesting test case for conducting N-of-1/SCD research in community settings, highlighting the necessity for guidelines/standards that recognise this reality [26; 41] and provide guidance on how difficulties may be mitigated [18; 42]. For instance, given the common methodological issues plaguing befriending service evaluation [ 43 ], such as variability in the content and dose of the befriending intervention; confounding; participant selection bias, it seems clear that designing a community-compatible process evaluation to record the implementation of befriending interventions is critical. The difficulties described above posed additional challenges to methodological rigour in our attempts to meet the WWC standards for quality in N-of-1/SCDs. For 19/33 participants who completed the study as planned, two baseline data points were present rather than the recommended three-to-five [21; 22; 44]. This difficulty is faced by many in applied or community settings. In one systematic review analysing the characteristics of SCDs, 10.5% of included studies (n = 61) had one or two points in the first baseline phase [ 45 ]. As suggested [ 41 ], the negative impact of overly stringent minimum requirements should be considered across a multitude of settings. For settings where the minimum number of data-points cannot be achieved, statistics can be helpful in supporting a more comprehensive analysis [42; 46]. We used visual and statistical analysis for data that met the standard of at least three data-points in each phase and used GAM where data did not meet those standards. When compared to visual analysis alone, this resulted in more nuanced information about the intervention effects. Implications for Practice Overall, the current study’s results indicate that the (information removed for review) befriending service had a therapeutic impact on HR-QoL. Results have implications for service delivery. In 2015, an evaluation of the (information removed for review) befriending service showed that only 4% of total referrals to the service were made for reasons of poor physical health ((information removed for review)). Informing healthcare practitioners and other referrers of the positive impact of befriending on HR-QoL is critical so that older adults who stand to benefit from befriending are referred. Results have implications for the delivery of the ongoing befriending service by (information removed for review) and also for those designing or commissioning new befriending services. Conclusions Our results contribute to mixed existing findings concerning the positive impacts of befriending interventions. To date, little is known about the impact of befriending interventions on physical health: our study provides preliminary evidence that they may be therapeutic for HR-QoL scores, and may also reduce the negative association between loneliness and health. Further research is required to corroborate these pilot findings. Declarations Statement of Ethical Approval The study received approval from the Trinity College Faculty of Health Sciences Ethics Board. Statement of Funding This work was supported by the Irish Health Research Board (grant number APA2017004). The research partner organisation ALONE supplied co-funding through this grant. The Health Research Board had no role in any aspect of the research. ALONE co-author SM reviewed the manuscript for critical content, and facilitated study recruitment. Acknowledgements: Thanks to the staff of ALONE, Trinity College Dublin, and Maynooth University who facilitated this research, and to all study participants involved. This research was funded by the Health Research Board under the award APA 2017 004. Declaration of interest statement: All authors confirm they have no conflict of interest to disclose. Declaration of contribution of authors JMcHP led funding acquisition, study conception and design, data collection, data analysis and interpretation, and article drafting, revision, and submission. EH was involved in acquisition of data, analysis of data, and in the writing, revision, and final approval of the article. BL was involved in funding acquisition, study conception and design, consulted on data collection, revised the article critically for important intellectual content, and approved the final version to be submitted. FK was involved in funding acquisition, study conception and design, consulted on data collection and data analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. TS was involved in funding acquisition, study conception and design, consulted on data collection and analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. CW was involved in study conception and design, consulted on data analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. SM was involved in funding acquisition, data acquisition, revised the article critically for important intellectual content, and approved the final version to be submitted. ME was involved in data collection and analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. CH was involved in funding acquisition, study conception and design, co-led data collection, was involved in data analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. Author Note Results were initially presented, in a preliminary format, at the European Geriatric Medicine conference in London in 2022. The study was preregistered and available here: https://hrbopenresearch.org/articles/3-60 References Perlman, D., & Peplau, L. A. (1981). Toward a social psychology of loneliness. Personal relationships, 3, 31-56. Baarck, J., & Kovacic, M. (2022). The relationship between loneliness and health. Hawkley, L. C., Buecker, S., Kaiser, T., & Luhmann, M. (2022). Loneliness from young adulthood to old age: Explaining age differences in loneliness. International journal of behavioral development, 46(1), 39-49. Beckers, A., Bücker, S., Casabianca, E., & Nurminen, M. (2022). Effectiveness of interventions tackling loneliness. Cacioppo, J. T., & Hawkley, L. C. (2009). Perceived social isolation and cognition. Trends in Cognitive Sciences, 13(10), 447-454. Peplau, L. A., & Perlman, D. (1982). Loneliness: A sourcebook of curent theory, research, and therapy. New York: Wiley. Gardiner, C., Geldenhuys, G., & Gott, M. (2016). Interventions to reduce social isolation and loneliness among older people: an integrative review. Health & social care in the community. Preston, C., & Moore, S. (2019). Ringing the changes: the role of telephone communication in a helpline and befriending service targeting loneliness in older people. Ageing & Society, 39(7), 1528-1551. Lee, C., Kuhn, I., McGrath, M., Remes, O., Cowan, A., Duncan, F., Baskin, C., Oliver, E. J., Osborn, D. P., & Dykxhoorn, J. (2022). A systematic scoping review of community‐based interventions for the prevention of mental ill‐health and the promotion of mental health in older adults in the UK. Health & Social Care in the Community, 30(1), 27-57. Siette, J., Cassidy, M., & Priebe, S. (2017). Effectiveness of befriending interventions: A systematic review and meta-analysis. BMJ Open, 7, e014304. Cohen, S. (1988). Psychosocial models of the role of social support in the etiology of physical disease. Health Psychol, 7(3), 269-297. Brown, E. G., Gallagher, S., & Creaven, A. M. (2018). Loneliness and acute stress reactivity: A systematic review of psychophysiological studies. Psychophysiology, 55(5), e13031. Zawadzki, M. J., & Gavrilova, L. (2021). All the lonely people: Comparing the effects of loneliness as a social stressor to non-lonely stress on blood pressure recovery. International Journal of Psychophysiology, 167, 94-101. Freak-Poli, R., Ryan, J., Tran, T., Owen, A., McHugh Power, J., Berk, M., Stocks, N., Gonzalez-Chica, D., Lowthian, J. A., & Fisher, J. (2022). Social isolation, social support and loneliness as independent concepts, and their relationship with health-related quality of life among older women. Aging & Mental Health, 26(7), 1335-1344. McDonald, S., Vieira, R., & Johnston, D. W. (2020). Analysing N-of-1 observational data in health psychology and behavioural medicine: a 10-step SPSS tutorial for beginners. Health psychology and behavioral medicine, 8(1), 32-54. Krasny-Pacini, A., & Evans, J. (2018). Single-case experimental designs to assess intervention effectiveness in rehabilitation: A practical guide. Annals of physical and rehabilitation medicine, 61(3), 164-179. Dallery, J., & Raiff, B. R. (2014). Optimizing behavioral health interventions with single-case designs: from development to dissemination. Translational behavioral medicine, 4(3), 290-303. Sullivan, K., Shadish, W., & Steiner, P. (2015). An introduction to modeling longitudinal data with generalized additive models: Applications to single-case designs (Vol. 20). American Psychological Association. Graham, J. E., Karmarkar, A. M., & Ottenbacher, K. J. (2012). Small sample research designs for evidence-based rehabilitation: issues and methods. Archives of physical medicine and rehabilitation, 93(8), S111-S116. Campbell, M., Katikireddi, S. V., Hoffmann, T., Armstrong, R., Waters, E., & Craig, P. (2018). TIDieR-PHP: a reporting guideline for population health and policy interventions. Bmj, 361. Tate, R. L., Perdices, M., Rosenkoetter, U., Wakim, D., Godbee, K., Togher, L., & McDonald, S. (2013). Revision of a method quality rating scale for single-case experimental designs and n-of-1 trials: The 15-item Risk of Bias in N-of-1 Trials (RoBiNT) Scale. Neuropsychological rehabilitation, 23(5), 619-638. Kratochwill, T. R., Hitchcock, J., Horner, R., Levin, J. R., Odom, S., Rindskopf, D., & Shadish, W. (2010). Single-case designs technical documentation. What works clearinghouse. Kazdin, A. E. (2020). Single-case experimental designs: Characteristics, changes, and challenges. Journal of the Experimental Analysis of Behaviour, 115(1), 56-85. McDonald, S., Quinn, F., Vieira, R., O’Brien, N., White, M., Johnston, D. W., & Sniehotta, F. F. (2017). The state of the art and future opportunities for using longitudinal n-of-1 methods in health behaviour research: a systematic literature overview. Health Psychology Review, 11(4), 307-323. Burke, L. (2015). ALONE Befriending Service Evaluation Report. Dublin, Ireland: Liz Burke Communications. Kratochwill, T. R., Horner, R. H., Levin, J. R., Machalicek, W., Ferron, J., & Johnson, A. (2021). Single-case design standards: An update and proposed upgrades. Journal of School Psychology, 89, 91-105. Kratochwill, T. R., Horner, R. H., Levin, J. R., Machalicek, W., Ferron, J., & Johnson, A. (2023). Single-case intervention research design standards: Additional proposed upgrades and future directions. Journal of school psychology, 97, 192-216. Central Statistics Office. (2017). Population 2017 - Chapter 7: Ethnicity and Irish Travellers. Dublin, Ireland: Central Statistics Office. Rabin, R., Gudex, C., Selai, C., & Herdman, M. (2014). From translation to version management: a history and review of methods for the cultural adaptation of the EuroQol five-dimensional questionnaire. Value in Health, 17(1), 70-76. Feng, Y.-S., Kohlmann, T., Janssen, M. F., & Buchholz, I. (2021). Psychometric properties of the EQ-5D-5L: a systematic review of the literature. Quality of Life Research, 30(3), 647-673. Hobbins, A., Barry, L., Kelleher, D., Shah, K., Devlin, N., Goni, J. M. R., & O’Neill, C. (2018). Utility values for health states in Ireland: a value set for the EQ-5D-5L. Pharmacoeconomics, 36(11), 1345-1353. Walters, S. J., & Brazier, J. E. (2005). Comparison of the minimally important difference for two health state utility measures: EQ-5D and SF-6D. Quality of life research, 14(6), 1523-1532. McCaffrey, N., Kaambwa, B., Currow, D. C., & Ratcliffe, J. (2016). Health-related quality of life measured using the EQ-5D–5L: South Australian population norms. Health and quality of life outcomes, 14(1), 1-12. Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). A short scale for measuring loneliness in large surveys results from two population-based studies. Research on aging, 26(6), 655-672. Wood, S. (2006). Generalized additive models: An introduction with R. Boca Raton, FL.: Chapman & Hall/CRC. What Works Clearinghouse. (2017). What Works Clearinghouse Standards Handbook V4, Appendix A: Pilot Single-case Design Standards. USA: US Department of Education. Wood, S. (2018). Mixed GAM computation vehicle with GCV/AIC/REML smoothness estimation and GAMMs by REML/PQL. R package version, 1.8-23. Shadish, W. R., Zuur, A. F., & Sullivan, K. J. (2014). Using generalized additive (mixed) models to analyze single case designs. Journal of school psychology, 52(2), 149-178. Cohen, S., & McKay, G. (1984). Social support, stress, and the buffering hypothesis: A theoretical analysis. In A. Baum, J. E. Singer & S. E. Taylor (Eds.), Handbook of Psychology & Health. Hillsdale, New Jersey: Erlsbaum. Kwasnicka, D., & Naughton, F. (2020). N-of-1 methods: A practical guide to exploring trajectories of behaviour change and designing precision behaviour change interventions. Psychology of Sport and Exercise, 47, 101570. Harris, K., Stevenson, N., & Kauffman, J. (2019). CEC Division for Research position statement: Negative effects of minimum requirements for data points in multiple baseline designs and multiple probe designs in the What Works Clearinghouse standards handbook, version 4.0. CEC Division for Research. Manolov, R., Gast, D. L., Perdices, M., & Evans, J. J. (2014). Single-case experimental designs: Reflections on conduct and analysis. Neuropsychological rehabilitation, 24(3-4), 634-660. Laermans, J., Scheers, H., Vandekerckhove, P., & De Buck, E. (2023). Friendly visiting by a volunteer for reducing loneliness or social isolation in older adults: A systematic review. Campbell Systematic Reviews, 19(4). Ledford, J., Zimmerman, K., Schwartz, I., & Odom, S. (2018). Guide for the use of single case design research evidence. Division for Research of the Council for Exceptional Children. Shadish, W. R., & Sullivan, K. J. (2011). Characteristics of single-case designs used to assess intervention effects in 2008. Behavior Research Methods, 43(4), 971-980. Kazdin, A. E. (1978). Methodological and interpretive problems of single-case experimental designs. Journal of Consulting and Clinical Psychology, 46(4), 629. Footnotes Initially the aim of the study was to evaluate the impact of the befriending intervention on both HR-QoL and cognitive function as per our protocol (information removed for review) but no effects were identified for cognitive function, so for brevity we exclude reporting of this outcome in this manuscript. Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryMaterialsJuly2024.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4807512","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":332139688,"identity":"369b5471-e165-4269-8a34-75b574c66bb5","order_by":0,"name":"Joanna McHugh Power","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDUlEQVRIiWNgGAWjYBAC9gYIndggkcDAzPgHxGY+ACQkcGphhGophmhpA7HZEojSUo+khccAr8MY25ufPS6oYcjtn5Fj/LqwzSaxX/rM1808DBZyOLX0HDM3nnGMIXfG/Tdm1jP/pCXO7MvddpuHQcIYp5YZCWbSPGwMuRskcsyMeRsO5244wwvWktiAU0v6N2mefwyJcC37z/A8A2mpx6VFcEaOmTRvG1iL8WOwLTw8bCAtCbgcJs1zpkyat08iccaNtDJm3oa0+hln2MxuzjGQMMRlCx97+zZpnm/AgJqRvPkzb4ONMX8P87Mbbyrq5HHZAgXgiGNDij78UQMHzB+IUzcKRsEoGAUjDQAAyuFWCn355pkAAAAASUVORK5CYII=","orcid":"","institution":"Maynooth University","correspondingAuthor":true,"prefix":"","firstName":"Joanna","middleName":"McHugh","lastName":"Power","suffix":""},{"id":332139689,"identity":"13b9727e-b735-4403-b0ec-f90175c6bcca","order_by":1,"name":"Eimile Holton","email":"","orcid":"","institution":"Maynooth University","correspondingAuthor":false,"prefix":"","firstName":"Eimile","middleName":"","lastName":"Holton","suffix":""},{"id":332139690,"identity":"36a15635-55e4-40d2-8f6f-d674342e1259","order_by":2,"name":"Brian Lawlor","email":"","orcid":"","institution":"Trinity College Dublin","correspondingAuthor":false,"prefix":"","firstName":"Brian","middleName":"","lastName":"Lawlor","suffix":""},{"id":332139691,"identity":"20007a9b-c808-45a3-8fc8-90796e9f54f2","order_by":3,"name":"Frank Kee","email":"","orcid":"","institution":"Queen's University Belfast","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Kee","suffix":""},{"id":332139692,"identity":"243b00c7-9eb6-4f89-8d65-f6804b6c5407","order_by":4,"name":"Thomas Scharf","email":"","orcid":"","institution":"Newcastle University","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Scharf","suffix":""},{"id":332139693,"identity":"31c11943-ba3d-427b-921a-403fb2095d9c","order_by":5,"name":"Michelle Kelly","email":"","orcid":"","institution":"National College of Ireland","correspondingAuthor":false,"prefix":"","firstName":"Michelle","middleName":"","lastName":"Kelly","suffix":""},{"id":332139694,"identity":"c941a7e8-442c-4c5d-8a00-5c00348ce9ab","order_by":6,"name":"Caoimhe Hannigan","email":"","orcid":"","institution":"National College of Ireland","correspondingAuthor":false,"prefix":"","firstName":"Caoimhe","middleName":"","lastName":"Hannigan","suffix":""}],"badges":[],"createdAt":"2024-07-26 10:55:12","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4807512/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4807512/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":61810742,"identity":"274b7bb8-168d-4024-ba06-0686632815ec","added_by":"auto","created_at":"2024-08-05 20:22:06","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":62756,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eFlowchart of Sample Recruitment (Registration continued until n = 86 participants were registered onto the study.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4807512/v1/5873ab66d506aafe78ec3a19.png"},{"id":61810743,"identity":"9b81624c-549b-411b-8ca0-7e370afa598d","added_by":"auto","created_at":"2024-08-05 20:22:06","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":37339,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eInteraction between Time and Phase for Disutility Scores over time (from M3). In the Baseline Phase (a), disutility scores improve slightly until timepoint 6, then they start to increase, i.e. health gets worse. In the Treatment Phase (b), overall, there is a downward trend (i.e. health gets better) in disutility scores. This indicates a therapeutic effect of the befriending intervention on disutility scores.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4807512/v1/082abf1b8df59b4707226ba4.png"},{"id":61811997,"identity":"2cfd7210-dbd5-4268-9732-fa4f75ca070a","added_by":"auto","created_at":"2024-08-05 20:38:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":139369,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eContour plots of the interaction between Time and Loneliness on disutility scores in the a) Baseline and b)Treatment Phase. Lines connect areas of equal disutility scores over time, and yellow/areas of lower colour intensity indicate higher disutility scores (worse health), while red/areas of higher intensity indicates lower disutility scores (better health).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4807512/v1/f8c387deb158e3bd3899db18.png"},{"id":61810744,"identity":"6e2e4df8-ffa5-450a-a7df-07f9bd7de083","added_by":"auto","created_at":"2024-08-05 20:22:06","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":26803,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eLetter fluency over time during a) baseline phase and b) treatment phase.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4807512/v1/84513534b0cf42f56a6fa05e.png"},{"id":61811998,"identity":"271162a8-dfc5-4ad5-bf3b-69ecc4fdd961","added_by":"auto","created_at":"2024-08-05 20:38:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":658700,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4807512/v1/6f2d4231-ea78-484a-a30b-dacc3967dec2.pdf"},{"id":61810747,"identity":"c3a48b60-b916-42e9-a594-b30e31a98a69","added_by":"auto","created_at":"2024-08-05 20:22:07","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":365218,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialsJuly2024.docx","url":"https://assets-eu.researchsquare.com/files/rs-4807512/v1/77c418d4e2930236f5bb8cb4.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eThe Impact of a Befriending Service on Health-related Quality of Life in Older Adults: An Interventional N-of-1 Pilot Study.\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLoneliness is a painful insufficiency in the quality or quantity of social relationships [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The link between loneliness and health among older adults is well established [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. This link is concerning given the high levels of loneliness observed among older populations [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. As interventions on loneliness typically yield small effects [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] research prioritising effective intervention and evaluation is warranted.\u003c/p\u003e \u003cp\u003eOne intervention often offered to alleviate loneliness is befriending. Befriending interventions are designed under the assumption that providing companionship and social support alleviates loneliness, as per major theoretical approaches [5; 6]. Befriending interventions do not reliably reduce loneliness [7; 8] but they may improve older adult health [9; 10].\u003c/p\u003e \u003cp\u003eBefriending interventions may indirectly affect health by buffering the negative impact of loneliness. We invoke the stress-buffering model as comparison [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] \u0026ndash;in which social support \u0026ldquo;buffers\u0026rdquo; the individual from the negative impact of stress on health. Social support provided in a befriending intervention may buffer the individual from the impact of loneliness on health, since loneliness is a stressor [12; 13]. The aim of this study was to test whether a befriending intervention improved health outcomes of older adults, either directly and/or by reducing the impact of loneliness on health.\u003c/p\u003e \u003cp\u003eWe focused on health-related quality of life (HR-QoL). Although limited evidence exists for a therapeutic effect of befriending interventions on HR-QoL [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] loneliness is associated with HR-QoL [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWhen evaluating a community service, feasible research design is an important consideration. Control groups are inappropriate when individuals are seeking a community service. N-of-1 designs (single case designs, SCDs) offer practical, ethical, person-centred alternatives to between-group designs [15; 16], and are well-established in health literature [17; 18]. Interventional N-of-1s include baseline observations and systematic manipulations of an independent variable, and sample sizes tend to be small (\u0026gt;\u0026thinsp;10), with use of visual analysis common [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. We hypothesised that a befriending intervention would a) improve HR-QoL among older adults who experienced loneliness, and b) buffer the negative impact of loneliness on HR-QoL.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eDesign\u003c/h2\u003e \u003cp\u003eWe followed TIDieR and RoBiNT guidelines [20; 21], and What Works Clearinghouse (WWC) documentation [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] in reporting our interventional N-of-1 AB design, with outcomes measured during baseline (A-phase) and intervention (B-phase). This approach is suitable for community settings where flexibility is required [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] and there is an ethical issue involved in using a control group or wait-listing service seekers (i.e. in the context of a befriending service, which has known benefits). The sample were new seekers of an existing community-based befriending service which matches older adults to volunteers based on availability, resulting in varying baseline lengths. The service provider agreed to allow two timepoints to go by before participants were matched with a befriender (thus marking the beginning of the intervention, which would then run weekly for the duration of the data collection and beyond, at the preference of the participant). Because matching occurred based on availability, baseline length varied across participants with no systematic manipulation of intervention start-points (AB designs are therefore considered pre-experimental; [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. The study was approved by the local faculty Research Ethics Committee. All participants provided informed written consent.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eIntervention\u003c/h2\u003e \u003cp\u003eThe evaluated intervention is the (removed for review) volunteer-led befriending service in Ireland, which has been run by (removed for review), a non-governmental organisation, since 2009. The stated aim of the service is to alleviate the negative impact that loneliness has on health [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Older adults can self-refer or be referred by others (typically healthcare professionals). A volunteer is paired with a service user to provide weekly social contact until either party discontinues it, or until the service user requests a new match. There is no guidance on the content of the befriending visit \u0026ndash; service user and volunteer pairs are free to engage how they wish, and volunteers are advised to visit for one hour. Volunteers receive regular phone calls from (removed for review) to catch issues such as disengagement, issues in the partnership, or concerns about the service user\u0026rsquo;s welfare.\u003c/p\u003e \u003cp\u003eParticipants entered the study at least two weeks before their intervention. During the intervention phase, volunteer visits took place on a weekly basis. Since this is a voluntary, community-based intervention, no regular checks of procedural fidelity or adherence were completed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eRecruitment\u003c/h2\u003e \u003cp\u003e Participants were new users of the befriending service who responded to the screening question \u0026ldquo;do you feel lonely\u0026rdquo; with either \u0026ldquo;all the time\u0026rdquo; or \u0026ldquo;at least some of the time\u0026rdquo;. Users were informed that the study was an evaluation of the effect of the befriending service on HR-QoL\u003ca class=\"FNLink\" href=\"#Fn1\" id=\"#FNLinkFn1\"\u003e\u003c/a\u003e, and that their decision to engage had no bearing on their service receipt. Interested users were contacted by the research team, who established eligibility, sent information by post, and followed up after one week.\u003c/p\u003e \u003cp\u003eEligibility criteria were: aged 60+, ability to provide informed consent, and living within the greater Dublin area. Exclusion criteria were: self-reported receipt of a diagnosis by a medical doctor of: intellectual disability, psychotic disorder, dementia or serious memory impairment. Participants were made aware that the terms of the study meant a delay of two weeks in receiving a befriender match (although this was not, in practice, different to the length of time normally taken).\u003c/p\u003e \u003cp\u003eIn total, 246 individuals were referred to the research team as potential participants between September 2018 and June 2021, of a possible 2,485 individuals who approached or were referred for befriending. Recruitment challenges included service staff buy-in, research staff illness (which resulted in participants not being assessed prior to being matched to a befriender, thus yielding many participants who were \u0026ldquo;matched too soon\u0026rdquo; in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), and the COVID-19 pandemic, which prevented recruitment for a period of three months to allow service staff to focus on existing service users.\u003c/p\u003e \u003cp\u003eOf these 246 individuals, 76 declined (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), 23 were ineligible (20 could not provide informed consent, two were in an ineligible location, one was too young). 147 participants gave consent and were registered to the study. Of these, 21 dropped out, two died, 10 discontinued the befriending intervention, and 61 were matched to a befriender before two weeks had passed. 86 participants in total were registered. 53 participants completed the study, but because 20 participants were not matched to a befriender during the study, full datasets with baseline and intervention data suitable for inferential analyses were only obtained from n\u0026thinsp;=\u0026thinsp;33 participants (\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;2 baseline data-points). Visual analysis was also conducted from a subset (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;14) who met WWC design standards (see Supplementary Materials ([22; 26].\u003c/p\u003e \u003cp\u003eFor the inferential analysis, an a priori sample size requirement of 85 participants was calculated (information removed for review) based on 13 observations per participant, once fortnightly for 26 weeks (power in N-of-1s is a function of number of observations and of participants; [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. While the dataset of 33 complete cases fell short of this, the data for visual analysis (n\u0026thinsp;=\u0026thinsp;14) exceeded the typical 2\u0026ndash;3 participants in an N-of-1 [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Given the complexity of the study design, the ultimate analyses are a compromise between the feasibility of the research and the standards outlined [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] and are best considered as a pilot study.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eSample Characteristics\u003c/h2\u003e \u003cp\u003eThe mean age of participants was 75 (48.7% male; See Supplementary Materials). 83% of the sample lived alone (compared with 27% of those aged 65\u0026thinsp;+\u0026thinsp;in the general population, [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]). We compared HR-QoL outcomes to Irish norms (from participants aged 65 and over; Irish EQ-5D-5L survey 2015\u0026ndash;2016; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003ca href=\"https://hrbopenresearch.org/articles/3-60\" target=\"_blank\"\u003ewww.ucd.ie/issda\u003c/a\u003e\u003c/span\u003e\u003cspan address=\"http://www.ucd.ie/issda\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). 30% of our sample had no problems with mobility relative to 53% in the normative sample. For self-care, 61% of our sample (compared to 85% of the normative sample) had no issues. For engaging in usual activities, 38% had no issues with engaging in usual activities, relative to 68% of the normative sample. 31% of our sample had no issues with pain, relative to 45% of the normative sample; 31% had no issues with anxiety or depression in our sample compared to 82% of the normative sample. Overall, despite our sample having a lower age threshold than the normative data sample, moderate and severe issues in HR-QoL domains were more frequent.\u003c/p\u003e \u003cp\u003eA logistic regression was conducted to predict attrition. This analysis showed that those with a lower level of education (OR\u0026thinsp;=\u0026thinsp;0.85, CI\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;.76, \u0026minus;\u0026thinsp;.95), and those who were separated (OR\u0026thinsp;=\u0026thinsp;.17, CI\u0026thinsp;=\u0026thinsp;.04, .72), single (OR\u0026thinsp;=\u0026thinsp;.27, CI\u0026thinsp;=\u0026thinsp;.09, .79), or widowed (OR\u0026thinsp;=\u0026thinsp;.31, CI\u0026thinsp;=\u0026thinsp;.14, .71) relative to divorced were less likely to be lost to attrition.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e*FIGURE \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e HERE*\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section4\"\u003e \u003ch2\u003eProcedure and Setting\u003c/h2\u003e \u003cp\u003eThe first data collection point was performed at a home visit, but all data collection switched to phone after the beginning of the COVID-19 pandemic (March 2020). Data collection was scheduled for all participants every fortnight for 26 weeks (yielding max. 13 timepoints). Written consent was returned by post. Data collection was conducted by trained research assistants, who used a script. If participants could not be reached, two more attempts were made to contact them, before the datapoint was marked as \u0026ldquo;missed\u0026rdquo;.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eMeasures\u003c/h2\u003e \u003cp\u003eHR-QoL\u003c/p\u003e \u003cp\u003eThe EQ5D-5L was used to evaluate HR-QoL [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. This index is responsive to changes in health over time, correlates strongly with global health measures and measures of functioning, and has excellent test-retest reliability [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Utility scores can be calculated for five domains of health (see Supplementary Materials). Age norms for Ireland were used to convert responses into utility scores [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. A utility score of 1 indicates full health with a score of 0 meaning a state as bad as being dead, with negative values possible. A minimally important difference of .074 was identified [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. The utility scores were converted to a disutility index in order to reduce problematic skewness and negative values [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eLoneliness was measured using the three-item UCLA Loneliness scale [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e] which is well validated in older populations. Baseline covariates age, gender, years in education, living status (living alone or with others) were also recorded.\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003eAnalytic Plan\u003c/h2\u003e \u003cp\u003eGeneralised additive modelling (GAM) was used to evaluate trends in outcomes over time for each participant, controlling for covariates. GAM is a semi-parametric regression [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] where the data inform the researcher about the trend, as opposed to other single-case analytic methods that assume linear or no trend [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The \u0026ldquo;wiggliness\u0026rdquo; of the trend is measured using estimated degrees of freedom or \u0026ldquo;edf\u0026rdquo;; while values of 1 indicate a linear trend, higher values indicate non-linearity. Visual analysis was also conducted (see Supplementary Materials) following guidelines [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. We used the package \u0026ldquo;mgcv\u0026rdquo; in R Studio for conducting the analyses [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. The hypotheses (and implied models) below were preregistered ((information removed for review)). Data are not available for further analysis due to constraints in the ethical approval of this study.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cstrong\u003eHypothesis 1\u003c/strong\u003e \u003cp\u003e \u003cem\u003eThe befriending service has a therapeutic (positive) impact on HR-QoL.\u003c/em\u003e \u003c/p\u003e \u003c/p\u003e \u003cp\u003eWe followed the steps outlined [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e] in applying GAM to single-case design data from the 33 participants with sufficient data (using a robust maximum likelihood estimator given the small sample size). A Gaussian distribution with a logarithmic link was chosen. First, a model was created with Time (1\u0026ndash;13), Phase (0\u0026thinsp;=\u0026thinsp;baseline, 1\u0026thinsp;=\u0026thinsp;treatment), and Case (individual ID) as predictors (M0). Then a smoothing term on Time was added (M1), which marginally improved model fit (AIC = -104 to -101). Smoothing terms allow one to define a trend as non-linear, but without interpolation, yielding a line that fits the data well. A third model (M2) was run without the case predictor, which disimproved model fit (AIC\u0026thinsp;=\u0026thinsp;219) so the case predictor was retained in M3, which also specified that Time terms varied by Phase, in effect evaluating an interaction between Time and Phase which could identify whether the befriending intervention exerted a therapeutic effect on disutility scores. M3 had an improved fit (AIC = -126; adjusted R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;.724). Disutility scores increased albeit nonlinearly (i.e. health got worse) in the baseline phase over time (edf\u0026thinsp;=\u0026thinsp;3.893, F\u0026thinsp;=\u0026thinsp;3., p\u0026thinsp;=\u0026thinsp;0.002; see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In the treatment phase, disutility scores decreased (i.e. health got better), and were highly nonlinear (edf\u0026thinsp;=\u0026thinsp;5.98, F\u0026thinsp;=\u0026thinsp;2.98, p\u0026thinsp;=\u0026thinsp;.008). Result patterns indicated that while health declined in the baseline phase, the befriending intervention could mitigate decline. When covariates were added, fit improved again (AIC = -128, adjusted R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;.73). Of the covariates, only living status had an impact (F\u0026thinsp;=\u0026thinsp;3.6, p\u0026thinsp;=\u0026thinsp;.02) such that those living alone had poorer HR-QoL than those living with spouses.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e*FIGURE \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e HERE*\u003c/h2\u003e \u003cp\u003eVisual analyses [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] were conducted (See Supplementary Materials). Based on these analyses, we concluded that the intervention had therapeutic effects on disutility scores, since for 5/14 participants, the minimally important difference of 0.074 was met (although for other participants there was decline). For 6/14 participants, an effect took place within three timepoints, indicating that the effect of befriending on HR-QoL occurs quickly.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis 2\u003c/strong\u003e \u003cp\u003e \u003cem\u003eThe befriending service mitigates (buffers) the negative impact of loneliness on disutility scores.\u003c/em\u003e \u003c/p\u003e \u003c/p\u003e \u003cp\u003eModel M4 modelled the differential association between loneliness and disutility scores over time, to gauge whether the befriending intervention reduced the magnitude of this association. We used a tensor product smoother specifying an interaction between (smoothed) time and (smoothed) loneliness (the \u0026ldquo;ti\u0026rdquo; function in mgcv), specified separately for each intervention phase (Baseline, Treatment). Covariates and main effects were included. Model fit improved (AIC = -161; Adjusted R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;.77). There were significant interactions in the baseline (edf\u0026thinsp;=\u0026thinsp;9.7, F\u0026thinsp;=\u0026thinsp;2.88, p\u0026thinsp;\u0026lt;\u0026thinsp;.001) and treatment phase (edf\u0026thinsp;=\u0026thinsp;9.48, F\u0026thinsp;=\u0026thinsp;5.50, p\u0026thinsp;\u0026lt;\u0026thinsp;.001) - in both phases, there was an association between loneliness and disutility scores over time. Being male was associated with lower disutility scores (i.e., better health; F\u0026thinsp;=\u0026thinsp;8.48, p\u0026thinsp;=\u0026thinsp;.003), as was living alone (F\u0026thinsp;=\u0026thinsp;9.29, p\u0026thinsp;\u0026lt;\u0026thinsp;.001).\u003c/p\u003e \u003cp\u003eTo visualise these effects, we used contour plots (in the \u0026ldquo;vis.gam\u0026rdquo; function of mgcv) such that the \u0026lsquo;Time by Loneliness\u0026rsquo; interaction on disutility score outcomes was plotted separately for each phase (see Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Hereafter low disutility scores are described as good health, and high disutility scores as poor health. Red areas indicate better health, while the yellow areas indicate worse health. In both contour plots, health was worst at higher initial levels of loneliness. Health declined over time in the baseline phase, and for participants with high levels of loneliness, health was poor and stayed poor throughout the study. However, in the treatment phase, health was stable for longer (getting worse at the end of the study, but this happened for all participants). As such, the intervention reduced the impact that loneliness had on health over time. For participants who had lower levels of loneliness at the study onset, health worsened over time in the baseline condition, but did not get worse over time in the treatment condition until the end \u0026ndash; that is, for people with low and moderate levels of loneliness, the intervention prevented decline in HR-QoL over time.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e*FIGURE \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e HERE*\u003c/h2\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe explored whether a befriending intervention could improve HR-QoL in a sample of older adults and reduce the impact of loneliness on HR-QoL. While HR-QoL generally deteriorated over six months, as could be expected for a group of older adults with worse than usual functioning at baseline, it deteriorated less after the onset of the befriending intervention. Furthermore, while loneliness was associated with worsening HR-QoL over time, this association was buffered by the befriending intervention. Thus our findings are compatible with prior literature on the buffering effect [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], and with literature suggesting that loneliness itself is a stressor [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAlthough high in ecological validity, conducting an evaluation of a community-based intervention was challenging. We hope that sharing these insights may assist future researchers who are planning similar evaluations. To begin, we experienced considerable recruitment issues. When participants were recruited, our design requirements were sometimes at odds with the usual procedures of the service providers. A considerable proportion of participants were matched with a befriender too soon, which meant that they were not eligible for study registration.\u003c/p\u003e \u003cp\u003eBecause of problematic missingness and issues with matching, only 33/86 participants registered to the study had sufficient data for inferential analysis, meaning that our results are best interpreted as a pilot study. That said, our sample size is large for an interventional N-of-1. Future researchers should consider the feasibility of an interventional or experimental design and ensure that community partners or service providers are clear on the requirements for that design (see [24; 40]).\u003c/p\u003e \u003cp\u003eWhen planning the study, our intention was to use an experimental multiple baseline design (MBD) with baseline (A) and intervention (B) phases, but the MBD would require planned sequential introduction of the intervention across participants (see [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]). In reality, matching occurred based on availability of befrienders, so ultimately we could not establish the experimental control required for an MBD. This rendered our design a pre-experimental AB design. While this is not uncommon in the N-of-1 literature [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], further research is required to determine if the effects of befriending observed can be replicated in more controlled design contexts.\u003c/p\u003e \u003cp\u003e Public health guidelines also presented challenges for evaluation, since visits were paused during COVID-19 lockdowns, and contact was maintained by phone for most participants and their volunteers. The methodological constraints we experienced present an interesting test case for conducting N-of-1/SCD research in community settings, highlighting the necessity for guidelines/standards that recognise this reality [26; 41] and provide guidance on how difficulties may be mitigated [18; 42]. For instance, given the common methodological issues plaguing befriending service evaluation [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e], such as variability in the content and dose of the befriending intervention; confounding; participant selection bias, it seems clear that designing a community-compatible process evaluation to record the implementation of befriending interventions is critical.\u003c/p\u003e \u003cp\u003eThe difficulties described above posed additional challenges to methodological rigour in our attempts to meet the WWC standards for quality in N-of-1/SCDs. For 19/33 participants who completed the study as planned, two baseline data points were present rather than the recommended three-to-five [21; 22; 44]. This difficulty is faced by many in applied or community settings. In one systematic review analysing the characteristics of SCDs, 10.5% of included studies (n\u0026thinsp;=\u0026thinsp;61) had one or two points in the first baseline phase [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. As suggested [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e], the negative impact of overly stringent minimum requirements should be considered across a multitude of settings. For settings where the minimum number of data-points cannot be achieved, statistics can be helpful in supporting a more comprehensive analysis [42; 46]. We used visual \u003cem\u003eand\u003c/em\u003e statistical analysis for data that met the standard of at least three data-points in each phase and used GAM where data did not meet those standards. When compared to visual analysis alone, this resulted in more nuanced information about the intervention effects.\u003c/p\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eImplications for Practice\u003c/h2\u003e \u003cp\u003eOverall, the current study\u0026rsquo;s results indicate that the (information removed for review) befriending service had a therapeutic impact on HR-QoL. Results have implications for service delivery. In 2015, an evaluation of the (information removed for review) befriending service showed that only 4% of total referrals to the service were made for reasons of poor physical health ((information removed for review)). Informing healthcare practitioners and other referrers of the positive impact of befriending on HR-QoL is critical so that older adults who stand to benefit from befriending are referred. Results have implications for the delivery of the ongoing befriending service by (information removed for review) and also for those designing or commissioning new befriending services.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOur results contribute to mixed existing findings concerning the positive impacts of befriending interventions. To date, little is known about the impact of befriending interventions on physical health: our study provides preliminary evidence that they may be therapeutic for HR-QoL scores, and may also reduce the negative association between loneliness and health. Further research is required to corroborate these pilot findings.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eStatement of Ethical Approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study received approval from the Trinity College Faculty of Health Sciences Ethics Board.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatement of Funding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Irish Health Research Board (grant number APA2017004). The research partner organisation ALONE supplied co-funding through this grant. The Health Research Board had no role in any aspect of the research. ALONE co-author SM reviewed the manuscript for critical content, and facilitated study recruitment.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e Thanks to the staff of ALONE, Trinity College Dublin, and Maynooth University who facilitated this research, and to all study participants involved. This research was funded by the Health Research Board under the award APA 2017 004.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of interest statement:\u003c/strong\u003e All authors confirm they have no conflict of interest to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of contribution of authors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJMcHP led funding acquisition, study conception and design, data collection, data analysis and interpretation, and article drafting, revision, and submission. EH was involved in acquisition of data, analysis of data, and in the writing, revision, and final approval of the article. BL was involved in funding acquisition, study conception and design, consulted on data collection, revised the article critically for important intellectual content, and approved the final version to be submitted. FK was involved in funding acquisition, study conception and design, consulted on data collection and data analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. \u0026nbsp;TS was involved in funding acquisition, study conception and design, consulted on data collection and analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. CW was involved in study conception and design, consulted on data analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. SM was involved in funding acquisition, data acquisition, revised the article critically for important intellectual content, and approved the final version to be submitted. ME was involved in data collection and analysis, revised the article critically for important intellectual content, and approved the final version to be submitted. CH was involved in funding acquisition, study conception and design, co-led data collection, was involved in data analysis, revised the article critically for important intellectual content, and approved the final version to be submitted.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Note\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResults were initially presented, in a preliminary format, at the European Geriatric Medicine conference in London in 2022. The study was preregistered and available here: https://hrbopenresearch.org/articles/3-60\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003ePerlman, D., \u0026amp; Peplau, L. A. (1981). Toward a social psychology of loneliness. Personal relationships, 3, 31-56.\u003c/li\u003e\n \u003cli\u003eBaarck, J., \u0026amp; Kovacic, M. (2022). The relationship between loneliness and health.\u003c/li\u003e\n \u003cli\u003eHawkley, L. C., Buecker, S., Kaiser, T., \u0026amp; Luhmann, M. (2022). Loneliness from young adulthood to old age: Explaining age differences in loneliness. International journal of behavioral development, 46(1), 39-49.\u003c/li\u003e\n \u003cli\u003eBeckers, A., B\u0026uuml;cker, S., Casabianca, E., \u0026amp; Nurminen, M. (2022). Effectiveness of interventions tackling loneliness.\u003c/li\u003e\n \u003cli\u003eCacioppo, J. T., \u0026amp; Hawkley, L. C. (2009). Perceived social isolation and cognition. Trends in Cognitive Sciences, 13(10), 447-454.\u003c/li\u003e\n \u003cli\u003ePeplau, L. A., \u0026amp; Perlman, D. (1982). Loneliness: A sourcebook of curent theory, research, and therapy. New York: Wiley.\u003c/li\u003e\n \u003cli\u003eGardiner, C., Geldenhuys, G., \u0026amp; Gott, M. (2016). Interventions to reduce social isolation and loneliness among older people: an integrative review. Health \u0026amp; social care in the community.\u003c/li\u003e\n \u003cli\u003ePreston, C., \u0026amp; Moore, S. (2019). Ringing the changes: the role of telephone communication in a helpline and befriending service targeting loneliness in older people. Ageing \u0026amp; Society, 39(7), 1528-1551.\u003c/li\u003e\n \u003cli\u003eLee, C., Kuhn, I., McGrath, M., Remes, O., Cowan, A., Duncan, F., Baskin, C., Oliver, E. J., Osborn, D. P., \u0026amp; Dykxhoorn, J. (2022). A systematic scoping review of community‐based interventions for the prevention of mental ill‐health and the promotion of mental health in older adults in the UK. Health \u0026amp; Social Care in the Community, 30(1), 27-57.\u003c/li\u003e\n \u003cli\u003eSiette, J., Cassidy, M., \u0026amp; Priebe, S. (2017). Effectiveness of befriending interventions: A systematic review and meta-analysis. BMJ Open, 7, e014304.\u003c/li\u003e\n \u003cli\u003eCohen, S. (1988). Psychosocial models of the role of social support in the etiology of physical disease. Health Psychol, 7(3), 269-297.\u003c/li\u003e\n \u003cli\u003eBrown, E. G., Gallagher, S., \u0026amp; Creaven, A. M. (2018). Loneliness and acute stress reactivity: A systematic review of psychophysiological studies. Psychophysiology, 55(5), e13031.\u003c/li\u003e\n \u003cli\u003eZawadzki, M. J., \u0026amp; Gavrilova, L. (2021). All the lonely people: Comparing the effects of loneliness as a social stressor to non-lonely stress on blood pressure recovery. International Journal of Psychophysiology, 167, 94-101.\u003c/li\u003e\n \u003cli\u003eFreak-Poli, R., Ryan, J., Tran, T., Owen, A., McHugh Power, J., Berk, M., Stocks, N., Gonzalez-Chica, D., Lowthian, J. A., \u0026amp; Fisher, J. (2022). Social isolation, social support and loneliness as independent concepts, and their relationship with health-related quality of life among older women. Aging \u0026amp; Mental Health, 26(7), 1335-1344.\u003c/li\u003e\n \u003cli\u003eMcDonald, S., Vieira, R., \u0026amp; Johnston, D. W. (2020). Analysing N-of-1 observational data in health psychology and behavioural medicine: a 10-step SPSS tutorial for beginners. Health psychology and behavioral medicine, 8(1), 32-54.\u003c/li\u003e\n \u003cli\u003eKrasny-Pacini, A., \u0026amp; Evans, J. (2018). Single-case experimental designs to assess intervention effectiveness in rehabilitation: A practical guide. Annals of physical and rehabilitation medicine, 61(3), 164-179.\u003c/li\u003e\n \u003cli\u003eDallery, J., \u0026amp; Raiff, B. R. (2014). Optimizing behavioral health interventions with single-case designs: from development to dissemination. Translational behavioral medicine, 4(3), 290-303.\u003c/li\u003e\n \u003cli\u003eSullivan, K., Shadish, W., \u0026amp; Steiner, P. (2015). An introduction to modeling longitudinal data with generalized additive models: Applications to single-case designs (Vol. 20). American Psychological Association.\u003c/li\u003e\n \u003cli\u003eGraham, J. E., Karmarkar, A. M., \u0026amp; Ottenbacher, K. J. (2012). Small sample research designs for evidence-based rehabilitation: issues and methods. Archives of physical medicine and rehabilitation, 93(8), S111-S116.\u003c/li\u003e\n \u003cli\u003eCampbell, M., Katikireddi, S. V., Hoffmann, T., Armstrong, R., Waters, E., \u0026amp; Craig, P. (2018). TIDieR-PHP: a reporting guideline for population health and policy interventions. Bmj, 361.\u003c/li\u003e\n \u003cli\u003eTate, R. L., Perdices, M., Rosenkoetter, U., Wakim, D., Godbee, K., Togher, L., \u0026amp; McDonald, S. (2013). Revision of a method quality rating scale for single-case experimental designs and n-of-1 trials: The 15-item Risk of Bias in N-of-1 Trials (RoBiNT) Scale. Neuropsychological rehabilitation, 23(5), 619-638.\u003c/li\u003e\n \u003cli\u003eKratochwill, T. R., Hitchcock, J., Horner, R., Levin, J. R., Odom, S., Rindskopf, D., \u0026amp; Shadish, W. (2010). Single-case designs technical documentation. What works clearinghouse.\u003c/li\u003e\n \u003cli\u003eKazdin, A. E. (2020). Single-case experimental designs: Characteristics, changes, and challenges. Journal of the Experimental Analysis of Behaviour, 115(1), 56-85.\u003c/li\u003e\n \u003cli\u003eMcDonald, S., Quinn, F., Vieira, R., O\u0026rsquo;Brien, N., White, M., Johnston, D. W., \u0026amp; Sniehotta, F. F. (2017). The state of the art and future opportunities for using longitudinal n-of-1 methods in health behaviour research: a systematic literature overview. Health Psychology Review, 11(4), 307-323.\u003c/li\u003e\n \u003cli\u003eBurke, L. (2015). ALONE Befriending Service Evaluation Report. Dublin, Ireland: Liz Burke Communications.\u003c/li\u003e\n \u003cli\u003eKratochwill, T. R., Horner, R. H., Levin, J. R., Machalicek, W., Ferron, J., \u0026amp; Johnson, A. (2021). Single-case design standards: An update and proposed upgrades. Journal of School Psychology, 89, 91-105.\u003c/li\u003e\n \u003cli\u003eKratochwill, T. R., Horner, R. H., Levin, J. R., Machalicek, W., Ferron, J., \u0026amp; Johnson, A. (2023). Single-case intervention research design standards: Additional proposed upgrades and future directions. Journal of school psychology, 97, 192-216.\u003c/li\u003e\n \u003cli\u003eCentral Statistics Office. (2017). Population 2017 - Chapter 7: Ethnicity and Irish Travellers. Dublin, Ireland: Central Statistics Office.\u003c/li\u003e\n \u003cli\u003eRabin, R., Gudex, C., Selai, C., \u0026amp; Herdman, M. (2014). From translation to version management: a history and review of methods for the cultural adaptation of the EuroQol five-dimensional questionnaire. Value in Health, 17(1), 70-76.\u003c/li\u003e\n \u003cli\u003eFeng, Y.-S., Kohlmann, T., Janssen, M. F., \u0026amp; Buchholz, I. (2021). Psychometric properties of the EQ-5D-5L: a systematic review of the literature. Quality of Life Research, 30(3), 647-673.\u003c/li\u003e\n \u003cli\u003eHobbins, A., Barry, L., Kelleher, D., Shah, K., Devlin, N., Goni, J. M. R., \u0026amp; O\u0026rsquo;Neill, C. (2018). Utility values for health states in Ireland: a value set for the EQ-5D-5L. Pharmacoeconomics, 36(11), 1345-1353.\u003c/li\u003e\n \u003cli\u003eWalters, S. J., \u0026amp; Brazier, J. E. (2005). Comparison of the minimally important difference for two health state utility measures: EQ-5D and SF-6D. Quality of life research, 14(6), 1523-1532.\u003c/li\u003e\n \u003cli\u003eMcCaffrey, N., Kaambwa, B., Currow, D. C., \u0026amp; Ratcliffe, J. (2016). Health-related quality of life measured using the EQ-5D\u0026ndash;5L: South Australian population norms. Health and quality of life outcomes, 14(1), 1-12.\u003c/li\u003e\n \u003cli\u003eHughes, M. E., Waite, L. J., Hawkley, L. C., \u0026amp; Cacioppo, J. T. (2004). A short scale for measuring loneliness in large surveys results from two population-based studies. Research on aging, 26(6), 655-672.\u003c/li\u003e\n \u003cli\u003eWood, S. (2006). Generalized additive models: An introduction with R. Boca Raton, FL.: Chapman \u0026amp; Hall/CRC.\u003c/li\u003e\n \u003cli\u003eWhat Works Clearinghouse. (2017). What Works Clearinghouse Standards Handbook V4, Appendix A: Pilot Single-case Design Standards. USA: US Department of Education.\u003c/li\u003e\n \u003cli\u003eWood, S. (2018). Mixed GAM computation vehicle with GCV/AIC/REML smoothness estimation and GAMMs by REML/PQL. R package version, 1.8-23.\u003c/li\u003e\n \u003cli\u003eShadish, W. R., Zuur, A. F., \u0026amp; Sullivan, K. J. (2014). Using generalized additive (mixed) models to analyze single case designs. Journal of school psychology, 52(2), 149-178.\u003c/li\u003e\n \u003cli\u003eCohen, S., \u0026amp; McKay, G. (1984). Social support, stress, and the buffering hypothesis: A theoretical analysis. In A. Baum, J. E. Singer \u0026amp; S. E. Taylor (Eds.), Handbook of Psychology \u0026amp; Health. Hillsdale, New Jersey: Erlsbaum.\u003c/li\u003e\n \u003cli\u003eKwasnicka, D., \u0026amp; Naughton, F. (2020). N-of-1 methods: A practical guide to exploring trajectories of behaviour change and designing precision behaviour change interventions. Psychology of Sport and Exercise, 47, 101570.\u003c/li\u003e\n \u003cli\u003eHarris, K., Stevenson, N., \u0026amp; Kauffman, J. (2019). CEC Division for Research position statement: Negative effects of minimum requirements for data points in multiple baseline designs and multiple probe designs in the What Works Clearinghouse standards handbook, version 4.0. CEC Division for Research.\u003c/li\u003e\n \u003cli\u003eManolov, R., Gast, D. L., Perdices, M., \u0026amp; Evans, J. J. (2014). Single-case experimental designs: Reflections on conduct and analysis. Neuropsychological rehabilitation, 24(3-4), 634-660.\u003c/li\u003e\n \u003cli\u003eLaermans, J., Scheers, H., Vandekerckhove, P., \u0026amp; De Buck, E. (2023). Friendly visiting by a volunteer for reducing loneliness or social isolation in older adults: A systematic review. Campbell Systematic Reviews, 19(4).\u003c/li\u003e\n \u003cli\u003eLedford, J., Zimmerman, K., Schwartz, I., \u0026amp; Odom, S. (2018). Guide for the use of single case design research evidence. Division for Research of the Council for Exceptional Children.\u003c/li\u003e\n \u003cli\u003eShadish, W. R., \u0026amp; Sullivan, K. J. (2011). Characteristics of single-case designs used to assess intervention effects in 2008. Behavior Research Methods, 43(4), 971-980.\u003c/li\u003e\n \u003cli\u003eKazdin, A. E. (1978). Methodological and interpretive problems of single-case experimental designs. Journal of Consulting and Clinical Psychology, 46(4), 629.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e Initially the aim of the study was to evaluate the impact of the befriending intervention on both HR-QoL and cognitive function as per our protocol (information removed for review) but no effects were identified for cognitive function, so for brevity we exclude reporting of this outcome in this manuscript.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"078e8e97-d6fe-4d78-95a7-cf5c8ab2d98f","identifier":"10.13039/100010414","name":"Health Research Board","awardNumber":"APA 2017 004","order_by":0}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Maynooth University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"intensive longitudinal, N-of-1, quality of life, loneliness","lastPublishedDoi":"10.21203/rs.3.rs-4807512/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4807512/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose:\u003c/strong\u003e Befriending interventions are unlikely to reduce loneliness, but they may provide social support which buffers the negative impact of loneliness on health outcomes of older adults. An interventional N-of-1 design was used to assess the impact of a befriending intervention on health-related quality of life (HR-QoL) among older adults, and whether such intervention attenuated the impact of loneliness on HR-QoL.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e Participants were \u003cem\u003en\u003c/em\u003e = 33 new users of the service, aged 60+. Outcomes were measured at 13 timepoints across 26 weeks, and data were analysed using generalised additive modelling (GAM) with a subset of data analysed using visual analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Results indicate that the befriending service may reduce decline of HR-QoL (i.e., health declined in the baseline phase over time: edf = 3.893, F = 3.0, p=0.002, while in the treatment phase, health remained more stable: edf = 5.98, F = 2.98, p=.008). The befriending intervention also suppressed the association between loneliness and HR-QoL. \u003cstrong\u003eConclusion: \u003c/strong\u003eWe supported our hypothesis, that befriending interventions may moderate the impact of loneliness on HR-QoL. Interventional N-of-1 designs however carry considerable recruitment and participant burden, which should be considered prior to onset. This research provides an insight into practical difficulties when evaluating existing community-based services, particularly in relation to adhering to best practice design guidelines.\u003c/p\u003e","manuscriptTitle":"The Impact of a Befriending Service on Health-related Quality of Life in Older Adults: An Interventional N-of-1 Pilot Study.","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-05 20:22:02","doi":"10.21203/rs.3.rs-4807512/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"962a16b6-013d-4b84-a53d-c6188282cc4d","owner":[],"postedDate":"August 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":35258022,"name":"Psychology"}],"tags":[],"updatedAt":"2024-08-05T20:22:02+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-05 20:22:02","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4807512","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4807512","identity":"rs-4807512","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}