López et al. BMC Geriatrics (2019) 19:44 https://doi.org/10.1186/s12877-019-1055-9 STUDY PROTOCOL Open Access Community intervention to reduce social isolation in older adults in disadvantaged urban areas: study protocol for a mixed methods multi-approach evaluation María José López1,2,3*, Carolina Lapena4,5,6, Alba Sánchez7, Xavier Continente1,2,3, Ana Fernández1 and the Evaluation Group of the “Schools of Health for older people” Abstract Background: The proportion of older people has dramatically increased in recent decades. Moreover, social and demographic trends show a global increase of older people at risk of loneliness and lack of social relationships. The objective of this study was to evaluate the process, the effectiveness and the cost-effectiveness of a planned 22 weekly group sessions called School of Health for Older People to reduce social isolation. Methods: This is a mixed methods multi-approach evaluation that includes: 1) A qualitative evaluation among coordinators and participants taking part in the intervention, through in depth-interviews and focus groups, respectively. The main topics covered will be positive and negative aspects of the intervention, suggestions for its improvement, opinions on different aspects of the intervention, and perceived benefits; 2) A quantitative quasi-experimental design, comparing a group of individuals taking part in the intervention with another group with similar characteristics not receiving the intervention. Data will be collected at the beginning and at the end of the intervention. Social support will be measured through questions drawn from the Medical Outcomes Study and the National Social Life, Health, and Aging Project questionnaires. Psychological morbidity will be measured through Goldberg’s General Health Questionnaire, and Health-related Quality of Life will be measured through the EuroQoL questionnaire. Information on visits to the primary care center in the years before and after the intervention will be obtained from the electronic records of the primary care centers; 3) A cost-utility analysis, which will be conducted from a health system (primary care) perspective, including direct costs of the program and the primary care health services used. The effects of the intervention will be measured on quality-adjusted life years. Discussion: There is an urgent need for studies assessing the effectiveness and the efficiency of potential interventions to reduce social isolation among older persons. The results of this study will help to fill the knowledge gap in this area and might be especially useful for the development of social and public health policies and programs for older people in disadvantaged neighborhoods in urban areas. Trial registration: NCT03142048 retrospectively registered (April 11, 2017). Keywords: Social isolation, Aging, Older people, Evaluation * Correspondence: mjlopez@aspb.cat 1 Public Health Agency of Barcelona, Lesseps, 1, 08023 Barcelona, Spain 2 Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Madrid, Spain Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. López et al. BMC Geriatrics (2019) 19:44 Page 2 of 8 Background multi-approach evaluation study designed to assess The proportion of older people has dramatically increased this intervention. in recent decades [1]. In the city of Barcelona, 21.2% of the population is 65 years or older, and projections indicate that this percentage will increase to 23.9% in 2031 [2]. Hypothesis Social and demographic trends show a global increase The intervention evaluated in this protocol will signifi- of older people at risk of loneliness and lack of social cantly improve social support, mental health, general relationships. This may be due to certain aging-related health status and distinct dimensions of quality of life in factors such as retirement or the loss of a partner or the intervention group (IG) compared with the compari- close friends [3]. Previous studies have found higher son group (CG). Among participants in the IG, the inter- rates of loneliness in deprived urban areas [4]. Further- vention will also significantly reduce the number of more, other studies have found that some of the va- visits to their primary care centers. Furthermore, this riables significantly associated with loneliness in older intervention will be cost effective. adults included poor income and lower educational level [5], and that living in a deprived area adds barriers to so- cial engagement [6]. Overall, recent studies confirm that Objectives rural residents reported less social isolation and more General objective social relationships than urban residents [7]. Therefore, To evaluate the process, effectiveness and cost-effectiveness disadvantaged urban areas need to be studied, and eva- of an intervention to reduce social isolation and its conse- luated interventions in these areas should be prioritized. quent negative impact on health of older people. Social relations are associated with good mental health, while their absence is linked to a significant increase in morbidity and mortality [8–10]. Because of its high preva- Specific objectives lence and the evidence of its impact on health and well- being, social isolation is an important public health issue. 1) To evaluate the implementation process of the Based on the aforementioned evidence, interventions intervention (participant profile, positive and are needed to reduce social isolation and its negative negative aspects of the intervention, barriers and effects on quality of life. Although preventing and ad- facilitators in its implementation, quality of the dressing social isolation in older people is a priority in intervention, and satisfaction). health policies, there is a clear lack of evidence on the 2) To assess the impact of the intervention on social effectiveness of health promotion activities in this field, support, self-perceived health status, mental health as well as wide heterogeneity in the interventions and and quality of life among participants, and visits to their quality [11–15].Therefore, studies evaluating the the primary care center. effectiveness of interventions designed to reduce the 3) To determine the cost-effectiveness of the impact of social isolation on wellbeing and quality of life intervention. in older people are much needed. Detailed protocols on how to evaluate this complex interventions can help to standardize the evaluation across interventions and Methods homogenize the evidence on this area. In this sense, this Study design protocol might be useful as a tool for future interven- This is a mixed methods multi-approach evaluation that tions to be evaluated. includes: Since 2008, an intervention called the “Schools of Health for Older People” has been implemented in some deprived – a qualitative evaluation among coordinators and neighborhoods of Barcelona. This intervention is part of participants who underwent the intervention the comprehensive action Health in the neighborhoods through in-depth interviews and focus groups, [16], focused on reducing health inequalities in the city of respectively; Barcelona by implementing community interventions – a quantitative quasi-experimental design, comparing in the most deprived neighborhoods. The main goal of a group of individuals taking part in the intervention the “Schools of Health for Older People” is to reduce (IG) with another group with similar characteristics social isolation in this collective, as well as its poten- not receiving the intervention (CG). Data will be tial harmful health effects. As shown in numerous collected at the beginning and at the end of the studies [17, 18], these effects can be especially impor- intervention; tant in low-income older people. The main objective – an economic evaluation, which includes a cost-utility of this protocol is to describe the mixed methods analysis. López et al. BMC Geriatrics (2019) 19:44 Page 3 of 8 Study setting and participants and recruited through primary care centers, social services Qualitative evaluation and civic centers for older people. Once individuals have The study population will include coordinators of the accepted to participate, there will be a pre-registration School of Health (community nurses from the Public process that will record their names and telephone num- Health Agency of Barcelona) and participants who will bers. Later, the participants will be contacted by telephone attend the School of Health in the intervention neigh- to arrange an appointment, in which the person will be borhoods. In-depth interviews will be carried out with informed individually about the study. After agreement, the coordinator of each IG, who will be selected through he/she will sign the informed consent form and complete intentional opinatic sampling (non-randomized sampling a questionnaire. in which the researcher selects a sample based on their To calculate the sample size, we took into account knowledge about the study and population) [19]. that 45% of people aged 65 years or more in Barce- School of Health participants will be adults aged 65 lona reports fair or poor health [21] and assumed years or older residing in two disadvantaged neighbor- that the intervention would reduce this prevalence by hoods of Barcelona (Spain) where the School of Health 17 points. Accepting an alpha risk of 0.05% and a will be implemented. Theoretical sampling will be de- power of 80% in a unilateral contrast, and assuming a signed to select participants from the School of Health. loss of 10%, we estimated that a theoretical sample of For each neighborhood, two groups of participants will 79 participants in the IG and 79 participants in the be selected according to their risk of experiencing loneli- CG will be needed. ness. The risk of loneliness will be measured through We expect to recruit 80 individuals in the IG and 80 various questions extracted from the National Social in the CG. To have similar groups, the recruitment set- Life, Health, and Aging Project (NSHAP) study [20], tings (primary care centers, social services and civic cen- included in the questionnaire used in the quantitative ters) would be taken into account, by selecting the same study. Individuals who report feeling a lack of compa- percentage of people from each setting. nionship often, regardless of their cohabitation status The exclusion criteria are the following: (with relatives or alone), will be categorized as ‘at high risk’ and those reporting hardly ever feeling alone and – Participants with difficulties in maintaining living with relatives will be classified as ‘low risk’. Thus, participation for 6 months. four focus groups (two per neighborhood) with 6 to 10 – Participants with difficulties in understanding or assistants will be assembled. expressing themselves in Spanish or Catalan. For each focus group, we will select participants pro- viding the widest range of profiles to ensure a variety of Economic evaluation discourses according to age, sex, marital status and edu- The study population and the selection criteria will be cational level. Community nurses will be asked about the same as those in the quantitative quasi-experimental the ability of the selected participants to follow the dy- design. namics of a focus group. Then, the selected participants will be contacted by telephone, informed about the study and asked if they are willing to participate. Those refus- Participant timeline ing to take part in the study will be replaced by other in- The participants will be contacted by telephone to ar- dividuals with similar characteristics. Those who accept range and appointment, in with the person will be to take part in the focus group will have to sign an in- informed individually about the study. After agreement, formed consent form. he/she will sign the informed consent form and complete the baseline questionnaire (December–January Quantitative quasi-experimental design 2015). Those participants not able to arrange an ap- The study population will consist of adults aged 65 years pointment, will receive the consent form and complete or older residing in the disadvantaged neighborhoods of the baseline questionnaire during the first session of Barcelona (Spain) selected for the study. The interven- the intervention (just before starting). The intervention tion neighborhoods (n = 2) and comparison neighbor- will last 22 weeks (January–June 2015). At the end of hoods (n = 2) will be selected on a convenience basis, the last session the participants will complete the ensuring similar socioeconomic characteristics, inclu- follow-up questionnaire (June 2015). Regarding the ding the percentage of people with primary studies or qualitative analysis, the in-depth interviews and focus less, unemployment rates and disposable household groups will be carried out between 6 to 9 months after income rates. the intervention (January–March 2016). The intervention will be offered to adults aged 65 years The participant flowchart of the study procedure is or older, living in one of the two selected neighborhoods, outlined in Fig. 1. López et al. BMC Geriatrics (2019) 19:44 Page 4 of 8 Fig. 1 Flow chart of enrolment, allocation and follow-up Description of the intervention A brief description of the contents of one “School of The intervention “School of Health for Older People” Health for Older People” is shown in Table 1. consists of 22 weekly group sessions of 1.5 h each, in which issues related to health—including both biological Outcome measures and psychological issues—and social topics, are discussed. Qualitative evaluation Attendance is free and the sessions will be held in com- The main topics covered in the qualitative evaluation munity centers of the selected neighborhoods. In addition will be positive and negative aspects of the intervention, to helping participants learn about different health issues, suggestions for its improvement, opinions on different the intervention encourages interaction among partici- session contents and logistic aspects of the intervention, pants and works on skills in different fields, such as nutri- and perceived benefits derived from the intervention. tion, management of emotions, self-medication, sexuality or physical activity. Furthermore, some sessions included visits to public spaces of the neighborhood and leisure Quantitative quasi-experimental design activities adapted to older people. Furthermore, most sessions are led by professionals Main outcomes Social support will be measured who are experts on the topic covered and work in the through questions taken from the Medical Outcomes neighborhood (professionals from the health services, Study: Measures of Quality of Life Core Survey (MOS) social services, markets or neighborhood associations), [22] and NSHAP questionnaires [20]. Psychological making it easier to inform participants of the neighbor- morbidity will be measured through Goldberg’s Gen- hood’s available resources. The aim of the intervention is eral Health Questionnaire (GHQ-12) [23], and to decrease social isolation and loneliness and, therefore, health-related quality of life will be measured through to improve mental health, self-perceived health and well- the EuroQoL (EQ-5D) questionnaire [24]. Objective being (see the explanatory model of the potential effects information will be obtained on visits to the primary of the intervention in Fig. 2). care center during the year before and after the López et al. BMC Geriatrics (2019) 19:44 Page 5 of 8 Fig. 2 Explanatory model of the potential effects of the “Schools of Health for Older People” intervention through access to the electronic records of the primary care centers. All these variables will be measured simultaneously in individuals in the IG and CG. Data will be collected through baseline and post-intervention questionnaires administered by trained researchers. Table 1 Example of the contents of sessions in the “School of Intervention-related variables We will register attend- Health for Older People” ance of participants at each session in the IG. Addition- Session 1 Introduction and group instructions. Colds and flu ally, at the end of the School of Health, we will measure Session 2 Adequate nutrition and fluid intake participants’ satisfaction with questions such as “please Session 3 Elimination of body waste. Hygiene rate your general satisfaction with the School of Health Session 4 Laughter therapy from 0 to 10;”“ please rate your satisfaction with the Session 5 Health hazards in the household and first aid speakers, place/space, frequency, schedule, and length of Session 6 Self-medication: the less the better sessions from 0 to 10” “ Would you recommend the School of Health to a friend? ” These questions will be Session 7 Sexuality: enjoying it at any age included in the post-intervention questionnaire. Session 8 Preventing scams and thefts Session 9 Circus workshop Sociodemographic variables Sociodemographic infor- Session 10 Going out: how to do it in the safest way mation will only be collected in the baseline question- Session 11 Memory and the five senses naire including the following variables: age, sex, marital Session 12 Emotions status (single/ married/ divorced/ widowed), cohabit- ation status (living alone/living with someone else), edu- Session 13 Physical activity at the park cational level (illiterate /primary / secondary / university) Session 14 The value of older age and residential neighborhood. Session 15 How to become a volunteer at our neighborhood Session 16 Our household remedies: let’s share! Economic evaluation Session 17 Diarrhea and Vomiting: what to do? The economic evaluation will be conducted from a Session 18 Heat waves health system (primary care) perspective, including di- rect costs of the program and the primary care health Session 19 Songs of our times. Farewell services used. The time frame will be 6 months and López et al. BMC Geriatrics (2019) 19:44 Page 6 of 8 consequently no discounting will be applied to the t test, and the Mann-Whitney U test will be used in calculations. non-normally distributed variables. In case of diffe- rences, a further multivariate effectiveness analysis will Costs The costs of the program will be calculated based be conducted adjusting for these variables. on the hourly wages of the public health professionals Further descriptive analyses will be conducted for re- involved in the design and implementation of the inter- sults at baseline and at follow-up. Categorical outcomes vention, obtained from the standard professional wages will be described through percentages, and continuous in the Official Gazette of the Catalan Government; the outcomes will be described through mean scores and wages of the community nurses involved in the coordin- standard deviations or the median and interquartile ation and development of the workshops, obtained from range, as required. Differences between pre- and post- the Catalan Institute of Health retributions report; and intervention measurements will be assessed and com- the costs of the venues and materials used throughout pared using the McNemar test, paired t-test, or sign test, the intervention. The costs of the primary care medical according to the type of variable and distribution. and nurse visits either at the health center or at the pa- To assess the effectiveness of the intervention, a tient’s home will be drawn from the Official Gazette of two-sample McNemar test will first be conducted to the Catalan Government on healthcare costs for the last analyze differences in changes in pre- and post- published year. intervention results between the IG and the CG for each main outcome. In those outcomes showing differ- Effects The effects of the intervention will be measured ences between the groups at the baseline measurement, in quality adjusted life years (QALY), through the Poisson regression models with robust variance will be EQ-5D questionnaire. Spanish tariffs will be used to esti- built. In each model, the independent variables will be mate the utility of health states described by the partici- the group (IG or CG), the pre-intervention measure- pants. QALYs [25] will be assessed at baseline and at 6 ments for the outcome assessed and any relevant char- months (end of the intervention), and QALYs will be acteristics that differ between the groups at the baseline calculated by multiplying utility by the amount of time a assessment. patient spent in a particular health state. Linear interpolation will be used for transitions between health states. QALYs will be assessed at baseline and at 6, 12 Economic evaluation and 18 months. The incremental cost-utility ratio (ICUR) will be calcu- lated by subtracting the costs of the CG from the costs Data analysis of the IG and dividing the result by the QALYs for the Qualitative evaluation CG subtracted from the QALYs for the IG. The incre- A thematic analysis will be carried out with the support mental costs and the incremental utility will be modeled of ATLAS.ti.software. In-depth-interviews and focus by generalized linear models, taking into account the groups will be recorded and transcribed literally. Inter- distribution of both costs and effects. viewers will keep a diary in which any reaction to events To estimate the uncertainty related to the ICUR, we occurring during the research will be recorded. will calculate 95% confidence intervals for the mean cost The analysis will combine the inductive and deductive differences using the non-parametric confidence interval. definition of codes. Preanalytical intuitions will be for- Bootstrapping with 1000 replications will be performed. mulated after successive readings of the transcriptions Bootstrapped cost effects pairs will be plotted on and the observation notes. Next, three multidisciplinary cost-effectiveness planes and used to estimate cost- investigators will create an initial analytical plan based utility acceptability curves (CUACs). In the cost-utility on the most relevant topics (codification). The creation planes, the ‘x’ axis represents the difference in QALYs of categories by grouping the codes will be based on the and the ‘y’ axis the difference in costs. CUACs demon- criterion of similarity in relation to the objectives of the strate the probability that an intervention is cost effect- study and the emerging elements. ive at a specific ceiling ratio, which is the amount of money society is willing to pay to gain one extra unit of Quantitative quasi-experimental design effect. Willingness to pay values will range from 0€ to A descriptive analysis will be conducted to rule out 50,000€ [26]. significant differences between baseline characteristics The robustness of the estimates will be addressed by (sociodemographic variables and outcomes studied) in conducting sensitivity analyses. This analysis will com- the IG and the CG. The chi-square test will be used to prise a variation in the unit cost of the primary care compare qualitative variables. Continuous variables with medical or nurse visits, in the costs of the workshop normal distribution will be compared using the Student leaders’ hourly wage and in the venue rental cost. López et al. BMC Geriatrics (2019) 19:44 Page 7 of 8 Ethics considerations The results of this study will help to fill the knowledge Participants will be informed both verbally and in writing gap in this area and might be especially useful for the about the aims, methods, procedures and measures per- development of social and public health policies and formed during the study. They will be also informed about programs for older people in disadvantaged neighbor- ethics issues such as confidentiality, their right to ask any hoods of urban areas. questions during the study, and their right to withdraw at Abbreviations any time without penalty. To ensure that all participants CG: Comparison group; CUAC: Cost-utility acceptability curves; EQ- have received proper information about the study and 5: EuroQoL-5D; GHQ-12: Goldberg’s General Health Questionnaire; have agreed to participate, all participants will be asked to IG: Intervention group; ICUR: Incremental Cost-Utility Ratio; MOS: Measures of Quality of Life Core Survey; NSHAP: National Social Life, Health, and Aging sign a written consent form. The research team is com- Project; QALY: Quality Adjusted Life Years mitted to performing this study in accordance with the Good Clinical Practice Guidelines of the Declaration of Acknowledgements Helsinki. This protocol was approved by the Comité Ético Evaluation Group of the “Schools of Health for Older People”: Alba Sánchez, Ana Fernández, Anna Pérez, Camila Astorga, Carme Cortina, Carolina Lapena, de Investigación Clínica Parc de Salut Mar (code n° 2015/ Cristina Rey, Elia Diez, Enriqueta Pujol, Ferran Daban, Gemma Castillo, Iolanda 6500/I). Robles, Lucía Artazcoz, María José López, Mariona Pons, Marc Marí- Dell’Olmo, Montse Petit, Natalia Sagarra, Nuria Calzada, Olga Juárez, Xavier Bartoll and Xavier Continente. Discussion Strengths and limitations Funding A potential limitation of the study is the length of This work was supported by the 8th Institute for the Investigation in Primary Healthcare (IDIAP) Jordi Gol Grant Concession ‘Convocatòria d’una estada a follow-up, since the post-intervention measurements will una Unitat de Recerca de l’IMIM o l’ASPB o l’IDIBAPS [grant number 2013/01], be carried out immediately after the end of the interven- funded by the Catalan Health Institute. tion (6 months). Therefore, we will not be able to assess its long-term effects. Another limitation of the quantita- Availability of data and materials Not applicable. tive part of the study is that, because the design is quasi-experimental, there may be some differences Authors’ contributions between the IG and the CG. However, a statistical com- MJL was responsible for the research protocol design, which was developed parison of the main sociodemographic variables will be by all the authors. AS, CL, XC and AF contributed to the manuscript preparation. All authors edited, read and approved the final version of the assessed and if any differences are found, they will be manuscript. taken into account in the adjusted regression models. A strength of the study is that, as far as we know, it is Ethics approval and consent to participate This protocol was approved by the Comité Ético de Investigación Clínica Parc the first to assess the effectiveness and cost-effectiveness de Salut Mar (code n° 2015/6500/I). Participants will be informed both of an intervention focused on reducing social isolation verbally and in writing about the aims, methods, procedures and measures in older people in deprived neighborhoods in an urban performed during the study. They will be also informed about ethics issues such as confidentiality, their right to ask any questions during the study, and environment. Importantly, this is a multi-approach eva- their right to withdraw at any time without penalty. To ensure that all luation assessing the process, effectiveness and cost- participants have received proper information about the study and have effectiveness of an intervention while using mixed agreed to participate, all participants will be asked to sign a written consent form. methods (quantitative and qualitative). Furthermore, the tests applied to measure the main outcomes (social Consent for publication support, psychological morbidity and health-related qual- Not applicable. ity of life) are validated tests that not only allow for assess- Competing interests ment of the potential impact of the intervention but also The authors declare that they have no competing interests. for comparison with other studies. Publisher’s Note Implications Springer Nature remains neutral with regard to jurisdictional claims in Aging is one of the greatest social and economic chal- published maps and institutional affiliations. lenges for European societies. At the same time as cities Author details are growing, their share of older residents is increasing. 1 Public Health Agency of Barcelona, Lesseps, 1, 08023 Barcelona, Spain. By 2025, more than 20% of European citizens will be 2 Spanish Consortium for Research on Epidemiology and Public Health 65 years or older [27]. Ensuring the integration of older (CIBERESP), Madrid, Spain. 3Sant Pau Institute of Biomedical Research (IIB Sant Pau), Barcelona, Spain. 4Centre d’Atenció Primària Sanllehy, Gerència d’Àmbit people in society is essential to promote the wellbeing d’Atenció Primària Barcelona Ciutat, Institut Català de la Salut, Barcelona, of older urban residents. Therefore, studies assessing Spain. 5Unitat de Suport a la Recerca Barcelona Ciutat, Institut Universitari the effectiveness and efficiency of potential interven- d’Investigació en Atenció Primària Jordi Gol (IDIAP Jordi Gol), Barcelona, Spain. 6Universitat Autònoma de Barcelona, Bellaterra, Cerdanyola del Vallès, tions to reduce the social isolation of older people are Spain. 7Preventive Medicine Department, University Hospital of Bellvitge, urgently needed. Hospitalet de Llobregat, Barcelona, Spain. López et al. BMC Geriatrics (2019) 19:44 Page 8 of 8 Received: 6 July 2017 Accepted: 31 January 2019 25. Sassi F. Calculating QALYs, comparing QALY and DAY calculations. Health Policy Plan. 2006;21(5):402–8. 26. Sacristán JA, Oliva C, Del Llano J, Prieto L, Pinto JL. ¿Qué es una tecnología sanitaria eficiente en España? Gac Sanit. 2002;16(4):334–43. References 27. European Commission. 2015. Ageing policy. http://ec.europa.eu/health/ 1. Beard JR, Bloom DE. Towards a comprehensive public health response to ageing/policy/index_en.htm. Accessed 30 June 2016. population ageing. Lancet. 2015;385(9968):658–61. 2. Idescat. Projeccions de poblacions. 2014. http://www.idescat.cat/cat/idescat/ publicacions/cataleg/pdfdocs/pp2013-2051pr.pdf Accesed 15 June 2016. 3. Domènech-Abella J, Lara E, Rubio-Valera M, Olaya B, Moneta MV, Rico-Uribe LA, Ayuso-Mateos JL, et al. Loneliness and depression in the elderly: the role of social network. Soc Psychiatry Psychiatr Epidemiol. 2017. https://doi.org/ 10.1007/s00127-017-1339-3. 4. Scharf T, Philipson C, Smith AE. Poverty and social exclusion: growing older in deprived urban neighbourhoods. In: Holland C, editor. Growing older: quality of life in old age. Buckingham: Open University press; 2004. p. 81–106. 5. Savikko N, Routasalo P, Tilvis RS, Strandberg TE, Pitkälä KH. Predictors and subjective causes of loneliness in an aged population. Arch Gerontol Geriatr. 2005;41(3):223–33. 6. Barnes M, Blom A, Cox A, Lessof C. The social exclusion of older people: evidence from the first longitudinal study of ageing (ELSA). London: Office of the Deputy Prime Minister; 2006. 7. Henning-Smith C, Moscovice I, Kozhimannil K. Differences in social isolation and its relationship to health by rurality. J Rural Health. 2019. https://doi. org/10.1111/jrh.12344 [Epub ahead of print]. 8. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Med. 2010;7(7):e1000316. 9. Steptoe A, Shankar A, Demakakos P, Wardle J. Social isolation, loneliness, and all-cause mortality in older men and women. Proc Natl Acad Sci U S A. 2013;110(15):5797–801. 10. Tabue Teguo M, Simo-Tabue N, Stoykova R, Meillon C, Cogne M, Amiéva H, Dartigues JF. Feelings of loneliness and living alone as predictors of mortality in the elderly: the PAQUID study. Psychosom Med. 2016;78(8):904–9. 11. Dickens AP, Richards SH, Greaves CJ, Campbell JL. Interventions targeting social isolation in older people: a systematic review. BMC Public Health. 2011;11:647. 12. Cattan M, White M, Bond J, Learmouth A. Preventing social isolation and loneliness among older people: a systematic review of health promotion interventions. Ageing Soc. 2005;25(01):41–67. 13. Coll-Planas L, Nyqvist F, Puig T, Urrútia G, Solà I, Monteserín R. Social capital interventions targeting older people and their impact on health: a systematic review. J Epidemiol Community Health. 2016. https://doi.org/10. 1136/jech-2016-208131. 14. Gardiner C, Geldenhuys G, Gott M. Interventions to reduce social isolation and loneliness among older people: an integrative review. Health Soc Care Commun. 2016. https://doi.org/10.1111/hsc.12367. 15. Hagan R, Manktelow R, Taylor BJ, Mallett J. Reducing loneliness amongst older people: a systematic search and narrative review. Aging Ment Health. 2014;18(6):683–93. 16. Salut als barris. 2016. http://www.aspb.cat/documents/barcelona-salut-als- barris-memoria-2017/. Accesed 12 September 2016. 17. Sadana R, Blas E, Budhwani S, Koller T, Paraje G. Healthy ageing: raising awareness of inequalities, determinants, and what could be done to improve health equity. Gerontologist. 2016;56(Suppl 2):S178–93. 18. Niedzwiedz CL, Richardson EA, Tunstall H, Shortt NK, Mitchell RJ, Pearce JR. The relationship between wealth and loneliness among older people across Europe: is social participation protective? Prev Med. 2016;91:24–31. 19. Pla M. El rigor en la evaluación cualitativa. Aten Primaria. 1999;24(5):295–8. 20. Cornwell EY, Waite LJ. Measuring social isolation among older adults using multiple indicators from the NSHAP study. J Gerontol B Psychol Sci Soc Sci. 2009;2009. https://doi.org/10.1093/geronb/gbp037. 21. Agència de Salut Pública de Barcelona. [La salut a Barcelona 2015]. 2015http://www.aspb.cat/wp-content/uploads/2016/11/Informe-Salut-2015. pdf. Accessed 15 January 2017. 22. Andrade E, Arce C, Seoane G. Adaptación al español del cuestionario «Perfil de los Estados de Ánimo» en una muestra de deportistas. Psicothema. 2002;14(4):708–13. 23. Goldberg DP, Oldehinkel T, Ormel J. Why GHQ threshold varies from one place to another. Psychol Med. 1998;28(4):915–21. 24. Rabin R, de Charro F. EQ5D: a measure of health status from the EuroQol group. Ann Med. 2001;33(5):337–43.