Blancafort Alias et al. BMC Public Health (2021) 21:84 https://doi.org/10.1186/s12889-020-10094-9 RESEARCH ARTICLE Open Access Promoting social capital, self-management and health literacy in older adults through a group-based intervention delivered in low-income urban areas: results of the randomized trial AEQUALIS Sergi Blancafort Alias1,2*, Rosa Monteserín Nadal2,3, Irene Moral2,3, Marta Roqué Fígols1,2, Xavier Rojano i Luque1,2 and Laura Coll-Planas1,2 Abstract Background: Evidence is scarce on how to promote health and decrease cumulative inequalities for disadvantaged older people. Downstream complex interventions focusing on intermediate factors (self-management, health literacy and social capital) may have the potential to mitigate the inequitable impacts of social determinants in health. The aim of the AEQUALIS study was to assess the effectiveness of a group-based intervention to improve self-perceived health as indicator of health inequality. Methods: Pragmatic randomised clinical trial addressed to older adults (≥ 60 years) living in urban disadvantaged areas with low self-perceived health. The intervention was delivered in primary care settings and community assets between 2015 and 2017 and consisted in 12 weekly sessions. The primary outcome was self-perceived health assessed in two ways: with the first item of the SF-12 questionnaire, and with the EQ-5D visual analog scale. Secondary outcomes were health-related quality of life, social capital, self-management, mental health and use of health services. Outcomes were assessed at baseline, post intervention and follow-up at 9 months after the end of the intervention. (Continued on next page) * Correspondence: sergi.blancafort@uab.cat 1 Fundació Salut i Envelliment UAB. Casa Convalescència, Sant Antoni Maria Claret, 17, 4a planta, 08041 Barcelona, Spain 2 Institute of Biomedical Research (IIB Sant Pau), Sant Quintí, 75-77, 08041 Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 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 in a credit line to the data. Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 2 of 12 (Continued from previous page) Results: 390 people were allocated to the intervention group (IG) or the control group (CG) and 194 participants and 164 were included in the data analysis, respectively. Self perceived health as primary outcome assessed with SF-12-1 was not specifically affected by the intervention, but with the EQ-5D visual analog scale showed a significant increase at one-year follow-up only in the IG (MD=4.80, 95%CI [1.09, 8.52]). IG group improved health literacy in terms of a better understanding of medical information (− 0.62 [− 1.10, − 0.13]). The mental component of SF-12 improved (3.77 [1.82, 5.73]), and depressive symptoms decreased at post-intervention (− 1.26 [− 1.90, − 0.63]), and at follow-up (− 0.95 [− 1.62, − 0.27]). The use of antidepressants increased in CG at the follow-up (1.59 [0.33, 2.86]), while it remained stable in the IG. Conclusions: This study indicates that a group intervention with a strong social component, conducted in primary health care and community assets, shows promising effects on mental health and can be used as a strategy for health promotion among older adults in urban disadvantaged areas. Trial registration: ClinicalTrials.gov, NCT02733523. Registered 11 April 2016 - Retrospectively registered Keywords: Social medicine, Public health, Primary health care, Healthcare disparities, Aging, Self-management, Health literacy, Social capital Background dependency effect on socioeconomic inequalities in A number of factors are related to health inequities. It is health [16]. These factors are interrelated, as it has been well established that a low socioeconomic level is associ- suggested that older adults with low health literacy ated with poor health [1], with urban areas concentrat- would be less prone to maintain a social network or ing higher health inequalities, and ageing being also an make use of social resources [17]. Likewise, some au- axe of inequality [2]. Consequently, older people living thors have suggested that associations between health in disadvantaged urban areas should be a specific focus literacy and health behaviours are mediated by social for health equality research and policies, since they espe- cognitive factors such as self-efficacy [18]. Accordingly, cially suffer its accumulative effects [3]. However, there these three modifiable intermediate factors (i.e., self- is still a need for better evidence on how to address management, health literacy and social capital) might es- those inequalities [4]. Ndumbe-Eyoh & Moffat [5] clas- tablish synergies with each other increasing their cap- sify interventions addressing health inequalities as “up- acity to reduce health inequalities. Moreover, their stream” (structural and system-level changes), synergies might reinforce older people’s wellbeing and “midstream” (community or organizational level) and might also be appropriate to addressed mental health. “downstream”. This latter include behavioural or psy- When designing and evaluating interventions involving chosocial factors that occur at individual level and have several interacting components, emerging methodologies the potential to mitigate the inequitable impacts of social that address complex interventions are valuable [19]. determinants. These methodologies also take in account dimensions of Health-related behaviours have been pointed as a core complexity such as the number and difficulty of behav- element to consider in interventions promoted by local iours required by those delivering or receiving the inter- primary health care agents [6]. An extensive body of evi- vention, the number of groups or organisational levels dence exists on the association between unhealthy be- targeted, the number and variability of outcomes and haviours and the burden of morbidity and mortality [7], the degree of flexibility or tailoring of the intervention as well as low health-related quality of life and mental permitted. Another key question in evaluating complex illness [8, 9]. From a salutogenic perspective, the concept interventions is how an intervention works. Hence, of positive mental health or mental well-being, which which are the active components within the intervention implies “feeling good”, has emerged as a strong predictor and how do they reach their effect [20]. As current evi- of overall health [10, 11]. Likewise, promising evidence dence is weak on the effectivemess of policies and inter- correlates health-related behaviours with better self- ventions to reduce health inequalities, more rigorous management and mental well-being [12, 13]. evidence-based approaches are needed to inform policy- Health literacy, is also strongly associated with health making in this area [21]. Thus, the effectiveness of these inequalities, for instance linking low socio-economic complex interventions should be robustly assessed level with health choices related with worse outcomes through randomised clinical trials [22]. [14]. Social capital, an umbrella concept that includes To this end, we developed a complex intervention social support and participation in the community [15], called “Sentir-nos Bé” (“Feeling well”) to promote self- has been identified as having both a buffer and a management, health literacy and social capital in older Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 3 of 12 adults living in disadvantaged urban areas in order to re- Interventions duce health inequalities [23]. The main objective of the The intervention of interest was a community program AEQUALIS study was to evaluate the effectiveness of with multiple components, involving/targeting a broad the programme “Feeling well” to improve self-perceived range of outcomes, and new skills and behaviours to be health compared to usual care. Self-perceived health was acquired by those delivering and receiving the interven- chosen as a useful indicator since it correlates with gen- tion. Thus, a complex intervention [19]. The interven- eral health status, mortality and morbidity, as well as tion is described in accordance with the TIDieR health inequalities [24–26]. Secondarily, we assessed the guidelines [28]. It was aimed at promoting social support effectiveness of the intervention in the improvement of and participation (i.e., social capital), self-management health-related quality of life, social capital, self- and health literacy as intermediate factors between social management, health literacy, mental health and the use determinants and health outcomes with potential to of health resources. mitigate health inequalities. The intervention consisted in 12 sessions held weekly for 2 h and facilitated in groups up to 15 people. Nine of the 12 sessions were de- Methods livered in the primary care centre, and three consisted in The study protocol has been previously reported, and no local outings to public spaces to practice physical activ- relevant changes were made to the planned methods ity, a supermarket or a market, and a community equip- during the trial [23]. The trial is reported according to ment offering social activities that could be of interest to CONSORT 2010 statement [27]. the participants. All group interventions were run be- tween September 2015 and April 2017 and were facili- tated by nine health and social care professionals (five Study design, settings and patients nurses, two social workers and two general practitioners) The trial had a pragmatic multicentric, parallel, previously trained as group facilitators by the research individually-randomised controlled design with a 1:1 al- team on how to apply the intervention guide. At least location ratio between intervention and control. Partici- one health or social care professional from each site par- pants were recruited from people attending 16 primary ticipated as an observer of the intervention, completed care centers belonging to low-income neighborhoods. an observation log including quantitative and qualitative Selection of centers was based on convenience sampling. measures of implementation such as fidelity and adher- Individuals were considered eligible to participate in the ence. Participants in the control group were offered the study if: they were community-dwelling, aged 60 years or intervention at the end of the study (delayed-interven- above, and perceived their health as fair or poor. Partici- tion group). pants were excluded if: they needed help to go to the primary care center; had cognitive impairment or diag- Outcomes nosed dementia; had a medical condition that contra- The primary outcome was self-perceived health, and it indicate physical activity; had any severe mental health was assessed in two ways with the first item of the SF-12 problem that hinders participation in a group dynamic; questionnaire [29] (dichotomized into excellent-very or had an end of life situation. good-good and fair-poor self-perceived health), and with the EQ-5D visual analog scale of the EuroQOL tool [30]. Secondary outcomes included health-related quality of Randomisation and blinding life, self-management, health literacy, social capital (so- Once participants were included in the study, assigned cial support, social participation), mental health (emo- an identification code, and completed the study baseline tional well-being, loneliness, depressive symptomatology, assessment, they were randomised to an intervention anxiolytics and antidepressants consumption), and use group (IG) or control group (CG). Concealed random- of health resources (visits to nurse, general practitioner, isation was conducted centrally at Fundació Salut i social worker, emergency units, and hospitalizations). Envelliment UAB, using a computer-based random- The quality of life was measured by SF-12 questionnaire. block randomisation scheme, stratified by primary care Self-management was measured by the Appraisal of Self- centre. Participants and professionals conducting the Care Agency Scale (ASA-R) [31]. The Health Literacy group-based intervention remained unblinded. Blinding Scale HLS-EU-16 [32] was used to measure some com- of outcome assessors was intended as assessments were ponents of health literacy. Especifically, we estimated the conducted by professionals not involved with the inter- difficulty in understanding medical information, finding vention delivery or observation; however, this blinding out about activities that are good for mental well-being was hard to sustain, as participants could easily reveal and assessing healthy lifestyles. Social support was mea- their group allocation. sured by the Social Resources Inventory in Older Adults Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 4 of 12 [33], and social participation was measured by Este II with a baseline assessment regardless of their perman- Subjective Social Participation Index [34]. Emotional ence in the study or their loss to follow-up, withdrawal well-being was measured by the Warwick-Edinburgh or drop-out. All statistical analyses used two-sided tests Mental Wellbeing Scale (WEMWBS) [35]. Loneliness of statistical significance; estimates of the size of treat- was measured by the 11-item De Jong Gierveld Loneli- ment effects were presented with confidence intervals ness Scale [36], comprising subscales for emotional and and significance level. Significance levels were set to a social loneliness. Depressive symptomatology was mea- 5% level. All analyses were performed with STATA/MP sured by the Geriatric Depression Scale (GDS-5) [37]. 14.0. Anxiolytics and antidepressants consumption, as well as use of health resources were obtained through electronic Results primary care records. Study participants and baseline characteristics An English language version of the questionnaire used A total of 480 persons were screened for eligibility, of as part of the study design is included as a supplemen- whom 90 were excluded because they did not meet in- tary file. All outcomes were measured at the baseline clusion criteria (n=52) or refused to participate (n=38). (T0), after the intervention was completed (T1), and 9 The 390 included participants were allocated to the IG months after the end of the intervention (T2). The use (195 participants) or the CG (195 participants). 358 par- of health services (visits to nurse, general practitioner, ticipants could be assessed at baseline (194 and 164), social worker, emergency units, and hospitalizations) re- and included in the data analysis. Figure 1 illustrates the ferred to the last 12 months, and were measured at T0 flow of participants through each stage of the study. Par- and T2. Assessments were conducted by health or social ticipants were recruited between 15/07/2015 and 31/01/ care professionals from the primary care centre not in- 2017. The last wave of 9 months follow-up assessments volved with the intervention delivery or observation. All after the intervention ended on June 30th, 2018. outcome assessors received training in interview skills Descriptive values for each group of baseline socio- and outcome measure administration by the research demographic, psycho-social and clinical characteristics group prior to the start. of the study group participants are presented in Table 1. Participants were mostly women (81.4%) and their mean Sample size age was 73.6 ± 6.9 years. The majority of participants A sample size of 390 participants was needed to detect a (82.2%) had received no formal education or completed clinically relevant benefit in self-perceived health after 3- only primary studies. Slightly more than a fifth (21.4%) month intervention, defined as a 10% increase in the were born in the same city where currently living. Four prevalence of participants who considered their health as out of ten participants were widow (41.1%, mean of good, according to the first item of SF-12. Details of years of widowhood = 10.2) or lived alone (40.8%). The computation have been previously published [23]. treatment and control groups were well balanced at baseline with regards to most socio-demographic, Data analysis psycho-social and clinical characteristics. However in re- First of all, a descriptive analysis was conducted to lation to CG, IG had on average 3.64 point less in men- characterize the intervention and the control group re- tal health (SF-12 score on the mental component), a garding socio-demographic, health and psychosocial var- higher use of antidepressants (risk difference 15.54%) iables. Categorical variables were described as and 1.84 visits less to the nurse and more participants frequencies and percentages. Continuous variables were understanding medical information (risk difference described as mean and standard deviation, or as medians 9.82%). and quartiles (Q1 and Q3) if they were clearly asymmet- rical. Baseline comparability of the study arms was assessed by means of chi-square test for qualitative vari- Results of multilevel models on the effectiveness of the ables, and Student’s t-test and Wilcoxon tests for quanti- intervention over time tative variables. In a second step, changes between Primary outcome baseline, post-intervention and 9 months after end inter- Between baseline and T1, good/very good self-perceived vention comparing study arms (experimental and con- health assessed with SF-12 increased for participants of trol group) were carried out using multilevel mixed- both groups: in the IG (mean change 5.15 [95%CI: 3.34, effects models; linear, logistic or ordered logistics regres- 6.97]) and the CG (3.27 [1.89, 4.66]). At follow-up, the sions were used when applicable. Repeated measures improvement from baseline was maintained in both were considered nested at two levels, at center level and study arms, but the increase was higher for the IG with at individual level. It followed the intention-to-treat an absolute rise of 27.27% (5.01 [3.21, 6.82]), compared principle, thus including all randomized participants to a 16.14% (3.13 [1.71, 4.56]) in the CG (Table 2, Fig. 2). Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 5 of 12 Fig. 1 Flow chart of participants through each stage of randomized clinical trial However, no differences between study groups were ob- (P=0.036). No further significant differences were found served at any timepoint (Table 2). between the two study grups in the remaining secondary Likewise, self-perceived health assessed with the visual outcomes at this time point. When comparing changes analog scale also increased from T0 to T1 in both between baseline assessments and T1, the physical com- groups IG (8.29 [4.91, 11.67]) and CG (5.49 [1.80, 9.18]) ponent of the SF-12 improved in both groups: IG (1.57 (Table 2, Fig. 2) and no differences were found between [0.16, 2.97]) and CG (2.95 [1.41, 4.50]). However, only study groups. Nevertheless, only the IG sustained the the IG improved the mental component (3.77 [1.82, improvement until T2 (4.80 [1.09, 8.52]). 5.73]), showed higher levels of understanding medical information (− 0.62 [− 1.10, − 0.13]) and decreased de- Secondary outcomes pressive symptomatology (− 1.26 [− 1.90, − 0.63]). When When comparing the groups at T1, depressive symp- comparing changes between baseline assessments and tomatology was higher in the CG compared to the IG T2, higher levels of self-management were reported by Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 6 of 12 Table 1 Characteristics of participants at the baseline by treatment group Total (n= Intervention group (n= Control group (n= P value 360) 194) 164) Demographic characteristics Women, % (n) 81.39 (293) 82.65 (162) 79.88 (131) 0.500 Age, in years, mean (SD) 73.63 (6.92) 73.41 (6.92) 73.88 (7.02) 0.531 Education level, % (n) (without or only primary studies) 82.23 (296) 84.18 (165) 79.88 (131) 0.548 Born in the same city where currently living, % (n) 21.39 (77) 23.47 (46) 18.90 (31) 0.301 Marital status 0.369 Widow, % (n) 41.11 (148) 44.39 (87) 37.20 (61) Years of widowhood, median (Q1-Q3) 7,8 (3–15) 6 (3–14) 10 (3–16.5) Same tabulation as "marital status" Living alone, % (n) 40.83 (147) 40.31 (79) 41.46 (68) 0.715 Same tabulation as "marital status" Caring for other people, % (n) 46.94 (169) 50.51 (99) 42.68 (70) 0.198 Health status Self-perceived health (SF-12-1)a, % (n) 0.291 Very good 0.28 (1) 0.51 (1) 0 (0) Good 2.22 (8) 1.02 (2) 3.66% (6) Fair 77.78 (280) 78.06 (153) 77.44% (127) Poor 19.72 (71) 20.41 (40) 18.90% (31) Self-perceived health (EQ-5D VAS)b, mean (SD) 51.31 (18.59) 49.97 (18.62) 52.89 (18.48) 0.140 a SF-12 Physical component, mean (SD) 36.49 (9.55) 36.59 (9.62) 36.36 (9.50) 0.829 SF-12a Mental component, mean (SD) 40.04 (13.49) 38.38 (14.11) 42.02 (12.44) 0.013* Self-management (ASA-R)c, mean (SD) 67.94 (12.06) 66.93 (12.07) 69.30 (11.97) 0.124 Psychosocial characteristics Social support Extension of social networkd, median (Q1-Q3) 4 (3.5–5) 4 (3–5) 4 (4–5) 0.288 e Emotional support , median (Q1-Q3) 3 (2–4) 3 (2–4) 3 (2–4) 0.872 Instrumental supporte, median (Q1-Q3) 1 (0–1) 1 (0–1) 1 (0–1) 0.421 e Informational support , median (Q1-Q3) 1 (0–2) 1 (0–2) 1 (0–2) 0.062 Social participationf, % (n) 0.950 Low 29.44 (106) 28.57 (56) 30.49 (50) Moderate 43.06 (155) 46.43 (91) 39.02 (64) High 21.94 (79) 19.39 (38) 25.00 (41) Same tabulation as "social participation" Emotional Wellbeing (WEMWBS)g, 53.78 (8.94) 52.93 (8.65) 54.91 (9.24) 0.135 mean (SD) Loneliness intensityh (De Jong Gierveld Loneliness Scale), % (n) 0.508 Without loneliness 33.61 (121) 32.65 (64) 34.76 (57) Moderate 63.06 (227) 63.27 (124) 62.80 (103) Severe 2.50 (9) 3.57 (7) 1.22 (2) Very severe 0.56 (2) 0.51 (1) 0.61 (1) Same tabulation as "social participation" Depressive symptomatology 57.22 (206) 59.69 (117) 54.27 (89) 0.505 (GDS-5)i, % (n) Number of chronic diseases, median (Q1-Q3) 5 (3–8) 5 (3–8) 5 (3.75–8) 0.913 Number of chronic medications, median (Q1-Q3) 6 (4–8) 6 (4–8) 6 (4–8) 0.493 Anxiolytics consumption, % (n) 42.22 (152) 45.92 (90) 37.80 (62) 0.300 Antidepressants consumption, % (n) 30.28 (109) 37.24 (73) 21.95 (36) 0.006** Use of health services (last 12 months) Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 7 of 12 Table 1 Characteristics of participants at the baseline by treatment group (Continued) Total (n= Intervention group (n= Control group (n= P value 360) 194) 164) Visits to nurse, median (Q1-Q3) 5.5 (2–10) 5 (2–9) 6 (3–11) 0.088 Visits to general practitioner, median (Q1-Q3) 9 (5–13) 9 (5–13) 8 (5–13) 0.912 Visits to emergency units, median (Q1-Q3) 0 (0–1) 0.5 (0–2) 0 (0–1) 0.120 Boldface indicates between groups statistical significance at baseline *P < 0.05 **P < 0.01 a SF-12: Short Form 12 Health Survey b Visual Analog Scale of the EuroQOL: Range 1–100. Higher values indicate higher self-perceived health c Appraisal of Self-Care Agency Scale (ASA-R): Range 24–120. Higher values indicate higher self-management d Social Resources Inventory in Older Adults. Number of sources of relationship from 0 to 5 (partner, children, other relatives, neighbours and friends) e Social Resources Inventory in Older Adults. Number of sources of relationship providing emotional/instrumental or informational support from 0 to 5 (partner, children, other relatives, neighbours and friends) f Este II Subjective Social Participation Index g Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS). Higher scores indicate a higher level of mental well-being h 11-item De Jong Gierveld Loneliness Scale i Geriatric Depression Scale (GDS-5). A score of 2 or more indicates depressive symptomatology both IG (2.86 [0.77, 4.95]) and CG (4.09 [1.72, 6.47]). in the visits to nurses in the IG, though it was also signifi- Similarly, emotional loneliness decreased in both IG (− cantly lower at the baseline and reached the same levels as 0.34 [− 0.60, − 0.08]) and CG (− 0.41 [− 0.70, − 0.13]). the CG. The use of antidepressants increased significantly from The promising results reported by this study in mental 21.95 to 29.70% in the CG (1.59 [0.33, 2.86]), while it health are supported by previous reseach which showed remained stable in the IG. Visits to nurse increased in that the participation in a community-based intervention the IG (1.72 [0.43, 3.02]) reaching a similar number of reduced depressive symptoms in older adults with comor- visits than the CG at baseline. Finally, when comparing bidity [38, 39]. Similarly, multi-domain interventions changes between T1 and T2 assessments (results not among community-living older persons and in primary shown in the table), social loneliness decreased in the IG care settings also reduced depressive symptomatology and (− 0.30 [− 0.49, − 0.11]). The decrease of depressive demonstrated a significant improvement in mental health symptomatology observed at T1 was maintained in the [40, 41]. The lack of results in self-management and social IG until T2 (− 0.95 [− 1.62, − 0.27]). capital differ from previous studies [42] which shown that strengthening social support in group-based interventions Discussion might improve self-management behaviours among socio- This study assessed whether a complex intervention im- economically deprived patients. In contrast, other re- proved self-perceived health in older adults living in urban searchers have reported that financial constraints and low disadvantaged areas, and its impact on self-management, socioeconomic status are barriers to effective self- health literacy, social capital, mental health, health-related management [43, 44]. The use of existing social support quality of life, and use of health resources. The primary groups has been also underlined as an acceptable and at- outcome self perceived health assessed using SF-12-1 was tractive method of delivering a self-management interven- not specifically affected by the intervention. Remarkably, tion to older people in socio-economically disadvantaged the same outcome measured with the EuroQOL visual areas [45]. The results obtained in health literacy are analog scale showed a significant increase at one-year in line with those reported from similar group-based follow-up only in the IG. Among secondary outcomes, the interventions [14]. Finally, there are several possible intervention seemed to improve health literacy in terms of explanations for the unexpected results in the use of understanding medical information and the mental com- health resources. Previous findings [46], showed a de- ponent of quality of life, and to decrease social loneliness crease in visits to the general practitioner in a group- and depressive symptoms. It is also important to note that based intervention targeting older people suffering antidepressants consumption increased at the follow-up from loneliness. Regarding the higher number of visits only in the CG, although we should be cautious in inter- to the nurse, other studies in the same context [47] preting to what extent the programme prevents the pre- found the same result and suggested that it might be scription of antidepressants by addressing depressive due to greater trust with these professionals as a con- symptoms with a non-medical intervention. Our research sequence of their role as observers of the interven- was unsuccessful in proving that the intervention had any tion, or due to a higher awareness of health problems effect on self-management, social capital, or visits to GP by participants thus increasing indeed the number of and emergency units. Unexpectadly, we found an increase appropriate visits. Table 2 Adjusted predictions for outcomes by group allocation and time of assessment (n = 358) Baseline (T0) Intra-group changes from T0 to T1 Intra-group changes from T0 to T2 Intervention Control Intervention Control Intervention Control Primary outcome (Self-perceived health) Excellent/Very Good/Good (SF-12-1)a, % (n) 1.53 (3) 3.66 (6) 5.15 (3.34, 6.97)** 3.27 (1.89, 4.66)** 5.01 (3.21, 6.82)** 3.13 (1.71, 4.56)** b EQ-5D visual analog scale 49.97 (18.62) 52.89 (18.49) 8.29 (4.91, 11.67)** 5.49 (1.80, 9.18)** 4.80 (1.09, 8.52)* 2.34 (− 1.74, 6.43) Secondary outcomes Health-related quality of life (SF-12)a Physical component, mean (SD) 36.59 (9.62) 36.3 (9.50) 1.57 (0.16, 2.97)* 2.95 (1.41, 4.50)** 1.22 (−0.34, 2.79) 1.36 (− 0.35, 3.08) Mental component, mean (SD) 38.39 (14.11)* 42.03 (12.45)* 3.77 (1.82, 5.73)** 0.38 (−1.77, 2.53) 1.13 (− 1.05, 3.31) 0.62 (− 1.76, 3.01) Blancafort Alias et al. BMC Public Health Self-management (ASA-R)c, mean (SD) 66.93 (12.07) 69.30 (11.97) 1.69 (−0.23, 3.62) 1.25 (−0.95, 3.44) 2.86 (0.77, 4.95)** 4.09 (1.72, 6.47)** Health literacy (HLS-EU) Understand medical information, % (n) 25.51 (50)** 15.85 (26)** −0.62 (−1.10, − 0.13)* −0.45 (− 0.99, 0.08) −0.25 (− 0.80, 0.29) −0.32 (− 0.92, 0.29) Find out about activities for mental well-being, % (n) 32.65 (64) 25.61 (42) −0.45 (− 0.92, 0.02) 0.34 (− 1.17, 0.85) −0.31 (− 0.82, 0.21) 0.18 (− 0.38, 0.74) (2021) 21:84 Assess healthy lifestyles, % (n) 17.35 (34) 14.03 (23) −0.24 (− 0.72, 0.23) −0.00 (− 0.51, 0.51) −0.08 (− 0.60, 0.44) 0.09 (− 0.48, 0.66) Social capital Extension of social networkd, mean (SD) 3.95 (0.86) 4.04 (0.86) −0.00 (− 0.09, 0.08) −0.02 (− 0.12, 0.07) −0.05 (− 0.15, 0.05) 0.00 (− 0.11, 0.11) Low social participatione, % (n) 29.71 (52) 28.66 (45) −0.23 (− 0.74, 0.28) −0.08 (− 0.65, 0.49) 0.21 (− 0.35, 0.78) −0.50 (− 1.14, 0.14) Mental health Emotional well-being (WEMWBS)f, mean (SD) 52.93 (8.65) 54.91 (9.23) 1.49 (−0.01, 2.98) 0.11 (− 1.60, 1.82) − 0.25 (− 1.86, 1.35) 0.08 (−1.84, 2.01) Emotional loneliness subscaleg, mean (SD) 2.63 (1.75) 2.53 (1.62) −0.12 (− 0.35, 0.11) −0.07 (− 0.32, 0.19) −0.34 (− 0.60, − 0.08)* −0.41 (− 0.70, − 0.13)** g Social loneliness subscale , mean (SD) 1.22 (1.32) 1.15 (1.22) 0.12 (− 0.04, 0.29) 0.02 (− 0.16, 0.20) −0.18 (− 0.36, 0.01) −0.13 (− 0.34, 0.08) Depressive symptomatology (GDS-5)h, % (n) 59.69 (117) 54.27 (89) −1.26 (− 1.90, − 0.63)** 0.03 (− 0.64, 0.70) −0.95 (− 1.62, − 0.27)** −0.67 (− 1.42, 0.08) Anxiolytics consumption, % (n) 45.92 (90) 37.80 (62) −0.37 (− 1.16, 0.43) 0.27 (− 0.62, 1.16) 0.54 (− 0.34, 1.41) −0.38 (− 1.40, 0.63) Antidepressants consumption, % (n) 37.24 (73)** 21.95 (36)** −0.20 (− 1.17, 0.77) 0.83 (− 0.31, 1.97) 1.05 (− 0.08, 2.17) 1.59 (0.33, 2.86)* Use of health resources Visits to nurse, mean (SD) 6.88 (7.04) 8.72 (9.26) N/A N/A 1.72 (0.43, 3.02)** −0.19 (− 1.64, 1.26) Visits to general practitioner, mean (SD) 9.66 (5.55) 9.68 (6.03) N/A N/A −0.36 (− 1.27, 0.54) − 0.12 (− 1.13, 0.89) Visits to emergency units, mean (SD) 1.24 (1.86) 0.91 (1.41) N/A N/A −0.18 (− 0.55, 0.19) 0.31 (− 0.10, 0.72) Boldface indicates between groups statistical significance at baseline *P < 0.05 **P < 0.01 a SF-12: Short Form 12 Health Survey b Visual Analog Scale of the EuroQOL: Range 1–100. Higher values indicate higher self-perceived health c Appraisal of Self-Care Agency Scale (ASA-R): Range 24–120. Higher values indicate higher self-management d Social Resources Inventory in Older Adults. Number of sources of relationship from 0 to 5 (partner, children, other relatives, neighbours and friends) e Este II Subjective Social Participation Index f Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS). Higher scores indicate a higher level of mental well-being g 11-item De Jong Gierveld Loneliness Scale: subscales for emotional loneliness score 0–6 and social loneliness score 0–5. Higher scores indicate higher levels of loneliness h Page 8 of 12 Geriatric Depression Scale (GDS-5). A score of 2 or more indicates depressive symptomatology Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 9 of 12 Fig. 2 Average predicted values for outcomes by treatment and time of assessment Strengths and limitations with previous results [52] that considers this indicator of To the best of our knowledge, this is the first randomised great importance for the design of programmes aimed at clinical trial addressing health inequalities by promoting improving older’s people health. self management, health literacy and social capital and We feel strongly that the study also provides a showing promising effects on mental health. It is plausible novel approach in terms of the participating actors, that some limitations might have influenced the results the problems tackled and the strategy applied. This obtained. Firstly, both grups differed at baseline in the reinforces previous findings in the literature [53] and mental component of SF-12, understanding medical infor- fairly supports the idea that a community-based col- mation, antidepressants consumption and number of visits laboration with primary care providers can improve to the nurse. However, randomization is the most robust health strategies, and would appear to indicate that method of preventing the selection bias that occurs when linking organized primary health care with the group which receive the intervention differ systematic- community-based interventions may be a promising ally from the control group, in ways likely to affect out- direction for research and practice. This strategy comes [48, 49]. Another possible source of bias is the high might be especially relevant considering the increasing loss to follow-up specially in the control group. Re- workload of primary care professionals with the grow- searchers have shown that recruitment and retention of ing proportion of older people with chronic diseases older adults in clinical trials is usually a challenging task [54]. In this sense, practical effectiveness has been due to their comorbidities as well as social and cultural appointed as a key issue when assessing a complex barriers, especially in low-income contexts [50, 51]. intervention [55]. Finally, this intervention can guide Hence, further research should address challenges such as initiatives aimed to improve a perception of health drop-out and lack of adherence as a common shared which is based on their strengths and needs and goes problem with other group interventions of health promo- beyond the chronic disease, and be transferred to tion in primary care settings, specially in deprived socio- other settings and become a useful resource to which economically contexts and older people. Thirdly, a major health professionals may refer older people, specially those source of contamination comes from the fact that partici- suffering psychosocial problems. To this aim, professionals pants could easily revealed their group allocation and thus from 16 primary care centers were trained as potential fa- blinding outcome assessors were hard to sustain, and thus cilitators of the intervention, and the Institut Català de la it might lead to biased estimates of treatment effect. Fi- Salut (Catalan Institute of Health), an organization nally, as the main eligibility criteria of the study, self- dependent on the Health Department of the Catalan Gov- perceived health could be misleading in terms of recruit- ernment, commissioned the research team to train during ment, given that health may have different meanings for 2019 and 2020 nearly a hundred professionals from Barce- older people and professionals. Our hypothesis is in line lona urban area. Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 10 of 12 Conclusions Received: 13 July 2020 Accepted: 20 December 2020 The results of this study provide evidence to inform pol- icy makers how to promote health among older adults in urban disadvantaged areas while addressing inequity. References 1. Borrell C, Ru M, Pasarn MI, Benach J, Kunst AE. La medición de las It has shown how a group intervention with a strong so- desigualdades en salud. Gac Sanit. 2000;14:20–. cial component, conducted in primary health care and 2. Walker A. Why is ageing so unequal? In: Cann PP, Dean M, editors. Unequal community assets, can lead to promising effects on men- ageing: the untold story of exclusion in old age. Bristol: The Policy Press; 2009. p. 141–58. tal health and may be used as a model for its biopsycho- 3. Artazcoz L, Rueda S. Social inequalities in health among the elderly: a social approach on health, with an emphasis on challenge for public health research. J Epidemiol Community Health. 2007; salutogenesis by means of promoting self-management, 61(6):466–7. https://doi.org/10.1136/jech.2006.058081. 4. Borrell C, Malmusi D, Artazcoz L, Diez E, Rodríguez-Sanz IP, Campos P, et al. health literacy and social capital. Propuesta de políticas e intervenciones para reducir las desigualdades sociales en salud en España. Gac Sanit. 2012;26:182–9. 5. Ndumbe-Eyoh S, Moffatt H. Intersectoral action for health equity: a rapid Supplementary Information systematic review. BMC Public Health. 2013;13:1056. https://doi.org/10.1186/ The online version contains supplementary material available at https://doi. 1471-2458-13-1056. org/10.1186/s12889-020-10094-9. 6. Berenguera A, Pons-Vigués M, Moreno-Peral P, March S, Ripoll J, Rubio- Valera M, et al. Beyond the consultation room: proposals to approach health Additional file 1. promotion in primary care according to health-care users, key community informants and primary care Centre workers. Health Expect. 2017;20(5):896– 910. https://doi.org/10.1111/hex.12530. Abbreviations 7. Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, IG: Intervention group; CG: Control group; MD: Mean difference; et al. A comparative risk assessment of burden of disease and injury CI: Confidence interval attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the global burden of disease Acknowledgements study 2010. Lancet. 2012;380(9859):2224–60. https://doi.org/10.1016/ The authors would like to thank health professionals from the following S0140-6736(12)61766-8. primary care centers for their support and involvement during the 8. Franco OH, Wong YL, Kandala NB, Ferrie JE, Dorn JM, Kivimäki M, Clarke development of the trial: Besòs, Blanes, Chafarinas, Cotxeres, Dr. Lluís Sayé, A, Donahue RP, Manoux AS, Freudenheim JL, Trevisan M, Stranges S. Horta, Horts de Miró, La Marina, Les Hortes, Marià Fortuny, Nova Lloreda, Cross-cultural comparison of correlates of quality of life and health Sardenya, Sant Ildefons, Sant Martí, Sant Pere, Torrassa. status: the Whitehall II study (UK) and the Western New York health study (US). Eur J Epidemiol. 2012;27(4):255–65. https://doi.org/10.1007/ Authors’ contributions s10654-012-9664-z. SB, LCP and RM coordinated the study. IM, MR and XR analyzed data. All 9. Murray CJ, Atkinson C, Bhalla K, Birbeck G, Burstein R, Chou D, et al. authors contributed to the interpretation of data and writing of the The state of US health, 1990-2010: burden of diseases, injuries, and manuscript. All authors provided feedback on the drafts of this paper and risk factors. JAMA. 2013;310(6):591–608. https://doi.org/10.1001/jama. have read and approved the final manuscript. 2013.13805. 10. Chida Y, Steptoe A. Positive psychological well-being and mortality: a Funding quantitative review of prospective observational studies. Psychosom Med. The study was funded by the Programme RecerCaixa (Call 2014), a joint 2008;70(7):741–56. https://doi.org/10.1097/PSY.0b013e31818105ba. initiative by the Associació Catalana d’Universitats Públiques (ACUP) and 11. Huppert FA. Psychological well-being: evidence regarding its causes and Obra Social la Caixa Grant number 2014ACUP00207. Study sponsor and consequences. Appl Psychol Health Well Being. 2009;1(2):137–64. https://doi. funders had no role in study design; collection, management, analysis, and org/10.1111/j.1758-0854.2009.01008.x. interpretation of data; writing of the report; and the decision to submit the 12. Araújo-Soares V, Hankonen N, Presseau J, Rodrigues A, Sniehotta FF. report for publication. Developing behavior change interventions for self-Management in Chronic Illness: an integrative overview. Eur Psychol. 2019;24(1):7–25. https://doi.org/ 10.1027/1016-9040/a000330. Availability of data and materials 13. Stranges S, Samaraweera PC, Taggart F, Kandala NB, Stewart-Brown S. Major The full trial protocol can be accessed in:https://bmcpublichealth. health-related behaviours and mental well-being in the general population: biomedcentral.com/articles/10.1186/s12889-018-5219-x#Bib1 the health survey for England. BMJ Open. 2014 Sep 19;4(9):e005878. https:// The datasets and materials used and/or analysed during the current study doi.org/10.1136/bmjopen-2014-005878. are available from the corresponding author on reasonable request. 14. Visscher BB, Steunenberg B, Heijmans M, Hofstede JM, Devillé W, van der Heide I, Rademakers J. Evidence on the effectiveness of health literacy Ethics approval and consent to participate interventions in the EU: a systematic review. BMC Public Health. 2018;18(1): The study design was approved by the Clinical Investigation Ethics 1414. https://doi.org/10.1186/s12889-018-6331-7. Committee of the IDIAP Jordi Gol. Participants were given a full explanation 15. Nyqvist F, Forsman A. Social capital as a health resource in later life: the of the study and signed written informed consent before their enrollment. relevance of context. Berlin, Heidelberg: Springer; 2015. p. 267. 16. Uphoff EP, Pickett KE, Cabieses B, Small N, Wright J. A systematic Consent for publication review of the relationships between social capital and socioeconomic Not applicable. inequalities in health: a contribution to understanding the psychosocial pathway of health inequalities. Int J Equity Health. 2013;12:54. https:// Competing interests doi.org/10.1186/1475-9276-12-54. The authors declare that they have no competing interests. 17. Geboers B, Reijneveld SA, Jansen CJ, de Winter AF. Health Literacy Is Associated With Health Behaviors and Social Factors Among Older Adults: Author details Results from the LifeLines Cohort Study. J Health Commun. 2016;21(sup2): 1 Fundació Salut i Envelliment UAB. Casa Convalescència, Sant Antoni Maria 45–53. https://doi.org/10.1080/10810730.2016.1201174. Claret, 17, 4a planta, 08041 Barcelona, Spain. 2Institute of Biomedical 18. Geboers B, de Winter AF, Luten KA, Jansen CJ, Reijneveld SA. The Research (IIB Sant Pau), Sant Quintí, 75-77, 08041 Barcelona, Spain. 3Equip association of health literacy with physical activity and nutritional d’Atenció Sardenya, EAP Sardenya, Sardenya, 466, 08025 Barcelona, Spain. behavior in older adults, and its social cognitive mediators. J Health Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 11 of 12 Commun. 2014;19(Suppl 2):61–76. https://doi.org/10.1080/10810730.2014. and gender surveys. Eur J Ageing. 2010;7(2):121–30. https://doi.org/10.1007/ 934933. s10433-010-0144-6. 19. Campbell NC, Murray E, Darbyshire J, Emery J, Farmer A, Griffiths F, Guthrie 37. Ortega Orcos R, Salinero Fort MA, Kazemzadeh Khajoui A, Vidal Aparicio S, B, Lester H, Wilson P, Kinmonth AL. Designing and evaluating complex de Dios del Valle R. Validación de la versión española de 5 y 15 ítems de la interventions to improve health care. BMJ. 2007;334(7591):455–9. https://doi. Escala de Depresión Geriátrica en personas mayores en Atención Primaria org/10.1136/bmj.39108.379965.BE. [Validation of 5 and 15 items Spanish version of the geriatric depression 20. Michie S, Abraham C. Interventions to change health behaviours: evidence- scale in elderly subjects in primary health care setting]. Rev Clin Esp. 2007; based or evidence-inspired? Psychol Health. 2004;19:29–49. https://doi.org/ 207(11):559–62. https://doi.org/10.1157/13111585. 10.1080/0887044031000141199. 38. Markle-Reid M, Ploeg J, Fraser KD, Fisher KA, Bartholomew A, Griffith LE, 21. Mackenbach JP. Bakker MJ; European network on interventions and policies Miklavcic J, Gafni A, Thabane L, Upshur R. Community program to reduce inequalities in health. Tackling socioeconomic inequalities in improves quality of life and self-Management in Older Adults with health: analysis of European experiences. Lancet. 2003;362(9393):1409–14. diabetes mellitus and comorbidity. J Am Geriatr Soc. 2018;66(2):263–73. https://doi.org/10.1016/S0140-6736(03)14639-9. https://doi.org/10.1111/jgs.15173. 22. O'Cathain A, Croot L, Duncan E, Rousseau N, Sworn K, Turner KM, Yardley L, 39. Saito T, Kai I, Takizawa A. Effects of a program to prevent social isolation on Hoddinott P. Guidance on how to develop complex interventions to loneliness, depression, and subjective well-being of older adults: a improve health and healthcare. BMJ Open. 2019;9(8):e029954. https://doi. randomized trial among older migrants in Japan. Arch Gerontol Geriatr. org/10.1136/bmjopen-2019-029954. 2012;55(3):539–47. https://doi.org/10.1016/j.archger.2012.04.002. 23. Coll-Planas L, Blancafort S, Rojano X, Roqué M, Monteserín R. Promoting 40. Ng TP, Nyunt MSZ, Feng L, Feng L, Niti M, Tan BY, Chan G, Khoo SA, self-management, health literacy and social capital to reduce health Chan SM, Yap P, Yap KB. Multi-domains lifestyle interventions reduces inequalities in older adults living in urban disadvantaged areas: protocol of depressive symptoms among frail and pre-frail older persons: the randomised controlled trial AEQUALIS. BMC Public Health. 2018;18(1): randomized controlled trial. J Nutr Health Aging. 2017;21(8):918–26. 345. https://doi.org/10.1186/s12889-018-5219-x. https://doi.org/10.1007/s12603-016-0867-y. 24. Miilunpalo S, Vuori I, Oja P, Pasanen M, Urponen H. Self-rated health status 41. da Dantas Bruno A. S, de Miranda Jessica MA, Cavalcante Anna CV, as a health measure: the predictive value of self-reported health status on Toscano Gislani a. da S, Torres Larissa SS, Rossignolo Simone C. de O, the use of physician services and on mortality in the working-age et al. impact of multidimensional interventions on quality of life and population. J Clin Epidemiol. 1997;50(5):517–28. https://doi.org/10.1016/ depression among older adults in a primary care setting in Brazil: a s0895-4356(97)00045-0. quasi-experimental study. Braz. J. Psychiatry. 2020;42(2):201–8. https:// 25. Burström B, Fredlund P. Self rated health: is it as good a predictor of doi.org/10.1590/1516-4446-2019-0577. subsequent mortality among adults in lower as well as in higher social 42. Vissenberg C, Nierkens V, van Valkengoed I, Nijpels G, Uitewaal P, classes? J Epidemiol Community Health. 2001;55(11):836–40. https://doi.org/ Middelkoop B, Stronks K. The impact of a social network based intervention 10.1136/jech.55.11.836. on self-management behaviours among patients with type 2 diabetes living 26. Solé-Auró A, Alcañiz M. Educational attainment, gender and health in socioeconomically deprived neighbourhoods: a mixed methods inequalities among older adults in Catalonia (Spain). Int J Equity Health. approach. Scand J Public Health. 2017;45(6):569–83. https://doi.org/10.1177/ 2016;15(1):126. https://doi.org/10.1186/s12939-016-0414-9. 1403494817701565. 27. Schulz KF, Altman DG, Moher D. CONSORT Group. CONSORT 2010 43. Bayliss EA, Steiner JF, Fernald DH, Crane LA, Main DS. Descriptions of statement: updated guidelines for reporting parallel group randomised barriers to self-care by persons with comorbid chronic diseases. Ann Fam trials. Version 2. BMJ. 2010;340:c332. https://doi.org/10.1136/bmj.c332. Med. 2003;1(1):15–21. https://doi.org/10.1370/afm.4. 28. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. 44. Liddy C, Blazkho V, Mill K. Challenges of self-management when living with Better reporting of interventions: template for intervention description and multiple chronic conditions: systematic review of the qualitative literature. replication (TIDieR) checklist and guide. BMJ. 2014;348:g1687. https://doi. Can Fam Physician. 2014;60(12):1123–33. org/10.1136/bmj.g1687. 45. Lawlor ER, Cupples ME, Donnelly M, Tully MA. Promoting physical activity 29. Vilagut G, Valderas JM, Ferrer M, Garin O, López-García E, Alonso J. among community groups of older women in socio-economically Interpretación de los cuestionarios de salud SF-36 y SF-12 en España: disadvantaged areas: randomised feasibility study. Trials. 2019;20(1):234. componentes físico y mental [Interpretation of SF-36 and SF-12 https://doi.org/10.1186/s13063-019-3312-9. questionnaires in Spain: physical and mental components]. Med Clin (Barc). 46. Pitkala KH, Routasalo P, Kautiainen H, Tilvis RS. Effects of psychosocial 2008;130(19):726–35. Spanish. https://doi.org/10.1157/13121076. group rehabilitation on health, use of health care services, and 30. Badia X, Schiaffino A, Alonso J, Herdman M. Using the EuroQoI 5-D mortality of older persons suffering from loneliness: a randomized, in the Catalan general population: feasibility and construct validity. controlled trial. J Gerontol A Biol Sci Med Sci. 2009;64(7):792–800. Qual Life Res. 1998;7(4):311–22. https://doi.org/10.1023/a: https://doi.org/10.1093/gerona/glp011. 1024933913698. 47. Coll-Planas L, Del Valle GG, Bonilla P, Masat T, Puig T, Monteserin R. 31. Alhambra-Borrás T, Durá-Ferrandis E, Garcés-Ferrer J, Sánchez-García J. The Promoting social capital to alleviate loneliness and improve health among appraisal of self-care agency scale - revised (ASA-R): adaptation and older people in Spain. Health Soc Care Community. 2017;25(1):145–57. validation in a sample of Spanish older adults. Span J Psychol. 2017;20:E48. https://doi.org/10.1111/hsc.12284. https://doi.org/10.1017/sjp.2017.52. 48. Collins R, MacMahon S. Reliable assessment of the effects of treatment on 32. Nolasco A, Barona C, Tamayo-Fonseca N, Irles MA, Mas R, Tuells J, et al. mortality and major morbidity, I: clinical trials. Lancet. 2001;357(9253):373– Alfabetización en salud: propiedades psicométricas del cuestionario HLS-EU- 80. https://doi.org/10.1016/S0140-6736(00)03651-5. Q16 [health literacy: psychometric behaviour of the HLS-EU-Q16 49. Eccles M, Grimshaw J, Campbell M, Ramsay C. Research designs for questionnaire]. Gac Sanit. 2018;34(4):399–402. https://doi.org/10.1016/j. studies evaluating the effectiveness of change and improvement gaceta.2018.08.006. strategies. Qual Saf Health Care. 2003;12(1):47–52. https://doi.org/10. 33. Diaz-Veiga P. Evaluación del apoyo social [Assessment of social support]. In: 1136/qhc.12.1.47. El F-BR, editor. Ambiente: Análisis Psicológico [the enviroment: 50. Mody L, Miller DK, McGloin JM, Freeman M, Marcantonio ER, Magaziner psychological analyses]. Madrid: Pirámide; 1987. p. 125–49. J, Studenski S. Recruitment and retention of older adults in aging 34. Rubio R, Rubio L, Pinel M. Un instrumento de medición de soledad social, research. J Am Geriatr Soc. 2008;56(12):2340–8. https://doi.org/10.1111/j. Escala Este II. Madrid: IMSERSO; 2009. 1532-5415.2008.02015.x. 35. Castellví P, Forero CG, Codony M, Vilagut G, Brugulat P, Medina A, 51. McMurdo ME, Witham MD, Gillespie ND. Including older people in et al. The Spanish version of the Warwick-Edinburgh mental well- clinical research. BMJ. 2005;331(7524):1036–7. https://doi.org/10.1136/ being scale (WEMWBS) is valid for use in the general population. bmj.331.7524.1036. Qual Life Res. 2014;23(3):857–68. https://doi.org/10.1007/s11136-013- 52. Girón P. Is age associated with self-rated health among older people in 0513-7. Spain? Cent Eur J Public Health. 2012;20(3):185–90. 36. De Jong GJ, Van Tilburg T. The De Jong Gierveld short scales for emotional 53. Leveille SG, Wagner EH, Davis C, Grothaus L, Wallace J, LoGerfo M, Kent and social loneliness: tested on data from 7 countries in the UN generations D. Preventing disability and managing chronic illness in frail older Blancafort Alias et al. BMC Public Health (2021) 21:84 Page 12 of 12 adults: a randomized trial of a community-based partnership with primary care. J Am Geriatr Soc. 1998;46(10):1191–8. https://doi.org/10. 1111/j.1532-5415.1998.tb04533.x. 54. Contel JC, Muntané B, Camp L. La atención al paciente crónico en situación de complejidad: el reto de construir un escenario de atención integrada [Care of the chronic patient in a complex situation: the challenge of building an integrated care scenario]. Aten Primaria. 2012;44(2):107–13. Spanish. https://doi.org/10.1016/j.aprim.2011.01.013. 55. Haynes B. Can it work? Does it work? Is it worth it? The testing of health care interventions is evolving. BMJ. 1999;319(7211):652–3. https://doi.org/10. 1136/bmj.319.7211.652. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.