Puig-Barrachina et al. BMC Public Health (2020) 20:345 https://doi.org/10.1186/s12889-020-8441-2 RESEARCH ARTICLE Open Access How does a targeted active labour market program impact on the well-being of the unemployed? A concept mapping study on Barcelona “Employment in the Neighbourhoods” Vanessa Puig-Barrachina1,2* , Davide Malmusi2,3, Lucía Artazcoz1,2,4,5, Xavier Bartoll1,4, Eva Clotet6, Imma Cortès-Franch1,2,4,7, Lorena Ventura6, Ferran Daban1,4,8, Èlia Díez1,2,4 and Carme Borrell1,2,4,5 Abstract Background: The “Employment in the neighbourhoods” program is an innovative, tailor-made Active Labour Market Program that has been implemented in 12 neighbourhoods in Barcelona (Spain). Its goal is to get people from deprived, high-unemployment neighbourhoods back to work. The aim of this study was to describe the effects of the program on participants’ quality of life, and identify the mechanisms underlying these effects, according to their own perception and the perception of technical staff who assisted them. Methods: We used Concept Mapping, a mixed methods approach combining qualitative and quantitative analysis, to develop a conceptual map of the participants’ and technical staffs’ perceptions about changes in the participants’ quality of life. Data collection occurred within the generation and structuring steps where participants brainstormed answers to a focus question, and then rated and sorted the responses. To create maps, we used Concept Systems Incorporated software, which conducted two main forms of analysis, a multidimensional scaling analysis, and a hierarchical cluster analysis. Results: Study participants reported several positive effects on mental health and emotional wellbeing, including self-esteem and empowerment, and considered that this was achieved through strengthened social networks, skills acquisition, emotional coaching, and personalized technical assistance. They also described some negative impacts, mainly related to the labour market situation. We observed marked gender differences in the discourses of program participants. Conclusions: The results obtained have allowed us to identify different perceived effects and mechanisms by which the “Employment in the Neighbourhoods” Active Labour Market Programme can influence quality of life of participants from the most deprived areas of Barcelona. Keywords: Active labour market programs, Unemployment, Barcelona City (Spain), Deprived neighbourhoods, Gender, Quality of life, Concept mapping * Correspondence: vpuig@aspb.cat 1 Agència de Salut Pública de Barcelona, Pl. Lesseps, 1, 08023 Barcelona, Spain 2 CIBER de Epidemiología y Salud pública (CIBERESP), Madrid, Spain Full list of author information is available at the end of the article © The Author(s). 2020 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. Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 2 of 16 Background effects are primarily due to poverty resulting from lost Introduction income and financial strain. Psychosocial factors include Since 2008, unemployment has risen sharply in many the loss of non-economic resources provided by employ- European countries due to the global economic crisis ment. According to Jahoda’s “latent functions” theory and the prolonged recession. In addition, the Euro- [9], while people primarily engage in paid work to earn pean Union imposed a political agenda of austerity their living (the manifest function of employment), they on Southern European countries in exchange for fi- also profit from five latent functions that are unintended nancial assistance, and this contributed to exception- by-products of employment, namely time structure, so- ally high unemployment rates in Spain, Italy, Greece, cial contact, collective purpose, social identity/status, and Portugal [1]. Unemployment has multiple strik- and activity. Moreover, uncertainty about one’s future ing social effects, including poor physical and mental work situation is also a stressor that leads to physio- health, and low subjective well-being and life satis- logical changes, risky health behaviours, and conse- faction [2], and is one of the most important social quently, poorer health [10, 11]. determinants of health inequalities [3]. During the 1990s, and for economic reasons, many ALMPs and health: effects, pathways and mechanisms countries introduced Active Labour Market Programmes ALMPs are thought to ease the negative effects of unemploy- (ALMPs) in an effort to reduce unemployment [4]. ment on health and wellbeing by promoting participation in ALMPs aim to reintroduce participants into the labour work-related activities [12]. ALMPs alter unemployed peo- market through vocational training courses, improved job ple’s environment by providing an opportunity to meet, search skills, work experience (internships and subsidized socialize, and strengthen social networks, and offer daily ac- jobs), and behavioural programmes, among others [4, 5]. tivities, routine and structure [12]. In terms of status, voca- In terms of health and wellbeing, ALMPs are thought to tional training courses may provide the perception of ease the negative effects of unemployment through differ- involvement in a legitimate alternative status to employment, ent psychosocial mechanisms [6]. namely education status [12]. Improved job search and re- A recent review showed that most studies that assessed lated skills, such as writing a CV and learning to respond in the health effects of ALMPs focused on ALMPs imple- a job interview, could increase one’s feeling of control over mented in Anglo-Saxon and Nordic countries [6]. However, one’s life course, and workplace participation through work the effects of the recent economic crisis were most strongly experience and training in a regular market could reduce the felt in Southern European countries, but to date little is psychosocial need for employment by providing a more known about the health impact of ALMPs in these countries. employment-like life situation [12]. To our knowledge, only one study has evaluated the health Based on the above theoretical knowledge, we can gen- effects of an ALMP implemented in Spain [7], and no similar erally expect involvement in an ALMP to have a positive study has been conducted in other Southern European coun- impact on health and wellbeing compared to open un- tries [6]. In addition, few reports on this topic discuss the employment situations (i.e. being unemployed not being mechanisms involved in this improvement. enrolled in any ALMP). A recent scoping review includ- “Employment in the Neighbourhoods” (EiN) is an ing 36 quantitative and qualitative studies conducted ALMP implemented in Barcelona, Spain, that aims to re- from the 1990s to the present day, showed mainly posi- duce social inequalities through labour market insertion tive health impact of ALMPs in high-income countries. of people living in neighbourhoods with the highest un- 80.6% of the studies included in the review found a posi- employment rates in the city, and who have the greatest tive impact on health or quality of life, and the rest material needs. A pre-post intervention study using showed no effects [6]. Of all types of the ALMPs imple- quantitative research methods showed improved mental mented, the only type that always showed a positive ef- health among participants, while self-perceived health fect on mental health was the one that offered job status remained stable or worsened [8]. search assistance accompanied by a psychological com- The general objective of this paper is to describe the effects ponent, such as improving self-confidence, self-image, or of EiN on quality of life and the mechanisms involved, ac- self-efficacy [6]. This result was the same for the differ- cording to the perception of the program’s participants and ent countries where it was implemented (Australia, the technical staff who assisted them. Technical staff mem- EUA, Finland). The rest of types showed controversial bers are those professionals who accompany participants results. The heterogeneity of contexts where these pro- through the entire process as personal guidance counsellor. grams were implemented and the heterogeneity of pro- grams’ content could partly justify these controversial Unemployment and health: pathways and mechanisms results. For example, the diverse effects of such pro- Unemployment has been shown to cause poor health grams on health in the same context are evidenced by through material and psychosocial pathways. Material Strandh et al. [12], who showed that in Sweden at the Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 3 of 16 end of the 90’s, unemployed individuals involved in open unemployment [19]. The authors argue that pater- workplace participation reported better mental health nalistic programmes that work through directive methods than those in open unemployment; however, those in- and impose unilateral obligations can affect individual’s volved in vocational training did not experience im- interpretations of their role and status in society. proved mental health. Gender and family also have a fundamental influence Few of the studies mentioned above explored the on the unemployment experience, with some authors mechanisms underlying the health changes caused by suggesting that non-work is less damaging in societies participation in the ALMP. Quantitative evaluation of where there is a stronger emphasis on family life [20, programs offering job search assistance and psycho- 21], such as for example the Southern European coun- logical component (e.g. inoculation against setbacks or tries. In these countries, where the Welfare State is increasing self-efficacy) reported positive results in terms weaker in terms of universal social policies, additional of Fryer’s theory of gaining mastery and control over material and non-material support is provided by the one’s life course [13]. Sense of mastery and a general in- family. In that case, we could hypothesise that ALMPs crease in labour market engagement could be related to are not considered to be as important for social support improved mental health [14]. This type of program has as in other countries, if the traditional family plays its only been implemented in the USA [15], Finland [14] role. On the other hand, in these countries, where the and Australia [16]. Similarly, qualitative evaluation of the Welfare State is less developed, women are still most re- programs offering job search activities to young people sponsible for informal care, while men still play the in Finland [17] reported positive results in terms of breadwinner role, at least in archetypal terms. It has Jahoda’s model of latent function, by providing a struc- been shown that in countries where gender roles are still ture of life [9]. The study showed how this type of pro- traditional and differentiated, unemployment has a grams help transform unemployment into a meaningful greater effect on males, because it impairs their individ- pattern of time use, especially through work-like activ- ual/social identity [3]. In that sense if ALMPs contribute ities or education. Moreover, in line with the idea of to supporting this identity, they could play a very im- sense of mastery and economic control, participants re- portant role among unemployed men. ported seeing new possibilities in the labour market, and In terms of culture, it has been proposed that the having economic support from the program. moral virtues of work characteristic of the Protestant ethic are central to the Nordic work ethic. In this cul- Contextual factors matter ture, work provides good role models and stimulates The impact of ALMPs may depend on various social and good behaviour [20]. Work ethics has an important in- political factors, such as the “model of activation” [18], gen- fluence on the significance of employment, probably on der roles, family as a social institution, and the culture of the unemployment experience, and thus, in the meaning work. “Model of activation” refers to the policy strategy and effects of ALMPs. aimed at reducing unemployment, as well as the correspond- In this sense, it is particularly relevant that most evalua- ing social protection system for the unemployed. There are tions of ALMPs were conducted in Anglo-Saxon and Scan- two major activation models, “work-first” or “liberal type” dinavian countries, and only one study has been carried out and the “universalistic type”. In “work-first” or liberal type, in Eastern Europe countries [22] and one in Southern [7]. the role of ALMPs and social policies is limited to inciting Thus, it is at least reasonable to question whether the health individuals to seek work, providing quick information and outcomes and the underlying mechanisms of ALMPs are simple matching services, investing in short-term vocational transferable to different contexts. There is a lack of empirical training, inciting people to be as active as possible through- and theoretical studies addressing these issues. out life, and accepting any job on offer [18]. On the other hand, the “universalistic type” of activation guarantees rela- ALMPs in Spain tively high standards of living for unemployed people and Spain is one of the countries that spends the smallest does not push them to accept any job in the market [18]. fraction of its GDP on ALMPs, which is thought to be The liberal model is typical of Anglo-Saxon countries, and an important determinant of health and wellbeing the universalistic of Scandinavia, while most other countries, among unemployed people [23]. Spain’s unemployment such as continental and Southern European countries, oscil- benefits are less generous than Nordic ones, and unlike late between these two. the liberal model of activation, unemployment benefits For example, in Germany “One-Euro-Jobs” is an are not compulsorily linked to ALMPs. In terms of gen- ALMP based on subsidized employment that includes der and family, Spain is characterized by a strong coercive elements in line with those of the liberal model. “familialism”, with a family solidarity model based on an Individuals who participate in this program did not per- asymmetric gender division of work and low female par- ceive improved integration in comparison to those in ticipation in the labour market. Spain has residual family Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 4 of 16 policies with lack of support for families, making them Thus, we expect EiN to ease the negative effects of un- rely on unpaid work [24]. To our knowledge, ALMPs in employment on participants’ health and wellbeing by Spain have not been evaluated in detail, with the excep- providing an opportunity to socialize and strengthen so- tion of one study [7] that evaluated the effects of the cial networks, and, for those who participate for at least Madrid Government’s minimum income program (IMI) several weeks, a routine and structure through daily ac- on health and wellbeing. tivities. More specifically, we expect them to have greater control over their life course through job search activities, the perception of involvement in a legitimate “Employment in the Neighbourhoods”: an ALMP to alternative to employment through professional skills reduce social inequalities in Barcelona, Spain training, and a reduction of the psychosocial need for In 12 neighbourhoods of Barcelona (Catalonia, Spain), employment through social integration support. an innovative tailor-made ALMP called “Employment in In order to describe the effects of EiN on quality of life the Neighbourhoods” (EiN) (“Treball als Barris” in Cata- and the mechanisms involved, we developed two con- lan) was implemented to get people from deprived, ceptual maps with a cluster approach based on the per- high-unemployment neighbourhoods back to work, and ceptions of the participants and technical staff using the to promote socioeconomic revitalization of these neigh- Concept Mapping technique. bourhoods. This program is part of a broader initiative, “The Neighbourhoods Law”, launched in 2004, to reduce Methods social inequality in Catalonia by improving the quality of The study uses a phenomenological design to explain life in specific neighbourhoods primarily through urban how people describe social phenomena through their ac- renewal projects (DPTOP, 2009). tual experiences [25]. We used Concept Mapping (CM), EiN is run by Barcelona City Council through its em- a mixed methods approach that combines qualitative ployment promotion agency, “Barcelona Activa”, and is and quantitative analysis [21]. Through CM we devel- partly financed by the Catalan Employment Service (Au- oped a conceptual map based on the perceptions of the tonomous Government of Catalonia). Enrolment in the participants and technical staff of EiN; EiN managers program is voluntary, and participants are required to be were excluded. CM has proven to be a useful method registered as a job seeker and to reside in one of the for evaluation, which can be used directly with partici- neighbourhoods where the program is implemented. pants to map how programs have affected them [26]. The program prioritises people at risk of exclusion, i.e. The study was conducted between March and July 2016 those aged > 50 years who are long-term unemployed, on Barcelona Activa premises. those under 35 years who left school early, those with The sample design followed a theoretical plan that de- low professional qualifications, women who have suf- termined the typologies of profiles of the sample units to fered gender violence, women who want to re-join obtain as much discursive variability as possible, thereby labour market, and people with physical disabilities. guaranteeing that different conceptual meanings of the The program enrols 1500–2000 people per year, most study phenomenon are collected [25]. We constructed of whom have a basic or no formal education. A high per- typologies according to the characteristics we considered centage of participants are foreigners, mostly non-EU na- would influence the meaning of the phenomenon under tionals, and most are long-term unemployed who are not study. In the case of participants, we also introduced two eligible for contributory unemployment benefits. Un- segmentation criteria, their employment situation after employed persons who wish to join the program first at- the program (unemployed vs working or having worked tend a personal interview with a trained professional to for at least three months), and their gender. We decided evaluate their occupational status and develop a tailored to preserve the gender stratification during brainstorm- job-search plan. The duration of the program depends on ing sessions due to possible differences in communica- each specific job-search, and can range from a few weeks tion patterns between genders. In mixed-gender public to a few months, with a maximum of one year. gatherings, men are known to spend more time talking The program’s methodology is based on customized than women and are more likely to interrupt, such that integration pathways that combine several insertion women generally have less confidence in their ability to measures. It mostly provides job search guidance and make arguments. Moreover, due to the gender division professional skills training to increase employability, and of labour, which is still present to a lesser or greater ex- promotes social and professional integration support for tent in most societies, the mechanisms that link the people with special difficulties, considering the specific intervention to health and well-being may differ between needs and potential of each neighbourhood and working men and women. Thus, we created four groups of EiN in close coordination with local associations and the participants: unemployed females, employed females, un- main social agents (See Table S1 for specific actions). employed males, and employed males. Within each of Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 5 of 16 the above-mentioned groups of participants, we aimed staff’s work would not be evaluated according to the ef- to achieve heterogeneous groups in terms of country of fects on participants’ health, in order to avoid reporting birth and neighbourhood of residence. Professionals and mostly positive effects. program users came from the nine decentralised job- Quality of life dimensions were considered as a start- search information points located in different neighbour- ing point for the discussion to describe effects and also hoods of Barcelona. mechanisms. Quality of life contains different dimen- In order to recruit the participants, two members of sions that can be considered mechanisms within a chain the research team went to the premises of “Barcelona of causality when considering unemployment situation Activa” to ensure confidentiality of personal data, where and the effects of ALMP. The research team did not they obtained the complete list of participants in the EiN introduce technical aspects, such as which items could during 2015. Participants who were still enrolled in the be considered effects and which considered other mech- program in 2016 were discarded. The lists were obtained anisms. However, as moderators of the brainstorming according to sex, neighbourhood of residence and coun- session, when two items were linked as a causal chain, try of birth. We randomly called person number 1, then we asked participants to verify this relationship. 11, then 21 and so on for each stratum, e.g. females born A minimum of two people from the research team in Spain from neighbourhood number 1. During the conducted the sessions. One moderated the session by telephone call, in addition to explaining the study and trying to allow everyone to participate and express their inviting them to participate, they were asked whether, opinions, covering most of the dimensions of the con- after carrying out the programme, they had worked for cept of quality of life, and facilitating both positive and at least three months. Recruitment stopped when 20 negative effects. The other took notes on the board and participants were obtained for each group. The ideal size commented on the ideas. All sessions were audio re- for the first phase of the CM, brainstorming, is about 10 corded and photographs were taken. participants for each group, and we expected a 50% loss. We collected participants’ demographic characteristics, Regarding the group of technical staff, we invited to including age, gender, educational level, employment status, participate, all who were currently working at the time both currently and after having participated in EiN, and the of the research study or had recently been working for a number of months of employment in case they were work- long period in EiN. ing. For technical staff the questionnaire collected data on age, gender, educational level, the number of months/years Concept mapping methodology employed as EiN professionals, and the number of years of Concept mapping comprises three data collection phases: experience as occupational staff. Participants received a 10- brainstorming, idea structuring, and participatory inter- journey-card for public transport as compensation and to pretation [26]. Data analysis is conducted between idea overcome the participation barrier. structuring and participatory interpretation. Qualitative Once all brainstorming sessions were completed, the re- data are collected during the brainstorming and interpret- search team reviewed the final statements to avoid repeti- ation steps, and quantitative analysis is then used to con- tion. Two lists of statements were created, the first from struct the maps in order to understand these ideas. ALMP participants and the second from technical staff. We systematically condensed all brainstorm responses Brainstorming from all participants, i.e. the lists from unemployed men, At the outset of the project, researchers agreed on the fol- unemployed women, employed men and employed lowing focus statement for the brainstorming activity: women, into a unique list of 53 statements; we condensed “One way the Employment in the Neighbourhoods pro- the responses from technical staff into 43 unique items. gram has influenced my quality of life is …” ; for ALMP technical staff, the focus statement was slightly modified Idea structuring (sorting and rating) as follows: “One way the ALMP has influenced the quality Approximately 2–3 weeks after the initial session, a sec- of life of “Employment in the Neighbourhoods” partici- ond session of approximately 2 h was held for each pants is …” . We held a brainstorming session of approxi- group to sort and rate responses. Separate sessions were mately 1 h for each of the five groups. Key terms were organized for EiN participants and technical staff. Dur- further clarified; the term “quality of life” was defined as a ing this session, participants worked individually, and 2 multidimensional construct that includes emotional well- to 4 researchers were present to assist with literacy is- being, interpersonal relations, material wellbeing, personal sues or other questions. First, participants were asked to development, physical wellbeing, self-determination, social sort the statements presented on cards into groups or inclusion, and rights [27]; influence was defined as either themes that made sense to them, and to name each pile positive or negative. The research team encouraged dis- created. ALMP participants sorted their own list of 53 cussions on negative experiences, and highlighted that statements, and technical staff sorted their list of 43. Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 6 of 16 Then, ALMP participants were given a rating sheet with the brainstorming is represented by a dot and identified by same corresponding statements and asked to rate the im- its corresponding number. The position of all statements portance in relation to quality of life of each corresponding tends to reflect the computed distances between them. statement on a 5-point Likert-type response scale: from [1] From here, the next step is to form clusters of state- little importance to [5] extremely important. In addition, par- ments that capture more general themes or concepts. ticipants were asked to rate the frequency with which they The Euclidian distances between points on the map are have experienced the statement from [1] never occurred to then grouped using Hierarchical Cluster Analysis (HCA) [4] occurred frequently. Technical staff were also given a rat- to form clusters of statements using Ward’s algorithm ing sheet to rate importance of their statements on the same [30]. Agglomerative HCA gives as many possible cluster Likert-type response scale; and the frequency of occurrence solutions as there are statements. The task for the ana- according to their professional knowledge, from [1] infre- lyst is to decide the number of clusters that the state- quent to [5] very frequent. ments should be grouped into for the final cluster solution. There is no mathematical way to select the Data analysis final cluster solution. Researchers choose the final clus- Using data from the second session, we generated maps and ter numbers solution based on the consistency between rankings using software from Concept Systems Incorporated the clusters’ contents, their knowledge on the issue, and [28]. To create the maps, the software conducts two main a group discussion with some participants on the per- forms of analysis, first a multidimensional scaling (MDS) ceived changes. In our case, the final group discussion, analysis followed by hierarchical cluster analysis [26]. The also known as participatory interpretation, was carried analysis begins by transforming the sorted piles of statements out with the group of technical staff only. Cluster names into a symmetric binary matrix of similarities for each par- were derived from the descriptive labels participants ticipant. A total similarity matrix is obtained by summing were asked to provide and the research team’s know- across the individual matrices. This similarity matrix shows ledge on the subject [26]. Finally, the average ratings re- the number of participants who sorted each pair of state- garding importance and frequency of each statement, ments together. As it is impossible to visualize k-dimensional and the average ratings for all statements in a cluster space, a multidimensional scaling (MDS) algorithm is used were calculated for EiN participants (stratified by sex) to transform the high-dimensional data into two- and for EiN professionals. dimensional space, while trying to preserve the pairwise dis- tances as much as possible [29]. The extent to which the ori- Participant characteristics ginal relative distances are preserved in the two-dimensional Four groups of ALMP participants were formed from a space is measured by the “stress” index. The more the MDS total of 41 participants (Table 1). These 41 individuals algorithm successfully preserved pairwise distances, the lower attended the brainstorming session, while only 33 of them the stress index. For concept mapping projects, stress values attended the sorting and rating session. Most foreign-born usually range from 0.205 and 0.365 [26]. High stress values participants came from South and Central America, imply greater complexity in the similarity matrix than can be followed by those born in Morocco, and a small represen- represented well in two dimensions, and that there was con- tation from Asia and Sub-Saharan Africa. Most partici- siderable variability in the way participants grouped the state- pants had elementary or post-elementary education, a few ments, or both. Stress index values were calculated for each had university-level education, and some were illiterate. of the point maps. Another issue considered when construct- The group of professionals included 9 members of tech- ing the map was the choice of cut-off level. The cut-off level nical staff of EiN. 8 women and 1 man attended the first is used to refine the analysis results by identifying instances brainstorming session, and 7 people attended the second of “outlying” sorts, on the item level. It is of great value when session (sorting and rating). Most of the technical staff it is difficult to interpret results with all data included. Thus, was trained in psychology, pedagogy, and social work. cut-off level filters out spurious relationships between state- There was high variability in their time and experience in ments. Only associations above the cut-off level are mapped. EiN, from 8 months to more than 8 years, as well as vari- For example, if a value of 2 is specified, this means that if ability in time-experience as labour market coaches. only two or fewer users placed two statements together in a The numbers and characteristics of participants are pile, the analysis will treat it as though no users placed the shown in Table 1. statements together. The choice of a high cut-off level sim- plifies the map by reducing the number of associations Results found; a low cut-off level may result in a complex map that Influence of the “Employment in the Neighbourhoods” is difficult to interpret. program on quality of life: participants’ perception Thus, the result of this step is a two-dimensional map A set of statements were generated by each group of of points, where each statement generated during the ALMP participants representing changes in their quality Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 7 of 16 Table 1 Demographic and socioeconomic characteristics of participants in all groups Women Men Unemployed (N = 7) Working (N = 11) Unemployed (N = 11) Working (N = 12) ALMP Participants (N = 41) Age Minimum 34 32 32 32 Maximum 57 55 56 62 Median 45 43 39 50 Country of birth Spain 2 6 3 7 Others 5 5 8 5 Number of hours in the program Minimum 7 9 3 14 Maximum 86 86 104 116 Median 11 27 23 55 Technical Staff (N = 9) Sex Female N=8 Male N=1 Age Minimum 31 Maximum 45 Median 43 Years of experience as a coach in insertional programs Minimum 5 Maximum 25 Median 10 Months of experience in “Employment in the Neighbourhoods” Minimum 7 Maximum 97 Median 36 of life as a result of the EiN program. The four groups of We observed marked differences between the dis- ALMP participants brainstormed a total of 53 statements, courses of males and females during the brainstorming which are organized in clusters in Table 2, ordered by fre- sessions. The female discourse emphasized psychological quency and importance rating, and stratified by sex. Fig- and emotional improvement (Cluster 1), in both the ure 1 illustrates the resulting cluster map. Since the employed and unemployed groups. Women highlighted similarity between participants’ perceptions resulting in the work of the professionals as one of the key factors similar items, we constructed a single map for all partici- that influenced their quality of life, with statements such pants. The final map showed 8 clusters: 1) “Improvement as, “having someone who cares about them” (SN 15), in mental health and well-being”, 2) “Emotional coaching”, “feeling accompanied” (SN 43), “feeling protection” (SN 3) “Strengthening of social network”, 4) “Personalized 44), “technical staff knows you personally” (SN 9), and technical assistance”, 5) “Acquisition of skills and re- “technical staff provides emergency support” (SN 36). sources to find a job”, 6) “Improvement in everyday life”, They related these items with “increased self-esteem and 7) “Negative emotions”, and 8) “Satisfaction in achieving confidence” (SN 1), “decreased depression” (SN 3), and goals”. The final point plot had a stress index of 0.34, with “decreased anxiety and stress” (SN 40). They also re- a similarity cut-off filter of 3. ported increased self-esteem and confidence (SN 1, Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 8 of 16 Table 2 Statements and clusters resulting from group sessions held with EiN participants and their scores based on criteria of importance (I), from [1] little importance to [5] extremely important and frequency (F) according to their own experience, from [1] never occurred to [4] occurred frequently Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 9 of 16 Fig. 1 Cluster map derived from participants perceptions of the impact of “Employment in the Neighbourhoods” on their quality of life Cluster 1) as a result of the learning process, through as they felt calmer and were able to forget about other prob- courses in new technologies (SN 13, Cluster 5), and lems while working. Consistent with this, one of the most interviewing and professional skills, among others (SN 6, important ideas among males was the feeling of frustration Cluster 5). Women highlighted the programs’ ability to when they did not achieve their objective (SN 23, Cluster 7). reactivate their social networks (Cluster 3). Being en- Notwithstanding their main objective, they appreciated rolled in the program and having someone waiting for other benefits derived from being enrolled in the pro- them acted as “an obligation to leave home” (SN 17). In- gram. They highlighted the need to be active, to occupy creasing their social network had a positive effect on men- their time and have a structured day (SN 32, Cluster 3), tal well-being, decreasing levels of anxiety and stress (SN which was met by the program. Another main idea 40, Cluster 1). Both groups of women also highlighted in- highlighted by men was the program’s lack of continuity creased employability after the program (SN 11, Cluster (SN 46, Cluster 7). They appreciated the program’s posi- 5), and having a job (SN 41, Cluster 8) as a key factor for tive effect on their emotional well-being but noted that quality of life. However, the women who had found a job this was not stable over time. Some of them referred to emphasised the precarious employment conditions (SN the 6 months duration of subsidized jobs, and 35, Cluster 7) and the consequent demoralization and loss highlighted the need for longer occupation plans and of mental well-being. During the rating session, women continued technical support and emotional counselling. again highlighted, as the most important and commonly During the men’s rating session, the most important and experienced items, increased employability (SN 11), an commonly experienced items were learning job-seeking improved attitude with their family (SN 53), and a feeling techniques (SN 6), satisfaction when finding a job (SN of enthusiasm when receiving job offers (SN 25). 14), and learning new technologies (SN 13). The male discourse during the brainstorming sessions fo- All participants (men and women) emphasized emo- cused on the importance of finding a job (SN 41) through tional proximity as a specific and important aspect of the program as the main source of well-being. Pathways to the EiN program, compared to other employment inte- improved quality of life were derived from being employed. gration programs they had previously attended. Accord- Economic stability was recognized as a source of material ing to all groups, this was the program’s most important well-being, i.e. meeting living standards such as housing. added value: having a person of reference who knows Men highlighted having a job as a source of identity (SN 52, them personally and could help them when needed. Cluster 8), and emphasized that they had a better relation- ship with their families when they had a job (SN 53, Cluster Influence of the “Employment in the Neighbourhoods” 1), due to their role of breadwinners. According to them, program on quality of life: perception of the technical their value as a family member increased when they could staff provide economic resources to their families. Moreover, go- The technical staff group generated a total of 43 statements ing to work helped improve their psychological well-being, on their perception of how participants’ quality of life is Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 10 of 16 influenced by the program. The Point Plot achieved an opti- impacts on mental health and emotional wellbeing. In mal stress index of 0.293, with a similarity cut-off filter of 1. contrast to the research team’s perspective, technical The concept mapping analysis mapped the 43 statements staff considered that, although the general objective of which are organized in clusters in Table 3, ordered by fre- EiN is reemployment, the program’s specific aims were quency and importance rating. The resulting cluster map is to improve participants’ confidence, self-esteem, and illustrated in Fig. 2. empowerment through emotional coaching and specific According to the technical staff, the program had a activities, as a path toward other employment insertion positive effect on participants’ “Mental health and emo- activities and increasing employability. Note that EiN tar- tional well-being” (Cluster 1) by decreasing anxiety and gets groups with significant difficulty in labour market in- despair (Statement Number, SN, 4), increasing autonomy sertion, such as people at risk of social exclusion and (SN 18), and improving their mood (SN 37) (among those in long-term unemployment, some of whom are re- others, See Table 3). It also improved their “Confidence, cruited through various social agents in the community, self-esteem and empowerment” (Cluster 2), and such as social services and street educators. Thus, some of “Strengthened their social networks” (Cluster 3) by pro- the participants are in need of more extensive assistance, moting positive relational attitudes (SN 25), and increas- beyond reemployment. In this respect, any improvement ing the sense of belonging to a group (SN 22). in our target population should be considered a success. Participants’ quality of life was also improved through “Acquisition of skills” (Cluster 6) and related items, in- Discussion cluding: a) technical skills, such as learning new tech- The results obtained using the concept mapping meth- nologies (SN 39) or professional skills (SN 35); b) daily odology have allowed us to identify different mecha- life skills, such as the ability to set specific goals (SN 6), nisms and effects by which the “Employment in the creating a daily routine (SN 29) and regular schedules Neighbourhoods” ALMP can influence the quality of life (SN 38), and improving transversal competencies, of participants from the most deprived areas of Barce- among others; and c) job-seeking skills, such as aware- lona according to their perceptions and those of profes- ness of job-seeking resources (SN 16), increased negoti- sionals. The effects described were perceived improved ation skills (SN 12), awareness of their labour rights (SN mental health and emotional wellbeing. Improved men- 31), as well as increasing employability (SN 36). In- tal health has also been reported following a pre-post creased employability was located near the cluster enti- evaluation using quantitative methodology [8]. Partici- tled “Achievements and satisfaction” (Cluster 7). pants’ also described negative emotions such as frustra- Technical staff also considered that EiN improved partici- tion and mood disorders linked to their failure to find a pants’ quality of life through “Emotional coaching and job or the cessation of subsidized employment. personalized technical assistance” (Cluster 5). They per- Some of the suggested mechanisms for the perceived im- ceived that participants feel accompanied (SN 32), and ap- provement on mental health and wellbeing were in line with preciate that someone listens to them (SN 33) and offers Jahoda’s theory [9], namely strengthened social networks, support without judging them (SN 24). Finally, there was daily structure, activity, and social identity. Some other also a cluster of negative impacts for participants’ well- mechanisms were in line with Fryer’s theory of increasing being, “negative emotions” (Cluster 4), which mainly re- sense of control [13], namely increased empowerment, self- sulted from frustration in not getting a job (SN 21). esteem, and confidence. One of the particularities of this The most important clusters, as rated by technical staff, study is that empowerment is manifested through increasing were Cluster 1 (improved mental health and emotional negotiation skills and the awareness of labour rights, in well-being) and Cluster 7 (achievement and satisfaction). addition to increasing several skills which has been suggested The clusters rated as most frequent were Cluster 5 (emo- by other studies [12]. This opinion was shared by profes- tional coaching and personalized technical assistance, after sionals and participants of EiN, although further developed excluding SN 7, “dependence on technical staff”), and by professionals. Cluster 7 (achievement and satisfaction). A somewhat new psychosocial factor highlighted by During the final group discussion, also known as par- both professionals and participants was the feeling of ac- ticipatory interpretation, which was only carried out with companiment. One of the specific characteristics that the group of professionals, the research team suggested differentiate EiN from other ALMPs, even in Barcelona, that two types of clusters could be interpreted from the is the use of a personal guidance counsellor to accom- data analysis. This first type of clusters is those related pany participants through the entire process. These re- to the specific program objectives (acquisition of skills, sults are consistent with those of previous qualitative emotional coaching and personalized technical assist- research, where the attitude of personnel was essential, and ance, and achievements), and second type is those re- where participants felt strong support from staff [17] and lated to other effects of the program, such as positive perceived them to be welcoming and non-judgmental [31], Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 11 of 16 Table 3 Statements and clusters resulting from group sessions held with technical staff and their scores based on criteria of importance, from [1] little importance to [5] extremely important and frequency of occurrence according to their professional knowledge, from [1] infrequent to [5] very frequent *SN 7 in Cluster 5 was considered the only negative statement in the cluster. The average frequency and importance of the Cluster without this SN are 4.23 and 4.10, respectively **SN 19 was related to the skills acquisition cluster for some of the participants and with improving confidence, self-esteem and empowerment cluster for some others. The final consensus was to leave the item in between these clusters. Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 12 of 16 Fig. 2 Cluster map derived from technical staff perceptions of the impact of “Employment in the Neighbourhoods” on their quality of life ideas that also appeared in our study. The reported conse- an indicator of health-related quality of life [37], individ- quences were not feeling alone, improved self-confidence uals assess self-perceived health as physical health. Phys- and self-esteem, and increased optimism and joy [31]. Social ical health is much more difficult to change than mental support has been suggested as an important health- health due to adaptive disorders. protecting factor among the unemployed [32]. EiN provided social support through both the technical staff’s support and Gender differences of ALMPs on health a strengthened social network. Although there is extensive literature on the differential Finally, economic stability, as a material pathway, was effects of unemployment between genders, less is in highlighted by individuals who took subsidized employ- known about the gender differential effects of ALMPs ment, which also had a negative facet, namely its short- on health. One of the few published studies, a cross- term duration. sectional study from the UK, showed that life worth and Although we observed mostly positive perceptions re- happiness was higher among employed than un- lated to quality of life, the quantitative pre-post interven- employed male ALMP participants, and among females, tion study evaluation showed that self-perceived health participation had a marked effect on life satisfaction [5]. status remained stable or worsened [8]. Few studies have Rojdalen et al. [38] also designed a non-experimental analyzed the effect of ALMPs on self-perceived health study based on retrospective judgments in Sweden, tak- status, and those that have done so also found no statis- ing gender differences into account. Mental health was tically significant changes in self-perceived health status more often judged by women and less well-educated following the program [33–35]. Our first hypothesis is people as being positively effect of training. Yet neither that changes in self-perceived health could take more of these studies provided a sociological reflection on the time to emerge. EiN had a maximum duration of 1 year, mechanisms. A qualitative study among women from and most of participants were only spending a few weeks East London [39], where most programs are compulsory, or months. The post-intervention questionnaire was car- showed that due to the temporary nature of the pro- ried out one year after the beginning of the intervention. grammes, participants experienced feelings of insecurity, Our second hypothesis is that individuals assess self- financial strain, role conflicts, and concerns about non perceived health as a combination of information from maternal childcare. Mothers felt guilty for leaving their specific health problems, general physical functioning children during the training. and health behaviors [36]. Although the WHO definition In our study, the women related their improvement to of health as “a state of complete physical, mental and so- the ability of the program to reactivate their social net- cial well-being and not merely the absence of disease work, to their perceived increase in self-esteem and con- and infirmity” would categorize self-perceived health as fidence through the learning process, and to their feeling Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 13 of 16 of accompaniment and protection by the technical staff. better than no job” [9]. The simplistic dichotomy between However, they also highlighted the health disadvantages employment and unemployment is rather more complex of entering the labour market, given the health impacts than viewing unemployment as “bad” and employment as of precarious employment, especially long working “good”. This is particularly pertinent given evidence that the hours, which could enter into conflict with their family labour market is often dominated by low pay and limited responsibilities. On the other hand, men related their sustainability, such that reemployment in a precarious pos- improvement to the greater ability to find a job. Eco- ition could be more detrimental for health and quality of life nomic stability is recognized as one of the most import- [42]. Thus, there is a need for a comprehensive program for ant pillars for men’s well-being, and they highlight their unemployed individuals, including economic benefits and ac- job as a source of identity, again highlighting the bread- tivities to reintroduce them into the labour market. A com- winner role. This appears to be specific to Spanish social prehensive program could trigger both material and context, and as we hypothesized, could be attributed to psychosocial mechanisms to cushion the effects of un- gender roles and the configuration of the family in Spain. employment. These measures should be accompanied by ef- No less important is the fact that both men and women fective policies to improve job quality. mentioned how the ALMP changed their family relation- ship, which highlights how important it is for ALMPs in Strengths and limitations this specific context to account not only for the individual Concept mapping is a strong method for analysing per- but also the relationship with the family, in order to be ceptions. Our study introduces the double perspective of more effective, at least in terms of quality of life. professionals and participants, which enriches the ana- lysis and highlights gender differences. Moreover con- Labour market situation cept mapping is very useful evaluation technique, as it In some circumstances, programs that promote reem- can help to develop a model or theory for change. The- ployment are not feasible, either because of a lack of jobs ory of change explains how the activities undertaken by due to economic conditions or a lack of skills in specific a program, such as an ALMP, contribute to a chain of disadvantaged populations. The Spanish labour market results that lead to the perceived effects. Thus concept is characterized by persistent high unemployment and mapping enables us to evaluate a complex intervention precarious employment conditions, including more tem- and complements the quantitative assessment to better porary work, low salaries, vulnerability, and inability to understand its results [8]. Finally, maps are easy to read exercise rights [40]. Labour market reforms imple- and interpret by decision-makers and the general public. mented during the last years of the current economic re- To our knowledge, this is one of the few health-related cession have worsened this situation, increasing social evaluations of ALMPs in Spain, and in other Southern inequality. Immigrants and manual and poorly-educated European countries. Thus, our study makes an import- workers, as well as women and young people, are over- ant contribution because, compared to Northern Europe, represented in the most precarious jobs [41], and these Southern European countries have different welfare sys- groups are the target population of EiN. In this scenario, tems and different gender and family regimes. the program highlighted negative emotions of frustration This study also has some methodological limitations, and discouragement when participants were not reem- the most important one being the participants’ difficulty ployed, or reemployed in precarious jobs. These negative in reading comprehension and conceptualization skills emotions emerged as a specific cluster in both the tech- due to their educational background, or because they nical staff and participants groups, and were more im- were non-native speakers of Spanish or Catalan. Al- portant in the latter. In addition, finding a job in though this was not problematic for the brainstorming precarious employment improves wellbeing at first, but session, it was the main drawback during the rating and this is disrupted by the reality of precarious labour mar- sorting session, in that some participants were not able ket conditions, or at the end of the 6–month term of to sort the statements with respect to their similarity to jobs subsidized by the program. In this sense, one of the some overarching concept. This is reflected in the clus- participants’ specific demands was that the program be ter map, in that the distance between statements in each continued, and that they maintain the counsellor or per- cluster are longer in the participants’ map, than in the son of reference after completing the insertion itinerary. professionals’ one. Moreover, the use of a cut-off filter of This suggests that the positive effects of EiN are prob- 3 in the final map indicates the difficulty in interpreting ably not maintained beyond the end of the programme. it due to spurious relationships between statements. We However, this would be a hypothesis to be tested by suspect that there was a lot of noise in the sorts, and means of a cohort study. that it would be useful to re-compute maps filtering out Considering labour market characteristics and health and these spurious relationships, leaving only the most fre- wellbeing, we need to reflect on the notion that “any job is quently made associations. Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 14 of 16 Also, participants’ difficulty in reading comprehension to construct a quantitative tool for measuring the mecha- and conceptualization skills was the main reason for not nisms underlying ALMPs, and to validate this in terms of conducting the final group discussion with EiN partici- psychometric properties. This could be implemented in a pants, i.e. the participatory interpretation session. In this quantitative longitudinal cohort to properly evaluate the session, researchers shared different map solutions with mechanisms and effects of EiN and similar programs. participants, discussed their interpretation, and possible cluster names. This last step in the methodology, al- Final remarks though not compulsory, increases the exchange of know- The impact of ALMPs in relation to their costs is often ques- ledge between researchers and participants. Thus, this last tioned. Of course, the criteria for success of an ALMP de- step was only conducted with professionals but not EiN pend on the theoretical perspective on which the evaluations participants. In this case, the research team choose the are based. These criteria could be the degree of participation final map and the names of the clusters. On the other in activation programmes, which would increase participants’ hand, the EiN participants showed extremely rich gener- chances of reemployment [43]. From a public health per- ation of ideas, most of which were consistent with the spective, and from a “health in all policies” approach [44], technical staff’s perceptions. After finalizing the study, we improved health and wellbeing should also be one of the cri- would recommend other qualitative techniques when teria for success. There is generally limited scientific evidence working with this specific population; nonetheless, it was in this field, especially in relation to labour market policy- an excellent technique for working with professionals. makers’ inclusion of health and wellbeing as a measure of Regarding internal validity, while we emphasised the the success of ALMPs. Clearly, there is a societal benefit to importance of discourse variability, our results may be evaluating programs and policies, not purely in economic prone to self-selection bias. Participants who decided to terms, but also in terms of health and wellbeing. take part in the study could be those ones with the high- est satisfaction, although it was not possible to control Conclusions this, as no register of those who left the programme was The “Employment in the Neighbourhoods” program showed available. Moreover, professionals may feel pressure to important advances in the quality of life of participants ac- point out more positive effects than negative ones. cording to their perceptions and those of technical staff, Regarding external validity, this study may not be represen- mainly in terms of better perceived mental health and emo- tative for the entire population participating in EiN. The aim tional wellbeing. The mechanisms for these perceived im- of the technique is to achieve maximum discourse variability provements were skills acquisition, strengthened social to ensure that different conceptual meanings are collected. To networks, emotional coaching and personalized technical as- accomplish this aim, we conducted theoretical sampling and sistance, and providing social support. Moreover, our study ensured that socio-demographic characteristics were similar to provides some insights into the strengths and weakness of those of the entire population. Moreover, we conducted two the program, which would be useful for improving it. different concept mappings, one with participants and one with professionals, with similar results. However, a more Supplementary information complete concept to increase external validity might be Supplementary information accompanies this paper at https://doi.org/10. 1186/s12889-020-8441-2. achieved by repeating the same procedure several times, in dif- ferent years, and with different groups. Note that discourse Additional file 1: Table S1. Summary of actions in the “Employment in saturation was reached after conducting the four groups of the Neighbourhoods” program ALMP’s participants. Abbreviations Recommendations ALMP: Active Labour Market Programme; GDP: Gross Domestic Product; MDS: Multidimensional Scaling; SN: Statement Number The mechanisms related to ALMPs is still an underexplored topic, especially from the epistemological perspective of situ- Acknowledgements ated knowledge, creating opportunities for challenging re- The authors thank Patricia Altimira, Brenda Biaani León, Victoria Porthé, María search. There are two immediate research challenges. First, Salvador-Piedrafita, and Hugo Vásquez for their contributions. based on the results of the concept mapping, we suggest Authors’ contributions conducting in-depth interviews with EiN participants in Conception and design: VPB, CB, DM, LA, XB, EC, IC, LV, FD, ED. Collection and order to further explore the chain of mechanisms and trig- assembly of data: VPB, CB, DM, XB, EC, IC, FD, ED. Analysis and interpretation of the data: VPB, CB, DM. Drafting of the article: VPB. Critical revision of the article gers. It would be extremely useful to conduct a qualitative for important intellectual content: VPB, DM, CB, LA, IC. Final comments and longitudinal study to follow participants from the moment approval of the article: VPB, CB, DM, LA, XB, EC, IC, LV, FD, ED. before entering the program, at different cut-points during Funding the program, and a few months after it ends. Second, using The research leading to these results has received funding from RecerCaixa information from the qualitative study, it would be possible and CIBER de Epidemiología y Salud pública (CIBERESP). The funding bodies Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 15 of 16 have not participated in any phase of the study (design, collection of data, 14. Vuori J, Silvonen J, Vinokur AD, Price RH. The Työhön job search program in analysis and interpretation, writing the manuscript). Finland: benefits for the unemployed with risk of depression or discouragement. J Occup Health Psychol. 2002;7(1):5–19. Availability of data and materials 15. Vinokur AD, Schul Y. Mastery and inoculation against setbacks as active The datasets used and/or analysed during the current study are available ingredients in the JOBS intervention for the unemployed. J Consult Clin from the corresponding author on reasonable request. Psychol. 1997;65(5):867–77. 16. Harry JM, Tiggemann M. The Psychological Impact of Work Re-entry Ethics approval and consent to participate Training for Female Unemployed Sole Parents. Aust J Soc Issues. 1992;27(2): All participants gave written informed consent, and data collection was 75–91. Available from:[cited 2018 Feb 20] http://doi.wiley.com/https://doi. approved by the Clinical Research Ethical Committee of Parc de Salut Mar org/10.1002/j.1839-4655.1992.tb01279.x. (Number 2015/6032/I). 17. Björklund O, et al. Young Finnish unemployed men’s experiences of having participated in a specific active labor market program. Am J Men’s Health. Consent for publication 2017;11(5):1426–35. https://doi.org/10.1177/1557988315604020. Not applicable. 18. Barbier J-C. Systems of social protection in Europe: two contrasted path to activation, and maybe a third. In: Lind J, Knudsen H, Jörgensen H, editors. Competing interests Labour and employment regulation in Europe. Brussels: P.I.E. Peter Lang; The authors declare that they have no competing interests. 2004. p. 233–53. 19. Gundert S, Hohendanner C. Active labour market policies and social Author details integration in Germany: do “one-euro-jobs” improve individuals’ sense of 1 Agència de Salut Pública de Barcelona, Pl. Lesseps, 1, 08023 Barcelona, social integration? Eur Sociol Rev. 2015;31(6):780–97. Spain. 2CIBER de Epidemiología y Salud pública (CIBERESP), Madrid, Spain. 20. Paugam S, Russel H. The effects of Employment Precarity and 3 Ajuntament de Barcelona, Barcelona, Spain. 4Institut d’Investigació Unemployment on Social Isolation. In: Gallie D, Paugam S, editors. Welfare Biomèdica (IIB-Sant Pau), Barcelona, Spain. 5Universitat Pompeu Fabra, Regimes and the Experience of Unemployment in Europe. Oxford: Oxford Barcelona, Spain. 6Barcelona Activa, Barcelona, Spain. 7Program in Univ Press; 2000. p. 243–64. Methodology of Biomedical Research and Public Health, Universitat 21. Tattarini G, Grotti R, Scherer S. The buffering role of the family in the Autònoma de Barcelona, Barcelona, Spain. 8Program in Biomedicine, relationship between job loss and self-perceived health: Longitudinal results Universitat Pompeu Fabra, Barcelona, Spain. from Europe, 2004–2011. Heal Place. 2018;52:55–61. https://doi.org/10.1016/ j.healthplace.2018.04.006. Received: 19 August 2019 Accepted: 2 March 2020 22. Bonin H, Rinne U. “Beautiful Serbia” - objective and subjective outcomes of active labour market policy in a transition economy. Econ Transit. 2014;22(1):43–67. 23. Stuckler D, Basu S, Suhrcke M, Coutts A, Mckee M. The public health effect References of economic crises and alternative policy responses in Europe: an empirical 1. Zamora-Kapoor A, Coller X. The effects of the crisis: why southern Europe? analysis. Lancet. 2009;374(9686):315–23 Available from: http://www.ncbi.nlm. American behavioral scientist; 2014.. nih.gov/pubmed/19589588 [cited 2015 Jan 13]. 2. Buffel V, Van de Velde S, Bracke P. The mental health consequences of the 24. Ferrera M. The’Southern model’of welfare in social Europe. J Eur Soc Policy. economic crisis in Europe among the employed, the unemployed, and the 1996;6(1):17–37. non-employed. Soc Sci Res. 2015;54:263–88. 25. Patton MQ. Qualitative research and evaluation methods. 3rd ed. London: 3. Puig-Barrachina V, Malmusi D, Marténez JM, Benach J. Monitoring social Sage Publications, Inc.; 2002. p. 598. determinants of health inequalities: the impact of unemployment among 26. Kane M, Trochim W. Concept mapping for planning and evaluation. Applied vulnerable groups. Int J Health Serv. 2011;41(3):459–82 Available from: Social Research Method Series, vol. 50. California: Sage Publications; 2007. p. http://www.ncbi.nlm.nih.gov/pubmed/21842573 [cited 2016 Aug 29].. 200216. 4. Filges T. Active Labour Market Programme Participation for Unemployment. 27. Verdugo MA, Schalock RL, Keith KD, Stancliffe RJ. Quality of life and its Campbell. [Internet]. 2015; Available from: http://www.campbellcollaboration.org/ measurement: important principles and guidelines. J Intellect Disabil Res. lib/download/3654/Filges_ALMP_Review.pdf [cited 2015 Oct 13]. 2005;49(10):707–717. Available from: [cited 2017 Jan 18] http://doi.wiley. 5. Sage D. Do Active Labour Market Policies Promote the Well-Being, Health com/https://doi.org/10.1111/j.1365-2788.2005.00739.x. and Social Capital of the Unemployed? Evidence from the UK. Soc Indic 28. Concept Systems Incorporated. The Concept System Global Max [Internet]. Res. 2014;124(2):319–37 Available from: http://link.springer.com/10.1007/s112 Ithaca; 2016. Available from: http://www.conceptsystemsglobal.com/. 05-014-0788-2 [cited 2016 Feb 19]. 29. Bar H, Mentch L. R-CMap—an open-source software for concept mapping. 6. Puig-Barrachina V, Giro P, Artazcoz L, Bartoll X, Cortès-Franch I, Fernandez A, Eval Program Plann. 2017;60:284–92. https://doi.org/10.1016/j.evalprogplan. et al. A scoping review of the impact of active labour market policies on 2016.08.018. health outcomes. Eur J Pub Health. 2020;1(30):36–42. 30. Everitt B. Cluster analysis. 2nd ed. London: Heinemann Educational Books; 1980. 7. Ayala L, Rodríguez M. Health-related effects of welfare-to-work policies. Soc 31. Stjernswärd S, Bernce R, Östman M. “Young Women”: The Meaning of a Sci Med. 2013;93:103–112. Available from: http://dx.doi.org/https://doi.org/ Collaborative Program Supporting Young Women’s Rehabilitation and 10.1016/j.socscimed.2013.05.037. Reintegration into the Labor Market. Soc Work Public Health. 2013;28(7): 8. González-Marín P, Puig-Barrachina V, Bartoll X, et al. Employment in the 672–84 Available from: http://www.ncbi.nlm.nih.gov/pubmed/24074131 neighborhoods of Barcelona: health effects of an active labor market [cited 2016 Dec 29]. program in Southern Europe [published online ahead of print, 2019 Dec 32. Milner A, Krnjacki L, Butterworth P, LaMontagne AD. The role of social 15]. J Public Health (Oxf). 2019;fdz169. https://doi.org/10.1093/pubmed/ support in protecting mental health when employed and unemployed: a fdz169. longitudinal fixed-effects analysis using 12 annual waves of the HILDA 9. Jahoda M. Employment and unemployment. A social-psychological analysis. cohort. Soc Sci Med. 2016;153(C):20–6. Cambridge: Press Syndicate of the University of Cambridge; 1982. p. 158. 33. Huber M, Lechner M, Wunsch C, Walter T. Do German welfare-to-Work 10. Levi L. Psychosocial environmental factors and psychosocially mediated Programmes reduce welfare dependency and increase employment? vol. effects of physical environmental factors. Scand J Work Environ Health. 12. Leibniz: German Economic Review; 2011. 1997;23(Suppl 3):47–52. 34. Reynolds C, Barry MM, Nic Gabhainn S. Evaluating the Impact of the 11. Sverke M, Hellgren J, Naswall K. No security: a meta-analysis and review of job Winning New Jobs Programme on the Re-employment and Mental Health insecurity and its consequences. J Occup Health Psychol. 2002;7(3):242–64. of a Mixed Profile of Unemployed People. Int J Ment Health Promot. 2010; 12. Strandh M. State intervention and mental well-being among the 12(2):32–41. Available from: [cited 2018 Feb 20] https://www.tandfonline.com/ unemployed. J Soc Policy. 2001;30(1):57–80. doi/full/https://doi.org/10.1080/14623730.2010.9721812. 13. Fryer D. Employment deprivation and personal agency during 35. Sage D. Do active labour market policies promote the subjective well-being unemployment: a critical discussion of Jahoda’s explanation of the of the unemployed? Evidence from the UK National Well-Being Programme. psychological effects of unemployment. Soc Behav. 1986;1:3–23. J Happiness Stud. 2015;16(5):1281–98. Puig-Barrachina et al. BMC Public Health (2020) 20:345 Page 16 of 16 36. Quesnel-Vallée A. Self-rated health: caught in the crossfire of the quest for “true” health? Int J Epidemiol. 2007;36(6):1161–4. 37. CDC. Healthy Days Methods and Measures | HRQOL | CDC [Internet], vol. 11; 2018. p. 83. Available from: https://www.cdc.gov/hrqol/methods.htm [cited 2020 Feb 18]. 38. Röjdalen G, Gelin G, Ivergård T. Self-assessed changes in mental health and employment status as a result of unemployment training. Appl Ergon. 2005; 36(2):145–55. 39. Coutts AP. Active Labour Market Programmes ( ALMPs ) and health : an evidence-bas. Review prepared for the Strategic Review of Health Inequalities in England Post 2010 ( Marmot Review ), vol. 2010; 2009. 40. Vives A, Amable M, Ferrer M, Moncada S, Llorens C, Muntaner C, et al. Employment precariousness and poor mental health: evidence from Spain on a new social determinant of health. J Environ Public Health. 2013;2013:978656. 41. Puig-Barrachina V, Vanroelen C, Vives A, Martínez JM, Muntaner C, Levecque K, et al. Measuring employment precariousness in the European Working Conditions Survey: the social distribution in Europe. Work. 2014;49(1):143–61 Available from: http://www.ncbi.nlm.nih.gov/pubmed/23803435 [cited 2016 Feb 22]. 42. Van Aerden K, Gadeyne S, Vanroelen C. Is any job better than no job at all? Studying the relations between employment types, unemployment and subjective health in Belgium. Arch Public Heal. 2017;75(1):55 Available from: http://www.ncbi.nlm.nih.gov/pubmed/28852501 [cited 2017 Sep 12]. 43. Breidahl KN, Clement SL. Does Active Labour Market Policy have an Impact on Social Marginalization? Soc Policy Adm. 2010 ;44(7):845–864. Available from: [cited 2017 Jan 23] http://doi.wiley.com/https://doi.org/10.1111/j.1467-9515.2010.00747.x. 44. Cook S, Leppo K, Ollila E, Peña S, Wismar M. Health in all policies: seizing opportunities, implementing policies. Malta: Ministry of Social Affairs and Health in Finland; 2013. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.