Individual, Group and Organised Experiential Expertise in Recovery from Intimate Partner Violence and Mental Health Problems in Finland

This paper examines recovery from intimate partner violence and mental health prob-15 lems through the perspective of experiential expertise (EE). The aim is to investigate the formation of EE in the context of empowerment-oriented recovery in non-governmental organisations (NGOs). This triangulated study includes both quantitative and qualitative data. The quantitative part was acquired through a survey on people with personal experience of mental health problems ( N ¼ 133) and the qualitative 20 part was collected from a development project including sixty female victims of intimate partner violence and nine professionals who worked with the women. First, the data were analysed separately through statistical analysis and hermeneutic close reading. Second, a triangulated synthesis was made. There are three types of EE: individual, group and organised. These provide a basis for personal and social 25 empowerment. EE consists of emotional and influential ways of acting, which enable renewed action as well as the reformation of a self-image. Our findings indicate that the relation between EE and recovery is reciprocal and mutually beneficial. We argue that especially group and organised EE should be integrated more firmly into social work practices, preventive social policy and structural social work.


Introduction
This paper examines recovery from intimate partner violence and mental health problems through the perspective of experiential expertise (hereafter referred to as EE). There are differences between experiencing 5 and confronting intimate partner violence and mental health problems, and the phenomena are laden with nuances. However, they have shared and complementary characteristics, and they can be interrelated. Mental health difficulties are often caused by violence and they can expose a person to victimisation (e.g. Humphreys and Thiara, 2003;Ramon et al., 10 2015). Both of these phenomena are tabooed, sensitive and intertwined with other social problems, thus affecting people's lives to a great extent and calling for a holistic approach within social services. Being based on these features, recovery from intimate partner violence and mental health problems is often challenging, and it is difficult to recognise the 15 issues that support it (e.g. Anderson et al., 2012;Skogens and von Greiff, 2014).
EE has mainly been studied in the context of forming interactive expertise through collaboration between a client and professionals (e.g. Collins and Evans, 2002;Cowden and Singh, 2007;Hill, 2009;Vennick 20 et al., 2014), or in the context of recovery (e.g. Loumpa, 2012;Shera and Ramon, 2013). Both contexts emphasise the perspective and knowledge of the client or service user and a holistic instead of a service-, professional-or problem-led approach to their situations. The lived experiences are seen as a valid way of knowing (Glasby and Beresford, 2006) 25 and as an adequate base for expertise. Also, professionals can hold experiential knowledge, and EE and professional knowledge are complementary rather than distinct areas (e.g. Nowotny, 2003;Collins et al., 2010). Our focus is on EE of victims of intimate partner violence and people with mental health difficulties in non-governmental organisations (here-30 after referred to as NGOs). We approach EE as a shared process, which is primarily run by and for the people involved and which can be utilised for developing professional knowledge and reaching common purposes.
Recovery is often perceived and studied as an individual process in which people set goals, search their strengths and resources by self- 35 reflection, and seek a new, valued sense of self (Deegan, 1988). It is seen as a process that empowers service users to cope in and to take control of their everyday lives. Our research on EE emphasises the group, community and societal levels of recovery, which have been sparingly implemented in social work practices and policies (Shera and 40 Ramon, 2013). There, recovery is seen as a collective process that is based on the support of the community and peers and involves sociocultural and professional factors (e.g. Anderson et al., 2012;Loumpa, 2012;Shera and Ramon, 2013).
From the political viewpoint, the role of experts by experience (e.g. McLaughlin, 2009) is highlighted in service design and delivery. In the recent decades, the focus has been on service user involvement (e.g. Beresford and Croft, 2004;Hodge, 2005;Glasby and Beresford, 2006;5  , user participation (e.g. Carr, 2007;Raitakari et al., 2014) and new governance and deliberative democracy (e.g. Solomon, 2004;Martin, 2011). Similar agendas are being introduced widely in Finland as well; for example, in the 'Finnish National Plan for Mental Health and Substance Abuse Work 2009-2015' (Partanen et al., 2010), 10 experts by experience legitimised their status in developing the social and health services. However, the participation of clients and citizens has been enhanced particularly by NGOs and non-profit organisations, which have invested on developing EE and mobilised the idea of recovery-oriented practices. 15 We approach EE and recovery from the viewpoint of personal and social empowerment. Empowerment entails the idea of a person as a subject who has a right to become heard when constructing a change in one's life situation. People need to seize the power to determine and reconstruct assessment, agency and action (Adams, 1996). As a social pro-20 cess, empowerment focuses on changing society and people's everyday life settings rather than their behaviour or attributes (e.g. Ferguson and Woodward, 2009;Weldon, 2011;Hokkanen, 2014a).
The purpose of the paper is to explore the formation of EE in two Finnish NGOs. We focus on (i) the types of EE in them and (ii) the 25 ways in which the types are interrelated. On the basis of this, we consider the role of EE in promoting recovery. This paper contributes to the research discussion on recovery and EE by deepening our knowledge of the various types of EE and their internal relations.
In the sections that follow, we first describe our data and method of 30 analysis. Second, we present our findings concerning the types of EE. Third, we examine the interrelations of these types. Finally, we discuss the way in which they enable empowerment-oriented recovery in social work. We argue that EE should be considered as a more integral part of social work than it presently is.

Method
The two NGOs retical underpinnings of the research interests and data collection were the same: EE and empowerment. By combining the data from the NGOs, we searched for a deeper insight into EE that promotes empowerment-oriented recovery (Sandelowski et al., 2012).
The quantitative data on mental health 10 The data concerning mental health focused on EE-based activities as a way to recovery. The data were collected through a survey from member associations of the Finnish Central Association for Mental Health that vary in terms of their size, resources, location, sociocultural environment and relationship to authorities. This was done even though it was evi-15 dent that some of the associations would not be able to convene or otherwise organise the distribution of the questionnaire during the responding period. Personal and social empowerment as a shared, change-oriented and strengths-based process was considered when forming the questions concerning the resources, methods and outcomes of 20 EE-based activities (Hur, 2006;Hokkanen, 2012Hokkanen, , 2014a. The questionnaire was constructed and piloted with twenty-one active members of the target population, seven employees of the associations and two senior researchers of the topic. After the piloting, minor changes were made. The questionnaire was sent to 161 contact persons of the member 25 associations in 2013. The contact persons were asked to distribute the questionnaires to their key members who had personal experiences of mental health problems. It is impossible to define the response rate because the size of the target population is not known. Altogether, 382 members replied to the questionnaire, 133 of whom acted as experts by 30 experience. According to the contact persons and the respondents themselves, they are key players in peer-based activities and identify themselves as experts by experience. Their answers are seen as a valid source of knowledge (Nowotny, 2003) and they are analysed in this study. The other respondents were engaged in other peer support activities, for ex-35 ample as leaders of peer support groups or as members of the board of their association. Associations with fixed premises and at least one employee were more likely to respond than others. However, the coverage of the survey was extensive and EE activity was found even in associations functioning with very limited resources (Hokkanen, 2014b). 40 Table 1 summarises the background information of the respondents.
In this paper, we present the results of four multi-item questions focused on the tasks and the audiences of the EE-based action, the resources for the action, the ways of acting and the outcomes of acting as an expert by experience. The questions were: The items under Q1 are summarised in Figure 1. The items under Q2 were: experiences of life itself, experiences of mental health issues, a stable life situation, experiences of overcoming troubles, personality, absorbing skills from peers, participating in activities, training, supervision 15 and absorbing skills from professionals. The Q3 items are presented in Table 2 and the Q4 items in Table 3. The response options under Q2 and Q3 were: not important, somewhat important, important and extremely important; under Q4, they were: decreased on the part of many, decreased on the part of some, neutral, increased on the part of some 20 and increased on the part of many.
Factor analysis was utilised with Q3 and Q4. When the variables correlated with one another, based on polychoric correlations, the maximum likelihood extraction method along with the oblique rotation promax were used in the factor analysis. The sum scores were formed 25 by calculating the mean of the variables with very high loadings ( 0.5) (in bold and underlined in Tables 2 and 3) and by rescaling them to the distribution of the original variables. The factor analyses were made using R; the other analyses included correlations, cross-tabulation, sum scores, group analysis and other basic statistical methods us- 5 ing SPSS.

The qualitative data on intimate partner violence
The data concerning intimate partner violence were connected to the development project. The project was conducted by three local associations of the Federation of Mother and Child Homes and Shelters. 10 These local associations had their own premises, paid employees and established service production. The project included sixty victims who acted as experts by experience and nine professionals. There were twenty-three open and closed support groups in the project (from three to ten participants in each closed group). The data consist of three sets 15 of material: (i) interviews with victims of violence (five individual and one group interview) focusing on their action as experts by experience and on their thoughts of the meaning of EE; (ii) interviews with professionals (four individual and one group interview) dealing with their views on the value of EE in terms of professional knowledge and prac-20 tices; the interviews lasted from fifty minutes to two hours and forty minutes, and they were transcribed (150 pages); (iii) written material produced throughout the project by the experts by experience (e.g. feedback from the project activities) and the professionals (reflective diary notes and summaries) for the purposes of the research (altogether 25 118 pages). The analysis of the qualitative data began by reading the material closely, focusing on the contents, spaces and phases of action to gain knowledge of EE and the meanings given to it. The experiences were regarded as subjective and unique, and they were understood hermeneutically (Laitinen and Nikupeteri, 2013). The quotations presented 30 in the article illustrate some of the central points of the three types of EE. They are anonymised and indicated as 'professional' or 'expert by experience'.

Triangulated synthesis
We utilised triangulation to achieve more reliable and transferable 35 knowledge about EE (Denzin, 1978;Thorne et al., 2004;Sandelowski et al., 2012). The study was triangulated in terms of the data (quantitative data from the survey versus qualitative data from the interviews and writings), the methods (statistical analysis versus hermeneutic close reading) and the researchers (their backgrounds covering the third sector, peer support, EE, recovery and sensitive issues) (Denzin, 1978). Additionally, the synthesis was enriched by the diverse nature of the phenomena of intimate partner violence and mental health issues (Sandelowski et al., 2012).
We first analysed the two data-sets separately without seeing each other's raw data and then interpreted and theorised the findings in synthesis by examining them against the related literature on EE through a dialogic conversation (Thorne et al., 2004). The joint analysis started by examining the contradictions and similarities of the data. As for contra-10 dictions, we noticed that the data-sets cover different dimensions regarding EE. The data on intimate partner violence describe the formation phase of EE and the knowledge base was facilitated by professionals, while the data on mental health problems depict the advanced phase and the knowledge base rested on peer support. 15 In terms of similarities, we first found out that the sharing of experiences, peer support and the change-oriented recovery process were joint features. By combining these findings, we discovered three types of EE: individual, group and organised. Later, we found similarities concerning the different ways (emotional and influential) and outcomes (renewed 20 self-image and action) of acting as an expert by experience. These types and the different ways and outcomes of acting as an expert by experience are the results of the synthesis, and they will be examined further in the empirical sections. Additionally, we utilised the findings of the qualitative data in naming and interpreting the results of the factor anal- 25 ysis. The triangulated synthesis enabled us to achieve a multidimensional, valid and tested understanding of EE.

Ethical considerations
We adhered to ethical guidelines throughout the research (Finnish Advisory Board on Research Integrity, 2016). The boards of the related 30 associations and federation approved the studies, including their ethical principles. The research permits also included permission for subsequent use of the data for research purposes. In conducting the survey, each association made its own decision on whether or not to deliver the questionnaires to its key members, and each individual member decided 35 whether or not to respond. The background and purpose of the research, the utilisation of the data and the researcher's contact information were included in the covering letter to the associations and their key members. The respondents returned the questionnaires directly to the researcher. In the qualitative research, the participants gave their 40 informed consent, and they had the option to withdraw at any phase of the research. The integrity of the participants was ensured and their experiences and choices respected. The professionals close to the experts by experience made sure that the women received professional help both during and after the data collection.

Limitations
Despite the strengths of the data, there are limitations as well. The ma-5 terial was collected from NGOs whose ideology is to involve people in societal decision making and service delivery. Data collected from agencies in the public sector could produce other outcomes regarding the effectiveness of EE. Positive experiences may be over-emphasised in the survey because the receivers of the questionnaires were selected by the 10 associations themselves. In the research on intimate partner violence, there were professionals present in the data collection. Thereby, the client-professional relationship may have had an impact on what the respondents wanted to share. These limitations are related to the sensitivity of the two research topics. Additionally, the services provided 15 by the associations are considered to be significant by the service users. According to Hirschman's (1992) concepts of an individual's relationship with a person or organisation, these people find it challenging and demanding to exit a relationship. As a result, some loyalty towards the service provider may exist, and one should therefore proceed with caution 20 in transferring the results concerning EE to other areas of social work.

Findings
The types of experiential expertise

Individual experiential expertise
In the survey, the respondents estimated the importance of resources 25 (Q2) for acting as an expert by experience. Three resources were related to individual EE: experiences of life itself (71 per cent found it extremely important), of mental health issues (70 per cent) and of overcoming troubles (69 per cent). In the qualitative data, the interviewed women perceived living in a violent relationship and confronting related 30 experiences to be grounds for acting. We refer to these as individual experiences, which create the basis for EE. In order to convert individual experiences to individual EE, people needed to find their experiences as strengths and orient themselves to change a problematic situation.
The individual EE present in the helping process was constructed in 35 the interaction between a professional or peer and the person receiving help. According to the survey, 64 per cent of the experts by experience had a mutual support relationship with their peers. Based on the qualitative data, many women had interacted with a therapist or social worker before sharing their experiences with peers in a group. Building trusting relationships, being involved in a process as an actor and respecting the sensitiveness of a situation contributed to the personal experience. The professionals' understanding of the women's situations appeared signifi-5 cant in transforming experience into expertise and action: There are three women who are over seventy, have lived more than fifty years in a violent relationship, and are not willing to divorce. They are living in very difficult circumstances. They all have a mission: They wish that the new generations of women aren't subjugated under the same 10 circumstances as they have been in their lives. Therefore, instead of concentrating on their traumatic backgrounds I focused on a working model that respects their experiences and allows them to produce knowledge of how they have recovered and what the driving force has been. They are very skilful and foster various types of knowledge. There 15 are several issues that have supported their survival in the middle of everything. So, we begun to strengthen those aspects (Professional).
In the context of mutual aid, the help-seeker started acquiring a new identity. Thereby, one began to regain resources to control the life situation. In order to support recovery, it was essential to cultivate individual 20 EE by sharing one's experiences with peers.

Group experiential expertise
In group EE, personal experiences were taken to the group level, which constructs and strengthens personal, shared and proxy agency (Roth, 2004;Hokkanen, 2012;Loumpa, 2012;J€ arvikoski et al., 2013). Three of 25 the resources (Q2) found useful in acting as an expert by experience were related to cultivating individual EE and turning it into group EE: participating in activities (64 per cent found it extremely important), absorbing skills from peers (50 per cent) and absorbing skills from professionals (42 per cent). Cultivating experiences through EE-based activities 30 also appeared in the qualitative data. The starting point of the project was professionals' lack of knowledge and methods to intervene in and to help victims of lesser-known forms of violence (post-separation stalking, intimate partner violence among the aged, violence in religious communities). Thus, the professionals formed groups of victims to produce 35 knowledge, and to support recovery. The groups offered supportive and multilateral social relations, a sense of belonging and joint action for the women.
According to the survey, most of the experts by experience acted as leaders of peer support groups (68 per cent) or as facilitators in meeting 40 places (56 per cent). Such tasks enabled them to reconstruct a multifaceted picture of a shared issue. The interviewed women reported that, in peer groups, their experiences of belonging and sharing violent experiences were empowering: Every time I come here, I feel stronger. Alone I easily slip into the feeling that I'm a weak loser and guilty of everything. Sometimes it feels 5 like my history stigmatizes me. It is great that this place where everybody knows the saddest things of my life frees me to be myself. I receive knowledge, possibilities for acting and being. It gives a sense of belonging to a group that really tries to change things -not just to moan about our miserable destiny (Expert by experience). 10 The quotation indicates that a respectful atmosphere made it possible to reflect on experiences, which was necessary before taking further action. Listening to a variety experiences increased an understanding of the joint, partly shared and unique features of the experiences as compared to the sharing experiences by two people. When intimate partner vio- 15 lence occurred in a religious community, confidentiality was multidimensional: it was meaningful between the women and the professional who did not belong to the community, as well as among peers and in their relations with the community. The professionals offered a channel for reflection by considering the supportive sides of the community as well as 20 the fact that the community overpowers women's experiences and limit their agency (also Hurtig, 2013). Thus, confidential power-mutual trust between the victims, professionals and group members (Laitinen and Nikupeteri, 2013)-reinforced the women's recovery processes.
The idea of group EE is based on interaction and reciprocity. When 25 sharing experiences, people reintegrate with the community and society, and place their own experiences in a cultural and institutional context. Group EE enables societal change-oriented action as a path to collective recovery, which again strengthens individual recovery capital (Masterson and Owen, 2006;Skogens and von Greiff, 2014).

30
Organised experiential expertise Organisation increases the conspicuousness, continuity and credibility of EE. The associations included in this research promoted EE and thus facilitated its cultivation. According to the survey, the training (54 per cent found it extremely important) and the supervision (36 per cent) ar-35 ranged by an association were found useful resources (Q2) in carrying out the task related to organised EE. In both NGOs, the experts by experience carried out a large number of tasks: they gave advice, guided, helped, gave lectures, collaborated, evaluated and advocated. In order to carry out these tasks, the activities 40 were organised by the association. The activities were also directed to many audiences: peers, students, professionals, agencies, service providers and the public. The experts by experience mostly worked with peers, indicating the importance of peer support in EE, but they usually had other collaborative partners, as well (Figure 1, based on Q1).
Peer leaders and professionals in the associations encouraged people, created facilities and settings for action, and provided examples and sup-5 port for people especially in controversial situations. They also used power in a positive way and situated themselves in the nexuses of relationships to create and maintain collaboration between the experts by experience and the authorities and politicians. One professional recalled the principles and ethics of social work in supporting organised EE in action: 10 The main point is to think about the channel of action, what to choose, and to go forward. It's necessary to think and rethink, where and how, because mere yelling and screaming won't cut it. There needs to be an ethical justification for the action in support of human dignity, the right to be heard and treated equally and humanely (Professional). 15 Organised EE in the NGOs made it possible to integrate into EE the individual experiences of those participants who did not transform their own experiences into individual expertise. Its co-operational testing and formally ordered structure made organised EE more resilient and resistant to changing resources, actors or circumstances and it was easier to 20 approach by people who reach out to peers. It was also a more efficient means to develop services and society because it was based on shared power and intention. Organised, group and individual experiential expertise stimulate one another Investigating the types of EE showed that organised EE legitimised and supported other types of EE, whereas group EE empowered individuals 5 and contributed to their personal and collective recovery. Next, we investigate more deeply the interrelation between three types of EE.
In the survey, the respondents were also asked to estimate the importance of the ways of acting as an expert by experience (Q3). Factor analysis was utilised to clarify the interrelations between the ways of acting. 10 Two factors were defined (Table 2). We referred to the first one as the emotional way of acting. This factor, where close attention was paid to other persons, represented 39 per cent of the total variance. The second factor was referred to as the influential way of acting. It stood for 20 per cent of the total variance. The influential way of acting focuses on 15 changing things and creating new understanding.
The emotional way of acting and the influential way of acting were considered highly important by 50 per cent and 70 per cent, respectively, of the respondents. No significant differences regarding the backgrounds of the respondents (see Table 1) were found. 20 The influential way of acting in EE was aimed at developing individual, group-based and organised expertise and designing social services through them. Thereby, it set off wide social and political changes (also Hyv€ ari, 2001;Dugan and Reger, 2006;Weldon, 2011). Advocating individual experiences for wider contexts required an activist stance from the 25 experts by experience. At the group and organised levels, the emotional way of acting appeared as courage to share individual experiences, which created the basis for acting and influencing. For example, in the project on intimate partner violence, the victims of post-separation stalking gave comments on the forthcoming provision on anti-stalking in the Criminal Code and worked out an initiative to the municipal council on providing 5 alarm systems for victims of stalking. The coexistence of the influential and emotional ways of acting in the practices of the NGOs enabled them to change discriminative social structures and stand up for their rights. Also, the outcomes of acting intertwined with the types of EE. The respondents of the survey evaluated the outcomes of EE-based action in 10 terms of experts by experience in general-not just themselves (Q4). Factor analysis was employed to find out whether the outcomes of EE activities formed groups. Two factors were defined: renewed self-image, explaining 42 per cent of the total variance; and renewed action, explaining 24 per cent of the total variance (Table 3). 15 There were no significant differences between a renewed self-image and renewed action regarding the background variables (see Table 1). Renewed action and a renewed self-image were considered to be very common among experts by experience by 64 per cent and 60 per cent of the respondents, respectively. This appeared in the qualitative data as 20 well: both renewed self-image and renewed action were considered essential for recovery. The women concurred that joint activities gave them new insights into violence and strengthened their experience of being useful resources in acting against violence in a wider context. An empowered perception of oneself and of the others remodelled their 25 agency and action as individuals and as a collective.

Discussion
The research findings indicate that the three types of EE-individual, group and organised-form an integral part of the empowerment-oriented recovery process. The types of EE are not mutually exclusive; rather they 5 are interconnected and parallel ( Figure 2). The most recognised type of EE in social work is individual EE. Normally, it is perceived as client-oriented or client-centric practice aimed at changing the client's life situation by converting their individual experiences into individual EE. However, individual EE can be a 10 misleading basis for service evaluation or for the reconstruction of a helping system if it is separated from group and organised EE. Paying attention to an individual's experience may also obscure weaknesses and deficiencies in the service system and at the policy level (also Cowden and Singh, 2007). Individual EE promotes recovery especially in terms 15 of the client's identity and role as a service user (also Loumpa, 2012). Our findings at group EE level emphasise the importance of peerbased sharing of experiences (also Solomon, 2004;Masterson and Owen, 2006;Loumpa, 2012) in promoting recovery. Equally important is collective agency organised by NGOs who are committed to empowerment 20 (also Roth, 2004;Solomon, 2004;Carr, 2007;Hokkanen, 2014a). In Figure 2 Social work's linkages to the formation of experiential expertise group EE, experiences are cultivated by sharing, renaming and reflecting on problematic situations with peers. Group agency increases the possibilities for personal and collective recovery, as it presupposes safety and a supporting community as well as treating the diversity of people's ex- 5 periences as an asset (Skogens and von Greiff, 2014), even in contradictory issues (Dugan and Reger, 2006).
The strength of organised EE is in its diverse knowledge forms that are based on longitudinally confirmed experiences. The knowledge is constructed and renewed in peer collaboration and in co-operation with 10 service providers, authorities and politicians. This form of EE promotes recovery especially through its wide influencing possibilities (legislation, local political decision making and services) in a supported and safe context, and by recognising structural changes (e.g. legislation, funding decisions). Acknowledging organised EE augments earlier studies regarding 15 interactional expertise (e.g. Collins and Evans, 2002), knowledge-based practice (e.g. Glasby and Beresford, 2006) and the radical edge of user involvement (e.g. Barnes and Cotterell, 2012).
Without organisational EE, it may prove challenging to continue with personal and social empowerment and to find arenas for recovery in the 20 long run. As earlier research shows, experts by experience may be invited to evaluate and develop services but, in the important decisions, they are often faced with a glass ceiling (Staniszewska et al., 2012). This can be seen in managerialistic and professional-led user involvement (Hodge, 2005) and in trading zones and interactional expertise (Collins 25 et al., 2010) if it is the officials who decide when experts by experience are used, for what purposes they are used and who those persons are (Barnes and Cotterell, 2012).
The findings suggest that it is essential to take advantage of the different types of EE when reconstructing the service design and delivery. 30 Together, they create a space for the emotional and influential ways of acting, which are important in promoting empowerment-oriented recovery-forming a renewed self-image and taking renewed action. Moreover, group and organised EE contribute to dialogical, collaborative and change-oriented recovery processes. Group and organised EE 35 highlight the importance of empowerment orientation and recall radical, structural, community-sensitive and citizen-oriented social work (Adams, 1996;Ferguson and Woodward, 2009;Barnes and Cotterell, 2012).
Our findings are in line with earlier studies emphasising that recovery requires more than adequate services, professional expertise and conven-40 tional clienthood in which the social worker is a service provider and the client a receiver (Deegan, 1988;Barnes and Cotterell, 2012;Vennick et al., 2014). Peer support and NGOs committed to empowerment are essential in constructing and cultivating expertise to achieve a sustainable recovery process. Without them, experts by experience will face without the support of significant others, which diminishes the effects of their action (Hodge, 2005;Cowden and Singh, 2007;Shera and Ramon, 2013).

5
In this paper, we have argued for three types of EE-individual, group and organised-in promoting recovery. These types of expertise create a space for emotional and influential way of acting, they renew the selfimage and action of participants as persons and as a collective, and they interact with each other. EE in NGOs that are committed to empower-10 ment is intentional and future-oriented, and it involves joint agency of actors and positive use of power. The relation between EE and recovery in NGOs is reciprocal: they strengthen each other. EE relates to the essence of social work, which is to affect people's life situations by changing their social and societal living environments. Social workers can 15 develop their theoretical expertise and practices by utilising clients' expertise that is cultivated in groups and organisations. To achieve this, professionals need to restore the rights of people, support their participation, and act as advocates with and for those whose agency is limited by creating possibilities for them to discover their personal and group 20 agency. Furthermore, the role of social workers in the context of recovery is to facilitate social change by empowering persons, groups and communities in their social environments. In our view, the utilisation of EE with an empowerment orientation should be increased and considered as an integral part of social work practices, preventive social policy 25 and structural social work.