In the 1980s, in the Länsi-Pohja Hospital District in Southwest Lapland, research began on how family and psychotherapeutic treatments could be applied in the care of severe mental health problems (Seikkula & Alakare, 2004). It was observed that predefined interventions often did not function effectively, as people's problems were inherently complex.

Surprisingly, positive developments seemed to occur when practitioners entered treatment situations without predetermined goals, roles, or tasks. Instead, they aimed to create a safe environment by adopting a basic attitude of curiosity, actively listening, asking open-ended questions, and varying the rhythm of the conversation—so that all voices could be heard equally. As a result, human encounters became more dialogical, meaning more reciprocal, enabling people to reflect on their situations from multiple perspectives. Even those experiencing psychosis no longer had to defend or justify their delusions in the same way as before, which often improved collaboration, calmed situations, and made individuals’ described experiences seem less incomprehensible—or psychotic—because they could now be better contextualized. In such a dialogical space (see Figure 1), the shared understanding that emerged could then be used, in accordance with the needs-adapted treatment model, to flexibly apply more specific treatment and symptom-relief methods.

Figure 1. Dialogical space (Arnkil, 2019)

In the Länsi-Pohja region, a complete reorganization of the mental health care system was initiated so that the system would support entering into a dialogical space regardless of the situation. Predefined treatments and referrals were abandoned, as they were not only unnecessary from the perspective of dialogical encounters but could also hinder the possibility of engaging in open dialogue. Consequently, the distinction between primary and specialized mental health care became unnecessary, and help requests could be responded to immediately by bringing the relevant people together and, when needed, inviting others into the care network.

Since achieving dialogue required a sense of safety, meetings were arranged in locations where the people involved felt safest. In severe crises like psychosis, this sense of safety required that meetings occur as intensively as needed, and that participants had sufficient knowledge about what was happening in the care process and whom to contact in case of problems. To ensure this, the same professionals took responsibility for organizing the dialogical care process across service boundaries. This helped maintain the shared understanding developed in dialogical interaction, preventing the need to restart the help process with different people each time.

In addition to providing safety for those receiving care, professionals also needed to feel secure. This is a challenge in dialogical work, as professionals often find safety in predetermined definitions and interventions, which can simultaneously hinder the development of a dialogical space. In Länsi-Pohja, to foster a sense of safety and ensure sufficient staffing, all mental health workers were trained as family psychotherapists in in-house psychotherapy programs throughout the 1980s and 1990s. These programs especially emphasized dialogical work and the cooperation of care teams. Furthermore, work was mostly organized in pairs to prevent an overwhelming burden from falling on one individual, which can often create insecurity and hinder dialogue.

As a result of these efforts, a new kind of psychiatric service system emerged in the Länsi-Pohja Hospital District, where mental health service resources were dedicated to supporting dialogical care processes. Once dialogical responses to help requests were ensured, the shared understanding developed in those encounters could be used to support more defined mental health care or other forms of help. This approach does not exclude more targeted interventions or acute symptom relief when considered necessary after a dialogical encounter, in prolonged cases, or when the symptoms are too severe to allow for proper interaction. This type of care system and way of responding to people seeking help in mental health services came to be known as the Keropudas model, later more widely recognized as the Open Dialogue approach.

My Research on Open Dialogue in the Treatment of Psychosis

I began exploring the Open Dialogue approach by starting a psychology internship at the Keropudas psychiatric hospital. I was impressed by the flexibility, humanity, and groundedness of the services, although I was initially skeptical about the effectiveness of the model, mainly because the work culture was nearly the opposite of what I had been trained in.

After the internship, I stayed in the region to work as a clinical psychologist while also studying the Open Dialogue approach more closely. I inherited psychologist Markku Sutela’s office, where I found his draft of a follow-up study on psychosis patients treated under Open Dialogue. The idea intrigued me, and I began wondering if I could develop it further. My internship supervisor, psychologist Tapio Salo, introduced me to Professor Jaakko Seikkula, who had previously worked in the area and conducted extensive research on the approach. I shared my idea with Jaakko, who seemed pleased that someone wanted to examine it more deeply. I then began designing a follow-up study to examine the long-term effectiveness of Open Dialogue compared to other forms of psychosis treatment. Initially, I thought I would write one article that would look good on my CV and help my job search when I returned to Southern Finland. But things turned out differently.

Through the secondary use of social and health care registers, we were able to track service use and mortality of all individuals treated for a first episode of psychosis in the region during the 1990s, from the start of care until 2015. We also created a comparison group of patients treated for psychosis in other parts of Finland during the same period, resulting in a unique average follow-up time of 19 years. The data was supplemented with information from local research and health care records.

According to the results (Bergström et al., 2017), most patients entering care through Open Dialogue received an immediate response to their help requests, the treatment consisted of intensive home visits, and in nearly all cases, especially early on, members of the social network participated. In longer treatments, individual psychotherapy became more prominent.

Based on the results, medication and hospitalization were used mainly when symptoms needed to be quickly alleviated due to prolonged insomnia, risk of violence, or lack of improvement in intensive outpatient care (Bergström et al., 2017). Compared to conventional treatment, both tranquilizers and antipsychotics were used significantly less in Open Dialogue (Bergström et al., 2023). However, those who were prescribed medication did not differ significantly in dosage or long-term psychotropic medication use (Bergström et al., 2023). Antipsychotic use was associated with longer treatment duration, partly independent of GAF scores and diagnosis at intake (Bergström et al., 2020), but the influence of symptoms on treatment initiation and long-term outcomes cannot be fully controlled in such a design.

Nevertheless, the results suggest that the dialogical approach can identify a subgroup of patients who do not require long-term antipsychotic medication. This can help reduce long-term side effects, such as physical and psychological adverse effects, and withdrawal symptoms that are often mistaken for psychotic relapse (Bergström et al., 2022; Bergström & Gauffin, 2023).

The topic has not yet been extensively studied, and there have been concerns that need-adapted medication use might worsen outcomes or increase mortality. Such concerns may stem from generalizations based on isolated cases or decontextualized interpretations of cross-sectional data. However, local experiences and longitudinal data do not support these concerns (Bergström, 2024; Bergström, 2020).

In fact, in the Open Dialogue system, medication can be started much earlier when needed, because untreated psychosis episodes tend to be shorter due to immediate response (Bergström, 2020; Seikkula et al., 2011). If symptom-based medication or individual pathology-focused interventions were started immediately, it would result in over-medicalization, since the low threshold of the services attracts a broader range of people, not all of whom meet strict diagnostic criteria in long-term follow-up. It's also often unclear who the actual "patient" is in a network-based intervention.

When clinical and demographic background factors were controlled, patients treated outside the Open Dialogue system were more likely to still be in treatment (OR: 2.2; 95% CI: 1.3–3.7) and/or disabled (OR: 2.6; 1.6–4.3) at the end of the 19-year follow-up (Bergström et al., 2018). Mortality rates adjusted for age and gender were slightly lower in the dialogical group (3.4 vs. 2.9), but the difference was not statistically significant for illness-related or suicide mortality. Service use, including repeat hospitalizations, was sometimes higher in the early phases of dialogical care due to the system’s fluid nature, but this reversed over the long term (Bergström, 2020).

Lived Experiences and Qualitative Understanding

However, registry data does not reveal how psychosis and its dialogical treatment are experienced at the individual level. For this reason, together with the region’s former chief psychiatrist Birgitta Alakare, we decided to contact people treated in the 1990s and ask them directly. I traveled around Finland interviewing former and current patients. At the beginning of each meeting, I asked them to share their life story in their own way. Many were touched that someone wanted to listen.

The interviewees did not typically label their experiences as psychosis, nor did they emphasize the treatment’s role in their recovery (Bergström et al., 2019). Many shared stories of difficult life events, such as loneliness, bullying, or repeated disappointments in relationships. Some described themselves as sensitive by nature—not necessarily a negative trait. Many recounted how various challenging events accumulated at a certain life stage, increasing their stress, causing insomnia, and eventually leading to such confusion that they could no longer make sense of what was happening. At this point, someone usually became concerned and care began.

Some recalled being surprised when strangers came to their home to talk. In hindsight, many viewed those conversations positively. Some also found medication helpful in calming down and regaining sleep (Bergström et al., 2022).

However, treatment was not seen as miraculous (Bergström et al., 2019). Rather, difficult life situations gradually improved. Stressors eased, some formed new relationships, and others felt they had learned something new. Some were helped simply by the understanding and kindness of others.

A small number had similar experiences again later, usually in response to new life stressors. Many of them later tried to see these episodes in a more positive light, even though they had been frightening and disruptive at the time.

Conclusion

Researching the Open Dialogue turned into a long-term journey, and my short visit to Lapland became a multi-year career in psychiatric specialized care. Eventually, I had enough publications to compile a dissertation, where I examined both qualitative and quantitative results.

In my interpretation, dialogical processes that integrated mental health care into the context of people’s real lives—and did not automatically pathologize their experiences—helped individuals retain their sense of agency and form new meanings for experiences labeled as psychosis. I believe this is reflected in the registry data as better preserved work and functional capacity and lower average service use. This remains a hypothesis I hope will one day be further tested.

My dissertation was awarded the highest honors in 2020. Since then, I have continued as a part-time postdoctoral researcher at the University of Jyväskylä's Department of Psychology, also exploring the Open Dialogue model in youth psychiatry.