Research Protocol for the STAR_T Couples Intervention Phase: A Multidisciplinary Cluster Randomized Controlled Trial on Stress and Resilience During Relationship Transitions

Stress is a universal human experience, a natural reaction to the demands of daily life. Yet, when stress becomes chronic or poorly managed, it transforms from a temporary alarm into a persistent burden that can erode mental health, disrupt sleep, and weaken the body's immune system. While individuals often face these pressures alone, human beings are fundamentally social creatures, and our closest relationships play a decisive role in how we handle these challenges. A stable, supportive partnership can act as a buffer, dampening the body's stress response and providing a sense of safety that fosters resilience. Conversely, conflict within a relationship can amplify stress, triggering physiological reactions that increase the risk of anxiety, depression, and long-term health issues. This dynamic does not stop at the couple; the tension between parents frequently spills over into the lives of their children, affecting the emotional and physical well-being of the entire family unit. Understanding how to repair these fraying connections is not just a matter of relationship advice; it is a critical public health question with the potential to protect the mental health of generations.

In response to this challenge, researchers at Heidelberg University and the University of Zurich have launched a large-scale investigation known as the STAR_T project. This study is designed to test whether a specific online training program, called PAIRwise, can help families navigate relationship crises and reduce the stress that permeates their daily lives. The researchers are not merely asking couples how they feel; they are measuring the biological reality of stress to see if the intervention works on a deeper level. By combining self-reported questionnaires with biological samples, the team aims to determine if improving a couple's relationship can lower stress hormones in both parents and their children, and whether adding a layer of human support to the online program makes a significant difference in the outcome.

The study focuses on families in Germany where parents are experiencing relationship conflict and have at least one child between the ages of eleven and seventeen. These families are recruited through counseling centers, social media, and psychological practices, and they are invited to participate in a randomized trial. Once enrolled, the families are divided into two groups. One group receives the PAIRwise program in a self-guided format, where parents work through seven core modules and select four optional modules at their own pace. The other group receives the same program but with the added support of an eCoach. These eCoaches are trained clinical psychologists who provide personalized feedback on the parents' reflections after each module, offering a human connection to the digital experience. The program itself is built on established principles of couple therapy, teaching skills in communication, emotion regulation, and problem-solving, while also addressing how parental conflict affects children.

To measure the effectiveness of this intervention, the researchers track the families over a period of twelve months. They assess mental health at the beginning of the study and again at two, six, and twelve months. For the parents, the primary measure of success is a reduction in symptoms of depression, anxiety, and stress. For the children, the researchers look for improvements in behavioral and emotional difficulties. Beyond these standard questionnaires, the study digs deeper by collecting saliva samples from both parents and children at specific intervals. These samples are analyzed for cortisol, a hormone that serves as a biological marker for stress. The researchers also use a method called ecological momentary assessment, where participants receive prompts on their smartphones several times a day over a weekend. These prompts ask about their current mood, their interactions with family members, and their immediate stress levels, providing a real-time snapshot of family life that goes beyond what people can recall in a single survey.

The study is structured to compare the self-guided version of the program against the version with eCoach support, aiming to see if the extra human interaction leads to better results. The researchers expect that the intervention will reduce the mental health burden for parents and improve relationship satisfaction. They also hypothesize that these improvements will have a "spillover effect," meaning that when parents manage their stress and conflict better, their children will also show signs of reduced stress and improved well-being. The inclusion of cortisol measurements allows the team to see if these subjective improvements are matched by a biological shift in the body's stress response. If the program is successful, it would demonstrate that a digital tool, even without face-to-face therapy, can effectively lower stress levels in a family system.

This research is particularly significant because it addresses a gap in available mental health resources. While online interventions exist, there is a lack of evidence-based, German-language programs specifically designed for couples in distress that also consider the impact on children. The STAR_T project aims to fill this void by providing a scalable, low-threshold option for families who might not otherwise seek help. The study is funded by an internal grant from Heidelberg University and is being conducted in strict accordance with ethical guidelines, ensuring that participant data is kept secure and that families can withdraw at any time. The researchers are careful to monitor for any potential negative outcomes, such as an escalation of conflict or relationship dissolution, acknowledging that relationship interventions are not without risks.

The team has set a target of enrolling sixty families, comprising approximately one hundred and eighty individuals, to ensure they have enough data to draw meaningful conclusions. While this sample size is sufficient to detect moderate effects, the researchers acknowledge that it may not be large enough to identify very small effects or complex interactions between different family types. They also note that because families are self-selecting into the study, the results may not perfectly represent all families in the general population. However, by combining rigorous scientific methods with a practical, accessible intervention, the study offers a clear path toward understanding how relationship stability can be restored and how that restoration can heal the entire family. The findings from this trial will help determine whether digital tools can effectively support families in crisis, offering hope that even in times of conflict, resilience can be cultivated and stress can be managed.