The end of a romantic relationship is often not a sudden or unexpected event. A new study published in the Journal of Personality and Social Psychology has found that relationship satisfaction tends to decline gradually at first, followed by a much steeper drop around one to two years before a breakup. This “terminal decline” pattern was consistent across large samples in four national studies, highlighting how the unraveling of a relationship often begins long before the final decision to separate.
The researchers were interested in better understanding how romantic relationships change over time—specifically, what happens in the lead-up to a breakup. Although previous studies have looked at how relationship satisfaction tends to drop in the early years of a partnership, the focus has generally been on time since the relationship began.
But in this study, the researchers asked a different question: What happens when we look at satisfaction based on how close a couple is to ending things? Their idea was inspired by research on aging, which has shown that well-being often drops sharply in the years just before death—a pattern referred to as “terminal decline.” The team applied this concept to romantic relationships, wondering if something similar might happen in the years leading up to a separation.
To explore this idea, researchers Janina Larissa Bühler and Ulrich Orth analyzed data from four large-scale, nationally representative longitudinal studies conducted in Germany, Australia, the United Kingdom, and the Netherlands. These studies followed thousands of individuals over many years, collecting annual information about their romantic lives, including their satisfaction with their relationships and whether they had separated from a partner. In total, the study analyzed data from over 15,000 individuals who experienced a breakup and compared them to a matched group of individuals who stayed in their relationships.
To make sure their comparisons were fair, the researchers used a statistical technique called propensity score matching. This method helped them account for pre-existing differences between people who eventually broke up and those who didn’t, such as age, personality traits, income, education level, and initial relationship satisfaction. This allowed them to more confidently attribute any observed changes in satisfaction to the process of separating, rather than other factors.
The main focus of the analysis was to examine how satisfaction changed in relation to the timing of a breakup. Instead of looking at how satisfaction changed over the total length of a relationship, they focused on the number of years leading up to the separation—a measure they called “time-to-separation.” When they plotted this data, they found a clear and consistent pattern: satisfaction declined in two distinct phases. First came a slow, gradual drop, lasting for several years. Then came a steeper decline that began roughly one to two years before the relationship ended. This second phase, which the researchers labeled the “terminal phase,” marked a period of rapid deterioration.
Across the different datasets, the timing of this terminal phase varied slightly, with the sharp decline typically beginning between seven months and a little over two years before the breakup. These findings suggest that many relationships begin to unravel long before a couple officially ends things. The researchers also compared this pattern to changes in life satisfaction. While people’s overall life satisfaction did tend to decrease before a breakup, the change was much smaller and less consistent than what they saw in relationship satisfaction. This supports the idea that terminal decline is more specific to the relationship itself rather than a general drop in well-being.
Interestingly, when the researchers compared this approach to the more traditional method of looking at time since the relationship began, they found that time-to-separation was a much better predictor of changes in satisfaction. In other words, knowing how close a couple is to breaking up offers more insight into their declining relationship satisfaction than knowing how long they’ve been together.
The study also explored some key differences between people who initiated the breakup and those who were on the receiving end. In one of the datasets, individuals who initiated the separation showed signs of dissatisfaction earlier, beginning their terminal decline more than a year before the breakup. Those who were left by their partner, on the other hand, often didn’t show a sharp decline in satisfaction until much closer to the end—sometimes just a few months beforehand. But once their satisfaction began to drop, it did so more quickly than for the initiators.
The researchers also looked at whether other factors—like age, gender, marital status, and previous relationship experience—played a role in how satisfaction changed. They found that age at separation and marital status sometimes influenced the patterns of decline, although the results were mixed across different studies. For example, in some cases, being married appeared to buffer against a steep terminal decline, while in others, it seemed to make the decline steeper. Gender and prior relationship experience didn’t consistently affect the results.
These findings suggest that while many couples do experience slow erosion in satisfaction over time, the shift toward an inevitable breakup often accelerates significantly in the final one to two years. This period of rapid decline may reflect growing conflict, emotional disengagement, or the realization that efforts to fix the relationship are no longer working.
Although the study was comprehensive, it had some limitations. All four datasets came from Western countries, so the findings may not apply in the same way to other cultural contexts where norms around marriage, divorce, and long-term relationships differ. Additionally, while the data covered long timespans and included many participants, it was still based on self-reported measures, which can be affected by memory biases or changes in how people interpret their experiences.
The researchers also noted that more frequent data collection—such as monthly instead of yearly assessments—might give a clearer picture of how the final stages of a relationship unfold. And while this study focused on average trends, future work could explore how these patterns differ between individuals, including those who experience personal growth, increased life satisfaction, or new romantic relationships after a breakup.
The study, “Terminal Decline of Satisfaction in Romantic Relationships: Evidence From Four Longitudinal Studies,” was authored by Janina Larissa Bühler and Ulrich Orth.
