“It has been known for a long time that couples tend to be more similar than dissimilar to each other, but there are still many possible explanations for this,” explained Csajbók, an assistant professor of psychology at Charles University in Prague. “Also, similarity at one point in time does not tell much about how they are changing over time in relation to each other. This was particularly interesting in the context of mental health similarities, in which we can observe considerable fluctuations over time (people get better, seek help, bad things happen and they get worse, etc).”
The researchers obtained data from a six-wave longitudinal study called the Survey on Health, Ageing and Retirement in Europe (SHARE). The study included an assessment of depressive symptoms every two years for a period of 12 years. Csajbók and her colleagues restricted their analysis to heterosexual couples who had completed at least three assessments of depressive symptoms, leaving them with a sample of 11,136 couples (average age of 60) from 16 European countries.
The study authors used a statistical technique called dyadic growth mixture modeling to examine how couples’ depressive symptoms changed over time. Depending on their trajectories, the couples were grouped into one of four classes. The largest group included 76.9% of the couples and was characterized by both partners having consistently low depressive symptoms. Next, in 8.1% of couples, female partners had consistently high depressive symptoms and male partners had consistently low symptoms. Finally, in 7.2% of couples, both partners had decreasing depressive symptoms, and in 7.8% of couples, both partners had increasing symptoms.
Notably, the two groups with increasing and decreasing symptoms followed the same direction of change in symptoms — both partners’ symptoms increased or both partners’ symptoms decreased. However, the rates of change in symptoms were not the same. For example, in the group where both couples’ symptoms decreased, the male partners began with higher depressive symptoms and then saw a stronger decline in symptoms compared to their female partners.
The researchers say that their findings did not reveal any couples with truly diverging symptoms — for example, there was no class where one partner’s symptoms increased over time while the other partner’s decreased. Additionally, no group was identified where both partners had consistently high depressive symptoms. The authors say this may suggest that highly depressed people tend not to form a relationship together, or if they do start a relationship, it is unstable and short-lived.
“What we can confidently say so far is that we are making efforts to understand how couples affect each other’s well-being,” Csajbók told PsyPost. “We hope that this will help provide better treatment, since it directs practitioners’ attention to long-term patterns of a bigger unit than the patient, patterns on the couples’ level. It is not new that the network of familial relationships affects the individuals’ well-being and vice versa. But this research is going to help better identify those who are at risk.”
As might be expected, couples belonging to the group where both partners’ symptoms increased over time were more likely to report a bereavement or a break-up. They also had lower well-being and physical health compared to couples who were consistently nondepressed.
Csajbók and colleagues say their findings have implications for the future, for both clinical practice and social sciences. They suggest that treating all couples the same is ill-advised since couples vary in their depressive symptom trajectories. These differences mean that couples likely require different intervention approaches.
“Not all couples show a consistently similar long-term pattern, and this might be the reason why we could not identify yet which mechanisms are responsible for couples’ similarity – it can be attributed to different mechanisms in different cases,” Csajbók said. “We are curious to know how these patterns developed and what mechanisms are responsible for them, but this is open for a next study. We believe that the future of this field is to investigate the heterogeneity of our samples. When aggregating the whole sample together in research, not all results are applicable for everybody, and we expect to gain higher precision by exploring these inconsistencies.”
Limitations to the study include a sample that was comprised of people from Western, industrialized societies who tend to be wealthier and more educated. The findings may not generalize to people from other backgrounds. “We also tested a generally older population who were already established couples – this poses limitations to knowing how these relationships were initiated, how long it took to end up on the trajectory we see now in their 50s, 60s, and 70s,” Csajbók said. “We will address these issues in our upcoming studies.”
“It is important to mention that these are observed patterns which we captured, but we do not know enough about the individual couples to explore more closely what happened to them that caused their depression patterns,” she added.
