A quarter-century of data from more than 3,600 American adults is challenging one of the most familiar assumptions in psychology: that people who appear to be doing well will continue to do so in a stable, predictable way. A new study published in Nature Mental Health finds that psychological well-being follows several distinct long-term paths rather than a single universal pattern. Some adults remained broadly well across decades, while others experienced gradual deterioration, late improvement or dramatic gains that proved difficult to sustain. The research also shows that depressive symptoms at the beginning of the study were associated with sharply different long-term outcomes, suggesting that depression may influence not only how people feel in the present but also the direction their well-being takes over time.
The analysis used information from the Americans’ Changing Lives study, a large US cohort followed for approximately 25 years. The researchers examined data from 3,617 adults and focused on psychological well-being, a broad measure that captures positive functioning rather than simply the absence of mental illness. Well-being can include a sense of purpose, personal growth, autonomy, positive relationships and the ability to manage life’s demands. This distinction is important because someone may not meet criteria for depression while still experiencing low fulfillment, limited purpose or poor emotional vitality. Conversely, a person with depressive symptoms may later recover and develop stronger psychological resources.
To identify patterns hidden inside the average trends, the researchers applied growth mixture modeling, a statistical technique designed to detect groups of individuals who follow different developmental trajectories. Conventional longitudinal analyses often calculate an overall increase or decrease across a population, potentially masking major differences between people. Growth mixture modeling instead estimates several unobserved subgroups, each characterized by its own starting point and pattern of change. In practical terms, the method asks whether the population is best understood as one broad trend or as a collection of distinct pathways unfolding at different speeds and in different directions.
Four major trajectories emerged. The largest group, representing about 60 percent of the sample, showed persistent well-being across the observation period. This was the healthiest overall pattern, but it did not mean that every member of the group felt consistently positive at every measurement. The researchers detected substantial intra-individual variability, meaning that individuals within this apparently stable category could still fluctuate considerably over time. A subgroup maintained persistently low well-being despite belonging to the broader persistent pattern, highlighting a central finding of the study: stability in a statistical trajectory does not necessarily equal high psychological health.
The remaining trajectories revealed more dramatic change. One group experienced decelerated declines in well-being, with psychological functioning worsening over time but the rate of deterioration slowing. Another showed gradual increases, suggesting a steady accumulation of positive change across the years. The final group displayed marked but unsustained increases: well-being rose sharply, but the improvement did not remain at the same level. These patterns illustrate why a single snapshot of mental health can be misleading. Two adults with identical well-being scores at one point in life may be moving in opposite directions, with one beginning a prolonged decline and the other approaching a sustained recovery.
Depressive symptoms at baseline helped predict which path participants were more likely to follow. Adults who reported higher levels of depressive symptoms had greater odds of entering the decelerated-decline trajectory. This association is intuitive in one respect: early psychological distress may leave individuals vulnerable to future challenges, particularly when depressive symptoms interfere with motivation, relationships, physical health or access to supportive resources. Yet the results also contained a surprising counterpoint. Elevated baseline depressive symptoms were likewise associated with greater odds of belonging to the group that experienced marked improvements in well-being, even after accounting for people’s initial level of well-being.
That finding suggests that depression is not a simple forecast of unchanging decline. Individuals beginning with more depressive symptoms may have more room for improvement, but the statistical association remained after the researchers considered baseline well-being, indicating that depressive symptoms contributed information beyond a person’s starting position on the well-being scale. The result does not mean depression is beneficial or that severe symptoms reliably lead to recovery. Rather, it points to divergent pathways: similar initial distress can be followed by worsening, slowing deterioration or substantial improvement, depending on the broader psychological, social and health conditions that shape a person’s life.
The study’s emphasis on divergence has implications for both research and clinical practice. If well-being changes along multiple trajectories, then interventions should not be evaluated only by asking whether symptoms have fallen below a diagnostic threshold. A person may experience fewer depressive symptoms but still lack purpose, social connection or a sense of growth. Another person may show modest symptom change while steadily building well-being through relationships, meaningful activity, improved health or greater autonomy. Measuring positive functioning alongside distress could therefore help clinicians identify whether treatment is producing genuine psychological recovery or merely reducing one dimension of suffering.
The findings also reinforce the importance of repeated assessment. Long-term well-being cannot be fully understood through a single questionnaire or a single clinical visit. Tracking people across years can reveal whether an apparent improvement is durable, whether a decline is accelerating or slowing, and whether a person who appears stable is actually experiencing repeated fluctuations. The authors caution, in effect, that the trajectories are associations rather than proof that depressive symptoms directly cause later changes in well-being. Life circumstances, physical illness, socioeconomic conditions, treatment, personality and social support may all contribute to the pathways identified by the model. Even so, the study offers a powerful population-level message: psychological well-being is dynamic, depression can precede radically different futures, and protecting mental health requires more than eliminating symptoms—it requires helping people sustain the capacity to live well over time.
Subject of Research: Long-term psychological well-being trajectories and the role of baseline depressive symptoms in US adults
Article Title: Depressive symptoms predict divergent trajectories of well-being in US adults
Article References: Lyman, C., Ong, A. & Rottenberg, J. Depressive symptoms predict divergent trajectories of well-being in US adults. Nature Mental Health (2026). https://doi.org/10.1038/s44220-026-00705-6
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s44220-026-00705-6
Keywords: psychological well-being, depressive symptoms, mental health, longitudinal study, growth mixture modeling, Americans’ Changing Lives, adult development, mental health trajectories, depression and recovery
