Is mental health harmed by earlier emotional stress?

Thursday April 4 2013

“Daily stresses take toll on long-term mental health,” The Daily Telegraph reports. It goes on to say, “that little row with your spouse or stress of finding yourself stuck in a traffic jam may be taking a heavier toll on your mental health than previously thought”.

So should we obsess about daily stress, or should we maintain a firm British stiff upper lip? While the news correctly reflects the conclusions of a decent piece of research, there are problems drawing any major conclusions from the science.

The Telegraph story is based on a long-term US study that measured what people said were their daily stress triggers (‘stressors’), and how they said they emotionally responded to them. It then compared these findings with symptoms of mental health conditions a decade later.

Experiencing negative mood and emotions at the beginning of the study was linked with whether people had symptoms indicative of major depression, dysthymia (a form of long-term depression) or generalised anxiety disorder after 10 years. People’s reported emotional response to daily stressors predicted whether people reported they had mental health conditions, but not whether they had a diagnosis based on symptoms of these conditions.

Overall, the study’s limitations – such as low participation and high drop-out rates – make it difficult to conclude that our mood today predicts our mental health in 2023.

Where did the story come from?

The study was carried out by researchers from the University of California Irvine, California State University Fullerton and Pennsylvania State University in the US, and was funded by the US National Institutes of Health.

The study was published in the peer-reviewed journal, Psychological Science.

The research was reported appropriately by the Telegraph.

What kind of research was this?

This was a prospective cohort study that assessed the association between response to daily stressful events and common mental health disorders 10 years later.

Being prospective in nature, this study reduces the likelihood that any associations are found due to individuals incorrectly recalling their past emotional responses to stress. Having such a long follow-up period does, however, increase the risk that many people who participated at the beginning of the study may drop out by the end of the study, which could potentially bias the results.

What did the research involve?

Researchers recruited people aged 25 to 74 in 1995 and 1996. These people completed questionnaires assessing their physical and mental wellbeing, their experience of daily stress triggers (which the researchers call ‘stressors’), and their affective (emotional) reaction to these stressors. Ten years later they were followed up and their mental health assessed, including whether they suffered from common mental health disorders such as depression and generalised anxiety disorder.

The questionnaires at the beginning of the study measured general affective distress, where participants reported how often during the previous 30 days they had felt worthless, hopeless, nervous, restless or fidgety, and how often they felt that ‘everything was an effort’ or that they were ‘so sad that nothing could cheer them up’. Participants were also asked at this time whether they had experienced or been treated for ‘anxiety, depression or some other emotional disorder’ during the previous 12 months. Researchers then employed a commonly used tool, called the Composite International Diagnostic Interview – Short Form (CIDI-SF) to diagnose major depressive disorder, dysthymia (a form of long-term depression) or generalised anxiety disorder (GAD).

Another questionnaire asked about participant experiences with daily stressful events. This was completed every evening for eight days at the beginning of the study, and included items such as having an argument; a problem at work or home; having a friend, associate or someone else within one’s social network experience an upsetting problem; and other potentially stressful situations. Participants also reported their mood and emotions on these eight days. The items included were identical to those described above, but related only to the previous 24 hours instead of the previous 30 days. The researchers estimated emotional response to stressors (or ‘affective reactivity’) by calculating the difference in mood and emotion on days without stressors compared to days with stressors.

In their analyses, the researchers controlled for several potential confounders including gender, education and age, as well as negative feelings on non-stressor days.

What were the basic results?

During the first wave of the study, 1,483 people completed the questionnaires on their general emotional distress (affective distress) over the previous month, what daily stressors they had experienced, and daily emotional distress.

Ten years later, only 793 participants (53.4%) completed the follow-up questionnaires. A further 82 participants were not included in the analysis because data was incomplete on either the initial or follow-up questionnaire. This left 711 people who were included in the analysis of the association between emotional response to daily stressors and long-term mental health.

Among those participants included in the analyses, 12.2% reported symptoms at the beginning of the study consistent with either major depressive disorder, dysthymia or generalised anxiety disorder. At the follow-up interviews 10 years later, 10.3% met the same criteria.

The researchers found that people with a symptom-based diagnosis of depression, dysthymia or generalised anxiety disorder at baseline were likely to have one of these diagnoses 10 years later (odds ratio (OR) 3.98, 95% confidence interval (CI) 2.03 to 7.81). Negative emotions on days that were free from stressors at the beginning of the study also predicted a symptom-based diagnosis of these conditions 10 years later (OR 1.31, 95% CI 1.05 to 1.63).

While emotional response to daily stressors did not significantly predict mental health outcomes 10 years later (OR 1.25, 95% CI 0.92 to 1.70), it did predict a self-reported experience or diagnosis of such disorders (OR 1.56, 95% CI 1.21 to 2.01). The average number of daily stressors reported at baseline was not predictive of a diagnosis either (OR 0.91, 95% CI 0.65 to 1.28).

How did the researchers interpret the results?

The researchers conclude that, ‘the average levels of negative affect that people experience and how they respond to seemingly minor events in their daily lives have long-term implications for their mental health’.


This study suggests that the daily experience of negative emotions may predict the presence of common mental health disorders a decade later.

The researchers highlight the theory that, ‘affective responses to seemingly minor daily events have long-term implications for mental health’, and that their findings support this. While the study did reportedly include a large, nationally representative sample of adults, there are several limitations. Many of these limitations were reported by the study authors, and they include the facts that:

  • The researchers did not report how many people were initially invited to participate, only how many people agreed to participate. If there was a big discrepancy in numbers and the characteristics of those who did and did not agree to participate, it could mean there was an initial selection bias, but we cannot tell if this is the case because the figures were not reported.
  • There was a very high loss to follow-up over the course of the study, with 46.6% of participants dropping out of the study for various reasons. The researchers did not report whether, or how, the people who could not be followed up differed from those who continued to participate in the study.
  • Diagnoses of major depression, dysthymia and generalised anxiety disorder were based on self-reported symptoms over the previous year, and such reports may not be completely reliable.
  • Information was collected only at the start of the study and 10 years later. It is not known whether similar levels of negative mood persisted throughout the study period, or when disorder symptoms first appeared. Information was also collected on negative affect (emotions) and negative daily events – it is not known how positive mood and events influence this association.
  • The individuals included in the final analyses were more likely to be of European American descent and tended to have higher education levels. Assuming the results hold – regardless of the other limitations – they may not be generalisable to groups other than educated individuals of European descent.

Overall, the authors say that their results suggest that daily stress triggers (stressors) cause ‘wear and tear’ on emotional wellbeing. They say that this is consistent with theories that suggest the manner in which people experience negative emotions and respond to negative events in their lives has an impact on their future mental health.

Further studies that address some of the drawbacks of this study, especially the potential for selection and misclassification bias, would help to more firmly establish the link between our current emotional state and our future mental health.

Analysis by Bazian
Edited by NHS Choices