Resilience May Be Overrated

By Jane Farrell

Natural resilience – the ability to bounce back from major life stressors – may not be as common or as easy as once thought, according to new research from Arizona State University.

The study questions prior claims that resilience is the “usual” response to major life stressors by looking at longitudinal data in a more nuanced way and making less generalization about the human response to such dramatic events.

A paper detailing the research, “Resilience to major life stressors is not as common as thought,” is published in the current issue of Perspectives on Psychological Science.

“We show that contrary to an extensive body of research, when individuals are confronted with major life stressors, such as spousal loss, divorce or unemployment, they are likely to show substantial declines in well-being and these declines can linger for several years,” said Frank Infurna, an ASU assistant professor of psychology and co-author of the new study.

“Previous research largely claimed that individuals are typically resilient to major life stressors,” he said. “Whereas when we test these assumptions more thoroughly, we find that most individuals are deeply affected and it can take several years for them to recover and get back to previous levels of functioning.”

Infurna and co-author Suniya Luthar, an ASU Foundation Professor in psychology, were seeking to replicate prior work that showed among adults, resilience — which is described as stable, healthy levels of well-being and the absence of negative outcomes during or following potentially harmful circumstances — is the prototypical trajectory after potentially traumatic events.

Previous work by others in the field involving people going through traumas ranging from bereavement and deployment in military service to spinal-cord injury and natural disasters had reported that resilience is the most common response following significant negative life events.

“Our findings go against the grain and show there can be more to the picture than that,” Infurna said. “It may not be the case that most people are unperturbed and doing fine.”

Infurna and Luthar used existing longitudinal data from Germany (the German socioeconomic panel study), which is an ongoing survey that began in 1984 and annually assesses participants over a wide range of measures. The outcome that they focused on was life satisfaction, which assesses how satisfied individuals are with their lives, all things considered, as they pass through years of their lives.

Essentially, Infurna and Luthar documented that “rates of resilience” vary substantially based on assumptions applied while running the statistical models.

They explain that in essence, the question that was addressed in previous studies was not, “How many people are resilient?” But instead, “Assuming A and B, how many people are resilient?”

And what were the A and B assumptions applied in previous studies?

One was about how much the groups (resilient and others) differed but within one another. Previous studies assumed that whereas resilient and non-resilient groups differed in life-satisfaction changes over time — steady and high in the former but not the latter — trajectories of change were the same for all people within all of the groups. To illustrate with four hypothetical people, this would mean that Rita and Ralph, in the resilient group, both showed the same steady, high life satisfaction over time; whereas Norma and Nate, both in a non-resilient group who showed declines as a function of their major life event, showed declines exactly at the same time, and then rebounded at exactly the same time. Infurna and Luthar allowed for the possibility that Nate might have recovered two years after the adverse event and Norma immediately after the event (for example, when divorce signaled release from a particularly unhappy marriage).

The second assumption in earlier studies was that “peaks and valleys” over time would be the same within the resilient and non-resilient groups, that is, the degree to which people showed extreme highs and lows around the average of their own subgroups. Back to the illustrative example, this assumption would mean that in prior studies, life-satisfaction scores across all 10 years ranged between 4 and 8 (out of 10) for resilient and for non-resilient groups. Infurna and Luthar, by contrast, allowed for the possibility that Ralph and Rita may have stayed within the range of 6 to 8 over 10 years (that is the definition of resilience — stable good functioning) but that Norma and Nate may have been as low as 2 in one or two years, and as high as 10 in others; again, by definition, these people are “not stable.”

Merely removing the restrictive assumptions applied in previous studies dramatically changed the percentage of people found to be resilient. Using exactly the same database, rates of resilience in the face of unemployment were reported to be 81 percent. With the restrictive assumptions removed, Infurna and Luthar found the rates to be much lower, around 48 percent.

“We used previous research as a basis and analyzed the data based on their specifications,” Infurna explained. “Then we used our own specifications that we feel are more in line with conceptual assumptions and we found contrasting results.”

“The previous research postulated that most people, anywhere from 50 to 70 percent, would show a trajectory characterized by no change. They are largely unperturbed by life’s major events,” Infurna said. “We found that it usually took people much longer, several years, to return to their previous levels of functioning.”

A finding that means giving a person time alone to deal with the stressor might not be the best approach to getting him or her back to full functionality, Infurna said.

“These are major qualitative shifts in a person’s life, and it can have a lasting impact on their lives,” he said. “It provides some evidence that if most people are affected, then interventions certainly should be utilized in terms of helping these individuals in response to these events.”

The findings have implications not just for science but for public policy. According to Infurna, sweeping scientific claims that “most people are resilient” carry dangers of blaming the victims (those who do not rebound immediately), and more seriously, suggest that external interventions are not necessary to help people hit by traumatic events.