By Frank J. Infurna, Ph.D.
What does it mean to lose your spouse or partner, to become unemployed, or to receive a cancer diagnosis? Such events, commonly referred to as major life stressors, result in a qualitative shift in one’s life circumstances. The consequences of these major life stressors are not only confined to the individual, but can reverberate through one’s family and community. In the case of spousal death, the surviving spouse must adjust to a life without reliance on their partner for tasks of daily living or having to engage in common shared interests alone. Facing unemployment brings up anxieties about where the next paycheck will come from to help support one’s family.
Intuition (backed by empirical evidence) may incline you to believe that these major life stressors should lead to substantial declines in well-being and mental health, with individuals likely to bounce back over time. However, not all individuals succumb to the detrimental consequences of major life stressors. The resilience literature is built on the premise that there is a great deal of heterogeneity in individuals’ ability to adapt following adversity, with some being able to maintain functioning, whereas others initially decline, but recover over time.
The past 15 years of research in the resilience literature in adulthood and old age has portrayed contrary findings. In fact, research has shown that when confronted with significant adversity (irrespective of type), most individuals are remarkably resilient, defined as showing stable, high levels of functioning. This has been shown for a wide range of adversities, including spousal loss, divorce, unemployment, and health adversities, such as heart attack and cancer diagnosis, as well as natural disasters and military deployment.
This research has been coupled with advances in contemporary methods of analysis—in particular, growth mixture modeling (GMM). GMM is a statistical method of analysis that enables illumination of discrete trajectories (i.e., resilient, recovery, growth, and grief) when applied to large data sets that include individuals exposed to significant adversity.
Recently, we challenged notions that resilience is the common response to major life stressors, showing that this is largely due to data analytic and measurement choices (Infurna & Luthar, 2016). To do so, we used existing longitudinal data from the German Socio-Economic Panel Study, an ongoing survey initiated in 1984, and annually assesses participants over a wide range of measures. The outcome measure of interest was life satisfaction and the major life stressors we examined were spousal loss, divorce, and unemployment.
From our review of the literature that has utilized GMM to study resilience, we noticed two methodological assumptions that were commonly held, but never thoroughly tested: (1) homogeneity of variance across trajectories and (2) slope variances set to 0. The first assumption implies that the amount of within-group variability is the same between trajectories. By contrast, we allowed for the fact that resilient individuals, as a group, are more stable, whereas those in the other groups, show more variability. Furthermore, by estimating the slope variances to 0, prior analyses assumed that all individuals in the particular trajectory show the same rate of change in the outcome of interest. We allowed for the possibility that some individuals may take longer to recover, whereas others may recover more quickly (e.g., 3 years versus 1 year to return back to normalcy).
What we found when we relaxed these methodological assumptions was unexpected and surprising. Removing the restrictive assumptions applied in previous studies dramatically changed the proportion of people who were resilient. Rates of resilience in the face of unemployment were reported to be 81 percent. With the restrictive assumptions removed and having applied specifications of the model that are more in line with conceptual assumptions, the rates were much lower, around 48 percent. Focusing on divorce, using the same methodological assumptions as before, 85 percent were considered resilient, whereas when the restrictive assumptions were removed, 36 percent were likely to belong to the resilient trajectory. Lastly, for spousal loss, 75 percent were deemed resilient using the same methodological assumptions as before, whereas when the restrictive assumptions were removed, 47 percent were likely to be resilient.
Given the widespread use of GMM in the examination of resilience to major life stressors, we believe that there are important extensions of our findings. The most pressing direction we are focusing on is examining the multidimensionality of resilience. Most studies, overwhelmingly, have solely included one outcome, making it difficult to ascertain whether resilience is across-the-board. Put differently, if most individuals are found to be resilient in the well-being domain, does this transfer to other areas of life, such as health, sleep disturbances, or substance use? We tested this in a recent study where we examined resilience to spousal loss in five outcomes (Infurna & Luthar, in press). We found that resilient trajectories largely differed across outcomes, with 66 percent, 19 percent, 26 percent, 37 percent, and 29 percent showing a resilient trajectory in life satisfaction, negative affect, positive affect, general health, and physical functioning, respectively. When considering the multidimensional nature of resilience across five key outcomes, only 8 percent of bereaved individuals were deemed resilient across all outcomes, whereas 20 percent did not show a resilient trajectory in all five outcomes examined. This multidimensional approach permits for studying cross-domain variability and allows for determining whether resilience in specific outcomes co-exist with declines in others.
Our findings have important implications not just for science, but for public policy. Sweeping scientific claims that “most people are resilient” carry dangers of blaming the victims and, more seriously, suggest that external interventions are not necessary to help people hit by traumatic events. We believe that our findings are cause to reconsider the commonness of resilience to major life stressors, and that instead, most individuals are deeply affected, but are able to recover over time. For both empirical and conceptual reasons, it is unwise for scientists to make any definitive statements about the commonness of resilience in the face of major life stressors because, as we have shown, it depends on data analytic choices and the outcomes examined.
References
Infurna, F. J., & Luthar, S. S. (2016). Resilience to major life stressors is not as common as thought. Perspectives on Psychological Science, 11, 175-194.
Infurna, F. J., & Luthar, S. S. (in press). The multidimensional nature of resilience to spousal loss. Journal of Personality and Social Psychology.
New OppNet FOA: Because Life Is a Series of Chronic Conditions
About the Author
Frank J. Infurna, Ph.D., is an assistant professor of psychology at Arizona State University. He received his Ph.D. in human development and family studies from The Pennsylvania State University and his B.A. in psychology and brain and cognitive science from the University of Rochester. He is a developmental psychologist with a general interest in studying psychosocial and health development in adulthood and old age from a lifespan perspective. More specifically, his research examines how older adults adapt resiliently to the developmental challenges and other life stressors that they face. He is also interested in identifying psychosocial factors that promote healthy aging outcomes and has extensive experience in the use of contemporary methods of longitudinal analysis, including latent growth curve (multilevel) modeling, growth mixture modeling, and survival models.
