Comparing Biological and Non-Biological Measures of Stress

Background and Purpose: Investigators have paid increasing attention to stress, based on evidence of its connection with disease. There is confusion, however, about what the term “stress” means and how it should be measured. Stress has been defined and measured in three ways that are used as proxies for one another. One is as perceived stress, usually measured by self-report scales such as the Perceived Stress Scale (Cohen, et al., 1983). Stress also is defined as life stressors, and measured by counting those stressors using scales like the Life Events Questionnaire (Hobson, et al., 1998) Lastly, stress is measured as biological changes, using biomarkers. These can be divided into those that measure immediate stress and those that measure long-term stress. Immediate stress is measured by diurnal salivary cortisol at keys times during the day. Long-term stress, called allostatic load, is measured here using serum cortisol. In reality the three conceptualizations of stress measure different things. Using the measures interchangeably can produce error that affects research outcomes. We hypothesize that measures of stress taken in the same sample of women will not be identical, and vary statistically significantly from one another.

Methods: The study sample was 230 African-American women living in 15 neighborhood areas of Chicago. Data were collected in three ways. Interviewers visited women in their homes to assess a variety of psychosocial and other factors, during which they were given the Perceived Stress Scale and the Life Events Scale. Women were taught to collect salivary cortisol that later was retrieved by RAs. Collection was four times per day for three consecutive days, as per standard protocol. Blood samples were taken during clinic visits. The ability to collect multiple measures from the same group of women allows comparison across measures of stress.

Results: Although immediate stress and long-term stress were highly correlated (r = 0.76, p < .01), neither was positively correlated with either perceived stress (r = -0.26, p <.05; r = -0.39, p < .01) or life events (r = -0.11, p = .37, r = -0.15, p =.19). This suggests that perceptions of stress do not reflect the physiological stress that women are experiencing. There was, however, a significant association between perceived stress and life events (r= 0.20, p <0 .05) suggesting that the two measures are related.

Conclusions and Implications: The two measures of biological stress were associated with one another as were the two non-biological measures. Yet biological and non-biological measures were not significantly related. This suggests that measures of stress cannot be used interchangeably and that researchers should note that outward measures of stress do not reflect biological processes. This is important for social work researches, because stress is strongly associated with disease and must be measured precisely.

References

Cohen, S. et al. 1983. A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385-396.

Hobson, C. J., et al. 1998. Stressful life events: A revision and update of the social readjustment rating scale. International Journal of Stress Management, 5(1), 1-23.