Stress, Health, and Well-Being across the Adult Life Span: A Multiple Time-Scale Approach

Restricted (Penn State Only)
Koffer, Rachel Elyse
Graduate Program:
Human Development and Family Studies
Doctor of Philosophy
Document Type:
Date of Defense:
July 10, 2018
Committee Members:
  • Nilam Ram, Dissertation Advisor
  • David Manuel Almeida, Committee Chair
  • Nilam Ram, Committee Member
  • Timothy Raymond Brick, Committee Member
  • Idan Shalev, Outside Member
  • intraindividual variability
  • daily stress
  • entropy
  • lifespan development
  • aging
Daily stressors—minor hassles that disrupt daily life—and the biological and psychological responses to those stressors are associated with both proximal and distal health and well-being outcomes. Major stress process theories (e.g., stress and appraisal theory, allostatic load theory) suggest daily stressors may “pile up” over time (Lazarus & Folkman, 1984; McEwen, 1998). However, pathways that account for these associations across the adult life span are largely unknown and require a multi-timescale biopsychosocial approach to daily stress. At micro time-scales, new methods are needed to capture fluctuation in repeated measures (i.e., intraindividual variability, or IIV) and associated constructs. For example, how can we capture the fluctuation of stressor types an individual experiences across a typical week? How can we characterize robustness of diurnal cortisol regulation, a well-known biological marker of daily stress? At macro time-scales, empirical tests of theoretical biopsychosocial stress processes are needed to understand developmental change in stressor IIV and its long term outcomes. This dissertation consists of three studies that advance methods for analysis of daily stressor and cortisol IIV, examine developmental differences in stressor and cortisol IIV, and test stress process theory linking these IIV constructs to long-term biological and psychological outcomes. Study 1 illustrates how an intraindividual variability (IIV) framework can be used to interpret categorical repeated measures data. Experience sampling studies often collect categorical repeated measures, including stressor types, activity types, and social partner types. Such variables are often analyzed as raw frequency counts, aggregate sums, or collapsed binary indicators. Using 8-occasion categorical data on daily stressors from the National Study of Daily Experiences (N= 1,499, MAge = 46.74, SDAge = 12.91) I derive and compute six IIV metrics that invoke numeric and nominal measurement of the central tendency, dispersion, and asymmetry of individual’s stressor experiences. I then examine how these metrics of IIV—relative dominance, diversity, log-skew and mode, spread, order—are related to age and interindividual differences in negative affect. Results demonstrate the utility of the numeric and nominal categorical IIV metrics; age gradients in the three numeric IIV stressor metrics reflect the changing roles and selective experiences that accompany socioemotional aging and five of six IIV metrics map differences in negative affect. Findings highlight how the unique constructs measured by these six metrics of categorical IIV may be used to examine dynamic process, study interindividual and age-related differences, and expand the variety of developmental research questions that may be answered using categorical repeated measures data. Study 2 utilizes metrics from Study 1 to examine links between “chronic stressors,” as indicated by high daily stressor exposure and low stressor diversity (i.e., many stressor experiences concentrated in one or few types), and chronic inflammation. A growing body of research has identified characteristics of stressor experiences (e.g., duration, controllability) that contribute to higher inflammation. The combination of high stressor exposure and low stressor diversity may lead to allostatic load due to failure to adapt to repeated stressors. Older adults may be at greater risk, given tendency to have lower stressor diversity than younger adults. Data for this study come from 1,011 adults aged 34-84 years who provided eight consecutive days of self-reported stressor experiences in the National Study of Daily Experiences and blood samples assayed for six inflammatory markers in the Biomarker Project of the Midlife in the United States study. A structural equation model was used to examine the associations among systemic and endothelial inflammation, stressor exposure, stressor diversity, and age. As expected, results indicated higher stressor exposure combined with lower stressor diversity was related to higher endothelial inflammation. This combination is discussed as failure to adapt to consistently repeating stressor experiences. Study 3 uses an IIV approach to examine the relation between Hypothalamic-Pituitary-Adrenal axis (i.e., HPA; anatomical structures that mediate physiological stress response) dysfunction and immune function. HPA axis function has a distinct diurnal pattern, peaking 30 minutes after waking, and declining thereafter until flattening in the evening. Links between cortisol trajectories and health have primarily examined individuals’ typical cortisol levels and slopes. However, individuals differ in the extent that they deviate from their typical diurnal pattern. To describe each individual’s typical diurnal trajectory, person-specific four-part linear spline models were fit to repeated measures of salivary cortisol (4 samples of salivary cortisol on 4 consecutive days, total of 26,191samples) obtained from N = 1705 participants (Age Range: 33-84 years) across 6,714 days. Summary measures of variability describe the extent that each individuals’ repeated measures of cortisol fluctuated (i.e., cortisol variability) around his/her typical diurnal trajectory. A structural equation model was then used to test associations among diurnal cortisol trajectory slopes, cortisol variability, age, and two latent factors of inflammation (systemic and endothelial). Contrary to expectations, the combination of steeper slopes and low variability in diurnal trajectory related to higher endothelial inflammation. Implications of cortisol variability as an indicator of HPA dysregulation is discussed, particularly as the present findings did not replicate those of the new cortisol variability literature. Together, these papers aimed to elucidate associations among psychosocial stressor experiences, affect, and biological dysregulation at multiple time-scales in ways that support and inform major stress process theories. Each paper highlighted the usefulness of an IIV approach, and presented a novel method of examining the day-to-day dynamics of the stress process. Each paper additionally offered mathematical operationalization of stressor and stress response “pile-up” suggested in Lazarus and Folkman’s (1984) psychological stress and appraisal and McEwen’s (1998) allostatic load theories. Age differences in many of the forwarded metrics suggest developmental and aging-based changes across the stress process, further supporting the need for multiple time-scale designs.