Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Psychological and Brain Sciences

Degree Program

Clinical Psychology, PhD

Committee Chair

Sephton, Sandra

Committee Co-Chair (if applicable)

Salmon, Paul

Committee Member

Salmon, Paul

Committee Member

Mast, Benjamin

Committee Member

Woodruff-Borden, Janet

Committee Member

Newton, Tamara

Committee Member

Cash, Elizabeth

Author's Keywords

cortisol; lung cancer; positive affect; cancer patient


Positive psychobiological processes within lung cancer patients are drastically understudied. This dissertation explores the nature of positive affect (PA) and potential associations with diurnal cortisol among lung cancer patients, given the prognostic significance of diurnal cortisol rhythms. Theoretical underpinnings and current literature involving PA, cancer, and diurnal cortisol are first reviewed. An original integrated model of PA and cortisol among cancer patients is then presented, from which the proposed dissertation study and analyses are derived. Sixty-one non-small cell lung cancer patients provided self-report assessment of mood (PANAS PA and NA subscales, CES-D PA subscale), medical and demographic characteristics, and 10-day salivary cortisol. Aim 1 tested hypotheses that: (1) patients will experience moderate PA, and more PA than NA, (2) PA and NA will emerge as separate factors in factor analyses, and (3) higher PA will correlate with variables reflecting lower disease burden. Aim 1 was assessed through descriptive statistics, correlations, t-tests, and exploratory factor analyses. Aim 2 tested hypotheses that: (1) higher PA will relate to lower cortisol means, (2) higher PA will relate to steeper diurnal ���� slope, (3) PA will relate more strongly to overall mean cortisol than diurnal slope. Aim 2 was tested through hierarchical linear regressions and path analyses. Aim 1 results showed that patients generally held positive emotions and endorsed PA items that reflected determination and resilience. They also reported more PA than NA; these two constructs emerged as separate and distinct factors. Race, smoking, and current treatment all significantly related to PA. Aim 2 revealed that PA did not significantly associate with mean cortisol variables or diurnal slope. However, higher NA was associated with flattened slopes, after excluding patients taking corticosteroids. Although the relationship between PA and mean cortisol was consistently stronger than with diurnal slope in path analyses, the associations were non-significant. Patients reported experiencing positive emotions that may reflect resilience and adaptive coping. Positive affect did not have strong associations with cortisol, which may be due to pre-existing cortisol dysregulation or small sample size. Future studies should continue to explore mind-body associations of positive psychological processes in lung cancer patients.