Background: Self-esteem and optimism have been associated with appraisal and outcomes in a variety of situations. The degree to which the contribution of self-esteem and optimism to outcomes over time is accounted for by the differences in threat (primary) or resource (secondary) appraisal has not been established in persons with human immunodeficiency virus (HIV).
Objectives: To examine the longitudinal relationship of personality (self-esteem and optimism) on primary and secondary appraisal and outcomes of well-being, mood, CD4 + T-lymphocyte count, and selected activities.
Methods: Men (n = 56) and women (n = 42) infected with HIV completed eight self-report measures twice over 18 months. Hierarchical Multiple Regressions were used to examine the relationship of personality variables on appraisals and outcomes. The mediating effects of primary and secondary appraisals were explored.
Results: Self-esteem uniquely accounted for 6% of the variance in primary appraisal and 5% in secondary appraisal. Optimism accounted for 8% of the unique variance in secondary appraisal. Primary and secondary appraisal mediated differently between personality and outcome variables. A strong predictor of well-being, mood disturbance, and activity disruption at Time 2 was participants' initial level of these variables. Socioeconomic status was a strong predictor of mood.
Conclusions: Self-esteem and optimism are important but different resources for adapting to HIV disease. Strategies for reducing threats and increasing resources associated with HIV may improve an individual's mood and sense of well-being.
Elizabeth H. Anderson, PhD, RN, is an Assistant Professor, University of Connecticut School of Nursing, Storrs, Connecticut.
Accepted for publication December 1, 1998.
Supported in part by a predoctoral NRSA (F31 NRO6291), the American Nurses Foundation, University of Rochester School of Nursing Alumni Seed Fund and Susan B. Anthony Faculty Research Grant.
The author thanks Kathleen King, PhD, RN, Professor, University of Rochester School of Nursing, Rochester, New York, for chairing the dissertation committee and Jean E. Johnson, PhD, RN, Professor Emeritus, University of Rochester, School of Nursing, Rochester, New York, for postdoctoral mentorship. Special thanks to research assistants Judith Quinn, MSN, and Ann Abrahamson, MSN, University of Rochester, and Mei-Ling Yeh, PhD, University of Maryland, and to statisticians who reviewed this paper, Barker Bausell, PhD, University of Maryland. and Stephen Owen, PhD, University of Connecticut.
Address reprint requests to Elizabeth H. Anderson, PhD, RN, Assistant Professor, University of Connecticut School of Nursing, 231 Glenbrook Road, U-2026, Storrs, CT 06269-2026; E-mail: firstname.lastname@example.org.