Publication Date



Open access

Embargo Period


Degree Name

Master of Science (MS)


Psychology (Arts and Sciences)

Date of Defense


First Committee Member

Gail Ironson

Second Committee Member

Rick Stuetzle

Third Committee Member

Kent Burnett


Objective: The purpose of the present study is to examine whether level of written emotional expression (EE) and emotional/cognitive processing (ECP) for traumatic events predict health status (CD4 and VL), Category B symptoms, depression and anxiety in an HIV+ population over four years. Specifically, two different scoring methods of two variables within ECP (cognitive appraisal and self-esteem) will be compared to see if a change score (SMCHANGE) or a final score (SMFINAL) better predict outcomes. The possible mediating role of ECP in the relationship between EE and outcomes will also be explored. Methods: This longitudinal study assessed 169 HIV+ and diverse men and women in the midrange of illness as indicated by a CD4 number between 150 and 200 and no previous AIDS-defining symptom. EE/ECP data was gathered during baseline assessment and participants attended follow-up assessments every 6 months for a period of 4 years. Hierarchical Linear Modeling was used to examine change over time in CD4, VL log, Category B symptoms, depression and anxiety controlling for age, gender, ethnicity, education, anti-HIV medication and baseline values for each outcome. In addition, analyses for CD4 and VL log were rerun controlling for medication adherence. Results: Positive EE was found to be significantly related to only CD4 and Category B symptoms slopes. Negative EE was not related to any outcome. ECP was found to be related to CD4, VL log and Category B symptoms slope. No relationships were found between EE/ECP and depression and anxiety. SMFINAL scores on ECP subscales were found to predict CD4 and VL log slope better than SMCHANGE, but SMCHANGE scores predicted Category B symptoms slope better than SMFINAL. Within meditational analyses, ECP was found to mediate the relationship between positive EE and CD4 slope controlling for adherence. Positive EE mediated the relationship between ECP and Category B symptoms slope. Conclusions: Higher engagement in positive EE and ECP within emotional writing about a trauma contributes to beneficial changes in health outcomes over time within HIV+ individuals. SMFINAL seems to be more related to CD4 and VL log slope while SMCHANGE seems to be more related to Category B symptoms slope, indicating that both scoring methods within ECP seem to be valuable. Findings support the meditational role of ECP between EE and CD4, and provide new evidence that positive EE plays a meditational role between ECP and Category B symptoms. These findings can be used to help improve health for patients in future studies or in CBT therapies.


HIV, Emotional Expression, Cognitive Processing, Scoring, Trauma, Writing