Abstract
Psychosocial factors play a role in HIV progression. Expressive writing (EW) interventions have demonstrated positive effects on psychosocial and disease-related outcomes in HIV. Linguistic properties of written essays have been related to psychological outcomes, although such research is sparse in the HIV literature. This project examines the relationship of word usage to psychosocial and disease-related factors in HIV. As part of a larger randomized, controlled trial of EW, this project examined the subset of subjects (n = 121) randomized to write about trauma. Linguistic Inquiry and Word Count (LIWC) software was used to analyze word usage (pronouns, positive, negative, social, and religious words) in relationship to psychosocial (depression, PTSD, social support, and stress) and physical (ART adherence, physical HIV symptoms, CD4 count, viral load) outcomes over 6 months of follow-up. Negative word usage predicted increased physical HIV symptoms over 6 months, while positive word usage predicted decreased symptoms over the same period, controlling for baseline HIV symptoms, CD4, and viral load. There was a trend toward social word usage predicting decreased HIV symptoms over 6 months, controlling for baseline HIV symptoms. Word usage did not predict ART adherence, CD4 count, viral load, or any psychosocial factors. There was a trend toward social word usage predicting lower stress levels at 6 months. In conclusion, positive word usage in EW essays predicted decreased HIV symptoms while negative word usage predicted greater HIV symptoms, over 6 months, controlling for baseline HIV symptoms. Future studies of written EW essay content are warranted in HIV.