Wendi Da

Date of Award

Fall 2019

Document Type

Open Access Dissertation


Health Promotion, Education and Behavior

First Advisor

Shan Qiao

Second Advisor

Xiaoming Li


With prolonged lives thanks to antiretroviral therapy, parents living with HIV (PLH) face challenges regarding telling their HIV-positive status to children (i.e., parental HIV disclosure). With aims to assist PLH in making a well-planned and developmentally appropriate HIV disclosure to their uninfected children, a theory-driven intervention was conducted among 791 PLH with children aged between 6-15 years in Guangxi, China.

Guided by the stage model of the Health Action Process Approach (HAPA), three critical disclosure stages were defined: 1) the pre-intention stage, in which people have not yet decided to disclose; 2) the intention stage, in which people have decided to disclose but have not yet started action; 3) the action stage, in which individuals make actual disclosure event. Accordingly, people at the three stages were defined as pre- intenders, intenders, and actors.

Using secondary data from the baseline (W1) and the first two follow-ups (W2 at 6-month and W3 at 12-month) data, this dissertation evaluated the intervention effect on parental HIV disclosure stages and examined the roles disclosure-related psychosocial factors play in the process. Three major research questions were addressed: 1) is there any intervention effect on HIV disclosure stage transition between W2 and W3? 2) is there any intervention effect on disclosure-related psychosocial factors from W1 to W2 and from W2 to W3? 3) do disclosure-related psychosocial factors at W2 yield stage- specific predictive effects on disclosure stage transition from W2 to W3?

To address the first question, a multigroup first-order manifest Markov Chain method was conducted to assess intervention effect on disclosure stage transitions between W2 and W3. Among pre-intenders at the first follow-up, those in the intervention group were more likely to progress to the action stage rather than being static (OR = 3.43, 95% CI 1.17, 10.01). However, no statistically significant intervention effect was detected in promoting progression from pre-intention to intention (OR = 1.02, 95% CI 0.47, 2.20). Among intenders at the first follow-up, no statistically significant intervention effect was detected in preventing backward transition to pre-intention (OR = 0.71, 95% CI 0.35, 1.43) or promoting forward transition to action (OR = 2.01, 95% CI 0.84, 4.79).

To address the rest two questions, a proportional latent change score (LCS) method was used to assess intervention effect on disclosure-related psychosocial factors including knowledge, outcome expectancy (perceived costs and rewards), self-efficacy, and planning. Predictive effects of these psychosocial factors on disclosure stage transitions were examined by treating these factors as covariate in the Markov chain. At the first follow-up, statistically significant intervention effect was detected for disclosure knowledge, action self-efficacy, and action planning but not for disclosure outcome expectancy. Stage-specific predictive effects of action self-efficacy and action planning on HIV disclosure stage transitions were detected.

Findings from the dissertation have significant implications for future studies.

First, the 3-stage HAPA model can be applied to the evaluation of HIV disclosure interventions. Second, the identification of stage-specific psychosocial predictors of stage transition allows the development of stage-matched interventions tailored to their needs.