How reliable are self-reports of HIV status disclosure? Evidence from couples in Malawi

Introduction: The majority of research on human immunodeficiency virus (HIV) disclosure utilizes the perspective from a single individual, which cannot be substantiated in the absence of supporting data such as from a primary partner. Objectives: The objectives of this study were to evaluate: (1) the extent to which self-reported HIV disclosure was confirmed by a primary partner; (2) individual and relationship-level predictors of self-reported versus confirmed disclosure; and (3) whether confirmed disclosure was a stronger predictor of correctly assessing a partner's HIV status compared to self-reported disclosure. Methods: As part of an 8-wave longitudinal study from 2009 to 2011 in southern Malawi, 366 individuals (183 couples) were interviewed about their primary relationship (wave 3), individually tested for HIV (wave 4), and then asked whether they disclosed to their primary partner (wave 5). Results: While 93% of respondents reported that they disclosed, only 64% of respondents had confirmed reports from their partner. Having communicated with partner about HIV was positively associated with self-reported disclosure; this association remained significant but became more precise in the models for confirmed disclosure. Confirmed disclosure, but not self-report, was a significant predictor of correctly assessing a partner's HIV status. Being male, having lower perceived partner infidelity, having higher relationship unity, and testing HIV-negative were positively and significantly associated with correct assessment. Dyadic data from two partners provide an improved measure of disclosure as compared to a single individual's self-report and could be used to identify behavioral and biomedical opportunities to prevent HIV transmission within couples. (C) 2015 Elsevier Ltd. All rights reserved.