Evaluation of accuracy and performance of self-reported HIV and antiretroviral therapy status in the Nigeria AIDS Indicator and Impact Survey (2018)

Abstract

Background Data on awareness of HIV status among people living with HIV (PLHIV) are critical to estimating progress toward epidemic control. To ascertain the accuracy of self-reported HIV status and antiretroviral drug (ARV) use in the Nigeria HIV/AIDS Indicator and Impact Survey (NAIIS), we compared self-reported HIV status with HIV rapid diagnostic test (RDT) results and self-reported ARV use with detectable blood ARV levels. Methods On the basis of responses and test results, participants were categorized by HIV status and ARV use. Self-reported HIV status and ARV use performance characteristics were determined by estimating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Proportions and other analyses were weighted to account for complex survey design. Results During NAIIS, 186,405 participants consented for interview out of which 58,646 reported knowing their HIV status. Of the 959 (weighted, 1.5%) who self-reported being HIV-positive, 849 (92.1%) tested HIV positive and 64 (7.9%) tested HIV negative via RDT and polymerase chain reaction test for discordant positive results. Of the 849 who tested HIV positive, 743 (89.8%) reported using ARV and 72 (10.2%) reported not using ARV. Of 57,687 who self-reported being HIV negative, 686 (1.2%) tested HIV positive via RDT, with ARV biomarkers detected among 195 (25.1%). ARV was detected among 94.5% of those who self-reported using ARV and among 42.0% of those who self-reported not using ARV. Overall, self-reported HIV status had sensitivity of 52.7% (95% confidence interval [CI]: 49.4%–56.0%) with specificity of 99.9% (95% CI: 99.8%–99.9%). Self-reported ARV use had sensitivity of 95.2% (95% CI: 93.6%–96.7%) and specificity of 54.5% (95% CI: 48.8%–70.7%). Conclusions Self-reported HIV status and ARV use screening tests were found to be low-validity measures during NAIIS. Laboratory tests to confirm self-reported information may be necessary to determine accurate HIV and clinical status for HIV studies in Nigeria.

Brief Report: Self-Reported HIV-Positive Status but Subsequent HIV-Negative Test Results in Population-Based HIV Impact Assessment Survey Participants—11 Sub-Saharan African Countries, 2015–2018

“…However, this proportion is smaller than several other studies that report up to 10% of discordancy between self-report and HIV-positive status. 7,8,12 Various factors may contribute to this discordancy which include (1) wrong self-report of HIV status, (2) initial false-positive HIV diagnosis, or (3) misunderstanding of the questionnaire resulting in wrong response during the survey. Although HIV diagnostic tests are highly reliable and accurate, studies have reported false-positive HIV diagnoses in all settings.…”

“…For example, Ibrahim and colleagues reported positive predictive value of 92% for self-reported HIV-positive status in a large HIV-focused survey in Nigeria. 7…”

“…However, the sensitivity of self-reported HIV status, using detected antiretroviral (ARV) as the gold standard, is generally .90%. 7,8 Less commonly reported is the positive predictive value or the proportion of persons who report a prior positive HIV test who are subsequently determined to be HIV-negative. For example, Ibrahim and colleagues reported positive predictive value of 92% for self-reported HIV-positive status in a large HIV-focused survey in Nigeria.…”

“…For example, Ibrahim and colleagues reported positive predictive value of 92% for self-reported HIV-positive status in a large HIV-focused survey in Nigeria. 7 We used aggregated data of individuals from nationally representative Population-Based HIV Impact Assessment (PHIA) surveys conducted in Cameroon, Côte d'Ivoire, Eswatini, Ethiopia, Lesotho, Malawi, Namibia, Tanzania, Uganda, Zambia, and Zimbabwe from 2015 to 2018 to determine the proportion and characteristics of participants who self-reported being HIV-positive but subsequently tested HIV-negative.…”

HIV prevalence, risk behaviour, and treatment and prevention cascade outcomes among cisgender men, transgender women, and transgender men who sell sex in Zimbabwe: a cross-sectional analysis of programme data

“…127 The low self-reported ART coverage among people who knew their status (15-25%) is in sharp contrast to the estimated 91% 18 coverage among the Zimbabwean general population, 67% among FSW, 36 45% among MSM 45 and 34% among transgender women. 45 While the accuracy of self-reported ART status can vary greatly and is often subject to quite extreme under reporting in surveys, 47,48 misreporting ART use within the context of a programme is likely much less as it results in immediate and supported linkage to ART.…”