Establishing an empirical cut-off on the 12-item Brief Berger HIV Stigma Scale to screen psychosocial vulnerability among PLHIV in Nigeria

Technical Summary: Establishing an Empirical Cut-off on the 12-item Brief Berger HIV Stigma Scale

Problem Statement
Despite the widespread use of the 12-item Brief Berger HIV Stigma Scale to measure HIV-related stigma, there is a critical gap in clinical practice: the absence of a validated, empirically derived threshold to identify individuals at high risk for psychosocial vulnerability. In Nigeria, where HIV prevalence is substantial and stigma poses significant barriers to care, the lack of a specific cut-off score limits the utility of stigma assessment tools for early intervention. This study addresses the need to establish a practical, data-driven cut-off score for the scale within the Nigerian context to screen for psychosocial vulnerability, defined as moderate-to-severe depression or anxiety.

Methodology
The researchers conducted an analytical cross-sectional study between May and August 2024 involving 285 people living with HIV (PLHIV) across three tertiary healthcare centers in Lagos State, Nigeria (LUTH, LASUTH, and NIMR). Participants were selected via proportionate stratified sampling.

• Instruments: The study utilized the 12-item Brief Berger HIV Stigma Scale (divided into four subscales: personalized stigma, negative self-image, disclosure concerns, and public attitudes) to assess stigma. Psychosocial vulnerability was defined as a composite outcome: a score of ≥10 on the Patient Health Questionnaire-9 (PHQ-9) indicating moderate-to-severe depression, or ≥10 on the Generalized Anxiety Disorder-7 (GAD-7) indicating moderate-to-severe anxiety.
• Statistical Analysis: The primary objective was to determine an optimal cut-off score using Receiver Operating Characteristic (ROC) curve analysis and the Youden Index. The study evaluated diagnostic accuracy metrics (sensitivity, specificity, predictive values) and the Area Under the Curve (AUC).
• Validation and Modeling: Internal validation was performed using 10-fold cross-validation and 1,000 bootstrap resamples to assess overfitting. Decision curve analysis (DCA) was employed to evaluate clinical net benefit. Multivariable logistic regression was used to identify independent predictors of vulnerability among the stigma subscales, adjusting for sociodemographic variables (age, sex, marital status) and utilizing backward elimination based on the Akaike Information Criterion (AIC).

Key Results

• Sample Characteristics: The cohort (n=285) had a mean age of 47.1 years, was predominantly female (72.6%), and 80% had disclosed their HIV status.
• Prevalence: 44.9% of participants met the criteria for psychosocial vulnerability.
• Diagnostic Accuracy: The Brief Berger Scale demonstrated acceptable discriminatory power with an AUC of 0.717 (95% CI 0.658–0.777).
• Optimal Cut-off: A score of ≥30 was identified as the optimal threshold. This cut-off yielded high sensitivity (87.5%) and a strong negative predictive value (82.8%), though specificity was modest (49.0%).
• Internal Validation: Cross-validation (AUC 0.703) and bootstrap correction (AUC 0.701) confirmed the stability of the model with minimal overfitting.
• Predictors: In multivariable analysis, "Concerns with Public Attitudes" was the strongest independent predictor of psychosocial vulnerability (adjusted OR 1.68, p<0.001). Conversely, "Disclosure Concerns," despite a near-universal prevalence (96.1%), showed no independent association, likely due to insufficient variance. "Negative Self-Image" showed a trend toward significance but did not reach it in the final model.
• Subgroup Analysis: Psychosocial vulnerability was significantly higher in females (49.8%) than males (32.1%). However, the diagnostic performance (AUC) of the stigma scale was comparable across sexes, and no significant interaction by sex was found.

Key Contributions

1. Establishment of a Clinical Threshold: The study provides the first empirically derived cut-off (≥30) for the 12-item Brief Berger HIV Stigma Scale specifically for screening psychosocial vulnerability in a Nigerian tertiary care setting.
2. Identification of Key Drivers: The research highlights that "concerns with public attitudes" are a more potent predictor of mental health vulnerability than internalized stigma or disclosure concerns in this specific population, suggesting that societal stigma may be a more immediate target for intervention than internal psychological states in this context.
3. Validation of Screening Utility: The findings support the use of the scale as a high-sensitivity "rule-out" tool. The high negative predictive value suggests it is effective for identifying patients who do not require immediate intensive psychosocial assessment, thereby aiding triage in resource-limited settings.

Significance and Claims
The authors claim that the ≥30 cut-off offers a practical, sensitive tool for identifying PLHIV in need of psychosocial assessment within resource-limited settings. They emphasize that while the tool shows acceptable discrimination, its modest specificity necessitates confirmatory assessment (via PHQ-9/GAD-7) for positive screens.

The paper maintains a modest stance regarding its broader applicability. It explicitly states that the findings are limited to urban, tertiary-level, treatment-engaged cohorts and may not generalize to rural or lower-level facilities. The authors stress that the cut-off is provisional and requires external validation in independent cohorts before widespread clinical adoption. Furthermore, they note that the cross-sectional design prevents causal inference and that the study's reliance on a single dataset for both deriving and evaluating the cut-off underscores the necessity of future validation studies. The study concludes that while the tool is a step toward integrated HIV care, systemic barriers such as workforce shortages and stigma must be addressed alongside tool implementation.