Assessing Suicide Risk in Children Affected by HIV/AIDS With the House-Tree-Person Drawing Test

Suicide is a serious public health concern, particularly among Chinese children affected by HIV/AIDS. However, these risks are often subtle and obscured by social desirability bias and cultural stigma. This study aims to evaluate the predictive validity of the House-Tree-Person (HTP) drawing test in assessing suicide risk among children affected by HIV/AIDS.

Conducted in western China, this study included 650 children in discovery set and 243 children affected by HIV/AIDS in validation set. Participants completed demographic questionnaires, a suicide risk scale, and the HTP drawing test. Chi-square tests compared drawing characteristics between children with and without suicidal ideation, binary logistic regression identified predictive HTP features, and receiver operating characteristic (ROC) curves assessed accuracy, with the area under the curve (AUC) indicating discriminatory power.

Among children affected by HIV/AIDS, suicidal ideation and behaviour prevalence was 20.3% and 8.2%, respectively. Eleven drawing characteristics significantly predicted suicidal ideation (p < 0.05), and 8 predicted suicidal behaviour (p < 0.05). In the validation set, predictive accuracy reached 88.4% for ideation and 96.0% for behaviour. ROC analysis yielded an AUC of 0.906 for ideation (p < 0.001, 95% CI [0.859, 0.953]) and 0.981 for behaviour (p < 0.001, 95% CI [0.964, 0.997]).

The HTP drawing test proves a valuable, nonverbal tool for early identification of suicide risk in children affected by HIV/AIDS. Its integration into routine assessments can enhance timely interventions, improving mental health outcomes in this high-risk group.