Voice Quality in Individuals With Acquired Immunodeficiency Syndrome.

Only 29 people were studied and the groups were unequal: 20 with AIDS against 9 controls. The authors say this limits how far the findings can be generalised. Everyone in both groups had metabolic conditions such as dyslipidaemia, and everyone with AIDS also had lipodystrophy, so this is not a comparison of AIDS against healthy ageing. The voices were recorded once, so nothing here shows whether the pattern changes over time or how it develops. The authors state that the pattern reflects the combined effect of long-term antiretroviral therapy, opportunistic disease, lipodystrophy, ageing and metabolic change, and that the contribution of any one of these cannot be separated out. The acoustic measures separated the groups poorly and differed between groups only in the male subgroup. Recordings were made in 2012. The authors argue the stored data remain valid, but the treatment context has changed since then. The study was not designed to test any treatment, and the authors say the findings do not support proposing a therapy protocol.

Funded by the Brazilian National Council for Scientific and Technological Development (CNPq), grant 311596/2022-23. The authors declare no conflict of interest.

Do not read these voice patterns as caused by AIDS, by antiretroviral therapy or by lipodystrophy. Everyone in the AIDS group had all of these at once, plus ageing and metabolic conditions, and the authors state that the findings reflect their combined effect and that no single factor can be separated out. Do not read the 98.68% and 98.31% figures as a measure of how well anyone can spot these voices by ear in clinic. They describe how well a statistical model sorted already-labelled samples using three trained judges' descriptions.