Sunny Tang’s Post

𝐍𝐞𝐰 paper in press: 𝐬𝐩𝐞𝐞𝐜𝐡-𝐛𝐚𝐬𝐞𝐝 𝐝𝐢𝐠𝐢𝐭𝐚𝐥 𝐛𝐢𝐨𝐦𝐚𝐫𝐤𝐞𝐫𝐬 𝐟𝐨𝐫 𝐧𝐞𝐠𝐚𝐭𝐢𝐯𝐞 𝐬𝐲𝐦𝐩𝐭𝐨𝐦𝐬 Assessment of negative symptoms in schizophrenia spectrum disorders (SSD) remains challenging with traditional clinical methods. Digital health technologies, including automated speech analysis, offer a potential solution, but progress has been hindered by heterogeneity in methods and uncertainty about which speech features have the greatest clinical utility. Michael Spilka, Sunny Tang and colleauges systematically evaluated 𝟑𝟖 acoustic and linguistic speech features in 𝟔𝟐 individuals with SSD across baseline and follow-up visits. The features were examined in relation to: 💡 Test–retest reliability 💡 Associations with clinician-rated negative symptoms 💡 Convergent and discriminant validity 💡 Specificity to negative symptom severity 💡 Clinical and demographic characteristics 𝐊𝐞𝐲 𝐬𝐩𝐞𝐞𝐜𝐡-𝐛𝐚𝐬𝐞𝐝 𝐦𝐚𝐫𝐤𝐞𝐫𝐬 Three features emerged as robust speech-based markers across tasks and visits: 💫 Speech proportion 💫 Speech rate 💫 Unfilled pauses These markers demonstrated adequate or better reliability, consistent associations with negative symptoms, discriminant validity, specificity, and no associations with antipsychotic medication or extrapyramidal symptoms. Speech rate additionally showed convergent validity with an alternative negative symptom measure and was the most reliable feature at the level of a single task administration. Schizophrenia International Research Society (SIRS) First version of this paper can be read here: 🔗 https://lnkd.in/eaGyWN57