Jacobson N, O’Cleirigh C. (2021). Objective digital phenotypes of worry severity, pain severity and pain chronicity in persons living with HIV. The British Journal of Psychiatry. 218(3): 165–167. doi: 10.1192/bjp.2019.168
Researchers recruited adults living with HIV (n = 68) to participate in a study to determine whether objective digital biomarkers developed from participant movement data could accurately predict symptoms of pain and worry in this population. Recruitment occurred in person at a clinic queue in a South African hospital. At baseline, participants reported symptoms of pain and worry. Participants with chronic pain also reported worst pain severity. Each participant received an actigraph which they wore on a belt over their right hip during the day and at night. Accelerometer sensors in the actigraph recorded frequency and intensity of movement. Researchers collected one week of participant actigraphy data. Using the Differential Time-Varying Effect Model and analysis of oscillations and movement distribution, researchers created digital biomarkers (objective, quantifiable physiological and behavioral data) from participant actigraphy data. Researchers compared observed symptom severity values reported in baseline measures with symptom severity values predicted by digital biomarkers to determine whether the digital biomarkers could accurately predict symptom severity. Analysis revealed significant strong correlations between predicted and observed values for worry symptom severity and worst pain severity. Agreement between predicted and observed chronic pain status was moderate (74.63% accuracy). Findings suggest that digital biomarkers created from actigraphy data can accurately capture pain severity, pain chronicity, and worry severity among individuals living with HIV. Future randomized controlled trials could examine whether objective digital biomarkers would be more effective in directing individuals with HIV to specific psychopharmacological therapies or psychotherapies instead of opioid treatments, compared with traditional symptom self-report methods.