Estimating the Risk of Depression and Care Burden among Dementia Caregivers: A Feasibility Study

Omid Ghadami, Kruthika Gaddam, Mahesh Moodukonaje, Hyeju Jang, Hee-Tae Jung

Confirmation: I have read and agree with the IEEE BHI 2025 conference submission's policy on behalf of myself and my co-authors.

Keywords: Early intervention, mental health, predic tion, depression, burden, informal caregivers, dementia, Patient Health Questionnaire, Zarit Burden Interview

TL;DR: This paper demonstrates how analyzing caregivers' data using NLP and machine learning can estimate depression risk and care burden, enabling early intervention to improve caregiver well-being and care for PwD.

Abstract: Caregivers of people living with dementia (PwD) are highly susceptible to depression due to the substantial care burden they experience. While caregivers often neglect their own mental health and rarely seek necessary medical services, they often communicate their perceived burdens and depressive symptoms to social workers, who serve as critical points of contact for their loved ones. Thus, accurately estimating the risk of depression and care burden through these conversational interactions may facilitate early screening and intervention. This feasibility study explored the effectiveness of using caregivers' demographic information and their narrative descriptions of caregiving experiences to estimate depression risk and caregiver burden. Utilizing Natural Language Processing (NLP) and machine learning techniques, we trained estimation models based on data from 65 caregivers, using clinical screening measures---the Patient Health Questionnaire-8 (PHQ-8) and Zarit Burden Interview (ZBI)---as reference standards. The best-performing models achieved F1 Scores of 0.74 and 0.78 for depression and burden estimation, respectively. These results demonstrate the promise of leveraging conversational and demographic data for early identification of caregiver distress, which could inform timely interventions and ultimately enhance both caregiver well-being and the quality of care provided to PwD.

Track: 4. Clinical Informatics

Registration Id: LBN5VMYM4PD

Submission Number: 179