The goal of this observational study is to learn about the ability of a point risk score prediction model, developed using electronic medical record data, to predict the risk of progression from geriatric depression to dementia in older Chinese adults. The main questions it aims to answer are:

Does a higher point risk score increase the risk of developing dementia in older adults with depression?

What is the accuracy of the point risk score prediction model in identifying individuals at high risk of dementia among older adults with depression?

Participants will receive their usual medical care as they normally would. No new treatments, tests, or procedures will be performed specifically for this study. The research team will collect data from their electronic medical records, including depression diagnoses, dementia diagnoses, comorbidities, medication records, and follow-up information. The point risk score will be calculated based on these routinely collected clinical data.

Inclusion Criteria:

• Adults aged ≥50 years.
• A diagnosis of depression clearly recorded by a clinician in the electronic medical record at baseline (based on structured or unstructured diagnostic documentation formed through routine clinical practice).
• Complete baseline electronic medical record data available, including at least demographic information (age, sex), clinical diagnoses, comorbidities, and medication records.
• At least one follow-up record available in the electronic medical record system to enable determination of incident dementia outcomes.
• Informed consent provided for the use of routine medical data for this research analysis.

Exclusion Criteria:

• Any type of dementia diagnosis recorded in the electronic medical record at baseline (including Alzheimer's disease, vascular dementia, and other types of dementia).
• Medical record documentation indicating that the diagnosis of dementia preceded the diagnosis of depression, or an inability to clearly determine the chronological order of the two diagnoses.
• Presence of other neurological diseases at baseline that may independently cause severe cognitive impairment (e.g., Parkinson's disease, multiple sclerosis, brain tumor, normal pressure hydrocephalus).
• Missing key baseline electronic medical record data that would preclude subsequent risk model analysis.
• Incomplete follow-up information or inability to clearly confirm incident dementia outcomes through the electronic medical record system.
• Refusal to participate in the study or withdrawal of informed consent.

The study population consists of older adults aged ≥50 years with a baseline diagnosis of depression, derived from a prospective cohort at the data center of the Second Affiliated Hospital of Nanchang University, China (2010-2025). All participants are patients receiving routine clinical care at the hospital. Data are prospectively collected through the hospital's electronic medical record system, including demographic information, clinical laboratory indicators, depression records, and incident dementia diagnoses during follow-up, to identify preclinical risk factors for progression from geriatric depression to dementia and to develop a point risk score prediction model.