Older Adults Have the Highest Suicide Rates in the United States: Can Risk Algorithms Help Identify Those at Highest...

Area of Science:

• Geriatric Medicine
• Public Health
• Data Science in Healthcare

Background:

• Older adults, particularly those 75 and older, face elevated suicide risks.
• Existing suicide risk algorithms show promise but haven't been evaluated across different age strata of older adults.
• Emergency departments are critical contact points for identifying suicide risk in older populations due to high rates of treatment and readmission.

Purpose of the Study:

• To develop and evaluate suicide risk prediction models for adults aged 65 and older using electronic health records.
• To compare the performance of these risk algorithms across distinct age groups: 65-74, 75-84, and 85 years and older.

Main Methods:

• A retrospective cohort study utilized electronic health records from the Cosmos database (2019-2024).
• Data included demographics and International Classification of Diseases version 10 (ICD-10) codes for 856,132 adults aged 65+ seen in US emergency departments in December 2023.
• Models predicted suicidal behaviors (nonfatal or fatal) in 2024 using historical data (2019-2023).

Main Results:

• Suicide risk algorithm performance declined with increasing age.
• The area under the receiver operating characteristic curve (a measure of performance) was 0.85 for ages 65-74, 0.83 for 75-84, and 0.81 for ages 85+.

Conclusions:

• Current suicide risk algorithms are less effective in identifying risk among the oldest adults (85+).
• Enhanced strategies are necessary to improve suicide risk detection for older adults in general healthcare settings like emergency departments.