|
Abstract
|
Introduction
Many aging adults fear developing Alzheimer's disease or a related dementia (ADRD). After being diagnosed with ADRD, feelings of loss, anger, and uncertainty are common. At a time when the U.S. suicide rate among older adults has accelerated and the prevalence of ADRD is increasing, research is needed to quantify suicide risk in persons with newly diagnosed dementia and to identify patient characteristics that are associated with short-term risk for suicide. Past research is mostly limited to small samples with cross-sectional or case series designs. Prior cohort studies primarily come from Denmark or Asia and are limited by narrowly selected samples and inconsistent accounting for time since dementia diagnosis. We present results from a U.S. national study of suicide risk in adults aged >65 years during the first year following newly diagnosed dementia. Competing risk regression was used to identify risk factors of suicide. Standardized mortality ratios (SMRs) were calculated to quantify suicide risk using the U.S. Compressed Mortality File.
Methods
We extracted a cohort of adults aged >65 years with newly diagnosed ADRD (2011-2016) from Medicare claims linked to the National Death Index. After identifying patients with an ADRD diagnosis on an outpatient or inpatient claim, we limited the cohort to those without any claim for an ADRD diagnosis in the prior 12 months. Dementia included Alzheimer's, vascular, Lewy bodies, frontotemporal, or unspecified. Suicide was defined as cause of death coded as intentional self-injury/suicide (ICD-10 codes X60-X84, Y87.0, U03). Recent mental health and substance use disorders were defined as >1 inpatient or >2 outpatient diagnoses within 180 days prior to index ADRD diagnosis date. Severity of medical comorbidity was evaluated with the Elixhauser Comorbidity Scale, excluding mental health, substance use, and neurological disorders. Chronic pain was classified using a previously developed research algorithm for administrative data. Crude suicide rates per 100,000 person-years were estimated for the first year after ADRD diagnosis. Time-to-event analysis for suicide mortality was conducted using Fine and Gray's competing risks analysis.
Results
Of the 2,667,987 older adults with ADRD, 62.2% were women, 82.5% were non-Hispanic White, and 46.5% aged 85+ years. The suicide rate was 26.42 per 100,000 person-years. After adjusting for demographic variables, rural residence, recent mental health and substance use disorders, and chronic pain conditions were associated with increased risk. Relative to the general population, the overall SMR for suicide was 1.54 (95%CI=1.42, 1.65) with the highest risk among ADRD adults aged 65-74 years (SMR=3.43, 95%CI=2.96, 3.89). Approximately half of all suicide deaths occurred within 90 days.
Conclusions
In this national U.S. study of over two million older adults, we found that the number of observed suicide deaths within the first year of dementia diagnosis was approximately 54% higher than expected when compared to the general geriatric population. The risk of suicide mortality was particularly elevated among adults aged 65-74 years and within the first 90 days of dementia diagnosis. Our study identified a number of specific factors associated with higher risk of suicide, which may assist health care providers in decisions regarding appropriate, potentially life-saving care during the critical period following initial diagnosis of dementia. Limitations include our inability to validate the accuracy of dementia diagnoses in Medicare claims data, which may not include all cases of dementia. Second, our analyses did not include Medicare Advantage enrollees or older adults who receive health care from the Veterans Health Administration. Third, information was not available concerning additional suicide risk factors, such as other proximal stressful life events, social disconnection (e.g., marital status, loneliness), and access to lethal means. Third, the data used to estimate SMRs were based on mortality patterns from January 1, 2011, to December 31, 2016, which was a period marked by increasing U.S. suicide rates for males and females aged 65-74 years, and may not generalize to other epochs. Finally, competing mortality risks may complicate detection of and contribute to underestimated counts of suicide deaths in older adults, particularly in the >75 years of age group. The clinical and policy implications of these results highlight the importance of suicide risk screening and support at the time of diagnosing incident dementia, particularly for patients who are younger (i.e., aged 65-74 years) and for those with a history of chronic pain or mental health or substance use disorders. These findings also support mental health care, thorough assessment of patient and caregiver needs, initiating referrals for services and supports, and lethal means restriction and safety counseling during the critical transitional period following diagnosis of dementia.
Funding
This research was supported by grants from NIMH (R01MH107452-02S1) and American Foundation for Suicide Prevention (YIG-0-148-19).
Language: en |