Koo et al. (Reference Koo, Kõlves and De Leo2017) compared differences in sociodemographic variables, recent life events, and mental and physical illnesses between three age groups (65–74 years, 75–84 years, and 85 years and over) who died by suicide. Psychiatric problems, suicidal behavior, legal and financial stressors, and relationship problems decreased with age. In contrast, physical conditions and bereavement increased with age. Suicide rates were increasing with age for males, but not for females. In addition, significant differences in the prevalence of potential risk factors within the three different age groups existed. I have some concerns about their study.
First, Yeh et al. (Reference Yeh, Ng and Wu2017) conducted a conditional logistic regression analysis to specify risk factors of suicide in the older people. They classified subjects into the same categories of age with those by Koo et al., and crude suicide mortality rates in the young-old, middle-old, and oldest old were 30.57, 35.88, and 43.46 per 100,000 people. In addition, crude suicide mortality rates in male and female were 46.52 and 22.32 per 100,000 people. They adopted propensity score matching for a one-to-one ratio on gender, age, and residence between suicide people and control. Statistical methods of two reports differ, and risk assessment on suicide in older adults should be conducted by setting appropriate controls and by multivariate analysis, although I accept the need of aging effect in older adults.
Second, Diehl-Schmid et al. (Reference Diehl-Schmid2017) summarized risk factors of suicidal intention and behavior in dementia, and patient's cognitive impairment profile, behavioral disturbances, social isolation, and psychiatric comorbidities, such as depression, should be evaluated for predicting suicide intention. There are some different factors in relation to suicide intention, attempt, and completion (Joo et al., Reference Joo2016; Simon et al., Reference Simon2016). In addition, there is a difficulty of identifying risk factors for suicide in patients with dementia. Further study is needed to know factors in older adults, who died by suicide, with stratifying by age.
Conflicts of interest
Not applicable.