Choosing life

Adolescents with Type I diabetes have been reported to have higher rates of mental health problems, including depression and suicidal ideation (1, 2). These results are in line with studies of other chronic illnesses, which increase risk of psychiatric disorder via systemic effects, and by interfering with adolescents’ pursuit of developmentally appropriate activities (3). Yet, while Goldston and colleagues found higher rates of suicidal ideation, they also found lower rates of suicide attempts in adolescents with Type 1 diabetes. The results of the survey reported in this issue by Dr. Radobuljac and colleagues are even more surprising: boys with Type I diabetes had lower rates of suicidal ideation, suicide attempts, and non-suicidal self-injury (NSSI), compared to male controls, whereas the rates of ideation, attempts, and NSSI were similar in female diabetics and controls (4). Why would diabetes be protective against selfdestructive behavior? The authors posit that the influence of diabetes on family lifestyle, through a focus on diet, activities, and medical check-ups, may account for the reported lower rates of health risk behaviors in youthful diabetics compared to their non-diabetic peers (5, 6). Among the strongest protective factors against a wide range of health risk behaviors, including suicidal behavior, violence, and substance abuse, are parental monitoring and supervision, and families eating meals and spending leisure time together (7–9). It is possible that the impact of diabetes compels the family to engage in monitoring and cohesion that are incompatible with the development of health risk behaviors that can predispose to suicidal behavior. Moreover, more frequent medical attention may also account for the higher rates of referral for specialty mental health care in diabetes vs. controls reported in this paper, which in turn may attenuate risk for selfharm behaviors. Interestingly, by the time diabetics reach adulthood and are no longer under the protective umbrella of their families, this putative protective effect is no longer present (10). Why would the protective effect of diabetes be stronger in boys than in girls? Whereas both boys and girls with self-harm behaviors are likely to have depression, the contribution of violent behavior and alcohol and substance abuse to suicidal risk is more prominent in boys (11, 12). Higher parental monitoring and family togetherness are particularly protective against these risk behaviors that differentially plague boys. Unfortunately, this paper does not report on many of the common correlates of suicidal behavior, such as depression, anxiety, alcohol and substance abuse, violence, family cohesion, and family conflict (13). Without knowing the prevalence and severity of these factors in this sample, it is hard to tell whether there is also a protective effect in girls. For example, if diabetic girls had particularly high rates of depression compared to controls, but still showed similar rates of self-harm indices, then, adjusted for background risk, diabetic girls would also be considered to be at lower risk for suicidal ideation, attempts, and NSSI. There may be another aspect of the experience of diabetes that may affect a young person’s attitude about suicide and suicidal behavior. That is, relative to non-diabetic young people, diabetic youth are forced to confront their mortality at a much younger age. All of the life-style requirements, and constant monitoring of glucose, HbA1C, and concomitant possible health complications of poorly controlled diabetes force the diabetic youngster to make an active choice to live on a daily basis. In contrast, the majority of youth who attempt suicide do not clearly think through whether they want to live or die. It is relatively common among youth who attempt suicide by jumping to report that as soon as their descent began, they realized that they had made a terrible mistake. A person who is actively choosing life every day may not be as likely to jump first and only afterward consider the consequences. Awareness of mortality is one thing, whereas preoccupation with an early death is another. Believing that one’s lifespan is considerably foreshortened can have the opposite effect, that of increasing the likelihood of health risk behaviors (14). Some two decades ago, my colleagues and I decided to study the problem of suicidal contagion by interviewing the members of adolescent suicide victims’ social network. We hypothesized that these youth would have the biggest ‘‘dose’’ of exposure to suicide and be the most likely to make an imitative attempt subsequently. Instead, these youth, who had much higher