Adult Outcomes of Childhood Bullying Victimization

In “Tom Brown’s School Days” (1) the evil bully Henry Flashman roasts young Tom Brown in front of the great fire in the Rugby school common room until Tom’s trousers are seared onto his thighs. Flashman is soon expelled from Rugby—but for drunkenness, not for bullying. He goes on to become Sir Harry Paget Flashman, VC, KCB, KCIE, an “illustrious Victorian soldier” (according to Wikipedia) and the hero of a dozen novels.

Thomas Hughes, the early Victorian author of the Tom Brown stories, has Tom go to Oxford, fall in love, and come to no harm from his experiences at Flashman’s hands. In this, Hughes’s attitude is one held by most parents and teachers until quite recently. As part of the process of growing up to be “a brave, helpful, truth-telling Englishman, and a gentleman” (p. 69), being bullied is inevitable if not actually a good thing.

The Arseneault group (Takizawa et al. [2]) brings this Victorian view of bullying up to date. They use data from a 50-year-old study to document the adult consequences of being bullied as a child. Parents of a representative birth cohort of British children born in 1958, interviewed when their child was 7 years old, and again at 11, reported whether their child was bullied by other children never, sometimes, or frequently. After adequate controls for attrition and potential confounds, answers to this single question predicted many kinds of poor biopsychosocial functioning decades later: psychological distress at ages 23 and 50, depression at age 45, poor physical health at ages 23 and 50, and poorer cognitive function at age 50. It is worth noting that these are the ages at which these variables were measured; absence of effects at other ages means that variables were not measured, not that the results were nonsignificant.

So bullying predicts poorer functioning up to 40 years later. But how much does this matter in real-world terms? The authors have tested this by comparing the effects of bullying to those of other childhood problems known to bode ill for adult functioning: placement in public or substitute care; unattractive physical appearance; and being in the worst quartile on a scale of poor parenting, physical or sexual abuse, poverty, and parental mental illness or drug problems. The risk of problems in adulthood that were created by bullying was of the same magnitude in each case. Furthermore, the outcomes remained significant after controlling for a wide range of correlated risks.

In a 2010 study of monozygotic twins discordant for bullying (Bowes et al. [3]), the Arseneault group showed that there is a large genetic component to both bullying and being bullied. Thus, the present study, with its long timeline, suggests that being bullied may affect the environment in which the next generation of genetically vulnerable children is raised, so that children of bullied parents are doubly at risk of being bullied themselves.

How can bullied children be helped? Being bullied is not an illness or disorder; rather, it is an exposure to which some children are more vulnerable than others. Standard public health approaches to a dangerous exposure are generally grouped into three categories: 1) reduce the areas, places, times, or settings where people might be exposed (primary or universal prevention); 2) increase the resilience of those exposed (secondary or targeted prevention); and 3) reduce the damage caused by the exposure (tertiary or indicated prevention). Progress has been made in the first category, especially through various types of school programs, although the results are patchy (4). Professor Arseneault and colleagues have provided an example of secondary prevention in their elegant study of monozygotic twins concordant for being bullied (3). With genetic (and many environmental) factors held constant, the twin who received more parental warmth had fewer behavioral problems. Tertiary or indicated prevention of mental illness following bullying first of all requires clinicians to ask patients and parents about bullying, intervene if necessary, and treat the anxiety, depression, and sometimes even suicidality that bullying can cause.

A limitation of the article, caused by the restricted questioning 40 years ago, is that the authors can say nothing about the long-term consequences of being a bully. Nor can they separate out the children who were both bullies and victims, who have been shown to be at highest risk of later psychiatric disorders and poorer adult functioning (5, 6). It is to be hoped that the authors will use the range of data sets at their disposal to explore the long-term consequences of being a bully rather than a victim. This is a group to whom psychiatry has paid less attention, but who raise many interesting questions. Some research suggests that being a bully (but not a bully/victim) increases the likelihood of delinquent behavior and substance abuse in the late teens, but not all longitudinal studies agree (7).

Another important theme for psychiatric research emerges from this article: the value of longitudinal studies. Most clinicians will be familiar with the Framingham Heart Study, set up in 1948 and still going today (8, 9). This study of a population cohort is “a prospective study concerned with the incidence of coronary heart disease and with the study of those factors, both host and environmental, which may contribute to its development” (8; p. 539). The physicians who started the Framingham study were looking for “the cause” of coronary heart disease; what they found was best described in a risk algorithm that combined age, sex (male), systolic blood pressure, antihypertensive treatment, smoking, diabetes, high-density lipoprotein, and body mass index (10). The algorithm became more accurate as time went on and the participants aged.

The Framingham Heart Study completely changed the medical approach to coronary heart disease, from one based on treatment to one based on prevention. Would a “Mental Health Framingham” have similar results? Unfortunately, Framingham is one of only a handful of truly longitudinal studies funded by the National Institutes of Health. The United States has to rely on other countries, with perhaps a more patient view of public health, to provide data that we can use to answer questions that need time. The British 1958 birth cohort study used by the Arseneault group was the second such commitment by the United Kingdom, and several others have followed it. Fifty years later, the birth cohort study is still going strong and producing important information about childhood predictors of adult morbidity and mortality (11). This study of the long-term effects of bullying is just one example.

The Takizawa et al. study also provides an example of the value of serendipity in public health research. Over time, researchers were able to study the health effects of many events and interventions undreamed of when birth cohort studies began, e.g., the surge in out-of-wedlock births in the past 50 years (12). It has also been possible to link a series of birth cohort studies in the United Kingdom to test for differences in causal relationships over time, e.g., changes in the risk to children associated with having a teenage mother (13). The productivity of these birth cohort studies has been prodigious and their value beyond calculation. The Takizawa et al. article offers us another major contribution, by demonstrating the frightening long-term costs of childhood bullying.