Bullying Behaviors and Associations

BULLYING AND PHYSICAL HEALTH PROBLEMS

Abstract

This review examines the extant literature on peer victimization for an association between bullying and physical health problems across three distinct populations: children & adolescents (2-21 years), early to middle adulthood (22-59 years), and older adults (> 60 years). Bullying effects were evaluated across these populations for peers, co-workers, and spouses. All studies were published between 1992 and 2008 and reported physical health problem measures alongside psychological and emotional outcomes. Findings suggested that the greatest associations between health problems and bullying showed a dose response relationship in that greater frequency and severity of victimization was associated with more physical health problems, especially for females. There are few longitudinal studies identifying clear physical health consequences linked to bullying across the lifespan. However, an association between bullying and physical health outcomes was demonstrated. Methodological differences and limitations, including the nature of peers changing across the lifespan, are discussed as well as implications for future research to aid our understanding of the entwined nature of physical and psychological health outcomes for victimized individuals.

Key words:

Bullying, social rejection, mobbing, social isolation, children, adolescents, adults, physical health, somatic, psychosomatic problems.

KEY POINTS

* Victimization and its relation to physical health outcomes is mostly studied among the adolescent population

* Victims and bully-victims report more somatic complaints than bullies or uninvolved individuals

* Few longitudinal studies exist in developing the relationship of bullying and health outcomes across populations, in part due to the nature of peers changing (e.g. from classmates to co-workers and caretakers: from romantic partners to spouses)

* Physical health problems are an important factor that need to be included when predicting long term outcomes of chronic poor social relations (e.g. bullying)

* Victimization and its physical health correlates across time is an understudied phenomenon that will be important for future health care providers to understand

* Not all recipients of poor interpersonal relations are destined to suffer from poor physical or psychological health outcomes as there are many important contributing factors

A Review of Bullying Behaviors and Associations with Physical Health Outcomes Across the Lifespan

Researchers have long argued that belonging to social groups is a fundamental human need, and that people strive to maintain complex and rewarding social relationships with others (Bowlby, 1973; Baumeister & Leary, 1995; Maslow, 1968, Taylor, 2007; MacDonald & Leary, 2005; Williams & Zadro, 2001). This need to belong is thought to drive individuals to form and maintain positive, significant, lasting relationships across the lifespan. This review examined whether disruptions of these interpersonal relationships, namely victimization, or bullying are associated with poorer health outcomes across the lifespan. This loss or disruption of interpersonal bonds is collectively termed interpersonal dysfunction and includes bullying (aggression), chronic ostracism, and social isolation. To assess bullying effects across time this review is divided into distinct life stages, namely 1) childhood and adolescence (6 - 21years), 2) young and middle adulthood (22-59 years), and 3) older adults (60+ years). As health outcomes may not be directly visible until years after interpersonal social stressor influences have presented themselves, I included research that assessed whether social dysfunction in an earlier stage affected health in the current stage (e.g. Allison, Roeger, & Reinfeld-Kirkman, 2009).

Bullying Defined

Bullying involves repeated and intentional aggressive actions of one or more peers designed to intimidate or physically harm another person who is perceived to be unable to defend him/herself (Olweus, 1993). Victimization or bullying is a universal phenomenon with many negative correlates (Olweus, 1991, 1993; Perry, Kusel & Perry, 1988; Rigby & Slee, 1993) including delinquency, depression, and suicide ideation (Forero, McLellan, Rissel, & Bauman, 1999; Rigby, 2000; van der Wal, de Wit, & Hirasing, 2003). It is a prominent form of social rejection that cuts across all age groups, race, gender, and socioeconomic status. Peer victimization, or bullying has been shown to be a stable phenomenon over extended time frames, often starting in preschool, peaking in early adolescence (Brendgen, Vitaro, Bukowski , Doyle, & Markiewicz, 2001; Nansel, Haynie, & Somnons-Morton, 2003), and can remain stable throughout high school (Williams & Guerra, 2007). Peer victimization or bullying not only refers to children but includes peers who are of relatively equal status and involved in somewhat open relationships such as classmates, co-workers, or spouses. As the nature of these relationships changes across the lifespan so do the terms for victimization. What was once considered teasing in kindergarten grows into bullying during adolescence, harassment in adulthood (bullying or mobbing in the workplace), and social isolation or elder abuse (aggression) for older individuals. Therefore, I have included phenomenon such as school bullying, workplace mobbing (bullying), and peer-to- peer elder abuse (aggression) and social isolation under the criteria of dysfunctional interpersonal relationships.

What We Know About Bullying

There is ample empirical support for different types of aggression involved in this dysfunctional social behavior including direct physical and verbal aggression (e.g., pushing, hitting, kicking, name calling, or teasing), and indirect and relational aggression, (e.g., excluding from a group on purpose, rumor spreading, and ignoring (Crick et al., 2002; Duncan, 1999; Egan & Perry, 1998; McDougall, Hymel, Vaillancourt, & Mercer, 2001). For many individuals these experiences can have damaging emotional and psychological consequences (Olweus, 1991, 1993, Perry, Kusel, & Perry, 1988; Rigby & Slee, 1993). These include emotional and behavioral problems and risky behaviors, such as substance abuse and alcoholism (Archer & Coyne, 2005; Crick & Grotpeter, 1995; Dodge, 1989; Egan & Perry, 1998; Kupersmidt, Coie, & Dodge, 1990). Although prevalence rates vary across countries studies suggest that 35% of youth are exposed to some degrees of bullying involvement (World Health Organization, 2004; Nansel, Overpeck, Pilla, Ruan, Simons-Morton, & Scheidt, 2001; Olweus, 1994). However, being a persistent target of aggression is problematic. Previous evidence from Australia, England and the United States suggested 10 % of children were victimized on a regular basis (Slee, 1992) and more recent, approximately one third of American children report being repeatedly victimized by peers (Grills & Ollendick, 2002; Limber & Small, 2003). As a result of an international interest in victimization outcomes negative psychological risks and effects of chronic peer victimization are well documented. However, very little research has been directed toward the effects of peer aggression (bullying) on physical health outcomes for children or adults.

Adult bullying is a complex phenomenon and often includes, for example, physical abuse and sexual harassment which are not included in this paper. Conversely, workplace bullying is comparable to peer victimization in school, where an employee is persistently picked on or humiliated by fellow employees or supervisors (also known as mobbing), and has been recorded in several work domains (Einarsen & Raknes, 1995; Einarsen & Skogstad, 1996; Matthiesen et al., 1989). In the US workplace, bullying reports have ranged from 38% of employees experiencing some type of bullying in a given year and 42% witnessing the bullying of others (Quinn, 1999). Likewise, Bjorkqvisst, Osterman, and Hjelt-Back (1994) found that 30% of men and 55% of women reported being the recipient of some form of workplace bullying. Although there is increasing evidence for workplace bullying and the negative health correlates (e.g. Kivimaki, 2004) there is a lack of information regarding our elderly population in relation to victimization. There are a growing number of senior citizens who are very socially active, but little seems to be known about their relationships with regards to aggressive interpersonal behaviors (other than abuse research, which is outside the scope of this paper). However, indirect aggression has been found to be prevalent in older adults in retirement communities, suggesting that even older adults are not immune from being either targets or perpetrators of peer victimization (Walker, Richardson, & Green, 2000). In order to pursue these bullying effects over time this review is a first attempt to examine research on bullying and its associations with physical health consequences across different age groups throughout the lifespan.

Methods

Search Strategy and Inclusion Criteria

Two research assistants and I independently reviewed articles and came to a consensus on those that would be included in this review based on our criteria. PsycINFO, MEDLINE, and Google Scholar were searched for the key terms: bullying, mobbing, victimization, peer neglect, ostracism, rejection, social exclusion, divorce, elder abuse, peer-to-peer aggression, somatic, psychosomatic, and physical health problems. This paper focused only on interpersonal dysfunction specifically involving bullying. Searches were limited by the English language and peer reviewed work. Only studies that examined chronic or continual dysfunction were included. Papers that examined laboratory manipulations of acute social problems (e.g., laboratory manipulated ostracism or social rejection) were not included. Furthermore, articles were only selected if they specifically examined or included physical health problems as outcomes of bullying involvement. That is, mental health problems (e.g., emotional problems) and cognitive impairment that were the result of interpersonal dysfunction were excluded from this paper. I excluded the extensive body of research on abuse and maltreatment for children and adults due to gross imbalance of power between the abuser and abused. I also excluded elder abuse research that focused on nursing homes or abuse of the cognitively infirm (e.g., Alzheimer's or dementia patients), as again, this encompassed a large power imbalance among victims and perpetrators. This review concentrated on peer relationships that have a comparable power base such as classmates, co-workers, and spouses, where the individuals themselves are of more equal status, even though one might argue there is always a power imbalance involved in bullying behaviors (e.g., Olweus 1993). Health was assessed via common physical complaints (e.g., colds/flu, headaches, stomach aches, sore throats) and psychosomatic symptoms, (e.g., bed wetting, sleep disorders, eating problems). All studies included some somatic symptoms and psychosomatic symptoms. However, the same symptoms were not systematically reported in all studies. Results were separated by sample population (i.e., children, adolescents, and adults). As there were variations among articles in terms used for victimization such as bullying, peer rejection, ostracism, and direct vs. indirect aggression, I used interpersonal dysfunction as an all-inclusive term for relational and direct aggression (bullying) leading to physical health problems across the lifespan.

Findings

A total of 33 articles met the inclusion criteria. The general characteristics of the studies are represented in Tables 1, 2, and 3, and include sample size, participant details (if known), health measures and outcomes related to health. As health problem definitions changed with specific studies I have included research that categorized health with terms such as physical health, somatic health problems, or psychosomatic health problems.

Bullying and Physical Health Problems in Childhood & Adolescence

Children under 12 years old

Bullying and victimization has been widely investigated internationally (Due et al., 2005; Glew, Rivara, & Feudtner, 2000; Smith et al., 1999), but research for pre-school to mid-elementary bullying is much less developed than research among older children (Hanish, Kochenderfer-Ladd, Fabes , Martin, & Denning, 2004; Kochenderfer & Ladd, 1996; Monks, Ortega Ruiz, & Torrado Val, 2002). As a result I found no documented evidence linking bullying to physical health problems in younger children (under 6years old). However, bullying in kindergarten is well documented as a serious problem (Crick, Casas, & Ku, 1999) and may well be a pre-cursor to a continuous victimization cycle and later poor physical health outcomes. Only within the last 15 years have studies started to examine the relationship between peer victimization, namely bullying, and physical health problems for children. This review identified eight (8) studies within the criteria for children between the ages of 6 -12 years (See Table 1). In general all of the included studies identified a positive relationship with frequency and severity of bullying leading to more frequent and more severe health symptoms or a dose response relationship. Within this age group (under 12 years) three studies used clinical interviews or a combination of interview and surveys to collect data.

One of the frequently cited studies is Williams, Chambers, Logan, and Robinson (1996). Semi-structured standard health interviews were conducted by school nurses in an inner London school district for the 1992-93 academic year. They established that British children who reported being harassed by peers “sometimes” or “frequently” were nearly two and half times more likely than their non-victimized peers to report somatic complaints such as headaches or stomach aches to a school nurse. There was also a significant linear trend of increasing risk of symptoms with increased frequency of bullying (dose-response relationship) shown for all reported health and psychosomatic symptoms: not sleeping well ({chi}2 = 61.8; p<0.0001), bed wetting ({chi}2 = 13.5; p = 0.0002), feeling sad ({chi}2 = 26.2; p<0.0001), headaches ({chi}2 = 38.0; p<0.0001), and tummy aches ({chi}2 = 38.6; p<0.0001).

In 2001, Wolke, Woods, Bloomfield and Karstadt, implemented individual structured interviews with children between the ages of 6 and 9 years, and parents completed health questionnaires for the child for the previous six months. This study differentiated between bullies, victims, and bully-victims (those individuals who are ineffectual aggressors act as both victims and perpetrators of bullying) and examined the relationship between direct and relational bullying with common health problems. Wolke et al. (2001) reported that direct victims and bully/victims are at the most risk for developing subsequent health problems related to ongoing victimization. Victims (especially girls) reported more psychosomatic health symptoms than non affected children (M =2.34, [95% CI =2.22-2.46]; neutral: M =2.10. [95% CI =1.99-2.21]). Interestingly, Wolke and colleagues found no relationship between relational bullying and health outcomes. This may have been due in part to the age range of the children. Björkqvist (1994) argues that choice of methods for expressing aggression is linked to maturational changes throughout the lifespan. Until children develop more mature verbal skills and cognitive strategies they use more direct methods of aggression (e.g., pushing, hitting) that are more tangible to them during this preadolescent phase. As with other forms of victimization (e.g., race and weight discrimination), younger children are often more concrete in their thoughts and actions (Black-Gutman, & Hickson, 1996; Rex-Lear, 2007). Moreover, a certain degree of impulsivity exists in physical aggression whereas indirect bullying requires some strategic planning that usually improves with age, verbal skills, and cognitive ability.

The third interview based study assessed bullying prevalence among young children in India Kshirsagar, Agarwal, and Bavdekar (2007) using semi-structured interviews of children and parents as well as similar health items in Williams et al. (1996), found the same general pattern of results. They reported that everyone who claimed to be bullied reported at least one health symptom compared to those not bullied. Kshirsagar et al. (2007) also reported indirect relational bullying was the most common form of aggression related to health outcomes in contrast to Wolke et al. (2001).

The next 4 studies in this age group employed self report measures of health outcomes and found strong connections between bully-victims and physical health problems. In 2004, Fekkes, Pijpers, and Verloove-Vanhorick studied bullying behavior associated with a wide variety of health complaints and depression in a large Dutch study. They concluded bullying is indeed associated with many somatic complaints and bully-victims are at even greater risk of developing depression and somatic problems than bullied children alone. Gini (2008) reported a significant relationship for victims and bully-victims to poorer health outcomes and overall adjustment for elementary school aged children. Previously, Houbre, Tarquinio, and Thuiller (2006) conducted three studies with 291 French students between the ages of 9 years 12 years (M = 10.14). As this review is only focused on physical health and bullying relationships I only refer to Study 2 of Houbre et al's. (2006) work. In two phases they first assessed how a child's bullying status (e.g. bully, victim, bully/victim, and non bullied, control) affects the manifestation of psychosomatic symptoms, and secondly, the potential link between behavioral problems and the onset of psychosomatic symptoms. Their psychosomatic scale incorporated a comprehensive cross section of 44 items including cognitive difficulties and multiple psychosomatic components as follows:- neuro-vegetative disorders [(1) dizziness, vision problems, tingling sensations, (2) heart palpitations, trouble breathing], sleep disorders, digestive disorders, somatic pain, eating disorders, skin conditions, vegetative symptoms and dysuria , diarrhea and constipation. For younger children this is one of the more comprehensive scales across the studies in this review. Bullies had a higher psychosomatic symptom level than controls, particularly symptoms of digestive and neurovegetative disorders, which corroborates earlier findings (Natvig et al., 2001; Kaltiala-Heino, Rimpela, Rantanen, & Rimpela, 2000). Furthermore, consistent with the findings of other studies (e.g., Rigby, 1998,1999; Williams et al., 1996), victims experienced multiple health problems, in particular cognitive difficulties, neuro-vegetative disorders, somatic pain, and skin conditions. Houbre and colleagues demonstrated, that bully/victims reported more psychosomatic problems than all other groups (e.g. victims, bullies); most affected by neuro-vegetative disorders, digestive problems, somatic pain, and skin conditions. In addition, there was a positive link between behavioral problems and the onset of psychosomatic disorders.

In contrast, Kumpulainen and colleagues (1998) identified victims, particularly boys, displayed the most psychosomatic symptoms. This may have been due, in part, to the age of the sample, (i.e., Kumpulainen's mean participant age was 8 versus 10 years) , or the scale used. Kumpulainen only used two items to assess psychosomatic health symptoms versus Houbre et al's, 44 item scale. Nonetheless, these findings combined underline the need to recognize bullying problems and their relation to both psychological and somatic health problems in younger children much quicker than is currently the case.

There is clearly an issue of identifying causation in concurrent studies. Some research suggests peer rejection and victimization can lead to ill health (Boivin, Hymel, & Bukowski, 1995), while others posit that ill health, especially psychosocial disorders and anxiety may leave children vulnerable to victimization later on (Hodges & Perry, 1997; Schwartz, Dodge, & Coie, 1993). In order to better specify causation, larger studies such as the one conducted by Fekkes, Pijpers, Fredriks, Vogels, and Verloove-Vanhorick (2006) are needed. In Fekkes etal. (2006) they followed a cohort of children over 6 months with health and bullying measurements taken at the beginning and end of the school year. This study was unique in that it assessed whether health symptoms both precede and/ or follow school interpersonal dysfunction. It is worth noting that this study excluded children at the beginning of the study who reported either health symptoms or problems with victimization. This decreased the true numbers of children involved in bullying and/or health symptoms. However, even after implementing these criteria Fekkes et al. (2006) found children who were bullied at the beginning of the year had a significantly higher risk of developing new health symptoms during the course of the school year. For example, odds ratios were particularly high for depression (4.18), anxiety (3.01), bedwetting (4.71), abdominal pain (2.37), and feeling tense (3.04). Being bullied had a strong relation to the development of abdominal pain but only for girls (odds ratio: 4.98; [CI: 2.17-11.43]; p < .001). Psychosocial problems at the beginning of the school accounted for risk of victimization later on, but physical health symptoms such as headaches and abdominal pain did not. In sum, their results did not support the hypothesis that ill health may precede subsequent victimization. Conversely, their work supports the conclusions of other concurrent studies that bullying influences health outcomes. However, more longitudinal studies would be prudent before we draw strong conclusions on the directionality of poor health and victimization.

Overall these studies of elementary school aged children provide evidence that bullying is alive and well in younger children and affects more than their psychological well-being.. Those involved either as victims, bully-victims, or bullies are indeed more susceptible to physical health problems than uninvolved children. This would suggest that children already exposed to dysfunctional interpersonal relations before they reach adolescence may well be at a distinct health disadvantage during adolescence and possibly their adult years. However, there are several limitations to this body of work on the association between bullying and physical health outcomes that are worth mentioning.

Limitations

First, although some studies used some form of (semi) structured interview (probably due to reading comprehension of this age range), the criteria used for each is unclear and difficult to compare. Williams et al. (1996) reported a lower prevalence of bullying compared to previous British studies. This may be due to the unstructured verbal interview technique and the provision of an explanation of bullying. Providing definitions (explanations) of bullying has been linked to lower reporting rates of bullying, sometimes as much as a 30-10% decline (e.g., Baldry, 2004; Jensen-Campbell, & Rex-Lear (unreported); Vaillancourt et al., 2008). Self report peer victimization is often preferred as subjective perceptions are considered more relevant than peer or teacher perceptions when examining psychological and physical well-being (Juvonen, Nishina, & Graham, 2001; Panak & Garber, 1999). Conversely, data based solely on self reports creates a risk that children may over estimate symptoms and report more problems. This common method variance might lead to overestimates of associations between bullying and health problems, and inflated effect sizes. However, self report data is considered valid and reliable for collecting sensitive victimization information (Ladd, Kochendorfer & Ladd, 2002) but not sufficient to develop a complete picture. Multiple independent raters are required, such as, teachers, peers, parents to help prevent possible inflated correlations between health and bullying. Multiple raters and methods such as recent study Knack, Jensen-Campbell and Baum (2009) included physiological, health, and behavioral measures. These types of comprehensive methodological studies may provide answers as to why the link between victimization and negative health outcomes exists. In turn this will create a much stronger case for parents, teachers, and policy makers to take notice of dysfunctional interpersonal relationships that may be damaging to an individual's health in the short and longer term.

Second, time frames for assessments across or within studies varied dramatically from some health measures pertaining to the previous two weeks, six months, or academic school year (e.g. Houbre, 2006). In one study respondents completed an Affective Pain Index for the previous two weeks while their use of medical services was measured from retrospective self reports for the previous month. This could be problematic as retrospective health reporting can become mismatched over time and may not explicitly be linked with bullying, but other coexisting social stressors, such as fear of being at school, or other behavioral issues. Indeed, Reis and Wheeler (1991) suggest that without interval controls, retrospective self reporting is prone to substantial distortion from: selection of representative events, - it is difficult for most people to store and retrieve detailed information about repetitive or mundane events in long term memory; recall of the characteristics of those events are subject to distortions of (e.g., recollection biases), or motivated defense mechanisms (e.g., selective perception, dissonance reduction – which may be particularly salient in victimization linked memories), and aggregation across multiple events. It is not well known how people combine data from multiple incidences to arrive at a single global impression, which can be skewed by atypical events (e.g. bullying). In sum, it would be wise to collect health data over a shorter, specific time frame, permitting researchers to rule out alternative explanations, thereby pinpointing the correlates of victimization and physical health problems with greater precision. Thirdly, health assessments for this type of research need refining. Many studies limited their inclusion criteria for somatic health symptoms and psychosomatic health symptoms by only including headaches and stomachaches as dependant variables, whereas others included a much wider range of symptoms. Varying criteria across studies made direct comparison difficult. Whether these limited criteria are enough to suggest victimized children are at increased health risk later on is questionable.

Fourthly, many studies did not separate types of bullying for example, relational and overt aggression (e.g., Williams et al., 1996) which can impact studying male and female bullying tactics and outcomes. It has been reported that girls' verbal skills develop earlier than boys providing them a head start for employing relational, indirect aggression tactics much earlier than boys (Macoby & Jacklin, 1974). Furthermore, there is some evidence that boys and girls cope differently under stress thus different types of bullying may have differential relations to social adjustment and personal health outcomes (Crick & Bigbee 1998; Vaillancourt et al., 2008). Finally, psychosocial factors (e.g., ethnicity, family environment, social support, number and quality of friendships) were not systematically accounted for across studies. To date, no particular variable(s) have been consistently pinpointed for why some people are involved in bullying and not others, including heritable traits, such as personality and negative affectivity; or environmental variables, such as abusive home environments. To understand the mechanisms of why there is significant association between victimization and health outcomes future research will be wise to recognize particular characteristics, and how much variance is accounted for factors affecting adaptive versus maladaptive functioning for those involved in poor interpersonal relationships. Nonetheless, the bullying evidence suggests that physical health problems are prevalent in children and the effects are widespread across age, race, and gender, in different countries around the world.

Summary

As multiple physiological and psychological pathways may lead to poor physical health outcomes the cross sectional nature of the majority of these studies limits inferring causality and directionality of effects. So the picture is far from clear cut as to how victimization impacts individuals' physical health. However, these studies support a definitive pattern of physical health problems that co-occur when bullying behaviors are present and persistent. To this end, parents, teachers and health care professionals need to consider bullying when confronted with unexplained psychosomatic and somatic complaints in younger children. Bullying cannot be ignored as a contributing factor to detrimental health outcomes and as young children's bodies and brains and immune functions are developing rapidly they need to be protected before they reach adolescence.

Adolescents between 12 and 21 years old

Until recently there has been little focus on direct physical health symptoms associated with poor interpersonal relationships during adolescence, this next section highlights some of the pertinent findings so far. I identified sixteen (16) studies measuring the relationship between victimization and health problems in young people from the age of 11 to 21 years. Of these three (3) examined the predictive link between peer victimization and health from a longitudinal perspective.

Longitudinal Studies

In a sample of eighth- and ninth-grade Australian students Rigby (1999) demonstrated that severe victimization was associated with poor physical health. Self-reports of peer victimization predicted physical complaints 3 years later, even after controlling for initial levels of physical problems. In particular, victims were found to suffer more from sleep disorders, bedwetting, headaches, stomachaches, and feeling unhappy. The symptoms also varied according to the victimized child's gender. Boys tended to have more headaches and backaches, and to be more irritable than girls, who were more nervous and had more sleep disorders. Moreover, the greater the exposure to bullying showed a dose-response effect in the number of symptoms reported. The number of symptoms appears to be additionally dependent on the distress level, and also on the social support provided by the teacher for girls, and by peers for boys. The less support received, the greater the symptoms.

Multiple facets of global well being and general function have been assessed in relation with victimization and in every domain we find negative outcomes. For example, Nishina, Juvonen and Witkow (2005) examined associations between psychosocial, somatic and scholastic consequences of peer harassment among of sixth graders over a 1 year period. Perceived peer victimization predicted greater psychosocial and physical problems which in turn predicted reduced academic functioning. However, in agreement with previous work (Fekkes et al. 2006) existing physical conditions did not predict later victimization. Wilkins-Shurmer et al. (2003) assessed global well being and physical functioning among bullied adolescents (mean age 13.6 years) and determined physical functioning scores (i.e. lower physical fitness) were associated with experiencing higher frequency of victimization, especially for girls over a short time frame (5 days). However, the long term relationship between bullying and physical fitness was not significant for boys or girls.

Sex Differences Associated with Bullying Outcomes

Indeed, Rigby (1998) noted a strong connection between being bullied and headaches. Self report questionnaires administered toward the end of the school year, assessed both mental and physical health problems associated with adolescent peer rejection for victims, bullies, and bully-victims in 819 secondary school children. Health measures assessed both mental and physical health problems including a 21 item list of physical complaints such as: headaches, stomachaches, sore throats, coughing, wheezing, ‘thumping' in chest, and sleeping problems. Rigby reported that frequency of bullying was linked to an increase in psychological and physical health complaints – and gender makes a difference. Girl victims reported more colds, headaches and sore throats, while boy victims reported more physical injuries from accidents. The health implications are different for boys and girls, such that boy bullies have poorer mental health outcomes than others, and girl bullies have better mental and physical health than girl victims. However, the study did not support the view that bully-victims experience greater physical or mental health problems than others.

It is apparent then that the consequences of bullying have different outcomes for boys and girls. To look at this in more detail Gruber & Fineran (2007) examined the impact of bullying and sexual harassment on middle and high school girls. These two types of negative behaviors are usually studied separately. Previously Pellegrini (2001) and Pepler et al. (2002) analyzed these as a combined phenomenon, but without physical health outcomes. They focused on the perpetrators of these negative behaviors and found that bullies also tend to be engaged in sexual harassment. As bullies' interest in the opposite sex increases so does their proclivity to become sexual harassers. Gruber and Fineran (2007) measured somatic complaints reported by girls in both middle and high school in relation to this phenomenon. In contrast to the bullying literature that shows bullying decreases with age (Nansel et al., 2001), (as previously discussed in this review), this particular study revealed high school girls experience bullying and sexual harassment more frequently than their middle school peers. For example, ridiculing increased from 51.5% in middle school to 58.2 % in high school (e.g., upset for the fun of it, 42% to 53%; spreading sexual rumors, 33% to 53%). However, middle school girls were more adversely affected with more physical health problems by the bullying/harassment and the impact was greater. A lesser impact on older girls' physical health outcomes suggests that they have developed strategies such as, more complex coping mechanisms and social support networks to insulate themselves against potentially adverse health effects of unwanted behavior. Overall, high school girls demonstrated significantly poorer physical health outcomes (higher scores) in high school than their middle school counterparts, physical health (M = 6.23 vs. 7.32). For girls, the pressure to conform to peer groups, compete in attractiveness and weight, and dating, may contribute to excess stress and thus negative health outcomes. As adolescence brings a host of biological, psychological and social challenges that can negatively affect their health, individual's coping strategies and social support systems will come into play. Also, the use of alcohol or drugs by bullied or harassed girls suggests that they use these substances as a coping mechanism for upsetting events in the same ways that adults do (Gruber & Fineran, 2007). Unfortunately, this study did not assess boys simultaneously as an integral component of sexual harassment. They often play a significant role as perpetrators, especially in high school, and we might have seen some interesting health correlates.

In 2008, Gruber and Fineran conducted a follow up study comparing the same bullying and sexual harassment effects for adolescent boys and girls. Bullying and sexual harassment had particularly large effects on physical health (b=.296 and b=.335, respectively). In sum, though bullying and sexual harassment were pervasive among boys and girls the adverse effects were more noticeable for girls. Recent work by Vaillancourt, deCatanzaro, Duku, and Muir (2009) found sex differences in the way stressful situations are interpreted and processed. Vaillancourt and colleagues posit that this may be due to different coping strategies employed by males and females under bullying stress. Girls are more concerned with their peer evaluations than boys are generally, and will tend to internalize abuse, whereas boys will tend to externalize their abuse by becoming more aggressive themselves. This new approach may provide one possible explanation as to why victimization and health outcomes are stronger for girls.

Victims versus Perpetrators

Forero, McLellan, Rissel, and Bauman (1999) conducted a cross-sectional study analysis with Australian primary school age children, (mean age =11.88 years), and secondary school age children in academic year 8 (mean age =13.96 years) and academic year 10 (mean age=15.97). Interestingly, they found that bullies and bully-victims were more likely to report moderate (1.6; 1.63) and frequent (1.8; 1.97) psychosomatic and psychological symptoms whereas ratios for victims were not significant. These results are consistent with some studies that suggest bullies are unhappy and suffer from low self esteem, or lacking in friends and antisocial behavior (Rigby & Slee, 1993; Rigby, 1998; Zubrick et al., 1997). However, there is also contradictory evidence suggesting that bullies are not necessarily unhappy, or unhealthy (e.g. Perren & Alsaker, 2006; Wolke et al., 2001).

Baldry (2004) did not find that bullies experienced physical health problems. She examined the extent to which victimization at school over a period of 12 months affected the mental and physical health of Italian children. The somatic health measures were extracted from the Italian version of the CBCL for 11–18–year-old youths (Child Behavioral Check List, Youth Self-report for ages 11–18, Achenbach and Edelbrock, 1983), frequently used in the assessment of psychological welfare. Baldry used hierarchical regression models to assess that bullying has a significant impact on children's health, after controlling for socio-demographic variables including age, gender, socio-economic status, and relationship with parents. Somatic complaints such as aches, pains, feelings of tiredness and sickness, were significantly predicted by both direct and indirect victimization, whereas bullying others was not associated with physical health. Being a victim of indirect bullying at school was the strongest predictor for somatic complaints - the full model accounted for 14.7% of the variance. Later, Baldry and Winkel (2004) examined the same sample of Italian children for the independent effects of indirect and direct bullying at home and at school on mental and physical health outcomes. They again found the largest predictor for health problems was still indirect victimization at school.

Finally, in 2006, Srabstein, McCarter, Shao and Huang analyzed a large cross section of US data gathered from the World Health Organization Health Behavior In School-aged Children survey. They included over 15,000 students from 6th -8grades and included multiple commonly reported physical ailments and the associations with bullying behavior. In contrast to previous studies, they concluded that frequent involvement in bullying behaviors in any capacity was associated with poor health status. This suggests that under particular circumstances, still not fully understood, all this involved in poor interpersonal relations are at risk for poorer health.

Moderating and Mediating Effects on Bullying/Health Associations

Aligned with both earlier and more recent studies Natvig, Albrektsen and Qvarnstrøm (2001), noted that 71% of bully-victims (boys) and 50% girl bully-victims reported the most psychosomatic symptoms consistent with that obtained by Houbre et al. (2006), and Forero et al. (1999) with their sample of younger students. Natvig et al. (2001) found strong consistent associations with different symptoms. This dose-response relationship suggests a causal relationship between increasing bullying behavior and its association with a significant increase in number of psychosomatic symptoms. However, Natvig et al. (2001) adjusted their risk assessments for social, personal and other stress related factors. In so doing, their risk assessments were weakened, even though they remained significant. This weakening effect might suggest that as in previous studies, social relationships, personal resources, friendship structures and coping mechanisms may buffer against the negative health consequences of victimization (Knack et al., 2009, Malcolm, Jensen-Campbell, Rex-Lear 2006; Natvig et al., 1999).

By examining the predictive link between peer rejection and physical health moderated by reactivity/negative emotionality Brendgen and Vitaro (2008) have added a further dimension to this body of research. They took a diathesis-stress approach suggesting that peer rejection may contribute to negative health outcomes for those individuals already experiencing some preexisting vulnerability such as, reactivity/ negative emotionality. Reactivity/negative emotionality is associated with the temperamental quality of neuroticism (Rothbart, 2000). Individuals high in neuroticism often experience more negative affectivity (i.e. fear, frustration, sadness, and low soothability) when exposed to stress (Pennebaker 1993), and report more pain than others in response to the same standardized stimuli, (Bollmer Harris, Milch, 2006).

To explore this idea Brendgen and Vitaro (2008) assessed students over a two year period from grades 7-8 (mean age =13.10 in grade 7). They included measures of peer rejection during 7th grade (T1) via peer nominations and depression at TI, and then reassessed self reports of health, peer rejection and reactivity in 8th grade (T2); family adversity measures were completed by the children's' parent(s), and reactivity/negative emotionality measures from teachers. Brendgen and colleague used hierarchical linear regression to predict the effects of peer rejection on health outcomes at T2 with the moderating effects of reactivity/negative emotionality. The study found that peer rejection early on only influenced health symptoms later if reactivity/negative emotionality was significant at T1 (В=.27, CI .02-.51, p=<.05), but only for girls. This predicative link between peer rejection and somatic problems provides evidence that supports the diathesis-stress model of disease, which suggests that an environmental stressor such as peer rejection or victimization can lead to detriments in health, but predominantly in those individuals showing preexisting vulnerabilities. This goes some way to explain why not all victimized children do not suffer poor physical health outcomes.

While examining whether being bullied predicts health outcomes Knack, Iyer, and Jensen-Campbell (2010) controlled for personality differences among college age adolescents. Indeed, for adults Einarsen and Skogastad (1996) suggest that personality (e.g., conscientiousness) may actually be a major contributing factor to workplace bullying (and not vice versa). This mirrors Jensen-Campbell and Malcolm's (2007) finding that conscientiousness is a major predictor of being bullied in adolescence. Furthermore, neurotic individuals are expected to have more negative interpersonal relationships overall (Jensen-Campbell & Malcolm, 2007). In line with this thinking Knack et al. (2010) in a short term longitudinal approach found that victimization did predict increases in frequency and severity of reported health problems over time after controlling for individual differences. They also determined, that previous health problems do not predict changes in victimization over time (e.g., Fekkes et al., 2006). Given that the link is not associated with personality, Knack et al. suggest that repeated experiences of peer victimization act as a chronic stressor and can alter physiological functioning such as the HPA axis and neuroendocrine functioning (see Knack, Gomez, & Jensen-Campbell, in press; Knack et al. in press).

Along with commonly reported ailments children involved in frequent bullying behaviors as bullies and/or victims have also been associated with greater risk of injuries, obesity, risk taking behaviors, and suicide ideation (Fekkes et al, 2004; Janssen, Craig, Boyce & Pickett, 2004). Also, involvement with school bullying is associated with reduced school performance, lower feelings of belongingness to the school environment, and fear of being at school (Glew, Fan, Katon, Rivera & Kernic, 2005). An even greater concern is that suicide ideation has been significantly related to peer victimization even after controlling for levels of depression and social support (Kaltialo Heino., 1999, Rigby & Slee, 1999; Van der wal et al., 2003). Bond et al. (2001) found that being victimized has a significant negative impact on the future emotional well-being of young adolescent girls, independent of their social relations, but the same does not apply for boys. As with other forms of abuse, girls tend to react internally, and therefore feel more depressed and sad, whereas boys tend to overtly react to problems encountered at school (Duncan, 1999).

Few studies control for existing levels of mental health or physiological functioning before assessing bullying effects on physical health outcomes. However, Rigby (1999) and CONTROLSGreco, Freeman, and Dufton (2007) found that even after controlling for initial health differences, bullied adolescents are still at risk for poorer health outcomes. Greco et al. (2007) created a no-pain control group for their study with adolescents and examined the link between abdominal pain, victimization, and social functioning. Abdominal pain is one of the most common chronic pain complaints reported in childhood and interferes with daily functioning as well as increased school and social problems such as absenteeism, academic difficulties and perceived lower social competencies (Perquin et al., 2003; Scharff, 1997). As this often exists in the absence of medical diagnosis it has been suggested that stress, peer victimization, in this instance, may be partially related to this condition (Greco et al, 2007). Greco and colleagues found that even after controlling for abdominal pain reports, children in the pain group (especially boys) experienced higher levels of both overt and relational bullying. In turn, these adolescents reported a greater occurrence of health problems including headaches, abdominal pain, sore throats, colds, and mouth sores compared to non-bullied adolescents.

An interesting cross sectional study undertaken by Schnohr and Niclasen (2006) examined bullying trends from 1994, 1998, and 2002 among Greenlandic adolescents who were 11, 13 and 15 years of age respectively. Using logistic regression they simultaneously examined the strengths of relationships between bullying behaviors and physical health symptoms. They also included psychological well-being and risky behaviors, such as drinking and smoking. They determined that overall there has been an increase in bullying prevalence and significant changes between the types of bullying among Greenlandic children since 1994. The proportions of children who experienced bullying behavior in some capacity significantly increased (p>.001). For victims the proportion increased from 6.7% of the total population in 1994, to 11.4 % in 2002, for bullies the proportion increased from 5.3% in 1994, to 6.6% in 2002 and bully-victims increased from 4.6% to 7.8%. Only bullies reported an increase in poor health behaviors, such as, smoking tobacco and drinking alcohol (even after controlling for age and gender). Furthermore, only bullies significantly reported an association with stomach aches and sleeping difficulties. However, Schnohr and Niclasen suggest that headaches, stomach aches and sleeping difficulties are general characteristics of children engaged in any aspect of bullying behavior, whether as victims, bullies, or bully-victims. An important contribution of this study is the emphasis that if bullying behaviors are increasing (at least in Greenland) then we are likely to see an increase in negative health outcomes in later years, if not directly, then from risky health behaviors such as smoking and drinking alcohol.

The College Years

Despite the developing body of research on the harmful effects of bullying during school years there is little research into bullying at the college level. Although retrospective studies have assessed long term psychological outcomes of childhood bullying with college students (Duncan, 1999) few have actually looked at the prevalence of bullying during college. Contrary to the idea put forward by some that bullying declines and disappears by age 16 (Eslea & Smith, 1994, cited by Eslea & Rees, 2001) it likely becomes more sophisticated in nature and may not be recognized on less complex bullying scales. Recent research with college age students suggests that peer bullying is still occurring during university years. One study suggested that bullying does graduate from school to college with 6% reporting having been bullied by peers and 5% bullied by teachers (Chapell et al., 2004). These data do not fit with previous research suggesting that bullying declines with age (Olweus 1984). Conversely it is consistent with the data positing that bullying is alive and well beyond the school yard and fairly common in many workplace environments (Hallberg & Strandmark, 2006; Hansen et al., 2004; Mikkelsen & Einarsen, 2001). Recent work by Knack, Gomez, and Jensen-Campbell (in press) with college age students considers the pathways of chronic social stressors (e.g. victimization) influencing later sensitivity to actual physical pain. In particular, peer victimization is positively associated with physical pain symptoms such as stomach aches, headaches and possible cardiac problems. Moreover, in a large international comparative cross-sectional study Due et al. (2005) examined frequency of twelve health related (8 psychosomatic) symptoms associated with bullying among school age children from 11-15 years of age. Across 28 countries in Europe and North America they found the same dose-response pattern of risk increased symptoms increased with increasing exposure to bullying. This study adjusted for age, affluence, and country on an individual level. Even though proportions of bullying prevalence varied considerably across countries, the prevalence of health symptoms consistently increased with frequency of bullying regardless of between country differences. Due and colleagues also suggest that victimized adolescents are more likely to use medication for physical pain than are non-victimized adolescents (Due et al., 2007).

Limitations

As with the child studies many of the same limitations apply to adolescent studies of poor interpersonal relations and associations with poorer health outcomes. The meditational role of physical symptoms is a new contribution to the proximal and distal consequences of poor peer relations and still not clearly defined and operationalized. Often health terms are broadly used thus, reported health deficits may be indicative of other underlying key psychological or physical health problem. For example, Rigby (1998) included anorexia and bulimia as physical health measures. These two related disorders are normally categorized as psychological with physical health outcomes, thus blurring the lines of what might be categorized as physical health. Other examples include impaired immune functioning. – Kumpulainen et al's. (1998) study found that absenteeism from a school seemed to be related to bully-victims, victims, and to depression in girls. Conversely, Knack, Jensen-Campbell, and Baum (2009) examined absenteeism in context with physical illness and bullying and suggest that absentees may actually be physically sick from their victimization experiences and not just avoiding school. Other research has found that chronic stressful events increase stress hormones (e.g., cortisol) which in turn suppress immune system functioning (e.g., Kemeny, 2003; McEwen, 2000). Therefore, experiencing chronic victimization may well have other underlying physiological immune impairment and these individuals may be more likely to get colds and other illnesses that prevent them from going to school. Furthermore, clustered absences might be more indicative of prolonged illness, whereas staggered absences may be more indicative of frequent minor illness (e.g., colds) or parent-approved school avoidance.

Also, it is apparent that many research questions do not control for previous levels of bullying in childhood. This puts us at a disadvantage for understanding the continuous link between environmental, heritable, and physiological components, and under what conditions they might combine to influence physical health outcomes over time. An example of this are self reports of health provide no way to validate current health reports with previous health records or alternative sources. Furthermore, differences in significant dose- response relationships do not always remain significant after controlling for various social and personal factors. These differences may lie in modeling of data between studies. This again, makes drawing conclusions difficult as to whom or why certain individuals experience greater health problems than others after experiencing victimization in some capacity. Additionally, new avenues of bullying research such as that of Knack et al. (2009) provide important positive steps towards understanding why the victimization-poor health link is so robust. They also provide preliminary evidence suggesting a link between bullying, cortisol variability and negative health outcomes.

Consequently, experiencing victimization may be particularly problematic during the teen years (Kumpulainen & Rasanen, 2000; Rigby, 1998). If peer victimization is considered a social stressor, it is a logical suggestion that bullying may have negative implications on physical health outcomes. Various studies support this hypothesis, as research reveals the association between bullying and risky health behaviors including substance abuse, alcohol abuse, violence, and suicidal ideation (Kaltiana-Heino et al., 2000; Nansel et al., 2001., Olweus, 1994). These factors may well contribute to increased morbidity and mortality in youth.

Summary

Adolescence is a major developmental milestone marked by physiological and behavioral changes that are equaled only in the development of the fetus and during infancy (Insel & Roth, 2008). This period has also been deemed as a critical juncture in development and consolidation of behaviors that may affect health later on (La Greca, Prinstein, & Fetter, 2001; Shribman, 2007; Williams, Holmbeck, & Neff Greenley 2002). Therefore, experiencing difficulties with peer relations, namely bullying, may be especially problematic during teen years (Kumpalainen & Rasanen, 2000; Rigby, 1998).

Despite variability across definitions and methods of assessment most would agree that peer victimization among adolescents is a prevalent and persistent problem. This review has only highlighted the physical health symptoms reported in the studies mentioned but by no means do I ignore the multitude of psychological problems that are documented in relation to this type of interpersonal dysfunction.

This group of studies emphasizes common ailments youth report on a regular basis, and they do indeed demonstrate a positive relationship with the prevalence and severity of bullying. However, different methodologies, terminology, and statistical analyses, make it difficult to infer causality and directionality of bullying and health outcomes, particularly over time. This highlights a gaping hole in our understanding of the entwined effects of dysfunctional social relations and physical health outcomes. However, the extant research does provide a stepping stone toward building our bridge of understanding of how victimization can influence our physical health, as well as our mental and emotional well being over time. The next section reviews bullying in the workplace research and whether the same associations between victimization and health still exist in adulthood.

Adults Between 22 and 59 years of age

Bullying in the workplace has been given many names including harassment (Brodsky, 1976), scapegoating (Thylefors, 1987), workplace trauma (Björkqvist, Österman, & Hjelt-Bäck, 1994), abusive behavior/emotional abuse (Keashly, Trott, & MacLean, 1994; Keashly, 1998), bullying (Adams, 1992; Rayner, 1997; Einarsen, 1996; Vartia, 1996), aggression (Kaukainen et al. (2001), and mobbing (Zapf, 1999) among others. They all seem to refer to variations of the same phenomenon – victimization. Comparable to bullying definitions for children and adolescents this phenomenon is systematic and repeated negative or aggressive behaviors from one or more colleagues or superiors, in a situation where an individual cannot defend him/herself or escape the behaviors. These behaviors can cause severe social, psychological and physical (psychosomatic) health problems for the victims or those exposed to such negative acts. As with the child and adolescent review this section only included studies incorporating health measures in association with bullying behaviors.

Bullying Prevalence at in the Workplace

Although the psychological ramifications of bullying behavior involvement are well documented Gemzøe Mikkelsen and Einarsen (2001) believed that Danish research had largely ignored these issues in recent years. Consequently, they examined the previous and current prevalence of bullying in two studies: Study 1 with postgraduate students (n= 90), and Study 2 within four Danish organizations (N=675). They assessed whether exposure to bullying behaviors at work is (still) related to self-reported psychological and psychosomatic stress symptoms. They measured bullying in 2 ways: with a self report question and a 23-item Negative Acts Questionnaire (see Einarsen and Raknes, 1997). A seven-item Stress Profile (Setterlind & Larsson, 1995) measured psychosomatic complaints in Study 2. Overall, they did find low levels (2-4%) of bullying prevalence among their samples. Moreover, they also determined the operational method of measuring prevalence may be the most reliable way to estimate prevalence figures. In accordance with previous research on bullying behavior at work and associations with increased health problems (e.g. Einarsen & Raknes, 1997; Zapf, Knorz, & Kulla, 1996) the data revealed correlation coefficients of up to .39 between reports of bullying behaviors and psychosomatic symptoms. Consequently, bullying may have a low prevalence in Danish work-life, but exposure to bullying behaviors at work, no matter how frequent, has the capacity to elicit stress symptoms in workers.

A slightly different approach taken by Finnish researchers assessed what they termed as “everyday aggression” – interpersonal aggressive behavior that can occur frequently in the workplace and can affects employees' global well-being; which might also be called victimization. Kaukainen and colleagues (2001) defined aggression as the intent to injure another person, and were concerned with workplace aggression in a general sense more in line with bullying acts (which might be defined as harassment) rather than severe assault. Interestingly, Kaukainen et al. (2001) also examined the perception by employees that their health and well being was attributed to the aggression in the workplace. Studying a cross-section of occupations and a wide age range (20-60 years) they collected self reported data on overt and covert aggression and measures of subjective well being, including a ten-item physical symptom scale.

As is the case with the child and adolescent literature indirect styles of aggression were deemed the most frequently used type of aggression, while males' lack of well-being was positively correlated with all types of experienced aggression. Overall 12.4% of the employees considered themselves to be frequently victimized and subjectively experienced greater psychosocial and somatic problems than those who had been victimized to a lesser extent or not at all. Furthermore, the victimized group tended to attribute their psychosocial symptoms to be being a target of workplace aggression more often than the lesser victimized group (M=3.42 vs. 0.91, t = 3.79, df = 165, P < .001). This also speaks to the idea that although people are experiencing physical symptoms in relation to bullying exposure, they may not be recognizing the relationship between the two. This has implications for how we might tackle the bullying and health relationship for all those concerned whether they are children, adolescents, or adults.

The final study in this section highlights a small qualitative study (N=22) explored perceived health consequences of workplace bullying via open interviews with informants (Hallberg & Strandmark, 2006). The interviews comprised of open introductory questions, and relevant probing follow-up questions. Bullying was associated with both psychological and physical symptoms only months after bullying had been experienced. Psychosomatic symptoms reported included, for example, headache, gastric catarrh, hypersensitivity to sounds, respiratory and cardiac complaints, hypertension, and general body pain. These subjective reports again suggest a dose-response pattern, in that over time the somatic complaints were reported as becoming chronic, and informants related their occurrence to the work situation. However, as a qualitative study, and with only seven male informants, the findings must be interpreted with caution as they may not be transferable to other settings, and no causal relationships can be established. There are several limitations to this type of study including no clear definitions or descriptions of health problems, and time frames for associations between bullying and health outcomes are subjectively reported. Also, there are no controls for previous mental and physical health, even though some individuals reported previous health conditions such as diabetes, asthmas, and hypertension. There were no provisions for individual differences in personality or emotionality prior to experiencing poor interpersonal relations at work. As reported previously these factors may have moderating or mediating effects on perceived bullying outcomes. Conversely, this study provided a personalized insight into some of the many stressors those involved in bullying subjectively experience whether from superiors (downward bullying), or from colleagues (horizontal bullying) and subsequent health problems they personally experience that may affect them into their latter years.

Dispositional Effects on Bullying Outcomes

In the adolescent research Jensen-Campbell et al. (2010) found that after controlling for personality variables, victimization independently and significantly predicted health outcomes. Other evidence suggests that the ability to tolerate negative stimuli depends on personality (e.g., Costa & McCrae, 1980). This psychosomatic model of bullying suggests that bullying leads to negative affect, which in turn, leads to physical symptoms.

Workplace studies have examined the psychosomatic model of bullying by testing the associations between personal dispositions such as negative affect (NA) and self efficacy and bullying outcomes (Gemzøe Mikkelsen & Einarsen, 2002). In their study of 433 Danish manufacturing employees Gemzøe Mikkelsen and Einarsen determined that exposure to workplace bullying correlated moderately with psychosomatic complaints (r=.32, p<.01, explaining 10% of the overall variance in psychosomatic complaints). Simultaneously, bullying behaviors correlated with state NA (but not self efficacy) which acted as a partial mediator between bullying and psychological and psychosomatic health outcomes. Self efficacy only partially moderated the relationship between bullying and health outcomes, providing further evidence that bullying behaviors are detrimental to one's well being above and beyond the influence of personal characteristics (Bjorkvist et al. 1994; Niedl, 1996; Zapf et al. 1996; Kivimaki et al. 2004).

Likewise, the central research question of Hansen et al. (2006) was to examine if negative affect might influence the appraisal of potential stressors such as victimization in the workplace. Hansen and colleagues assessed five diverse workplace domains, within a wide age range (18-65 years). They determined that negative affect partially mediated the effects of bullying on health outcomes. In other words, bullied individuals may well experience more nervousness and anxiety in this particular negative environment. However, this could only be ascertained as a state effect rather than a trait effect due to the cross-sectional nature of the study, and not generalized to other domains of employees' lives. Simultaneously, Hansen and colleagues found support for their hypothesis that bullied employees experience a lower physiological stress response than non-bullied employees independent of other factors. Like Knack et al. (2009), they also found a different diurnal profile with lower concentrations of awakening cortisol levels for bullied respondents. These findings along with higher rates of self reported physical and mental health symptoms suggest bullied workers experience more chronic health symptoms as well. However, this particular study did not assess how long victims had been exposed to bullying, and thus cannot evaluate the effects based on the length of exposure. Although correlational, this study still provides strong empirical evidence that workplace bullying is detrimental to victims' health and well being independent of individual differences in disposition or personality.

In a Similar approach as the adolescent research, Djurkovic, McCromick and Casimir (2006) assessed the particular ole of neuroticism on poor interpersonal relations and health outcomes in the workplace. Theoretically, neuroticism should moderate the perceived effects of workplace bullying on both psychological and physical health outcomes. For instance, neuroticism is associated negatively with stress tolerance (Dornick & Ekehammar) 1990) and has strong positive correlations with anxiety, depression, and medically unexplained physical symptoms (e.g. De Gucht, Fischler, & Heiser, 2004).

FILL THIS IN WITH WHAT THEY DID – FINDINGS AND SHORTCOMINGS

Sex Differences

Hoel, Farragher, and Cooper (2004) assessed the lingering effects of bullying on workplace individuals, as well as the effects on witnesses of bullying behaviors in a large British national study. In line with adolescent studies of bullying Hoel et al. (2001, 2004) found gender differences in bullying experiences and outcomes. Interestingly, they also found the effects of being a victim appeared to increase with age, while those bullied in the past were less affected than recent targets. The latter finding speaks to the idea that retrospective reporting (over distant time frames) dilutes possible bullying and health outcome effects (see Reis & . Hoel et al's (2004) higher reporting of physical symptoms for males also corroborates findings in the children's bullying literature that boys are prone to reporting more physical symptoms (Rigby 1998 and others).

Moderating and Mediating Effects on Bullying/Health Associations

While examining the relationship between different types of abuse and health outcomes Striegel-Moore et al. (2002) also assessed whether bullying by peers is associated with an increased risk for developing binge eating disorder in black and white women. They interviewed women already diagnosed with binge eating disorder, a healthy group, and a psychiatric comparison group (N= 520). Although the development of eating disorders has been linked with certain types of abuse, (particularly sexual abuse) findings have been inconsistent (Kearney-Cooke & Striegel-Moore 1994). However, Striegel-Moore and colleagues hypothesized that binge eating may well be a coping mechanism for individuals experiencing extreme environmental, and/or personal stressors (Wilfley, Pike, Dohm, Striegel-Moore, & Fairburn, 2001). They found that both black and white women with binge eating disorder were significantly more likely than healthy comparison women to report a history of bullying by peers. This is a consistent with previous findings of eating disorders and discrimination among children and adults (Puhl & Brownell, 2001). However, reporting of bullying in childhood is not uniquely associated with the development of eating disorders but is considered a risk factor for the development of later mental disorders (Striegel-Moore, et.al., 2002 and others). I included this study as it concentrates on one specific disorder other than common physical ailments such as colds, flu etc. with bullying involvement. Conversely, the study focused on other forms of abuse, not specifically bullying, thus the types of bullying, frequency, duration, and severity are unknown and it would be unwise to draw causal conclusions. It is however, further evidence that the experience of bullying can have far reaching effects long after the immediate experience. In turn this lends support to the hypothesis that victimization experiences at one point in life do indeed have the capacity to affect physical health outcomes at a later life stage.

Of the few studies to consider biological characteristics, Kivimäki et al. (2004) completed a large prospective study of (N= 4791) Finnish hospital workers (1998-2000). They prospectively examined the occurrence of work stressors influencing the onset of fibromyalgia, a functional somatic syndrome with no known etiology; characterized by multiple symptoms including widespread muscle pain, tenderness, and sleep disturbance (Kivimäki et al., 2002). In addition to considering high work load and low decision latitude in the workplace Kivimäki and colleagues assessed the role of workplace bullying as a predictor for stress related health symptoms, (e.g., fibromyalgia, osteoarthritis, sciatica). They controlled for age, sex, income, smoking, comorbid psychiatric disorders, and body mass index. The strongest associations were found between workplace bullying and incidence of fibromyalgia. Furthermore, the 243 respondents who reported workplace bullying were more than four times as likely as non-bullied employees to develop fibromyalgia. However, a possible limitation could be that perceptions of bullying may have been influenced by undiagnosed physical pain (i.e., fibromyalgia) at baseline. In sum, this study highlights again, that prolonged bullying stress may well affect underlying immune system functioning via reduced cortisol levels (a characteristic associated with fibromyalgia patients) that can result in chronic health conditions later on.

LIMITATIONS

SUMMARY

Senior Citizens older than 60 years of age

It is interesting how bullying among social groups, whether they comprise of children or adults, has been tolerated for so long. Although there is growing evidence for the effects of bullying and health outcomes in younger people, there is still very little systematic investigation directed toward the effects of peer victimization in the elderly. Children and adults alike will use physical and relational aggression in calculated and systematic ways to gain control over those they perceive to be weaker. However, older people are likely to employ adaptive bullying strategies that can maximize on the effects of bullying, but minimize personal risk (Campbell et al*****From Lauri on desktop). Indirect aggression has been found to be prevalent in older adults in retirement communities, especially when they are involved in large and loosely connected networks of acquaintances (Walker, Richardson, & Green, 2010). However, these effects go largely unseen, and little research has assessed health outcomes of bullies and bullied individuals in this age group.

Only a handful of studies have even looked at the extent of abuse experienced by older people (Grossman & Lundy 2004; Teaster & Roberts 2004). Collectively these studies leave us with questions for example a) What types of victimization do older people experience, b) How often does this victimization occur, c) What are the health correlates, if any of elderly victimization? Given the paucity of elder abuse studies and the physical health correlates it is not surprising that our team of researchers could find no studies focusing on peer victimization as defined by this review - the everyday type of peer victimization that we know exists but appears not to be documented. For example, when a member of a Women's Red Hat Society group is explicitly and systematically targeted as a nuisance, ignored at functions, and not invited to attend social gatherings – is this still bullying? Older people often refer to being bullied into making decisions about their finances or possessions, etc, by family members and peers. Perceived social support matters to elderly people in that studies have recognized their level of social involvement does have an impact on health outcomes (*****). Consequently, one might expect senior citizens to be somewhat vulnerable to social demands of others in order to maintain that level of perceived social support. Socio-selectivity theory (****) posits that as individuals age they seek out meaningful social groups and individuals who meet their socialization needs and tend to exclude those they deem as peripheral in their social hierarchy. On that premise one might argue that peer victimization at this age is a non-issue. I believe that as people age bullying still occurs but we mask it as different things – and it may manifest in research as abuse. Physical and sexual spousal abuse is well documented in older people, as well as nursing home and care abuse and aggression. However, information on health outcomes of those non-institutionalized civilians that still actively function within society is practically non-existent. One avenue of related research by Walker Richardson and Green (2000) documented how seniors adopted different techniques of aggressive behaviors in order to manipulate and control individuals within their social circles. It makes sense that older individuals will employ the least costly of techniques i.e. non- physical tactics in order to get what they want.

According to the US Census Bureau (2008) 23.4% of US citizens are over 55 and 17.4% are over 60, this translates into many people are likely to be involved in bullying phenomenon either as victims, perpetrators, or bully-victims but research to date has never approached bullying as a lifespan phenomenon. Therefore, I posit that older people who firstly, may have been bullied as children or as working adults may be susceptible to experiencing or promoting similar behaviors as older citizens. Secondly, I propose that some older citizens are still affected by bullying processes and may experience poorer physical health outcomes than those who are not currently nor have never been bullied. Thirdly, I expect that victimization stress will have had an impact on seniors' telomere growth and decline, and thus will have affected their biological aging process.

It is still unclear whether the dose-response relationship between bullying experiences and physical health problems demonstrated repeatedly throughout this review is determined by the proximal effects of victimization or whether there are more distal corollaries. There is strong suggestion from current research that effects of chronic or repeated exposure to stress at different time periods in life depend on the brain areas that are either developing or declining at the time of exposure. (see Lupien, McEwen, Gunnar, & Heim, 2009). Therefore, poor interpersonal relations during period of rapid brain development may have adverse effects on particular brain structures as well as cognitive, emotional, and physical deficits later on (e.g., Gunnar & Donzella, 2002;). This has been demonstrated in avenues of research that have assessed early stressors influencing later psychological health outcomes, such as Holocaust trauma research (Kielsen1992; Yehuda et al., 2007) childhood familial adversity, and adverse childcare stress (e.g. Gunnar & Quevedo, 2007; Lupien, King, Meaney, & McEwen, 2000). Most researchers would agree that stress has a direct effect on a number of physiological systems including the nervous system, endocrine system, and hypothalamic-pituitary-adrenal (HPA) stress system which then affect other systems in the body (e.g., Cannon, 1929; Carver, 2007; Dougall & Baum, 2001; Luecken and Lemery, 2004; Mason, 1971; Repetti et al., 2002; Selye, 1956; Troxel and Matthews, 2004). One of the most frequently studied components of the HPA axis is the glucocorticoid hormone cortisol. This system is designed to enable the quick and responsive secretion of cortisol necessary to adaptively respond to challenge. McEwen (1998) suggested different processes in the stress response can ultimately result in overexposure or underexposure to stress hormones, which fail to adequately prepare the individual to meet situational demands (McEwen and Wingfield, 2003) subsequently increase susceptibility to poor health outcomes. This pathway is a complex process because health outcomes may not be directly visible until years after particular stressors are presented. It is posited that persistent and altered activation of the endocrine and immune system resulting from chronic childhood and adolescent stress may contribute to later risk for fatigue, pain, cardiovascular disease, and other health related problems in adulthood (Heim, Newport, Mletzko, Miller, & Nemeroff, 2008) and especially harmful during childhood when physiological regulatory mechanisms are still developing (Gunnar and Quevedo, 2007; Lupien, McEwen, Gunnar, & Heim, 2009).

Request the removal of this essay...

Struggling?

A custom
essay
can
help...







We write essays!