Sage Journals: Discover world-class research

Abstract

Objective: The purpose of the present study was to determine the proportion of adult South Australians who report having experienced school bullying and to examine the relationship between past victimization and adult health-related quality of life.

Method: A representative sample (n=2833) of metropolitan and country South Australian adults were asked in a face-to-face interview whether they had experienced bullying when they were at school. Health-related quality of life was measured using the Medical Outcomes Study Short Form 36-item health survey questions (SF-36). Regression analyses (linear and logistic) were performed, taking into account survey weights.

Results: Nearly one-fifth of adults reported having experienced bullying when they were at school. Older persons and those born overseas were less likely to report having been bullied. Those reporting that they had been bullied experienced significantly poorer mental and physical health compared to those who had not been bullied.

Conclusions: Adults commonly reported experiencing bullying while at school and these reports were associated with lower health-related quality of life in adulthood. School bullying needs further investigation as a preventable cause of mental health problems across the lifespan.

Keywords

bullying health-related quality of life

School bullying is associated with high levels of distress in childhood, and the cross-sectional evidence for the ill effects of school bullying on children's mental and physical health is strong. Bullied children are often socially anxious, and their lives are characterized by feelings of loneliness, depression and low self-esteem [1]. School bullying is also associated with many of the common health complaints of childhood including headaches, abdominal pain, nausea, recurrent upper respiratory tract infections, sore throats and palpitations [2–4]. Reflecting the impact of bullying on overall quality of life, bullied children and adolescents report poorer scores on health-related quality of life (HRQoL) questionnaires [5,6].

There is emerging evidence that the effects of school bullying can have long-lasting effects on emotional functioning. In a longitudinal study of a population sample of Finnish children born in 1981, bullying during the early school years was found to be predictive of psychiatric disorder in early adulthood even after controlling for childhood levels of psychiatric symptoms [7]. Adults who recall school bullying also experience higher rates of emotional disorders. Gladstone et al. found that one-quarter of adults surveyed in an outpatient depression clinic reported being bullied in childhood and they experienced higher rates of comorbid anxiety [8]. Lund et al., from a general population survey, found that adult reports of bullying at school were associated with higher levels of depressive symptoms and a significantly increased risk of being diagnosed with depression between ages 31 and 51 years [9]. This suggests that school bullying could be a preventable cause of mental health problems in the adult population at least into mid-life.

The findings from these studies suggest that school bullying could be aetiologically significant in increasing the risk of adult depression and anxiety. It is not known, however, whether the psychosomatic symptoms and poorer HRQoL experienced by bullied children also persist into adulthood. The current study sought to extend the research into school bullying by surveying HRQoL and past histories of peer victimization among a representative sample of South Australian adults. Specifically we hypothesized that those adult respon dents who reported having been bullied at school would experience both poorer physical and mental health compared with adults who did not report having been bullied at school.

Methods

Data were collected in the South Australian Health Omnibus Survey (HOS). The HOS sampling procedures have been described in detail previously [10,11] but, briefly, a large representative sample of South Australians aged ≥15 years is surveyed in face-to-face interviews. In Autumn 2008, 4830 households were sampled and a response rate of 62.8% was achieved. In the present study, data from participants in the HOS who were ≥ 18 years of age and who were no longer in school (n=2833) were used in the analyses.

Self-reported bullying at school was determined from the following question: ‘When you were at school did you experience traumatic bullying by peers that was particularly severe, for example, being frequently targeted or routinely harassed in any way by ‘bullies’?’ Responses to this question were coded as no = 0, not bullied; and yes= 1, bullied.

Physical and mental health were measured using the Medical Outcomes Study Short Form 36-item health survey questions (SF-36). The SF-36 has eight subscales (physical function, role physical, bodily pain, general health, vitality, social function, role emotion, mental health) and two summary components (one each for physical and mental health). For ease of comparison, normed t-scores (mean = 50, SD=10) of the SF-36 sub-scales were used in analyses. Data were weighted according to the algorithms for use with the South Australian population [12].

Analyses were performed using Stata/IC version 10.0 (Statacorp, College Station, TX, USA). Survey data were weighted by gender, age and demographic variables (country of birth, marital status, educational attainment and household income) in order to be representative of the South Australian population [13]. Some small discrepancies in frequency and percentage totals reported in the tables occur due to rounding in the presence of weighted data. Logistic regression was used to identify the independent predictors of the characteristics of adults reporting that they had experienced school bullying and multiple regression was used to identify the impact of school bullying on Mental and Physical health SF-36 Component Summary scores.

Ethics approval for including the bullying survey question in the HOS was provided by the South Australian Department of Health–Human Research Ethics Committee.

Results

Overall, 18.7% of participants reported recalling bullying at school. Table 1 shows the percentage of persons who reported having experienced bullying at school vs demographic variables. Higher proportions of younger people, those born in Australia and people who had never married reported that they had experienced bullying at school.

Table 1.

Reporting of bullying vs demographic variables

	Not bullied		Bullied		Total
	n	%	n	%	n	%
Gender
Female	1194	81.9	264	18.1	1458	51.5
Male	1110	80.7	265	19.3	1375	48.5
Total	2304	81.3	530	18.7	2833	100.0
Age group (years)
18–24	229	76.6	70	23.4	299	10.6
25–34	349	76.7	106	23.3	455	16.1
35–44	414	76.9	125	23.2	539	19.0
45–54	427	80.0	107	20.0	534	18.8
55–64	372	85.3	64	14.7	436	15.4
65++	513	89.8	58	10.2	571	20.2
Total	2304	81.3	530	18.7	2833	100.0
Household income (AUD)
<40 000 p.a.	657	80.0	164	20.0	822	34.9
>40 000 p.a.	1244	81.1	291	18.9	1535	65.1
Total	1901	80.7	455	19.3	2356	100.0
Area
Metropolitan	1697	81.1	395	18.9	2093	73.9
Regional	606	81.9	134	18.1	741	26.1
Total	2304	81.3	530	18.7	2833	100.0
Employment status
Working	1331	81.0	312	19.0	1643	61.4
Not Working	861	83.4	171	16.6	1032	38.6
Total	2192	81.9	483	18.1	2675	100.0
Country of birth
Australia	1680	79.7	429	20.3	2109	74.4
UK and Eire	271	84.6	50	15.4	321	11.3
Other European	180	87.6	26	12.4	206	7.3
Asian	110	86.9	17	13.1	126	4.5
Other	63	87.3	9	12.7	72	2.5
Total	2304	81.3	530	18.7	2833	100.0
Schooling
Left before 15/Trade qualification	645	81.6	145	18.4	791	27.9
Left after 15	686	81.0	161	19.0	847	29.9
Tertiary	971	81.3	224	18.7	1194	42.2
Total	2302	81.3	530	18.7	2832	100.0
Indigenous
No	1623	79.7	413	20.3	2037	97.1
Yes	48	78.5	13	21.5	61	2.9
Total	1671	79.7	426	20.3	2097	100.0
Marital status
Married/de facto	1555	81.9	344	18.1	1899	67.0
Separated/divorced/widowed	356	82.6	75	17.4	430	15.2
Never married	393	78.0	111	22.0	504	17.8
Total	2304	81.3	529	18.7	2833	100

Small discrepancies in frequency and percentage totals due to rounding of the weighted data.

The results from a series of univariate logistic regressions (Table 2) showed that those aged 55–64 years (odds ratio (OR) = 0.56, 95% confidence interval (95%CI) = 0.37–0.85) and ≥65 years (OR=0.37, 95%CI=0.24–0.56) were approximately half as likely to report recalling that they were bullied at school compared with respondents aged 18–24 years of age. Respondents born in the UK and Eire (OR=0.71, 95%CI=0.51–1.0) or other European countries (OR=0.56, 95%CI=0.35–0.87) were also less likely to report having been bullied at school compared with Australian-born respondents. In a multivariate analysis that included both variables (age group and country of birth) simultaneously, both variables retained their significance.

Table 2.

Predictors of self-reported experience of school bullying

		Univariate		Multivariate
	n	OR	95%CI	OR	95%CI
Gender
Female (R)	1458
Male	1375	1.08	0.88–1.33
Age group (years)
18–24 (R)	299
25–34	455	0.99	0.66–1.50	1.01	0.67–1.53
35–44	539	0.99	0.67–1.46	1.01	0.68–1.49
45–54	534	0.82	0.55–1.22	0.83	0.55–1.24
55–64	436	0.56∗	0.37–0.85	0.58∗	0.38–0.88
65++	571	0.37∗	0.24–0.56	0.38∗	0.25–0.58
Household income (AUD)
>40 000 p.a. (R)	1535
<40 000 p.a.	822	1.07	0.85–1.34
Area
Metropolitan (R)	2093
Regional	741	0.95	0.74–1.22
Employment status
Working (R)	1643
Not working	1032	0.85	0.68–1.05
Indigenous Australian?
No (R)	1623
Yes	48	1.07	0.54–2.13
Country of birth
Australia (R)	2109
UK and Eire	321	0.71∗	0.51–1.00	0.85	0.61–1.20
Other European	206	0.56∗	0.35–0.87	0.74	0.47–1.18
Asian	126	0.59	0.34–1.04	0.52∗	0.29–0.93
Other	72	0.57	0.27–1.20	0.53	0.25–1.14
Schooling
Left before 15/Trade qualification	791
Left after 15	847	1.04	0.79–1.37
Tertiary	1194	1.02	0.80–1.31
Marital status
Married/de facto (R)	1899
Separated/divorced/widowed	430	0.95	0.73–1.24
Never married	504	1.28	0.97–1.68

CI, confidence interval; OR, odds ratio; R, reference group; ∗p<0.05.

The mean scores on SF-36 subscales were compared between those who reported being bullied and those who had not (Table 3). Participants who reported having been bullied had statistically significantly poorer outcomes on seven of the eight SF-36 subscales: bodily pain, role physical, general health, vitality, social function, role emotion and mental health.

Table 3.

SF-36 subscales as a function of having experienced bullying

	Not bullied		Bullied
	Mean	SE	Mean	SE	t	p
SF-36 subscales
Physical function	49.79	0.24	49.60	0.45	−0.22	0.82
Role physical	49.09	0.24	47.54	0.53	2.66	0.008
Bodily pain	49.79	0.23	47.04	0.49	5.12	<0.001
General health	49.28	0.23	47.46	0.48	3.41	<0.001
Physical summary	49.51	0.24	49.15	0.44	0.72	0.469
Vitality	49.43	0.23	46.64	0.53	4.86	<0.001
Social function	49.57	0.24	45.72	0.59	6.02	<0.001
Role emotion	49.55	0.24	45.25	0.67	6.03	<0.001
Mental health	49.69	0.22	45.23	0.56	7.34	<0.001
Mental summary	49.68	0.23	44.83	0.61	7.45	<0.001

SF-36, Medical Outcomes Study Short Form 36-item health survey questions.

The negative effects of bullying on Mental Health Component Summary Scores was assessed in a multivariate model adjusting for demographic factors associated with mental health (Table 4). A series of univariate analyses showed that age, gender, income, employment, education, marital status and identification as an Indigenous Australian were related to Mental Health Component Summary Scores. These variables were simultaneously entered into a multiple regression in addition to the variable ‘Bullied’. After adjusting for demographic variables the effect of bullying remained significantly associated with poorer levels of self-reported mental health (t=−5.58, p<0.001).

Table 4.

Factors associated with SF-36 Mental Health Component Summary Scores

	Coefficient	t	p	95%CI
Gender
Female (R)
Male	1.30	2.43	<0.05	0.25–2.34
Age group (years)
18–24 (R)
25–34	−1.09	−0.94	NS	−3.35 to 1.18
35–44	−0.59	−0.56	NS	−2.68 to 1.50
45–54	−0.20	−0.19	NS	−2.43 to 2.01
55–64	1.49	1.28	NS	−0.81 to 3.78
65++	5.90	4.32	<0.001	3.22–8.57
Household Income (AUD)
>40 000 p.a. (R)
<40 000 p.a.	−3.12	−4.11	<0.001	−4.61 to −1.63
Employment status
Working
Not working	−2.10	−2.71	<0.05	−3.62 to −0.58
Schooling
Left before 15/Trade (R)
Left after 15	1.30	1.69	NS	−2.21 to 2.81
Tertiary	1.13	1.64	NS	−0.22 to 2.47
Indigenous
No (R)
Yes	−3.78	−1.92	NS	−7.66 to 0.09
Marital status
Married/de facto (R)
Separated/divorced/widowed	−2.51	−2.86	<0.05	−4.23 to −0.79
Never married	−1.97	−2.40	<0.05	−3.58 to −0.36
Bullied
No (R)
Yes	−3.90	−5.58	<0.001	−5.27 to −2.53

CI, confidence interval; R, reference group.

The effects of bullying on Physical Health Component Summary Scores was assessed in a multivariate model adjusting for demographic factors (Table 5). A series of univariate analyses showed that gender, region, age group, household income, employment status, country of birth, schooling and marital status were related to Physical Health Component Summary Scores. These variables were simultaneously entered into a multiple regression in addition to the variable ‘Bullied’. After adjusting for demographic variables (and in particular the variable age group), the effect of bullying was significantly associated with poorer levels of self-reported physical health (t = −2.41, p<0.05).

Table 5.

Factors associated with SF-36 Physical Health Component Summary Scores

	Coefficient	t	p	95%CI
Gender
Female (R)
Male	1.77	4.46	<0.001	0.99–2.55
Age group (years)
18–24 (R)
25–34	0.31	0.43	NS	−1.08 to 1.69
35–44	−0.64	−0.87	NS	−2.07 to 0.80
45–54	−2.99	−3.78	<0.001	−4.54 to −1.44
55–64	−4.46	−5.25	<0.001	−6.13 to −2.79
65++	−6.36	−6.11	<0.001	−8.41 to −4.32
Household income (AUD)
>40 000 p.a. (R)
<40 000 p.a.	−0.89	−1.54	NS	−2.03 to 0.24
Employment status
Working
Not working	−3.01	−4.95	<0.001	−4.21 to −1.82
Area
Metropolitan (R)
Regional	−0.56	−1.21	NS	−1.46 to −0.34
Schooling
Left before 15/Trade (R)
Left after 15	2.59	4.63	<0.001	1.59–3.70
Tertiary	1.89	3.44	<0.005	0.82–2.97
Country of birth
Australia (R)
UK and Eire	−0.85	−1.16	NS	−2.29 to 0.59
Other European	−0.67	−0.68	NS	−2.63 to 1.28
Asian	3.31	4.58	<0.001	1.89–4.73
Other	3.77	3.65	<0.001	1.74–5.79
Marital status
Married/de facto (R)
Separated/divorced/widowed	−0.64	−1.03	NS	−1.85 to 0.57
Never married	−0.45	−0.88	NS	−1.45 to 0.55
Bullied
No (R)
Yes	−1.16	−2.41	<0.05	−2.11 to −0.22

CI, confidence interval; R, reference group.

Discussion

School bullying is common in the community and in the present survey nearly one-fifth of South Australian adults reported recalling experiences of bullying while they were at school. The results from the present study indicate that reports of early exposure to bullying were associatedwith an increased risk of emotional and psychosomatic disorders. Adults recalling school bullying experienced more psychiatric symptoms and greater impairment across all of the SF-36 mental health scales. They reported more anxiety and depressive symptoms such as feeling nervous, downhearted and depressed, together with low vitality and poorer functioning due to emotional problems. Although several demographic factors (gender, income, employment status and marital status) had significant relationships with overall mental health, the relationship between school bullying and mental health remained significant after adjusting for demographic variables.

It is only recently that systematic evidence is becoming available to estimate the strength of the association between past histories of school bullying and adult adjustment. These studies have been mostly restricted to small clinical samples but they do show increased rates of mid-life depression and higher levels of depressive symptoms for adults who recall being bullied at school [6,9,14]. The present findings extend this line of research and indicate that reports of school bullying are associated with significant adult mental health problems in the general community.

As far as we can determine, the current study is the first to examine the effects of school bullying on physical health as part of a standard assessment of HRQoL in the adult population. This is an important step because childhood studies have repeatedly found that school bullying is linked to higher rates of psychosomatic complaints. The present survey found poorer adult physical health in three of the four physical health subscales of SF-36: General Health, Role Physical and Bodily Pain. This pattern of increased symptoms of pain and frequent sickness is consistent with reports from childhood studies in which higher levels of headaches, abdominal pain and frequent illness are thought to be stress-related conditions secondary to the trauma of the bullying.

The ability of adults who reported school bullying to accomplish specific everyday physical functions such as climbing stairs, carrying groceries or heavy lifting, however, was not significantly affected. Again this is similar to the childhood pattern in which it is more common to find mild to moderate increases in physical symptoms rather than severe disability. The differences between the bullied and non-bullied groups on the HRQoL sub-scales were statistically significant but they were not large enough in absolute terms (around 2 points) to be above the clinical threshold for the SF-36 scales (>3 points different) [15].

The present study found that younger respondents were more likely to report school bullying than older respondents. This could indicate that the frequency of bullying has increased over the second half of the 20th century. Alternatively, younger respondents might have a broader interpretation of the concept ‘traumatic bullying’ and label a more extensive range of negative peer behaviour as bullying. Given the considerable public health focus on the issue in recent decades it is also possible that younger respondents feel more comfortable (or less stigmatized) reporting that they have been bullied. Further research is required to better understand the effect of age on the reporting of specific kinds of bullying behaviour.

A major limitation of the present study is that it is cross-sectional and it is not possible to determine the direction of causality between reports of past victimization and adult HRQoL. It is possible that reports of past bullying could be affected by current mood rather than the reverse. The accuracy of retrospective recall, however, has been examined in some depth in other studies and there is reasonable support for the validity of adult recall. Brewin et al. reviewed the evidence for retrospective reports and, on the basis of both experimental and naturalistic studies, concluded that recall of significant past events is not greatly affected by mood state [16]. In a longitudinal study Rivers found that retrospective recall of bullying was reasonably stable and consistent over time [17].

Further study of school bullying and adult health and well-being seems justified in the light of the findings of this population survey. School bullying could represent a preventable cause of increased adult anxiety, depression and stress-related health problems. This line of research might be useful in the investigation of the long-term potential for anti-bullying programs to improve health and well-being across the lifespan. The current study was part of a large population survey that covered a wide range of health-related behaviours. Given the restrictions on the length of the survey, the amount of information available about school bullying is limited and did not cover the full range of bullying behaviour. The survey used an item from a clinical study of school bullying among depressed outpatients [8], which was designed to capture more severe forms of peer victimization. Future studies could explore the full extent of victimization and involvement in bullying behaviour, including more specific information about the responses to bullying such as reactive aggression; involvement with bully/victim behaviour; and the overall quality of peer relationships.

Footnotes

Acknowledgement

This study was supported by the Australian Government's Primary Health Care Research, Evaluation and Development (PHCRED) Strategy.

References

1. Hawker

DSJ

Boulton

. Twenty years’ research on peer victimization and psychosocial maladjustment: a meta-analytic reviewof cross-sectional studies. J Child Psychol Psychiatry 2003; 41:444–455.

2. Williams

Chambers

Logan

Robinson

. Association of common health symptoms with bullying in primary school children. BMJ 1996; 313:17–19.

3. Wolkea

Woodsa

Bloomfielda

Karstadtb

. Bullying involvement in primary school and common health problems. Arch Dis Child 2001; 85:197–201.

4. Rigby

. The relationship between reported health and involvement in bully/victim problems among male and female secondary schoolchildren. J Health Psychol 1998; 3:465–476.

5. Svavarsdottir

Orlygsdottir

. Health-related quality of life in Icelandic school children. Scand J Caring Sci 2006; 20:209–215.

6. Wilkins-Shurmer

O’Callaghan

Najman

Bor

Williams

Andersen

. Association of bullying with adolescent health-related quality of life. J Paediatr Child Health 2003; 39:436–441.

7. Sourander

Jensen

Ronning

. What is the early adulthood outcome of boys who bully or are bullied in child hood? The Finnish ‘from a boy to a man’ study. Pediatrics 2007; 120:397–404.

8. Gladstone

Parker

Malhi

. Do bullied children become anxious and depressed adults? A cross-sectional investigation of the correlates of bullying and anxious depression. J Nerv Ment Dis 2006; 194:201–208.

9. Lund

Nielsen

Hansen

. Exposure to bullying at school and depression in adulthood: a study of Danish men born in 1953. Eur J Public Health 2009; 19:111–116.

10.

10. Taylor

Dal Grande

Wilson

. The South Australian Health Omnibus Survey 15 years on: has public health benefited? Public Health Bull 2006; (3):30–32.

11.

11. Wilson

Wakefield

Taylor

. The South Australian Health Omnibus Survey. Health Promot J Aust 1992; 2:47–49.

12.

12. Hawthorne

Osborne

Taylor

Sansoni

. The SF-36 version 2: critical analyses of population weights, scoring algorithms and population norms. Qual Life Res 2007; 16:661–673.

13.

13. Australian Bureau of Statistics. Census of population and housing: Selected Social and housing characteristics. Catalogue no. 2068. Canberra: Australian Bureau of Statistics, 2006.

14.

14. van Dijk

Imhof

Moll

. Quality of life of adult retinoblastoma survivors in the Netherlands. Health Qual Life Outcomes 2007; 5:30.

15.

15. Hayes

Morales

. The RAND-36 measure of health-related-quality-of-life. Ann Med 2001; 33:350–357.

16.

16. Brewin

Andrews

Gotlib

. Psychopathology and early experience: a reappraisal of retrospective reports. Psychol Bull 1993; 113:82–98.

17.

17. Rivers

. Retrospective reports of school bullying: stability of recall and its implications for research. Br J Dev Psychol 2001; 19:129–141.