Commentary on work stress and mental health

Reichenberg & MacCabe (pp. 1073–1074) provide a commentary on two papers in this issue on work stress and mental health (Melchior et al., Agerbo et al.). Both are prospective studies and confirm previous studies showing associations between work-related stress and poor mental health, specifically depression, anxiety and suicide. While noting that both studies mark methodological advances on previous research, the authors highlight the need for future research to disentangle the mechanisms that underpin the observed associations.

Set-shifting ability in eating disorders

Roberts et al. (pp. 1075–1084) report findings from a systematic review and meta-analysis of 15 studies of set-shifting ability in eating disorders. While the effect sizes varied from small (0·36) to large (1·05) depending on the set-shift task, all studies showed a decline in set-shifting ability in those with eating disorders. The authors note that such deficits are not specific to eating disorders and have been found in other disorders. The authors conclude with suggestions for further research.

Eating disorders

This issue contains three further papers on aspects of eating disorders. In the first, Tseng et al. (pp. 1085–1096) examined the prevalence and correlates of eating disorders (EDs) in a population of gifted dance (n=655) and non-dance (n=1251) high-school students in Taiwan, using a two-phase survey design. They found that the prevalence of all EDs was much higher in the dance (8%) than non-dance (2%) students. The correlates of EDs were: being in a dance class, high concern regarding body image, and lower family support; lower parental education was associated with EDs for non-dance students only.

Duncan et al. (pp. 1097–1107) investigated the clustering of ED symptoms in a sample of 378 young adult twins drawn from the Missouri Adolescent Female Twin Study using latent class analysis. They identified five latent classes, which the authors label: unaffected; low weight gain; weight concerned; dieters; and ED. Those in the ED class had higher levels of co-morbid psychiatric disorders, suicidality and major depression, compared with a number of the other classes. When those in the ED class were separated into those with a diagnosed ED and those without, few differences were observed between them. The authors conclude that there are many subjects with significant ED symptoms who are not being identified by DSM-IV ED categories.

Bulik et al. (pp. 1109–1118) examined the prevalence of ED pre-pregnancy, and patterns of remission, continuation and incidence during pregnancy in a sample of 41157 pregnant women from the Norwegian Mother and Child Cohort Study. Pre-pregnancy prevalences were: anorexia nervosa, 0·1%; bulimia nervosa (BN), 0·7%; binge-eating disorder (BED), 3·5%; ED not otherwise specified (EDNOS), 0·1%. The most common pattern for BN and EDNOS during pregnancy was remission and new incident cases were rare. In contrast, the most common pattern for BED was continuation and a high rate of new cases.

Work stress and mental health

Three further papers examine aspects of work stress and mental health. In the first, Melchior et al. (pp. 1119–1129) investigated the impact of work stress on major depressive disorder (MDD) and generalized anxiety disorder (GAD) in young working adults (n=891) drawn from the Dunedin study. They found that those exposed to high psychological job demands had a two-fold increased risk of MDD and GAD, independent of socio-economic position. In longitudinal analyses, high-demand jobs were associated with the onset of MDD and GAD in those without any pre-job history of diagnosis or treatment for either disorder.

Agerbo et al. (pp. 1131–1140) examined risk of suicide by occupational group in a nested case-control study of 3195 suicides and 63 900 matched controls drawn from Danish population registers. Risk of suicide varied widely, with highest rates in doctors, the unemployed and nurses and lowest rates in architects and the armed forces. With the exception of doctors and nurses, most of the excess rates of suicides in other occupations were due to socio-economic characteristics. For doctors and nurses, much of the excess was due to the increased use of self-poisoning.

Nickel et al. (pp. 1141–1149) report findings from a randomized controlled trial of a behavioural/psycho-educational group intervention for those chronically stressed at work. In a sample of 72 males, the authors report a reduction in the intervention group (n=36) in systolic blood pressure, salivary cortisol concentration, and self-reported stress relative to those receiving a placebo intervention (n=36). The authors conclude that the intervention appears to be effective in the treatment of men suffering chronic stress due to overwork.

Depression

Three papers investigated aspects of depression and other disorders. Lawlor et al. (pp. 1151–1161) examined associations between body mass index (BMI) and risk of psychiatric hospital admission in a cohort of 7036 men and 8327 women aged 45–64 followed up over a median 29 years. They found that a greater BMI was associated with a decreased risk of admission for psychoses and depression/anxiety. These findings were similar for men and women, and held after adjustment for baseline psychological distress and total cholesterol.

Fanous et al. (pp. 1163–1172) examined the relationships between major depression (MD), neuroticism (N) and extraversion (E) in a sample of 3030 male twins using a longitudinal, structural equation twin model. They found that E was negatively correlated with MD. N predicted the onset of MD, and was predicted by current and past MD. The authors further found that covariation between N and MD was due to additive genetic and individual specific environmental factors shared by both traits and a direct causal path between MD and N.

Naz et al. (pp. 1173–1181) investigated remission and relapse in a cohort of 87 subjects admitted to hospital for a first time with psychotic depression followed over 4 years. During the follow-up, 69% achieved a period of full remission. Predictors of a longer time to remission were: a long latency between onset and first admission; a lower GAF score pre-first admission; and lack of medical insurance. Of those who remitted, 43% subsequently relapsed. No predictors of relapse were identified.

Other topics

This issue concludes with three papers examining a variety of topics. In the first, Lloyd-Richardson et al. (pp. 1183–1192) examined the prevalence, correlates and functions of non-suicidal self-injury (NSSI) in a community sample of adolescents (n=633). The prevalence of any NSSI was 47%. The prevalence of moderate/severe NSSI was 28%. Those whose NSSI was moderate/severe were more likely than other groups to have a history of psychiatric treatment, hospitalisation and suicide attempts, and current suicidal ideation. The most common reasons given for NSSI were to influence the behaviour of others and to manage internal emotions.

Silberg et al. (pp. 1193–1202) investigated developmental differences in the aetiology of anti-social behaviour using longitudinal, genetically informed data on 1037 male twins from the Virginia Twin Study of Adolescent Behavioural Development. The authors report a number of trends: (1) a single genetic factor influencing anti-social behaviour at age 10 through young adulthood; (2) a shared environmental effect beginning in adolescence; (3) a transient genetic effect at puberty; and (4) a genetic influence specific to adult anti-social behaviour.

Lubman et al. (pp. 1203–1209) investigated the motivational significance of drug cues in opiate dependence in a sample of 14 methadone-maintained subjects and 14 matched controls. They used active visual oddball tasks to explore the impact of opiate-related cues on central attentional processes. They found that the P300 elicited by opiate stimuli in the oddball task was significantly larger than that elicited by the neutral stimuli in the methadone-maintained group, but not in the controls. The authors conclude that enhanced P300s to drug cues may provide an important biological marker for psychological mechanisms relevant to addiction.