There are ongoing questions about whether unemployment has causal effects on suicide as this relationship may be confounded by past experiences of mental illness. The present review quantified the effects of adjustment for mental health on the relationship between unemployment and suicide. Findings were used to develop and interpret likely causal models of unemployment, mental health and suicide.

A random-effects meta-analysis was conducted on five population-based cohort studies where temporal relationships could be clearly ascertained.

Results of the meta-analysis showed that unemployment was associated with a significantly higher relative risk (RR) of suicide before adjustment for prior mental health [RR 1.58, 95% confidence interval (CI) 1.33–1.83]. After controlling for mental health, the RR of suicide following unemployment was reduced by approximately 37% (RR 1.15, 95% CI 1.00–1.30). Greater exposure to unemployment was associated with higher RR of suicide, and the pooled RR was higher for males than for females.

Plausible interpretations of likely pathways between unemployment and suicide are complex and difficult to validate given the poor delineation of associations over time and analytic rationale for confounder adjustment evident in the revised literature. Future research would be strengthened by explicit articulation of temporal relationships and causal assumptions. This would be complemented by longitudinal study designs suitable to assess potential confounders, mediators and effect modifiers influencing the relationship between unemployment and suicide.

To identify food sources of fat, to compare food and nutrient intakes at different levels of relative fat intake, and to examine the contribution of different food groups to the variation in relative fat intake. Relative fat intake was expressed as energy contributed by fat in percentage of non-alcohol energy.

Cross-sectional analysis of baseline data from the Malmö Diet and Cancer Study. An interview-based diet history method, a structured questionnaire and anthropometric measurements were used to obtain data. Analysis of variance compared food and nutrient intakes across quintiles of relative fat intake. Stepwise regression examined the contribution of food groups to the variation in relative fat intake.

Baseline examinations were conducted between 1991 and 1996 in the city of Malmö, southern Sweden.

A sub-sample of 7055 women and 3240 men of the Malmö Diet and Cancer cohort.

The major fat sources were dairy products, margarines, meat & meat products, and cakes & buns. Most plant foods, especially fruit, vegetables and breakfast cereals, were negatively associated with fat intake. Low fat consumers had significantly higher intakes of dietary fibre, vitamin C, β-carotene, folic acid, iron, zinc and calcium. Intakes of all types of fatty acids and fat-soluble vitamins were positively associated with fat consumption.

The results suggest that many food groups and nutrients may confound the associations between relative fat intake and disease. Plant foods, especially, are important to consider in studies of fat intake and disease risk.

To compare vitamin B6 status indices with each other and with potential confounding factors, in the datasets from two national British surveys and associated studies.

Vitamin B6 status was measured by plasma pyridoxal phosphate (PLP) and plasma pyridoxic acid (PA) in both surveys, and also by erythrocyte aspartate aminotransferase activation coefficient (EAATAC) in one of the surveys. Plasma α1-antichymotrypsin was measured as an index of acute phase status; plasma creatinine was measured as an index of renal function; and plasma total alkaline phosphatase activity was measured as a proxy for PLP hydrolase activity.

The survey of people aged 65 years and over was carried out in 80 postcode sectors across mainland Britain during 1994–95 and the survey of young people was carried out in 132 postcode sectors across mainland Britain during 1997.

Blood samples from c. 1000 subjects of both sexes in each survey permitted measurements of plasma PLP and PA. There were also measurements of EAATAC in the young people's survey.

According to published limits of normality, only 5% or less of the young people had unacceptable vitamin B6 status as measured by plasma PLP. About half had apparently unacceptable status by EAATAC, but this observation is difficult to interpret. The young people had considerably higher plasma concentrations of PLP and lower concentrations of PA than the older people. In both surveys, plasma PLP was strongly correlated with plasma PA and in the young persons’ survey it was also correlated, although much less strongly, with the basal activity and activation coefficient of aspartate aminotransferase. Both plasma PLP and EAATAC (but not PA nor basal aspartate aminotransferase activity) were influenced by acute phase status in young people, as indicated by significant correlations with α1-antichymotrypsin. In people aged 65 years and over, PA (but not PLP) was correlated with renal function, as indicated by its relation with plasma creatinine; however PLP (but not PA) was correlated with plasma alkaline phosphatase activity.

Several potential confounders — acute phase reaction, kidney malfunction and hydrolase activity — may influence vitamin B6 status indices, although differently for different indices and different age groups. Since older people have relatively poor vitamin B6 status, which may have important health implications for them, more reliable vitamin B6 status indices are needed.