Oslo Mortality Study
Although contemporaneous effects of both individual (composition) and environment (context) on health have been demonstrated, we do not yet have a full understanding as to how the effect of context at different stages of the life course may influence subsequent health outcomes. We are conducting a series of studies to improve our understanding of this subject, in collaboration with Dr Øyvind Næss from the Norwegian Institute of Public Health in Oslo, based on the Oslo mortality study (a cohort of all inhabitants in Oslo aged 30 - 69 years in 1990, linked to the Censuses and various registers from 1960 - 1990, and to the death register from 1990-1998).
Unlike situations in which population health is determined by infectious diseases with a short induction time, most chronic diseases in developed countries probably develop over many years. It should be possible to demonstrate longitudinally any effect of area of residence on ill health, and we have emphasised the need to take account of the interaction between contextual exposures and the stage of the life course. We are using the Oslo mortality study to examine the relative influence of area of residence on mortality risk along the life course in different age groups and, in particular, looking to see if this differs for cardiovascular, psychiatric and violent causes of death. Current work is focusing on the extent to which current and historical compositional and contextual factors may account for the observed variation between areas at different time points.
We have shown that, among younger men, the most recent area of residence is of the greatest importance; this finding coincided with greater importance being attached to the area of residence closest to death for violent and psychiatric causes of death. For older age groups the area of residence at all stages of the life course appeared to have a roughly equivalent impact on mortality. At older ages cardiovascular deaths predominated; such deaths were shown to be related to the area of residence in earlier as well as later life.
A number of studies have demonstrated adverse health effects attributable to exposure to air pollution. However, this exposure is not distributed evenly throughout populations but is patterned according to area deprivation which is, in turn, related to individual socio-economic circumstances. We have shown air pollution to be associated with several neighbourhood-level indicators of deprivation; these area deprivation measures explained some of the effect of air pollution fine particulates (PM2.5) on mortality even after adjusting for individual measures of socio-economic status.
Neighbourhood socio-economic conditions have been found to have an influence on morbidity and mortality even after individual characteristics have been taken into account. However, to measure fully the impact social conditions may have on the risk of mortality, the whole life course must be taken into account since mortality risk increases cumulatively over the life course. Few studies have combined ecological and life course factors to see if contextual effects may be explained by social conditions earlier in life at the individual level. Most earlier studies of neighbourhood effects have had a cross sectional design or with short follow up; the Oslo mortality study includes information about individuals and areas from 1960 with a follow-up until 1998. Using this study, we showed that earlier life social conditions at the individual level can be an important residual confounder when investigating the effect of neighbourhood on mortality risk shortly before death. A significant effect of neighbourhood educational level became insignificant after adjustment for individual deprivation throughout the life course. For deaths among younger men, the most recent area of residence was of greatest importance for subsequent mortality; at older ages we found evidence of the accumulation of the impact of residential exposures, with a broadly equal contribution from areas of residence across the lifecourse.