|
An approach to modelling the effects of lifetime socioeconomic
circumstances
| Title |
A structured approach to modelling the
effects of binary exposure variables over the life course. |
| Authors |
Mishra G, Nitsch D, Black S, De Stavola BL, Kuh D, Hardy
R.
|
| Ref |
International Journal of Epidemiology 2009; 38: 528-537. |
| Lay Summary |
There is growing evidence that the duration of time spent in
unfavourable
circumstances across life is related to an increased risk of
chronic disease
and premature mortality. Analyses that only consider the
association of a
cumulative score derived, for example, from summing binary
indicators of
social class at different ages with a later outcome may produce
misleading
results as they do not consider the full range of potential life
course models
that may be operating.
We described a model-fit approach to disentangle the different life
course
hypotheses, given the assumption of no measurement errors and
provided
the study has sufficient power.51 We recommended comparing a set
of
nested models – each corresponding to the accumulation, critical
period and
effect modification hypotheses – to an all-inclusive (saturated)
model.
Our approach has highlighted that researchers should not adopt an a
priori
hypothesis without testing to see if other life course models fit
the data
equally well. |
| |
|
| |