Welcome to the hotseat. We've prepared a guide if you'd like to read more about how it works.

Problems associated with an investigation of mortality by occupation

0 votes
asked Nov 5, 2017 in BUS 3024S - Contingencies by anonymous


From the Test 3 paper (2000), one of the questions asked what the problems were with an investigation of mortality by occupation. One of the answers in the memo was given below:

"Census/death register investigations have severe limitations and the alternative approach, through longitudinal studies, following up a group of workers in one particular industry may produce better results in the longer term. This type of investigation would embrace not just mortality, but also sickness absence which occurs nearer the onset of occupational influence on health"

I understand that introducing census every year is very costly (hence making it limited) and it could be better to track a group of individuals and see what the mortality is in that particular group. However, I am not sure what the last sentence in the memo's answer is trying to convey. Is it trying to say that we could also get morbidity rates from this new approach?


1 Answer

0 votes
answered Nov 6, 2017 by Njabulo.Dube (2,950 points)
What the last sentence in the memo is conveying is the notion that under a longitudinal study, where measurements for the same response variable (i.e. sickness) are taken at several occasions for each subject, it would be easier to estimate morbidity rates than under a cross-sectional study (i.e. census every so often).

This seems reasonable because when following the same individuals across time one can determine when they fall ill and how long a particular illness lasts. Under a cross-sectional study the effects of censoring are heightened as census usually take place less often, making estimates of morbidity more difficult.