Fries predicted in 1980 a continuing rectangularization of the survival curve, and trends toward delayed age of onset of chronic disease and toward compression of morbidity in late life. Others have presented evidence that challenges the first two predictions. However, direct evidence regarding the last one has been lacking.
The study reported here analyzed death certificate data on the longest interval between the onset of any condition listed as causing or contributing to the death and the occurrence of the death, from a representative sample of 500 Ontario death certificates in 1975 versus 1985, to deceased individuals aged 65 +. There was a statistically significant decrease between 1975 and 1985 in the frequency of the shortest intervals (under a month) and a not statistically significant increase in the frequency of the longest intervals (5+ years or “years”). Adjustments regarding not-stated intervals and attempts to control for a confounding factor did not change the picture. This study did not yield any statistically significant evidence of a compression of morbidity in late life, and did yield statistically significant findings that were inconsistent with that hypothesis.