January 14, 2015

The growing evidence on compression of morbidity

Health, functioning, and disability in older adults—present status and future implications

There is currently a wide debate about chronic care and multimorbidity. Some messages appear that this is strictly connected with ageing, and forget the details. Though disease process have to be tackled, we have to ask ourselves about wellbeing in later life. A key issue is to understand its impact on functioning and disability. This is precisely what a recent article in The Lancet offers. The research question:
Will the years gained be productive and healthy, or will elderly people live longer lives in conditions of ill health? Three main hypotheses have been proposed to address this question.2 The compression of morbidity hypothesis posits a situation for which the age of onset of morbidity is delayed to a greater extent than life expectancy rises, thereby compressing morbidity into a short period at a late age.3 The expansion of morbidity hypothesis maintains the opposite, that increases in life expectancy are matched or exceeded by added periods of morbidity.4 Both compression and expansion of morbidity could happen in absolute or relative terms—ie, changes in the absolute number of years lived with disability—or in terms of healthy life expectancy as a proportion of total life expectancy.
And a conclusion:
 Our systematic examination of the scientific literature shows that support for morbidity pattern hypotheses varies mainly according to the type of health indicator. Disability-related or impairment-related measures of morbidity tend to support the theory of compression of morbidity, whereas chronic disease morbidity tends to support the expansion of morbidity hypothesis.
This is an article to read and file for the future. The basic approach is defined, the difficulty is about the data. My impression is that we need to use morbidity adjusted life expectancy measures, as those I presented in this blog some months ago. The advantages are clear compared to healthy life expectancy that needs a lot of hypothesis and are based on surveys and self perceptions. Morbidity adjusted measures use  disease codes directly. Why not apply them widely?

PS. The whole series on ageing in The Lancet.