Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

On clustering

Hands up anyone who is average

On clustering

Older people are often a delight, sometimes frustrating. There are lots of them about, and there will be more before long, and at some time they will include us. So it is perhaps time to think about them and their problems a little more. For professionals helping with their health problems, a little extra light and evidence would probably help as well.

Although the medical care of older people is similar to that of younger age groups, we have surprisingly limited knowledge even of what is normal in the elderly. Clinical trials of treatments rarely include older patients and trials specific to older patients are uncommon. Diagnosis and treatment for this very important group of people depends largely on informed guess work based on evidence from younger patients, plus some dollops of experience and intuition.

While younger people in good or ill health are often broadly similar, there is more of a spectrum in older people, and fit 90 year olds are not the same as sick 70 year olds. Perhaps there is a tendency to cluster. Bandolier was taken by a description of a typical patient in an arthritis clinic: older woman, significantly overweight, trapped in a negative pattern of continual weight gains and pain, accompanied in turn by decreasing activity and functional capacity.

Hands up anyone who is average

Bandolier 95 highlighted the propensity of some people in trials to do well, and some to do not so well. Yet often we are given average results where few if anyone is average. Perhaps trials are being reported better, but this clustering into good or not good responders is something seen more often. It is a challenge to trialists, and especially to folks doing meta-analyses, if the average values they do sums with don't actually mean very much. Bandolier will return to clustering in more detail in future. In the meantime we would welcome any thoughts, insights, or key references that you know of that we have missed.

next story