- G Bugeja, specialist registrara,
- A Kumar, specialist registrara,
- Arup K Banerjee, consultant physiciana
- Correspondence to: Dr Banerjee
- Accepted 7 August 1997
Conclusions reached in studies of young people cannot be extrapolated to elderly people.1 It is thus essential that elderly people are included in clinical studies. We set out to discover whether and to what extent studies published in medical journals exclude this section of the population.
Methods and results
We examined all the original research papers in all the issues of the BMJ, Gut, the Lancet, and Thorax between 1 June 1996 and 1 June 1997. We excluded case studies and meta-analyses. We looked for a mention of age limit first in the abstract or the methods section of the papers, but we sometimes found it mentioned in other parts of the papers instead. We then categorised the papers as studies (a) specifically of elderly people (aged 75 years or more); (b) excluding elderly people for justifiable reasons (where long follow up was planned);(c) excluding elderly people with no justifiable reasons; and (d) not setting an age limit. We excluded papers that were concerned with children and adolescents; pregnancy and reproduction;sexually transmitted diseases; animal work; cell and tissue studies;or employment related research.
The results are given in the 1. We found 1012 papers and excluded 522. Of the remaining 490 papers, 18 were specifically about elderly people, 37 excluded the elderly for justifiable reasons, 170 excluded them unjustifiably, and in 265 no age limit was set.
A third of the original research papers in major medical journals excluded elderly people without justification. Although the journals were not selected randomly and we included a small number of papers published within a short period of time, we believe that our findings are important and relevant to current practice and behaviour in clinical research.
Elderly people may be excluded from clinical research for many reasons. Often the decision to participate in research cannot be taken by the elderly person alone. Family members and caregivers take part in the decision to participate in the study, and the investigator has to establish rapport with both the patient and family members. This is usually a time consuming and complicated task. Many investigators may avoid these problems by excluding elderly people from their study.2 Or they may exclude them because they fear that frailty and comorbid conditions may put elderly people at increased risk from the study, but we believe that such risks are best investigated in the rigorous setting of a clinical trial. Selected elderly patients can enter trials without being put at an increased risk of more severe or frequent side effects.3
Applying the results of research that excluded elderly people to the management of elderly people is, we believe, inappropriate and possibly dangerous. More elderly people should be included in clinical trials. Study populations should be stratified to include a sufficient number of elderly people to permit valid conclusions on effects on elderly people. If elderly people are excluded from any research work then this must be made clear to readers of the paper and the reasons for such exclusion justified.
Conflict of interest: None.