RESEARCH - Narrow eligibility criteria threaten applicability of clinical trial results

[Guest guest]

Narrow Eligibility Criteria Threaten Applicability of Clinical Trial Results

NEW YORK (Reuters Health) Mar 21 - Randomized controlled trials are

essential for effective medical policy decisions and for maintaining

up-to-date medical practice, but the applicability of the results is

threatened when eligibility criteria are too strict, exclusive, or

inadequately defined, researchers report in the Journal of the American

Medical Association for March 21.

Dr. A. Fowler, at the University of Toronto, and his associates note

that women, children, elderly individuals, ethnic groups, patients with

chronic medical illnesses, and those taking other medications have often

been excluded from pivotal trials. Other reasons for being barred from a

trial have included socioeconomic status, language or communication barrier,

participation in other trials, and ethnicity.

To evaluate exclusion criteria among randomized controlled trials published

in nine major medical journals, Dr. Fowler's team searched MEDLINE for

trials published between 1994 and 2006. Of 4827 articles, 283 were chosen

for analysis using a series selection technique.

They found that exclusion criteria were omitted in 12% of trial reports. Of

all exclusion criteria, the authors graded only 47% as strongly justified.

" Reasons for exclusions were frequently not justified in the context of

individual randomized controlled trials, " the authors report, particularly

in multicenter drug evaluations and trials sponsored by industry.

Dr. Fowler and his team " recommend that every reasonable effort be made to

minimize exclusions of specific patient populations when such patients will

likely form a group to which the results are generalized. "

For exclusions deemed necessary, they advise that they be described by

objective criteria and clearly listed within a table apparent to the reader.

JAMA 2007;297:1233-1240.

http://www.medscape.com/viewarticle/553981

Not an MD