The FDA issued new guidelines in 1993 (Federal
Register, 1993), perhaps spurred by its own find-
ings in 1989, and confirmed by the General
Accounting Office (GAO), that in only 50% of
submissions were gender analysis discussed in
NDA submissions. Temple (1992) reported that
two FDA surveys demonstrated that women were
included routinely and in proportion to the pre-
sence in the treatment population, and young
women in large numbers (Bushet al., 1993). Not
recorded were his concluding remarks, in which he
said many NDAs did not adequately discuss gender
difference, which would be addressed in the new
amended guideline. The FDA, in its discourse in
the 1993 guidelines,Revised Policy on Inclusion of
Women of Childbearing Potential in Clinical
Trials, mentions that it was swayed by a legal
precedent. In 1991, the US Supreme Court found
on behalf of the plaintiff workers union that their
pregnant members had been unfairly excluded
from jobs by the Johnson Control Company,
because the working conditions exposed their
fetuses to potential risk. The court wrote: ‘Welfare
of future children should be left to the parents...
rather than to employers who hire them’. Although
not quite the same circumstances, the FDA was of
the mind that this opinion would also apply to
pregnant (informed) women, giving them the
right to enter drug trials irrespective of phase of
development.
The FDA revised guidelines on this and ethnic
differences which appeared in July 1993 in the
Federal Register, in essence abolished the prior
ban on women of childbearing age from phase I
and phase II studies, and stipulated additional
topics, including the embryotoxic and teratogenic
risk potential, to be covered in the patients’
informed consent.
Earlier, the NIH had issued its own guidelines to
its staff, grant applicants and academic centers it
supported. It called forallresearch on human sub-
jects concerning drugs, devices, epidemiology,
nondrug device studies and treatment outcomes,
to include both genders and minority representa-
tives whenever possible. In phase III studies,
‘women and minorities and their subpopulations
in sufficient numbers should be included, such that
valid analyses of differences can be accomplished’.
It stipulated that ‘cost was not an acceptable reason
for exclusion, and that programs and support for
outreach efforts to recruit these groups be under-
taken’ (NIH, 1986). Failure to ensure adequate
effort to implement could be reason for grant rejec-
tion or loss of financial support.
To amplify the female view, both the FDA and
NIH during the last decade have appointed women
to significant roles. Dr Bernadette Healy headed
the NIH and created the Office of Research in
Women’s Health; Dr Henny led the FDA until
2001 and within the FDA, Dr Janet Woodcock
and Dr Kathy Zoon were appointed to head
CDER (drugs), and CBER (biologics), respec-
tively, two of the largest centers perhaps partly in
response to an article by LaRosa and Pinn (1993),
both women bemoaning the exclusion of women in
decisions of research.
The industry is now encouraged by the FDA to
include women earlier in the clinical development
program, but there are also still good reasons why
the FDA might deny inclusion of women of child-
bearing potential – insufficient toxicology data; a
disagreement over the interpretation of such data;
agency knowledge of another company’s confiden-
tial data indicating a potential risk with a drug
class-related compound and, finally, an FDA revie-
wer’s individual comfort level with ‘high-risk
population exposure’. Such an event has now
become rarer.Pharmaceutical industry practice
In July 1991, a surveywas completed by this author
for the Pharmaceutical Manufacturers Association
(PMA), Special Populations Committee on the
current practice of the industry in handling
gender and minority data (Edwards, 1991). Vice-
Presidents of headquarters, clinical and regulatory
affairs were contacted at 46 companies; 33 com-
panies responded (nearly all the major companies).
All 33 responding companies collected gender-
related data on the participant patients in clinical
studies. Over three-quarters of the companies
reported that they deliberately recruit ‘representa-
tive’ numbers of women. It should be noted that the
term ‘representative’ has not been defined by the214 CH16 DRUG DEVELOPMENT RESEARCH IN WOMEN