|
Editorial JULY 2006
The Practice of
Evidence-based
Medicine
Dorothy Kammerer-Doak, MD
The practice of medicine is both an art and a science. One cannot
discount years of experience and patient interaction and rely solely
on science. Additionally, there is often no concrete scientific
evidence on which to base medical practice. However, a critical,
current review of the literature should guide how physicians care
for their patients.
The randomized, controlled trial (RCT) has been touted as the “gold
standard” for comparing two medical or surgical treatments
regarding both outcomes and complications. During my relatively
short time as a physician in obstetrics and gynecology (with a
subspecialty in urogynecology), I have seen changes in the way
that the specialty is practiced. When I was in training, one of
the standard, accepted surgical procedures for treating stress
urinary incontinence (SUI) was the modified Kelly-Kennedy plication/anterior
repair. Multiple randomized trials compared this procedure with
Burch retropubic urethropexy (RPU), and uniformly demonstrated
the superior success rate of RPU over anterior repair. Thus, anterior
repair is no longer a first-line surgical treatment for SUI. More
recently, RCTs have compared midurethral sling procedures such
as tension-free vaginal tape (TVT) with the gold-standard Burch
RPU. These trials have demonstrated similar success rates, but
decreased complications with TVT. So today, the Burch RPU and TVT
have been critically evaluated using the scientific gold-standard
RCT, and have become the standard of care for the surgical treatment
of SUI.
Furthermore, there has been a plethora of “new” procedures
utilizing various graft materials and “kits” to treat
both urinary prolapse and incontinence. As a result, many physicians
have adopted them into routine use with little scientific evidence
to evaluate their success and complications compared with the standard
surgical procedures. It is not yet known whether the use of grafts
to augment anterior or posterior repairs yields higher or lower
success and complication rates than routine repair without grafting,
or whether a TVT sling is more or less successful than a transobturator
tape sling, as there have been few RCTs to compare outcomes. There
are many observational studies, but few “head-to-head” comparisons
to guide practice. Similarly, there are no RCTs comparing different
techniques of vaginal apical suspension with and without grafting.
Physicians cannot know which surgical practices will produce
optimal outcomes, in terms of success or complication rates, until
new procedures are randomly compared with current surgical practices.
Therefore, as a medical community, we need to encourage the companies
developing these new technologies to fund RCTs so that we will
have the science we require to practice evidence-based medicine.
New technology is not always better, as demonstrated by the introduction
of continuous fetal-heart rate monitoring: Its use became the standard
of care without any evidence to prove better fetal outcomes compared
with intermittent fetal auscultation. When RCTs were finally conducted
to compare the two techniques, it was found that continuous electronic
monitoring conferred a limited benefit, but at the expense of increased
operative deliveries.
However, RCTs are not without fault, and are subject to bias
depending on their design. The Women’s Health Initiative
postmenopausal hormone therapy (HT) trial and the Term Breech Trial
are examples.1,2 Both were flawed in their methodology, and consequently
their conclusions may not be completely valid. In the HT trial,
scientific scrutiny of the methodology demonstrates inadequate
participation by minorities, an overrepresentation of older women
remote from menopause, and the inclusion of unhealthy women with
preexisting illnesses, indicating that the results may not be applicable
to healthy, recently postmenopausal women or to different ethnic
groups. The Term Breech Trial randomized subjects to either planned
vaginal or cesarean delivery. Review of the data notes many problems,
such as the inclusion of inappropriate subjects (eg, twins, stillbirths,
infants with congenital anomalies), failure to randomize eligible
patients, unavailability of clinicians experienced in breech vaginal
delivery, and inclusion of perinatal mortality unrelated to mode
of delivery. In fact, a review of outcomes at 2 years noted no
differences between the two groups.
In conclusion, the science of the practice of medicine may be
imperfect, but it is instrumental to furthering our knowledge and
providing the best outcomes for our patients. Physicians must remain
critical about the introduction of new practices, and not abandon
the traditional treatments without careful study of risks versus
benefits. To practice the highest quality of medicine, we must
maintain the delicate balance between art and science.
Dorothy Kammerer-Doak, MD
Editorial Advisory
Board Member
back
to top
References
- Rossouw JE, Anderson GL, Prentice RL, et al.
Risks and benefits of estrogen plus progestin in healthy postmenopausal
women: principal results from the Women’s Health Initiative
randomized controlled trial. JAMA. 2002;288(3):321-333.
- Hannah ME, Hannah WJ, Hodnett ED, et al. Outcomes at 3 months after planned cesarean vs planned vaginal delivery for breech presentation at term: the international randomized Term Breech Trial. JAMA. 2002; 287(14):1822-1831.
back
to top
|
