|
1
|
|
|
2
|
- A health concern gets substantial public attention
- Pressure builds from all quarters to reach a policy decision
- High quality RCT data are not available to conclusively answer the
question
- What to do? The example of ephedra
|
|
3
|
|
|
4
|
|
|
5
|
- In China: used as part of a
concoction for a variety of symptoms
- In Europe: approved in Germany
to treat asthma and congestion
- In USA:
- To lose weight
- To improve athletic performance
|
|
6
|
- “FDA targets herbal product”
(Financial Times, London, 8/19/2002)
- “Petition urges U.S. to ban supplements with ephedra” (NYT, 9/6/2001)
- “Ephedra is now under intense scrutiny” (LA Times, 8/2/2002)
- “For young athletes, limited gains--at a high risk (Washington Post,
4/9/02)
- “Ephedrine: safe or lethal?” (USA Today, 11/8/2001)
- Public Citizen asks Secretary of HHS to ban ephedra as a threat to
public health
- Ephedra is an “imminent hazard…”
Senator Durbin
|
|
7
|
- HHS Secretary Thompson announces that RAND will conduct comprehensive
review of the existing science on ephedrine alkaloids (6/14/2002)
- “The results of the RAND review
will help FDA's scientists
to develop future regulatory actions on dietary supplements containing
ephedrine alkaloids.”
Lester M. Crawford, DVM, PhD. Acting Commissioner, FDA (10/8/2002)
- AHRQ, with funding from Office of Dietary Supplements and National
Center for Complementary and Alternative Medicine, sponsor an Evidence
Report by the Southern California Evidence Based Practice Center at
RAND.
|
|
8
|
|
|
9
|
|
|
10
|
|
|
11
|
- Ephedra alone and in combination
is associated with weight loss
- No long-term studies of effectiveness
- Few head-to-head comparisons with alternative therapies
- Ephedra likely to be less effective than weight loss drugs
|
|
12
|
- 8 studies identified
- None on Ephedra
- 6 by Bell and colleagues
- All studied particular interventions and outcomes, and could not be
pooled statistically
- All studied small numbers of young, fit males and assessed immediate
performance
- Studies report improved performance only for ephedrine + caffeine
|
|
13
|
- Does ephedra use have adverse effects?
- Does using ephedra increase the risk of heart disease or other serious
and life-threatening events?
|
|
14
|
|
|
15
|
- No deaths, heart attacks, or other serious adverse events reported in
clinical trials
- Clinical trials studied only a few thousand people
- This number is too small to detect a rate of adverse events smaller
than 1 per 1,000
|
|
16
|
|
|
17
|
|
|
18
|
- Sentinel Event
- Documentation of the adverse event
- Documentation that ephedra or ephedrine was consumed within the prior
24 hours (by report or toxicology examination)
- Documentation that other potential causes of the adverse event were
evaluated and excluded
- Possible Sentinel Event
- Documentation of the adverse event
- Documentation that ephedra or ephedrine was consumed within the prior
24 hours (by report or toxicology examination)
- Other potential causes exist or were not effectively excluded
|
|
19
|
|
|
20
|
|
|
21
|
- Neither side - the critics or the supporters - is completely right or
wrong
- Ephedra does promote short-term weight loss
- But long-term effects are unknown
- Ephedrine plus caffeine does have modest short-term effects on athletic
performance
- After use of a single dose
- Ephedra is associated with some side effects
- Upper GI symptoms, autonomic hyperactivity, psychiatric symptoms,
palpitations
|
|
22
|
- Ephedrine acts on receptors in the cardiovascular system and in the
brain
- Chemically related compounds have been proven to cause these same
adverse events
- Phenylpropanolamine, amphetamines
- The sentinel events we identified are a signal that ephedra may also
cause serious adverse events
- Recent case-control study reported an increased risk for hemorrhagic
stroke with ephedra use
- Adjusted odds ratio of 3.95 (95% CI: 0.7-18.0; p=0.07)
|
|
23
|
- Clinical trials
- Usual limitations for clinical trial meta-analysis
- No deaths or other serious adverse events reported but insufficient
sample size to detect a rate of adverse events smaller than 1 per 1000
- Case report adverse event data
- Under-reporting?
- Many reports did not contain sufficient documentation
- No estimate of the number of people using ephedra
- Study design considered insufficient to warrant conclusions regarding
causality
- In order to assess safety at the conventional threshold for statistical
significance and inference, a definitive hypothesis-testing study is
needed.
|
|
24
|
- There is no provable certainty; therefore, the question is what are the
ramifications of a policy decision or non-decision, and how much
uncertainty can we tolerate?
- In the case of ephedra, the degree of uncertainty regarding a causal
relationship between use and serious adverse events did not meet the
conventional “p<0.05” threshold.
|
|
25
|
- The time needed to conduct a hypothesis-testing study of ephedra would
be 2-5 years.
- If ephedra is causally related to an increased risk of serious adverse
events, then during the time period needed to conduct a study some
number of ephedra users would succumb.
- If ephedra is not causally related to an increased risk of serious
adverse events, then a decision to withdraw it from the market would
deny some potential ephedra users the possibility of weight loss
benefits we observed.
|
|
26
|
- Policy decision was made to withdraw ephedra from the market.
|
|
27
|
|
|
28
|
|