Docstoc

Journal Club

Document Sample
Journal Club Powered By Docstoc
					     Journal Club

Alcohol and Health: Current Evidence
         July-August 2006



            www.alcoholandhealth.org   1
             Featured Article

      Effect of oral acamprosate on
   abstinence in patients with alcohol
     dependence in a double-blind,
   placebo-controlled trial: the role of
            patient motivation

Mason BJ, et al. J Psychiatr Res. 2006;40(5):382–392.

                    www.alcoholandhealth.org       2
         Study Objective

To evaluate the…

• efficacy of acamprosate to maintain
  abstinence in U.S. patients with alcohol
  dependence




               www.alcoholandhealth.org      3
                 Study Design
• First randomized controlled trial of acamprosate in the
  United States

• Occurred in 21 alcohol treatment clinics

• Subjects, recruited primarily by newspaper ads, were
  assigned to receive daily…
   – 2 g of acamprosate (n=258),
   – 3 g of acamprosate (n=83), or
   – placebo (n=260)

• Also received self-help materials and 8 sessions of brief
  counseling

                      www.alcoholandhealth.org              4
    Are the Results of the Study Valid?
Primary Guides:

• Was the assignment of patients to treatments randomized?
  Yes, and there were no statistically significant differences
  between groups at baseline.

• Were all patients who entered the trial properly accounted
  for and attributed at its conclusion? Yes.

• Was follow-up complete? 99% of subjects provided data
  for efficacy analyses; 13% were lost to follow-up, and 51%
  of subjects withdrew from the study.

• Were patients analyzed in the groups to which they were
  randomized? Yes, for the main analysis.
                       www.alcoholandhealth.org            5
Are the Results of the Study Valid? (cont.)

 Secondary Guides:

 • Were patients, health workers, and study personnel
   "blind" to treatment? Yes, patients and study personnel
   were blinded to treatment assignment.

 • Were the groups similar at the start of the trial? Yes.

 • Aside from the experimental intervention, were the
   groups treated equally? Yes.




                       www.alcoholandhealth.org              6
         What Were the Results?
   -How large was the treatment effect?
   -How precise was the estimate of the
            treatment effect?
The groups did not significantly differ in unadjusted
analyses. The study sample size was calculated to detect
an effect size of 0.30.

              Group                              % Days Abstinent
                                                     Overall
Placebo                                                 54
2 g of acamprosate                                      56
3 g of acamprosate                                      61

                      www.alcoholandhealth.org                      7
       What Were the Results? (cont.)
• The researchers decided to perform post-hoc analyses
  adjusted for factors associated with treatment
  outcomes including…
   – baseline goal of total abstinence,
   – alcoholism severity,
   – stage of readiness to change,
   – psychological antecedents (e.g., psychiatric
     hospitalizations or suicide attempts),
   – addiction severity,
   – treatment exposure, and
   – study site.

                     www.alcoholandhealth.org        8
     What Were the Results? (cont).
              Group                        Adjusted % Days
                                              Abstinent

    Placebo                                         52
    2 g of acamprosate                              58
    3 g of acamprosate                              63

  P=0.04 for 2g vs placebo
  P=0.01 for linear trend across the 3 groups

• The adjusted analysis was then repeated in subjects
  committed to total abstinence (n=115 for placebo; n=100 for
  2 g; n=26 for 3 g).

                         www.alcoholandhealth.org            9
      Guidelines for Interpreting
         Subgroup Analyses
• Is the effect suggested by comparisons within rather than
  between studies?

• Did the hypothesis precede rather than follow the analysis?

• Was the subgroup effect one of a small number of hypothesized
  effects tested?

• Is the magnitude of the effect large?

• Was the effect statistically significant?
• Is the interaction consistent across studies?

• Is there indirect evidence that supports the hypothesized
  interaction?
                           www.alcoholandhealth.org             10
Is the effect suggested by comparisons
 within rather than between studies?


 • The analysis compared randomized groups
   within a single study.

 • It directly compared 3 groups of subjects
   committed to abstinence.




                  www.alcoholandhealth.org     11
Did the hypothesis precede rather than
         follow the analysis?


• The hypothesis followed the analysis.

  – After finding that most patients did not have
    abstinence as a treatment goal, the researchers
    chose to examine results from the 241 patients
    with this goal.



                   www.alcoholandhealth.org       12
Was the subgroup effect one of a small
number of hypothesized effects tested?

 The authors do not state explicitly the number
  of subgroup analyses performed.

 However, they reported only 1 subgroup
  analysis.




                  www.alcoholandhealth.org     13
Is the magnitude of the effect large?
• The magnitude of the effect was large. There was an
  absolute increase of 12% in percent days abstinent.

                   Group                       % Days Abstinent Among
                                                  Subjects With an
                                                  Abstinence Goal
     Placebo                                                     58

     2 g of acamprosate                                          70

     3 g of acamprosate                                          73


 Analyses were adjusted for baseline variables and treatment exposure
                                   www.alcoholandhealth.org             14
Was the effect statistically significant?


  • Yes, the P value for the comparison of 2 g
    of acamprosate vs placebo was 0.02.

  • Similarly, the P value for the linear trend in
    abstinence across the 3 groups was 0.02.




                   www.alcoholandhealth.org          15
 Is the interaction consistent across
               studies?

• A meta-analysis of international studies found
  that acamprosate improves odds of
  abstinence.

  – Patients in these studies were usually abstinent for
    5 days and entered the studies directly from
    inpatient detoxification.

  – These studies are consistent with others that show
    acamprosate has efficacy in patients with more
    stable initiation of abstinence.

                    www.alcoholandhealth.org          16
  Is the interaction consistent across
             studies? (cont.)

• Another U.S. study (Anton et al, 2006) that did
  not require substantial abstinence as an
  eligibility criterion also found no efficacy for
  acamprosate.

• Two other studies involving patients who were
  actively drinking reported no efficacy for
  acamprosate.
                   www.alcoholandhealth.org      17
Is there indirect evidence that supports
     the hypothesized interaction?

• The effect was consistent with previous
  studies of acamprosate.

• The effect was also present for other
  outcomes, such as GGT levels, abstinence
  >=90% of days, and abstinence <=10% of
  days.


                  www.alcoholandhealth.org   18
Is there indirect evidence that supports
 the hypothesized interaction? (cont.)
• Brain hyperexcitability during alcohol withdrawal and
  protracted abstinence is hypothesized to be a particularly
  sensitive target for acamprosate’s therapeutic actions (al
  Qatari et al., 1998; Spanagel and Zieglgansberger, 1997).

• The drug appears to act primarily by restoring normal N-
  methyl-D-aspartate (NMDA) receptor tone in the
  glutamate system (Spanagel and Zieglgansberger, 1997).

• Acamprosate has been shown to have a specific dose-
  dependent effect on decreasing excessive alcohol intake
  in animal models of alcohol dependence.

                      www.alcoholandhealth.org          19
                   Summary
• In a 3-group randomized controlled trial of
  acamprosate, the drug was not efficacious.

• Investigators noticed that…
  – their subjects had not been required to abstain before
    study entry (unlike in most previous studies), and
  – most did not have abstinence as a goal.

• Investigators performed adjusted subgroup
  analyses and found that acamprosate had
  efficacy in subjects committed to abstinence.
                     www.alcoholandhealth.org          20
              Summary (cont.)
• Although caution should be used when interpreting these
  post-hoc subgroup analyses…
   – the findings are biologically plausible, consistent with
     previous studies, and from a direct within study
     comparison, and

   – the magnitude of the effect is large and statistically
     significant (although whether other subgroup analyses
     were tested is unclear).

• Although confirmation would be useful, it is likely that
  acamprosate has efficacy in patients motivated to be
  abstinent.

                       www.alcoholandhealth.org              21

				
DOCUMENT INFO
Categories:
Tags:
Stats:
views:0
posted:3/7/2012
language:
pages:21