Assignment 1 - Sample Practice Critique “Gold Standard”

 

Article Reference:

Stoll AL, Severus WE, Freeman MP et al. Omega 3 Fatty Acids in Bipolar Disorder: A Preliminary Double-blind, Placebo-Controlled Trial.  Arch Gen Psychiatry 1999; 56: 407-412.   OVID Accession Number - 00062674-200000000-00304.AN

 

Research Question:

What are the subacute mood-stabilizing effects of w3 fatty acids in patients with unstable bipolar disorder?

 

Variables:

The independent variable was categorical with two nominal categories: w3 fatty acid treatment or placebo.  The w3 fatty acid treatment was seven capsules/day of 440 mg of eicosapentanoic acid and 240 mg docosahexanoic acid.  The placebo group received seven capsules containing olive oil.  The study was designed as double-blind in that neither the subjects nor clinicians were aware of which group as subject was assigned.

 

The dependent variables (outcome measures) were categorized by the researchers as primary and secondary outcome measures.  The primary outcome measure was duration of time in the study.  The secondary outcome measures were scores on Young Mania Rating Scale, Hamilton Rating Scale on Depression, Clinical Global Impression Scale, and Global Assessment Scale.  The instruments are described as standard instruments; presumably reliable and valid (references were given).  Finally, adverse effects were assessed - apparently by clinical observation.

 

Control variables included gender, the presence or absence of concurrent lithium treatment, and the presence or absence of rapid cycling (of the manic and hypomanic phases).  These variables were controlled by stratification in that patients in each stratum or category of the variable were randomly assigned to either treatment or control groups in a balanced manner.  For example, for the gender variable, equal numbers of men and women were assigned to treatment or placebo.

 

An extraneous variable which possibly could have affected the results, was a “fishy” aftertaste reported by subjects, more often for the w3 group.  Thus, a potential placebo effect, though small, did exist.

 

Hypotheses:

The researchers listed these hypotheses:

1.      Orally administered w3 fatty acids would exhibit inhibitory effects on signal transduction in human neuronal membranes.  This hypothesis was not directly tested in the study.

2.      High-dose w3 fatty acids would stabilize mood in bipolar disorder. 

This hypothesis can be operationally stated (in two statements) as:

A.                The w3 fatty-acid group will stay in remission longer than the placebo group (directional alternative).

B.         The w3 fatty-acid group will show more positive scores than the placebo group on the measures of mood: Young Mania Rating Scale, Hamilton Rating Scale on Depression, Clinical Global Impression Scale, and Global Assessment Scale (directional alternative).

 

Research Design:

The design was described as a parallel-group, placebo-controlled double-blind pilot study.  Based upon the study hypothesis, variables, and methods, the study design can be classified using the approach to understanding research designs presented in The Whole Art of Deduction.  The research question is a differences question.  The design is an elaboration of the Pre/Post w/Control Group. 

 

 

The study design is diagrammed as:

 

 

The study design is more complex than the diagram indicates in that the subjects were assessed at weeks 2, 4, 6, 8, 12 and 16.  Thus, the design could be diagrammed as:

 

 

A possible threat to the internal validity of the study was  a possible placebo effect in that the fishy aftertaste of the w3 fatty-acid capsules may have allowed those subjects to determine they were in the treatment group.  However, inherent in any treatment study is that as subjects become aware of improvement in their condition, they are likely to believe they are in the treatment group.

 

Population:

The population of interest was adult (age 18 to 65 years) men and women who met DSM-IV criteria for bipolar disorder (types I or II) and were free of notable psychiatric or medical comorbidity.  Patients were required to have at least one manic or hypomanic within the past year.  The sample was 30 patients recruited at Brigham and Women's Hospital, Boston, MA and Baylor College of Medicine, Houston, TX.  This is termed a sample of convenience, not a random sample. These patients, volunteering to participate in the study, may not be representative of all adults with bipolar disorder.  In their commentary Calabrese, Rapport & Shelton describe a potential threat to the external validity of the study.  It was not a typical acute or maintenance study: the generalizibility of the results to more seriously ill patients or their long-term remission is in question..

 

 

Data Gathered:

Analyzable data were obtained for the 30 subjects who met the criteria of at least 30 days participation in the study.  Comparison of the duration of time in the study (also referred to as duration of remission) between the two groups was analyzed using the Kaplan-Meier survival analysis (Mantel-Cox log-rank statistic).  Categorical data were analyzed with the Fisher exact test.  Continuous data were analyzed with the Mann-Whitney test.  Level of significance a was set at .01. 

 

The w3 fatty-acid group had a significantly longer period of remission than the placebo group (Kaplan-Meier analysis). The w3 fatty-acid group performed significantly better than placebo on the Clinical Global Impression Scale and the Hamilton Rating Scale on Depression.  The groups did not differ significantly on the Young Mania Rating Scale, Global Assessment Scale or adverse effects. 

 

Results:

The authors concluded that w3 fatty-acid treatment improved the short-term course of bipolar disorder.  The planned interim analysis (after 20 patients) showed significant differences leading to an early termination of the study after 30 patients completed.  The researchers do describe the study as preliminary – it did not permit an evaluation of the anitmanic effects of w3 fatty acids.  The researchers also acknowledged the possible placebo effect of the fishy aftertaste of the w3 fatty-acid capsules.  As stated above, in their commentary Calabrese, Rapport & Shelton state that the present study was not a typical acute or maintenance study thereby limiting the generalizibility of the results to more seriously ill patients or their long-term remission.  They also question the use of the Kaplan-Meier survival analysis. The authors acknowledge the criticisms but believe that the preliminary study provides a basis more larger, long-term studies.