Clinical Site:
Bangor, Maine:
- Marshall-Blum: Clinical Outcomes Specialists (parent company)
Herbal Research Clinic
Independent Medical Research Center
James M. Blum, PhD, Study Coordinator, Epidemiologist and Biostatistician
Medical Director: Peter Marshall, MD
Medical Advisory Team: Ed Kelmenson, MD, and Ron Chicoine, MD
Design:
• Prospective, randomized, double blind placebo-based clinical trial with an additional ‘facet’
• This design was chosen over a traditional crossover design because it was felt that the herbs have the potential of actually restoring the body. Under this theory, subjects randomized to receive the active product initially, would make an inappropriate control.
• The additional ‘facet’ will be for those controls that ‘fail to respond’. If they so desire, they will be crossed over to the active product at the end of their placebo phase. This is an added incentive to those randomized into the control/placebo group, to continue in the study. Occasionally, subjects become agitated and frustrated that their symptoms have not changed and without this incentive, they are lost-to-follow-up. In this manner, we can encourage subjects to complete the control phase. We fully realize that subjects who are crossed-over will not be blinded at that time with respect to their therapy, and thus will represent an ‘open-label’ trial.
• The duration for the initial trial will be approximately one-to-two months on product or placebo. The reason for this ambiguity is that the onset of headaches or migraines is not predictable. Since this product is for the treatment of already existing headaches or migraines, we would need each subject to experience a headache, in order to test the efficacy and safety of the product.
• The ratio of subjects on product to control was two-to-one. Those subjects who complete the control phase and crossover to the product will remain on the actual product for one-to-two additional months.
Product Formulation:
Homeopathic blend, All 3.0x: Feverfew, Goldenseal, Dandelion, Polyporus, distilled water, vegetable glycerin, with potassium sorbate used as a preservative
Product Usage:
• Administer ten (10) full sprays under the tongue and hold for a least thirty (30) seconds before swallowing. Repeat initial dose after five (5) minutes as required. Do NOT use as a preventive measure.
Other instructions included:
• Please report any difficulties with the product or ask any questions you might have as soon as possible.
• Please inform us of any change in your health status.
• Bring your spray bottle with you for each visit.
• Please call the offices to set up an appointment after one migraine, if the product failed, or two migraines, if the product worked.
FOR THOSE TAKING PRESCRIPTIONS, SUCH AS IMITREX:
Use the experimental spray and wait five minutes; if no relief, spray again and wait two minutes. If the spray fails, then please use your Imitrex.
Placebo Product:
The placebo was nearly identical to the active product except that it did not contain any of the active compounds. The spray bottles were identical for both placebo and product and the liquid appeared to be the same for both. We had no reports of subjects suspecting that the placebo was the inactive spray.
Inclusion Criteria:
• Men or women with tension, migraine, or other types of headaches
• Subjects could do the study in conjunction with prescription medication
• Ages 18 - 65
• Subjects who passed a compliance screening test
• Subjects who were able to tolerate the active product and placebo
• Subjects who signed a consent form and passed all compliance tests
Exclusion Criteria:
• Individuals with organic neurological illness
• Subjects who were non-compliant with testing and treatment regimens
• Subjects who had problems with the treatment herbs
• Subjects under 18 or over the age of 65
• Subjects with moderately severe co-morbid disease, that includes cardiac, pulmonary, renal, hepatic, or active cancer (this determination is subject to the study physician)
Confounding Factors:
• Gender
• Type of Headaches (migraine, cluster, tension, sinus, food-sensitivity (MSG), altitude, etc.)
• Co-morbid conditions
• History of trauma (e.g., accidents, head injuries, and the like)
Protocol:
Subjects were recruited from various physician practices and from our general population. They met the inclusionary and exclusionary criteria and passed various compliance tests. Subjects were dropped during the study if they repeatedly failed the compliance standards. Subjects were required to sign an informed consent, subject to the criteria set forth from the Fox’s Human Right’s Committee. The Study Coordinator instructed all subjects as to the protocol and other details of the study. Subjects were instructed precisely how and when to take the spray and how to report adverse reactions, complications of their disease, and any other pertinent information. Subjects were randomized to either receive the actual MigraSpray or placebo in the first phase of the study. For every subject who is randomized to receive the placebo spray, two will be randomized to receive the herbal preparation. Potential subjects were interviewed and screened into the study at the Marshall-Blum clinic. They received instructions on how and when to complete the survey instruments and be given one bottle of spray (actual product or placebo). Any adverse events were reported to NatureWell and to Fox Commercial IRB using standard reporting format.
Statistical Analysis:
Randomization was performed using a 9-patient block design using standard randomization schemes to ensure proper randomization. All data was entered into a Microsoft Access database, while all statistical analyses were performed using the PC version of Statistical Analysis System (SAS System of Cary, North Carolina). Baseline characteristics were compared using Fisher’s exact test chi-square test for categorical parameters and the two-sample t-test for continuous parameters. Statistical significance was set at five percent (or 0.05). Outcomes were analyzed using chi-square tests (categorical data) and standard t-tests (continuous data). Statistical significance will be established that a trend will be between 0.051 and 0.10 while true significance will be defined as equal to or less that 0.05.
Primary End-Points:
• Elimination of symptoms (pain)
• Reduction of symptoms
• No change of symptoms
• Time required to reduce or eliminate symptoms Secondary End-Points:
• Blood Pressure
• QoL Survey Tool
Analytical Methods:
Methods:
• The primary end-point was the self-report headache severity score, which was rated from 1 to 5 (5 being the worse headache imaginable)
• The data from this scale were collected during a baseline headache and during a headache while assessing the treatment spray (product or placebo)
• A difference variable was created by subtracting these two data points
• Differences from baseline were analyzed alone and as a percentage from baseline
Two Sets of Data:
• Set 1 includes only the initial phase; it does not include the placebo cross-overs
• Set 2 includes the original placebo subjects who crossed-over to take the product
Type 1: Difference in Means:
• The means for the difference variable for each group (placebo or active product) were calculated
• Compared statistically using Fisher’s Exact Two-Tail t-test
Type 2: Percent Improved:
• The percentage of those subjects who showed an improvement over baseline was calculated by placing subjects into various categories: any improvement (> 0% improvement), slight (> 10% improvement), full (> 30% improvement), and complete (> 45% improvement)
• All categories were analyzed using the Chi-Square test or Fisher’s Exact Two-Tail t-test (due to the small cell limitations)
• These category improvements were determined a priori by the medical advisory group
RESULTS:
Baseline Data:
• There were no differences in any of the baseline data between the treatment and placebo groups; this indicates that the two g roups were medically equal
• These parameters included: age categories, weight, diabetes, hypertension, back or pelvic injury, cancer, daily caffeine intake, weekly alcoholic intake, weekly exercise, employment, household income, adults per household, education, smoking, self-report general health, use of alternative therapies or herbal supplements, cardiac, liver, pulmonary, and renal disease
• There was no statistical difference between the groups regarding the baseline migraine severity scale (3.9 vs. 3.5: placebo vs. product groups)
• Drops and Adverse Events:
Only one patient in this study reported an adverse reaction. She stated that the product spray actually made her migraine worse in nature. She was instructed to stop using the product. Her subsequent migraines were no worse than previous episodes.
Difference Between the Means:
• The efficacy of MigraSpray was clearly demonstrated by the difference in means for the overall group:
Migraine Severity:
Product Group Improvement: 1.22 points
Placebo Group Improvement: 0.12 points
P < 0.00001
Migraine Severity:
Product Group Improvement: 36%
Placebo Group Improvement: 4%
P < 0.00001
Overall Percent Improvement (defined by the percentage of subjects in each group):
All Levels of Improvement: (Chi-Square Tests):
Product Group Improvement: 87.8%
Placebo Group Improvement: 30%
P < 0.00001
Full-to-Complete Level of Improvement:
Product Group Improvement: 65.9%
Placebo Group Improvement: 0%
P < 0.00015
Time to Relief:
Product Group: 6.38 minutes on average
Conclusions:
The efficacy of MigraSpray has been clearly demonstrated in this randomized, placebo-controlled trial for migraine headache relief drawn from a general population, reporting migraine headaches. The strength of this trial includes the design (baseline headache measurements, specific protocols, randomization, placebo-controlled, blinding, nursing assessments, cross-over’s), outside review, drawn from a general population, extremely high safety record, equal baseline characteristics between the two groups (placebo and product), and statistical significance in all the major end-point categories. The results of this study are statistically significant, at a P value of less than 0.01 in all parameters.
Statistical References:
Clinical Trials
Meinert, Curtis
Oxford University Press 1986
Practical Statistics for Medical Research
Altman, Douglas G.
Chapman and Hall, London 1993 pp. 181-191 & 232-250
Medical Uses of Statistics, Second Edition
Edited by: John C. Bailar III and Frederick Mosteller
NEJM Books (New England Journal of Medicine)
Boston, MASS 1992 Chapter 10: P Values
|
 |
