1.

DATA ANALYSIS AND STATISTICAL METHODS

1.1

ENROLLMENT

Subjects meeting all of the inclusion criteria and none of the exclusion criteria will be enrolled in
the study.
After the subject has completed the online screening visit (V1) and the telephone screening visit
(V1.5) and has met all of the inclusion criteria and none of the exclusion criteria, the subject will
be assigned the next available enrollment ID.
1.2

DATA MANAGEMENT

The monitoring team will ensure that the subjects are compliant. Parallel dual data reconstruction
and transformation will be done by data management personnel across all endpoints. Data
validation and reconciliation of parallel transformation will occur after the dual data
transformation process. Data files will be password-protected by the data management team for
data integrity and confidentiality purposes prior to any analyses made by the Biostatisticians.
1.3

EXAMINATION OF DATA

All data elements will be screened for reasonableness, and all missing, suspicious, or impossible
values will be referred back to the monitoring team for query generation and resolution. All
numerical variables will be tested for normality, and data found to be substantially non-normally
distributed will be analyzed by appropriate non-parametric methods.
1.4

DATA LOCK AND UNBLINDING

After all suspicious entries in the database have been resolved; a data lock form will be signed by
the Head of Scientific Data Management and Analysis team to formally lock the database.
Locked database will be free from any changes on the data sources and data entries. Data unlock
form should be issued and signed if there will be changes needed to reflect or integrate on the
database.

1.5

DESCRIPTIVE SUMMARIZATION

All variables under investigation will be summarized by time point. Endpoints in

interval/ratioscale will be presented as (n, mean, standard deviation and standard error).
Numerical variables will also be presented graphically, as plots of average value versus time.
For each endpoint in at least ordinal scale, the differences between time periods for each product
group will be tested for nominal significance using non-parametric test (Wilcoxon Signed Rank
Test or Sign Test). Difference in the distribution between arms will be tested using nonparametric Chi-square Test.
For those numerical endpoints that were found to have normally distributed data, comparison
between visitsproducts will be assessed using Independent t-test for group differences at each
time point and paired sample t-test will be used to assess within-group changes from baseline to
each subsequent time points. For non-normally distributed numerical endpoints, Wilcoxon
Mann-Whitney test, for comparison between products visits to assess between group differences
at each time point and Wilcoxon Signed Ranks test or Sign test will be used to assess withingroup changes from baseline to each subsequent time points.
For those numerical endpoints that were found to have normally distributed data, comparison
from Baseline visit to subsequent timepoints will be assessed using Paired Samples T-test. For
non-normally distributed numerical endpoints, Wilcoxon Signed Ranks Test or Sign Test will be
used to assess changes from baseline to other time points.
All tests of hypotheses will be done at alpha=0.05.
1.6

COMPLIANCE CALCULATION

Compliance will be calculated, as follows:

% Compliance = (Number of study product consumed/Number of days since last visit) x 100
Compliance will be adjusted if dose titration is used.
1.7

STATISTICAL HYPOTHESES

The null hypotheses tested for this study are the following:

There were are no significant changes in the VAS Pain and Soreness Scale, Stanford
Exercise Behavior Scale, and Self-Reported questionnaire from Baseline to other time
points.

There were are no significant changes in the POMS questionnaire from Baseline to other
time points.

There are no significant changes in the number of Adverse Events from Baseline to other
time points.

1.8

STATISTICAL ANALYSIS

Statistical Software for Social Sciences (SPSS version 19) will be used to run all descriptive and
inferential analyses for all endpoints.
Efficacy Analysis
A Modified per Intend to Treat (Mod ITT) analysis will be performed to assess the efficacy
variables of the study. Subjects with at least one post-dose visit completed will be included
in the analysis.
The primary objectives of this study are; 1.) to assess the efficacy of the Statinzyme Study
Product on providing relief of muscle aches and discomfort and on 2.) assess the efficacy
of the Statinzyme Study Product on improving memory and mood.

All numeric/continuous efficacy variables will be tested for normality and will be analyzed
by Paired Samples T-test to assess if significant changes from baseline to other time points
exist. Non-normally distributed data will be tested using Wilcoxon Signed Ranks Test or
Sign Test.
All numeric/continuous efficacy variables will be tested for normality and will be analyzed
by Analysis of Covariance (ANCOVA). In the analysis, the value of the efficacy variable at
every time point will be modeled as a function of the treatment group (predictor variable of
interest) and of the value of that efficacy variable at Baseline (covariate). The analysis will
result to significant efficacy if the coefficient of the treatment group variable is
significantly different from zero and in the right direction.
The ANCOVA approach is used to mathematically compensate for the subjects baseline
characteristics that happen to be substantially unbalanced between the two treatment
groups. It is more efficient than the simpler Student t-test since it adjusts for possible
situation like regression to the mean and floor effects.
Data transformation will be used to utilize ANCOVA approach for residuals of the raw data
that were found to be substantially non-normally distributed.
Safety Analysis
To obtain comparable documentation on AEs, the investigator will ask the subject a set of
open and standardized questions at each visit. Frequency and intensity of AEs and serious
AEs will be recorded in detail, based on the subjects interviews during each visit.
Recorded AEs will be grouped by general type of event. Differences in AE patterns
between product groups will be assessed by Fisher Exact test.
1.9

CONTROL OF INFERENCE ERRORS

Hypothesis testing for each of the efficacy endpoints under investigation will be tested at
alpha=0.05. In terms of interpreting statistical results, for each specific endpoint, there is a 5%
chance of rejecting the null hypothesis when in fact the null hypothesis (no significant
difference) is true. In this case, we may have a false positive result which indicates that even in
the absence of true efficacy with respect to a particular endpoint, probability of achieving
nominal significance will increase.
1.10 ANALYTICAL POPULATIONS
The Modified per Protocol (Mod PP) population consists of all subjects randomly assigned to
one of the treatments will be analyzed together, regardless of whether or not they completed the
whole study duration. It ignores non-compliance, protocol deviations, withdrawal, and anything
that happens after randomization.
1.11 FINAL REPORT AND FINAL RESEARCH CLOSEOUT BINDER
After data analysis, final clinical study report will be provided by the CRO to the sponsor. The
CRO will summarize the results which accurately reflect the clinical data of this study.
All essential documents which includes CRFs, study materials, adverse event pages for each
subject, protocol deviations, database, and the Final report will be enclosed in a Final Research
Closeout binder. The binder serves as a repository of the overall study process.