Interpretation of Data

Submitted by: Surakshya Bhandari

Masters in Clinical Pharmacy

Purbhanchal University (1st Year/ 1st Sem)

21 August, 2022
What is statistical treatment of data?
• Statistical treatment of data is when you apply some form of
statistical method to a data set to transform it from a group of
meaningless numbers into meaningful output.
• It is a way of removing bias by interpreting the data statistically
rather than subjectively.

2
 ANALYSIS AND INTERPRETATION OF DATA
The objective of data analysis is to obtain as
much information about the population as
possible based on the sample data collected. A
common method for analyzing data of a sample
population is to classify the data and then plot
The
the objective
frequencyofofdata analysis of
occurrence is to
all obtain as much information about the population as possible based on the
the samples.
sample data collected. A common method for analyzing data of a sample population is to classify the data and
then plot the frequency of occurrence of all the samples. For example, the frequency of weight distribution of
a classthe
For example, of students may
frequency be plotted
of weight in the form of a histogram that relates frequency to weight ().
distribution
of a class of students may be plotted in the form
of a histogram that relates frequency to weight ().

An important observation in this example is that the weight of most students lies in the middle of the
weight distribution. There is a common tendency for most sample values to occur around the mean.
This is described in the central limit theorem, which states that the frequency of the values of
measurements drawn randomly from a population tends to approximate a bell-shaped curve or normal
distribution. 3
 THE NORMAL DISTRIBUTION
If data are plotted according to the frequency of
occurrence, a pattern for the distribution of the data
is observed. Most data approximate a normal or
Gaussian distribution. The normal distribution is a
bell-shaped curve that is symmetric on both sides of
the mean.
The shape of the normal distribution is determined by
only two parameters, the population mean and the
variance, both of which may be estimated from the
samples. The variance is a measure of the spread or
variability of the sample.

The area representing probability between any two points on the normal distribution is
calculated from this graph. In practice, a cumulative standardized normal distribution table is
used to allow better accuracy.

4
 Two-One Sided Tests Procedure

• The two one-sided tests procedure is also referred to as the confidence interval approach.
• This statistical method is used to demonstrate if the bioavailability of the drug from the test formulation is too low or
high in comparison to that of the reference product.

• The objective of the approach is to determine if there are large differences (i.e., greater than 20%) between
the mean parameters.
• The 90% confidence limits are estimated for the sample means.
• The interval estimate is based on Student’s t distribution of the data. In this test, presently required by the FDA, a 90%
confidence interval about the ratio of means of the two drug products must be within ±20% for measurement of the
rate and extent of drug bioavailability.
• For most drugs, up to a 20% difference in AUC or Cmax between two formulations would have no clinical significance.
The lower 90% confidence interval for the ratio of means cannot be less than 0.80, and the upper 90% confidence
interval for the ratio of the means cannot be greater than 1.20.
• When log-transformed data are used, the 90% confidence interval is set at 80%–125%. These confidence limits have also
been termed the bioequivalence interval. The 90% confidence interval is a function of sample size and study variability,
including inter- and intra-subject variability.
5
Analysis of Variance

• An analysis of variance (ANOVA) is a statistical procedure used to test the data for differences within and between
treatment and control groups. A bioequivalent product should produce no significant difference in all pharmacokinetic
parameters tested.
• The parameters tested statistically usually include AUC(t-0), AUC(0-∞), and Cmax obtained for each treatment or
dosage form.
• The ANOVA may evaluate variability in subjects, treatment groups, study period, formulation, and other variables,
depending on the study design. If the variability in the data is large, the difference in means for each pharmacokinetic
parameter, such as AUC, may be masked, and the investigator might erroneously conclude that the two drug products
are bioequivalent.
• A statistical difference between the pharmacokinetic parameters obtained from two or more drug products is
considered statistically significant if there is a probability of less than 1 in 20 times or 0.05 probability (p ≤ .05) that these
results would have happened on the basis of chance alone.
• The probability, p, is used to indicate the level of statistical significance. If p < .05, the differences between the two drug
products are not considered statistically significant.

6
 Contd….

• To reduce the possibility of failing to detect small differences between the test products, a
power test is performed to calculate the probability that the conclusion of the ANOVA is
valid.
• The power of the test will depend on the sample size, variability of the data, and desired
level of significance.
• Usually, the power is set at 0.80 with a  = 0.2 and a level of significance of 0.05. The higher
the power, the test is more sensitive and the greater the probability that the conclusion of
the ANOVA is valid.

7
 PHARMACOKINETIC SOFTWARE
• Pharmacokinetic software consists of computer programs designed for computation and easy solution of pharmacokinetic
problems. Not all computer programs satisfy the user's full requirements, but many provide the following.

1. Fitting drug concentration-versus-time data to a series of pharmacokinetic models, and choosing the one that best
describes the data statistically
Typically, a least-squares program is employed, in which the sum of squared differences between observed data points
and theoretic prediction is minimized. Usually, a mathematical procedure is used iteratively (repetitively) to achieve a
minimum in the sum of squares (convergence). Some data may allow easier convergence with one procedure rather than
another. The mathematical method employed should be reviewed before use.

2. Fitting data into a pharmacokinetic or pharmacodynamics model defined by the user

This method is by far the most useful, because any list of prepared models is often limited. The flexibility of user-defined
models allows continuous refinement of the model as new experimental information becomes available. Some software
merely provides a utility program for fitting the data to a series of polynomials. This utility program provides a simple,
quantitative way of relating the variables, but offers little insight into the underlying pharmacokinetic processes.

8
 PHARMACOKINETIC SOFTWARE
3. Simulation
Some software programs generate data based on a model with parameter input by the user. When the parameters are
varied, new data are generated based on the model chosen. The user is able to observe how the simulated model data
matches the experimental observed data. Because pharmacokinetic processes are conveniently described by systems of
differential equations, the simulation process involves a numerical solution of the equation with predefined precision.

4. Clinical pharmacokinetic applications

Some software programs are available for the clinical monitoring of narrow-therapeutic-index drugs (i.e., critical-dose drugs)
such as the aminoglycosides, other antibiotics, theophylline, or anti-arrythmics. These programs may include calculations for
creatinine clearance using the Cockcroft†“Gault equation dosage estimation, pharmacokinetic parameter estimation for
the individual patient, and pharmacokinetic simulations.

9
 PHARMACOKINETIC SOFTWARE
• Various pharmacokinetic programs (software) are available on the Internet. Some popular commercially available computer software
programs are listed below. The descriptions may not represent the latest versions. New features are often added or old features improved.

 PCNONLIN
PCNONLIN is a powerful least-squares program for parameter estimation. Both a user-defined model and a library of over
20 compartmental models are available. The program accepts both differential and regular (analytical) equations. Users
may select the Hartley-modified or Levenberg-type Gauss†“Newton algorithm or the (Nelder and Mead) simplex
algorithm for minimizing the sum of squared residuals. Some training is needed

 WINNONLIN
WinNonlin is Windows-based software for pharmacokinetic, pharmacodynamic, and non-compartmental analysis. It is
designed for easy interfacing and secure data management with PkS Suite. WinNonlin can calculate individual
bioequivalence for all of the common replicated crossover designs. WinNonMix is associated software for population
pharmacokinetic analysis. WinNonlin's input and output data may be managed via Excel (Microsoft)-compatible
spreadsheet files. The Non-compartmental Analysis module computes derived pharmacokinetic parameters (AUCt 0, AUC0
∠, C max, cumulative excretion, etc.) PCNonlin's extensive library of models for nonlinear regression and parameter
estimation are included in this software.
Standard descriptive statistics and confidence intervals are determined from datasets.
10
 PHARMACOKINETIC SOFTWARE
RSTRIP
RSTRIP is menu-driven and very suitable for student use; it fits data to models, mono-, bi-, and triexponentials based on model selection criteria (Akaike Information
Criteria). A good statistics menu is available for AUC, C max, T max, and mean residence time. The program gives initial parameter estimates and final parameters after
iteration. However, the program does not handle differential equations or user-defined models. Plot outputs are available, as are pharmacokinetic curve stripping, and
least-squares parameter optimization. The original software was written for PC DOS but has now been replaced by a Windows version with additional features.

GASTROPLUS
GastroPlus is a computer simulation program that predicts the rate and extent of drug absorption from the gastrointestinal
tract. This innovative program was developed by a team of scientist-programmers under the direction of Dr. Michael B.
Bolger at Simulations Plus, Inc., in collaboration with Dr. Gordon L. Amidon.

11
References:
• Biopharmaceutics and Pharmacokinetics by Leon.pdf
• shargel (1).pdf
• https://www.discoverphds.com/blog/statistical-treatment-of-data

12
THANK YOU

13