The document describes an experiment examining the effects of alcohol consumption on recovery from traumatic brain injury in mice. 13 mice were trained on a task and then given either an acute dose of alcohol or saline before being given a brain lesion. Their ability to perform the task was tested over 60 days. Statistical models including logistic regression and nonlinear regression were used to analyze the data. Surprisingly, mice that received alcohol recovered more quickly, regaining the ability to perform the task an average of 5 days faster than mice that received saline. Further research is needed to understand the mechanisms at play and verify the results with a larger sample size.
Lecture on causal inference to the pediatric hematology/oncology fellows at Texas Children's hospital as part of their Biostatistics for Busy Clinicians lecture seriers.
Lecture on causal inference to the pediatric hematology/oncology fellows at Texas Children's hospital as part of their Biostatistics for Busy Clinicians lecture seriers.
Crimson Publishers: Reply To: Comments on "Transabdominal Preperitoneal (TAPP...CrimsonGastroenterology
Reply To: Comments on “Transabdominal Preperitoneal (TAPP) Versus Totally Extraperitoneal (TEP) for Laparoscopic Hernia Repair: A Meta-Analysis” by Feng Xian Wei in Gastroenterology Medicine & Research
Practical Methods To Overcome Sample Size ChallengesnQuery
Watch the video at: https://www.statsols.com/webinars/practical-methods-to-overcome-sample-size-challenges
In this webinar hosted by Ronan Fitzpatrick - Head of Statistics and nQuery Lead Researcher at Statsols - we will examine some of the most common practical challenges you will experience while calculating sample size for your study. These challenges will be split into two categories:
1. Overcoming Sample Size Calculation Challenges
(Survival Analysis Example)
We will examine practical methods to overcome common sample size calculation issues by focusing in on one of the more complex areas for sample size determination; Survival analysis. We will cover difficulties and potential issues surrounding challenges such as:
Drop Out: How to deal with expected dropouts or censoring. We compare the simple loss-to-follow-up method and integrating a dropout process into the sample size model?
Planning Uncertainty: How best to deal with the inevitable uncertainty at the planning stage? We examine how best to apply a sensitivity analysis and Bayesian approaches to explore the uncertainty in your sample size calculations.
Choosing the Effect Size: Various approaches and interpretations exist for how to find the effect size value. We examine those contrasting interpretations and determine the best method and also how to deal with parameterization options.
2. Overcoming Study Design Challenges
(Vaccine Efficacy Example)
The Randomised Controlled Trial (RCT) is considered the gold standard in trial design in drug development. However, there are often practical impediments which mean that adjustments or pragmatic approaches are needed for some trials and studies.
We will examine practical methods how to overcome common study design challenges and how these affect your sample size calculations. In this webinar, we will use common issues in vaccine study design to examine difficulties surrounding issues such as:
Case-Control Analysis: We will examine how to deal with study constraints and how to deal with analyses done during an observational study.
Alternative Randomization Methods: How best to address randomization in your vaccine trial design when full randomization is difficult, expensive or impractical. We examine how sample size calculations are affected with cluster or Mendelian randomization.
Rare Events: How does an outcome being rare affect the types of study design and statistical methods chosen in your study.
Minimizing Risk In Phase II and III Sample Size CalculationnQuery
[ Watch Webinar: http://bit.ly/2thIgmi ]. In this free webinar, Head of Statistics at Statsols, Ronan Fitzpatrick, addresses the issues of reducing risk in Phase II/III sample size calculations. Topics covered will include:
Sample Size Determination For Different Trial Designs
Bayesian Sample Size Determination
Sample Size For Survival Analysis
& more
Combination of informative biomarkers in small pilot studies and estimation ...LEGATO project
Background:
Biomarker candidates are defined as measurable molecules found in biological media. According to Biomarkers Definitions Working Group, 2001, biomarkers cover a rather wide range of parameters. Recently, biomarkers are used widely in medical researches, where single biomarkers may not possess the desired cause-effect association for disease classification and outcome prediction. Therefore the efforts of the researchers currently is to combine biomarkers. By new technologies like microarrays, next generation sequencing and mass spectrometry, researchers can obtain many biomarker candidates that can exceed tens of thousands. To avoid wasting money and time, it is suggested to control the number of patients strictly. However, pilot studies usually have low statistical power which reduces the chance of detecting a true effect .
Non-inferiority and Equivalence Study design considerations and sample sizenQuery
About the webinar
This webinar examines the role of non-inferiority and equivalence in study design
In this free webinar, you will learn about:
-Regulatory information on this type of study design
-Considerations for study design and your sample size
-Practical worked examples of
--Non-inferiority Testing
--Equivalence Testing
Duration - 60 minutes
Speaker: Ronan Fitzpatrick, Head of Statistics, Statsols
Watch the video at: https://www.statsols.com/webinars
Standard error is used in the place of deviation. it shows the variations among sample is correlate to sampling error. list of formula used for standard error for different statistics and applications of tests of significance in biological sciences
Experimental design cartoon part 5 sample sizeKevin Hamill
Part 5 of 5 - Experimental design lecture series. This one focuses on sample size calculations and introduces some of the commonly used statistical tests (for normally distributed data). Toward the end it covers type I and II errors, alpha/beta and reducing variability.
PU 515 Applied Biostatistics Final Exam 1. The.docxamrit47
PU 515
Applied Biostatistics
Final Exam
1. The following are body mass index (BMI) scores measured in 12 patients who are free of diabetes
and participating in a study of risk factors for obesity. Body mass index is measured as the ratio of
weight in kilograms to height in meters squared. Generate a 95% confidence interval estimate of
the true BMI.
25 27 31 33 26 28 38 41 24 32 35 40
2. Consider the data in Problem #1. How many subjects would be needed to ensure that a 95%
confidence interval estimate of BMI had a margin of error not exceeding 2 units?
3. A clinical trial is run to investigate the effectiveness of an experimental drug in reducing preterm
delivery to a drug considered standard care and to placebo. Pregnant women are enrolled and
randomly assigned to receive either the experimental drug, the standard drug or placebo. Women
are followed through delivery and classified as delivering preterm (< 37 weeks) or not. The data
are shown below.
Preterm Delivery Experimental Drug Standard Drug Placebo
Yes 17 23 35
No 83 77 65
Is there a statistically significant difference in the proportions of women delivering preterm among
the three treatment groups? Run the test at a 5% level of significance.
4. Consider the data presented in problem #4. Previous studies have shown that approximately 32%
of women deliver prematurely without treatment. Is the proportion of women delivering
prematurely significantly higher in the placebo group? Run the test at a 5% level of significance.
PU 515
Applied Biostatistics
Final Exam
5. A study is run comparing HDL cholesterol levels between men who exercise regularly and those
who do not. The data are shown below.
Regular Exercise N Mean Std Dev
Yes 35 48.5 12.5
No 120 56.9 11.9
Generate a 95% confidence interval for the difference in mean HDL levels between men who
exercise regularly and those who do not.
6. A clinical trial is run to assess the effects of different forms of regular exercise on HDL levels in
persons between the ages of 18 and 29. Participants in the study are randomly assigned to one of
three exercise groups - Weight training, Aerobic exercise or Stretching/Yoga – and instructed to
follow the program for 8 weeks. Their HDL levels are measured after 8 weeks and are summarized
below.
Exercise Group N Mean Std Dev
Weight Training 20 49.7 10.2
Aerobic Exercise 20 43.1 11.1
Stretching/Yoga 20 57.0 12.5
Is there a significant difference in mean HDL levels among the exercise groups? Run the test at a
5% level of significance. HINT: SSwithin = 21,860.
7. Consider again the data in problem #6. Suppose that in the aerobic exercise group we also
measured the number of hours of aerobic exercise per week and the mean is 5.2 hours with a
standard deviation of 2.1 hours. The sample correlation is -0.42.
a) Is there evidence of a s ...
Crimson Publishers: Reply To: Comments on "Transabdominal Preperitoneal (TAPP...CrimsonGastroenterology
Reply To: Comments on “Transabdominal Preperitoneal (TAPP) Versus Totally Extraperitoneal (TEP) for Laparoscopic Hernia Repair: A Meta-Analysis” by Feng Xian Wei in Gastroenterology Medicine & Research
Practical Methods To Overcome Sample Size ChallengesnQuery
Watch the video at: https://www.statsols.com/webinars/practical-methods-to-overcome-sample-size-challenges
In this webinar hosted by Ronan Fitzpatrick - Head of Statistics and nQuery Lead Researcher at Statsols - we will examine some of the most common practical challenges you will experience while calculating sample size for your study. These challenges will be split into two categories:
1. Overcoming Sample Size Calculation Challenges
(Survival Analysis Example)
We will examine practical methods to overcome common sample size calculation issues by focusing in on one of the more complex areas for sample size determination; Survival analysis. We will cover difficulties and potential issues surrounding challenges such as:
Drop Out: How to deal with expected dropouts or censoring. We compare the simple loss-to-follow-up method and integrating a dropout process into the sample size model?
Planning Uncertainty: How best to deal with the inevitable uncertainty at the planning stage? We examine how best to apply a sensitivity analysis and Bayesian approaches to explore the uncertainty in your sample size calculations.
Choosing the Effect Size: Various approaches and interpretations exist for how to find the effect size value. We examine those contrasting interpretations and determine the best method and also how to deal with parameterization options.
2. Overcoming Study Design Challenges
(Vaccine Efficacy Example)
The Randomised Controlled Trial (RCT) is considered the gold standard in trial design in drug development. However, there are often practical impediments which mean that adjustments or pragmatic approaches are needed for some trials and studies.
We will examine practical methods how to overcome common study design challenges and how these affect your sample size calculations. In this webinar, we will use common issues in vaccine study design to examine difficulties surrounding issues such as:
Case-Control Analysis: We will examine how to deal with study constraints and how to deal with analyses done during an observational study.
Alternative Randomization Methods: How best to address randomization in your vaccine trial design when full randomization is difficult, expensive or impractical. We examine how sample size calculations are affected with cluster or Mendelian randomization.
Rare Events: How does an outcome being rare affect the types of study design and statistical methods chosen in your study.
Minimizing Risk In Phase II and III Sample Size CalculationnQuery
[ Watch Webinar: http://bit.ly/2thIgmi ]. In this free webinar, Head of Statistics at Statsols, Ronan Fitzpatrick, addresses the issues of reducing risk in Phase II/III sample size calculations. Topics covered will include:
Sample Size Determination For Different Trial Designs
Bayesian Sample Size Determination
Sample Size For Survival Analysis
& more
Combination of informative biomarkers in small pilot studies and estimation ...LEGATO project
Background:
Biomarker candidates are defined as measurable molecules found in biological media. According to Biomarkers Definitions Working Group, 2001, biomarkers cover a rather wide range of parameters. Recently, biomarkers are used widely in medical researches, where single biomarkers may not possess the desired cause-effect association for disease classification and outcome prediction. Therefore the efforts of the researchers currently is to combine biomarkers. By new technologies like microarrays, next generation sequencing and mass spectrometry, researchers can obtain many biomarker candidates that can exceed tens of thousands. To avoid wasting money and time, it is suggested to control the number of patients strictly. However, pilot studies usually have low statistical power which reduces the chance of detecting a true effect .
Non-inferiority and Equivalence Study design considerations and sample sizenQuery
About the webinar
This webinar examines the role of non-inferiority and equivalence in study design
In this free webinar, you will learn about:
-Regulatory information on this type of study design
-Considerations for study design and your sample size
-Practical worked examples of
--Non-inferiority Testing
--Equivalence Testing
Duration - 60 minutes
Speaker: Ronan Fitzpatrick, Head of Statistics, Statsols
Watch the video at: https://www.statsols.com/webinars
Standard error is used in the place of deviation. it shows the variations among sample is correlate to sampling error. list of formula used for standard error for different statistics and applications of tests of significance in biological sciences
Experimental design cartoon part 5 sample sizeKevin Hamill
Part 5 of 5 - Experimental design lecture series. This one focuses on sample size calculations and introduces some of the commonly used statistical tests (for normally distributed data). Toward the end it covers type I and II errors, alpha/beta and reducing variability.
PU 515 Applied Biostatistics Final Exam 1. The.docxamrit47
PU 515
Applied Biostatistics
Final Exam
1. The following are body mass index (BMI) scores measured in 12 patients who are free of diabetes
and participating in a study of risk factors for obesity. Body mass index is measured as the ratio of
weight in kilograms to height in meters squared. Generate a 95% confidence interval estimate of
the true BMI.
25 27 31 33 26 28 38 41 24 32 35 40
2. Consider the data in Problem #1. How many subjects would be needed to ensure that a 95%
confidence interval estimate of BMI had a margin of error not exceeding 2 units?
3. A clinical trial is run to investigate the effectiveness of an experimental drug in reducing preterm
delivery to a drug considered standard care and to placebo. Pregnant women are enrolled and
randomly assigned to receive either the experimental drug, the standard drug or placebo. Women
are followed through delivery and classified as delivering preterm (< 37 weeks) or not. The data
are shown below.
Preterm Delivery Experimental Drug Standard Drug Placebo
Yes 17 23 35
No 83 77 65
Is there a statistically significant difference in the proportions of women delivering preterm among
the three treatment groups? Run the test at a 5% level of significance.
4. Consider the data presented in problem #4. Previous studies have shown that approximately 32%
of women deliver prematurely without treatment. Is the proportion of women delivering
prematurely significantly higher in the placebo group? Run the test at a 5% level of significance.
PU 515
Applied Biostatistics
Final Exam
5. A study is run comparing HDL cholesterol levels between men who exercise regularly and those
who do not. The data are shown below.
Regular Exercise N Mean Std Dev
Yes 35 48.5 12.5
No 120 56.9 11.9
Generate a 95% confidence interval for the difference in mean HDL levels between men who
exercise regularly and those who do not.
6. A clinical trial is run to assess the effects of different forms of regular exercise on HDL levels in
persons between the ages of 18 and 29. Participants in the study are randomly assigned to one of
three exercise groups - Weight training, Aerobic exercise or Stretching/Yoga – and instructed to
follow the program for 8 weeks. Their HDL levels are measured after 8 weeks and are summarized
below.
Exercise Group N Mean Std Dev
Weight Training 20 49.7 10.2
Aerobic Exercise 20 43.1 11.1
Stretching/Yoga 20 57.0 12.5
Is there a significant difference in mean HDL levels among the exercise groups? Run the test at a
5% level of significance. HINT: SSwithin = 21,860.
7. Consider again the data in problem #6. Suppose that in the aerobic exercise group we also
measured the number of hours of aerobic exercise per week and the mean is 5.2 hours with a
standard deviation of 2.1 hours. The sample correlation is -0.42.
a) Is there evidence of a s ...
Question 1.A group of researchers is replicating an earlier .docxIRESH3
Question 1.
A group of researchers is replicating an earlier experiment that indicated that participants who received task-specific feedback were more likely to persist at a task than participants who received more general, encouraging feedback. In an effort to ensure that participants are not treated differently based on the condition that they are in, the researchers automate all of the procedures and follow a written protocol when interacting with the participants. The researchers are trying to minimize:
placebo effects.
demand characteristics.
experimenter expectancy effects.
participant suspicion effects.
Question 2.
In a study examining the effects of heredity on intelligence, researchers compare the correlation of intelligence test scores of identical twins with the correlation of intelligence test scores for fraternal twins. In this experiment, the researcher is assuming that the comparison of identical and fraternal twins is a measure of heredity. This is an example of a ________________ inference.
construct
statistical
generalizability
Causal
Question 3
Researchers interested in studying the effect of happiness on various health outcomes randomly assign each person who comes in to the laboratory to one of two study conditions. However, several people in the study are friends and drove to the study together. The group of friends indicates that they need to be in the same condition of the study so that they can all leave at the together to get home. Accommodating the subjects' request might threaten validity because of the effect of:
regression to the mean.
attrition.
maturation.
selection.
Question 4
In an experiment on the effects of everyday stress on memory, a researcher has participants record every hour how much stress they are feeling and then complete a short-term memory task. The results of the study reveal that everyday stress may affect short-term memory. After evaluating the results of the study, however, the researcher is concerned that people who have high scores on neuroticism questionnaires are more likely to report stress and exhibit memory problems than people who have low scores. The researcher is worried about __________ validity.
construct
internal
statistical conclusion
external
Question 5
__________ validity concerns the generalizability of findings beyond the present study.
Ecological
Construct
Statistical conclusion
External
Question 6
A researcher is investigating the ability of aversive punishment to decrease students' disruptive behaviors in class. She is worried that the number of punishments will vary from student to student and thus will bias the results of the study. The researcher would do well to:
run a pilot test before conducting the study.
manipulate participants' knowledge about the study.
use a yoked control-group.
use a red herring technique.
Question 7
A psychologist is examini ...
3.A study wants to determine if taking fish oils can reduce depr.docxcameroncourtney45
3.
A study wants to determine if taking fish oils can reduce depressive symptoms. A group of 50 volunteers who suffered from mild depression were randomly divided into two groups. Each person was given a three-month’s supply of capsules. One group was given capsules that contained fish oils while the other group was given capsules that look and tasted like fish oils, but actually only contained sugar. Neither the participants nor the investigator knew what type of capsule they were taking. At the end of the month, a psychologist evaluated them to determine if their depressive symptoms had
changed
.
Therefore, we are comparing the “change in depressive symptoms” for individuals in two groups.
Explain whether each of the following terms applies or does not apply to this study.
Why or why not?
a.
observational study
A: Does not apply: the study is not being conducted in their natural settings/groupings since they are randomly assigned to a group.
b.
randomized experiment
A: Does apply:
creates differences in explanatory variable when randomly assigning groups.
c.
placebo
d.
placebo effect
e.
single-blind
f.
double-blind
g.
matched pairs (dependent samples)
h.
block design
i.
independent samples
j.
explanatory variable (What is it?)
k.
response variable (What is it?)
4.
Does the use of cell phones lead to a higher incidence of brain cancer? People with brain cancer were matched with people who did not have brain cancer on age, gender, and living environment. Each participant in the study was asked to answer questions about previous life experiences and exposures.
Determine whether or not each of the following terms applies or does not apply to this observational study.
Why or why not?
a.
prospective
b.
retrospective
c.
case-control study
5.
A study involving ten people wants to compare the effectiveness of two different brands of antihistamines with regard to enhancing sleep. Each person is randomly assigned to take either Antihistamine A or Antihistamine B on the first night.
Then each person takes the other antihistamine on the following night.
With each person, the hours of sleep were recorded for each night. Explain whether each of the following terms applies or does not apply to this study.
Why or why not?
a.
observational study
b.
randomized experiment
c.
carry-over effect (confounding)
d.
matched pairs (dependent samples)
e.
explanatory variable (What is it?)
f.
response variable (What is it?)
6.
Suppose the study found in the previous problem instead found that each person took Antihistamine A on the first night and Antihistamine B on the second night. What terms that did not apply to the previous problem now apply to this problem? Explain.
7.
Are you annoyed with spam e-mail? Suppose a random sample of 200 Penn State students was asked this question of which 80% said that they are annoyed. From the provided information we can find the following:
sample percent = 80% (sample proportion = .80)
.
DNP-816 Analysis & Applic of Health Data for ANPSTATISTICS QUIZ.docxgreg1eden90113
DNP-816: Analysis & Applic of Health Data for ANP
STATISTICS QUIZE
1. Which of the following research designs includes both an intervention and randomization?
Group of answer choices
Grounded theory research
Non-experimental research
Time series design
Experimental research
2. What is the initial question the researcher should ask when selecting a research design for a particular study?
Group of answer choices
What is the norm in the research topic area?
What type of data analysis techniques will be used?
What instruments will be used to measure the variables in the study?
What is the primary purpose of the study?
3. Which of the following research questions is the appropriate question for a correlational research design?
Group of answer choices
What is the experience of women with hyperthyroidism and resolution of sypmtoms after treatment?
What is the relationship between amount of exercise/week and arthralgia in women with estrogen receptor positive breast cancer who are being treated with an aromatase inhibitor?
What is the prevalence of heroin addiction amongst adults ages 18-45 in the Greater Cincinnati region?
In patients undergoing a total hip arthroplasty, which of the following treatments is most effective in the decolonization of MRSA: preoperative povidone iodine or posteroperative mupirocin?
4. Match the types of quantitative research listed below with the sample study titles.
Group of answer choices
Descriptive research
Correlational research
Quasi-experimental research
Experimental research
5. Bias is a term used to indicate that data in a study are being distorted or slanted away from reality by some influencing factor. Which of the following is true about bias in research?
Group of answer choices
Instruments that are valid for measuring the identified variables are a source of bias.
The researcher can not be a source of bias in a study because he/she is in control
Preconcieved ideas about what the finding of a study will be may lead to bias in intrepreting data.
It is the same as manipulation because the researcher determines the treatment to be given.
6. Manipulation is a term used in quasi and experimental research to mean:
Group of answer choices
An underhanded strategy designed to make subject behave as the researcher wants them to.
Controlling the environment in which the research takes place
An intervention or treatment introduced by the researcher to assess its impact on the dependent variable.
The ability of the researcher to be able to handle or use the equpiment needed to collect data in the study
7. We do not know whether the pattern of results found in our samples accurately reflects what is happening in the population or if it is the result of what type of error?
Group of answer choices
Representative
Distribution
Sampling
Mean
8. Extraneous variables may be controlled by:
Group of answer choices
Using a natural clinical setting
Selecting individuals that are relatively alike in relation to var.
Dr. Joseph C. Fleishaker - Pfizer Inc., Speaker at the marcus evans Discovery Summit Fall 2011, delivers his presentation on From In Silco to In Vivo – Modeling and Simulation Technologies, a Tool for Optimized Drug Development
STATISTICS PRACTICE QUIZ 3Question 1A telecommunications com.docxmckellarhastings
STATISTICS PRACTICE QUIZ 3
Question 1
A telecommunications company asserts that, according to a study they conducted, patients who have access to more cable channels during recovery from surgery are discharged sooner than those who receive only basic cable in their hospital rooms. Which of the following demonstrates clinical significance of the study?
Patients with more cable channels were discharged an average of one-half day sooner than those with fewer channels
The study used a sample size of 12
Patients with fewer cable channels requested additional nursing support an average of 3 times more during their recovery.
Patients with more cable channels tended to prefer nature and arts channels over history or news and current affairs channels
Question 2
After completing the power analysis, the researchers determine they need a sample size of 400 to have adequate power in their study. After enrolling subjects, they have a lower response rate than anticipated and they only enroll 320. Inadequate enrollment may increase the risk of:
A type II error
Systematic bias
A type I error
Statistical significance
Question 3
Researchers study the relationship between interpersonal violence and health in college age women. The researchers examined the average score on a psychological distress scale and compared the score for abused versus non-abused women. If the researchers report a statistically significant difference and are
incorrect about this conclusion what type of error could it be?
A type I error
A type II error
A clinical error
An error of omission
Question 4
Researchers studied the relationship between Vitamin B12 consumption and hair growth and report a p value of 0.56. The study was a pilot study with an alpha of 0.10. You know this means:
There is no statistically significant relationship between Vitamin B12 consumption and hair growth.
Vitamin B12 consumption is association with a 56% increase in hair growth.
The sample size was too small and a type I error was made
There is a statistically significant relationship between Vitamin B12 consumption and hair growth.
Question 5
Researchers study the relationship between interpersonal violence and health in college age women. They selected an alpha of 0.05. The researchers examined the average score on a psychological distress scale and compared the score for abused versus non-abused women. A p value of 0.016 is reported. Based on this information, you know:
There is a statistically significant difference in the average psychological distress score
This is a clinically significant result
There is no significant difference in the average psychological distress score.
In this study women who were abused were statistically more likely to report psychological distress.
Question 6
A study reports that administering vancomycin incorrectly is associated with red man syndrome. You know this means:
The p value is less than alpha
A type I error was made in administering the medic.
Extending A Trial’s Design Case Studies Of Dealing With Study Design IssuesnQuery
About the webinar
As trials increase in complexity and scope, there is a requirement for trial designs to reflect this.
From dealing with non-proportional hazards in survival analysis to dealing with cluster randomization, we examine how to deal with study design issues of complex trials.
In this free webinar, you will learn about:
Dealing with study design issues
Practical worked examples of
Non-proportional Hazards
Cluster Randomization
Three Armed Trials
Non-proportional Hazards
Non-proportional hazards and complex survival curves have become of increasing interest, due to being commonly seen in immunotherapy development. This has led to interest in assessing the robustness of standard methods and alternative methods that better adapt to deviations.
In this webinar, we look at methods proposed for complex survival curves and the weighted log-rank test as a candidate model to deal with a delayed survival effect.
Cluster Randomization
Cluster-randomized designs are often adopted when there is a high risk of contamination if cluster members were randomized individually. Stepped-wedge designs are useful in cases where it is difficult to apply a particular treatment to half of the clusters at the same time.
In this webinar, we introduce cluster randomization and stepped-wedge designs to provide an insight into the requirements of more complex randomization schedules.
Three Armed Trials
Non-inferiority testing is a common hypothesis test in the development of generic medicine and medical devices. The most common design compares the proposed non-inferior treatment to the standard treatment alone but this leaves uncertain if the treatment effect is the same as from previous studies. This “assay sensitivity” problem can be resolved by using a three arm trial which includes placebo alongside the new and reference treatments for direct comparison.
In this webinar we show a complete testing approach to this gold standard design and how to find the appropriate allocation and sample size for this study.
Duration - 60 minutes
Speaker: Ronan Fitzpatrick, Head of Statistics, Statsols
A real life example to show the added value of the Phenotype Database (dbNP)....Chris Evelo
NuGO has initiated the development of the Phenotype Database (dbNP). This database is developed together with several other consortia (e.g. Netherlands Metabolomics Centre) and is currently used within several European projects, such as Food4me, NU-AGE, Bioclaims and Nutritech.
The Phenotype Database (www.dbnp.org) is a web-based application/database that can store any biological study. We used this application to perform an analysis on a combination of several studies with the objective to test if it is possible to answer new research questions using a ‘virtual cohort’.
Study comparison:
The assessment of the health status of an individual is an important but challenging issue. Nowadays, challenge tests are proposed as a method to assess and quantify health status. We would like to find mechanistic explanations for differences in clinical subgroups and to develop a metabolomics platform based fingerprint at baseline that represents important parameters of the challenge test. Currently, there is not one single study available that includes enough subjects from specific clinical subgroups to develop such a fingerprint or study the biological processes specific for those subgroups. Therefore, we developed a toolbox that facilitates the combined analysis of multiples studies.
1. All Models Are Wrong, Some Are Useful
Mixed Effects Models for Biological Data
Sean Mc Carthy
Introduction
A pervasive issue for applied statisticians
and practicing scientists is the choice and
application of statistical models. Depending on
the hypothesis that are of importance to a
researcher, a variety of statistical models can
be appropriately used on a single data set. In
this project, we consider several models that
elucidate different facets of information from
mice that have received a traumatic brain
injury. The project involves results pertaining
to mixed effects generalized linear and
nonlinear modeling. Specifically, we will
consider logistic regression, mixed effects
logistic regression, three parameter nonlinear
regression, and mixed effects three parameter
Experiment and Data
The data analyzed here comes from an
experiment examining the effects that alcohol
consumption has on recovery from traumatic
brain injury. Thirteen mice were trained to
reliably grab food pellets from an experimenter.
Seven mice were randomly assigned to
receive an acute dose of alcohol and five mice
were randomly assigned to receive a dose of
saline solution. Upon completing the
treatment, the mouse was anesthetized and
given a focal brain lesion to motor cortex.
Over the next 60 days, the mice were tested
on how many food pellets out of 20 they could
successfully grab. The data is longitudinal as it
monitors the mice over time. Furthermore, the
Research Question
Traumatic brain injury associated with alcohol consumption is an unfortunate reality of the modern world.
While some brain injuries can be recovered from, the degree of recovery and speed of recovery are highly
variable. The experiment was designed to examine the length of time needed to regain the ability to perform
a trained task. Two questions were of primary interest. First, did the acute alcohol condition lead to a
difference in recovery when compared to the saline condition? Second, did one group recover in a shorter
time or was no difference in recovery detected? To answer these questions four statistical models were
employed. The models allow us to answer each research question and demonstrate the differences between
Statistical Methods
All analysis was completed using the open source software R and R Studio. The methods used were
logistic regression, mixed effects logistic regression, fixed effects three parameter logistic nonlinear
regression, and mixed effects three parameter logistic nonlinear regression.
Acknowledgment
Loyola Graduate School Research Experience for Master's Programs
Fellowship for the opportunity.
●
Tim O'Brien for mentoring.
●
Ian Vaagenes for data and R coding help.
Works Used
●
Pinheiro, J.C., Bates, D.M.. Mixed-Effects Models in S and S-
PLUS. Springer Inc, Verlag New York. 2000.
●
Bates, D.M.. lme4: Mixed-effects modeling with R. Springer Inc,
New York. 2010.
●
Ian C. Vaagenes , Shih-Yen Tsai, Son T. Ton, Vicki A. Husak,
Susan O. McGuire, Timothy E. O’Brien, Gwendolyn L. Kartje
(2015) Binge Ethanol Prior to Traumatic Brain Injury Worsens
Sensorimotor Functional Recovery in Rats. PloS ONE 10(3):
Results/Discussion
The conclusions from the analysis of the data are surprising. Mice that were
given a single high dose of alcohol before traumatic brain injury generally
recovered more quickly. The mice in the Saline condition averaged five (5) extra
days to recover the ability to successfully reach for food pellets 50% of the time.
These findings are counter to what would be expected. Common sense would
dictate that flooding an animal with a central nervous system depressant before
brain injury should lead to slower recovery and possibly less recovery. The results
found here are completely contrary to common sense. The one time inebriated
mice recovered more quickly and were just as if not more successful than the
sober mice.
The results from this experiment raise several questions:
● What is the mechanism at play in recovery?
● How does alcohol consumption affect neural recovery?
● Are the same results found for long term alcohol exposure?
● Is a sample of 13 mice large enough?
Mechanisms of neural repair have been studied. Like skin tissue, neurons do
show ability to repair and scar. What is less understood is how a onetime large
dose of alcohol could facilitate healthy neuron and brain recovery. It may be that
by delaying a healing response, the damaged neurons have time to regain some
physical proximity that was disrupted. By regaining physical proximity, the
amount of neural repair could be reduced. Further study is required to examine
the mechanisms at work.
As to long term alcohol use, a study by Vaagenes et. al. (2015) demonstrated
that repeated exposure to alcohol caused slower recovery of skills in mice.
While mixed effects modeling can give estimates on a population, an
appropriate sample size is needed. As the poster title indicates, no model is
completely accurate. To verify these results, a replication with a larger sample
size would be needed.
Fixed Effects Logistic Regression
Strictly speaking, using simple logistic regression is not appropriate here due
to the data being longitudinal. The model does not for the repeated
measurements.
●
Model coefficients for Saline treatment (-0.93, 95%CI (-1.11, -0.75)) and the
Saline by Time interaction (0.01, 95%CI (0.009, 0.02)) provide weak evidence
Mixed Effects Logistic Regression
For addressing the first research question about a difference in treatment
groups, this is a more appropriate model.
●
Model coefficients for Saline treatment (-1.09, 95%CI (-1.94, -0.25)) and the
Saline by Time interaction (0.02, 95%CI (0.01, 0.02)) provide stronger evidence
that the acute alcohol condition mice recover more quickly.
Fixed Effects Nonlinear Regression
This is the first model to address the nonlinear nature of the data. The 3-
parameter logistic allows for estimation of the difference in time between the two
groups achieving 50% accuracy in reaching for food pellets.
●
The Saline treatment mice were 4.05 days (95% CI (1.56, 6.9)) slower in gaining
the ability to successfully reach for the pellet 50% of the time.
Mixed Effects Nonlinear Regression
The mixed effects nonlinear model allows for more accurate estimates and
inference about the population of mice as a whole. Again the time difference in
gaining the ability to grab 50% of pellets is significant.
●
The Saline treatment mice were 5.26 days (95% CI (0.49, 10.02)) slower in
gaining the ability to successfully reach for the pellet 50% of the time.