Understanding Clinical Research Behind the Statistics

In this blog you will find the correct answer of the Coursera quiz Understanding Clinical Research Behind the Statistics mixsaver always try to brings best blogs and best coupon codes

Week- 1

Peer-graded Assignment: Week 1: Navigating Clinical Research

Week 2 Graded Quiz

1.
Question 1
In a questionnaire, patients are asked to rate their opinion to a statement with a ‘1’ referring to not agreeing at all, to a ‘5’ referring to total agreement. Twenty patients complete the survey and the results are tabulated under a variable named ‘Responses to statement ‘.

What is an appropriate data type for this variable?

1 point

Nominal categorical
Ordinal categorical
Interval numerical
Ratio numerical

2.
Question 2
What best describes continuous data?

1 point

It cannot be subdivided.
It has a discreet number of positive values.
It has infinite possibilities for subdivisions.
It can have infinity large values.

3.
Question 3
This and the following few questions pertains to the article: Amoxicillin plus temocillin as an alternative empiric therapy for the treatment of severe hospital-acquired pneumonia: results from a retrospective audit

In this article the authors were concerned about the outcome of patients admitted with severe hospital acquired pneumonia. This refers to severe bacterial infection of the lungs. The patients were infected whilst in hospital. Bacteria that exist in the hospital environment are usually more difficult to treat due to resistance of the bacteria to antibiotics. Patients who contract hospital acquired bacterial infections may have a worse outcome than those who contract bacterial infection in the community (outside the hospital). These patients are in hospital for a specific medical reason, which usually means that their ability to fight infection might be compromised. Hospital-acquired bacteria can also be more virulent. It is this combination of a weakened immune system and aggressive, difficult to treat bacteria that lead to a worse outcome.

In the healthcare setting pertaining to this setting a decision was made to change the standard antibiotic therapy for the treatment of patients with severe hospital-acquired pneumonia. These are difficult decisions to make due to the fact that it may directly impact the wellbeing of patients. A retrospective audit was conducted to assess the effect of this change.

What type of research was conducted in this study?

1 point

Meta-analysis
Observation study
Experimental study
Systematic review

4.
Question 4
Suppose a study is planned to compare the effectiveness of the two types of antibiotics by selecting 100 patients and randomly assigning 50 of them to receive a new antibiotic number and the other 50 to receive the antibiotic used routinely. Both the patients and the medical personnel (including those who evaluate the treatment and collect the data) are blinded as to which antibiotic the patients receive until after the results of the analysis is available.

What is this type of study?

1 point

Randomized, double-blinded, controlled trial
Randomized, self-controlled trial
Randomized, single-blinded, controlled trial
Case-control series

5.
Question 5
Data point values were collected pertaining to the variable, length of hospital stay. These values were captured in terms of days.

What data type does this represent?

1 point

Ratio-type numerical
Interval-type numerical
Discrete data type
Ordinal categorical

6.
Question 6
Standard deviation is used to describe the spread in the data point values for age. What does this represent?

1 point

It represents the square of the difference between each individual age and the mean (average) age
It represents the minimum and maximum values in the data set of age values.
It represents the upper and lower limits within which 95% of the values fall.
It represents the square root of the average of the squared differences of all the values from the mean age.

7.
Question 7
The actual data points that made up the possible values for the variable Clinical Outcome were: cure, failure, relapse, and intermediate.

What data type are these?

1 point

Interval-type numerical
Ratio-type numerical
Ordinal categorical
Nominal categorical

8.
Question 8
By what measure of central tendency would you express the data point values for the variable Clinical Outcome?

1 point

Mean
Mode
Interquartile range
Median

9.
Question 9
Researchers are interested in the use of screening mammography and it’s efficacy in preventing mortality from breast cancer. They form two groups of patients, pre- and post-menopausal. Patient files from a radiology unit were used in this study and included all patients seen in the last year.

What is the sampling strategy appropriate in this study for distinguishing between these two groups?

1 point

Simple random sampling
Clustered random sampling
Systematic random sampling
Stratified random sampling

10.
Question 10
Researchers require 20 patients for their analysis of a new drug. Sixty equally suitable volunteers signed up to be part of the study. Every third volunteer on the list was selected to make up the sample.

What is the sampling strategy appropriate in this study?

1 point

Simple random sampling
Systematic random sampling
Cluster sampling sampling
Stratified random sampling

Week 3 Graded Quiz

1.
Question 1
Read the article linked here: http://www.samj.org.za/index.php/samj/article/view/9298/6855

The researchers were concerned that the reference values for common blood cell counts used by South African laboratories were taken from research conducted on samples taken from population that did not reflect (could not be inferred to) the local population.

Answer this question and those that follow.

What type of study was this?

1 point

Cohort study
A double-blind randomized trial
A case-control series
Cross-sectional study

2.
Question 2
The method of dispersion used to describe the age variable in this study was the range. What does range refer to?

1 point

Range refers to the set of values that occurs most commonly.
Range refers to the difference between the mean (average) and the median.
Range refers to the difference between the highest and lowest value in a set of data point values for a variable.
Range refers to the difference between the first and third quartile values.

3.
Question 3
The blood samples were stored at 20 degrees Celsius, with a standard deviation of 4 degrees Celsius. If the data point values were normally distributed, with no statistical outliers, what does this imply?

1 point

The minimum temperature at which the blood sample were kept was 16 degrees Celsius.
That in almost 70% of cases the actual temperature of the container was between 16 and 24 degrees Celsius.
No blood samples were kept in containers warmer than 24 degrees Celsius.
The median temperature at which the blood sample were kept must have been quite different from 20 degrees Celsius.

4.
Question 4
In Table 1 the (median) values for the first variable, RBCs (red blood cells), is expressed in cells (times 10 to the power 12) per liter of blood. What data type value is represented here?

1 point

Discrete
Interval-type numerical
Nominal categorical
Ratio-type numerical

5.
Question 5
The measure of central tendency used to describe the number of each cell type is the median. Consider the distribution of the actual data point values.

1 point

A binomial dustribution
The values will follow a Z-distribution
The values will follow a t-distribution.
A distribution of the actual values (how often each occurs) might be bell-shaped or take on any size of skewness.

6.
Question 6
To be entered into the study, participants had to be healthy. One of the criteria for classification as healthy was no prescribed medication (the baby / child should not be on any medication). If the data entry tool identified children that were not on medication as 0 (zero) and those on medication as 1 (one). Identify the data type for this variable.

1 point

Ratio-type numerical
Ordinal numerical
Nominal categorical
Interval-type numerical

7.
Question 7
Consider a possible follow-up study to this research, in which the same children were to be re-evaluated in five and again in 10 years. In the follow-up study researchers looked at their educational success and rates of certain illnesses. What study type would this represent.

1 point

Case-control series
Cohort study
Case series

8.
Question 8
The participants in this cross-sectional study were taken from the health clinics in one of the numerous informal settlements around a large city. What sampling technique would be describe this selection.

1 point

Stratified sampling
Simple random sampling
Systematic random sampling
Clustered sampling

9.
Question 9
A p-value:

Multiple options are correct. Please select all that apply.

1 point

is based on the calculation of a geometrical area.
takes a value between 0 and 1.
is the probability of an event occurring.
draws a curve and calculates the area under a certain part of that curve.

10.
Question 10
If there are five equally possible outcomes to an experiment, which are marked, in order, A, B, C, D, and E, what is the probability of finding a result of either D or E?

1 point

0.6
0.4
0.2
0.8

Week- 4

Peer-graded Assignment: Week 4 Peer review

Click Here To Download

Week 5 Graded Quiz

1.
Question 1
This and the following questions pertains to this article (the same paper used in a previous graded quiz) :

Full article link: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4514907/ or Link to PDF

In this paper the authors were concerned about the outcome of patients admitted with severe hospital acquired pneumonia. This refers to severe bacterial infection of the lungs. The patients were infected whilst in hospital. Bacteria that exist in the hospital environment are usually more difficult to treat due to resistance of the bacteria to antibiotics. Patients who contract hospital acquired bacterial infections may have a worse outcome than those who contract bacterial infection in the community (outside the hospital). These patients are in hospital for a specific medical reason, which usually means that their ability to fight infection might be compromised. Hospital-acquired bacteria can also be more virulent. It is this combination of a weakened immune system and aggressive, difficult to treat bacteria that lead to a worse outcome.

Consider the Materials and methods section. What type of test was used to compare continuous data types?

1 point

Mann-Whitney-U test
Fisher’s exact test
Parametric test comparing two groups, assuming equal variances
Non-parametric test

2.
Question 2
Which of the following was assumed when deciding to use Student’s t-test as one of the statistical analysis?.

1 point

The data point values for continuous data variables had a small variance.
The data point values for continuous data variables had a small standard deviation.
There was an equal number of participants in each group.
The data point values for the continuous data variables in the sample of participants were taken from a population in which those same variables were normally distributed.

3.
Question 3
Under the assumptions of a normal distribution of the required variable in the population from which a sample of participants are taken (as well as all other assumptions), which parametric test would be appropriate if the study design included self-controls?

1 point

t-test for dependent groups
Mann-Whitney-U tests
t-test assuming unequal variances
t-test assuming equal variances

4.
Question 4
If the investigators added a third group, using yet another antibiotic regime and this new group still met the criteria for the use of parametric tests, which test should be used instead of Student’s t-test?

1 point

Mann-Whitney-U test
Analysis of variance (ANOVA)
t-test assuming equal variances
Chi-square test

5.
Question 5
Consider the question above (the investigators adding a third group, using yet another antibiotic regime). Which would be an appropriate test for comparing numerical data point values if the assumptions for the use of a parametric test are not met?

1 point

Chi-squared test
Wilcoxon-Mann-Whitney test
Wilcoxon signed-rank test
Kruskal-Wallis test

6.
Question 6
The study mentioned above found no statistical difference between the mean ages of the two antibiotic groups (table 1 on page 1695). The mean age of the newer antibiotic group (piperacillin/tazobactam) was 79 years and that of the other groups was 80 years. If the Student’s t-test referred to a two-tailed p-value, what would the alternative hypothesis be stated as?

1 point

There is no difference between the mean ages of participants in the two groups.
The mean age of the piperacillin/tazobactam (newer antibiotic regimen) groups is higher than mean age of the other group.
There is a difference between the mean ages of the patients in the two antibiotic groups.
The mean age of the piperacillin/tazobactam (newer antibiotic regimen) groups is lower than mean age of the other group.

7.
Question 7
Consider the duration of antibiotic treatment (in days) for the two groups (table 1 page 1695). The mean number of treatment days for the piperacillin/tazobactam group was 6.7 +- 1.6 days. How should we interpret the plus/minus 1.6 days?

1 point

Some patients were treated for shorter than and some for longer than 6.7 days. The average difference between each patient’s actual length of treatment and 6.7 days was indeed 1.6 days, called the standard deviation.
No patient was treated for longer than 8.3 days
The shortest length of treatment was 6.7 – 1.6 = 5.1 days
An error occurred when this article was printed as patients only receive treatment for whole days (a discrete number of days, i.e. two days, four days or eight days).

8.
Question 8
If, instead of the standard deviation, confidence intervals were used for expressing the length of treatment (in days), how would you interpret the following result? The mean length of treatment for severe hospital-acquired pneumonia with piperacillin/tazobactam is 6.7 days (95% confidence interval 6.0 to 7.4 days.

1 point

Ninety-five percent of the patients were treated between 6.0 and 7.4 days.
Two-and-a-half percent (0.025) of patients were treated for more than 7.4 days and 2.5% (0.025) of patients were treated for shorter than 6.0 days.
Given the 95% confidence level, there is a 5% chance that a given patient in the population would need treatment longer than 7.4 days or shorter than 6.0 days.
If this study was repeated many times we would find that the true length of treatment was indeed between the lower and upper limits given in 95% of the studies.

9.
Question 9
Researchers quoted a p-value of 0.02 for the difference in effect of two drugs. What is the correct interpretation of a p-value?

1 point

Given a true null hypothesis, the probability of finding this particular difference (or a more extreme difference), was 0.02.
There is an 80% likelihood of having found no difference.
This is absolute proof that one drug is superior to the other.
There was a 20% likelihood of having found this difference.

10.
Question 10
Researchers find a strong positive correlation between alcohol intake and the incidence of hernia recurrence in patients who have undergone inguinal (groin) hernia repairs (p-value = 0.049, significance level 0.05). Which one of the following statements do you agree with?

1 point

Excessive alcohol intake causes the recurrence of hernias.
Excessive alcohol intake does not cause the recurrence of hernias.
The strength of correlation is not proof that alcohol causes hernia recurrence.
The p-value of 0.049 was too close to 0.05 to draw a conclusion from this result.

Week 6 Final examination

1.
Question 1
This scenario refers to question 1, 2 and 3.

A research paper states the following in the Methods and Materials section: “Cohorts of an equal number of participants (60 in each group) were taken from each of the three units and tasked with completing the survey. Participants were allowed to choose a single answer from the seven-point Likert choices following each statement.”

Further along in the Results section they mention:

“Question 2 of the survey shows a statistically significant difference between the three groups.”

What type of research would you consider this to have been?

1 point

Meta-analysis
Cross-sectional study
Cohort study
Experimental study

2.
Question 2
Pertaining to the research above.

What was the data type of the data point values compared to reach this significant difference between the groups?

1 point

Ordinal categorical
Nominal categorial
Ratio-type numerical
Continuous type

3.
Question 3
Pertaining to the research mentioned in the previous question, what type of sampling did the researchers employ?

1 point

Clustered sampling
Random sampling
Stratified sampling
Systematic random sampling

4.
Question 4

The following scenario refers to questions 4, 5, 6, 7 and 8.

The figure below is indicated in the Results section of a research paper. It identified four groups of patients and illustrates their analysis of the patients’ white cell counts. The researchers used a standard antibiotic drug and three novel (new) antibiotics, administered to patients admitted with bacterial pneumonia. The values below were taken from blood samples taken 24 hours after the start of treatment.

What type of plot is displayed in the figure above?

1 point

Count plot
Violin plot
Histogram
Box-and-whisker plot

5.
Question 5
Pertaining to the research above, which of the following would be an appropriate measure of dispersion of the white cell count data point values?

1 point

Standard deviation
Confidence intervals
Mode
Mean

6.
Question 6
Pertaining to the research above, indicate which of the four groups showed the biggest discrepancy between their mean (average) and median white cell count values.

1 point

Group 1
Group 2
Group 3
Group 4

7.
Question 7
Considering research to identify the most appropriate antibiotic, which of the following study types should be chosen to minimize the risk of bias and confounding?

1 point

Meta-analysis that includes high quality double-blind randomized controlled trials
Case-control study
Double-blind randomized controlled trial
Cohort study

8.
Question 8
Pertaining to the research above, it was found that patients receiving the standard drug had an average length of hospital stay of 6.7 days versus 5.9 days for the novel antibiotic groups (p = 0.04). For a chosen significance value of 0.05, what can be deduced from this result? Please choose the most appropriate answer.

1 point

There was no statistically significant difference in length of hospital stay between the standard drug group and the novel antibiotic groups.
The average stay among all the groups is 3.15
More patients will have to be recruited for a larger study.
There was a statistically significant difference in length of hospital stay between the standard drug group and the novel antibiotic groups.

9.
Question 9
The following scenario refers to questions 9 and 10.

Suppose one of the novel (new) antibiotics becomes approved for commercial use. A researcher wants to compare it’s efficacy in reducing patient temperature compared to the standard antibiotic drug used in their setting. The null hypothesis states that there is no difference in the average temperature on day 3 between the two antibiotics. Which is the correct alternative hypothesis?

1 point

There is a difference between the average temperature on day 3 between the two groups.
The novel (new) antibiotic will have a lower average temperature on day 3 compared to the standard antibiotic drug.
The standard antibiotic drug will have a lower average temperature on day 3 compared to the novel (new) antibiotic.
The standard antibiotic drug and novel (new) antibiotic drug will have a similar average on day 3.

10.
Question 10
Pertaining to the research above, it was shown that the data point values for one of the groups were not taken from a population in which the variable values are normally distributed. Which test would you have chosen to compare the white cell count of the four groups to each other?

1 point

Kruskal-Wallis test
Mann-Whitney-U test
Student’s t-test
ANOVA

11.
Question 11
The following scenario applies to question 11 and 12.

Researchers identified four groups of participants and compared the concentration of a certain drug in the blood stream of those participants and correlated it to the level of a certain hormone. Consider the following four graphs which appears in the Results section of their paper:

Which of the four graphs (groups A, B, C, or D) indicate the strongest inverse correlation between drug concentration and hormone level?

1 point

Group A
Group B
Group C
Group D

12.
Question 12
In considering the previous research paper, what type of statistical analysis would most likely have been used to compare the drug and hormone levels in the blood streams of the participants?

1 point

Mann-Whitney-U test
Kruskal-Wallis test
Student’s t-test
Linear regression

13.
Question 13

A study finds the mean cholesterol lowering capability of drug A as 2.0 mmol/L (95% CI 1.5 to 2.5 mmol/L) and that of drug B as 2.5 mmol/L (95% CI of 1.8 to 3.2 mmol/L).

Which of the following statements would you agree with?

1 point

Even though drug B lowers cholesterol levels more than drug A, there is a probability that the reverse might reflect the true population finding
Drug B is definitely more capable in lowering cholesterol levels than drug A.
Both drugs lower cholesterol levels equally.
Drug A is definitely more capable in lowering cholesterol levels than drug B.

14.
Question 14

The following scenario refers to questions 14 and 15.

Participants in the research above all took drug A for 4 weeks and were tested. This was followed by a period of abstinence for a minimum of 4 weeks before commencing a 4 week cycle on drug B. In comparing the mean (average) cholesterol levels between drugs A and B, which would be an appropriate statistical test?

1 point

Paired t-test
Student’s t-test
Mann-Whitney U
Chi-squared test for independence

15.
Question 15
The design of the study was altered to randomly assign two groups to either drug A or drug B. A large difference was found between the mean (average) and the median cholesterol levels (omitting any outliers) of these two groups, which would be an appropriate test to compare the two groups?

1 point

Student’s t-test
Chi-squared test for independence
Wilcoxon signed-rank test
Mann-Whitney U

16.
Question 16

The following scenario refers to questions 16 and 17.

The following is noted in the Results section of a paper. A questionnaire was sent to trainees in a surgery department. For the particular set of questions indicated, participants could rate their agreement to statements on scale of 1 (do not agree at all) to 5 (totally agree).

Pertaining to the research above, which statistical test would you advise for comparing the choices of the groups of trainees for this question.

1 point

Student’s t-test
Chi-squared test for independence
Wilcoxon-signed rank test
ANOVA

17.
Question 17
Pertaining to the research above, for a different section of the survey, the following contingency table was shown:

Trainees A Trainees B
Choice A 3 6 9
Choice B 4 4 8
7 10 17
Considering this table, what is the most appropriate test?

1 point

Chi-squared test for independence
ANOVA
Kruskal-Wallis test
Fisher’s exact test

18.
Question 18
The following scenario pertains to questions 18, 19 and 20.

A study looked at the results of four tests to diagnose acute calculous cholecystitis (infection of the gallbladder due to gallstones). PPV and NPV indicate positive and negative predictive value respectively. A table in the Results section shows the following:

Test Sensitivity Specificity PPV NPV
A 90.0 85 35.0 90.0
B 91.0 60 23.5 91.0
C 85.4 65 45.0 93.5
D 60.0 94 31.0 98.0
Tasked with choosing the test that would correctly indicate most patients with the disease, which would you pick?

1 point

Test A
Test B
Test C
Test D

19.
Question 19
Pertaining to the research above, the tests seem to have a high sensitivity, but a low positive predictive values. Which of the following could be a possible cause for this discrepancy?

1 point

The specificity is low.
The sensitivity is high.
A low prevalence of the disease in the sample population.
The disease is very common in the sample population.

20.
Question 20
Suppose we consider a test with a very high sensitivity (i.e. test A) to test a patient for the presence of a disease and the result from the laboratory is negative. Which one of the following statements is true?

1 point