BME423 Fall 2015: Homework 4

| August 30, 2017

Question
BME 423 Fall 2015: Homework 4

Due by noon on Friday, December 4th, 2015. Please read the homework guidelines before starting your assignment. Statistical tables necessary for completing this assignment are posted on Blackboard.

1. Ten children with autism were enrolled in a study to assess the effectiveness of a new drug designed to reduce repetitive behaviors in this population. The amount of time that each child spent engaged in repetitive behavior during a three-hour observation period was measured both before treatment and then again after taking the new medication for one week (see table below). It has been determined that these data are NOT Normally distributed. Is there any evidence of a difference in duration of repetitive behaviors between the two observation periods?

Before Treatment After 1 Week of Treatment

87

76

72

51

44

52

65

43

82

20

74

67

53

38

89

95

36

19

68

47

2. A physician was interested in the effect of an anesthetic on reaction times. Two groups were compared: one group was given the anesthetic, while the other was given a placebo. Subjects had to press a button when they identified a specific visual stimulus and their reaction time (the time between when the stimulus was shown and when the subject pressed the button) was measured in milliseconds. The table below contains the data for the two groups.

With Anesthetic Without Anesthetic

245

139

135

142

191

163

178

147

256

138

141

230

145

143

267

189

167

171

165

156

244

144

142

131

289

268

Do these data support the hypothesis that there is a difference in reaction time between people given the anesthetic and people not exposed to it? Use an appropriate nonparametric test and the corresponding parametric test. Discuss any differences in the two conclusions. Note: you still need to “test” the null hypotheses, even though the question was phrased using an alternative hypothesis.

3. To assess the effects of expectation on the perception of aesthetic quality, an investigator randomly sorted 24 amateur wine aficionados into three groups (A, B, and C) such that each group was made up of eight subjects. One member of group A and two members of group C did not show up for their interview/testing session, dropping the total number of subjects to 21. The subjects who did show up for their interviews were each asked to rate the overall quality of a particular wine on a 10-point scale, with “1” standing at the bottom of the scale and “10” at the top. As it turns out, all subjects tried the same wine and the only difference among the groups was the style of the interview, which was designed to induce a relatively high expectation of quality in members of group A, a relatively low expectation in members of group C, and a neutral state/expectation in members of group B. Do the data in the table below support the hypothesis that there is a difference in reported quality/enjoyment among the three groups of tasters? Note: you still need to “test” the null hypothesis, even though the question was phrased using an alternative hypothesis.

Wine Ratings

Group A

Group B

Group C

6.4

2.5

1.3

6.8

3.7

4.1

7.2

4.9

4.9

8.3

5.4

5.2

8.4

5.9

5.5

9.1

8.1

8.2

9.4

8.2

9.7

4. A group of seven experts rated four brands of coffee in a taste-testing experiment on a scale of 4 (the lowest possible score) to 28 (the highest possible score). The table below displays these ratings; is there evidence of a difference in the ratings of the four brands of coffee?

.gif”>

5. A study was done on a treatment for preventing additional heart attacks in subjects who have already experienced one or more heart attacks. The treatment was administered as soon as each patient was diagnosed with a heart attack in the emergency room. The table below shows the time between diagnosis/start of treatment and a subsequent heart attack in these patients; censored (loss to follow-up) observations are included (indicated by a “+” next to their entry).

a. Construct a survival table and show the following columns: # without an event at beginning of each interval; # who experienced an event (i.e., a heart attack) at end of interval; fraction who did not experience an event in interval; cumulative fraction without an event.

b. Using the data from part (a), draw a Kaplan-Meier plot. What is the median time to event (i.e., what is the median time before another heart attack occurs)?

Patient Number

Time to Event (days)

1

95

2

132+

3

125

4

137

5

141

6

110+

7

137

8

185+

9

156

10

79

11

140

12

132

6 . Choosing the appropriate test

Choosing the appropriate statistical test is a critical first step in any biostatistics problem. Please give the name of the statistical test or procedure that should be used for each of the following experiments. The answer to most of the questions may be found in the table on the inside cover of your textbook. You will need to ask yourself the following questions:

i. Are there:

§ 2 groups?

§3 or more groups?

§ 2 random variables that may or may not be associated?

ii. If you have groups, do they consist of

§ Different subjects?

§The same subjects? iii.Are the data

§Interval? (e.g., age, height, blood pressure)

ØIf so, are they normally distributed?

§Nominal? (e.g., died/survived)

§Ordinal (e.g., pain rating measured from 1 to 5, satisfaction rating measured from

1to 10)

a. Fifty-two bipolar patients are enrolled in a clinical trial. Half of them are given olanzapine, and half are given lithium. The investigator reports whether there were any seizures during treatment for patients in both groups. After finding a difference between the two groups using an appropriate statistical test, you want to know how much of an association there is between seizures and lithium treatment as compared to olanzapine.

b. Eighteen patients with obstructive sleep apnea are fitted with an oral appliance to prevent apnea. Each patient’s apnea and hypopnea index (AHI) is measured with and without the appliance. The results are not normally distributed. You want to know whether there is a difference in AHI with and without the oral appliance.

c. You are studying a group of cancer patients, and you want to know whether a new medication prolongs their survival time as compared to subjects who receive the standard treatment.

d. Two groups of 13 teenagers each are asked to rank how happy they are after they watch TV (first group) or after they play a video game (second group). You want to know whether there is a difference in the perceived happiness of teenagers after watching TV and after playing a video game.

e. Migraine patients are divided into three groups and treated preventively for a week with acupuncture (group 1), aspirin (group 2), and indomethacin (group 3). You want to know whether the choice of treatment affects the proportion of patients who develop at least one migraine during the week of the treatment.

f. The blood pressure of 8 subjects is measured when they are asleep, when they are meditating, and when they are watching The Simpsons. All the data are found to be Normally distributed. You want to know if there is a difference in blood pressure in the three situations.

Get a 30 % discount on an order above $ 100
Use the following coupon code:
RESEARCH
Order your essay today and save 30% with the discount code: RESEARCHOrder Now
Positive SSL