Statistics In Medicine 4th Edition Robert H. Riffenburgh, Daniel L. Gillen - Solutions

Unlock the full potential of "Statistics In Medicine 4th Edition" by Robert H. Riffenburgh and Daniel L. Gillen with our comprehensive solution manual. Access the answers key and solutions PDF to tackle solved problems with ease. Explore detailed questions and answers, including chapter solutions and step-by-step answers that guide you through each problem. Utilize the instructor manual for a deeper understanding of the textbook content. Our test bank provides a wealth of resources to enhance your learning experience. Enjoy the convenience of an online platform with free download options to support your studies in medicine.

Shown in Fig. 27.3 are the preoperative versus postoperative breast implant plasma silicon levels from DB5. Interpret the results of the Bland-Altman plot.
One question in the liver tumor study was whether or not patients’ BMIs were related to tumor size. As part of the process, we wanted to describe the BMI distribution. Like tumor size, BMI is limited on the smaller size but may have some quite large values, rendering it right skewed; we use the
Database DB24 (access on line) gives data from pediatric snakebite records. We ask if age makes a difference between being bitten in the hand/arm area versus the foot/leg area by season (seasons 1, 2, 3, 4). (Snakes and children are more active in warm seasons.) Using 4 seasons, m 5 4.We could
Suppose you are working with James Lind in 1747, attempting to isolate the cause of scurvy.After a long period at sea, a number of crew members are suffering from scurvy, characterized by weakness, fatigue, listlessness, bleeding gums, loose teeth, and subcutaneous bleeding. You note that scurvy
Give another example each of an epidemiological phenomenon that could be detected using crosscorrelation without lag, cross-correlation with lag, and (lagged) autocorrelation.
Give one example each of an epidemiological phenomenon that could be detected using crosscorrelation without lag, cross-correlation with lag, and (lagged) autocorrelation.
Sketch a rough survival graph from the life table survival results of Exercise 23.6.
Data13,14 on the survival of women with untreated breast cancer give rise to a life table with format and essential data as in Table 23.3. Complete the table. What is the probability that a woman with untreated breast cancer will survive 5 years?
A 2003 article7 addresses the occurrence of acute gastroenteritis caused by Norwalk-like viruses among crew members on the US Navy vessels Peleliu (2800 crew members) and Constellation(4500 crew members) visiting ports in Southeast Asia. During a period of rapid increase in the disease, 28 cases
In a Canadian study including the effect of caffeine consumption on fecundability, 6 2355 women drank caffeine-containing drinks and 89 did not. Of those who consumed caffeine, 236 conceived within 6 months of beginning pregnancy attempts. Of those who did not consume caffeine, 13 became pregnant.
List the three major steps in the evolution of epidemiologic methodology.
Suppose you wish to design a clinical trial to compare the 6-month change in testosterone levels among hypogonadism patients treated with a new experimental testosterone replacement therapy versus placebo. In total, 240 patients will be randomized in a 1:1 fashion to the new therapy or placebo. To
Again consider the randomized clinical trial designed to assess a new topical hemostatic agent for the treatment of moderate bleeding during abdominal surgery (see Exercise 22.1). The study sponsor believes that there may be a “learning effect” with the two agents to be used because surgeons
Consider a randomized clinical trial designed to assess a new topical hemostatic agent for the treatment of moderate bleeding during abdominal surgery. The new treatment will be compared to an existing treatment where patients will be randomized to one of the two arms at the first identification of
In a study13 on radiographic screening in a correctional facility in New York City, the rate of tuberculosis among entering inmates was 0.00767. A number of 4172 entering inmates were screened. How many were detected by the radiographic screening? Of those detected, 25 had entered with a prior
An ophthalmologist suspects that the effect of a beta-blocker (timolol) on IOP diminishes with age.3 He takes a pilot sample of n 5 30 patients, recording their ages and their reduction in IOP 4 weeks after initiation of treatment. He calculates the sample correlation coefficient as r 5 2 0.23. How
A pediatric surgeon is studying indicators of patient condition following pyloromyotomy (correction of stenotic pylorus) in neonates.9 A past study demonstrated that 14 out of 20 infants, that is, a proportion of 0.7, experienced emesis after surgery. How large a sample will he require to ensure
At one time an eye surgeon was performing radial keratotomies using hand surgery and considered a laser device.6 His hand surgery record showed that 23% of eyes required surgical enhancement.He planned a prospective study in which patients would be randomized into the hand and laser groups. On how
At one time, radial keratotomy was performed by residents in ophthalmology in a certain hospital.A review of records showed that 18% of residents required enhancements in more than one-fifth of their cases.5 This frequency of enhancements was thought to arise from the surgery learning curve.
In DB7 the distribution of bone density is not far from normal. From these data we calculate m 5 154 and s 5 24. A new study is to be conducted to estimate the population mean. Using these pilot data estimates, determine n such that the half width of a 95% confidence interval for the population
In DB3 the mean for the 16 patients’ differences between baseline and 5-day serum theophylline level is 0.993, with standard deviation 3.485. For such a difference to be significant withα 5 0.05 and power 5 0.80, n 5 97 pairs would be required. Generating three random samples of 97 on a computer
Suppose we did not have the EIB and eNO data as in DB14, but found a 5-minute difference standard deviation of 7 ppb for both normals and asthmatics. We want to test for a difference in means between these two groups using α 5 0.05 and power 5 0.80, where we take the clinically relevant difference
Suppose we could take a set of INR readings from the clinic and another set from the lab, as in DB13, but not paired. We want to test for a two-sided difference between means, where a difference of 0.25 INR is clinically relevant. We take α 5 0.05 and power 5 0.90. σclinic 5 0.54 INR and σlab 5
In DB12 the extent (centimeters) of the carinal resection appears to affect survival. Using the standard deviations given for surviving (Died 5 0) and dying (Died 5 1) patients, estimate the minimum sample size required for a test between means of resection extent of 0.37 for those two groups with
DB7 contains very small samples of bone density measures for men and women. On the basis of the standard deviations of those samples, estimate the minimum sample size required to detect a difference of 10 units of bone density between the means for men and women with α 5 0.05 and power 5 0.80 (β
An emergency medicine physician wants to know if mean HR following two particular types of trauma is different.2 He considers a mean difference of 6 bpm clinically meaningful. From pilot data, he finds s1 5 6.13 bpm and s2 5 6.34 bpm, which he uses to estimate σ1 and σ2. How many patients will he
An emergency medicine physician wants to know if mean HR following a particular type of trauma differs from the healthy population rate of 72 bpm.2 He considers a mean difference of 6 bpm clinically meaningful. He takes σ to be the 9.1 bpm reported in a large study. How many patients will he need?
Assess the clinical effects of false-positive and false-negative outcomes and specify whether β should be the usual 4α in estimating minimum sample size for a test of (a) the effect of ondansetron hydrochloride on nausea following gall bladder surgery in DB2 and (b) a change in plasma silicone
Should minimum sample size estimation be based on a one-sided or a two-sided test, assuming the underlying distributions are normal, of (a) a difference in nausea scores between ondansetron hydrochloride and placebo in DB2 and (b) the difference between assay types in DB8?
Harold Shipman was a rural physician in Britain who was discovered in 1998 to have murdered a large number of his patients. A recent article12 provided data on the cumulative number of death certificates by sex that he signed in excess of the local average, totaling 224. Starting from 1978 with the
The systolic time ratio (STR) indicates the strength and efficiency of the heart. Does STR relate to HR? Fig. 20.7A displays STR graphed against 228 patients’ HRs in increasing order.6 A linear regression shows no relationship, which just says that a straight-line fit does not have a significant
The survival table for 274 diabetic women during the decade 197080 is given in Table 19.5.The equivalent data on 370 women for the decade 198090 appears in Table 19.4. A logrank test of the difference between the two curves yields χ2 5 0.63. Find the critical value, compare the χ2 values, and
Life table data for survival of diabetic women in period 197080 is shown in Table 19.3.Complete the life table. What is the estimate of probability that a woman with diabetes survives 10 years? Graph a survival curve. Calculate and graph the 95% confidence intervals on this survival curve.
Using the model estimates given in Table 18.2, compute a 95% confidence interval for the rate ratio comparing subjects with dementia to cognitively normal subjects, after adjustment for both age and education.
Using the model estimates given in Eq. (18.46), find the predicted rate of repeats of CHD for a 67-year-old cognitively normal participant with 16 years of education, and a 72-year-old demented participant with 14 years of education. In each case, scale your answers to an observation time of 50
In the UCI ADRC data set, there were a total of ANorm 5 3887 words stated across all N 5 80 normal participants. There were a total of ADem 5 852 words stated across all N 5 31 demented participants. Further, there were YNorm 5 187 and YDem 5 104 repeats observed among normal and demented
Using the model estimates given in Table 17.3, compute a 95% confidence interval for the odds ratio associated with BMI after adjustment for both sex and age.
Using the model estimates given in Table 17.3, conduct a level 0.05 test of the null hypothesis that BMI is not associated with CHD after adjustment for both sex and age.
A better prediction would probably account for a person’s BMI. Using the model estimates given in Eq. (17.13), find the predicted probability of CHD for a 67-year-old female with a BMI of 27.2 and an 83-year-old male with a BMI of 22.4.
Using the model estimates given in Eq. (17.12), find the predicted probability of CHD for a 67-year-old female and an 83-year-old male.
Using the descriptive statistics for the data given in Exercise 15.2, calculate the correlation coefficient between age and respiration rate. How is this correlation coefficient interpreted?
In Exercise 15.3 poses the question: Could change in SBP (y) caused by laryngoscopy be predicted by SVmR (x)?4 Required interim statistics are n 5 26, mx 5 0.1771, sxy 5 2.2746, sx 5 0.1553, and sy 5 20.0757. Find se and sb. Look up the 95% critical t. Test H0:β1 5 0. Is the slope significantly
Calculate the coefficient of determination for the data of Exercise 15.4. How is this coefficient of determination interpreted?
Calculate the coefficient of determination from the descriptive data in Exercise 15.2. How is this coefficient of determination interpreted?
As part of a study on circulatory responses to intubation during laryngoscopy, 4 the following question arose: Could changes in systolic blood pressure (SBP, y-axis) due to laryngoscopy be predicted by skin vasomotor reflex amplitude (SVmR, x-axis)? Readings (these readings from a graph may be off
It is believed that respiration rate in infants decreases with age. To examine that belief, respiration rate was recorded for 232 infants from 0 to 24 months.1 The scatter diagram of respiration rate against age is shown as Fig. 15.3. Identify the dependent and independent variables. Is there a
Is the esophageal thermometer reliable, that is, does it give internally consistent temperatures? The thermometer is read repeatedly five times on each patient. k 5 5 and n 5 48. 5(5 2 1)/2 5 10 possible pairs. The correlation coefficients for these pairs are 0.9069, 0.8676, 0.7769, 0.5406, 0.9797,
Results of a tumor board’s deliberations. Tumor boards are conferences used to provide multidisciplinary input into clinical decision-making for cancer patients. A tumor board composed of eight members considered the following case.6 A 68-year-old man had a gradual onset of dysphagia associated
Let us continue the clinic versus laboratory INR of Exercise 14.6. DB13 also presents paired in-range (0) versus below-range (21) and above-range (1) readings.We are interested in determining whether the clinic and laboratory agree on the reading, not on quality control (keeping the INR within
DB13 gives paired INR readings for 104 patients paired from the hospital laboratory and the Coumadin Clinic. For these data, r 5 0.900, showing a strong pattern of agreement. However, the mean of differences is 0.125. A paired t-test with the null hypothesis stating that the mean 5 0 yields p-value
For the 2 3 5 table in DB2 for nausea scores of 15 against drug versus placebo, calculate and interpret γ.
Let us consider the alterations in BP of pregnant women from changing position from upright to supine, given in Table 9.10. Is SBP or DBP more closely associated with a drop in BP from lying down? (Moving down in the table represents going from SBP to DBP.)
For the 2 3 2 table in Table 5.3 [constructed from database (DB) 2] illustrating the frequency of patients with no nausea versus nausea against drug versus placebo, calculate and interpret Φusing Eqs. (14.1) and (14.2).
Are the distributions of functionality the same for two types of ankle repair? An orthopedist installs hardware in the repair of broken ankles in n 5 19 patients.7 He randomly selects two groups: leave the hardware in place after adequate healing (n1 5 10) and remove it (n2 5 9). He judges the post
KS. Hematocrit's (Hct’s) need to be distributed normally to find a confidence interval. In the additional example of Section 11.8, we examined postoperative Hct on pyloromyotomized neonates6 and concluded that Hct was not different for laparoscopic versus open surgeries; thus we pool the data.
.)We want to test the hypothesis that the true mean temperatures are the same in the left (L) and right (R) ears (Exercise 11.2), but we want to investigate if the equal-variance t test may be appropriate. mL 5 96.41F, sL 5 0.88F, mR 5 97.86F, and sR 5 1.12F. At theα 5 0.05 level of significance,
Is the variability of the tympanic thermometer readings the same in both ears? We are investigating the reliability of a certain brand of tympanic thermometer (temperature measured by a sensor inserted into the patient’s ear). Sixteen readings (F), eight per ear, were taken on a healthy patient
.) In our clinical judgment a reliable thermometer will be no more than 1F off 95% of the time. This implies that 1F is about 2σ, so that σ 5 0.5. From the data, s 5 1.230. At the α 5 0.05 level of significance, is the variability of the tympanic thermometer reading on a patient unacceptably
Is the variability of readings from a tympanic thermometer too large? We are investigating the reliability of a certain brand of tympanic thermometer (temperature measured by a sensor inserted into the patient’s ear). Sixteen readings (F) were taken on a healthy patient at intervals of 1 minute,
Data on recovery after orthopedic surgery are given in DB10. We ask if triple hop time on the operated leg is not larger than (noninferior to) that on the nonoperated leg. We judge that recovery, that is, Δ, is defined as operated leg time no greater than 10% of nonoperated leg time. Do we have
Is experience associated with the rating of severity of cancer by radiologists reading radiographs? Three radiologists 2 a PGY1 resident (1), a PGY4 resident (2), and a senior staff member (3) 2 read radiographs of 10 cancer patients and rate the severity of disease on a 15 scale (DB35). For each
Overnight recordings of snoring loudness by 3776 sleep disturbed patients were grouped by BMI into normal ( # 25), overweight ( . 25 but # 30), and obese (. 30) (DB27). The groups were designated 1, 2, and 3, containing 363, 1106, and 2307 patients, respectively. The data were ranked by average dB
In Exercise 11.9, n1 5 n2 5 8 temperature readings were taken from a patient’s left and right ears, respectively.4 The question was, “Are the distribution centers for the two ears different (implying a two-tailed test)?” As an exercise, answer the question using the normal approximation to
n 5 8 temperature readings taken on a patient were 98.1, 95.8, 97.5, 97.2, 97.7, 99.3, 99.2, 98.1. When centered at 98.6F, is there sufficient evidence to conclude that the resulting distribution giving rise to these data is not symmetric about 0? As an exercise, answer the question using the
Does a certain tympanic thermometer measure the same in both ears? We are investigating4 the reliability of a certain brand of tympanic thermometer (temperature measured by a sensor inserted into the patient’s ear). Sixteen readings ( F), eight per ear, were taken on a healthy patient at
Normality of thermometer readings on a healthy patient. We are investigating the reliability of a certain brand of tympanic thermometer (temperature measured by a sensor inserted into the patient’s ear).4 Eight readings (F) were taken in the right ear of a healthy patient at 2-minute intervals.
(Unequal variances). Testing the effectiveness of a vaccine to inhibit human immunodeficiency virus (HIV). The vaccine was tested by randomizing HIV patients into unvaccinated(treatment 1) and vaccinated (treatment 2) groups and compared the number of HIV per mL of blood.5 The data format was as
Mean bone density for men versus women. In DB7, perform a t test to learn if there is evidence that the mean bone density is different between men and women. 1 2 Table 11.8 Human immunodeficiency virus levels for vaccinated versus unvaccinated patients. Patient 3 24 25 44 45 46 No. of 134 19,825
Theophylline level change for men versus women. In DB3, use the difference between theophylline levels at baseline minus at 5 days and perform a t test to learn if there is evidence that the mean reduction in level due to the drug differs between men and women.
Survival of carinal resection patients. In DB12 the sample is large enough to take the sample standard deviation as if it were a population standard deviation σ. For convenience, carry the calculations to only two decimal places. Perform a normal (z) test to learn if there is evidence that(1) mean
. As either ear may be higher, the alternative hypothesis is two sided. “L” denotes left, “R”, right. mL 5 96.41F, sL 5 0.88F, mR 5 97.86F, and sR 5 1.12F. At the α 5 0.05 level of significance, if we treat the samples within the patient as independent are the means of the left and
The original example that “student” used for his t. In W.S. Gossett’s classic 1908 paper3 introducing the t test, he used the following data. Two soporific drugs were used in turn on 10 insomniac patients and the number of hours of additional sleep each provided was recorded. Data were as in
Is loudness or duration of snoring a better predictor of the overweight influence on sleep disturbance? In an overnight sample of 3776 patients experiencing sleep disturbance, snoring in excess of 50 dB was recorded for both average loudness and duration.7 The patients were classified using body
Can white blood cell (WBC) identify strep A? In a sudden outbreak of Streptococcus pneumonia A (DB20), can WBC separate strep A from other strains? A ROC curve shows sensitivity and specificity for strep A for all recordings of WBC from 128 cases of infection (Fig. 10.2). What would be a reasonable
Confidence on the OR of coccidioidomycosis. From Table 10.2, we found OR 5 5.3040.Use the table data to put a 95% CI on the true OR excluding the 0.5 continuity correction).
Coccidioidomycosis dependent on dust exposure.6 For the data of Table 10.2, calculate and interpret the AR and rt.
on ligament repair4 (Table 9.6) to calculate LRR and se(LRR) and test the hypothesis that the LRR, λ, is equal to 0, that is, that there is no difference between the Modified Bostrom and ChrismanSnook methods of repair.
Is the quality of ligament repair significantly associated with the method of repair? Use the data in
Appendicitis diagnosis. Table 10.5 displays the Massachusetts General Hospital appendicitis data1 as a truth table. Calculate and interpret sensitivity, specificity, accuracy, PPV, NPV, RR, OR, LR, and AR.
Does the time of eating affect total cholesterol? The medical community has long believed that total cholesterol will be greater after a meal than after fasting. Is such an effect greater than would occur by chance? On different days, we measure the total cholesterol for three designs: after
Are boys more accident prone than girls? You suspect that boys between the ages of 3 and 5 years are more accident prone than girls. You identify families with a boy and a girl and check their medical records for accidents during those ages.4 Your data are displayed in Table 9.21.Use McNemar’s
Does blood-bank screening reduce hepatitis-infected blood? Studies through 1992 found the rate of viral hepatitis infection caused by red blood cell transfusions in the United States to be 1 per 3000 units transfused, or π 5 0.000333. Introduction of second- and third-generation hepatitis C
Large n: do boys and girls have limb fractures in equal numbers? Theoretically, the population proportion would be π 5 0.5 if the proportion were sex independent. We recorded the sex of the next 50 children presenting to the orthopedic clinic with limb fractures and found 34 boys.What do we
Small n: are more boys than girls born in this family? A small-town obstetrician has been delivering babies from a particular family for three generations. He notes a high proportion of male babies and suspects he may be seeing a genetic phenomenon. Of 16 babies born during his experience, 11 are
Change in blood pressure (BP) resulting from change in position. The influence of patient position during the measurement of BP has long been questioned. A 1962 article6 examined alterations in BP of 195 pregnant women resulting from a change from upright to supine positions. The data are given in
Do clinic versus laboratory tests have different out-of-range rates? In DB13, combine below-range with above-range result categories (last two columns) to form an out-of-range category.Form an in-range versus out-of-range 2 3 2 contingency table and use χ2 to decide if clinic and laboratory result
Does ondansetron HCl reduce nausea in gall bladder patients? In Table 5.3, test the 2 3 2 contingency table using chi-square to decide whether or not nausea score is reduced by the drug.
Are tattoos using titanium ink harder to remove? In DB6, test the contingency table using chisquare to decide whether or not titanium-containing tattoo ink is harder to remove than other inks.
Have protease inhibitors reduced pulmonary complications from HIV? In DB4, test the contingency table using chi-square to decide whether or not access to protease inhibitors has reduced the rate of pulmonary admissions.
An orthopedist is experimenting with the use of Nitronox as an anesthetic in the treatment of arm fractures in children.1,2 He finds that it provides a shorter duration of surgery and less pain than other procedures he has used. In addition, he expects that the shorter the procedure, the less the
, find 95% CIs for the standard deviation of the baseline data for each treatment. Are they similar?
Using the sample sizes and standard deviations given in
is concerned not only about the average heart rate of the trauma patients but also about the variability. Even if the mean is satisfactory, if the standard deviation is too large, outlying patients would be at risk. Recall that n 5 8, m 5 67.75 beats/min, and s 5 9.04 beats/min. Establish 95% CIs
The ED physician of
is concerned about the variability of time to full feeding in neonates. Even if the mean is satisfactory, if the variability is too large, the outlying neonates on the longer side of the scale would be at risk. Recall that n 5 20, m 5 6.56 hours, and s 5 4.57 hours. Establish 95% CIs on (a) the
The surgeon referenced in
The variability in platelet-derived growth factor (PDGF) has been estimated (DB9) as s 5 19,039.7. How far from this may a sample standard deviation deviate before we suspect the influence of a new factor? Find a 95% CI for the population σ.
An infectious disease specialist samples the white blood cell count of n 5 19 patients who contracted the same nosocomial infection in an orthopedic ward. She finds that the mean count is not unusual but suspects that the standard deviation may be larger than normal: s 5 6000. What is her 95% CI
Based on a sample of 226 patients, a 5-year recurrence rate of basal cell carcinoma when lesions measured more than 20 mm was 0.261.7 Compute a 95% CI for the population proportion, π.
In the kidney cooling data of DB15, the baseline (time 0) means are m1 5 34.40 for saline infusion and m2 5 36.12 for ice slush, each based on 12 readings. (Use 2 decimal places throughout for easier calculation.) The respective estimated standard deviations are s1 5 3.25 and s2 5 1.17. Find 95%

Showing 1 - 100 of 5416

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Last

Statistics In Medicine 4th Edition Robert H. Riffenburgh, Daniel L. Gillen - Solutions

Step by Step Answers