Practice Midterm.docx

1) When reading an outbreak investigation report, you would expect a case definition to include:  The date of peak prevalence, incidence rate, and age-specific incidence rate A set of criteria for the investigative team to use to determine whether a person has a particular disease or health conditio...

1) When reading an outbreak investigation report, you would expect a case definition to include:  The date of peak prevalence, incidence rate, and age-specific incidence rate A set of criteria for the investigative team to use to determine whether a person has a particular disease or health condition The risk of contracting the disease being investigated The average incubation period, possible vectors, and known treatment options 2) A cross-sectional study can be used to:  Identify what health risk factors require urgent attention in a population of interest Assess the association between exposures and rare outcomes Determine a temporal relationship between an exposure and outcome Establish a case definition during an outbreak 3) When a histogram is skewed heavily to the left, we can expect that:  The standard deviation will be relatively small There are many outliers in the data set with large values The median will be identical to the mode There are many outliers in the data set with small values 4) If an alcohol use screening is performed in a primary care clinic for all patients who report consuming 4 or more alcoholic drinks per week, this would be considered a ______________ screening.  Universal Targeted/ Indicated Clinical Diagnostic 5) Under what conditions would you expect the positive predictive value of a screening test to decrease?  When the cutoff score for the test is raised When the cutoff score for the test is lowered In a population where the prevalence of the disease in question is low In a population where the prevalence of the disease in question is high 6) Standard error can be described as:  A statistic that indicates how precise an estimate the sample mean is A synonym for standard deviation A synonym for the 95% confidence interval The average distance between each value in a sample and the sample mean 7) Suppose you compute the odds of having cats as pets among individuals with and without Toxoplasmosis to be OR = 2.6 (95%CI=1.8, 3.4). Would you consider this a statistically significant result?  Yes, because the confidence interval does not include 0. Yes, because the confidence interval difference is greater than 0. Yes, because the confidence interval does not include 1. No, because we don’t know what the null hypothesis was in this scenario. 8) If reporting the median value as the measure of central tendency, you should also report the:  Interquartile range Standard deviation Standard error p-value 9) What type of graph is depicted below, and for what purpose is it used? Open-response. 10) Which of the following are true statements about the distribution in the histogram pictured? Choose all correct answers.  The median is greater than the mean. The standard error is greater than the standard deviation. It is normally distributed.   It is left skewed. The mean is greater than the median. It is right skewed. 11) Looking at the histogram in question 10, which measures of central tendency and spread would be best to report for this variable? 12) A study comparing post-baccalaureate incomes between US-born and foreign-born graduates of public universities found that the confidence interval around the mean difference included zero. The researchers used ___________________ test to study the association and ___________________ the null hypothesis.  a one-sample t-test; failed to reject a two-sample t-test; failed to reject an ANOVA test; rejected a chi-square test; failed to reject 13) Match the descriptions below to the correct variable type. Some variable types may be used twice, and some may not be used at all.  Labeling theatre-goers as youth, regular adult, or seniors Assigning each student in a class a random three-digit ID number Recording the distance, in miles, that each student travels to the UW campus from home Coding UW students as “0” if an undergraduate student or “1” if a graduate student A  Nominal B  Dichotomous/Binary C  Continuous 14) Say you are interested in knowing the mean difference in hospital charges (in dollars) between persons with and without Medicaid, and whether this difference is statistically significant. Please outline your analytical plan, i.e. your hypotheses, how you would analyze the data to answer the question, what statistics you will use to interpret the results of your analysis, and how would you decide whether you can reject the null hypothesis. You do not need to actually run the tests or provide any software code as part of your response, and you do not need to create a dummy table. 15) In August 2016, 400 incoming 9th graders and their parents were approached for participation in a longitudinal study of anti-social behaviors and college enrollment among high school students. Of the 40% who enrolled in the study and participated in a baseline assessment, 80 students met the criteria for anti-social behavior. The remainder did not meet these criteria. Participating students were assessed annually each fall, and at the fourth assessment, researchers found that 100 students met the criteria for anti-social behavior. State whether or not the following measures of disease frequency can be calculated, and briefly explain your response: Prevalence of anti-social behaviors at the time of study enrollment Incidence of anti-social behaviors during the study period (9th-12th grade) 16)  Imagine a new emotional health screening tool has been developed for use in elementary schools. The test has a very high specificity but somewhat low sensitivity. Suppose the principal of a local school asks for your expert opinion on implementing this screening in their school.  How would you explain the strengths and limitations of this tool? What are two recommendations you would provide the principal regarding implementation in their school? What is one thing you could suggest to increase the sensitivity of the test? 17) Read the scenarios below and determine whether they apply best to a case-control study, cohort study, neither, or both. Place an “X” in the appropriate box.  Scenarios case- control cohort neither both The study sample has been selected on the basis of outcome status. The results table presents relative risk to indicate the association between exposure and outcome. The exposure and outcome of interest have already occurred. The results are especially prone to recall bias. You will randomly assign participants to either be exposed to a treatment or not exposed. You are studying a rare exposure. You are interested in the incidence rate of a disease in the exposed and non-exposed. 18) One year after you complete your study on the effectiveness of the Starting Right program, health officials in that Midwestern state ask for your assistance in investigating a new increase in preterm births statewide during the last six months. Residents throughout the state were encouraged to receive a new measles vaccine, and some health officials have become concerned that the increase in preterm births is an adverse effect of receiving the vaccine within a month prior to conception. You agree to assist the state department of health with a case-control study to investigate the association between preterm births and the measles vaccine, especially so they can get in front of potential public concern. a. What is the exposure of interest in this scenario, and what is the outcome of interest? b. Describe who will comprise your study sample. What measure of association will you calculate to measure the association between your exposure and outcome of interest, and why? What test will you run to test the statistical significance of that measure the association?  e. What would be your null hypothesis in this scenario? f. Briefly describe one advantage and one limitation of conducting a case-control study in this scenario, rather than a cohort study. g. Provide two reasons you would have difficulty concluding a causal relationship in this study, even if you find a strong association between administration of the measles vaccine and preterm births.  h. Suppose you find that among 400 individuals who received the new measles vaccine, 40 had preterm births and among 800 individuals who did not receive the vaccine, 50 had preterm births. Estimate the strength of association between the new measles vaccine and preterm births. Show your calculations. 19) You are analyzing data for a study of graduate students that examined the association between sleep and watching cartoons about an animated sponge near bedtime: Watch 2 hours before bedtime Watch right before bedtime Do not watch before bedtime Sleep duration   Low  100 250 110    Normal  600  250 690   Excessive  300 500 200  What test will you run to test the statistical significance of the association between level of sleep duration and watching cartoons before bed, and why? While you were collecting data about sleep duration, you found that in the overall sample (N=3000), mean self-reported sleep duration was 6.4 hours (95%CI 5.1 – 7.7). Please interpret the mean and 95%CI in words. 20) After the discovery of the human immunodeficiency virus (HIV), five companies were licensed to produce enzyme-linked immunosorbent assay (EIA, then called ELISA) test kits for detecting HIV antibody.  Validation testing by the FDA found that the sensitivity of the ELISA test was 94.0%, and the specificity was 98.0%. Assume that the actual prevalence of HIV antibody in a hypothetical patient population is 1.0% (1/100). Using a 2-by-2 table for a hypothetical population of 5,000 individuals, compute the number of false positives and false negatives you would expect when screening this population using the ELISA test. Please show your calculations.