EPID 4400 Final Examination Preparation with complete solutions., Exams of Epidemiology

Download EPID 4400 Final Examination Preparation with complete solutions. and more Exams Epidemiology in PDF only on Docsity! EPID 4400 Final Examination Preparation with complete solutions Case-fatality rate - ANSWER-Numerator - (Number of deaths from a disease) Denominator - (Number of cases of that disease) [Section 6] Exam 1 Important guidelines for evidence of a causal relationship. - ANSWER-- Dose-response relationship - Temporal Relationship (time) - Replication of findings - Strength of association EXCLUSION: Effect Modificaiton [Section 5] Exam 1 If the steady state assumption is satisfied, prevalence of a disease is mathematically related to... - ANSWER-- Average duration of the disease - Incidence of the disease [Prevalence = (incidence) x (average duration of the disease)] Exam 1 Choose rate or ratio that best describes the following statement: "At the initial examination, 31 persons aged 40 to 49 had coronary heart disease per 1,000 persons examined in this age group" - Age-adjusted mortality rate - Age-specific prevalence rate - Incidence rate - Prevalence rate - Standardized mortality ratio - ANSWER-Age-specific prevalence rate [This rate is distinguishable by the phrase per 1,000. The rate is broken down into an age group, which is an age-specific rate. However, no evidence of standardization is given. These are existing cases, not new cases.] Exam 1 In the classic or traditional sense, epidemiology is best defined as the study of the... - ANSWER-Distribution and determinants of health-related states or events in humans [Section 1] Exam 1 When a new treatment is developed that prevents death but does not produce recovery from a disease, the following will occur: - ANSWER-Prevalence of the disease will increase [Prevalence = (incidence) x (average duration). Prevention of death increases the average duration of disease, thus increasing prevalence.] Exam 1 A weighted average of the individual age-specific mortality rates in the study population, with the weights being the proportion of a standard population in each category, is the definition of the -Crude rate -Direct method of age-standardization -Indirect method of age-standardization -Proportionate mortality ratio -Standardized mortality ratio - ANSWER-Direct method of age-standardization [Definition of direct adjusted rate. Section 6] Exam 1 The sequence of studies in human populations typically begins with: - ANSWER-A descriptive study [Descriptive studies are hypothesis generating. The data and info collected here (especially person, place, and time) informs our analytical studies (hypothesis testing with a comparison group)] Exam 1 When calculating an age-standardized mortality ratio (SMR), what is needed? -Age-specific mortality rates in the sample (or study population) -Total number of observed deaths in the standard population -Both -Neither - ANSWER-Neither SMR = observed / expected The observed is the "number of deaths in the study population." The expected is calculated by taking age-specific rates of a standard population and applying these to the distribution of each study population. Exam 1 Which statement is not true? [The accuracy and detail (or lack thereof) of diagnoses listed as underlying causes of death is a disadvantage of relying on the cause of death data.] Exam 2 Most appropriate descriptive term? Febrile respiratory tract infections develop in approximately 80 percent of children infected with influenza. - Communicability - Immunogenicity - Incidence - Pathogenicity - Virulence - ANSWER-Pathogenicity [Refers to the number of cases of disease divided by the number of people infected. In this case, 80% of the children infected developed the disease.] Exam 2 A survey revealed that of 225 families in which there had been a known case of poliomyelitis, 56 had a dog as a family pet. In another survey, 30 of 99 poliomyelitis patients had a dog. The inference that dogs are a risk factor for poliomyelitis is... - Correct - Incorrect because a proportionate ratio is used when a rate is required to support the inference - Incorrect because of failure to distinguish between incidence and prevalence - Incorrect because of failure to recognize a possible cohort phenomenon - Incorrect because there is no control or comparison group - ANSWER-- Incorrect because there is no control or comparison group [Cross-sectional (survey) data is not a good study design for causal inference. There is no information on people who had dogs, but did not have poliomyelitis. Both groups of people discussed in the problem are groups of known cases of poliomyelitis and the number of those who had a dog as a pet. Although the survey data is on two groups, they are similar groups of patients. No comparison group of people who had dogs, but had no poliomyelitis is discussed. As this is cross-sectional data, we also do not have information on the temporality of the association (i.e. which came first the dog or poliomyelitis).] Exam 2 Effect modification - ANSWER-Occurs when the magnitude of the chosen measure of association between a causal agent and a disease differs according to the level of a third variable. Exam 2 In an observational analytic epidemiological study of a potential protective factor, the true risk ratio was 0.6, and the study had a bias away from the null hypothesis. The study risk ratio was most likely which of the following: - 0.3 - 0.7 - 1.0 - 2.0 - 3.0 - ANSWER-0.3 [Since we are told there is a bias away from the null (i.e. a value of 1.0), then the study value must be beyond the true value (away from 1). Since the true value is 0.6, it cannot be higher than 0.6, which would bias it toward the null hypothesis. Therefore, the answer must be 0.3, which is the only value provided that is further away from 1 on the protective side.] Exam 2 Most appropriate descriptive term? Approximately 50 percent of household contacts of a child who has a common cold become infected. - Communicability - Immunogenicity - Incidence - Pathogenicity - Virulence - ANSWER-- Communicability [Communicability refers to the ability of a pathogen to be transmitted from one host to another. The communicability of the common cold in this case is 50%. This means that 50% of the people with a cold will pass it along to someone else.] Exam 2 Usefulness of mortality data as an indicator of disease frequency is more accurate when the disease of interest... - ANSWER-- Has a high case fatality rate - Is relatively easily diagnosed Exam 2 The causes of death on a death certificate can be certified by the... - ANSWER-Medical examiner or coroner Exam 2 Cystic fibrosis is an autosomal recessive genetic disorder that is caused when an individual has mutations in both copies of their CFTR gene (i.e. has two mutations). Carries of cystic fibrosis have a mutation in only one copy of their CFTR gene (i.e. have only one mutation). Common screening panels for cystic fibrosis may not detect all types of genetic mutations of the CFTR gene that can occur (i.e. individuals with CFTR mutations in one or both of their CFTR genes may not be detected). A study comparing carries of cystic fibrosis is conducted where these two groups of patients are identified based on having only one mutation or two mutations detected by a common screening panel, respectively. What potential type of bias may occur here? - Surveillance Bias - Recall Bias - Misclassification Bias - Exclusion Bias - ANSWER-- Misclassification Bias [Misclassification bias may occur if some of the genetic mutations are missed by the screening panel. If only one mutation is detected in a subject who has cystic fibrosis (2 mutations), that individual will be misclassified as a cystic fibrosis carrier and not actually having cystic fibrosis. In other words, some of the identified "carriers" may actually have a second mutation that is missed by the screening panel, and should be correctly classified as a case of cystic fibrosis] Exam 2 A study of mortality patterns in 20 countries found a positive correlation that was statistically significant (p-value < 0.001) between cigarettes sold per capita and mortality rates from cancer of the oral cavity. This finding could be explained by... - Cigarette smoking being associated with cancer of the oral cavity - Countries with high cigarette consumption rates having populations that are more susceptible to cancer of the oral cavity for reasons other than smoking - Both - Neither - ANSWER-- Both [The significance of the association does not explain how the association occurred. Both explanations are viable. There is no information to indicate otherwise (i.e. confounders that have been included in the analysis, etc.).] Exam 2 Investigators examining the association between daily aspirin use (75mg) and colon cancer risk report a relative risk of 0.89 (95% CI 0.75-1.12) comparing aspirin users to non-users. From the information presented we can say that: - The authors found an 11% reduction in risk of colon cancer with daily aspirin use, that is unlikely due to chance alone. - The effect estimate is statistically non-significant at the 5% level (alpha 0.05) - Both - Neither - ANSWER-The effect estimate is statistically non-significant at the 5% level (alpha 0.05) [Section 10. If the finding was statistically non-significant, then we cannot conclude that option A (11% reduction in risk) was correct.] Exam 2 there will be a difference in how the groups are reporting their behavior (differential misclassification). This can lead to a bias either towards or away from the null, as the bias is likely differential between case groups.] Exam 2 To conclude that an observed association between an exposure and disease is internally valid, an investigator should evaluate or reject as alternative explanations for the research findings all of the following... - ANSWER-- Bias - Chance - Confounding - Interaction (effect modificatin) EXCLUDES: Generalizability [Section 10. Generalizability refers to external validity, not internal validity.] Exam 2 In a study of 50 cases and 50 controls, it is determined that the difference found with respect to a possible etiologic factor is not statistically significant at the 5% level. One may conclude from this finding that... - Chance or random error is not a possible explanation - Confounding has been eliminated - Observer or interviewer bias has been eliminated - The lack of statistical significance is solely the result of sampling variation - None of the above - ANSWER-- None of the above [Statistical significance refers to there being less than a 5% probability of observing a result as extreme as that observed due solely to chance. This means that the association is not likely to die to chance. If something is not statistically significant then chance could still be a viable explanation. Because of the limited information, one of the other options can really be concluded.] Exam 2 Most appropriate descriptive term? Death occurs in approximately 20 percent of cases of pneumococcal meningitis. - Communicability - Immunogenicity - Incidence - Pathogenicity - Virulence - ANSWER-- Virulence [Virulence refers to the cases of severe and fatal disease divided by all cases of disease. In this case, one out of five cases of pneumococcal meningitis resulted in death, an indication of virulence.] Exam 2 In a case-control study of the relationship between smoking and myocardial infraction, controls (those without MI) are matched to cases (those with MI) on the basis of race, sex and geographic location, is intended to decrease the chance of which of the following biases? - Confounding bias - Interviewer bias - Non-response bias - Recall bias - Surveillance bias - ANSWER-Confounding bias [Section 10. Part 2. Matching is a method used to control for confounding. It is incorporated into the study at the design stage and taken into consideration during analysis.] Exam 2 The CDC provide guidelines when investigating suspected cancer clusters and responding to community concerns. In which stage of this four-step process would an excess incidence be calculated? - ANSWER-Step 2 [Section 9, slide 14. Assessment of an excess in incidence would be undertaken in Step 2.] Exam 2 Consider a disease with two alleles, A and a. The mating types that could produce a heterozygous child are - Aa x aa - AA x aa - Aa x AA - Aa x Aa - All of the Above - ANSWER-All of the Above (Need a large A and small A to be heterozygous) Lecture 20 Clicker Quiz - Final Exam Consider a disease with two alleles, A and a. Which mating types give the highest proportion of heterozygous offspring? - Aa x aa - AA x aa - Aa x AA - Aa x Aa - All of the above - ANSWER-AA x aa [Each parent will give one of their alleles to their offspring. In an AA x aa combination, the only type of allele parent one can give is A, and the only type of allele parent two can give is a. This means that every combination possible for the offspring will be Aa, a heterozygous pairing. All other choices will have at least one homozygous pairing, either AA or aa.] Lecture 20 Clicker Quiz - Final Exam It is possible for a male to have received a sex chromosome from one of his four grandparents. For a male child, which grandparent could not have transmitted, via his parents, a sex chromosome? - Maternal grandfather - Maternal grandmother - Paternal grandfather - Paternal grandmother - ANSWER-- Paternal grandmother [A male has the sex chromosome pairing of XY, while a female has XX. Since we are talking about a male child, we are looking at how he inherits his XY pairing. The X will come from his mother, who could have received that X from either her mother or her father. Therefore, either his maternal grandfather or maternal grandmother may have contributed that X. The male offspring would have received his Y from his father. However, only males can pass the Y chromosome to offspring. Therefore, the Y the son received must have been passed on from his paternal grandfather. This means that the paternal grandmother could have contributed a sex chromosome.] Lecture 20 Clicker Quiz - Final Exam The risk of occurrence in first degree relatives of people with disease X is compared with first degree relatives of people without disease X. The type of study represented by this activity is best described as a (an) - Adoption study - Inbreeding study - Migrant study] - Recurrence risk - Twin study - ANSWER-- Recurrence Risk Lecture 20 Clicker Quiz - Final Exam In general, screening should be undertaken for diseases with each of the following characteristics - Diseases with high prevalence in identifiable subgroups of the population - Diseases for which case-fatality rates are high - Diseases with a natural history that can be altered by medical intervention - Diseases that are difficult to diagnose - Diseases for which treatment efficacy has been shown to be effective - ANSWER-- Diseases that are difficult to diagnose The same screening test is applied in a uniform manner in two populations. The prevalence of the disease being screened is 10% in population A and 1% in population B. Which of the following statements is true? - Reliability is higher in population A than in population B. - Sensitivity is higher in population A than in population B. - Specificity is higher in population A than in population B. P.S. 1 Describe the usual pattern of reasoning in epidemiology? (Match the ordered steps to their logical process of reasoning). - ANSWER-Step 1: Develop a hypothesis Step 2: Test the hypothesis by comparing an exposed human population and an appropriate control group Step 3: Systematically collect & analyze data Step 4: Assess the validity of any observed statistical associations Step 5: Judge whether the observed associations represent cause-effect relationships P.S. 1 During the past summer, a survey is conducted to determine the prevalence of aortic valvular stenosis and associated conditions within the general population of Iowa. What study design is this? -Case-control study -Randomized Controlled Trial (RCT) -Cohort study -Correlational study -Cross-sectional study - ANSWER-Cross sectional study P.S. 1 Inductees into the army are asked about their smoking history and current habits. Smokers and nonsmokers are subsequently compared in relation to the development of lung cancer and other chronic diseases. What type of observational analytic study is this? -Case-control study -Retrospective cohort study -Prospective cohort study -Ecologic study - ANSWER-Prospective cohort study P.S. 1 The smoking histories of patients entering a hospital with lung cancer are compared with the smoking histories of patients entering for other conditions requiring surgery. What type of observational analytic study is this? -Case-control study -Retrospective cohort study -Prospective cohort study -Ecologic study - ANSWER--Case-control study P.S. 1 Observation A: The age-adjusted incidence rate for breast cancer in Japanese women is considerably lower than the age-adjusted incidence rate for breast cancer in North American women. The proportion of Japanese women who smoke cigarettes is also much lower than the proportion of North American women who smoke cigarettes. Observation B: Within North America, cigarette smoking does not increase the risk for breast cancer, and within Japan, smoking does not increase the risk for breast cancer. Which of the answers given below do you believe is correct? - Observation A provides better evidence on whether or not a causal association between smoking and breast cancer exists than Observation B. - Observation B provides better evidence on whether or not a causal association between smoking and breast cancer exists than Observation A. - Observations A and B provide equally good evidence on whether or not a causal association - ANSWER-Observation B provides better evidence on whether or not a causal association between smoking and breast cancer exists than Observation A. P.S. 2 The association between cigarette smoking and lung cancer has been subjected to considerable scrutiny. Which of the following statements both strengthen the association between cigarette smoking and lung cancer and move the evidence toward the direction of a causal relationship? - The risk of lung cancer increases as the daily consumption of cigarettes increases and/or as the duration of smoking increases. - Ex-smokers have a lung cancer risk intermediate between those of nonsmokers and current smokers. - Animal experiments have shown an increased risk for precancerous lesions following tobacco smoke inhalation. - Prospective studies agree with retrospective cohort and case-control studies about the presence and direction of the association. - All of the above - ANSWER-All of the Above P.S. 2 In a case-control study evaluating occupational exposure to aluminum and risk of Alzheimer's disease, relative to no exposure, the odds ratios are:low exposure = 1.3 (95% CI, 0.9 - 2.0);moderate exposure = 1.7 (95% CI, 1.4 - 2.3);high exposure = 2.3 (95% CI, 2.0 - 2.9). Using this information, what is the most likely criterion for causality? -Biologic plausibility -Consistency of results -Correct temporal sequence -Presence of a dose-response relationship -Strength of association - ANSWER-Presence of a dose-response relationship P.S. 2 On January 14, 1960, a one-day census of mental hospitals was carried out in 12 Caribbean islands. The report of this census noted: When we examine the first admissions by length of stay, we find that 18 percent... have been hospitalized under one year, 19 percent between one and five years, and 63 percent for five years or longer. That is, the majority of first admissions are long-stay patients (Report on the Census of Mental Hospitals in the Caribbean, 1960). What type of study design was this? -Case-control study -Randomized Controlled Trial (RCT) -Retrospective cohort study -Prospective cohort study -Cross-sectional Study - ANSWER-Cross sectional study P.S. 2 On January 14, 1960, a one-day census of mental hospitals was carried out in 12 Caribbean islands. The report of this census noted: When we examine the first admissions by length of stay, we find that 18 percent... have been hospitalized under one year, 19 percent between one and five years, and 63 percent for five years or longer. That is, the majority of first admissions are long-stay patients (Report on the Census of Mental Hospitals in the Caribbean, 1960). What type of study would be necessary to answer questions about duration of stay in long-term hospitals? -Case-control study -Randomized Controlled Trial (RCT) -Cohort study -Cross-sectional Study - ANSWER-Cohort Study P.S. 2 Five published case-control studies evaluating the same exposure and disease produce the following odds ratios:1.4 (95% CI, 0.9 - 2.4)1.7 (95% CI, 1.3 - 2.5)2.4 (95% CI, 1.3 - 4.61.9 (95% CI, 1.5 - 2.7)2.0 (95% CI, 0.9 - 5.4) Using this information, what is the most likely criterion for causality? -Biologic plausibility -Consistency of results -Correct temporal sequence -Presence of a dose-response relationship -Strength of association - ANSWER-Consistency of results met regularly at local gyms, or to a control group. Both groups were evaluated at regular intervals for heart disease. The study design is best described as - Case-control study - Cross-sectional study - Intervention study - Prospective cohort study - Retrospective cohort study - ANSWER-Intervention study Lecture 03 Clicker Quiz - Exam 1 The variation in severity of illness to a given exposure is called the - Incubation or induction period - Latent period - Both - Neither - ANSWER-Neither [Variation in severity of illness addresses the biologic spectrum and variation in host response, not the incubation, induction or latent period.] Lecture 04 Clicker Quiz - Exam 1 Descriptive Epidemiology is - Primarily useful for hypothesis testing - The study of the amount and distribution of disease within a population by person, place, and time - Both - Neither - ANSWER-The study of the amount and distribution of disease within a population by person, place, and time Lecture 04 Clicker Quiz - Exam 1 The sequence of studies in human populations typically begins with - A case-control study - A clinical trial - A prospective cohort study - Descriptive studies - ANSWER-Descriptive studies [Descriptive studies consist of clinical observations and available data.] Lecture 04 Clicker Quiz - Exam 1 In the United States, state incidence rates of hepatitis C infection are associated with corresponding state mortality rates for liver cancer. Which of the following types of study design best fits this description? - A case-control study - A clinical trial - Cohort study - Correlational study - Cross-sectional study - ANSWER-Correlational Study [This question is looking at two population level data and associating the two rates. The only population level study listed is the correlational study. The other studies are all done at the individual level.] Lecture 04 Clicker Quiz - Exam 1 National per capita use of pesticides is associated with national non-Hodgkin lymphoma (NHL) mortality rates, but patients with NHL do not have higher pesticides levels than persons without NHL. This relationship is most consistent with - Birth cohort effect - Confounding bias - Ecologic fallacy - Recall bias - Selection bias - ANSWER-Ecologic Fallacy Lecture 04 Clicker Quiz - Exam 1 Types of study designs used for descriptive epidemiology include each of the following except Aggregate study Case-control study Case report Case series Correlational study - ANSWER-Case Control Study Lecture 04 Clicker Quiz - Exam 1 To establish internal validity in an epidemiologic study, you do each of the following except -Assure generalizability -Describe interaction -Exclude bias (systematic error) as an explanation -Exclude chance (random error) as an explanation -Exclude confounding as an explanation - ANSWER-Assure generalizability Lecture 05 Clicker Quiz - Exam 1 In evaluating types of causal relations for a given factor, the following are true: -A factor is considered sufficient when without it the disease never develops. -A factor is considered necessary when in its presence the disease always develops. -Both -Neither - ANSWER-Neither Lecture 05 Clicker Quiz - Exam 1 Factors A, B, or C can each individually cause a certain disease without the other two factors, but only when followed by exposure to factor Z. Exposure to factor Z alone is not followed by the disease, but the disease never occurs in the absence of exposure to factor Z. Factor B is: -A necessary and sufficient cause -A necessary, but not sufficient, cause -A sufficient, but not necessary cause -Neither necessary nor sufficient - ANSWER-Neither necessary nor sufficient [It is important to remember that we are evaluating factor B, and not factor Z. With factor B present, disease will still not occur if factor Z is not present. This makes factor B not sufficient, as disease will not always occur in its presence. Without factor B, disease can still occur if either factor A or C is present. This makes factor B not necessary, as disease still occurs in its absence. If the question had asked about factor Z, factor Z would be considered a necessary, but not sufficient cause.] Lecture 05 Clicker Quiz - Exam 1 Each of the following is an important criterion when making causal inferences except -Absolute risk -Biologic plausibility -Consistency of association in several studies -Dose-response relationship -Strength of association - ANSWER-Absolute Risk Lecture 05 Clicker Quiz - Exam 1 In a case-control study evaluating occupational exposure to aluminum and risk of Alzheimer's disease, relative to no exposure, the odds ratios (i.e., measures of association) are: For low exposure = 1.3; For moderate exposure = 1.7; For high exposure = 2.3.These data best fit which of the following guidelines for causality? -Biological plausibility -Consistency of results -Correct temporal sequence -Dose-response relationship -Strength of association - ANSWER-Dose-Response relationship [There is a clear correlation between the amount of exposure and odds ratio. As the amount of exposure increases, so does the odds ratio.] Lecture 05 Clicker Quiz - Exam 1 There is epidemiologic literature showing a positive association between aluminum exposure and Alzheimer's disease. At post-mortem examination, the pathologic lesions within brains of patients with Alzheimer's disease are shown to contain unusually high levels of aluminum. These data best fit which of the following guidelines for causality? The following vaccine trial was performed: 1,000 randomly selected children 2 years of age were given a vaccine against a certain disease and followed for 10 years. Of these, 80% were never afflicted with the disease. Which of the following is the most correct conclusion regarding the vaccine? -The vaccine is excellent because of the high rate of immunization. -The vaccine is not very effective because it should have produced a higher immunization rate. -No conclusion is possible because no follow-up was made of non-vaccinated children. -No conclusion is possible because no test of statistical significance was performed. -The significant figure is 100% - 80% = 20%, the rate of acquiring the disease. - ANSWER-No conclusion is possible because no follow-up was made of non-vaccinated children. [We do not know how the disease would have progressed without the vaccine. It is possible that 80% (or more) of children who were not vaccinated also were never afflicted with the disease. Without knowing the difference between the vaccinated children and the non-vaccinated children no real conclusion can be made.] Lecture 08 Clicker Quiz - Exam 1 Each of the following statements about the routine autopsy is true except -Its performance typically affects the timing of the funeral. -Its performance typically allows the opportunity to have an open casket at the funeral home. -Notes and photographs may be taken to document any important findings. -Small pieces of tissue may be collected for examination under a microscope. -The body may be x-rayed or scanned to look for problems like fractures, tumors, or gunshot wounds - ANSWER-Its performance typically affects the timing of the funeral. [Autopsies are done generally very soon after death, while a funeral takes several days to plan. Autopsies typically do not impact the ability to have an open casket. The incisions done during an autopsy are made in areas that are easily concealed] Lecture 08 Clicker Quiz - Exam 1 An autopsy requires consent of the family or legal guardian when performed -For legal reasons -For medical reasons -Both -Neither - ANSWER-For medical reasons [Family permission for legal reasons (ie. in the case of homicide or suicide) is not needed. The medical examiner can decided whether a autopsy should be done or not without family permission.] Lecture 08 Clicker Quiz - Exam 1 A medical examiner is a -Lay investigator -Pathologist or other health provider -Both -Neither - ANSWER-Pathologist or other health provider [Slide 8 gives a brief definition. A lay investigator would be a coroner. A medical examiner is a pathologist or other health care provider.] Lecture 08 Clicker Quiz - Exam 1 A 40-year old man, with a long history of hepatic cirrhosis due to a well-documented history of chronic alcoholism, died in the hospital of slowly progressive liver failure. Autopsy findings included jaundice, micronodular cirrhosis, ascites, and esophageal varices. No hepatocellular carcinoma was identified. What was the immediate cause of death? -Ascites -Chronic alcoholism -Cirrhosis of the liver -Esophageal varices -Liver failure - ANSWER-Liver Failure Lecture 08 Clicker Quiz - Exam 1 Strengths of cross-sectional studies include -Can be utilized as the starting point for a prospective cohort study -Tend to be carried out over a long time period -Both -Neither - ANSWER-Can be utilized as the starting point for a prospective cohort study Lecture 20 Clicker Quiz - Final Exam Descriptive studies of genetic epidemiology include each of the following except -Age at onset -Association with known genetic diseases -Migrant studies -Time trends in disease incidence -Twin studies - ANSWER-Twin studies Lecture 20 Clicker Quiz - Final Exam If a greater proportion of monozygotic twin pairs are found to be concordant for a certain disease than are dizygotic twin pairs, the observation suggests that the disease is most likely caused by -Environmental and genetic factors almost equally -Exclusively environmental factors -Exclusively hereditary factors -Gender differences in monozzygotic twins -Hereditary factors almost exclusively, with some nonhereditary factors possibly playing a role - ANSWER-Hereditary factors almost exclusively, with some nonhereditary factors possibly playing a role [Monozygotic twins are identical twins, sharing 100% of their genetics, which is of course more than the fraternal twins (dizygotic twins), who only share a portion of their genetics. A higher rate of monozygotic twin pairs developing a disease than dizygotic twins indicates that genetics probably play a big role. However, nonhereditary factors cannot be completely ruled out as a factor. Therefore, this observation suggests that the disease is most likely caused by hereditary factors almost exclusively, with some nonhereditary factors possibly playing a role.] Lecture 20 Clicker Quiz - Final Exam Concordance of a disease in twins reared together is compared with risk among twins reared apart. The study design represented by this activity is best described as a -Case-control study -Cohort study -Cross-sectional study -Twin study -Segregation study - ANSWER-Twin study [This is a particular form of twin studies. Generally anytime twins are used, it can be considered a twin study. While this is a good way to study the impact of genetics on particular diseases, it is difficult to find a large number of twins reared separately. For a real life example of twin studies, go to the Minnesota Center for Twin and Family Research. This center, particularly Dr. Thomas Bouchard has paid particular attention to twins reared apart.] Lecture 21 Clicker Quiz - Final Exam The recurrence risk in first degree relatives of persons with disease X is compared with first degree relatives of persons without disease X. Once identified, these relatives are followed in time to determine whether they develop disease X. The study design is best described as a Case-control study Cohort study Cross-sectional study Twin study Segregation study - ANSWER-Cohort study [In this example, our cases are the first degree relatives with a father who had the disease and our controls are those who's father did not have the disease. We are then comparing proportions of those in each group that had the disease, which is considered -Dependent on the size of the population being evaluation -Greater in test 1 -Equal in tests 1 and 2 -Greater in test 2 - ANSWER-Greater in test 2 [Diabetes severity increases with increased levels of blood sugar. So a test with a criterion of positivity at 160 mg/dl will detect fewer cases than 130 mg/dl (it will detect more severe disease because it will miss those at risk with a blood sugar level less than 160 mg/dl, but still high enough to classify as diabetic). Therefore test 2 set at 130 mg/dl will have a higher sensitivity and will detect more cases.] Lecture 22 Clicker Quiz - Final Exam When evaluating a screening program, the predictive value positive is -Neither -A process measure -Both -An outcome measure - ANSWER-A process measure [Predictive value positive is a process measure. This can be found on slide 46.] Lecture 22 Clicker Quiz - Final Exam When evaluating a screening program, major determinants of average lead-time include each of the following except -Distribution of maximum lead times -Frequency of screening -Sensitivity of the screening test -Specificity of the screening test - ANSWER-Specificity of the screening test [See slide 70 of section 17.] Lecture 23 Clicker Quiz - Final Exam To avoid length bias in a screening program, one should -Increase the prevalence rate of disease among the screened population -Repeat screening at shorter intervals -Both -Neither - ANSWER-Repeat screening at shorter intervals [See slide 74 of section 17.] Lecture 22 Clicker Quiz - Final Exam Strategies to reduce screening costs include each of the following except -Increasing the PV+ -Performing the screening test at an optimal interval -Shifting the criterion of positivity so that sensitivity is increased at the expense of specificity -Using multi-phasic screening -Using screening tests in series - ANSWER-Shifting the criterion of positivity so that sensitivity is increased at the expense of specificity [See slide 78 of section 17.] Lecture 23 Clicker Quiz - Final Exam The same screening test is applied in a uniform manner in two populations. The prevalence of the disease being screened is 10% in population A and 1% in population B. Which of the following statements is true? - Reliability is higher in population A than in population B. - Sensitivity is higher in population A than in population B. - Specificity is higher in population A than in population B. - The percent of all positive screening tests that have true positive results is lower in population A than in population B. - The percent of all positive screening tests that have false-positive results is lower in population A than in population B. - ANSWER-The percent of all positive screening tests that have false-positive results is lower in population A than in population B. [Prevalence does not have a direct effect on sensitivity or specificity or reliability. However, predictive value positive is greatly affected by prevalence. An increase in prevalence increases the PV+. Since population A has a higher prevalence, it would have more true positives and less false positives. Answer choice D has the reverse of the correct relationship, so E is the correct answer.] Lecture 23 Clicker Quiz - Final Exam Which of the following is the best index for concluding that an early detection program for breast cancer truly improves the natural history of disease, 15 years after initiation of the early detection program? - A decrease in the incidence of breast cancer. - An increase in incidence of breast cancer. - An increase in the prevalence of breast cancer. - A smaller proportionate mortality for breast cancer 15 years after initiation of the program compared to the proportionate mortality prior to its initiation. - Improved long-term survival rates for breast cancer patients that have been adjusted for lead-time bias and length bias. - ANSWER-Improved long-term survival rates for breast cancer patients that have been adjusted for lead-time bias and length bias. [While screening programs may result in a smaller proportionate mortality or cause an increase in incidence, these indicators may contain biases or not explain the whole picture. Evaluating long-term survival rates that have been adjusted for bias is best. It removes known bias that result from screening programs and helps to evaluate a screening program's ultimate goal of saving lives.] Lecture 23 Clicker Quiz - Final Exam Censoring in an intervention study whose end point is mortality (or death) occurs when - Subjects are excluded from the study before it begins because of the presence of a particular factor or complication. - Subjects die and the cause of death is unknown. - Subjects die before the planned termination of the study. - Subjects fail to comply with their assigned intervention. - They study ends before all subjects have died. - ANSWER-They study ends before all subjects have died. [In this case, A would be a poor choice because it was a decision made prior to the start of the study and therefore those subjects would not have any data to censor. B could be included in our results as another category or listed as unknown, if we wanted to include such a category. C is what we are looking for and we definitely want to include that in our results. D is something we cannot be sure of. We can't just assume that they didn't comply, they might have done other things. So E is really our best option. It is a very common use of censoring in survivor analysis.] Lecture 24 Clicker Quiz - Final Exam Common methods for survival analysis are - Kaplan-Meier analysis - Life table analysis - Both - Neither - ANSWER-Both [See slide 17 of section 18.] Lecture 24 Clicker Quiz - Final Exam Which of the following is an assumption of a standard life table analysis but not of a Kaplan-Meier life table analysis? - Groups compared are alike except for the exposure of interest. - Subjects who withdraw or are lost during the course of the study experience the same risk as those who continue. - The relative risk of the study outcome is constant over time. - The risk of the outcome is the same in 1 person followed for 10 years as 10 persons followed for 1 year. - Withdrawals occur evenly during intervals used in analysis. - ANSWER-Withdrawals occur evenly during intervals used in analysis. -Correct -Incorrect because age-specific incidence rates are not calculable in this type of study -Incorrect because gender-specific incidence rates are not calculable in this type of study -Incorrect because the relative risk is based on the comparison of two prevalence rates - ANSWER-Correct [This inference of the relative risk is correct. The opposite gender is used as the comparison group. A cohort study was done, therefore it was possible to calculate and use incidence rates. Incidence rates are used to calculate relative risks, making rates the appropriate choice for comparison] Exam 3 This quote "Allows direct measurement of incidence of disease in the exposed and nonexposed groups" is a strength of which of the observational epidemiologic study designs listed below? -Case series study -Cross-sectional study -Case-control study -Cohort study -Ecologic study - ANSWER-Cohort study [Cohort studies begin with exposed and unexposed individuals. These individuals are then followed to determine disease status, thus allowing for direct measurement of incidence of diease. Section 13] Exam 3 All of the following statements are true except: -Attributable risk percent is also referred to as an etiologic fraction -The magnitude of the relative risk does not predict the magnitude of the attributable risk -The odds ratio estimates the magnitude and direction of an association between exposure and disease -In a case-control study, incidence rates can be computed directly -The relative risk is the ratio of the measures of disease incidence for two populations - ANSWER-In a case-control study, incidence rates can be computed directly [Incidence rates cannot be directly computed in the exposed and nonexposed individuals. Case-control studies begin after the disease has been established. Section 12] Exam 3 In a case-control study of sunlight exposure and skin cancer, cases and controls were asked about participation in various outdoor activities and the use of sunscreen. Errors of recall in exposure occurred with equal frequency among cases and controls. Which one of the following occurred? -Confounding -Differential misclassification -Non-response bias -Non-differential misclassification -Selection Bias - ANSWER-Non-differential misclassification [Non-differential (random) misclassification occurred because the cases and controls were affected equally.] Exam 3 In a case-control study of exercise as a protective factor against the development of osteoporosis, controls tend to overestimate their level of exercise more than cases. The likely effect on the observed association as measured by the odds ratio is: -Impossible to predict from the information given -To have no effect on its magnitude -To shift its magnitude toward the value of 0 -To shift its magnitude toward the value of 1 - ANSWER-To shift its magnitude toward the value of 0 [The first thing to note is that exercise (our exposure) is a protective factor. This means that the OR will be less than 1. As we are hypothesizing a reduced risk of osteoporosis (cases) in those who exercise, if the controls (those without osteoporosis) over-estimate their exercise/exposure level, then the overall protective association of exercise will be strengthened, and shift the magnitude of the observed OR toward the value of 0.] Exam 3 Which of the following health statistics would be useful in selecting risk factors to target for intervention in order to have the greatest public health impact? Assume the planned interventions are equally effective but resources are limited. -Attributable Risk -Incidence of the disease among the exposed -Prevalence of the diease among the exposed -Proportiona mortality ratio -Relative risk - ANSWER-Attributable Risk [Attributable Risk has the greatest public health impact. This is because it estimates the numebr of cases that can be saved/prevented from intervening with the exposure/disease association. Section 13] Exam 3 A disadvantage for a clinical trial is -blinding of endpoints -lack of generalizability -lack of internal validity -protocol procedures and interventions specified -randomization - ANSWER-Lack of generalizability. [Section 14] Exam 3 A randomized trial comparing the efficacy of two drugs did not show a difference between the two (p-value > 0.05). Assume that in reality, however, the two drugs did differ. This is an example of -Type I error (a error) -Type II error (B error) -both -neither - ANSWER-Type II error (B error) [definition of type 2 error is failure to reject the null hypothesis when it is false] Exam 3 Compared to other phases of drug development what is the primary goal of phase III clinical trial? -controlled trial of efficacy -determine preliminary safety and efficacy -determine toxicity in animals -look at pharmacologic effects of the drug -post marketing surveillance - ANSWER-controlled trial of efficacy [see slide 52 of section 14] Exam 3 In 1945, there were 1,000 women who worked in a factory painting radium dials on watches. The incidence of bone cancer in these women up to 1975 was compared with that of 1,000 women who worked as telephone operators in 1945. Twenty of the radium dial painters and four of the telephone operators developed bone cancer between 1945 and 1975. When conducted in 1980 this study is an example of -case-control study -clinical trial -cross-sectional study -prospective cohort study -retrospective cohort study - ANSWER-retrospective cohort study Suppose a study of warfarin used in renal cell carcinoma results in an odds ratio of 2.4 with a 95% confidence interval of 2.1 to 2.7 after adequate consideration of bias, confounding, and effect modification. At the 5% level of statistical significance, the observed result is most appropriately interpreted as? -Warfarin users are 140% times as likely to develop renal cell carcinoma compared to nonusers, and the result was statistically significant -People with renal cell carcinoma are about 40% more likely to be warfarin users compared to people who don't have renal cell carcinoma -Warfarin users have 2.4 times the odds of developing renal cell carcinoma compared to non-users, the difference was statistically significant -Warfarin users have a significant 2.4% greater odds of renal cell carcinoma compared to non-users. - ANSWER-Warfarin users have 2.4 times the odds of developing renal cell carcinoma compared to non-users, the difference was statistically significant [Since the OR is greater than one, warfarin is a hazardous exposure. The confidence interval (2.1-2.7) excludes the value of 1 (our null hypothesis). This indicates a statistically significant increase risk. Patients exposed to warfarin have 2.4 greater odds of developing renal cell carcinoma, or are 140% more likely than nonusers to develop renal cell carcinoma] Exam 3 Advantages of prospective cohort studies over retrospective cohort studies include -they are less expensive -they can be completed faster -both -none of the above - ANSWER-none of the above [prospective cohort studies are more expensive and take longer than retrospective cohort studies slide 12 section 13] Exam 3 Disadvantages of using general population controls in a case control study include each of the following except: -Increase generalizability of study results -May be less motivated to participate -May not recall exposures with the same level of accuracy as cases -Population lists from which to select not always available -Usually more costly and time-consuming than hospital controls - ANSWER-Increase generalizability of study results [increased generalizability is considered an advantage, not a disadvantage. Section 12] Exam 3 When calculating a type of attributable risk for an exposure and disease, we assume the relationship between the exposure and the disease is causal when using -the case control study design -cohort study design -both -neither - ANSWER-both [attributable risk can be calculated for both cohorts and case-control studies. In order to calculate attributable risk for a specific exposure, we must assume the exposure causes the disease. We must assume the risk is real and present.] Exam 3 Which of the following is a disadvantage of a matched case control study -can examine multiple etiologic factors for a single disease -may provide control for confounding variables that are difficult to measure directly -cannot evaluate differences between cases and controls on matched variables -siblings as matched controls may provide comparability on a number of genetic and environmental factors -potential gain and accuracy of estimating the odds ratio - ANSWER-cannot evaluate differences between cases and controls on matched variables [Once matching is undertaken, the cases and controls will be more similar with respect to the matching characteristics, so you cannot evaluate differences between cases and controls on matched variables. section 12] Exam 3 In planning a case control or cohort study, you are interested in the smallest (or minimum) detectable odds ratio or relative risk as you consider your sample size. With this in mind, each of the following statements is true except -the smallest detectable odds ratio or relative risk moves further away from 1.0 as the level of significance changes from 0.05 to 0.01 -the smallest detectable odds ratio or relative risk moves further away from 1.0 as the power changes from 70% to 80% -the smallest detectable odds ratio or relative risk moves further away from 1.0 as the type one error changes from 0.05 to 0.01 -the smallest detectable odds ratio or relative risk moves further away from 1.0 as the type 2 error changes from 10% to 20% -the smallest detectable odds ratio moves further away from 1.0 as the number of cases and the number of controls decreases - ANSWER-the smallest detectable odds ratio or relative risk moves further away from 1.0 as the type 2 error changes from 10% to 20% [increasing type 2 error will cause the smallest detectable odds ratio to move toward 1.0, not further away. All other options are true.] Exam 3 Random (non differential) misclassification biases a measure of association (i.e., odds ratio or relative risk) -away from the null hypothesis -toward the null hypothesis -both -neither - ANSWER-toward the null hypothesis [bias toward the null hypothesis results in an underestimation] Exam 3 The following are all advantages of a longitudinal study, except? -Allows for repeated measures -Possibility of practice effects -Detection of temporal patterns and variables overtime -Avoids some of the concern regarding confounding -Can examine interactions with time - ANSWER-Possibility of practice effects [the possibility of practice effects is a distinct disadvantage of longitudinal studies (i.e., alteration of behavior/responses of subjects due to their awareness of being observed. section 15] Exam 3 (Screening) Three Levels of Prevention: 1. Primary 2. Secondary 3. Tertiary - ANSWER-Primary - altering susceptibility Secondary - Early detection and treatment (Screening) Tertiary - Alleviaiton of disability (Section 17 Handout, Letter A) Final Exam Uses of Diagnostic (Screening) Tests - ANSWER-1. Clinical diagnosis 2. Case-finding 3. Screening 4. Epidemiologic study (Section 17 Handout, Letter C) Final Exam Screening Definition - ANSWER--Presumptively ID an unrecognized disease by a test, examine, or procedure -Can be applied rapidly to sort people who are well and those who are not -Positive testing people must be further treated to diagnose [cross sectional studies are often referred to as prevalence studies. When you survey a population to determine who has disease/no disease and exposure/no exposure, you get an estimate of the prevalence of both disease and exposure.] Exam 4 Pretest Determine how the sensitivity and specificity will be traded off against one another -concordance rate -criterion of positivity -false negative percentage -predictive value positive -prevalence rate - ANSWER-criterion of positivity [as the criterion of positivity moves, so does the number of TP, FP, TN, and FN. As a result, sensitivity and specificity change. section 18 slides 34 and 35] Exam 4 Pretest The predictive value positive of a population based screening test can be increased by -increasing the prevalence of the disease in the screened population -increasing the sensitivity of the screening test while decreasing its specificity -both -neither - ANSWER-increasing the prevalence of the disease in the screened population [Positive predictive value is heavily influenced by other prevalence and specificity section 18 slide 51] Exam 4 Pretest Each of the following statements is true except? -Screening is conducted to reduce mortality and increase survival -The incidence of the target disease will likely increase when a new, effective screening test becomes widely used The predictive value positive is more influenced by the specificity than the sensitivity of the screening test -The purpose of screening is to identify asymptomatic cases of serious disease that can be effectively treated -Case control studies and experimental studies of screening effectiveness avoid selection bias - ANSWER-Case control studies and experimental studies of screening effectiveness avoid selection bias [Case control and experimental studies are study designs that are inherently difficult to control selection bias. Steps can be taken to reduce selection bias, but it is most likely not possible to completely avoid it] Exam 4 Pretest Which of the following is an assumption that underlies the person time analysis but not the Standard Life table analysis? -groups compared are alike except for the exposure of interest -subjects who withdraw or are lost during the course of the study experienced the same risk as those who continue -the relative risk of the study outcome is constant overtime -the risk of the outcome is the same in one person followed by 10 years as ten persons followed for one year -withdrawals occur evenly during intervals used in analysis - ANSWER-the risk of the outcome is the same in one person followed by 10 years as ten persons followed for one year [While this wasn't explicitly covered in lecture, we did discuss the concept of person years and option D is the reason we calculate person years period there is a more in depth coverage of this in chapter six of the textbook page 111] Exam 4 Pretest Where is the hypothesis or purpose for a study usually stated in a published manuscript -abstract -introduction section -both -neither - ANSWER-neither [the purpose of the abstract is to provide a brief summary of the purpose of the study, the results and major conclusions. The hypothesis should definitely be included here. The introduction is where the rationale for the study is stated. The hypothesis is also stated at the end of this section.] Exam 4 pretest When using screening tests for disease detection, the components of validity are -reliability and predictive value positive -sensitivity and specificity -sensitivity and reliability -specificity and predictive value positive -specificity and reliability - ANSWER-sensitivity and specificity [Slide 16, seciton 17] Exam 4 Pretest Familial aggregation studies of genetic epidemiology include each of the following except -adoption studies -inbreeding studies -migrant studies -recurrence risk -twin studies - ANSWER-migrant studies [slide 24, section 16] Exam 4 pretest It is impossible for a female to have received a sex chromosome from one of her four grandparents. For a female child, which grandparent could not have transmitted via her parents a sex chromosome -maternal grandfather -maternal grandmother -paternal grandfather -paternal grandmother - ANSWER-paternal grandfather [the female child has two X chromosomes, one from each of her parents. Her mother received one X chromosome from her mom and one X chromosome from her dad. Her father received one X chromosome from his mom and one Y chromosome from his dad. Therefore, the females father had pass on his only X chromosome, which had to have come from his mother. His Y chromosome wouldn't be passed on because his child is female] Exam 4 pretest The case control study is designed to determine whether various serum proteins are risk factors for a certain neurological disease. If after results are reported the laboratory determines that the reagent for measuring one of the proteins had expired, then this could result in -confounding bias -differential misclassification -ecological fallacy -random misclassification -selection bias - ANSWER-random misclassification [slide 74 section 13. The expired reagent will affect measuring of one protein in both the cases and the controls therefore the proportion of inaccurate responses is the same in both study groups and it is an example of random misclassification] Exam 4 pretest Which of the following is a possible outcome measure that could be used as an indicator of the benefit of a screening program aimed at early detection of the disease? -Improvement in quality of life and screened individuals. -Reduction of disease complications. -Both. -Neither. - ANSWER-Both [Slide 2, section 17] exam 4 pretest Each of the following is a true statement about limitations of epidemiologic information in the legal arena the courtroom except