hw 9

pdf

School

University of Illinois, Urbana Champaign *

*We aren’t endorsed by this school

Course

212

Subject

Medicine

Date

Apr 3, 2024

Type

pdf

Pages

12

Uploaded by DukeResolveMantis34

Report
10/23/23, 10 : 37 AM Quizzes - Results Page 1 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Results Gigi Yik — 3rd Attempt 15 Out of 15 points 02:10 Time for this attempt This assessment has unlimited attempts. Your Answers: 1 / 1 point Use this information to answer questions 1-4. Consider this fictional study, in which researchers have been studying possible health effects related to vaping. Since a relatively small percentage of the population vapes, the researchers recruited 1,551 who reported vaping regularly and 1,612 people who had not previously vaped and had no interest in Take Now Attempt History Results Points Score (Highest score is kept) Attempt 1 12 of 15 80% Attempt 2 14 of 15 93.33% Attempt 3 15 of 15 100% (Highest score) 100% 1
10/23/23, 10 : 37 AM Quizzes - Results Page 2 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results vaping. Over the next 12 months, the researchers asked participants to report any ongoing sicknesses or medical conditions. One condition in particular was a diagnosis of pneumonia. They noticed a difference in pneumonia rates for each group. Size Percent Reporting Pneumonia Vapers 1,551 3.93% Non-vapers 1,612 5.86% What kind of study is this? Feedback Based on answering correctly Good job! The researchers first identified two cohorts (vapers and non-vapers) and then monitored them for various response outcomes (pneumonia vs. no pneumonia) over some period of time. Note that the researchers did not assign the participants to vape or not vape, so this would simply be an observational study, rather than an experiment. 1 / 1 point Which statement best explains what this study reveals? An Experiment A Cross-Sectional Study A Cohort Study A Case Control Study Vaping is a likely preventative of pneumonia There is an association between those who frequently vape and a decreased risk for pneumonia, but it’s not clear whether vaping is causally related to pneumonia 2
10/23/23, 10 : 37 AM Quizzes - Results Page 3 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Feedback Based on answering correctly Good choice. We notice that the rate of pneumonia is lower for the vaping group, but there could be a confounder here. 1 / 1 point Which of the following describes a potential confounder to the relationship described in the previous question and be consistent with the data we found? Feedback Based on answering correctly That’s right! This is the only factor that explains why vaping would be associated with lower risk of pneumonia in the form of a confounder. The “Socializing” answer could describe a confounding relationship, but that explanation goes the opposite direction (the data said vapers are less likely to get pneumonia, but this explanation only fits if they are more likely). The “stronger lungs” factor would be mediating variable that explains how vaping Vaping is a likely cause of increased risk for pneumonia There is an association between those who frequently vape and increased risk for pneumonia, but it’s not clear whether vaping is causally related to pneumonia Younger people are more likely to vape, and younger people might be healthier and less likely to get pneumonia Vaping may lead to stronger lungs, and that may help prevent onset of pneumonia People who get pneumonia may have a less healthy diet People who start vaping may have done that because they socialize more and get respiratory viruses, making them more likely to get pneumonia than others 3
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
10/23/23, 10 : 37 AM Quizzes - Results Page 4 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results begins a causal chain to lower pneumonia risk. The “healthy diet” factor does not directly relate vaping habits, so it doesn’t explain a connection between the two. 1 / 1 point The researchers are trying to decide if they should compare the risk of pneumonia by calculating relative risk or an odds ratio. Does it matter which they use? Feedback Based on answering correctly Nice! Relative risks are typically appropriate to calculate in all design situations except case-control designs (when natural incidence is not preserved in the sampling method). In those situations, we might use an odds ratios to estimate the risk comparison. In all other situations, we could validly construct either relative risk or an odds ratio, but relative risk is often preferred since it’s more straightforward to interpret. 1 / 1 point An article in Medical Xpress reported a link between young people who take melatonin and a decreased risk of self-harm. https://medicalxpress.com/news/2023-03-melatonin- linked-decreased-self-harm-young.html One explanation is that melatonin promotes better quality sleep, and better quality of sleep results in better mental health and less inclination toward self harm. Does better quality of sleep help explain this link? If so, what type of explanation is it? Only a Relative Risk is appropriate to calculate for this particular design Only an Odds Ratio is appropriate to calculate for this particular design Both could be used, but Relative Risk is often preferred for being easier to interpret 4 5
10/23/23, 10 : 37 AM Quizzes - Results Page 5 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Feedback Based on answering correctly Nice work! In this explanation, melatonin is believed to cause better sleep, and better sleep then leads to better mental health/less self harm tendencies. Melatonin may not be directly affecting mental health/self harm tendencies, but sleep helps mediate the connection. 1 / 1 point Use this information for questions 6-7. A college study finds that students who use Adderall tend to have lower GPAs on average than students who don’t. Further investigation shows that people who struggle with executive functioning (like concentration) are more likely to use Adderall, and it’s actually their difficulties with executive functioning that hurt their grades, not the Adderall. The Adderall may actually help them perform better than they would otherwise. Does difficulty with executive functioning help explain this link? If so, what type of explanation is it? Feedback Based on answering correctly Yes, as a Confounder Yes, as a Mediator No, it does not explain the link Yes, as a Confounder Yes, as a Mediator No, it does not explain the link 6
10/23/23, 10 : 37 AM Quizzes - Results Page 6 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Nice job! Adderall doesn’t begin the causal chain here. Rather, it’s difficulty with executive functioning that leads people to try Adderall, and it’s difficulty with executive functioning that probably hurts grades. Adderall and lower grades are both outcomes of this confounder. 1 / 1 point Now consider this same example again. Further investigation shows that people who take Adderall are also more likely to have pets. Does having pets help explain this link? If so, what type of explanation is it? Feedback Based on answering correctly Good job! There is nothing in the description that explains why pets would relate to GPA. So this doesn’t explain why Adderall would be linked to lower grades. 1 / 1 point Use this information for questions 8-9. A doctor is studying whether having a particular gene variant might increase your risk of an astigmatism (the condition that makes people far-sighted or near-sighted). The researcher collected this data using an observational study format, where she compared people with the gene variant to people without the gene variant to see if there was a difference in the proportion had an astigmatism. Could this researcher have studied whether the gene variant causes increases risk of astigmatism using an experimental design? Yes, as a Confounder Yes, as a Mediator No, it does not explain the link 7 8
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
10/23/23, 10 : 37 AM Quizzes - Results Page 7 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Feedback Based on answering correctly Good choice! The explanatory factor we’re studying is a gene variant. But we can’t assign people to have a gene variant or not. We can only observe outcomes of people who happen to have the variant and people who don’t. 1 / 1 point Let’s say that the researcher finds that people with the gene variant are more likely to have an astigmatism. What could we reasonably conclude from this finding? Feedback Based on answering correctly Nice job! Since this is an observational study, we cannot confidently assume causality. All we have done is notice an association between the two. There could still be another confounder here that explains why both the gene variant and astigmatism commonly occur together. For example, maybe this gene variant commonly occurs with another gene variant that is the actual cause. Yes, with the right resources, this could be an experiment, and there are no ethical dilemmas here Yes, but it would not be ethical based on the potential for harm of the response variable No. It’s not possible to assign people to the explanatory factor in this investigation The gene variant does not give us any indication of whether an astigmatism is likely present or not in a particular person The gene variant is associated with increased risk for an astigmatism, but we don’t know if it is causally related The gene variant causes changes to the body that makes astigmatisms more likely 9
10/23/23, 10 : 37 AM Quizzes - Results Page 8 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results 1 / 1 point Use this information to answer questions 10-11. A car safety expert reviews accident reports to identify the relationship between the age of a driver and the severity of accident injuries. Accidents involving younger drivers are associated with higher rates of severe injuries. The safety expert is considering the possibility that younger drivers are more likely to engage in risky driving behavior (like texting), and whether that might explain the increase in severe injuries. Does “risky driving behavior” help explain this link? If so, what type of explanation is it? Feedback Based on answering correctly Good deduction! This explanation is framed to say that younger people engage in more risky driving behavior, and that reckless driving then leads to more severe injuries. Thus, the age of the driver is beginning the causal chain in this explanation. 1 / 1 point The safety expert decides to stratify the data by “vehicle type.” He notices that when comparing younger and older drivers of the same vehicle types, severe injuries are still more common in accidents with younger drivers—the differences are still the same. What does this mean? Yes, as a Confounder Yes, as a Mediator No, it does not explain the link The researcher identified vehicle type as a confounder to this relationship 10 11
10/23/23, 10 : 37 AM Quizzes - Results Page 9 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Feedback Based on answering correctly Good job! After stratifying by vehicle type, the researcher still saw an association. This means vehicle type didn’t explain away the difference. That said, there may still be other confounders that explain this connection! For example, younger people may be more likely to live in urban regions that naturally have more risky driving conditions. 1 / 1 point Researchers from the National Institute of Health surveyed nurses in hospitals and clinics across the country. They asked several questions including: “Do you current work in a clinic or hospital in a rural setting?” and “Do you expect to still work as a nurse 5 years from now?” They were curious if nurses who work in rural settings may be less likely to continue in their position as compared to nurses in more urban or suburban settings. What type of study is this? Feedback Based on answering correctly Good choice! This data is being collected at one point in time, rather than over time, The researcher has ruled out vehicle type as a confounder, but there may still be other confounders worth checking The researcher has evidence that driver age is causally linked to severe accident injuries An Experiment A Cross-Sectional Study A Cohort Study A Case Control Study 12
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
10/23/23, 10 : 37 AM Quizzes - Results Page 10 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results or through back-tracking previously collected data. The nurses weren’t recruited into the study based on their explanatory outcome or their response outcome (as far as we can tell). 1 / 1 point An epidemiologist (someone who studies infectious diseases) is examining patients who have been hospitalized due to a new strain of the flu. She compares these patients with others who had the flu but did not require hospitalization and is looking for explanatory factors that might reveal differences between these groups. One significant difference she finds is that those who were hospitalized tended to be much older on average than those who were not hospitalized. What type of study is this? Feedback Based on answering correctly Awesome! This study is first identifying people with a particular response (hospitalization), then looking for a potential causal agent. This study is also retrospective (common with case-control studies), meaning that we have to use records or additional patient surveying to identify possible causal factors. 1 / 1 point Non-steroidal anti-inflammatory drugs (NSAIDs) may increase the risk for indigestion and nausea, but it may depend on the specific drug used. To study this, a doctor assigns some patients seeking a pain relief drug to take ibuprofen while others are assigned to take An Experiment A Cross-Sectional Study A Cohort Study A Case Control Study 13 14
10/23/23, 10 : 37 AM Quizzes - Results Page 11 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results naproxen. Over 1 month, the doctor notes how many people in each group report indigestion or nausea as a side effects. What type of study is this? Feedback Based on answering correctly Nice job! The doctor assigned units to a treatment in this study. This allows the doctor to eliminate a lot of potential confounders and isolate the drug as the only systematic difference. This is in contrast to an observational study, where we can only observe what people already do. In those studies, groups may be different in more ways than the explanatory variable. 1 / 1 point If we have completed an observational study, is there any way we can attempt to determine whether the explanatory variable is causally linked to the response? Feedback An Experiment A Cross-Sectional Study A Cohort Study A Case Control Study No, but we can always complete an experiment with the same variables instead. Yes, but cautiously. We can stratify for possible confounders, but we they can never be sure that we have accounted for all possible confounders Yes. If the observational study has a large enough sample size, then a causal inference is appropriate. 15
10/23/23, 10 : 37 AM Quizzes - Results Page 12 of 12 https://illinoisedu.quiz-lti-iad-prod.instructure.com/taking/1441059/results Based on answering correctly Wonderful choice! It would be nice if we could always run an experiment instead, but there are a lot of conditions for which assigning units would not be ethical or possible. However, researchers can try to collect other potential confounding variables (sometimes called “covariates”) to see if there is still evidence for a difference, even after controlling for these other factors. This is in part how researchers determined that smoking greatly increased the risk for lung cancer.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help