Week 5 Worksheet - RCT cont., ecological study, random error, bias

pdf

School

Western University *

*We aren’t endorsed by this school

Course

1001/1002

Subject

Mechanical Engineering

Date

Dec 6, 2023

Type

pdf

Pages

Uploaded by BaronWaterJellyfish264

Name _____________________________ Week 5 Worksheet In last week’s worksheet, you were given the following problem and data: You design a study comparing the occurrence of headaches between two headache medications. You randomly assign 2170 people to take drug A, 437 of whom report headaches at the end of the study. You then randomly assign 1742 participants to take a placebo, of whom 162 reported headaches at the end of the study period. Headache No Headache Total Drug A 437 1733 2170 Placebo 162 1580 1742 Totals 599 3313 3912 Based on this information, you conducted an intent to treat analysis in which you analyzed everyone as if they stuck to their assigned treatment arm (Drug A or Placebo). You should have concluded that the risk of headaches was 2.17 times as likely/high among those who took Drug A compared to those who took the placebo. 1) After you do your analysis, you find out some more information about the participants. Some participants in your study were not taking the drugs/placebo as assigned. Instead, some people who were assigned drug A, took the placebo, and some people who were assigned the placebo took drug A. The table below shows additional information about which participants took the drugs as assigned and which didn’t. Please note that the data presented here may be in a different format than what we’ve used before, but the same concepts still apply. Calculate the appropriate measure of association using a per protocol analysis AND offer an interpretation. (10 points) Headache No Headache Total Drug A 437 - 400 took Drug A - 37 took placebo 1733 - 1000 took Drug A - 733 took placebo 2170 Placebo 162 - 62 took Drug A - 100 took placebo 1580 - 80 took Drug A - 1500 took placebo 1742 Totals 599 3313 3912

Name _____________________________ Week 5 Worksheet 2) Repeat your calculations AND offer an interpretation using an as treated analysis . (10 points) 3) Using the results from the as treated analysis, calculate the efficacy of drug A. Would you recommend this drug to individuals who suffer from headaches? Explain why or why not. (7 points) 4) Imagine you are interested in studying the relationship between recreational drug use and sexually transmitted infections (STIs). You gather a sample of 463 participants and meet with them to find out if they engage in recreational drug use (yes or no) in the past year and if they have had a STI (yes or no) in the past year. You hypothesize that recreational drug use will increase the occurrence of STIs. The data for your study are presented in the 2X2 table below. STI No STI Totals Recreational drug use 126 29 155 No recreational drug use 51 257 308 Totals 177 286 463 a. What study design is being used and how do you know? (5 points)

Name _____________________________ Week 5 Worksheet b. Write a null and alternative hypothesis for your study (6 points) i. H 0 (null): ii. H 1 (alternative): c. Calculate the appropriate measure of association AND offer an interpretation. (10 points) d. Let’s say the final results of your study get published. However, it is later proved by multiple other studies that there is no association between recreational drug use and STIs. What error did your study committee? (8 points) i. committed a type 2 error ii. made an error that is neither a type 1 error nor a type 2 error iii. committed both a type 1 and a type 2 error iv. committed a type 1 error v. made the correct decision 5) For each of the following scenarios, identify if the study would yield a significant relationship or nonsignificant relationship between exposure and disease. (10 points) a. RR= 1.5; 95% CI=1.1-1.9 Significant Non-significant (2 points) b. OR= 0.70; 95% CI=0.5-1.3 Significant Non-significant (2 points) c. RR= 1.5; 95% CI=0.9-1.8 Significant Non-significant (2 points) d. OR= 7.8; 95% CI=6.5-8.9 Significant Non-significant (2 points) e. RR=0.20; 95% CI=0.14-0.56 Significant Non-significant (2 points)

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

Name _____________________________ Week 5 Worksheet 6) Below are portions of the methods and the results sections of a study that looked at the association between county-level health factors and mortality rates in the United States. Based on this information, what type of study design do you think was used by the researchers? Explain your answer. (7 points). “ We conducted an analysis of county-level data from the 2017 County Health Rankings and the 2017 National Center for Health Statistics to examine the association between health factors and mortality rates in the United States. We estimated the association between health factors and mortality rates, adjusting for potential confounders such as age, sex, race/ethnicity, and poverty status. We found that counties with higher levels of health factors had lower mortality rates. The association was strongest for measures of social and economic factors, including educational attainment and employment, and weakest for measures of physical environment, such as air pollution and access to parks and recreational facilities.” Use the following information to answer the questions below: Suppose a study is conducted to investigate the association between eating spicy food and the development heart burn. You recruit 150 undergraduate students from UW who reported having heart burn in the past week. You also recruit another comparable group of 850 undergraduate students from UW who reported no heart burn in the past week. You have each participant come to your lab where your research assistants collect information on the participant’s eating habits pertaining to spicy food in the past week. Based on data from a systematic review, you hypothesized that those who ate spicy food would be more likely to have heart burn. The results of your study showed that of the 500 participants who reported heartburn in the past week, 100 reported eating spicy food in the past week. Additionally, of the 850 who reported no heart burn in the past week, 400 reported eating spicy food in the past week. A 2x2 table of the outcome and exposure information is listed below. Heart Burn No Heart Burn Total Eating spicy food 100 400 500 No spicy food 50 450 500 Total 150 850 1000 7) What study design did you use? (6 points) a. Ecological b. Metanalysis c. Cohort d. RCT e. Case control f. Cross sectional g. Systematic review Explain your answer:

Name _____________________________ Week 5 Worksheet 8) Write a null and alternative hypothesis for your study (6 points) a. H 0 (null): b. H 1 (alternative): 9) Calculate the appropriate measure of association for the study design used in the example AND provide an interpretation. (8 points) 10) Assume the p-value of your study was 0.02. Which of the following statements is correct? (7 points) a. Heart burn was associated with a significant increase in the odds of consuming spicy food in the past week. b. Heart burn was associated with a significant decrease in the odds of consuming spicy food in the past week. c. Heart burn was associated with a non-significant increase in the odds of consuming spicy food in the past week. d. Heart burn was associated with a non-significant decrease in the odds of consuming spicy food in the past week. e. Heart burn was not associated with consuming spicy food in the past week.