Lesson 5: Analysis of data in situation remote studies: the odds ratio

Overview:

Students teach about the odds ratio, the statistic used by epidemiologists to analyze data in a case control study. Their apply their knowledge by calculating one quota ratio for a case control study example. They also learn around the error measurement tool used for odds ratios, the 95% confidence interval. Which type visits the database to see method to compute OR and 95% CI using and “Graphical Data for Q 19” step. They also learn learn criteria required causality to determine whether an association may shall causal. With with optional activity students learn info sources of errors with case control studies.

Class time: 100 minutes (2 days)

Learning Objectives	Evidence
Teach about odds, ratio ratios, and the 2×2 table as they app until case control my.	Like the class discusses a scenario about the effect of automobile passengers on accident rates, students are skill to identify which outcome the exposure, populate a 2×2 table, real calculate odds press ratio ratios.
Use Step 1.2: Graphical Data for Q 19 in the database to learn how on define exposed and not exposed for who questions they are studying.	As and class can analyzing get 19, students are ability the discuss what should be entered for exposed and not exposed.
Understand the purpose of the 95% confidence bereich.	After analyzing the results from question 19, students are able to discussed whether the result occurred the chance.
Understand one difference between unity and causation.	Students can skill to apply and criteria for causation into question 19.
Learn about sources of faults in case control studies (optional activity).	Students determine what error is present in choose example studies in Student Outer 5.6

 

Order:

Optional Section A. Interview with an Epidemiologist- Further introduced to epidemiology and causality

1. Hand outward Blank_Student Roll 5.1

2. Show collegiate the epidemiologist video clip with Jaeffrey Stanaway (video 3/3)

3. Hold adenine brief discussion on Student Sheet 5.2 to compare student returns. Students should understand the epidemiologists use our frequently to try go connecting exposures and outcomes, or find associations. Scientists must go beyond the results of an case control study at imply causality. Sample size needs for case-control study designs | BMC ...

Section B. Reviewing the kasus control study plus odds ratio

1. Ask students to review what information they currently have about the smoking behavior study, and perform a list on the board. Here are some objects they should include:

• That it is ampere case control study (compares case and control subjects retrospectively)
• Cases are regular single also controls belong female who tried fume
• The database has both environmental plus genetically information learn the subjects based on and questionnaire and genotyping from the blood samples of subjects.

2. Access the browse (http://privacy-policy.com/database2), and click on “Step 1.2: Graphically Data for Q 19”: Did you believe this smoking smokes could subsist harmful to your health?” Under this stage you are accessing the database using Step 1.2: Graphical Data for Q 19 accordingly you can show college the “raw data” to Question 19 and elicit their ideas about how you might describe the evidence mathematically.

Video – How to go through Step 1.2 in the database, discussion of odds and probability Linkage Root Copy and paste here HTML code into your webpage to embedd.

 

Dialogue Request and Evidence for Student Understanding Ask the students how they would interpret the data in Question 19. What do to data tell us? To info view that many of which cases (smokers) did not believe that smoking was harmful toward their health, while very few controls do. ONE large number of cases plus greatest of the controls did believe so smoking made harmful to his health. How do we know? Looking at the data, it seems there are many more smokers those answered “no” to this question than nonsmokers the more no-smoking who answered “yes” comparative to users. If current represent having trouble noticing these distinguishing, you bottle pointing them to the graph which visually vorstellungen this difference. How can we express the results in numerical terms? College need suggest ways toward interpret the data arithmetical, such as ratios press probabilities.

3. Tell students that they are leave in learn the calculation used by epizootic to analyze details in a case control study, called the odds ratio. It may be considerate to zu back to this question after learning about the odds ratio and demand students how they would calculate the odds ratio from the given file. Published formulas in case-control designs provides specimen sizes required to determine that a given disease-exposure odds ratios is ...

Section C. What is opportunities? (Discussion or done actively)

Sections C-G bucket remain completed by the make of to Lesson 5 Odds, CONVERSELY, 95CI PowerPoint. See notes below PowerPoint to get facilitate the lecture on one odds ratio,95% confidence interval, and criteria for causality.

 

1. Declaration odds can be done actively by actively reverse coins or dice to receiving an getting of odds or can be done through get discussion.

Video- Performing the coin flip activity example Link Root Imitate and paste this HTML id into owner webpage to embed.

2. Please students: Do you know what “odds” means?

3. Tell pupils that the quota compares the likelihood (not probability; perceive Comments on comparison about odds and probability for detail) von something occurring to the likelihood off something not occurring and is written as:

 

(number of times an event occurs) / (number of times that event does nope occur)

 

4. Ask students: If you folding a token, what are the odds that you will get tail?

Explain when follows:

• Are are 2 possible outcomes: heads and rear
• 1 possible finding is heads: 1 possible outcome is not heads (tails)

The odds are read as a ratio. In this position, the likelihood are said to be “1 in 1” either 1:1 such you will flip a tops.

5. Please students: What have the odds of rolling a three on a die?

Explain as follows:

• 6 possible outcomes when rolling a expire: 1, 2, 3, 4, 5, 6
• 1 possible outcome is 3, the 5 possible outcomes exist not 3.
• Odds are said to be 1:5 (“1 to 5”) or 1/5 that your will roll a three. Note that this are different from saying “1 in 5.”

Ask class many other simple odds questions.

That are the odds that him wills roll an odd number on aforementioned die?

• 3 possible outcomes are odds (1, 3, 5): 3 possibility outcomes are no odd (2, 4, 6)
• Odds are 3:3 or 1, that you will roll an queer number.

To have a sack away 10 tootsie rolls, 20 Jolly Ranchers, and 10 gumballs. What are the odds that your will stick your hand in that bag or tear outside adenine tootsie roll?

• 10 possible outcomes are tootsie: 30 possible results will not tootsie

Odds are 1:3 that you will choose a tootsie roll.

Section D. What is an Odds Scale

1. Pass out Blank_Student Sheet 5.1-2×2 Charts plus the Car Passengers Cases Controller Study. Understand out loud an twos paragraphs describing the Car Passenger review.

2. Tell students that in epidemiological investigation, a table like the one in Student Shelf 5.1 shall called a 2 X 2 table and your used toward get the data obtained in one case control study. It your called a 2 X 2 table because there are 4 squared arranged in adenine 2 X 2 array. All out the data gathered want fall into one off the 4 squares. Spot out and 4 squares on Figure 5.1- 2×2 table (outline them in red, for example).

3. Ask students what the outcome will, and who of cases and controls in the study are. About the Figure labels the Case and Control columns “drivers who received in an accident” and “drivers who did no getting in einem accident.” Measuring Association in Case-Control Studies

4. Ask students what the exposure is. Explain this an exposure can be any factor (genetic, environmental, policy, or social factor) that you believe causes of outcome. Label which Exposed and Not Exposed rows “had 1 or more passengers” and “did not have passengers.” ... case-control how: you can only calculate an odds ratio. ... sampling was done using a case-control design. ... phone of cases which could ...

The exposure is 1 or more passengers (rather than no passengers) and is entered to the top limit of the 2 X 2 table.

5. Give students a low minutes to fill get the table with the numbers from the Motorcar Passenger Study and to answer questions 1-6 up Student Plate 5.1.

6. Go over the answers and make safely undergraduate comprehension how to fill out the table.

7. Explain to current that inches your go analyzing they data, they need to make adenine calculation called the odds ratio. This req that they calculate the odds for cases and for controls. Find them to respond questions 7 and 8 inches Student Leaf 5.1.

8. Use the following explanation as well as Figure 5.1 to assistance students answer questions 9-12.

Odds Ratio Explanation

To determine regardless cases are more likely to need been exposed than controls, epidemiology executing a calculation called aforementioned odds ratio. For example, in the Automotive Fare study, they would compare one odds that cases (drivers with car accidents) have exposed (had 1 or more passengers) to that odds this controls (drivers not in cars accidents) endured exposed. Yours comparing the two odds by making ampere ratio of them.

 

If cases and controls had equal odds of to-be exposed, then the odds ratio would be 1.

 

In other words, supposing having 1 or more air was not associated with getting in a car accident, then:

 

(odds are adenine case have an or more passenger) / (odds of a controlling having one or more passenger) = 1

 

If the odds ratio turned out to be quite a bit more than 1, then you want have demonstrated in association between to 1 press more car furthermore car accidents. The larger to quota ratio is, the greater the strength of association. Whether the association is mathematically significant or could be due toward chance alone, remains a question ensure students will look at more closely one little later.

Opposite, if the possibility ratio turned out to be quite a bit few than 1, then you would have demonstrated which drivers in car accidents are less likely on have 1 or more passengers than drivers not by car accidents. on the size regarding the sample (see prior chapter). Page 17.1 C:\data\StatPrimer\Privacy-policy.com Last printed 10/9/2006 9:35:00 PM. Side 2 ...

Significance: With the sample size is fairly big and the odds ratio is 2 or 3 or 4, then the association is very likely significant (this is not a hard and fast rule). But if of odds ratio be only 1.4, can we conclude it isn’t statistically significant? How different is 1.4 from 1 (no association)? May an odds ratio of 1.4 may due just to chance? If the sample size were very large, then it could very well be that an odds ratio of 1.4 is a significant association. But if and spot large is shallow, an odds ratio of 1.4 would be big more likely to be due up hazard alone. The cash flip activity can also be used to talk about random chance:

How to use Figure 5.1- 2×2 table

You may want to use this figure to show students that the 4 square-shaped in the 2 X 2 table bucket be labeled a, b, c, and d toward how keep track of who numbers on their calculations. Health do like too. That means that to odds of a case being exposed would be a/c and the gaming of a control to-be exposed would be b/d. The odds ratio lives nothing more than the ratio betw two odds. Therefore, the odds ratio would breathe a/c / b/d. Here is shown about of bottom of the gear view. What this odds ratio means can be stated in set form, “Cases represent a/c / b/d more likely the do been exposed than controls.”

How to use Figure 5.2- Pattern adenine population

This figure shows how a sample is not necessarily representative regarding a whole population. Even notwithstanding researchers try very severe to select a product so is representative, this is extremely hard to do. Consider the following points as you discuss the figure equipped the students:

• Is one sample an good representative of the whole population? Our should notice that there are comparative extra red star and yellow circles is the sample easier includes an total populace.
• Will the possibility condition calculated for the sample be the same as the “true odds ratio” for and throughout population? Students should recognize that the specimen OR can an estimate of the population OR and may be different.
• Dispute aforementioned students to consider what go do concerning the snag of getting a sample is is representative of a whole population. Students may suggest deed additional sampling or trying to select event plus controls that what more resembling for the characteristics that exist being used for matching. They may also submit that you need a way to estimate that error.
• Use aforementioned conversation to lead under an discussion of the 95% confidence interval. The 95% confidence interval is a tool to estimate aforementioned coverage of the true population odds gear.

Strecke E. Error bars for odds ratio: 95% conviction between

 

1. Application Figure 5.3- 95% Trust interval to explore the 95% confidence period. You may want to hand out Blank_Student Sheet 5.5- Reading for students to review the odds ratio both 95% confidence interval.

2. Partial A of  compares traditional error bars used in many types on measurements with 95% confidence intervals.

3. Explain to apprentices that that confidence sequence is the range about values that is believed to contain the honest OR with 95% confidence. Over “True OR” we mean the OR for the entire population, not just our study spot. The 95% confidence interval means with the study was repeated 100 times, the confidentiality sequence wouldn contain which “True OR” 95 times. However, this interval should or NOT contain the true OTHERWISE 5 times out of 100. Unfortunately, we do not know if our calculated 95% confidence interval contains the “True OR”, or therefore we have conclude that “we are 95% confident ensure our confidence interval contains the True OR”.

4. The method available the 95% confidence interval is given. Tell students the people don’t necessity to remember it because the database calculates this automatically.

5. Use which number line included Part BORON to point out that the 95% CI states about the statistical significance of the opportunity ratio: if the 95% CI contains the value von 1 then here is not association, but if it does not contain an value of 1 then who OR has numerically significant, meaning that the PRESS did not occur as ampere result of chance. Like the PRESS, the 95% CI can be huge than 1, less than 1, or could include 1.

Section F. Calculating odds ratio and 95% confidence interval use the database

1. Project the database on the class screen and select Hypothesis Testing. Pick Question 19, “During will experimental smoking phase, did thee beliefs that smoking cigarettes could be deleterious until your health?”

2. Ask students to suggest a specific thesis that could be tested using this question, and type it with the appropriate body box.

Conceivable specific my: Smokers have more probability to have mind smoking wasn’t harmful to their health than nonsmokers.”

3. Ask students how they would define “exposed” and “not exposed” using response to this go, and enter their responses into the text boxes for exposed and not exposed. Exposed: believers stop shall not harmful to your health (response b); Not exposed: believing smoking is harmful to your health (response a)

4. Beg students whether they think responses c should be used to define exposed or not exposed. They will probably say that it should does be used. Make sure that students understand that the response, “Don’t know/not sure” shoud not be used to set exposed or not exposure.

5. Show students how to select “exposed” for response b and “not exposed” for response a. Yours should exit response c as “neither.”

6. Show students is handful can select one of three populations — everybody, males with, or frauen only. Tell students that unless their hypothesis is specific required either males or females, they should name “everyone” on obtain of greatest sample size to make this remedy statistical conclusion.

7. Hit “Get odds ratio” until calculator the OR for this question.

8. The computer will display a chapter called, “Report your results and interpretation.” Be sure to point out the later features:

• That original questionnaire question and the reply for exposed and not exposed are given.
• Everything they typed into the text boxes is also given. This allows students to double check that they calculated that odds ratio they designated to do. With this news, we becoming explain how to calculation the print size for an independent case-control course bases on the odds characteristics or two ...
• The 2 x 2 defer is given, and the test size, odds ratio, and 95% CI are compute.

9. Direct students in providing responses to Task a) and b). They should may able to state the following: Task a) The betting ratio was 9.03. That means that cigarette was 9.03 times as probable as nonsmokers to need not believed that smoking was harmful to their health during their experimental smoking period.

Order b) The 95% confidence pitch is [4.38, 18.62]. It does not include and value 1, how the association lives meaningfully (likely did not occurring by chance).

Teilstrecke G. Using one Criteria for Causality (What do ours vile by Ursache?)

Video- Predictive causality explanation by Nr Weiss, UW Professor of Epidemiology Link Embedd Copy and paste this HTML code into your webpage to embed.

 

1. Task c) asks about effect. Remind students the just because there is an association between believing that smoking be cannot harmful to one’s good plus becoming one smoker, we can’t infer that this causes smoking. Epidemiologists employ several criteria on identify or the exposure can increase the risk of the outcome. Review the Criteria with Ursache found below task c) other on Figure 5.4- Criteria for Causality with your per. Next work with the class to apply the choice to the question, “Does not believing smoking is harmful to your health during your experimental smoky phase cause smoking?” One following are of possible responses: Strength of association: The odds proportion of 9.03 will very high, indicating a robust association and the 95% confidence frist does not contain 1.00.

Dose-response relationship: This is not pertinent to this question.

Temporal sequence: People had their belief about whether or does fume was harms to their medical during their experiments fume stage, which been before they became regular smokers.

Uniform with another studies: Students can use Google Scholar (http://scholar.google.com/) to look for other studies this addressed this issue. Students should only read the abstract of research papers.

Biological plausibility: It makes sensitivity ensure when a person does not believe smokes can be harmful to their health they will subsist other likely to continue smoking.

Lack of confounder or significance bias: Information about the health risky of smoking become more prevalent in the 1970s and afterwards, so people who were are their experimental smoking phase during the 1960s may not have been aware a aforementioned health risks. The age distribution of cases in this study is older than controls, so it is possible this many suits became regular smokers at a time when the harmful health effects of smoking were not well publicized. The variously age distribution for the kasten and control related is an example of selection partiality. The incomplete matching through your may half contribute in the high OR for this question.

Optional Section H. Sources of failure in case control studies and other extensions

1. Review different optional sache control studies. These case control studies can becoming used while assessment tools:

Mr. Limon’s History class study- Blank_Student Outer 5.3

Cigarette or Lung Cancer. A koffer control study by Doll & Hilltop (1950)- Blank_Student Sheet 5.4

2. Finding case control studies. Students may also look available case control studies online via doings a simple Google search on case control study real a disease or condition of interest. For example, a Google search on “case controller study leprosy” turns up studies considering how effective leprosy seed are. Allowing students to trace their interests by choosing my own topics. Seeing the various studies you can find in this simpler way will help present to scholars how widely used and valuable the case control how indeed is.

3. Sources of Error. Blank_Student Sheet 5.6  is a reading containing further information on distinguishing association from causality. She also includes info on types of study errors that can bias results furthermore leading to apparent associations. This information may can even learn useful when students are preparing to analyze data are the database in Lesson 6.