Quiz 3 Content

Question 1

Define epidemiology.

Answer 1

Study of the distribution of disease in relation to person, place and time, and measures of risk associated with exposures to disease

The study of how often diseases occur in different groups of people and why

Question 2

Define incidence.

Answer 2

Count of new cases over a period of time in a population ­

E.g. Number of new cases of diagnosed Type 2 Diabetes in BC in 2009

Question 3

Define prevalence.

Answer 3

Total number of cases of the disease in the population at a given time ­

E.g. Total number of BC residents with cancer in 2009

Question 4

Discern between mortality and morbidity.

Answer 4

Mortality - death rates

Morbidity – incidence of ill health/disease

Question 5

Describe epidemiological research.

Answer 5

Exposure → potential cause of an outcome; may be referred to as ‘independent variable’

Outcome → may be influenced by the exposure; also known as ‘dependent variable’ or ‘endpoint’

Question 6

Compare cross-sectional, cohort, and case-control studies regarding the timing of outcome and exposure.

Answer 6

Cross-sectional: O + E

Cohort: E (follow over time) O

Case-control: O (look back for) E

Question 7

What are the four steps of case-control study design? [4]

Answer 7

1. Investigator selects cases and controls

2. Data about exposure to possible etiologic factors collected in both groups

3. The frequency/amount of exposure among the two groups is compared

4. An odds ratio may be calculated to compare frequency of exposure among cases compared to controls

Question 8

How are cases and controls identified for a case-control study?

Answer 8

CASES: ­

Inclusion criterion: Have the outcome of interest

There should be a defined method for case ascertainment

CONTROLS:

­Inclusion criteria: Do NOT have outcome of interest

Must be selected to deal with potential confounding:

1. May choose controls with another condition to control for interaction with health care system

2. May match to cases for age, sex, ethnicity, education, income, area of residence, …

Question 9

Where are cases and controls selected from for a case-control study?

Answer 9

From the general population: ­

E.g. registries, households, telephone sampling ­

Can be costly and time consuming ­

Prone to low response rate

From clinic/hospital: ­

Often easier to identify ­

Typically higher response rate

Question 10

What are advantages of case-control studies? [4]

Answer 10

Useful for studying: ­

• Rare conditions ­
• Conditions with a very long lag between exposure and outcome
• Usually requires fewer subjects than cohort or cross-sectional studies
• Generally quicker and cheaper than cohort studies

Question 11

What are limitations of case-control studies? [6]

Answer 11

1. Difficult to establish temporality ­

Did exposure precede disease OR did diet/lifestyle change because of disease?

2. How far back?

Must determine appropriate time period in the past in terms of biological plausibility

E.g. For folic acid and neural tube defect – appropriate to ask about intake from 1 year ago BUT for ω-3 fatty acids and heart disease, how far back to ask about? – 1 year, 5 years, 10?

3. Reliance on Memory

Difficult to recall past practices accurately

4. RECALL BIAS

Cases more motivated than controls to search their memory to make sense of why they have the condition

Thus, cases may be more likely to report past exposure

5. Misclassification

Misclassification of cases

Misclassification of exposure status

6. Cannot calculate disease incidence or prevalence

Proportions of cases in the study sample ≠ proportions with the outcome in the population

Question 12

What is non-differential misclassification?

Answer 12

Exposure status of cases and controls equally likely to be misclassified ­

E.g. FFQ was not valid, therefore estimation of exposure was inaccurate in both cases and controls

­May affect ability to see an association when one exists

Question 13

What is differential misclassification?

Answer 13

Cases and controls differ in probability of misclassification ­

Method of assessing exposure leads cases to be more likely than controls to be misclassified ­

Eg. If cases and controls are assessed by different research assistants and one of the research assistants overestimates nutrient intake compared to the other ­

Bias may contribute to a false association

Question 14

What is the difference between an odds ratio and risk?

Answer 14

Risk (Probability) = likelihood of event of event A total number of events

Odds = likelihood of event A likelihood of alternate event

Example: Risk vs. Odds of rolling a 6 on a die

Risk = 1/6

Odds = 1/5

Question 15

If the risk is ½ (50%), what are the odds?

Answer 15

1:1

Question 16

If the odds are 3:1, then the risk is…?

Answer 16

¾ (75%)

Question 17

If the risk is 1/10 (10%), then the odds are…?

Answer 17

1:9

Question 18

If the odds are 1:99, then the risk is…?

Answer 18

1/100 (1%)

Question 19

What is an odds ratio?

Answer 19

= Odds of exposure among cases Odds of exposure among controls

Exposure refers to independent variable, could be nutrient intake, supplement intake etc.

Question 20

How is an odds ratio interpreted?

Answer 20

OR > 1 → exposure more likely in cases than controls (There is a positive association between the exposure and outcome)

OR < 1 → exposure is less likely in cases than controls (There is a negative association between the exposure and outcome OR there is a protective effect of the exposure)

OR = 1 → no difference in odds of exposure between cases and controls

Question 21

When is an odds ratio considered statistically significant?

Answer 21

An OR is considered statistically significant when the confidence interval (CI) excludes 1.0

The Confidence Interval is the range of values within which the true population value likely exists

Eg. 95% CI ­

95% chance that the true value for the population lies within the CI

Eg. 99% CI ­

99% chance that the true value for the population lies within the CI

For α = 0.05 (Type I error), a 95% (1 – α) CI is used

Eg. OR 3.01 (95% CI 1.86-4.85) ­

Sample odds ratio = 3.01 ­

95% chance that the true population value lies somewhere between 1.86 and 4.85

Generally, the larger the sample size, the smaller (and more precise) the CI

Question 22

Generally, the larger the sample size, the larger (and less precise) the confidence interval.

True or false?

Answer 22

False.

The larger the sample size, the smaller (and more precise) the confidence interval.

Question 23

What three questions should you ask yourself when you interpret an odds ratio?

Answer 23

Is the OR >1 or <1?

Does the CI include 1?

Is the CI very wide or very small?

Question 24

What is a multivariate (‘adjusted’) odds ratio?

Answer 24

Cases and controls may differ on variables other than the exposure variable that could act as confounders (e.g., age, BMI, income, etc)

To attempt to control for possible confounding, the other variables can be accounted for in analysis → This gives an adjusted or multivariate odds ratio

Non-adjusted odds ratio = crude odds ratio

Question 25

What is considered a ‘strong’ odds ratio?

Answer 25

\>4 or \<0.25 = strong 2-4 or 0.25-0.5 = moderate 1.5-2 or 0.5-0.67 = possible but weak \<1.50 or \>0.7 = possible but very weak

Question 26

Describe cohort studies. [4]

Answer 26

Exposure is assessed at baseline Participants followed over time Characteristics of those who develop the outcome are compared to characteristics of those who do not Disease incidence can be determined

Question 27

Describe the two types of cohort studies.

Answer 27

**Prospective** → at study onset outcomes have not yet occurred **Retrospective** → at study onset outcomes have already occurred (e.g., studying medical records to see if vaccinations are associated with decreased cancer risk)

Question 28

Compare retrospective cohort research with case-control research.

Answer 28

**Retrospective Cohort → Start with EXPOSURE.** Determine if exposure relates to development of outcome. ­ Example: Take medical records from 100 people who were vaccinated and 100 who were not. Determine if being vaccinated associated with developing cancer **Case-control → Start with OUTCOME.** Ask about previous exposure. ­ Example: Take 100 people who have cancer, and 100 who do not. Ask about previous vaccinations. Determine if being vaccinated associated with developing cancer

Question 29

What is a nested case-control study?

Answer 29

**Case-control study nested within a cohort study** Participants recruited for cohort study ­After some outcomes have been assessed, take a sample of the participants and analyze the data as a case-control study Eg. 1, 000 people recruited for cohort study on cancer 10 years into the study: case-control study on diabetes ­ 100 people with diabetes, 100 without ­ Determine whether differences in exposure between those who got diabetes in those who did not

Question 30

Which of the Bradford Hill Criteria for Causation cannot be shown with case-control or cross-sectional studies?

Answer 30

Temporality

Question 31

Describe the advantages of prospective cohort studies. [5]

Answer 31

* Can establish timing and directionality of events * Can calculate incidence (new cases) * Avoid potential for recall bias * May be easier (or at least more feasible) than a randomized controlled trial * If well designed, can assess many outcomes

Question 32

Describe limitations of prospective cohort studies. [4]

Answer 32

1. Confounding: Exposure may be related to another (unknown) factor that is associated with the outcome 2. Level of exposure may change over time 3. Loss to follow up: Particularly problematic if people who withdraw from the study have different characteristics than those who remain involved 4. Large sample size is needed if outcome of interest is rare

Question 33

A researcher wants to study whether drinking green tea can decrease the risk of kidney cancer using a prospective cohort design. The researcher determines that he needs to study **at least 100 people** that develop kidney cancer to have enough *_power_* to detect whether green tea is associated with cancer incidence. Assuming about 1 in 75 people are expected to develop kidney cancer in Canada, how many people must the researcher recruit into his study in order to find at least 100 people that develop kidney cancer?

Answer 33

At least 7,500 (1 in 75 develop kidney cancer = 100 people)

Question 34

What is absolute risk?

Answer 34

(Absolute) Risk of outcome in the whole sample = (Number with outcome) / (Total sample size)

Question 35

What is relative risk?

Answer 35

Likelihood that outcome occurs in exposed group compared to unexposed group Ratio of the risk among exposed to the risk among unexposed Null hypothesis → relative risk = 1 (no relationship) E.g., Subjects who took Vitamin C supplements were [RR] **as likely** to die from heart disease as subjects who did not take Vit C supplements

Question 36

How is relative risk interpreted?

Answer 36

**RR \< 1 →** outcome less likely to occur in exposed group than non-exposed group (Negative association between exposure and outcome) **RR \>1** → outcome more likely to occur in exposed group than non-exposed group (Positive association between exposure and outcome) **RR = 1 →** no relationship between exposure and outcome **With RR can say:** “Exposed are x times **as likely** to have the outcome” Eg. RR = 0.70, Subjects who took Vitamin C supplements were 0.70 times (or 70%) as likely to get heart disease Eg. RR = 2.2 Subjects who drink milk were 2.2 times as likely to have strong bones compared to those who did not drink milk

Question 37

Describe how relative risk may be expressed by the amount risk is increased/decreased in exposed compared to unexposed subjects.

Answer 37

**If RR is \<1, being exposed decreases the risk of having the outcome** Decreased risk = (1.00 - RR) x 100 Eg. RR = 0.70, (1.00-0.70 \* 100 = 30%) ­ exposed subjects were 30% less likely to get the outcome than unexposed subjects **If RR is \>1, being exposed increases the risk of having the outcome** Increased risk = (RR – 1.00) x 100 Eg. RR = 1.5, (1.5-1.0 \* 100 = 50%) ­ exposed subjects were 50% more likely to develop the outcome

Question 38

When is relative risk statistically significant?

Answer 38

RR is statistically significant when the confidence interval does NOT include 1

Question 39

What is risk difference?

Answer 39

Risk Difference = (Risk of outcome in exposed group) – (Risk of outcome in unexposed group) Null hypothesis: RD = 0

Question 40

What type of studies can odds ratios be calculated for? What about relative risks?

Answer 40

Odds Ratios may be calculated for cohort studies, case-control studies or experimental studies Relative Risks are used in cohort studies and experimental studies (where risk is compared between treatment group/control group) RR should NOT be calculated for case-control studies

Question 41

Describe “person years”.

Answer 41

In a population of 1000 healthy persons, 20 develop an outcome over two years of observation: Incidence = 20/1000 ­ Could also call this: 20 cases/2000 person-years Person years is way to incorporate both time and the population at risk in calculations of risk

Question 42

Describe the Framingham Heart Study

Answer 42

Cohort study Objective: Identify factors that contribute to CVD by following its development over time in generally healthy people Ongoing since 1948 Over 1000 research papers published

Question 43

Describe the nurses' health study.

Answer 43

Original cohort 1976: 122,000 married RNs aged 30-55 recruited to study effects of oral contraceptives. Health status, diet, etc. assessed every 2-4 yr NHS II 1989: ~117,000 nurses aged 25-42 NHS III 2010: Nurses aged 22-42, more culturally diverse

Question 44

Give two examples of Canadian cohort studies.

Answer 44

CHILD STUDY https://childstudy.ca/about/ CAN PATH https://canpath.ca/about/ https://canpath.ca/2019/06/cptp-celebrates-10-years-ofprogress/

Question 45

What is experimental research?

Answer 45

Researcher controls/manipulates the exposure of interest

Question 46

Define true experimental research. [3]

Answer 46

1. Treatment variable is controlled by researcher 2. Participants randomly assigned to groups 3. Control group for comparison

Question 47

What are the types of controls available?

Answer 47

**1.** Control group given nothing (no exposure/treatment) **2.** Placebo **3.** Another treatment: a. Used when treatment/exposure involves interaction with the researchers b. Used when a known standard treatment exists

Question 48

What are the benefits and limitations of a control group being given nothing (i.e., no exposure/treatment)?

Answer 48

Benefits: Easy to do Limitations: May be able to figure out who is in the treatment group and who is in the control group

Question 49

What are the benefits and limitations to placebo (i.e., inactive substance/treatment) as control?

Answer 49

Benefits: Enables blinding Limitations: May be difficult to find an appropriate placebo

Question 50

What is the difference between single and double blind?

Answer 50

Single blind → Participants do not know if they receive active treatment or placebo Double blind → Neither participants nor investigators know who gets treatment or placebo; important to prevent investigator bias

Question 51

Describe three challenges with placebos in nutrition research.

Answer 51

**1.** Not always possible to find a placebo that is similar to the treatment. Eg. Intervention = increase fruit & veg intake, placebo = ? **2.** Difficult (or unethical) to have a placebo group with “zero” intake of a nutrient **3.** Participants in placebo group may still obtain nutrient of interest from food; treatment and control group intake may overlap

Question 52

Describe the following control strategies: Another treatment: a. Used when treatment/exposure involves interaction with the researchers, or b. Used when a known standard treatment exists

Answer 52

**A.** If participants interact with researchers in the study, these interactions may influence health outcomes; control groups should have similar interactions as treatment group **B.** When a standard treatment or known therapy exists, this must be given to the control group; it is not ethical to withhold treatment with known benefits

Question 53

What is the goal of randomization?

Answer 53

To achieve baseline comparability between groups on factors related to outcome Achieve comparison groups that are similar in all characteristics EXCEPT for the treatment/intervention ­ Minimize potential confounding effects

Question 54

What is quasi-experimental research and when is it useful?

Answer 54

Lacks the key ingredient from “true” experimental designs – the random assignment of subjects Useful when subjects exist in natural groupings or random assignment is not possible

Question 55

What are the 4 types of experimental designs?

Answer 55

Post test-only control group design Pretest-post test control group design Solomon four-group design Crossover designs

Question 56

Describe when post-test only control group design would be useful and name potential limitations.

Answer 56

Appropriate when pretest is not convenient or possible Generally okay when groups are randomized (assume similar on most characteristics) _Potential limitations:_ Failure of randomization (groups differ) Inability to assess whether results vary depending on pretest level

Question 57

Describe the effects of pre-testing.

Answer 57

Taking a pretest could “sensitize” controls to the issue being investigated May affect knowledge and attitudes May affect final score, even in the control group E.g. Study to assess effect of education program on nutrition knowledge and attitudes. By giving pre-test, both controls and treatment group are exposed to the topic – controls may then seek info on nutrition on their own

Question 58

Describe solomon four-group design.

Answer 58

See image.

Question 59

Describe cross-over design.

Answer 59

Each subject receives both the treatment and control Washout period: period between treatments that allows for return to baseline values Compare each person’s responses to treatment versus control Each subject serves as their own control NOTE: Non-crossover study = parallel groups study

Question 60

Describe advantages and limitations of cross-over design.

Answer 60

Advantage: • Subjects serve as their own control, therefore, no differences between treatment and control group • Can use smaller sample size Limitation: • Can not use crossover design if treatment leads to a permanent change (Eg. education program)

Question 61

What is needed for a sample size calculation? [5]

Answer 61

1. Power 2. Significance level (α) 3. Effect size/difference between groups that we would like to be able to detect 4. Variability (Standard deviations) 5. One or two-sided tests

Question 62

What is power in sample size calculations?

Answer 62

Power to detect a difference among groups (if one exists) Probability of rejecting the null hypothesis if the null hypothesis is false Probability of NOT making a type II error

Question 63

Explain the influence of effect size on sample size calculations.

Answer 63

­ With smaller differences between groups, need larger sample size to detect the differences

Question 64

Explain the influence of variability on sample size calculations.

Answer 64

With greater variability, need more data points to represent the group