Misc

Question 1

Difference between ratio and proportions

Answer 1

proportions are related i.e. numerator is part of the denominator
Ratios are not related

Question 2

Rate

Answer 2

Rate = n/N * 10n (constant population)

n/N is a proportion

Question 3

Prevalence = Incidence x duration of disease

Answer 3

Similar if incidence high and duration short e.g. influenza

Question 4

Case control

Answer 4

Looks at outcome (disease), then retrospective attributable factros

Question 5

Cohort study

Answer 5

Looks at risk factor, prospective study

Question 6

Confounder

Answer 6

1. an independent risk factor for the outcome of interest; i.e. it must be a risk factor for the outcome even in the absence of the study factor, based on current knowledge;
2. associated with the study factor, in the data being analysed; and
3. not simply an intervening factor between the study factor and outcome; and
4. not part of the study hypothesis.

Question 7

Cross-sectional study

Answer 7

study that examines the relationship between diseases (or other health related characteristics) and other variables of interest as they exist in a defined population at one particular time. The temporal sequence of cause and effect cannot necessarily be determined in a cross-sectional study.

Question 8

Epidemic

Answer 8

The occurrence of disease clearly in excess of the level that is accepted as normal (ie the endemic level).

Question 9

Odds ratio

Answer 9

The ratio of the odds of an event in an exposed group to the odds of an event in the control group. Most commonly used as the effect measure in a case-control study.

When the incidence rate is small, odds ratios are very similar to relative risks.

Question 10

Phase I Trial

Answer 10

Usually performed on healthy volunteers without a comparison group to determine toxicity and appropriate dosing.

Question 11

Phase II Trial

Answer 11

looking for preliminary evidence of efficacy. Often performed on healthy volunteers. These are sometimes randomised controlled trials.

Question 12

Phase III Trial

Answer 12

Studies that are a full-scale evaluation of treatment. After a drug has been shown to be reasonably effective, it is essential to compare it to the current standard treatments for the same condition. Phase III studies are usually randomised controlled trials.

Question 13

Population attributable risk

Answer 13

The incidence of a disease in a population that is associated with exposure to the risk factor. It is equal to the incidence rate in the entire population minus the incidence rate in the non-exposed.
PAR = Ipop − IU

Note: to use this formula, the incidence of disease in the total population (not study population) must be known

OR PAR = Attributable risk x Prevalence of exposure in the population =(IE –IU)×pe

Where pe = proportion of exposed individuals in the total population (i.e. convert percentage to proportion for the calculation)

Question 14

Population attributable risk percent/fraction (aetiologic fraction)

Answer 14

The percentage of total disease incidence that is attributable to the exposure.

Question 15

Sensitivity

Answer 15

The proportion of people with a characteristic who are assessed as having it by the measuring device or test

Question 16

Specificity

Answer 16

The proportion of people without a characteristic who are assessed as not having it by the measuring device or test.

Question 17

Statistical power

Answer 17

The probability that the null hypothesis will be rejected if it is indeed false. In studies of the effectiveness of healthcare interventions, power is a measure of the certainty of avoiding a false negative conclusion that an intervention is not effective when in truth it is effective.

Question 18

Rate

Answer 18

An expression of how frequently an event occurs per head of population over time.

Question 19

Rate ratio

Answer 19

The ratio of the incidence rate in one group to the incidence rate in another. Although sometimes called a relative risk or risk ratio, rate ratio is the ratio of two rates, and risk ratio (or relative risk) is the ratio of two risks. The terms are frequently used interchangeably in the epidemiological literature, but strictly speaking they should not be. Although usually applied to incidence, you may see the term prevalence rate ratio for the ratio of one prevalence rate to another. A relative risk or rate ratio of one indicates no difference between comparison groups.

Question 20

Absolute risk is a statement that expresses the probability of a disease (or some other outcome of interest) occurring in the population. It is the same as the cumulative incidence of the disease and calculated in the same way.

Answer 20

For example, if 30 of 1000 heavy smokers (Exposed) developed lung cancer (Outcome) over a 10- year period, then the absolute risk (or probability) of lung cancer in this population would be 0.03 (= 30/1000) over 10 years.
This could also be expressed as an incidence rate i.e. 0.003 cases of lung cancer per year, or 300 cases per 100,000 population per year; or as an incidence density rate i.e. 300 cases per 100,000 person-years.

Question 21

Rate Ratio (Relative Risk) = IE / IU

Answer 21

where IE = incidence in exposed
IU = incidence in unexposed
IE = deaths among exposed total exposed
= a / (a+b) = 8 / 50 = 0.16
IU = deaths among unexposed total unexposed
=c/(c+d)= 5/50=0.10
Therefore the Rate Ratio = 0.16 / 0.10 = 1.6

Question 22

Odds ratio

Answer 22

An approximation of the rate ratio (relative risk) is the odds ratio (OR). The odds ratio is used as the measure of association in studies in which incidence rates (i.e. speed at which new cases occur) cannot be measured, such as in a case-control study. Because cases and controls are selected on the basis of having or not having a disease, it is not possible to measure disease incidence rates in a case-control study, and therefore the rate ratio cannot be calculated.

The odds ratio is a measure that compares the odds (or likelihood) of exposure to a factor of interest among those with disease (cases) and the odds of exposure to the same factor among those who do not have disease (controls).

Question 23

Attributable risk

Answer 23

Ie-Iu

AKA RR-1

Incidence is used to estimate the Attributable Risk (AR=IE – IU), therefore, AR cannot be directly estimated from a case-control study because incidence cannot be calculated from a case-control study

Question 24

Attributable risk percentage

Answer 24

AR% =
Attributable risk Incidence in exposed (IE)
= (IE -IU)/IEx100
= RR-1/RRx100

For case-control studies, an indirect estimate of AR% can be calculated from the OR

AR% = ((OR-1) × 100)/OR

Question 25

Type I error

Answer 25

False rejection of H0

While type I error is synonymous with sensitivity or false positives

Question 26

Type II error

Answer 26

False acceptance of H0
Type II error is synonymous with power, i.e. a false negative rate.
The way to minismise Type II errors is
o Bigger real difference between the groups
o More patients
o More events
o Less random measurement error

Question 27

Power

Answer 27

1-probability of making a Type II error

Question 28

Levels of evidence

Answer 28

Ia- evidence from meta-analysis of randomised controlled trials

Ib- evidence from at least one randomised controlled trial

IIa- evidence from at least one well designed controlled trial which is not randomised.

IIb- evidence from at least one well designed experimental trial

III- evidence from case, correlation and comparative studies.

IV- evidence from a panel of experts.

Grade of recommendations:

Grade A- based on evidence from at least once RCT (Ia,Ib)

Grade B- based on evidence from non-RCT (IIa,IIb,III)

Grade C- based on evidence from a panel of experts (IV).

Question 29

Relative risk reduction

Answer 29

(RRe-RRi)/RRe