Week 122 - Measurements in population I Flashcards Preview

MHS > Week 122 - Measurements in population I > Flashcards

Flashcards in Week 122 - Measurements in population I Deck (36):

What is incidence?

Proportion of a group that will develop a certain disease over a time period


What is prevalence?

Amount/proportion of people that have a disease at a given time


List the study order of hierachy of evidence?

1) Systematic reviews with at least one RCT

2) RCT

3) Non RCTs

4) Cohort studies

5) Case-control studies

6) Self controlled case series

7) Ecological studies

8) Cross sectional studies of incidence and prevalence

9) Case series and case reports


What is an RCT?

A randomised controlled trial

Randomisation of sorting people into groups removes any confounding factors between the intervention group and control group


What is a self controlled case series?

A study in which the incidence of a certain outcome is compared to the incidence of the outcome when the individual is not exposed

e.g. occurence of adverse reaction when given a drug compared to when not given


What is a cross sectional study?

A study that collects data on incidence/prevalence of a disease in a population at a specific point in time


What is an ecological study?

A study in which populations/groups are analysed to study effects of risk factor on health

e.g. rates of obesity in different countries


What is the disadvantage of an ecological study?

Hard to control due to confounding variables


What is a case report study?

A detailed report of a patient with a condition of interest


What is a case series study?

A descriptive study of patients with a known exposure or condition


What is a disadvantage of case-series study?

Can involve selection bias


List the two types of research studies?

Experimental studies

Observational studies


What is an experimental study?

Study which exposure is determined as part of study design


What is an observational study?

Study in which exposure is not determined as part of the study design

- already existing expossure and consequences are studied


Give an example of an experimental study?


Non-randomised trials


What is the null hypothesis?

Statement being tested in a test of hypothesis


What is the alternative hypothesis?

What is believed to be true if the null hypothesis is false


What is a case control study?

A study which starts with the identification of people:
- with the disease of interest, and a
- control group without the disease


What is a retrospective case control study?

A case control study which starts with the case and looks back at exposure to see if there is an associated risk


What is a prospective case control study?

A case control study which recruit individuals as they develop the disease of interest


What is a p-value?

Probability of observing a difference as large as previously seen in a study due to chance

p-value < 0.05 => statistically significant


What is a cohort study?

A study that starts by looking at a certain group of people and identifies those who are exposed to the risk factors and those who are not.

It then compares the incidence of disease development between the two exposure groups.


What is the difference between a prospective and retrospective cohort study?

Prospective = current exposure

Retrospective = identifies "past" exposure to risk factors


What are the advantages of a case control study?

Better for rare diseases

Retrospective case controls are quick and cheap


What are the disadvantages of a case control study?

Prone to bias

Time of first exposure to risk factor is difficult to judge

Incidence can't be determined


What are the advantages of a cohort study?

Better for rare exposures

View the multiple effects of one exposure

Less prone to bias

Incidence can be determined


What are the disadvantages of a cohort study?

Can be slow and expensive


What are the two types of random errors?

Type 1 = alpha error

Type 2 = beta error


What is a type 1 error?

A false positive


What is a type 2 error?

A false negative


How can type 1 errors be controlled?

By lowering the statistical significance level


How can type 2 errors be controlled?

Increase sample size and power of the study


What is sensitivity?

Measurement of how good a test is at diagnosing disease?
i.e. true positive rate

A high sensitivity reduced type 2 error


What is specificity?

Measurement of how good a test is at diagnosing no disease?
i.e. true negative rate

High specificity reduces type 1 errors


What is the formula for relative risk?

(Absolute risk when exposed) / (absolute risk when non-exposed)

i.e. [De / (De + He)] / [De' / (De' + He')]


What is the formula for odds ratio?

Odds of exposed / Odds of non-exposed

i.e. (De / He) / (De' / He') = (De He') / (He De')