Critical Appraisal Observational Studies

Question 1

What is an observational study?

Answer 1

A study in which no intervention is made

It provides estimates and examine associations of events in their natural settings without recourse to experimental intervention

Question 2

What is the role of an observational study in clinical research?

Answer 2

Draws inferences about the possible effect of a treatment on subjects, where assignment of treatment and control is out of control of researcher

Contrast with experimental studies, such as RCT where subject is randomly assigned to a treatment group or control

Question 3

What are some pros to a retrospective observational study?

Answer 3

inexpensive to conduct

Completed in shorter time period

Easier to access a larger number of subjects

Allows results to be obtained more quickly

Useful for studying exposure that no longer occur

Question 4

What are some cons to a retrospective observational study?

Answer 4

Subjects may not remember past information e.g. prescription history

Information and data may be less complete and inaccurate

Question 5

What are some pros to prospective observational studies?

Answer 5

Information and data may be more complete and accurate

Direct access to study subjects enhances reliability of data

Question 6

What are some cons to a prospective observational study?

Answer 6

Expensive

Completed over a longer period of time

More difficult to access subjects and usually requires a larger no. of subjects

Loss of subjects from the study over time may be substantial

Exposure status and diagnostic methods for disease may change

Question 7

What is a descriptive observational study? (give some e.g.)

Answer 7

Attempts to uncover or portray the occurrence of the condition or problem

Provides insight about the pattern of disease, drug use in a population

e.g. case reports, case series, cross sectional studies

Question 8

What is an analytical observational study? (give some e.g.)

Answer 8

Determine the cause of the condition or problem

Determine the cause-effect relationship

e.g. cohort studies, case control studies

Question 9

What are case reports?

Answer 9

A.K.A = spontaneous case reports or passive surveillance

Describes a single patient who was exposed to a drug and experienced a particular, usually adverse effect

Question 10

What are case series?

Answer 10

Collections of patients all of whom have a single exposure, whose clinical outcomes are the evaluated and described

Question 11

What do case series and case reports have in common?

Answer 11

No comparison groups

Question 12

What is a cross-sectional study?

Answer 12

Snapshot of pop at a specific point in time, allow observations across a wide pop –> asses prevalence of acute or chronic conditions or results of med intervation

Comprised of audits and questionnaires

A.k.a prevalence study

Question 13

What are some cons to a cross-sectional study?

Answer 13

Moderately expensive

Not suitable for rare/orphan disease

Difficulty in recalling past events may contribute to bias

Question 14

What are case control studies?

Answer 14

Assembles group of cases (w/ disease) and controls (w/out disease)

Examines exposure histories of cases and controls which are used to establish the extent of association between exposure and disease

Question 15

What are some strengths to case control studies?

Answer 15

Efficient in terms of cost and time compared to other analytical designs

Good design for studying rare outcomes and disease w/ long latency

Examine multiple potential risk factors for single outcome and possible interrelationships among factors

Question 16

What are some limitations to case control studies?

Answer 16

Prone to more biases compared to other analytical studies

Temporal relationship between exposure and outcome may be difficult to establish

Cannot directly derive incidence rates (unless population based)

Question 17

What is a cohort study?

Answer 17

Assembles a group of persons without the disease(s) of interest at onset of study, ascertains exposure status of individual

Follows cohort over time to determine the development of disease in exposed and non-exposed

A.k.a prognosis study

Question 18

What is a prospective cohort study?

Answer 18

Study pop and exposure status are identified at beginning of the study

Subjects are followed into the future to determine outcome

Question 19

What is a retrospective cohort study?

Answer 19

Study population and exposures status are determined using historical data

Outcome determined at beginning of study

Question 20

What are the strengths of a cohort study?

Answer 20

It allows direct comparison of outcome incidence in exposed and unexposed groups

can establish temporal relationship between exposure and outcome

More ethical than RCTs when the exposure is harmful

valuable when the exposure is rare

Question 21

What are the limitations of a cohort study?

Answer 21

For rare outcomes, large sample sizes or long follow up are necessary

Significant loss to follow-up/non-participation of the study subjects

Retrospective cohort studies require availability of data from pre-existing records

Question 22

In oncological research, what is a hazard and hazard ratio?

Answer 22

Hazard = an event

Hazard ratio (measure outcome) = a.k.a. relative hazard, calculated as the hazard of death for the intervention group divided by the hazard of death for the control

Reflects time survived to an event

Question 23

What does a hazard ratio of 1 mean?

Answer 23

equal event rate between groups

Question 24

What does a hazard ratio of 2 mean?

Answer 24

Twice as many patients in active group will have an event compared to the control in the next unit of time

Question 25

What does a hazard ratio of 0.5 mean?

Answer 25

half as many patients in active group are having the event compared to the control in the next unit of time

Question 26

In the context of oncology, what are survival curves?

Answer 26

Statistical pictoral of the survival experience of some group of patients in form of a graph showing the percentage surviving versus time

Best known way to estimate prognosis

Question 27

What does the vertical (y) axis represent on a survival curve?

Answer 27

proportion of people surviving

Question 28

What does the horizontal (x) axis represent on a survival curve?

Answer 28

Gives time after the start of the observation or experiment

Question 29

What does any point on the survival curve represent?

Answer 29

proportion or percentage surviving at a particular time after the start of an observation

Question 30

What is the Kaplan Meier (KM) curve/analysis?

Answer 30

Calculates the proportion surviving to each point that a death occurs –> provide more accurate estimate of survival curve

e.g. sees how many people are alive at 1 year after diagnosis

Estimates the survival curve in the presence of censored observations

Question 31

What are censored observations?

Answer 31

Refers to people who have dropped out of the study early or information on their survival time is incomplete

Question 32

What are some questions to consider when doing a critical appraisal for observational studies?

Answer 32

Is the study based on random or pseudorandom samples?

Are the criteria for inclusions in study clearly defined?

Were outcomes assessed using objective criteria?

If comparisons were made, was there a sufficient description of the group?

Was an appropriate stat analysis method used?

Question 33

What are some questions to ask when critically appraising the validity of a cohort study?

Answer 33

Is sample representative of patients in pop as a whole?

Are patients at a similar point in the course of their condition/illness?

Are outcomes assessed using objective criteria?

Was a follow up carried out over a sufficient time period?

Were the outcomes of people who withdrew described and included in the analysis?

Were outcomes measured in a reliable way?

Was appropriate statistical analysis used?

Question 34

What is the checklist for assessing the validity of case control studies?

Answer 34

Are confounding factors identified and strategies to deal with them stated?

Has bias been minimised in relation to selection of cases and of controls?

Were outcomes assessed using objective criteria?

Were outcomes of people who withdrew described and included in the analysis?

Were outcomes measured in a reliable way?

Was appropriate statistical analysis used?