CPH - Epidemiology (Whincup) Flashcards

1
Q

Appreciate the potential value of studying health and disease in populations. (5)

A

Population approach can show you:
- How much impact a disease /ilness has on health in community/population in terms of mortality and
morbidity - BUT clinical iceberg
- who may be at risk by seeing how illness varies between people/ places/ periods
- potential causes on wider scale
- how we can prevent disease
- management on population level - service optimisation, interventions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Define the terms epidemiology and public health

A

Both population level
Epidemiology - Study of the distribution and determinants of health and disease in human populations
Public health - Science & art of prolonging life, preventing disease, promoting health at pop level by organised efforts of society

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Explain the concepts of rates and risks as a tool for the study of disease in populations

A

Rate = number of cases / population at risk
Incidence rate = new cases of disease
Prevalence rate = presence of disease (new and old)
Mortality rate = occurrence of death

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Be aware of major differences in disease risks over time and between places.

A

PLACE - prevalence of a disease can vary in different countries, regions, boroughs, neighbourhoods
TIME - short and long term (seasons vs years)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Types of studies to measure differences between places (2)

A

Ecological study can look at population/community, to compare exposures and disease rates in different geographies
Best when causal factors are fairly consistent within population (not between populations being studied)
ECOLOGICAL FALLACY - Assumes that exposure and outcome are related in individuals (but it’s based on population data), may be CONFOUNDING factors

Migration studies assess whether differents are due to adult environment or genes / early environment
PITFALLS - Selection bias (are the people who migrated the same as the people in the original country), information bias (do migrants and non-migrants share info in the same way), potential influence of stress of migration

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Pitfalls of studying disease between places

A

PITFALLS - Are the geographical differences real? Differences in ASCERTAINMENT of medical care, differences in DIAGNOSIS (customs, equipment), differences in RECORDING information, differences in POPULATION STRUCTURE (age, gender, etc.) - can take into account, standardise diagnosis / recording, use population-wide surveys and more objective measures

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is seen when measuring over time? (two effects)

A

Period effect - related to a recent time period / cause

Cohort effect - related to the time of birth / early life cause

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the ecological fallacy?

A

Assumes that exposure and outcome are related in individuals (but it’s based on population data), may be CONFOUNDING factors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are the potential pitfalls when studying disease over time?

A

PITFALLS - Changes in ASCERTAINMENT, changes in DIAGNOSIS, changes in RECORDING of information, changes in POPULATION STRUCTURE - overcome with complementary study relating potential disease CAUSE to the disease outcome in individuals

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Be able to interpret geographic and temporal variations in disease, distinguishing between artefactual and real differences

A

Be careful of confounding factors
Further research required to overcome confounding and identify individual / causal relevance in some studies
Possibly use criteria for causality?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Appreciate the potential value of variations in disease risk to our understanding of disease causes and treatment patterns

A

May be able to identify causes and protective features of particular areas / cultures
Understand how changes at a particular moment may have an effect on that birth cohort OR on everyone living during that period

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Describe the importance of high blood pressure and its health consequences

A

Why is high blood pressure an important problem? Strongly associated with higher risks of CVD, CHD and stroke AND risk associated is very common in general population (BOTH SERIOUS AND COMMON)

(And how do we know high BP is bad - cohort studies looking at relationship between high BP and developing disease / dying)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Outline the main risk factors and causes of high blood pressure

A

Secondary - specific medical cause (about 1% of general population)
May be caused by coarctation of aorta, renal disease, adrenal diseases, pregnancy, drugs
Essential / primary - no specific medical cause
High BMI, alcohol intake, salt intake, low potassium, low fibre/high fat diet, physical inactivity, stress

Risk factors - older age, African-Caribbean, Family history, high BMI, high alcohol intake, NOT SEX/GENDER

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Explain the epidemiological basis underlying strategies for treatment of high blood pressure.

A

Greater BP reduction gives greater reduction in relative risk, due to common nature of high BP and negative implications, risk of CVD is still high even below treatment thresholds so we need….
POPULATION strategy - shift BP distribution with public policy intervention (treatment for high BP, treatment only for those with high CV risk, reduce salt in diet - manufacturing)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Be able to define the terms relative risk and attributable risk and appreciate their relevance to high blood pressure

A

RISK = PROBABILITY OF AN EVENT (# cases / # at risk)
RELATIVE - Ratio of risk - Risk in exposed group (high BP) / risk in unexosed group (normal BP)
ATTRIBUTABLE (excess risk) - Difference in risk - Risk in exposed - risk in unexposed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Understand basic concepts of causality, confounding and bias.

A

Causility - how strong are the releative risks?BUt more importantly for individuals, what is the attributable / excess risk? Attributable risk is those exposed - those note to indicate the extra risk

Confounding - other factors that may play a role at the same time (something associated with both exposure and outcome

Bias - a type of systematic error, either in the selection of cases/ controls, the collection of information, analysis

17
Q

Be able to distinguish between a causal factor and a risk factor (not determinants of causality)

A

Causal factor - A factor which increases risk occurring, exposure - determinants for causality
Risk factor - Any characeterstic which identifies a group at increased / decreased risk of disease now or later (not necessarily causal, independent or modifiable) - SUCH AS AGE OR SEX

18
Q

Summarize the types of study which can be used to investigate whether air travel is a cause of DVT/PE

A

Observational - CCS or cohort

Experimental - RCT

19
Q

Appreciate the design features of case control studies and the principles of analysis

A

Retrospective, looking back at links between exposure and outcome
PROS - Quick, cheap, good with uncommon, can look at several possible exposures
CONS - confounding factors, size of study, selection of cases / controls, study protocol, approach to analysis

20
Q

Appreciate the potential sources of bias in a case control study and ways to overcome them

A
  • when selecting cases, set standard definition and focus on more recent
  • when selecting controls, consider matching for age / sex
  • Assessment needs to be similar across groups (setting, instruments, informants, incentives)
  • Removing interviewers can help maintain blind hypothesis
  • When designing study, try to ensure same amount of exposure to confounding

INFORMATION BIAS AT INFORMATION GATHERING STAGE
SELECTION BIAS WHEN PICKING CASES/CONTROLS

21
Q

Consider strategies for prevention of air travel related DVT/PE.

A

Society / low risk strategy - exercise/walking on flights, hydration - improve structure of planes / flights to minimise risk??
High risk individual strategy - avoid flights, use of compression socks, aspirin in advance, LMWH?

22
Q

Appreciate the design features of cohort studies and the principles of analysis

A

Prospective study, follows groups of people with different exposure over time to see if they acquire the disease

Strengths:
Exposure is measured BEFORE disease develops (unlike CCS)
Can provid info on several parameters - relative risk, incidence and attributable risk
Can look at several exposures and several outcomes
NOT CHEAP OR QUICK

23
Q

Appreciate the main sources of bias in a cohort study and ways to overcome them

A

Confounding factors - identify and adjust for potential confounding factors
Size and statistical power
Selection bias - participants cannot have disease to begin with, if there is marked/selective loss to follow up
Information bias - Defining exposure, Need to ensure PARAMETERS, COLLECTION, ANALYSIS are all objective as possible

24
Q

Summarise the strengths of cohort studies compared with case control studies

A

Strengths:
Exposure is measured BEFORE disease develops (unlike CCS)
Can provid info on several parameters - relative risk, incidence and attributable risk
Can look at several exposures and several outcomes
NOT CHEAP OR QUICK

25
Q

Summarise the limitations of cohort studies compared with case control studies

A

Limitations:
Hypothesis, confounding factors
Size and statistical power of study
Selection of study population, to include a range of exposures
Baseline assessment of exposure and confounders, to identify (and potentially exclude) prevalent disease cases
Follow up methods
Approach to analysis

26
Q

Appreciate when an association is likely to be causal (Bradford Hill criteria)

A

SCATBIRDS
Strong (high releative risk)
Consistent (different studies, similar result)Analogy (parallels with other established disease models)(not the result of confounding)
Temporal sequence (exposure first, then outcome)

Biological plausibility (explained by biological mechanisms)
Independent
Reversible (remove cause, reduce risk)
Dose response (higher exposure, higher risk)
Specific (one exposure, one outcome)

27
Q

Appreciate the strength of evidence relating blood lipids and coronary heart disease

A

Strong

28
Q

List the main modifiable and non-modifiable risk factors that predispose a patient to coronary heart disease.

A

Modifiable - High BP, High cholesterol, smoker, overweight, T2DM
Non-modifiable - Age, sex, ethnicity, family history

29
Q

Describe the main design features of experimental studies including clinical trials (5)

A

Can test medical/surgical treatments, health promotion advice, dietary change, etc.

The study should be:
Controlled (placebo controlled)
Randomized
Double blind (or triple)
Large
Analysed by the 'Intention to treat' method
30
Q

Explain the value of randomized clinical trials in informing the management and prevention of diabetes.

A

xxx