I&P 2 > Epidemiology > Flashcards

Epidemiology Flashcards

(82 cards)

Study These Flashcards
1
Q

What is epidemiology?

A

Epidemiology- The branch of medicine which deals with the incidence, distribution, and possible control of diseases and other factors relating to health.

Epidemiology- study of the distribution (trends and risks) and determinants (identifying risk factor for COD) of health-related states or events in specified populations and the application (control of this problem) of this study to control the health problem

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

What is population thinking?

A

Population thinking: We cannot predict what will happen to an individual, but can predict the outcome of a group

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

What is group comparison?

A

Group comparison: To observe effects of treatment or exposure, compare those treated/exposed with those who were not

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

Why do we need epidemiology

A
  • Many potential causes of disease, for practical or ethical reasons, cannot be studied experimentally
  • Epidemiology allows us to monitor and explain changes in trends of disease within a population
  • Epidemiology allows us to quantify the potential impact of efforts to maintain and improve population health
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Reasons why the shift from infectious to chronic disease in the early 20th century posed a challenge to the discipline of epidemiology?

A
  • Chronic disease is not pathogenic, identifying the cause is harder
  • Chronic disease tend to be multifactorial so cause must be disentangled
  • Long latency means that temporal directions will be harder to establish
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the two ways to prevent disease? Why is one way favoured?

A

Geoffrey Rose- contributes to understanding how to prevent disease in the population. E.g. target prevention at high risk. Other approach is to target prevention at the wider group- more people at lower risk than higher risk of the disease

Higher BP, higher risk of CHD. Given the risk and size of the group is the amount we expect to die. Can see lower risk you save more.

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

What is the prevention paradox?

A

“A preventive measure that brings large benefits to the community offers little to each participating individual.”

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

What is the definition of incidence?

A

Incidence is the rate at which people contract a disease in the population. It counts the number of new cases over a given period of time, and divides it by the number of people who were at risk of contracting the disease over that time period. It is usually expressed as n cases per N population per t time.

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

Prevalence defintion?

A

Prevalence is a measure of how widespread a disease is at a given point in time, or over a given period of time. It counts the number of existing cases and divides it by the total number of people in the population. It is usually expressed as a point or period percentage, i.e., the number of cases per 100 population.

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

Difference between incidence and prevalence?

A

Prevalence = Incidence x Disease duration

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

Definition of absolute risk?

A

The absolute risk of a disease is the number of occurrences of the disease divided by the number of people at risk. We can therefore only calculate the absolute risk when the size of the population that produced the cases is known (i.e., in a population-based study).

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

What is relative risk?

A

The relative risk is a ratio of two absolute risks - the first in a group who were exposed or treated, and the second in another group who remained unexposed or untreated - and can therefore be used to quantify the effect of an exposure or treatment on a disease outcome.

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

What is risk difference?

A

A risk difference is the difference in absolute risk among those who were exposed or treated compared with those who remained unexposed or untreated. It is an estimate of the additional number of cases we may expect to arise due to a given exposure, or the number of cases we may expect to be able to prevent with treatment. As such, it can tell us something about the potential impact of a given exposure or treatment.

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

What is the equation for measure of occurence?

A

Measure of occurrence = numerator / denominator

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

What does measure of impact mean?

A

MEASURE OF IMPACT–> risk difference

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

What does measure of association mean?

A

MEASURES OF ASSOCIATION–> Odds ratio and relative risk

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is a numerator in terms of disease occurence?

A
  • Numerator: defines and counts the cases
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is the denominator for disease occurence?

A

Denominator: population that produced the cases and / or the time over which the cases arose

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What does occurence mean?

A

OCCURRENCE –> incidence and prevalence

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What can affect the numerator when couting disease?

A

Problems can arise when the conditions are difficult to diagnose

  • The classification that was used for the diagnosis –> take into account difference in diagnostic criteria / a change in classification
  • Take into account pathways to care when estimating disease prevalence based on primary or secondary data
  • Take into account stigmas-> will people always tell the truth? Will clinicians be less willing to diagnose a condition?
  • Ethnic differences
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

How can problems affect counting the denominator correctly?

A
  • Selective undercounting of certain groups in a population –> certain groups may be geographically mobile and so less likely to register an address (e.g., travellers)
  • Political reasons –> people may be less inclined
  • Deaths may not be recorded equally well in certain communities such as indigenous people
  • Denominator should not include anyone who cannot become a case –> e.g., if counting uterine cancer, it cannot include women who have had their uterus removed
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is the equation for prevalence?

A

Prevalence = incidence x disease duration

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Suppose that 2,000 people are screened for tuberculosis using chest x-rays: half of those screened live in town A and have higher incomes, the other half live in town B and have lower incomes. The x-ray screens are positive for 100 out of 1,000 people in town A, and for 60 out of 1,000 people in town B (Table 3).

Consider whether we can conclude that the risk of tuberculosis is higher in town A than in town B.

A

We cannot.

  • The reason is that the x-ray results give us a point prevalence (from first two columns).
  • They tell us about the number of cases of TB in both towns at the time of screening.
  • Because we do not know when the disease started for each individual (i.e., whether they are recent or chronic cases) we can’t conclude anything about the risk of TB in either town.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is the relative risk?

A

The relative risk is a ratio of two absolute risks.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
How do you calculate the relative risk?
To calculate it, we divide the incidence of disease in the exposed group by the incidence of disease in the unexposed group
26
What does the relative risk tell us?
This tells us something about the strength of association between the risk factor and the disease outcome.
27
What is risk difference?
Risk difference: difference between two absolute risks
28
How do you calculate the risk difference? What does it tell us?
To calculate, (incidence in unexposed group) – (incidence in exposed group) This tells us about excess risk of disease due to exposure
29
How do you calculate relative risk?
Calculated from: Using the data shown in the figure, we can calculate the relative risk for stroke associated with smoking by dividing the incidence among smokers (20 per 1,000 population) by the incidence among non-smokers (10 per 1,000 population). The relative risk of stroke associated with smoking is therefore (20/10=) 2.00) Relative risk of diabetes with smoking --\> 1.33
30
How do you calculate risk difference if the incidence of stroke smoking was 20 and non-smoking 10?
Risk difference for stoke associated with smoking: 20-10 = 10 per 100,000
31
Risk difference for stoke associated with smoking: 20-10 = 10 per 100,000 What does this mean?
What this means is that, if smoking were a cause of stroke and we knew how to eliminate it, we could potentially reduce the incidence of stroke by 10 per 1,000 among those who smoked.
32
Risk difference for stoke associated with smoking: 20-10 = 10 per 100,000. Relative risk of stroke with smoking --\> 2.00 Risk difference for diabetes associated with smoking = 20 per 100,000 Relative risk of diabetes with smoking --\> 1.33 What does this mean?
THEREFORE, RISK DIFFERENCE FOR DIABATES IS LARGER, EVEN THOUGH RELATIVE RISK OF SMALLER This is because Diabetes is more common than stroke Affects a larger absolute risk
33
What does risk difference tell us?
It tells us the POTENTIAL IMPACT of a risk factor in a population and the EXTENT BY WHICH DISEASE MIGHT BE PREVENTED if the factor were eliminated
34
What is attributable risk?
ATTIRBUTABLE RISK: gives us an indication of the proportion of risk in the population that is attributable to the exposure Because more women smoked than had high blood pressure, the attributable risk smoking in the population is greater, even though the relative risk of preterm birth with smoking is comparably smaller. From the point of view of the public health practitioner, a smoking cessation program is likely to produce the greatest benefit.
35
Why if smoking has a higher attributable risk should we focus our attention on it more than other causes of preterm birth?
Even though smoking is not as strongly associated with preterm birth as some of the other risk factors in this population, there are comparably many women who smoke. A smoking cessation program, despite having a smaller impact per individual, still has the greatest potential impact in terms of preventing preterm delivery in the population.
36
How can we limit cofounding bias via a control group?
Control group that resembles the treatment group in all ways except for treatment status.
37
How can we limit cofounding bias via exchangeability?
The concept of exchangeability can be used to evaluate whether the treatment and control groups are sufficiently similar for a estimating a difference in disease outcome due to exposure.
38
How can we limit cofounding bias via restriction?
These were: restriction (i.e., investigating the exposure-outcome association at a given level of a potential confounder)
39
How can we limit cofounding bias via stratification?
Stratification (i.e., calculating the exposure-outcome association at different levels of a potential confounder)
40
How can we limit cofounding bias via covariate adjustment?
Covariate-adjustment (i.e., using statistical models to adjust the exposure-outcome association for group differences in other characteristics).
41
How can we limit cofounding bias via conditioning?
Conditioning the comparison of disease outcomes between groups on other variables, we can minimise their potential confounding effects.
42
What is an interventional study?
Intervention study. It is one of the few exceptions in which epidemiology can resort to experimental methods: i.e., if contaminated water is a cause of cholera, then removing the cause should result in a decline in spread and mortality of the disease.
43
What does it mean a disease may be multifactorial?
Diseases are caused by different combinations of potential causes, and the same combination of potential causes may give rise to different diseases
44
What does necessary mean?
Necessary --\> without exposure, the disease develops
45
What does sufficency mean?
Sufficiency --\> in presence of exposure, the disease always develops
46
What does necessary but not sufficient mean?
1.Necessary but not sufficient Necessary --\> without exposure, the disease does not develop i.E.g., without having COVID, one cannot develop long-covid Not sufficient --\> not everyone who is exposed develops the disease i. Not everyone who has COVID develops long-covid ii. There is another factor here
47
What does sufficient but not necessary mean?
Not necessary --\> You can develop the disease without the exposure (aka, there are many ways to get the disease à it is multifactorial) i.E.g., radiation exposure is not the only thing which causes leukaemia you can get leukaemia from benzene exposure or even both Sufficient --\> in the presence of exposure, disease always develops i.Each factor is sufficient enough to cause the disease
48
What does neither necessary nor sufficent mean?
No single factor alone can cause the disease, and disease can develop from different combinations of factors Not necessary --\> You can develop the disease without the exposure Not sufficient --\> Not everyone who is exposed develops the disease E.g., smoking is neither necessary not sufficient of lung cancer - lung cancer can develop in people who do not smoke - not everyone who smokes develops lung cancer
49
The stronger associations the more likely to be causal.
The stronger associations the more likely to be causal.
50
How does consistency assess whether an association seen in observational data was likely to be causal?
Consistency- repeatedly observed by different persons, in different places, circumstances and times' is more likely to be causal
51
How does temporality assess whether an association seen in observational data was likely to be causal?
Temporality. For an exposure to be a cause of a disease, it must have preceded the disease. upward trend in smoking prevalence preceded a later rise in lung cancer deaths, so, at a population-level, the exposure preceded the outcome.
52
How does biological gradient assess whether an association seen in observational data was likely to be causal?
Temporality. For an exposure to be a cause of a disease, it must have preceded the disease. upward trend in smoking prevalence preceded a later rise in lung cancer deaths, so, at a population-level, the exposure preceded the outcome.
53
What is a biological gradient?
Biological gradient If risk of disease increases with increasing dose (exposure), association will be more casual. In the smoking-lung cancer debate, Hill noted that, among smokers, the risk of lung cancer increased linearly with the number of cigarette's smoked daily.
54
How does plausibility assess whether an association seen in observational data was likely to be causal?
Agent could cause disease based on knowledge
55
How does coherence assess whether an association seen in observational data was likely to be causal?
Coherence.If findings consistent with other data
56
How does experiment assess whether an association seen in observational data was likely to be causal?
Experiment . If experimental evidence is attainable, it provides stronger support for causality than the other viewpoints.
57
How does analogy assess whether an association seen in observational data was likely to be causal?
If current association with another exposure-disease relationship can be used to support causality. E.g.if smoking tobacco caused lung cancer, then we might expect by analogy that chewing tobacco (as was done historically in the United States) is associated with cancer of the mouth and throat.
58
A first approach that can be taken to evaluating whether a disease may have a genetic component is to assess what?
A first approach that can be taken to evaluating whether a disease may have a genetic component is to assess whether it is associated with a known genetic syndrome. For example, children with Down Syndrome are known to be at higher risk of leukemia
59
A second approach to identifying possible genetic components to causes of disease is by what?
A second approach to identifying possible genetic components to causes of disease is by studying the age at which symptoms typically begin
60
The reason that the genetic form tends to develop at an earlier age is what?
The reason that the genetic form tends to develop at an earlier age is that genetic susceptibility is present from birth, whilst in the non-genetic form of the disease the environmental causes will need time to build up.
61
Studies of twins- explain this grid
In cell a both twins have the disease; in cell b only twin 2 has the disease; in cell c only twin 1 has the disease; and in cell d neither of the twins have the disease.
62
If twins are concordant what does it mean?
If the twins are concordant, they either both have the disease (cell a) or both do not have the disease (cell d).
63
If twins are discordant what does it mean?
If they are discordant, only one of the twins will have the disease (cells b and c).
64
If we found that all the twin pairs we studied were concordant in terms of disease status (i.e., all pairs would fall into cells a and d), would this be consistent with genetic causation?
Yes: if the causes of the disease were entirely genetic and the twins are genetically identical, it follows that they should be concordant in terms of their disease status.
65
Would concordance be consistent with environmental causation?
Yes also: because identical twins will often be exposed to the same environment (i.e., reared by the same parents, shared a household, attended the same school, etc.) exposure to the same environmental factors could have resulted in concordant disease status. So, concordance in disease outcomes between identical twins doesn't really tell us much.
66
What then, if all of the identical twin pairs were discordant (i.e., all pairs would fall into cells b and c) in terms of their disease status? Is it consistent with genetic causation?
No: because the twins share the same genetic material they should have concordant disease outcomes if the etiology of the disease were entirely genetic. It therefore follows that there must have been a difference between the twins in environmental exposure that resulted in their discordant disease status.
67
However, the extent of discordance in identical twin pairs can tell us something about the likelihood that environmental factors play a role in disease causation. If discordance between the identical twins is low/ high what does that mean?
If discordance between the identical twins is low the disease is more likely to be genetic, if discordance in the identical twin pairs is high it is more likely that environmental causes are involved in the causation of the disease.
68
An alternative (and less restrictive) approach that can be taken in studies of twin pairs is to compare the extent of concordance between identical and fraternal twin pairs. If the origin of the disease was mostly genetic what would we expect a higher rate of concordance in?
An alternative (and less restrictive) approach that can be taken in studies of twin pairs is to compare the extent of concordance between identical and fraternal twin pairs. The rationale here is that, if the disease were mostly genetic in origin, we would expect a higher rate of concordance in disease status among identical twins (who share 100% of their genetic material) than among fraternal twins (who share only 50%).
69
What information do we need to understand public health problems?
* Demographic information- age or young- who does it affect, ethnicities- are some groups more at risk?, sex can cause problems * Socioeconomic data- wider determinants of health- unemployed, housing benefits, income support, free schools meals * Environmental data- air pollution (not available at smaller areas), drinking water quality, pests, crime statistics * Morbidity- limitation- only captures those that use healthcare services, censuses can capture information on health though * Life expectancy and mortality
70
What is demographic information?
Demographic information: for example, age profile (e.g., mostly young or elderly?) or ethnic composition (disease more or less common in different ethnic groups?). The health of a population will depend in part on these demographic characteristics. This is why rates are often age-standardized as we could not otherwise compare between populations.
71
What is socioeconomic data?
72
What is environmental data?
Environmental data: This may include air pollution, drinking water quality, noise pollution, pests, road traffic accidents, or crime statistics. Some of these data are more readily available than others. Air pollution for example, a hugely important environmental cause of health, is usually not available for smaller levels of geography where it would be the most useful.
73
What is morbidity data?
Morbidity: this could include infectious disease notifications, diagnoses made in primary or secondary care, hospital activity data, population screening test results, and health outcomes measured in large nationally representative surveys. The challenge with morbidity data is that, with the exception of population-based surveys or censuses, it can only reflect those using healthcare services. There may be many people in the population who are ill but do not seek healthcare, so service-use related data can only tell us so much.
74
What is life expectancy and mortality data?
Life expectancy and mortality: These are key indicators of population health and are useful for measuring change in population health over time, or differences in population health between locations. For example, the stagnation of life expectancy in the last decade in the UK whilst life-expectancy continued to increase elsewhere in Europe suggests that the UK population is less healthy than other European populations.
75
What does NHS information tell us about public health?
NHS- Primary and secondary care data Screening data Vaccination coverage
76
What does PHE information tell us?
PHE- Infectious disease notification Chronic disease surveillance Monitors health inequalities Environmental health hazards Vaccination coverage
77
What does ONS information tell us?
Office for national statistics: Census Demographic data Socioeconomic data Death certificates Birth registrations
78
What is validity tell us about data?
“Validity” or “appropriateness”- does it capture what you are interested in
79
What does accuracy tell us about data?
“Accuracy”- how precise is the information collected
80
What does completeness tell us about data?
“Completeness”- does cover information that interested in
81
What does timeliness tell us about data?
“Timeliness”- was collected recently enough to be meaningful
82
I&P 2
flashcards
Decks in class (5)
# Cards
Brainscape helps you reach your goals faster, through stronger study habits.
© 2026 Bold Learning Solutions. Terms and Conditions