Epidemiology 2 Flashcards

(65 cards)

1
Q

What is a primary prevention strategy?

A

The aim of a primary prevention strategy is to prevent the disease from occuring upstream at the first stageb but reducing the exposure or risk factors.

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

Example of primary prevention public health strategies?

A

Health promotion for healthy lifesyle e.g. healthy diet and stop smoking etc.
Fluoridation of drinking water
Childhood immunisation programmes
population level legistlation e.g. illegal to smoke in public places and plain packaging.

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

What is secondary prevention?

A

To aim to detect the disease early in order to decrease the progression/severity of the disease or stop the disease recurring

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

Examples of secondary prevention strategies?

A

Screening for diseases e.g. smear tests for cervical cancer, or mammograms to see if breast cancer.
Treatment with a drug to reduce further risk e.g. aspirin after heart attack stop another, give a pacemaker.

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

What is Tertiary prevention?

A

To minimise the impact after the event has happened by minimising disability and preventing complications.

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

Examples of tertiary prevention strategies?

A

Preventing death- e.g. in hopsital after a heart attack.
rehabilitation after a stoke.
A&E services after a crash

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

Car crash primary secondary and tertiary prevention strategies?

A

Primary: to stop the event e.g. speed limits, laws to ban drink driving, going on phones etc.
Secondary: Not stop event but reduce severity e.g. carseats for children, airbags, car crumple zones
Tertiary: stop death e.g. At A&E treatment, bandage wounds etc

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

How ar ethe primary, tertiary and secondary prevention strategies split further into 2? (examples)

A

Individual or population strategy
E.g. population strategies would be laws, health promotion activites, initiatives, screening programmes.

Individual strategies specific to individual e.g. identiying those high risk of diabetes- give healthy eating couselling, specific tests, drugs given to prevent further complications/death.

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

What are high risk approaches to prevention?

A

Involves identifying those at high risk and then tailoring the action for them- reduce exposure or provide a protection to.

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

Advantages of high risk approaches to prevention? (3)

A
  1. Focuses only on those at high risk, so has a high rate of return.
  2. Method chosen for the individual so likely to be appropriate and motivate them more.
  3. Physician motivated to help individual.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the prevention paradox?

A

A population prevention strategy which overall brings much benefit to the population (e.g. saves 300 lives a year), but if this is per 100,000, then 99,700 people will have no benefit.

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

What is risk compensation?

A

When the prevention method is counterrracted by other actions e.g. cyclist wearing helmet so cars see them as safer and may get closer. Or person on statins then eating more Cholesterol fatty food.

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

Advantages of population strategies? (3)

A
  1. Large potential for the population- doesn’t miss anyone at say medium risk.
  2. Behaviourally appropriate
  3. Doesnt involve screening for high risk people to target as just target everyone.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

disadvantages of population strategies? (3)

A
  1. Small benefit to most people- cost benefit analysis may be low.
  2. Poor motivation of individual or physician
  3. Can make health inequality worse- say healthy food and gym- more expensive and educated may see value more.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

disadvantages of high risk approaches to prevention? (4)

A
  1. Involves screening people- costs, recruiting people?
  2. Limited effect as it misses most of the popualtion. E.g. medium risk may hold most of the population so save the most lives, even if some people are at higher risk.
  3. Effects may be temporary- individual initially motivated to change but revert back.
  4. Now less public health more medicine.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

WHat is a screening test?

A

A process which sorts apparently well people into those who probably have the disease (or precursors/susceptable to) from those who probably dont.
Different to a diagnostic test which is a simple test that says yes or no.

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

Is a screening test primary, secondary or tertiary prevention?

A

Either primary or secondary.
E.g. Primary for breast cancer- genetic test have associated BRCA gene. Will be high risk- show susceptability.
E.g. Secondary for breast cancer- mammography- when no symptoms catch it early so can treat and stop progression.

Tertiary- will have symptoms so likely know have the disease so no point screening.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

How can we measure the efficiency of screening? (5)

A

Sensitivity, specificity, Positive predictive value, Negative predictive value, accuracy.

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

What is the sensitivity of a test?

A

The probability of someone with the disease having a positive test result. Left column.
True positive/ all with disease.

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

What is the specificity of a test?

A

The probability of someone without the disease having a negative test result. Right column.
True negative/ all those without disease.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

How are specificity and sensitivity usually related?

A

As the sensitivity increases, the specificity decreases. E.g. if the cut off for diagnosis is brought down, even more people with the disease will be identified, but also more wwithout the disease will get a positive result. So a balance is needed- if plot 1-spec vs sensitivity on a graph- the most top Left point- or add the two values and pick the highest number.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is a positive predictive value?

A

The probability that an individual with a positive test will have the disease.
Top column- True positive/all test positive

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

WHat is a negative predictive value?

A

The probability that an individual with a negative test will not have the disease. Bottom column.
True negative/ all test negative

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Accuracy of a screening test measured?

A

correct diagnosis/total screened.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Which tests for screening vary with the prevalance of the disease?
PPV and NPV: If rare disease the PPV will decrease, and the NPV will increase. E.g. too many people will be told may have the disease when don't (false positives) but false negatives will decrease. The screening test works less well- better for quite common diseases.
26
Which tests for screening dont vary with prevalance?
Sensitivity and specificity- these are tests for the screening test not disease.
27
When should a screening test be used?
1. When the condition is a prevalent health problem (not rare) 2. history of the condition known and has a detectable early stage. 3. Should be a treatment available for those at high risk and diagnostic test. 4. Appropriate health service provisions for the screening e.g. funding, workload. 5. Test suitable and acceptable. 6. Agreed policy on who to treat and repeat test intervals. 7. Costs benefit analysis weighed up.
28
What is the best way to test a screening test?
RCT- individual or cluster
29
Biases associated with screening test evaluation?
1. Selection bias 2. Lead-time bias 3. Length-time bias
30
WHy is selection bias a problem with screening tests evaluation?
There is a selection bias in who goes to the screening- often healthy organised people who have a higher socioeconomic background- more educated and empowered to go.
31
WHy is lead-time bias a problem with screening tests evaluation?
Lead-time is the length of time between detection from screening/ its clinical presentation/diagnosis and death. Bias comes from the fact that the diagnosis is earlier so it looks like the patients are living longer, when just started measuring earlier E.g. Found the disease a year early, so instead of patient living 3 years after diagnosis, seen to live 4 years- gained an extra year because of screening and finding early! No, just found the disease a year earlier
32
WHy is length-time bias a problem with screening tests evaluation?
This bias is caused by an overestimation of survival due to an excess of cases found that are asymptomatic and slow progressing e.g. cases only found in clinic after symptoms show when are likely agressive and fast progressing, whereas the screening catches the slow progression cases that will likely live longer anyway.
33
Screening programme types?
"mass screening" of whole pop. "opportunistic" screen while at the GP, but misses hard to reach people, but could reduce screening only to suspected higher risk patients. "occupational" e.g. Army Medical Commercial" e.g. DNA tests can buy communicable disease skin tests e.g. Heaf for TB exposure
34
Incidence definition?
Number of new cases of a disease in a timeframe, e.g. 8 per 1000people a year.
35
Prevalance defintion?
A measure of the proportion of the population that has a condition at a given time. Assesses the burden of disease.
36
Why is prevalence useful?
Useful to see the burden of disease, so planning for interventions- resource allocation etc.
37
Prevalance two types?
The number of cases at a given point in time (point prevalence) or period prevalance is the proportion of the population that has the condition at some time in a given time period and including those already have the condition.
38
Why is incidence useful?
Useful for exploring risk factors that cause the disease- it is unaffected by changes in treatments etc.
39
3 types of calulations for incidence?
Risk: new cases of disease/ population total Odds: New cases of disease/those in pop without the disease. Rate: New cases of disease/ person-time at risk
40
What factors lead to diseased state?
Addition of component factors to create sufficient cause. Some component factors are susceptability, infectious agent and vectors. Some of these may be essential (e.g. the virus), but may need other components to be sufficient cause.
41
Secondary attack rate calculation?
Number of new cases among contacts in given time period/total number of individuals at risk e.g. (contacts to the primary case) at the start of time period e.g. At a school 8 children developed chicken pox- they had a total of 15 siblings. 5 of them got the disease within a week. 5/15= 33% secondary attack rate
42
Relative measures of association are what?
They measure the strength of an assocition i.e compare the risk of getting disease with and without the exposure to see the ratio. exposed risk/unexposed risk.
43
What is odds ratio of exposure?
Used for case controls- opposite way round e.g. instead of.. risk of disease exposed/ risk unexposed it is exposed/unexposed for disease divided by exposed/unexposed for not diseased.
44
How do the ratios vary e.g. odds ratio vs risk ratio?
Very similar for rare outcomes, hence why odds can be used for rare, but can vary considerably for common. E.g. for rare the denominator becomes v similar in odds and risk
45
What is the downside to the risk ratios?
they dont give an indication of the real impact of the exposure e.g exposed/unexposed could be 6/1 vs 30/5 both RR=6 but A causes 7 deaths and B 35.
46
solution to the lack of indication of Risk Ratios on real impact?
Absolute Risk difference (stats), Attributable Risk (for epi)
47
What is absolute risk difference? (STATS)
I Risk in Exposed- Risk in unexposed I | Used to find NNT 1/ARD
48
What is attributable Risk?
The excess incidence that the exposure causes above the unexposed (attributable risk) i. e. Incidence in Exposed-Incidence in unexposed. e. g. 6-1=5 vs 35-5=30, can now see that B has a greater impact (x6)
49
What is attributable risk fraction?
Incidence exposed- unexposed/ exposed. What proportion of the incidence in the exposed is attributable to the exposure e.g. 35-5/35, 30/35= 86%
50
What to do about attibutable risk fraction if the exposure is a protective factor?
swap exposed and unexposed. incidence unexposed- exposed/ incidence unexposed. So refers to the proportion of the outcome that can be prevented by a protective factor e.g. 10-3/10 =70% can be saved (as only 3 got disease instead of 10). Preventable fraction=1-relative risk
51
What is population attributable risk?
the incidence of outcome in the population that can be attributed to the risk factor. Incidence in the pop-incidence in the unexposed. E.g. 94 cases in population, 14 cases are due to exposure. 94-4=80.
52
What is population attributable risk?
The percentage of the outcomes in the population that can be attributed to the exposure Incidence in pop- incidence in unexposed/ incidence in pop. e.g. 94-14/94 =80/94=85%
53
"within the population how many incidences are attributable to the exposure?"
Population attributable risk (PAR)
54
" what proportion of the disease in the exposed group is attributed to the exposure?"
Attributable risk fraction (ARF)
55
"What us the excess incidence due to the exposure?"
Attributable risk (AR)
56
"What is the proportion of the incidence in the population that is attributable to the exposure?"
Population Attributable risk fraction (PARF)
57
How can the confounding factor of age be dealt with in the analysis?
Age standardisation
58
Direct vs indirect age standardisation?
Direct uses your death rates on a standard population Distribution. (DD) Indirect uses standard death rates on your population distribution giving a ratIo (II)
59
What is death rate?
deaths/total population (eg for age standardisation in each given age group)
60
How do you do direct age standardisation?
Direct uses your death rates on a standard population Distribution. (DD) So take the death rate per age category and multiply this by the standard population size for each age category to get 'expected deaths'. =DIRECT STANDARDISED RATE (DSR) Add these and compare the actual deaths to the expected- explain the trend? Then standardised so can compare the expected deaths to other areas.
61
Advantages of direct standardisation? WHy do indirect?
It is a more statistically stable method, only do indirect if do not have access to deaths rates per age cateogy, or if th information is unstable (i.e. v small numbers or unusual event).
62
What standard population should be used for direct standardisation?
It doesnt matter as long as its consistant for all of the places so the DSRs are comparable. Could be UK, European or World etc
63
How do you do indirect standardisation?
Indirect uses standard death rates on your population distribution. Take a set of standard death rates for each age category and multiply this by the number of people in each age categories to get the 'expected deaths' Add up these to get the total expected deaths. STANDARDISED MORTALITY RATIO (SMR)= observed/expected.
64
SMR what? Calculation example?
Standardised mortality ratio for indirect age standardisation. SMR=Observed/expected e.g. Industrial city: 3800 observed/2800 expected =1.36, so saw 36% more deaths than expected. Whereas Seaside: 410 observed/480 expected= 0.85, so 15% less deaths than expect. (e.g. if a young population in a city may make the deaths look less than a seaside town where there are lots more elderly people, but if standardise can see seaside is more healthy.)
65
What are the two different outputs of direct vs indirect age standardisation?
``` Direct= Direct standardised Rate (DSR) for each area which can then compare rates. Indirect= Standardised Mortality Ratio (SMR) can see for each area the ratio of how much lower/higher than would expect. Can then compared these between areas too. ```