Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Poplhlth111 > Module 1 > Flashcards

Module 1 Flashcards

1
Q

Epidemiology definition

A

the study of dis-ease frequency in populations. A study’s starting point is always a population

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

risk equation

A

Occurrence ÷ Population ÷ years

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

Population/group

A

a group of people who share one or more common features. e.g country, ethnicity, gender/sex, diabetes.

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

Dis-ease

A

the absence of death, disease, disability, or the inability to do what matters most to you

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

why do you age standardise

A

ensures correct denominator for comparison
older people die more often
otherwise may cause confounding

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

PECOT

A

population
exposure group
comparison group
Occurance/outcome
time

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

Incidence

A

occurrence over time. data is counted forward, always categorical. counts events

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

Prevalence

A

occurrence at one point in time. Time is 1. can be categorical or numerical. Occurrence is counted at the point of allocation. can be counted backward. typically state

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

Categorical measure

A

yes or no, black and white outcomes

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

numerical measure

A

a number that you make into categories based on ranges or you do average for two different groups

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

Incidence strengths

A
  • clean measure that is only determined by risk in population
  • includes events, population, and time
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Incidence cons

A
  • can cost more
  • can take more time
  • can be difficult t measure at times
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Prevalence strengths

A

relatively easy to measure

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

prevalence cons

A
  • only includes events ad population, less info because less time
  • dirty measure that is determined by incidence, cure rate, and death rate
  • you lose data from from cures and deaths.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Unblinded study

A

researchers and participants both know who is which group

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

single blinded study

A

researchers but not participants know who is which group

17
Q

double blind study

A

neither researchers or participants know who is in which study

18
Q

Risk difference (RD)

A

EGO-CGO
when EGO=CGO, RD=0
units

19
Q

Relative risk (RR)

A

EGO÷CGO
when EGO = CGO, RR=1
no units but is generalisable

20
Q

Relative risk reduction (RRR)

A

when RR is less than 1, it can be expressed as relative risk reduction RRR. e.g 67 is 33 less than 100 so RRR = 33%

21
Q

Relative risk Increase (RRI)

A

When RR is more than one, it can be expressed as relative risk increase RRI, if RR=1.33, 133 is 33 more than 100 so RRI =33%

22
Q

Should you make decisions based on RR or RD

A

Decision should be based on RD, not RR. find the starting control exposure, then use relative risk to find what they would be under exposure, then calculate RD to make decision.

23
Q

Confounding

A

when exposure (eg high alcohol) is mixed with another factor (eg solvent use) that is also associated with the outcome (eg deaths)

24
Q

RAMBOMAN

A
  • Recruitment
  • Allocation ± adjustment in analyses
  • Maintenance
  • Blind or Objective Measurement of exposures and outcomes
  • ANalysis
25
Q

Regression to the mean

A

repeating measurements of studies with extreme results, multiple times, usually gives less extreme results

26
Q

Random sampling error

A

becasue identically designed and implemented studies will never have identical participants with identical charecteristics nor identical results.
- smaller sample = greater chance that sample is different to whole population

27
Q

how to reduce random sampling error

A

a larger sample or study can reduce random sampling error

28
Q

Random measurement error

A

measuringt he exact truth (especially in bio) is not possible
identical measurements of exposures and outcomes in the same or similar people can change moment to moment - especially if measurement is done by a human operator.

29
Q

95% confidence interval

A

means there is a 95% chance that the true value in a population lies within the 95% confidence interval. Assuming no ramboman and that it relates to the whole population
- a measure of the range of random error in our estimates of EGO, CGO, RR, RD in the whole population

30
Q

when is difference betwen EGO and CGO significant

A

when CI doesn’t overlap

31
Q

when is RD significant

A

when CI doesn’t overlap with no-effect line (RD=0)

32
Q

when is RR significant

A

when CI doesn’t overlap with 1

33
Q

Meta-analysis

A

when you can’t do a bigger study, do a systematic review, combine studies, and get combined RR (diamond shaped)

34
Q

individual participant study

A

every person in P is identified and individually allocated to EG or CG

35
Q

RCT - Randomised Control Trial

A

allocate randomly - prevalence or incidence:
- difficult to recruit representative P
- minimises confounding
- can be unethical with harmful exposures
- usually expensive or too small
- maintenance errors common
experimental: test intervention or confirm risk

36
Q

Cohort Study

A

allocate by measurement. done over time - incidence or prevalence
- easier to recruit representatively
- confounding can be common
- ethical to study harmful exposures
- maintenance error common for long term
- less expensive and can be large
non-experimental: investigate risk and exposures in groups of individuals

37
Q

Cross-sectional

A

allocate by measurement and done at a time. allocation and measurement, prevalence done at the same time.
- easy to recruit representative populations
- confounding can be common
- reverse causality
- no maintenance error
- less expensive and frequently large
- only prevalence
non-experimental: investigate risk and exposure prevalence

38
Q

reverse causality

A

When outcome causes person not to be exposed or that outcome causes exposure (e.g poor lung capacity as an outcome means people can do exercise as the exposure so it appears people who don’t do exercise will have poorer lung capacity)

39
Q

ecological study

A

P include multiple people or groups allocated to EG or CG
- large so low random error
- confounding common
- efficient for rare outcomes
- useful with majority of some populations are exposed but others are not
- measurement error common
experimental or not: investiagte risk and incidence/prevalence in different groups or populations

Poplhlth111 (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions