ecological and cross sectional studies

Question 1

when we have an observer that assigned the exposure they will be — studies by which we will either have — allocation which will lead to —- trials or — trials
however if the observer didn’t assign we will have —- studies, when comparison groups are used we will have —- which included:
when comparison groups are not used we will have — study

Answer 1

• intervention study
• random allocation
• randomised trials
• non randomised trials
• observational studies
• analytical study which include: ecological , case control , cohort , and cross sectional studies
• descriptive study

Question 2

-is an observational study and is the simplest study design by which no manipulation of variables is known as —
- they collect info from — at — in time
- they offer a — of disease aka — at – in time
- these can be also referred to as —

Answer 2

cross sectional studies
individuals
one point
snapshot
aka outcomes
one point
prevalence studies

Question 3

• cross sectional studies can be either — or —
• these are designed to measure — of the — or —- aka — which is —-
• often multibe outcomes are assessed as —- and —-
• consider several characteristics at — such as —-
• a —- of the outcomes is required
• it can be also used to examine —- between —- and — which is —-

Answer 3

descriptive or analytic
pervelance
outcomes or disease aka prevalence study
which is a descriptive
diabetes and hypertension
at once as income gender age etc
clear definition
examine association
exposure and outcomes
analytic

Question 4

population of interest —> representative sample —> measure outcomes with can lead to number which outcomes which can be calculate —- or it can lead to number without outcomes

Answer 4

calculate prevalence of the outcome which measure the outcome of disease frequency

Question 5

sources of data can be:
1- routine data sources which include :
2- population censuses
3- —- as SLAN 2007 by which the allow you to collect the — that you want by using —- methods , Often easier to examine associations between exposure and outcomes using —-

Answer 5

1- includes :
- death registration
- disease registers as : national cancer registry Ireland
- general practinorier data as ., Clinical Practice Research Datalink in the UK
- survey
- collection info
- rigorous methods
- survey data

Question 6

population of interest —> representative sample —> measure exposure and outcomes which can leads to : number with outcomes and numbers without outcomes among those exposed by which we can calculate —–
or it can lead to number with and numbers with our among those unexposed which we can calculate

Answer 6

• Calculate the prevalence of the outcome in the exposed
  -Calculate the prevalence of the outcome in the unexposed

Question 7

-Prevalence ratio or odds ratio and this is one way of — the — of the association between exposure and outcome
- prevalence ratio is the prevalence of the disease among — divided by the prevalence of the disease in the—
- PR of — reflects no association between the exposure and outcome, a PR of — indicates a positive association and of — indicates a negative association

Answer 7

• quantify the size
• exposed
• unexposed
• 1
• more than 1
• less than one
  ( check the PR formula in slide 13 so important ) + slide 15,17,18 for example SO IMPORTANT!

Question 8

-We use — when we have categorical variable (yes/no)

• If the exposure and outcome are measured on a numeric scale (e.g. blood pressure, age, BMI), we can summarise the strength of the linear relationship between them using a —
  which is donated by — and its used in — studies

Answer 8

• pervelence ratio
• correlation coefficient ( We typically use the Pearson correlation coefficient) and is donated by r
• used in cross sectional

Question 9

-Correlation coefficient always lies between — and —-
- the correlation coefficient of 0 means no —- between the 2 variables
- +1 refers to a —- relationship (as the value of one variable gets bigger the value of the other gets bigger)
- -1 refers to a —- relationship (as the value of one variable gets bigger the value of the other gets smaller)

Answer 9

-1 and +1
no linear association
perfect positive relationship
perfect negative relationship
( check the graphs in slide 20 so important LIKE ACC SO IMPORTANT )

Question 10

—– is the situation where an apparent relationship between an exposure and an outcome is due, in whole or in part, to a third factor meaning that the association of that exposure and outcome could really be due in whole or in part to other factors often referred to as a —-
which means the effects of the exposure under study on a given outcome are mixed in with the effects of an additional factor (or set of factors)
- example: Odds ratio of 2.88 suggests that boys who don’t eat before school are more likely to have obesity than those who do eat before school

Answer 10

• cofounding (meaning: is an unmeasured third variable that influences both the supposed cause and the supposed effect.)
  mixing of the effect

Question 11

criteria for a confounder - For a factor to confound the association between a potential cause (i.e., the exposure) and an outcome, it must:
1- be —- with the exposure
2- be — of the outcome
3- not to be in a — for the effect of exposure o the outcome

Answer 11

associated
predictor
casual pathway
( check slide 25,26 for the graph important )

Question 12

—- is a major issue in observational studies, which needs to be controlled for in the design and/or analysis.
- in order to control it we need to — and — it

Answer 12

confounding
identify and measure

Question 13

the advantages of the cross sectional study:
1- — study design
2- often — and relatively — to conduct
3- useful for — the health and heath care needs of population by measuring —- and —
4- can be used to —- association and therefor identify —- — relationship

Answer 13

simplest
quick
cheap
monitor
disease frequency
risk
examine
potentially casual relationship

Question 14

sources of errors in cross sectional studies: ( sources and the types of error basically )
1- source: section of participants as sampling is due to :
2-source: measurement in instruments as self-administered questionnaire, monitor, interview are due to :
3-source: measurement in observer is due to:

Answer 14

1- type:sampling error and selection bias
2- type:inaccuracy ( poor validity ) and poor reliability
3- type:between observers and within observers

Question 15

In cross-sectional studies, particularly need:
- a —- in cross-sectional studies in order to — sampling error and get a—- estimate of the prevalence
- a sample that is representative of the — in order to make inferences about the population based on the sample

Answer 15

large sample
reduce
precise
general population

Question 16

• Determining if an association is —- is very important in interpreting research evidence
• — and —- of evidence influences our ability to judge if an association is causal
• Before examining if an association is causal, should be quite confident the association is not due to —- , — or —-

Answer 16

casual
type and extent
random error bias or confounding

Question 17

• To be a causal association, the exposure must occur — the outcome
• As the exposure and outcome are measured at the —- , it’s not clear which occurred first
• If the exposure —- over time, not possible to establish causality from a cross-sectional study
• Depends on the nature of the —
• If the exposure does not change over time then a causal association is more – , even from a cross-sectional study

Answer 17

before
same time
change
expose
plausible

Question 18

limitation of cross sectional:
1- prone to — as —
2- Can be — and — to get a large enough sample to provide — results
3-Results may be due to —
4- Difficult to establish — exposure and outcome measured at the same time

Answer 18

bias as selecetion bias
costly
difficult
precise
confounding
causality

Question 19

• In most epidemiological studies we collect data on individuals and compare them in terms of exposure and outcome. However, individual data is not always available and this is why we conduct —– as its a study examines associations between units of grouped (or aggregated) data, such as electoral wards, regions or whole countries.

Answer 19

ecological studies

Question 20

-Unique feature of an ecological study is that data are for the — (e.g. a —- ), in contrast to cross-sectional studies, case-control studies and cohort studies, where we use data on exposures and outcomes for —-
- Meaning that we can only analyse aggregate data which we sometimes cannot really link back to a specific person

Answer 20

group
country
every individual
check slide 40 for example

Question 21

in ecological studies:
1- Collect data on an — for the group and compare to another group
Examples:
Pollution level in an area
Per capita salt intake
2- Collect data on an — for the group and compare to another group
Examples:
Rate of colon cancer in a country
Mortality rate – all cause or disease specific - in a country
3- . Examine the association between the —- and —- , e.g. using a scatterplot, statistical analysis

Answer 21

exposure
outcome
exposure and outcome

Question 22

advantages ecological study:
1- relatively — and —-
2- useful initial stage in the —– of —– and —-
3- identifies associations that are worthy of —– at the —- level

Answer 22

cheap and quick
epidemiological investigation of association and causation
further investigation
individual level

Question 23

limitation for the ecological study:
-Aggregate measures tell you nothing for certain about —
-An —– can occur when we ascribe to members of a group one or more characteristics they do not possess as individuals.
—> The consequence of an ecological fallacy is the association seen between aggregated data does not exist at the —-
—-> If the association is seen at the ecological level but not at the individual level, it cannot be —-
( example: Consider the Nobel prize winners study! although there is an association between chocolate consumption and Nobel prize winners we have no way of knowing the consumption of chocolate of the actual Nobel prize winners )

Answer 23

individual
ecological fallacy ( meaning: a logical error that occurs when the characteristics of a group are attributed to an individual)
individual level
causal