post quiz 1 Flashcards
1
Q
case report
A
- careful individual-level observations by health care providers of what they see during clinical practice
- provides a comprehensive and detailed description of a particular clinical phenomenon in a single patient
- the simplest study designs
- case report can be the basis by which other physicians identify and report similar cases from their practice
- when there are many similar case reports describing more than one patient with similar problems = case series
2
Q
case series
A
- same general principles of a case report except the focus is on more than one patient
- there are no comparison groups in either case report or series, so we can’t calculate measures of association
- but case reports and case series can be useful for public health in general
- ex: brief report describing 5 rare cases of pneumonia among 5 men in LA: first reported cases of HIV/AIDS
3
Q
cross-sectional
A
- great for initially investigating associations between a specific exposure and a disease of interest
- get data at one given point in time
- capturing exposure and outcome at same time
- any cases are prevalent because (a) they existed at time of study but we do not know their duration or temporality of exposure
- prevalence data
4
Q
cross sectional - sample
A
- not composed entirely of those at risk
- some of sample will have disease at baseline
- since no incidence, cannot calculate RR
- instead, calculate OR
5
Q
temporal bias
A
- when we assume the sequence of events between exposure and outcome
6
Q
survival/selection bias
A
- prevalence-incidence bias
- when exposure is related to duration of disease
- identifying only prevalent cases excludes those who died sooner after the disease developed but before study was carried out
7
Q
pros cross sectional
A
- allow for hypothese generation
- cheaper
- faster than other study designs
8
Q
goals of surveillance
A
- continued watchfulness over distribution and trends of incidence through systematic collection, cosolidation, and evaluation of morbity
- assess public health status
- evaluate programs
- define public health priorities
- stimulate research
9
Q
cohort study
A
- begins w target population
- contains both diseased and non-diseased
- can’t get the entire pop
- draw a sample
- sample must start w those at risk of outcome (non-diseased)
- assess exposure status of sample
- determine if they have exposure or not
- everyone is still at risk of developing disease
- follow participants for some length of time and observe incident cases
10
Q
cohort cons
A
- can take a long time
- selection bias: non participation/nonresponse, people who refuse to join or continue differ in characteristics from who is enrolled
- if people w disease are selectively lost to follow-up, incidene rates will be difficult to interpret
- information biases: if quality/extent of information obtained is diff for exposed persons than for unexposed
11
Q
cohort study example
A
- framingham heart study
- incident CVD cases identified every 2 years and by daily surveillance of hospitalizatios at the only hospital in framingham
12
Q
randomized control trials
A
- in observational, we observe exposure/outcome relationships
- experimental studies manipulate exposure to measure impact on outcomes
13
Q
difference between RCT and cohort
A
- both longitudinal, gather incidence data, and yield risk/rate ratios
- in RCT we assign expore while in cohort studies we observe those who are exposed/unexposed of their own choosing
14
Q
why randomize
A
- any subjective selection biases of researcher no longer have impact
- increase likelihood that groups will be comparable to each other in factors such as sex, age, race, and severity of disease
15
Q
cross over
A
- original groups: intentional to treat analysis
- final groups removing cross over: per protocol analysis
- final groups with crossover: as treated analysis
