Lecture 16 Flashcards
Study Design pt. II (22 cards)
cohort study designs basic
participants are chosen from a population that is AT RISK of developing outcome of interest
- participants are classified as exposed or unexposed
- - exposure is observed not assigned
- follow participants forward in time
population type (source v sample)
source - group of people which the study participants will be selected (Duquesne students)
sample - the people what are actually selected for the source population (students in epidemiology)
developing a hypothesis
PICO question
- investigators may want to learn about many things that are in a broad research area
- cohort studies are designed to study one or many hypotheses
steps in a cohort study
- state the research question
- objective and hypothesis - design the trial
- define exposures / outcomes
- define cohort - conduct the trial
- selection of subjects
- follow and collect data - analyze and report the data
cohort
a well -defined group of individuals who are at risk for the outcome of interest at the start of the follow-up period and are followed or traced over a period of time
birth cohort
a cohort defined by being born during a particular period of time
general population cohort
used if the goal of the cohort is to value the effects of exposures which commonly occur in large populations
- broad admission criteria
- investigators can assess a wide range of exposures and outcomes
special exposure
participants are intentionally selected on the basis of their exposure status
- good for rare exposures
closed cohorts
defined by a common start time, and no on his added to the cohort during the follow up period
- only exit the cohort If they get the outcome of interest, study ends or they die
- it is difficult to maintain
- analyzed by RISK
open (dynamic) cohort
allow members to enroll and leave at different times
- person-tine contributed by individuals vary
- analyzed by RATE
real time (prospective) cohorts
occurs when investigators identify and enroll a cohort of at-risk individuals and follow them forward in real time to see who develops outcome of interest
- pro: recruitment and data collection are controlled by investigators
- con: more expensive, latency period
historical (retrospective) cohorts
use historical records to understand what has happened to a cohort of individuals in the past
- pro: cheaper, faster
- con: must relay on exiting data which cannot change how it was collected
exposure groups
- should be as similar as possible to reference group
- confounders: any extraneous factors that may be associated with the outcome of interest
internal reference group
comprised of the unexposed members of the cohort
- used in general population cohorts
external reference group
comprised of unexposed individuals from some external group
- used in special exposure cohorts. since everyone has special exposure, your reference has to come from an outside group
risk period
time during which an exposure of interest could possibly cause an outcome
- includes induction and latency period
induction period
time it takes for an outcome to occur due to an exposure of interest
latency period
time during which an outcome exists but is not clinically or otherwise diagnostically detectible
induction and latency period
are important when calculating measures of incidence because they help define who is at risk of developing the outcome DUE to the exposure of interest
incubation period
reserved for infectious diseases and is defined as the time between when exposure to an infectious agent occurs and symptoms begin
strengths of cohort studies
- temporality
- given that the exposure status is observed and/or recorded prior to the occurrence of the outcome, the temporality between exposure and outcome is established by cohort design - study of rare exposures
- relatively well suited for study rare exposes or exposures that would occur infrequently int eh general population - study of several outcomes
- allow study of associations between multiple exposures on several outcomes allowing researchers to test multiple hypotheses - calculation of measures of outcome frequency
- allows investigators to calculate the risk or rate of the health outcome under study (researchers better understand changes in outcome frequency over time)
limitations in cohort studies
- loss to follow up
- can lead to selection bias - study of rare outcomes
- investigators may not observe enough people with outcomes to detect difference in their occurrence between exposure groups - study of exposure outcomes relationship with long induction period of outcomes with long latency periods
- investigators would have to wait a long time to collect data and draw conclusions - logistics
- can be expensive, time consuming and resource intensive