module 1 Flashcards
incidence - strengths
incidence is determined only by the disease risk in a population - clean measure of disease occurrence
incidence measure include events (N), population (D) and time (T)
Incidence - weakness
incidence can be difficult to measure as you have observe events over time
prevalence strengths
is easy to measure as you ‘stop time’ and count
prevalence weakness
measures include only events (N) and population (D) - less information than incidence
is determined by the incidence, cure rate and death rate (dirty measure of disease occurrence)
an epidemic
the occurrence of ‘disease’ is clearly in excess of normal
pandemic
an epidemic occuring in many countries
only one cg and many eg
Ecological studies
POPULATIONS/MULTIPLE GROUPS OF PEOPLE into eg and cg
Individual participation studies
allocate INDIVIDUALS to EG and CG & in triangle
prevalance
can be measured BACKWARDS <—-
RD
RD= EGO-CGO/CGO-EGO
EGO=CGO
RD = 0
EGO/CGO
RR
rr and rd
beware of the large RR but small RD
PECOT Triangle parts
Top: Is the SETTING well described
Mid: Is ELLIGABLE POPULATION as indentifiable, meaningful population?
Bot: Do the PARTICIPANTS represent the elligable population
in cohort studies, the participants are allocated by
measurement
cross sectional studies are allocated by
measurement to EG and CG
ramboman: maintence problem
long term study
Asking participants if they ___
subjective
Asking participants yes/no
objective
sub–>obj
make it anonymous
video of showing symptoms
double blind RCT
if everyone is “blind” it is less important if the outcome is not measured objectively
confounding: adjusting
sub-studies like this (2 seperate PECOT studies)
If allocated randomly to EG & CG
adjustment usually necessary
Measuring the truth
- because in biology the study participants are moving targets
- because we can never study everybody
we can only study a sample of all the people we would like to include in a study
- even identically
even identically designed & implemented studies will never include identical participants with identical characterstics
Goal of epidemiology
EGO
CGO
effects (rr and rd) in the whole population
Goals of epidemiology 2
but we usually estimate disease occurrence (ego & cgo) & effects (rr & rd) in 1 or 2 samples from population
95% confidence interval addition
assumes no non-random error in ramboman
wider confidence interval = more uncertainty
if 95% Cls for the RD (EGO-CGO) overlaps 0, no statistically significant difference
if 95% Cls for the RR (EGO/CGO) overlaps with 1, there is probably no statistically significant difference between them
RCT & COHORT
can be longitudinal studies, both arrow going down and across for time
cross sectional
only prevalence so only one arrow going across
rct
experimental study
cohort and cross sectional
observational study
systematic review
- REVIEW the literature SYSTEMATICALLY to FIND all relevant studies (a lot of pecot)
- assess the quality of studies and only KEEP the good ones (only few pecot)
usually done for rcts because most rcts are small
systematic review step 3
COMBINE results of good studies in a META ANALYSIS
Strengths & Weaknesses of IP RCT
Strengths & Weaknesses of IP cohort studies
Strengths & Weaknesses of IP cross-sectional
studies
population health epidemiology:
build fence
(prevent) on clifftop for people still healthy
