Revision powerpoint - everything we need to know Flashcards
describe the epidemiological transition
changes in level and causes of mortality
decline in total mortality
reduction in infectious disease
increases relative role of chronic non-communicable disease - cancer, CVS, chronic resp disease, diabetes
dynamic nature of the epidemiological transiotion
result of demographic, socioeconomic, technological, cultural , env and biological changes
small pox disappear, AIDS appear, TB denge re-emerge
decline in stomach cancer, rise and fall of lung cancer
shift from stroke to heart disease
most commonly diagnosed cancer
lung, breast, colorectal
most common cause of cancer death
lung liver and stomach
burden of non-communicable disease
1/3 cancers likely to be preventable through small number of lifestyle and environmental approaches
smoking largest preventable casue of cancer worls wide
burden shift to less developed countries
incidence caries between populations
cancer
major problem
>25% deaths world woide
in 2010 15.1% deaths - 8million died
canvcer rates in migrants converge to local rates
modifyable risk factors
take up to 20 years to appear
current rates affected by changes and exposure that took place in the past
cancer risk factors
smoking low intake fruit/veg alcohol unsafe sex obesity physical inactivity contaminated injections urban air pollution indoor smoke from household solid food use
cancer from infections
hepititus - liver H pylori - gastric HPV - cervical EBV - hodgkins lymphoma, stomach cancer HIV - AIDS defining malignancies, Kaposi sarcoma, non-Hodgkin lymphoma, cervical cancer schistosomes - bladder cancer
CVD
CHD and stroke 1st and 2nd cause specific mortality ww
29.5% all deaths - 15.6 million
more in developing world
low in japan
high and rising rates in formally socialist economies - Europe and middle east
rates higher in men than women - gap shrinking
trends declining in many countries recently
patterns suggest environment rather than genetics
risk factors for CVD
high bp
tabacco smoking
cholesterol level
commonest non-infectious cause of world mortality
ischemic heart disease
cerebrovascular disease
where are infectious disease the leading cause of death
sub sarharan Africa
6 commonest infectious cause of world mortality and some underlying high incidence
lower respiratory infections - 3.9m HIV 2.8m diarrheal disease - 1.8m TB - 1.6m malaria 1.2m measles 0.6m
why is there a change in mortality and incidence of infectious disease
treatment exposure diagnosis screening treat and diagnose early - mortality decrease but incidence increase
case
Person who has the disease, health disorder or suffers the event of interest
incidence
Number of new cases of a disease within a specified time interval - probability or risk of developing a disease in a specific time period
prevalence
Frequency of a disease in a population at a point in time (point prevalence) - proportion and measures status, for planning, compare burden between populations
mortality
Number of deaths attributed to a specific condition in a given time period - number deaths per 1000 individuals per year
morbidity
Number of cases of ill health, complications, side effects attributed to a specific condition over a particular time period.
- state of being unhealthy or diseased
drivers of aids epidemic, success and challenges of the response
access to anti-retroviral therapy
effective HIV prevention methods - safe sex
decline in prevalence in pregnant women
to reduce incidence further need a vaccine
incident increase because of better diagnosis and better treatment - incidence increase and mortality decrease
HIV infection - prevalence and mortality
incidence rising
mortaklity reduced because of HAART
duration of disease increasing
steep increase in prevalence
routine data
data routinely collected
recorded in ongoing systematic way - for administrative or statuary purposes without specific research questions
examples of routine data
deaths, hospital admissions, screening, immunisation uptakes, census counst, GP consultation data
major sources routine data in UK
2001 Census
Health Survey for England
NHS Inpatient Survey on patient experience
advantages of routine data
Relatively cheap Already collected and available Standardised collection procedures Relatively comprehensive – population coverage, large numbers Wide range of recorded items Available for past years
disadvantages
not answer qn incomplete ascertainment variable quality validity variable disease labling vary need careful interpretation
SMR
ratio between observed number of deaths in a study population and number of deaths would be expected - accounting for age and sex
SMR =
no observed deaths/no expected deaths if same age specific rates as standard population
age standardised death rates
measure how many people die each year and why they have dies assess effectiveness of a country’s health system
along with assessing how injuries and diseases affect the living - determine if focussing on the right thing that will prevent reducable deaths and disease
main approaches of intervention to improve health
clinical intervention - biomedical, prevention
health education
healthy public policy eg smoking
community development
primordial prevention
prevention of factors which promote emergence of lifestyles, behaviours, exposure patterns which contribute to the risk of disease
primary prevention
prevent the onset of disease - limit exposure of risk factor by individual behaviour change and actions in the community
health promotion and specific protection
secondary prevention
halt progression once illness is established - prompt effective treatment
special consideration for asymptomatic
tertiary prevention
rehab of people with established disease to minimise residual disabiliyu and complications
QOL
sample
group of people, objects or items taken from a larger population for measurement
variation
variation of observations in a single sample
what can you get from a sample
estimates of true underlying risks
problem with sample
risk that association is die to chance
p value
probability that the null hypothesis is tru
null hypothesis
hypothesis that there is no significance
attributable risk
difference in rate of a condition between exposed and unexposed population - guantifies the risk - fact
relative risk
ratio of the probability of an event occurring in exposed group to probability of event occurring in comparison in a non-exposed group
estimates magnitude of association
attributable risk =
incidence in exposed- incidence in unexposed
relative risk =
incidence in exposed/incidence in unexpoded
odds ratio
likelihood of having exposure if you have disease relative to having exposure if no disease
OR =
odds of exposure in cases/odds off exposure in controls
confounding
mixing of effects between exposure, the disease and a 3rd factor
dealing with confounding at desighn
randomisation - RCT
restriction/matching - case control
dealing with confounding in analysis
stratification - split analysis by age group
standardisation
regression - build statistical model
EBM
based on critical appraisal
dr should use critically appraised information in clinical practice for optimal care
developed over 30 years
why EBM matters to clinicians
Patient Medical Knowledge Practice-Based Learning and Improvement Interpersonal and Communication skills Professionalism
association
Association refers to the statistical dependence between two variables. Consider chance, bias, confounding, cause (CBCC)
chance
from samples not whole pop
bias
systematic - measurement/selection
casual effect
A simple way to remember the meaning ofcausal effectis: B happened because of A, and the outcome of B is strong or weak depending how much of or how well A worked.
Cause-effect relationship
Judgement is based on a chain of logic that addresses two main areas:
-Observed association between an exposure and a disease is valid
-Totality of evidence taken from a number of sources supports a judgement of causality
power calculations
to calculate minimum sample size for acceptable outcome
responder bias
prevented by single blinding
ecological studies
measure prevalence and incidence of disease, particularly when disease is rare.
- observationalstudydefined by the level at which data are analysed, namely at the population or group level, rather than individual level.
case report
follow up of an individual patient
descriptive/cross sectional
: Observationalstudythat analyzes data collected from a population, or a representative subset, at a specific point in time
strength
Measured by the magnitude of relative risk. Strong association is more likely to be causal and a weak association could be more easily the result of confounding or bias
consistency
similar results in similar studies
specificity - Bradford hill
particular exposure increasing the risk of a certain disease but not the risk of other diseases
temporal relationship
For a putative risk factor to be the cause of a disease, it has to precede the disease.
dose response relationship
Increasing levels of exposure lead to increasing risks of disease shows evidence of a causal effect.
plausibility
consistant with other studies
coherence
cause and effect doesn’t conflict with what is known with the natural history
experimental evidence
On humans (rarely available) or animals
analogy
Provides a source of more elaborate hypotheses about the association in question. Absence only reflects lack of imagination or experience.
clinical trial
A planned experiment in humans designed to measure the effectiveness of an intervention (usually a drug but can be surgical procedures, vaccine, complementary therapy etc)
features of a clinical trial
Define your intervention
Define your comparator • Placebo • Alternative treatment • Standard of care
Define your inclusion criteria
Define your exclusion criteria
Experimental study
Must contain a control group- to be sure why the outcome has occurred (could be due to the effectiveness of the new treatment or happened anyway)
Prospective: participants are followed through time
Patients are enrolled, treated and followed over the same period of time
Patients should be randomised –to eliminate allocation bias
block randomisation
assign people to group A or B randomly
stratification
by important characteristic
minimisation
adaptive stratification - calculates imbalance and allocates to maintain balance
biases in clinical trials and how to avoid them
allocation bias - randomisation
measurement bias - blinding
reporting bias - selective reporting
consort
consolidated standards of reporting trials - ensures papers about trials incude all relevant info
outcomes of a clinical trial
The outcomes are presented in terms of efficacy (the true biological effect of a treatment) or effectiveness (effect of a treatment when actually used in “normal” practice).
experimental event rate
incidence in intervention arm
control event rate
incidence in control arm
relative risk
eer/cer
relative reduction
cer-eer/cer
absolute risk reduction (ARR)
cer-eer
number needed to treat
1/ARR
what is included in a summary of information from a paper
Why did they do it? What did they do? What did they find? What did they conclude? In your opinion, was the study conducted well?
critical appraisal cchecklist
Question – Relevant? Hypothesis? Design – appropriate? Population – sample size? Methods – measure? Appropriate? Analysis – appropriate statistical tests? Confounders Bias – measurement, selection Ethics – ethical? Consent? Interpretation – casual inference (Bradford Hill
guidelines for different studies
RCT - consort
observational studies - STROBE
systematic reviews - PRISMA/MOOSE
all or non-case series
when all patients died before a new therapy was introduced, yet following its introduction, all patients receiving the treatment survived
why conduct a systematic review
high volume of data that needs to be considered
single studies are inefficient
meta analysis
combine the published estimates of effects from each study to generate a pooled risk estimate
what is a systematic review
review of a clearly formulated question that uses systematic and explicit methods to identify, select and critically appraise relevant research and analyse data from studies that are collected in the review
stages of systematic review
1 - planning
2 - identify research, select studies, study quaaltity assessment
3 - report and dissemination, study details abstracted and details tabulated to show summary of findings - estimate overall effect using a meta analysis
meta analysis
combine the published estimates of effects from each study to generate a pooled risk estimate
if too heterogenous might be inappropriate to pool
why do a meta analysis
more reliable and precise estimate of effect
difference between studies identified and explored
get pooled weighted average
presenting a meta analysis
forest plot
each study represented with a box and line
size of box = weight given to the study
line = 95% confidence interval
overall estimate is the diamond
centre of diamond and line - summary effect estimate
width of the diamond = CI around estimate
bias and limitation of systematic review
publication bias
inconsistency
low study quality
heterogeneity
publication biasd
cause data to be over or under estimated
explored with funnel plots - show if link between study size/precision and the effect of the estimate
what is screening
investigating apparently healthy individuals to detect unrecognised disease or its precursers so measures can be taken to prevent or delay the development of disease or improve prognosis
why screen
when early diagnosis = better prognosis
high risk - intervention = better prognosis
identification of those posing risk to others - eg infectious disease eg Hep B
mass screening programs in the UK
cervical cancer breast cancer bowel cancer abdominal aortic aneurysm diabetic eye screening
cervical screening
every 3yrs 25-49
5yrs - 50-64
breast screening
3yrds 50-70
>70 can self refer
bowel screening
2yrs men and women 60-74
abdominal aortic aneurysm
men 65th year
diabetic eye screening
people with T1/T2 DM >12
validity of screening test
ability to distinguish between those with the condition and those without
gold standard
a recognised way of determining who has the disease
sensitivity
ability of test to correctly identify people with the disease (true positives)
specificity
Ability of the test to correctly identify people without the disease (ability to exclude true negatives)
predictive value
Proportion of test results that are correct
dependent on the sensitivity and specificity and the prevalence of the condition in the population
+ve predictive value
likelihood that a patient with a positive test result will actually have the disease
-ve predictive value
likelihood that a patient with a negative test result will not have the disease
receiver operator characteristics curves
are used to determine a cut-off value for a diagnostic or screening test. Y axis- sensitivity vs 1-specificty (x-asis)
how do you choose cut off value
is informed by the attempt to maximise sensitivity and specificity
criteria for screening
feasibility effective selection bias lead time bias length time bias cost ethics
ethics
- May harm as well as benefit the individual
- Risks of the test
- Risks of subsequent diagnostic tests
- Risks of subsequent treatment
- False positive result causes unnecessary anxiety
- False negative result will give false reassurance
feasibity
easy to get attendance
acceptable
facilities for diagnostic tests
effectiveness
extent it affects subsequent outcomes
selection bias
people who participate are different to those that don’t
lead tiem bias
improvement really due to earlier date of diagnosis
length time bias
some conditions slower in developing to a health threatening phase - more likely to be detected at a stage that has favourable prognosis - false conclusion that screening is beneficial
cost
includes the diagnostic test and treatment in comparison to costs of treatment of more advanced disease
