Epidemiology Flashcards
What are the three goals of public health?
What perspective do they emphasise?
- health protection - protecting people’s health (for example from environmental or biological threats, such as food poisoning or radiation)
- health improvement - improving people’s health (for example by helping people quit smoking or improving their living conditions)
- health services are the most effective, most efficient and equally accessible
Population/group perspective
Maslow’s hierarchy of needs (bottom-up)
- P______ (5)
- S______ (4)
- L___/______ (3)
- E_____ (4)
- S___-a______ (4)
Physiological: Sex, sleep, warmth, food, water, breathing, shelter
Safety: Health, employment, family, home
Love/belonging: friends and family, sexual intimacy
Esteem: self-esteem, confidence, respect from others, achievement
Self-actualisation: morality, creativity, spontaneity, problem solving
Phase 3A public health definitions:
a) Epigenetics
b) Allostasis
c) Allostatic load
d) Salutogenesis
e) Emotional intelligence
a) genome expression dependent on environment
b) “stability through change”, physiological adaptation to react rapidly to environmental stressors and achieve stability (homeostasis)
c) long-term overtaxation of stress responses leading to impaired health
d) Favourable physiological changes secondary to experiences which promote healing and health
e) The ability to identify and manage one’s own emotions and those of others
Epidemiology definitions: a) Incidence b) Prevalence c) Absolute risk d) Relative risk e) Attributable risk (e.g. absolute risk difference) f Person time
a) No. of new cases over a period of time (amount of water filling bath from tap)
b) Total no. of cases at a particular point in time (total bath water)
c) Probability: no. of events (e.g. disease) divided by no. of people in group
d) Ratio of probability of outcome (e.g. disease) in exposed versus unexposed group
e) Risk that may be attributed to the exposure (incidence in exposed minus incidence in unexposed)
f Measure of time at risk (e.g. from study entry to disease onset)
Bias
a) Define
b) Selection bias - types
c) Measurement bias - i) types, ii) sources
a) Systematic error leading to distortion of the true underlying association
b) Selection of participants, allocation to groups
c) Measurement of exposure or outcome;
observer, participant (e.g. recall bias), instrument
Association and causation
a) 4 reasons for mistakenly assuming causation
b) What is confounding
a) Bias, chance, confounding, reverse causality
b) A factor Z is associated with the exposure X and independently influences the outcome Y but does not lie on the causal pathway
Prevention
a) Primary
b) Secondary
c) Tertiary
a) Prevent onset
b) Treat subclinical/early disease
c) Treat established disease to aid recovery/prevent morbidity
Bradford-Hill Criteria (mnemonic)
- BRADFORD
If it doesn’t fulfil these criteria - maybe it was caused by bias, confounding or chance!
Biologically plausible, Reverse causality, Association (strong - low P value), Definite, Faithful (consistent in different populations), Obscure, Reversibility Dose response
Example: study finds association between stress and HTN (but HTN could have increased stress levels: reverse causality)
Study designs
a) Cohort
b) Case-control
c) RCT design
d) Cross-sectional
e) Ecological - measure influence of what two variables?
a) People without disease assigned according to exposure; measure difference in outcome (e.g. disease onset); longitudinal; prospective inquiry direction
b) People assigned according to outcome (disease vs no disease) and assess differences in exposure; retrospective line of inquiry
c) Strict selection, random allocation, blinding of researchers and participants, control of other variables (only changing the experimental drug, etc.)
d) Prevalence - one point in time (e.g. survey)
e) Change according to time and place
Calculation of:
a) Absolute risk
b) Relative risk (risk ratio)
c) Odds ratio
f) NNT - calculation
a) Probability
b) Probability in exposed/unexposed
c) 1 - probability in exposed/1-probability in unexposed
d) 1/ARR
(number needed to harm = 1/AR excess)
Approaches to prevention
a) Population
b) High-risk
a) Seeks to shift the risk factor distribution curve (e.g. sugar tax, seatbelts in cars)
b) Identifies individuals above a certain threshold (e.g. BP 140/90) and treats them
Screening
a) Define
b) Types of screening
c) I Understand SCREEN (criteria for determining acceptability of programme)
a) Process that sorts out people who have a disease from those who don’t
b) Population (cervical screen), Opportunistic (HIV), Infective (MRSA), Occupational (occy health)
c) - Important,
- Understand natural Hx of Dx,
- Sensitive/Specific,
- Common,
- Risks outweighed by benefit,
- Early -stage identification,
- Expenses low,
- Non-invasive and acceptable
Test variables. (a = TP, b = FP, c = FN, d = TN)
a) Sensitivity
b) Specificity
c) PPV
d) NPV
e) True or false: prevalence affects each of the above 4
Sens = a/a+c (disease picked up) Spec = d/b+d (non-disease picked up) PPV = a/a+b (% positive results that have disease) NPV = d/c+d (% negative results that have non-disease)
False: affects only the PPV and NPV
What is lead time bias?
What is length time bias?
Interval between detection of (subclinical) disease and its presentation; they die at the same time but you just pick them up earlier
Screening more likely to detect slower progressing diseases (rapidly progressing conditions may result in death before screening date)
Cohort study
- advantages
- disadvantages
- rare exposures, longitudinal
- need a large sample; takes a long time
