Epidemiology in Public Health Flashcards
Describe prevention in public health
- Considers baseline risk, strength of relationship between risk factor and outcome, and prevalence in population (attributable risk and population attributable risk).
- Occurs at different levels of the population (individual/high-risk vs. population) and at different stages of developing an outcome (primary, secondary and tertiary).
Detail the different levels of prevention strategy
- Primary: aims to stop outcome by preventing or removing an exposure to a risk factor (via high risk or population approaches)
- Secondary: aims to interrupt outcome in early to mid-stages by early detection (screening) and treatment
- Tertiary: aims to reduce complications and severity of an established outcome through treatments and interventions. Treatment to reduce impact eg. diabetes
Distinguish between population and high-risk strategies and when they are each best used
Individual/High Risk Approach:
- focuses on reducing risk of individuals from high risk groups
- good when interventions are costly and there are limited resources
- those at greater risk are more likely to comply and get the greatest benefit
- LIMITATIONS: level of risk varies within group, relies on compliance,can be stigmatising and places burden of responsibility on high risk groups, may not seek to change circumstances of exposure to high risk factors (eg. vaccination rather than contaminated water source), smaller level of public health impact.
Population/Absolute Risk Approach:
- focuses on reducing absolute risk of outcome associated with exposure to a risk factor
- individuals have varying degrees of exposure for many outcomes
- many people exposed to low risk creates more cases than few people with high exposure - greater public health impact
Describe the prevention paradox
- Rose, 1992
- A measure that brings large benefits to the community offers little to each participating individual
- Often requires social attitudes/pressure, economic incentives or legislation
Describe the different types of dose-response profiles
Threshold dose
- Risk of adverse outcome increases rapidly after threshold point.
- Intervention should be targeted near the threshold point.
Linear
- Greater exposure = greater risk
- Population level behaviour change required to reduce risk via promotion and legislation
Curved Linear
- Risk of outcome increases with exposure but lower at low levels
- If outcome occurs at low exposure requires population approach
- If outcome occurs at high exposure requires targeted approach
J-shaped/U-Shaped
- Increased risk of outcome at high and low levels of exposure
- Requires targeted interventions as cannot shift the population too far either way.
What is the purpose of screening programs?
- Key preventative intervention strategy that enables early detection and treatment
- Can be high-risk or population
- Can be systematic or opportunistic
List criteria for assessing appropriateness of screening programs
- Availability of effective and reliable screening method
- Availability of intervention to reduce or improve outcome
- Safety and acceptability of the test to the individual
- Benefits must outweigh any harm
- Should consider cost of relative burden (cost of screening vs. cost saving for being fewer patients) and operational feasibility
How is the validity and reliability of screening methods evaluated?
Validity
- Based on ability of screening method to distinguish between individuals with and without the condition of interest
- How well can it predict disease - tested via sensitivity and specificity
Reliability
- Based on how consistent the results of the test are when the test is repeated (with the same person under the same conditions)
- Predictive value (negative and positive) - based on sensitivity and specificity of the screening method
Describe ‘sensitivity’
The proportion of those who have the outcome who are correctly identified
-> True positives
= Number of true positives / Number of those with the outcome
- Low sensitivity creates a high number of false negatives
- Need high sensitivity for infectious degree control to prevent transmission
Describe ‘specificity’
The proportion of those who do not have the outcome who are correctly identified
-> True negatives
= Number of true negatives / Number of those without the outcome
- Low specificity creates a high number of false positives (those incorrectly identified to have the outcome)
- High specificity is preferred when confirmation of outcome is invasive or expensive, or if the result is stigmatising
Describe ‘positive predictive value’
The likelihood of having outcome based on test result
= Number of true positives / Number of tested positives
- High PPV = high levels of specificity
- Less likely for someone with a negative result to have the condition
Describe ‘negative predictive value’
The likelihood of not having an outcome based on the test result
= Number of true negatives / number of tested negatives
- High NPV = High level of sensitivity
- Less likely for some with positive result not to have the outcome
Describe the relationship between predictive value and prevalence
- Predictive value is dependent on the prevalence of an outcome
- PPV decreases and NPV increases as prevalence decreases because overall probability of becoming a case has declined
Discuss some of the other considerations required when measuring effectiveness of screening
- Selection Bias: those who participate in screening often differ from those who don’t
- Lead-time Bias: early detection appears to prolong survival but has no effect on the outcome
- Length-time bias: screening is more likely to detect an outcome with a slow progression
- Over-diagnosis: may never become symptomatic or die of another diagnosis
List and briefly define the four main methods of public health surveillance
Routine:
- regular, systematic and accurate reporting of health outcomes
- establishes baseline frequencies
- can study incidence changes, assess interventions, hazards and cause-specific deaths
- civil registrations, census, population health indicators
Passive
- data collection at point of contact and/or practitioner reporting
- many factors affect data quality
- can lead to underreporting
- health facility data, outcome registries, notifiable outcomes
Sentinel
- use of selected reporting sites (with capacity) when passive surveillance does not provide high quality data
Active
- uses case-finding techniques, population surveys, review of clinical records
- commonly used for outbreaks of collecting data on prevalence of known risk factors
- useful for monitoring emergence or elimination of disease (particularly if lack of formal care)
