Public Health Flashcards
What are the 9 Bradford-Hill Criteria?
Temporality Strength of association Consistency Dose-response relationship Biological plausibility Specificity Coherence Experiment Analogy
What is a type 1 error
Rejecting the null hypothesis when it is true - saying there is an effect when there isn’t
What is a type 2 error
Accepting the null hypothesis when it is false - missing an effect that actually exists
A Pearson’s coefficient of 1 means…
Perfect correlation
What are the six steps of CDC Evaluation framework?
- Engage stakeholders
- Describe programme
- Design evaluation
- Collect data
- Justify conclusions
- Disseminate findings
6 Kass ethical framework domains i.e. process to work through when initially evaluating
- Proposed benefits
- Proposed programme
- Burdens
- Mitigation
- Distribution of benefits/burdens (equity)
- Balance of benefits/burdens (justice process)
4 Kass ethical burdens
- Privacy/Confidentiality
- Autonomy/Liberty
- Justice/Equity
- Individual health
7 general ethical considerations
- Producing benefits
- Avoiding/preventing harms
- Utility
- Distributive and procedural justice
- Respecting autonomy
- Protecting privacy and confidentiality
- Trust
5 Principles for resolving ethical conflicts
- Effectiveness - infringing moral considerations will improve health
- Proportionality - benefits must outweigh harms
- Necessity
- Least infringement - least burdensome alternative
- Public justification - explanation and transparency
Health Protection Amendment Act (2016) escalation pathway intervention
- Scope public health risk
- Voluntary measures (informal contact tracing)
- Formal contact tracing
- Directions and court orders
- Urgent public health orders
- Prosecution
Beauchamp and Childress Ethical Principles
- Beneficence
- Non-maleficence
- Autonomy
- Justice
Definition of public health
The science and art of preventing disease, prolonging life and promoting, protecting and improving health through the organised efforts of society
WHO Definition of health
A state of complete physical, mental and social well-being and not merely the absence of disease or infirmity
3 domains of public health
- Health Improvement
- Health Protection
- Health Service Delivery and Quality
DPSEEA framework for intervention
- Driving forces - car dependency
- Pressures - daily commuting
- State - average energy expenditure
- Exposure - individual energy expenditure
- Effect - obesity and NCDs
- Action - intervention at each level
11 WHO Healthy Cities qualities
- Clean, safe, high quality physical environment
- Stable and sustainable ecosystem
- Strong, mutually supportive and non-exploitative community
- Citizen participation and control
- Basic needs (food, water shelter, employment, income)
- Access to experiences and resources
- Diverse, vital and innovative economy
- Cultural and biological heritage
- An urban form that promotes all of the above
- Public health and sick care services
- High health status
4 purposes of epidemiology
- Causation
- Natural history
- Description of health status of population
- Evaluation of interventions
Define incidence rate
the number of new cases in a given period of person-time
Define cumulative incidence
number of people who develop disease during a specified time period, out of the total number of people who were at risk over that period and were free from disease to start with
Define point prevalence
proportion of population of people with disease at a specified point in time
Define period prevalence
proportion of population with disease at any time within a given period
Define Attributable risk
The difference in incidence between what the exposed and unexposed populations experience i.e. the extra burden that is due to the exposure
3 types of causes
- Sufficient cause - an event or exposure that, having occurred, ensures the outcome will occur
- Necessary cause - an event or exposure without which the outcome will not occur
- Component cause - an event or exposure which increases the likelihood of the outcome occurring, with an attributable fraction of <1 and >0
Epidemiologic triad
- Host
- Agent
- Environment
4 types of prevention
- Primordial - aims to alter the environments that allow unhealthy exposures to occur
- Primary - aims to reduce incidence of disease through exposure reduction
- Secondary - aims to cure or manage disease through early diagnosis and treatment
- Tertiary - aims to reduce complications or disability due to disease
3 Pros of ecological studies
Easy
Cheap
Uses existing datasets
4 Cons of ecological studies
- Cannot prove causation, can only infer it
- Has little explanatory power at the level of the individual
- Prone to confounding
- Ecological fallacy - inferring things about individuals in a population based on population characteristics e.g. men are more likely to get IHD, thus Tom has a higher risk of IHD than Mary
3 Pros of cross-sectional studies
- Useful for describing prevalence
- Good for qualitative explorations
- Quick and cheap
4 Cons of cross-sectional studies
- Can’t measure incidence as no time component
- Can’t establish causation as no temporality
- Difficult to use for acute or wax-and-wane disease as they’re less likely to be present at the time you ask
- Prone to selection bias
3 Pros of case-controls
- Relatively quick and cheap
- Good for rare diseases - don’t have to have a huge cohort to generate a few rare cases
- Can consider multiple risk factors
5 Cons of case-controls
- Not good for rare exposures
- Suffers heavily from recall bias
- Can’t calculate an actual risk estimate, or an incidence
- Prone to confounding
- Timelines unclear so difficult to prove causation
3 Pros of cohort studies
- As there is a clear timeline, can be used to establish causality, and doesn’t suffer from recall bias so much
- Can calculate incidence rate and an actual risk to presently disease-free people
- Can look at multiple outcomes - i.e. can determine a number of bad things that happen to smokers vs non-smokers
4 Cons of cohort studies
- Can be very difficult to retain people in the study - loss-to-followup is a big risk
- Can only study one exposure - this is the factor that separates the two groups
- Not good for rare diseases - need a really big group to get a few cases
- Expensive and time-consuming
2 purposes of randomisation
- Makes the two groups (intervention and control) as similar as possible at baseline, which is the most effective way to control for confounders (both known and unknown)
- Eliminates bias in treatment allocation - doesn’t allow for patient preference or willingness or investigator bias
3 parts of Triple blinding
- Participants are unaware of group allocation - controls for placebo effect
- Investigators are unaware of group allocation - controls for investigator/interviewer bias
- Analysts are unaware of group allocation - controls for bias introduced
4 reasons to do intention-to-treat analysis
- Gives an idea of the real-world performance of the intervention, because these issues occur in the real-world - no treatment is ever done under ideal conditions. I.e. effectiveness rather than efficacy
- Preserves the balance of baseline characteristics from randomisation - it may be that e.g. males are more likely to switch groups than females, which if you analysed the groups people ended up in would introduce confounding by gender
- Preserves your sample size and thus power - if you excluded dropouts, the sample size would shrink
- Prevents conscious and unconscious bias related to outcomes, or loss of data because people drop out
Reason for doing per-protocol analysis
Analysing only the people who completed the treatment as specified gives you the efficacy - how well the treatment works in the ideal, cleanest conditions
4 Pros of Community Trials
- Good for public health interventions where a community or system is the target for intervention rather than individuals
- Often more convenient - easier to randomise whole classes of children, rather than individual children in a class as this might be quite disruptive to the real-world functioning of the class
- Measures system effects
- Usually generalisable as studies whole communities
3 Cons of Community Trials
- Require long follow-up for public health interventions
- Inter-cluster correlation - clusters may be fundamentally different to each other due to differences in the way people associate themselves - need a large enough number of clusters to account for this and good inclusion criteria and work put into recruitment
- Often run as open cohorts - people enter and exit the study during the course, due to real-world influences
Steps of meta-analysis
- Define research question
- Systematic search of literature
- Quality assessment
- Pooling results
6 Limitations of meta-analysis
- Require homogeneity for pooling of results to be valid - can be very difficult to account for
- Dependent on quality of source data - bad inputs create bad output
- Can be seen as final word on a research question - doesn’t leave room for nuanced understanding
- Vulnerable to publication bias if this is not addressed
- Work best with access to primary data rather than just summary estimates and statistics
- Controversial to use in study types other than RCT
Criteria for confounding
- Associated with exposure
- Causes outcomes
- Not on causal pathway between exposure and outcome
Strategies for addressing confounding
- Randomisation
- Restriction
- Matching
- Stratification
- Standardisation
- Regression analysis
Define selection bias
Bias that arises from the way participants in a study are selected and from factors that influence participation in the study.
Causes of selection bias
- Unconcealed randomisation
- Volunteer bias
- Ascertainment bias - exposed more likely to be diagnosed than control
- Healthy worker bias
- Loss to followup
- Faulty selection criteria
Define measurement bias
Bias that arises when measuring tools do not accurately reflect reality
5 Ottawa Charter Actions (in descending order)
- Build Healthy Public Policy
- Create Supportive Environments
- Reorient Health Services
- Strengthen Community Action
- Develop Personal Skills
3 Ottawa Charter strategies
Advocate
Enable
Mediate
8 steps of Gosling Advocacy Framework
- Analyse the situation
- Identify advocacy issue
- Set goals/objectives
- Analyse policy and power
- Identify targets and influencers
- Develop message
- Build added strength
- Develop and implement action plan
5 elements of TWTW
- Taha Wairua - spirituality
- Taha Hinengaro - mental fortitude
- Taha Tinana - physical development
- Taha Whanau - family as basic unit of society
- Whenua (unseen) - land and whakapapa
4 stars and 2 pointers of Te Pae Mahutonga
- Mauriora - access to Te Ao Māori
- Toiora - healthy lifestyles
- Te Oranga - participation in society
- Waiora - environmental protection
Ngā manukura - community leadership
Te Mana Whakahaere - autonomy
5 Kaupapa Maori principles
- Tino Rangatiratanga
- Social Justice
- Te Ao Maori
- Te Reo
- Whanau