EPI 2 Flashcards
What are the 2 main things diagnostic tests must classify and the measure of these
- Correctly class diseased animals as diseased
- Diagnostic Sensitivity - Correctly class non-diseased animals as non-diseased
Diagnostic Specificity i
What does high sensitivity and high specificity imply
High Sensitivity implies a low number of false negatives
High Specificity implies a low number of false positives
Define sensitivity and specificity
Sensitivity -> number of true positives within the disease animals
Specificity -> the number of true negatives within the non-diseased animals
What are the 2 important questions when applying diagnostic tests and what equations answer this
- An animal has come up test positive to a particular condition, what is the probability of the animal really having the condition?
- Positive Predictive Value (PPV) - An animal has come up test negative, what is the probability that the animal is really free from the condition?
- Negative Predictive Value (NPV)
How to make a diagnosis and therefore how diagnostic tests work and their downfall
- Aim is to correctly identify abnormalities -> Normal Vs Abnormal -> for this wee need to define boundaries around normality - THE TEST
How to you separate the populations uninfected and infected?
Almost impossible for most tests always going to have false positives and false negatives
Define a test and list some examples
Any procedure that reduces uncertainty about the state of disease
- questions posed during history taking,
- clinical signs (examination and measurements) - CLINICAL EXAMINATION IS A TEST
- lab findings (haematology, serology, biochemistry, histopath),
- post mortem findings
- Gold standard is what we are sure will give us the result
What are 3 general types of tests and how to know which on to use and if choose wrong what will this affect
Categorized by what they measure
1) physiological or production parameter
2) agent, antigens or nucleic acid
3) animals immune response to challenge (antibodies)
Depends on the point of disease you are testing, unlikely to see antibodies at the start of disease but will see virus and vice versa
Effects SENSITIVITY - may not pick up the disease
What are the two general categories for things measured by diseases and which harder
- Binary outcome (+/-, pregnant versus not pregnant)
- Continuous outcome (heart rate, blood albumin, ab titre)
• need to set a cut-off to define what is normal versus not
What is the perfect diagnostic test, do they exists and how to measure the validity of the test
Perfect -> Would allow us to differentiate between disease positive and disease-negative individuals without error
• Does not exist – the tests that we use are imperfect
• To know how good they are we need to evaluate them against a reference
List and describe 7 characteristics that make for a valid diagnostic test
1) fitness for intended purpose -> accurate and precise
2) robustness (reliability) -> unaffected by small changes in test situation
3) repeatable -> within and between runs in one lab
4) reproducible -> between runs in different labs
5) Diagnostic sensitivity -> few false negatives
6) Diagnostic specificity -> few false positives
5 and 6 -> when tested in animal of known infection status in target population
7) thresholds -> cut-off selection
When is high sensitivity important
- a high cost from calling a diseased animal negative:
- Screening -> testing all animals -> pick up most as positive then later may do a high specific test to ensure that it is a true positive
- quarantine/testing new stock entering a herd)
- testing for something rare (<1% prevalence)
What is need high sensitivity and specificity what tests would you do in what order
When screening for a disease
1) Capture all animals that is positive -> don’t want to miss any -> High sensitivity
2) Confirmation -> follow-up work -> ensure all are true positive - High specificity
When is high specificty important
- a high cost from calling a non-diseased animal positive:
○ confirmatory testing
○ Depopulating (killing) based on test results
○ trade impacts when you get false positive results - ruling-in (confirmation):
What are 2 main ways of improving accuracy, the types within and what does this increase and therefore decrease
- Use several tests instead of a single test:
- Parallel testing(OR) –the animal is considered infected if it is positive to ANY test.
○ ↑ SENSITIVITY, ↓ false negatives -> decrease specific
- Serial testing (AND) –the animal must be positive to ALL test to be considered infected.
○ ↑ SPECIFICITY, ↓ false positives -> decrease sensitivity - Use a different cut-off point for the context
- But always a trade-off
○ Improving sensitivity will reduce specificity and vice versa.
Predicative valves what d they vary with and the 3 things needed to calculate
- PREDICTIVE VALUES VARY WITH PROBABILITY (PREVALENCE) OF THE CONDITION -not a constant trait of the test!
3 estimates: - Sensitivity - set for the test
- Specificity - set for the test
- Pre-test probability - variable
○ the probability of the animal having the condition PRIOR to testing (an estimate/guestimate of prevalence can be used)
How does positive and negative predictive values change with change in prevalence
- PPV goes down with low prevalence
- NPV goes down with high prevalence
what is post-test predicative value and how to get with negative and positive predictive value
Post-test predictive value is the value of the cow being infected given the predictive value
= positive predictive value and the remaining percent with the negative predictive value
What are surveys used for and what get when have a perfect test
- Surveys may be conducted to estimate the prevalence of a condition in a population
- What can be directly estimated is apparent prevalence, the % animals that test positive, not the real (true) prevalence
- True prevalence is the estimate you’d get with a perfect test
Define an outbreak
series of disease events clustered in space and time
List the 5 steps in problem solving
1) Evaluate solution
2) define and analyses problem
3) gather information
4) identify solutions
5) choose and implement solution
What are the 3 aims of an outbreak investigations and what type of approach is needed
- To identify the problem/cause -> problem solving, structured process
- To control it
- To prevent it
Systematic approach - Applied (field) epidemiology involves collecting information for action!
Aim to identify modifiable causal factors AND prevent further cases
List 3 contexts for investigations and what approaches are needed
1) Single agent clear cut outbreak (anthrax in Northern Victoria) -> diagnose, treat and prevent
2) Multi-agent/confused picture -> systematic investigation required (the 7 steps)
3) Emergency animal disease – 1 case is an outbreak -> contain and investigate (again the 7 steps)
With a single agent clear cut outbreak is occurring what are the 4 things success depends on
- true diagnosis being considered
- only one disease being present
- the results of examinations being definitive for the disease
- traditionally assumed risk factors apply
With a multi-agent outbreak what are 4 main challenges
•New, emerging, exotic or unfamiliar diseases may not be considered on the differential diagnosis list
○ Fear, unsure of how to proceed, higher stakes if zoonotic
•Misinformation, uncertainty, fear
•Time pressure
•Multifactorial disease complexes (e.g. bovine respiratory disease)
Stevenson’s First Law of Veterinary Science list 2
- Multiple clinical signs in an individual animal tend to have a single aetiology (cause/agent).
- A single syndrome in a population (e.g. infertility, decreased milk production) tends to have multiple aetiologies.
With emergence animals diseases what are the 5 objectives
- determine whether containment and eradication can be achieved
- identify other infected and suspected premises (IPs and DCPs)
- intervene immediately to achieve the most appropriate outcome
- identify system failures that lead to the outbreak
- prevent re-occurrence
What are the 3 main things involved with outbreak investigations
- clinical & environmental & management investigations
- laboratory investigations
- epidemiological investigations
What are the 7 steps taken in an outbreak investigation
- Verify the outbreak and the diagnosis
- Define, identify and count cases
- Describe outbreak according to individual, place & time
- Formulate working hypotheses
- Evaluate and test hypotheses
- Implement control and preventive measures
- Communicate findings
Step 1 in outbreak investigation: Verify outbreak & diagnosis what are the 4 features
1) Are you really dealing with a problem -> observed cases greater than expected
2) what is the problem -> definition of disease
3) other factors -> severity, public relations, political considerations, risk of spread zoonoses
4) make a diagnosis -> list differentials, history, clinical examinations
Step 2: Define, identify & count cases what need to do and what includes
Case definition -> clinical laboratory diagnostic criteria, information about affected region. specification of time during which the outbreak occurred
What does a good case definition ensure and allow you to do
- ensures consistency of diagnoses across place and time (specific time period)
- allows to calculate appropriate measures of disease frequency -> allows you to rule cases in or out
- helps to monitor responses to interventions
Step 3 of outbreak investigation: Individual, place and time what is needed for this step and what is involved
ONCE HAVE CASE DEFINITION
•Describe outbreak according to the individual:
•Describe outbreak according to place:
•Describe outbreak according to time:
List Risk factors within animal management and environment categories
Animal
Age, sex, breed, size, pregnancy
Management
Feed, water, drug treatment, pen location, stocking rate
Environment
Weather, climate, wind speed, proximity to wildlife
How do you describe the outbreak according to individuals
- compare affected animals with unaffected animals
- what are the characteristics of affected and unaffected animals in terms of exposure to potential risk factors, age, production level, stage of production cycle?
calculate attack rates
How do you describe the outbreak according to place
- where are the affected and unaffected animals located?
- because different groups or pens of animals often have different levels of exposures a dose-response relationship exists for many aetiologic agents, this is an important set of clues
1) random
2) clustered - most common
3) regular
How do you describe outbreaks according to time
- when in calendar time did the problem actually begin?
- be careful of ‘pseudoepidemics’ caused by the onset of producer awareness of a more chronic problem or caused by a change in problem definition (or change of staff)
- epidemic curves
1) Propagating epidemic
2) Point source epidemic
Step 4 of an outbreak investigation: Develop hypotheses how is this down
- by combining information formed from step 1-3
- prioritise your hypotheses by their likelihood and focus your efforts on those with the highest priority until they are either more fully supported or are refuted
Step 5 of an outbreak investigation: Evaluate & test hypotheses how is this done
1) predict what you should find in other animals, such as test results or production effects, and proceed accordingly to test your predictions
2) concentrate on cases in the middle of the epidemic curve (not the beginning or the tail end)
3) recall bias -> ask subjects about what they would do on a ‘typical’ day (as opposed to trying to solicit precise details for the period prior to the onset of illness)
Step 6 of an outbreak investigation: Implement control & preventive measures what involved
- hands on, practical advise
- short-term, long-term, prevention
- monitor progress
Step 7 of an outbreak investigation: Communicate findings how done
As professional provide a written report of findings - case report
- gives client evidence of your efforts (justifies the bill)
- provides a valuable record for future reference
Define monitoring and surveillance
Monitoring
- Regularly observing the health of a population to observe tends in disease
Surveillance
- observing the health/ disease status of a population and taking action when disease reaches a specified threshold
List 6 reasons to conduct surveillance
- Observe health trends -
- Rapid detection -> exotic or new emerging disease important
- Identify new and emerging disease -> within another species or area
- International trade obligations -> very important in terms of
- Plan and identify disease priorities -> which areas you need to focus on,
- Evaluate interventions -> need to monitor and evaluate
List 7 components of a surveillance program
1) objectives
2) case definition
3) target population
4) timing of sampling intervals
5) data management
6) methods for data analysis and triggers for action
7) feedback and dissemination of results
Notifiable diseases in Victoria what is the vets legal responsibility and in what time frame with 1) exotic diseases 2) serious 3) less serious
- Your legal responsibility to report on SUSPICION within define timeframes
○ Exotic animal disease and anthrax notify immediately
○ Serious but not exotic disease within 12 hours
○ Less serious notifiable disease - within 7 days to the Chief Veterinary Officer
World organisation for animal health list some objectives and the reason for global reporting
- Transparency
- Scientific information
- International solidarity
- Promotion of veterinary science
- Sanitary safety
- Food safety and animal welfare
Global reporting - All about risk management
- Countries know what is happening within their trade partners
- Can take measures to ensure countries don’t get certain diseases -> biosecurity insurance system
What are the 3 main types of surveillance programs and types within
- Active vs passive -> data collection
- General vs targeted -> disease focus
- Structures vs unstructured -> sampling method
Types within
1) Risk-based surveillance
2) Sentinel “”
3) Participatory “”
4) Syndromic “”
5) Rumour “”
Describe active vs passive surveillance
ACTIVE - main users of info make active efforts to collect the data -> surveys, specific testing programs
PASSIVE - main users of info take no action to initiate the collection of data -> secondary data is used -> reporting of clinical cases
Describe general vs targeted surveillance
GENERAL - monitors animal population to detect undefined and unexpected -> scanning
TARGETED - answers specific question about define diseases or condition
Give an example of a targeted surveillance system within Australia
- National TSE Surveillance Program (NTSESP): Mad Cow Disease (BSE) & Scrapie
○ Objective to demonstrate Australia free/early detection
○ Sampling adult cattle or sheep with neurological signs
§ Testing focused on samples from downer cows ○ Want the brain stem for samples
What occurs with sentinel surveillance
- Animals are susceptible to the disease
- Place in the area that unsure whether the disease is present in an area
- If seroconvert or get disease then disease in present in that area
§ How many seroconvert give you prevalence - Victorian Arbovirus disease surveillance use sentinel chickens
Participatory disease surveillance what type of surveillance and what are the 4 main features
- Forms of active surveillance
○ Community consultation
○ Resource intensive
○ Syndromic case definition -> more relying on symptoms and signs rather than laboratory evaluations
○ More qualitive underlying of context -> disease, impacts, responses at community level
Describe Syndromic surveillance and what can achieve
- automated data acquisition and generation of statistical alerts allows one to monitor disease indicators in real-time or near real-time
- case definition based on signs & symptoms, not on laboratory confirmation
- in theory allows earlier detection of outbreaks/disease than with traditional methods
List 5 things that are needed to conduct surveillance
1) Owners recognising and reporting disease in animals (communication. Compensation)
2) Veterinary infrastructure: sampling and recording outcomes
3) Lab infrastructure for diagnose
4) Political will and resourcing
5) Cooperation (national and international)
When monitoring disease frequency what is the goal and how achieve
accurate information is the goal ○ Accuracy implies freedom from error: § There are only two types 1. Systematic Error (Bias) 2. Random Error (Chance effects) - Monitoring systems should be designed to eliminate as many sources of bias as possible and to contain random error within tolerable bounds
List 6 attributes of effective surveillance systems
1) Simplicity - if too difficult decrease motivation to report and act
2) Flexibility - to detect emerging (unknown) diseases
3) Sensitivity - a high detection rate
4) Specificity - a low number of false (positive) alarms
5) Representativeness
6) Timeliness – detect and allow for action within a timeframe relative to the infectious cycle of the disease so that further cases can be prevented
What does disease control strategies suggest
- Disease control implies doing something to
○ Reduce morbidity due to disease
OR
○ Reduce mortality of the disease
List and briefly describe 9 disease control strategies
1) movement restriction on people, equipment and animals
2) quarantine - incubation and infectious periods
3) Biosecurity - isolation and testing
4) Outbreak investigation and surveillance
5) vaccination - ring or blanket
6) treatment - reduce shedding
7) vector and reservoir control
8) humane slaughter and disposal
9) do nothing -> wait and see NOT EXOTIC DISEASES
What are the 4 things that aradication implies you are doing
○ completely remove the disease from the population
○ prevalence↓ → no transmission
○ prevalence↓ → no major problem, but transmission
○ cause a disease to become extinct in a region or zone within a country (elimination)
What are the 7 WHO criteria that suggest eradication is a good option
1) epidemiologic susceptibility
2) effective practical intervention available
3) demonstrated feasibility
4. Consensus on perceived community cost to benefit
5. Synergy with other interventions
6. Need for eradication rather than control
7. Compensation and financial support for short-term losses
What are the 3 general things need for effective surveillance and eradication programs
technical - availability of resources, economic and social factors
Social -> farmers have reasons not to report, communication to increase awareness, enhance biosecruity is important (needs to be cost effective)
