Combined set - HM Final exam material

Definition of evidence based veterinary medicine

The conscientious and judicious use of the current best evidence in the health care of individuals and populations.

From strongest to weakest in terms of evidence, name the difference types of studies

Strongest:

- Systemic reviews

- Meta analysis

- Blinded RCTs

- Cohort studies

- Case control studies

- Case series

- Single case reports

- Editorials, opinions, consensus reports

- Comparative animal research

- In vitro "test tube" research

What is a PICO question?

Patient

Intervention

Comparison

Outcome

What are the issues with using practice records to find answers?

- Quality

- Completeness

- Lack of clear case definition

- Lack of follow up of outcomes and complications

Three types of experiential papers

DESCRIPTIVE:

- Case reports

- Case series

- Review papers

Types of explanatory papers

OBSERVATIONAL (we watch):

- Cross sectional

- Cohort

- Case-control

INTERVENTION (we play):

- Experiments (controlled environment)

- Clinical trials (real world - uncontrolled environment)

Three criteria to be a confounding variable

1. Associated to the outcome

2. Associated with the exposure

3. Not a result of the outcome

What defines a cross sectional study?

Sampling participants regardless of exposure or outcome status

Ex. Enrol a bunch of people, ask them if they smoke, ask them if they have lung cancer

What defines a cohort study?

Sampling participants based on EXPOSURE of interest

Ex. Enrol a bunch of smokers (exposure of interest), see if they get lung cancer

What defines a case-control study?

Sampling participants based on OUTCOME of interest

Ex. Enrol a bunch of people with lung cancer and without, ask them if they smoked

What do most explanatory studies prove

Association, but weak evidence for causation (cohort best for arguing causation because they are outcome free at the beginning of the study so you can follow to see if participants develop the outcome)

When is a cohort study best?

For a rare exposure (ex. radiation exposure)

When is a case-control study best?

For a rare outcome (ex. specific type of brain cancer)

How do we prove cause?

- Look for a temporal relationship

- Determine strength of association (RR or OR)

- Look for a dose-response relationship

- Determine if there is biological plausibility in the relationship

- Look for consistency or repetition

- Rule out

- Reversal

What does an odds ratio measure?

The STRENGTH of an association in an observational study

If there is no association between an exposure and an outcome, what would the OR be?

1 (or the 95% confidence interval includes 1)

Interpret the following: "OR of 2.2"

"Animals with the disease were 2.2 x more likely to have experience the exposure, than animals not experienced the exposure"

What makes a good 'gold standard' test?

Perfectly classifies individuals

What is another term for prevalence?

Pre test probability

What is another term for post test probablity

Negative and positive predictive values

Desirable outcomes of vaccination

- Fewer animals get sick

- Animals don't get as sick

- Animals aren't sick for as long

- Fewer animals die

- The vaccinated animals produce protective antibodies

What is the danger of a historical control in vaccine trials

There is huge variation in year to year especially if you are looking at death as an outcome

What is a fatal flaw in vaccine trials?

Not exposing the groups or challenging the groups to the disease of interest

Difference between incidence and prevalence

Incidence: NEW cases

Prevalence: NEW and OLD cases

What does a systematic review/meta analysis do?

Summarizes all available literature in a replicable and unbiased manner

In a meta analysis results from studies are combined statistically

What are the steps in a systematic review

1. Define the research question

2. Comprehensive search for studies

3. Select relevant studies from the search

4. Collect data from relevant studies

5. Assess risk of bias in relevant studies

6. Synthesize the results

7. Present the results

8. Interpret the results

What are the different types of questions systematic reviews can answer?

- Prevalence or indidence (descriptive)

- Diagnostic test accuracy

- Etiology (risk factors, exposure)

- Intervention (treatment efficacy) PICO type questions

What is the aim of data extraction in a systematic review/meta analysis

Extract study data on study results and characteristics that influence the external applicability, internal validity and limitations of the study

What type of outcome is of interest in a systematic review/meta analysis

An operational outcome (something that can be MEASURED), not a conceptual outcome

Conceptual outcome: Intelligence

Operational outcome: IQ score

Difference between a systematic review and meta analysis

Systematic review: Synthesis of quantitiative data when meta analysis is not feasible or advisable

Meta analysis: Statistical combination of data from multiple studies

Control of disease is based on

Management/manipulation of risk factors

What are the limitations of a systematic review?

They take a very long to complete, so by the time they are published there is usually more/new information

What is a strength of systematic reviews?

Written by a team of experts and follow a pre determined protocol

Infectious versus contagious

Infectious: Caused by or capable of being communicated by a microorganism in body tissues - it is about the agent

Contagious: Capable of being transmitted from individual to individual (person to person; animal to animal); communicable - it is about transmission

What is the difference between infected verus diseased

Infected is talking about the agent (has the agent taken up residency)

Diseased is talking about the host

Necessary versus sufficient cause

Necessary: NEED to be there to cause the disease (ex. e. coli is needed to cause coliform mastitis)

Sufficient cause: Sufficient to cause the disease (ex. e. coli + dirty stalls + not being vaccinated)

What are the three approaches to infectious disease control

1. Eliminate

2. Prevent

3. Control (live with the disease but keep it manageable)

What is the Reed-Frost model?

Number of cases over time = Susceptible individuals (1-probability of adequate contact)

Therefore, we can decrease the number of susceptible individuals and the probability of adequate contact to decrease the number of cases over time

How do we eliminate disease?

- Eradicate all infected animals (test and cull - requires good screening tests)

and/or

- Immunize all animals at risk

Three ways we prevent disease?

- Reduce exposure to an agent (ex. calf hutches)

and/or

- Reduce stress of the host

and/or

- Enhance host resistance (immune system, genetics, nutrition)

Passive versus active immunity

Passive: From mother (ex. in utero or colostrum)

Active: Vaccination

How do we control disease?

Strategic or routine medication (metaphylaxis, medicated feed, flea prevention, heart worm prevention)

What is risk?

The probability AND magnitude of an outcome

How do we decide how to deal with an infectious disease?

Depends on:

- Risk

- Magnitude of the problem

- Economic loss

- Zoonotic potential

What is biosecurity?

The protection of people, animals and ecological systems against disease and other biological threats - stop movement of things that cause disease!

How is biosecurity achieved?

Through systems that aim to protect public health, animal and plant industries, and the environment from entry, establishment and spread of unwanted pests and diseases

Three things we try to control with biosecurity?

1. Foreign animal diseases (emergencies)

2. Endemic diseases (day to day measures)

3. Zoonotic diseases

What is disease surveillance?

The continuous investigation of a given population to detect the occurrence of disease for control purposes, which may involve testing part of a population (can be clinical or subclinical disease)

Active disease surveillance

Initiate an active data collection mechanism

Ex. Testing all down cows for BSE

Passive disease surveillance

Use data derived from other sources

Ex. Accessing samples/data coming in through diagnostic labs

What is a challenge when it comes to day to day biosecurity?

- Programs are built on 'biological plausibility' not 'evidence'

- Measuring effectiveness of programs is hard

- Lack of a challenge

All perils versus targeted biosecuirty

All perils

- Difficult

- Not defined

- Likely not cost-effective

- Might not be maintained

- Ex. Swine

Targeted:

- Is possible

- Focused

- Can be cost-effected

- Can be maintained

- Ex. Cattle

How do you build a targeted biosecurity program?

- Identify disease(s) of concern (and why?)

- Determine current prevalence of disease

- Limit animal movements within and between farms

- Implement disease control program

- Monitor compliance with program

- Renew annually

What is a RAMP?

Risk assessment management plan

Tool for producers and veterinarians to assess individual farms and select three prevention strategies

Things to do to manage zoonotic risk?

- Prevent infectious disease

- Limit disease spread on and between farms

- Treat or cull sick animals

- Collect meat and milk hygienically

- Pasteurize all dairy products

- Maintain potable water

- Wash hands

Why is randomization in studies important?

Decreases bias by equally distributing known and unknown confounding variables

Two filarid nematodes of note?

1. Dirofilaria immitis (heartworm) - dogs and cats

2. Acanthocheilonema (dipealonema) reconditum - dogs - non pathogenic round worm

Five step life cycle of a filarid nematodes?

1. A mosquito ingests an infected animal, ingesting heart worm mcrofilariae (pre-L1 = early stage larva)

2. Microfilariae mature into heart worm larvae (L3) inside the mosquito for 10-14 days

3. Infected mosquito bites a dog transmitting the larvae (L3)

4. Larvae enter the dog's blood stream, migrate to the heart and lungs (70 days), and grow to a foot long and become sexually mature

5. Adult heart worms can live within the heart and lungs for 5-7 years

Definitive hosts of heart worm?

- Dogs

- Cats

- Ferrets

- Wild canids

- People (rare)

Predilection site of heart worm?

Right caudal lobar artery

Intermediate host of heart worm?

Mosquitoes

How long does it take to detect infection once a dog gets heart worm (aka the pre patent period)?

7-9 months

How long does it take for larvae to mature inside a mosquito?

18˚C = 29 days

26˚C = 12 days

30˚C = 8 days

*the warmer it is, the shorter it takes for the larvae to mature inside the mosquito*

In Southern Ontario, the earliest time for transmission is?

June 1st

In Southern Ontario, the latest time for transmission is?

October 8th

What is the prevalence of heart worm in Ontario?

0.13% (with 4/5 of cases occurring south of 401/402/403 highways)

What % of heart worm positive dogs show clinical signs?

12%

Of the heart worm positive dogs in Ontario, what % acquired heart worm in Ontario?

51%

What is the risk of heart worm in cats?

1/10 the risk it is in dogs

- often very few parasites

- adult parasites live 1-2 years

- microfilaremic for 1 month

Why does heart worm cause disease?

- Obstructs blood flow which cause pulmonary hypertension, right ventricular hypertrophy and heart failure due to the accumulation of fluid

- Parasites excrete products

- Dead/dying parasites cause pulmonary embolism

What do clinical signs of heart worm infection depend on?

- Number of adult parasites

- Size of dog (the bigger dog, the more they can handle)

- Exercise level of the dog

Clinical signs of heart worm infection?

Usual signs:

- Exercise intolerance

- Coughing

- Dyspnea

Severe signs:

- Loss of condition

- Hemoptysis

- Ascites

- Hydrothorax

Heart worm detection tests?

Examine blood:

- Wet blood film

- Stained blood smear

- Concentration methods

- Antigen tests (primarily used method)

- DNA test (not useful)

- Thoracic radiograph (look for right ventricle hypertrophy and tortuosity)

- Ultrasound (only useful if large number of parasites)

- Clinical signs (non specific)

- History (location, travel history, preventative medication, compliance)

If a dog has a LARGE burden of heart worm, what should you do before treatment?

Remove as much as you can surgically - treatment will kill heart worm and if there is a large burden there they will cause an embolism when they die

Screening versus diagnostic tests

Screening: For healthy populations

Diagnostic: For populations with clinical signs

Sensitivity (Sn)

The proportion of disease animals that test positive

- High Sn = SnOUT (can be confident in ruling disease out)

- Sensitivity is related to false negatives (higher the Sn, the lower the FN)

- Measure of how good a test is

Specificity (Sp)

The proportion of healthy animals that test negative

- High Sp = SpIN (can be confident in ruling disease in)

- Specificity if related to false positives (higher the Sp, the lower the FP)

- Measure of how good a test is

What does the prevalence of heart worm depend on?

- Location

- Travel history

- If the animal is on heart worm prevention

Why do we screen dogs in Ontario for heart worm?

1. To confirm the dog is heart worm free before putting it on prevention medicine (rule out), because the package says so - safety reasons that no longer exist for most prevention. If you do this and the dog gets heart worm while on prevention, the company will pay for treatment.

2. To remove reservoir of infection at the beginning of each transmission season

3. Annual testing is recommended by the American Heart worm Association

Give four reasons for not carrying out heart worm tests on dogs in southern Ontario

1. Southern Ontario is very low prevalence, therefore the likelihood of infection is low if the dog has not traveled

2. Low prevalence reduces the info gathered by the test and actually increases the risk of a false positive

3. If an animal is on heat worm prevention with good compliance, the likelihood is even lower that the animal could be infected, therefore the test doesn't provide any useful information

4. Given the above, it could be a waste of a client's money, in low income families especially

Give two reasons for carrying out an annual heart worm test on a dog in Southwestern Ontario

1. To screen for heart worm before starting heart worm prevention for dogs that haven't on heart prevention in the past (or didn't have good compliance)

2. For dogs that have travelled recently or regularly travel to higher prevalence areas and are not on heart worm prevention

Positive predictive value (PPV)

Given the dog has a positive test, what is the likelihood that this test result is true?

- Measure of how useful a test is

Negative predictive value (NPV)

Given that the dog has a negative test, what is the likelihood that this test result is true?

- Measure of how useful a test is

How do you calculate what you learned from a test in the case of a positive test result and a negative test result ?

Positive test result:

= PPV - prevalence

Negative test result:

= NPV - (100-prevalence)

What does prevalence effect?

Positive and negative predictive values

Why are heart worm tests not always useful?

Heart worm positive tests really only are 'certain' for populations where there is a high prevalence of heart worm

If you get a positive test for heart worm (without clinical signs), what should you do?

- Repeat the test (test may not have been done correctly)

- Send a blood sample to a different lab that before and do not tell them that the first sample was positive (another positive confirms the first test was done correctly, but does not confirm the animal is disease positive)

- Can look for microfilaria in blood

- If no signs and still positive, re-test in 6 months, make sure they are on topical monthly prevention which is safe in heart worm positive dogs

Where in the heart worm life cycle do heart worm preventions work?

The vast majority kill adults, the stage of the heart work that has been inoculated. You are killing the early stages of infection so that you never get adult parasites.

Minimum age to test a dog for heart worm?

7-9 months because that is how long it takes the parasite to mature

What is an outbreak

A cluster of events related in time and space with a markedly higher than expected incidence

Goals of an outbreak investigation

1. Reduce or stop the epidemic

2. Determine the reason(s)

3. Prevent a recurrence or a new outbreak with related causes

What are key determinants

Factors controllable by management that can be changed to affect disease rate - as vets we try to identify these and help the farmer make changes

When defining a problem, what should you do?

Come up with a specific case definition (ex. showing at least 3 of the following signs...) and try to dichotomize it

Ex. Yes they have mastitis, no they do not have mastitis (based on a strict/clear definition)

When gathering data, what should you do?

- Talk to everyone on farm (try to talk to them at the same time)

- Be non-judgemental

- Ask what their hypothesis is

- Observe the affected and unaffected animals

- Try to get an accurate time line

- Get info on policy (what should be happening) AND practice (what is actually happening)

- PMs are very useful tools

What is a 'method of agreement'?

What do most or all affected animals have in common?

What is 'method of difference'?

What is difference between affected and unaffected animals

How many samples should you collect per group

5-10

What samples can we take in an outbreak investigation

1. Blood (suck)

2. Poop (scoop)

3. Orficies (swab)

4. Necropsies (slice)

5. Feed (spoon)

6. Water (siphon)

And remember to identify

When our number of samples/observations are low, what happens to the confidence interval?

It will be wide (aka precision is low)

Main tasks in an outbreak investigation

1. Define the problem

2. Define the groups

3. Collect the samples

4. Establish a working diagnosis

5. Take action

6. Do the follow up