VETS1018

What is the attribution fraction?

% reduction or increase in risk of outcome in the exposed (or treated) group attributable to the exposure (or treatment). AF- relative measure RISK REDUCTION AF for case control studies: AF = (OR-1)/OR

What does SOAP stand for? (SOAP sheets)

- "S": Subjective (history) - "O": Objective (Physical exam) - "A": Assessment (problem list) - "P": Plan (Plan for diagnostics or therapy)

What are the causal models?

- Host-agent-environment triangle - component cause model - causal diagrams

What does a high temperature mean?

- Infection, inflammation or cancer - heat stress, scared or agitated

What is the confidence interval used for when there is more than one group?

- It is used to compare groups - CI for a comparison of groups represents the 'closeness' of the difference between groups to the difference between groups in the population - It is used to make a decision to REJECT OR ACCEPT THE NULL HYPOTHESIS

What things can you look for when feeling pulses?

- determine if pulses are synchronous - determine the pulse quality (poor, fair, strong (normal), brisk, bounding)

What should you NOT do in the case of seizures?

- do NOT put your hands in an animal's mouth! They do not swallow their tongues - do not hit, spray water on, or otherwise try to startle the pet out of a seizure. Seizure will end when it ends.

What is a narrative review? *

- no specific methods are used for selecting sources of evidence - more susceptible to unrepresentative reporting of evidence

How do you test CRT? (Capillary refill time)

- press finger onto mucous membrane - Release and immediately move fingers out of the way - count how long it takes for mucous membrane to regain colour - normal is 1-2 seconds

How does one randomise a control (clinical) trial

-researcher allocates animals into treatment groups based on random allocation -allocation to groups is blinded

What are the important legislations for vets in QLD?

1. Acts of veterinary science (Veterinary surgeons act 1936, Veterinary surgeons regulations 2002) 2. Care of animals (Animal care and protection act 2001 3. Use of drugs (Health act 1937, Health(drugs and poisons) regulation 1996 4. Use of radiation sources (Radiation safety act 1999, radiation safety regulation 2010) 5. Work place health and safety (Workplace health and safety Act 2011) 6. Care of wildlife (Nature Conservation act 1992, Nature conservation (wildlife management) regulation 2006)

If an association is shown in a study, what must it be due to?

1. Causation 2. Confounding 3. Bias 4. Chance variation

What are non-causal explanations in studies? / internal validity?

1. Confounding 2. Bias 3. Chance variation

What does a physical examination in diagnostic procedures consist of?

1. Distant inspection 2. Close examination- including vital signs, TPR, mucous membranes, pain etc.

What are the steps in evidence based medicine? QEAAO

1. Forumlate an answerable Question 2. Search for Evidence to answer the question 3. Critically Appraise the evidence (study design, epidemiological measures, causality, bias and confounding) 4. Apply the answer to your patient 5. Audit the Outcome

What are the steps in hypothesis testing?

1. From the observed data, a TEST STATISTIC is calculated Choosing a test statistic is a major topic in statistics - Each statistic has a probability distribution (e.g. t distribution, z distribution, F distribution etc.) 2. The probability (p‐value) of observing a test statistic as large or larger than that observed, if the H0 (null hypothesis is true is calculated - In reality statistical software provides the value of test statistic and the associated probability 3. The p‐value is compared to a cut‐off termed the 'level of significance' (called 'alpha') - Level of significance should be small, because we don't want to reject the null hypothesis when it is true (e.g. 0.05, 0.01, 0.001

What are the core communication skills of effective practitoners?

1. Introductions & establishing agenda 2. Non-verbal communication awareness 3. Using open and closed-ended enquiry 4. Pausing 5. Reflective listening 6. Demonstrating empathy 7. Great explanations 8. Sign-posting

What are the causal guidelines?

1. Koch's postulates 2. Hill's criteria 3. Evan's rules

How do we make diagnostic decisions?

1. Organise clues to come up with DIFFERENTIALS ( or differential diagnoses- ideas on what might be wrong with our patients) 2. We come up with the best DIAGNOSTIC TESTS to run to confirm or deny our suspicions

What is the general structure of the health/clinical examination?

1. Patient history 2. Patient examination a. Distant inspection b. Physical examination

What are Hill's criteria ?

1. Strength 2. Consistency 3. Specificity 4. Temporality 5. Biological gradient 6. Plausibility 7. Coherence 8. Experiment 9. Analogy

What are Halsted's principles of surgery?

1. Strict asepsis 2. Gentle tissue handling 3. Preservation of blood supply 4. Adequate haemostasis 5. Obliteration of dead space 6. No tension on sutures 7. Careful approximation of tissues 8. Debridement of dead tissue

What are Koch's postulates?

1. The agent must be present in every case of the disease. 2. The agent has to be isolated and grown in pure culture. Robert Koch (1843-1910) 3. The agent must cause specific disease. 4. Once isolated that agent must reproduce the disease and must be recovered from experimentally infected animals. PROBLEMS: Ignored ENVIRONMENTAL INFLUENCE, host factors, mixed infection, non-infectious diseases etc. 'single cause paradigm'

What are Evan's rules?

1. The proportion of individuals with disease should be higher in those exposed than those not 2. Exposure to cause should be more common in those with the disease than without 3. Number of new cases should be higher in those exposed than those not (in prospective studies) 4. Disease should follow the supposed cause 5. A spectrum of host responses from mild to severe should follow exposure 6. A host response (e.g. antibodies) should appear following exposure in those exposed but not in those not exposed 7. Experimental reproduction should occur more often in those exposed than those not exposed 8. Elimination of the cause should decrease the frequency of occurrence 9. Prevention/modification of the host response should decrease the disease 10. All relationships should be biologically and epidemiologically credible

What are the three ways of statistical testing?

1. hypothesis testing and p values 2. confidence interval approach to hypothesis testing 3. statistical tests- analysis of categorical data using chi squares - overview of other types of analysis

What are the clues you would collect to make a diagnosis?

1. signalment 2. history 3. Physical examination 4. Diagnostic test results 5. collective experience of others 6. Evidence from the literature 7. Our own instincts and experience

What is the diagnostic approach?

1. take CLUES 2. to make diagnosis 3. to make a treatment plan

What level of dehydration is an animal at if they have a dry mms, significant skin tenting, and a rapid heart rate?

10-11%

What is the normal bovine respiration rate?

10-40 br/m

what level of dehydration is an animal at if they have a dry mms, marked skin tent, eyes dull and sunken, signs of shock?

12% and higher!

What is the normal heart rate for cats?

120-180bpm

What is the normal canine respiration rate?

16-34 br/m

What is the normal feline respiration rate?

20-34br/m

What is the normal heart rate for horses?

24-44bpm

What is the normal equine rectal temperature?

37- 38.5 deg celcius

What is the normal canine and feline rectal temperature?

37.6- 39 deg celcius

What is the normal bovine rectal temperature?

38-39 deg celcius

What is the normal heart rate for cows?

48-84bpm

What level of dehydration is an animal at if they have a dry mucous membrane but are otherwise okay?

5% dehydration

What level of dehydration is an animal at if they have a dry mms and skin tenting?

7-9%

What is the normal heart rate for dogs?

70-120 bpm

What is the normal equine respiration rate?

8-20 br/m

How do you interpret this: 95% CI for RD: -0.123 to 0.303 (or -12.3% to 30.3%)

95% Confidence interval covers 0, and is RISK DIFFERENCE, therefore we conclude it is likely that the difference between groups occurred by chance - we do not reject the null hypothesis

How do you interpret this: 95% CI for % OR = 0.51 - 5.50

95% confidence interval covers 1, and is for RATIO, therefore we can conclude that it is likely that the difference between groups occurred by chance- we do not reject the null hypothesis

Explain the concept of hypothesis testing and interpret the meaning of a p‐value ****

??? help please

What is a case series?

A descriptive study that shows that a case can happen repeatedly. (whereas a case report describes something that happened one or a few times) - may provide valuable information about the prognosis of the condition - The describe: the usual clinical course of the condition of interest, common features amongst multiple cases, patterns of variability amongst cases eg. 123 cases of rodanticide poisoning in dogs. objective: to identify dogs with poisoning, to identity whether detected concentrations of the anticoagulants correlated with severity of clinical signs for dogs with poisoning and identify the most common disease processes present and the prognosis for those with negative screens eg. BSE and CJD (dementia)- describes 10 cases of CJD which occurred at a young age, showed distinctive features and showed extensive prions throughout brain, similar features to BSE

What is a sufficient cause?

A set of components that together is producing the disease- represents the whole pie

What is a confounder?

A variable is a confounder if: 1. It is causally associated with the OUTCOME and 2. It is non-causally (or causally) associated with the EXPOSURE, and 3. It and the exposure variable are on two separate causal pathways to the outcome

What is an advantage of an observational analytic study?

A wide range of hypotheses can be tested

What are the advantages and disadvantages of a cohort study?

ADVANTAGES- Provides disease incidence estimates - Can be used to study multiple effects (outcomes) of one exposure eg. smoking - Can be used to study rare exposures - Less recall bias as in case-control studies - Can be adjusted for loss-to-follow-up - can be used to demonstrate temporality: exposure occurs before the outcome Calculation of risk ratio, odds ratio, and incidence measures DISADVANTAGES - often require long duration (potential for confounding effects) - In the case of rare diseases, large numbers are necessary - losses of follow up often occur due to death or withdrawal - Can be expensive and resource intensive - groups may differ at beginning of study (problem of confounding)

What are the advantages and disadvantages of case-control studies?

ADVANTAGES: - well suited for: rare diseases, diseases with a long incubation period, conditions developing over a long period of time - quick to conduct - relative low cost - existing records can be used - can study multiple different risk factors (EXPOSURES) to the OUTCOME - require fewer subjects compared to other study types DISADVANTAGES - cannot provide info on disease FREQUENCY in population - not suitable for studies on rare exposures - Data collection relies on past records quality - Representativeness of the selection of animals to be studied is difficult to guarantee, especially for the control group - many sources of potential bias - calculation of odds ratio only

What are measures of impact in measures of association useful for?

Absolute risk

What are ABCs?

Airways, breathing, circulation

What is signalment?

An accurate description of the patient including: Age Gener Entire or desexed Breed Species eg. "Ruby Royal " is a 3 year old Hanoverian horse mare used for dressage

When do you start your distance inspection?

As soon as you greet the patient in the waiting room the moment you enter the property

What do we look at in mucous membranes?

Assess the colour, touch for moisture/tackiness, press and release for capillary refill time (CRT)

What are the two kinds of knowledge in veterinary medicine?

Background (broad, more fundamental) and foreground (specific) knowledge

What is BAR?

Bright, Alert, Responsive

How do we tell which kind of observational analytic study the study is?

By when the outcomes were determined. Before exposure was determined- case-control study (retrospective) At the same time as exposure- cross-sectional study Some time after exposure- cohort study (prospective)

What is the confidence interval used for when there is only one group?

CI for one group represents the 'closeness' of the sample mean to population mean

what does CRT stand for?

Capilllary refill time

What is a clinical trial?

Carried out under 'real-world' conditions Used for the evaluation of therapeutic or preventative drugs or interventions in the 'field' Animals exposed to all known and unknown factors of environment Animals followed overtime to measure one or more outcomes eg.

What are the kinds of descriptive studies?

Case report, Case series, surveys

What should you examine the chest for?

Check thoracic spine, ribs, sternum heart sounds (listen over apex beat of heart- palpate) - two heart sounds, 'lup dup' should be regular or follow respiration murmus lung sounds- wheezing, crackles, friction

What statistical test would we use for the following data?: • Both the outcome (e.g. Otitis externa) and exposure (e.g. treatment) are binary data. • Observations are independent of each other (i.e. no dogs were used twice, no two dogs came from the same household • The numbers of observations (e.g. dogs cured and not cured in each treatment group) are reasonably high.

Chi square test 1

What are component causes in the component-cause causal model?

Component causes are 'equal parts' in producing the effect (each piece of the pie is called a component cause)

What are systematic errors which occur in studies?

Confounding and bias

What is the component-cause model?

Consider causation by looking at how many pieces of pie are needed to have a full pie- once the pie is full, the disease occurs Component causes, sufficient causes necessary causes

What are the two major kinds of study designs?

Descriptive and analytic

What is the difference between descriptive and analytic?

Descriptive- aim is to describe a population Analytic- aim is to quantity a RELATIONSHIP between factors

What can you look for on the nose?

Discarge, moisture level,

What are the kinds of analytic study designs?

Experimental and Observational

What is the host-agent-environment triangle?

Factors that effect risk of disease can be partitioned into three groups: host, agent, environment LIMITS: does not account for temporality, does not show how factors might be interlinked, overemphasis on agent factors (not appropriate for toxins/non-infectious agents)

What is GDV?

Gastric Dilation and Volvulus- large deep chested breeds of dogs, acute emergency, require immediate surgery

What do you look for in a distant inspection?

General appearance, demeanour, posture

What do descriptive statistics include?

Graphical presentation of data, measures of CENTRAL TENDENCY (mean, median, mode), measures of SPREA (dispersion), standard error of the mean and confidence interval

What is variance?

How close each observed data value is to the mean sum of the squares of the differences of each of the n values from the mean, divided by degrees of freedom (n-1)

What is external validity?

How valid the study is to the other population

What are measures of strength in measures of association useful for?

Identifying risk factors, but they do not provide information on absolute risk

How do you interpret Risk Ratio? (RR)

If the risk ratio is ABOVE 1, the outcome is more likely if exposure occurred eg. cure is more likely if treatment occurred OR treatment increases risk/probability of cure eg. exposure to risk factor increase risk of disease If the risk ratio is 1, the exposure makes no difference to the outcome If the risk ratio is BELOW 1, the outcome is less likely if exposure occurred. Exposure reduces risk of disease Treatment reduces risk of cure OR treatment with new treatment is more successful than treatment with old treatment

What is the difference between an observational and experimental analytical study?

In an observational study, the investigator did not allocate the treatments/interventions, whereas in an experimental study, the investigator allocated the treatments/interventions. EG Observational- Are dogs with floppy ears more likely to get ear infections than those with pricked ears? Experimental- Does palladia cure mast cell tumours in dogs?

What does a low heart rate mean?

In shock or very ill, anaesthesia or sedation, athletic

What is the difference between an in-vitro and in-vivo diagnostic test?

In-vivo: observations and imaging (radiographs, ultrasound, endoscopy etc.), therapeutic trials In-vitro: Analysis of samples collected (tissue and blood samples, fluid)

How can you reduce chance variation?

Increase sample size

What is internal validity?

It describes the truthfulness of inferences about a study population. Relates to non-causal explanations

What is a case report?

It is a descriptive study design It describes some 'newsworthy' clinical occurrence based on ONE OR A FEW clinical cases. Eg. A case report reporting the infection of a veterinarian with hendra virus, speculating that it may have been caused by poor PPE, to warn other clinicians.

What is a survey?

It is a descriptive study design which quantifies the frequency and distribution of selected outcomes in defined populations - can provide hypotheses that lead to later analytical study - often info is collected from pre-existing sources (eg. birth, death or disease registries) eg. a survey on the occurrence of malignant fever in small ruminants in Northern Japan eg. --- conducted a descriptive epidemiological study on injuries in a part of America found a high prevalence of interpersonal injury in this area. IF INFO ON BOTH OUTCOME AND EXPOSURE ARE PROVIDED, it is not a survey, but a cross sectional study!

What is a laboratory trial?

It is carried out under strict laboratory conditions Provides good evidence of causation Often conducted in 'other' species Advantage: Complete control of experimental conditions Disadvantage: Relevance to 'real-world' conditions is doubtful eg. effects of anaesthesia on cognitive performance in a mouse model of alzeihmers disease

What is an advantage of an experimental analytic study?

It is possible to control potential confounders through the process of randomisation

What is a necessary cause?

It must be present for the disease to occur

What does the p value tell us?

It tells us whether the null hypothesis is likely or unlikely - can prove or disprove the null hypothesis Its size relationship with alpha tells us if the correlation is statistically relevant

What are the tests used for multivariable survival time data?

Kaplan-Meier curves Cox proportional Hazards regression

What can wheezing in lung sounds mean?

Large airway disease

What are the tests used for multivariable analysis?

Linear regression, logistic regression, kaplan-leier curves, Cox proportional Hazards regression

What is the test used for dichotomous (yes/no) outcome data which are multivariable?

Logistic regression

Interpret the confidence interval when the mean in one group is 10, and the mean in the other group is 6

Mean difference in weight = 10- 4 = 6 Mean (95% CI): 4.0 (2.04-5.96)- confidence interval found using t value Since the 95% confidence interval (2.0-5.96) DOES NOT COVER 0, we conclude it is unlikely that there is no difference between groups - we reject the null hypothesis

What are some negatives and positives of cross-sectional studies?

NEGATIVES- -only a 'snapshot' in time - difficult to determine if exposure occurred before outcome difficult to investigate cause-effect relationships - unsuited to the study of rare diseases (because a large sample size is needed to get enough cases) - unsuited to study of diseases with a short duration - incidence in exposed and unexposed individuals cannot be estimated - potential of confounding between risk factors Positives - When a representative sample is collected multiple parameters can be estimated (prevalence, proportion of exposed and unexposed, population structure) - Quick to conduct - moderate cost - allows to study multiple risk factors of disease - current records can be used occasionally calculation of risk and odds ratios

What does NAD mean?

No Abnormality detected

Does RR give an indication about absolute risk?

No! It shows us what effect the exposure has on the outcome (if it makes it more or less likely)

Does statistical significance always point towards clinical significance?

No- look at the example with two groups of cows. 200 from each pen were weighed, and the difference was found to be statistically relevant (p= 0.02). However, it was simply that the mean weight varied by 0.5kg

Can selection bias and miscalculation bias be controlled using analytical techniques?

No.

What are some other examples/types of selection bias?

Non-response bias (owner's nonresponse or refusal to participate in study) Lost-follow-up bias (owner's withdraw animals from study) Exclusion bias (animals develop health problems unrelated to trial and must be withdrawn) Survival bias (eg. introduction of insulin has lead to an increase in life span of those with diabetes, therefore the disease appears more prevalent)

How do you check respiration rate?

Number of breaths per minute (br/m) Watch the chest rise and fall Count the number of breaths in 15 seconds Multiply by 4 to get br/m Often the first part of your examination (before you even touch the patient, distant examination, while patient is in cage, race, crush, carrier or pen)

What do all observational analytical studies have?

P- Population E- Exposure C- Comparison O- Outcome

What do experimental analytical studies have?

P- Population I- Intervention C- Comparison O- Outcome

What do all study designs have?

P- a defined POPULATION O- OUTCOMES that are measured

What does PO/os, IV, SC/SQ and IM mean in regards to administration of drugs?

PO/ per os- oral IV- Intravenous SC/SQ- Subcutaneous IM- Intramuscular

How do you detect pain?

Palpation (feeling with fingers) and strategic questions, behavioural changes etc.

What tests do you use if data is continuously and normally distributed?

Parametric tests- t‐test, ANOVA, regression

What can observational studies be divided into?

Prospective or retrospective studies

What is the difference between prospective and retrospective?

Prospective-> The exposure status is determined in the present Retrospective-> look back in time at exposure status

What are the two major types of variables?

Qualitative and quantitative

What are the types of miscalculation bias?

Recall bias- cases are more likely to recall past exposure accurately Interview bias- Problem when interviewers are judgemental to the hypothesis under investigation Obsequiousness bias- (the clever Hans effect)- occurs when subjejcts systematically alter their responses in the direction they perceive to be desired by the investigator Differential- errors occur to a greater extent in one group non-differential- errors in one group independent of the other group

What is a systemic review? *

Relevant studies are systematically identified, appraised and summarised using explicit and reproducible methods (e.g. search terms, data bases)

When do you use a linear regression?

Requirements - Outcome continuous - One (simple) or more explanatory variables (multiple linear progression) - explanatory variables can be continuous or categorical - Outcome variable must be normally distributed - linear relationship between outcome and explanatory variables Time = race distance + sex

When do you use a logistic regression?

Requirements - outcome dichotomous/binary: 2 OUTCOMES - one or more explanatory variables - Explanatory variables can be continuous or categorical - Outcome variable must be normally distributed - Most common analysis method in the biomedical sciences - Outcome of analysis is expressed as Odds ratio, can be converted to probability eg. odds for heart disease (odds) heart disease (yes/no)= age (years) + family hisotory (yes/no) + smoking (years) Example: survival in dogs with soft-tissue sarcoma. Stratified by mitotic index (MI)

When do you use the Kaplan Meirer curve with log rank test?

Requirements -Outcome: time -one or more groups - used to describe and compare survival times - censored data ('loss' of animals) are included in calculations - Log-Rank test compares whether survival is different between groups

when do you use a Cox Proportional Hazards Regression?

Requirements: - used to compare survival times for more explanatory variables - A bit like logistic regression but uses additional data because it takes survival time into account - Helps to distinguish individual contributions of variables on survival - Outcome of analysis is expressed as a hazard ratio, which is interpreted like an odds ratio Example: Death from heart disease (yes/no) = age(years) + family history (yes/no) + smoking (years)

What is a meta-analysis? *

Results from all relevant studies are analysed statistically to give an overall summary result.

What is bias?

Results in observed effect estimates which differ from those which truly exist in the target population

What is the sample error of the mean?

SEM is an estimate of the average variation of the sample mean how close the sample mean is to the actual mean allows us to calculate confidence intervals SEM= Standard deviation/ squareroot(number samples)

How do you calculate SEM? (standard error of the mean)

SEM= standard deivaition/ squareroot(number sampled)

What are SOPs?

Safe Operating Procedures

How do you interpret the Odds ratio?

Say the odds ratio is 1.63, this means that the outcome is 1.63 times more likely to occur with exposure compared to without exposure Odds ratio of 1- Odds of cure/disease are similar in both groups Odds ratio larger than 1- Odds of cure/disease were more likely in the exposed group than the unexposed group. Odds ratio smaller than 1- Odds of cure/diseases were more likely in the unexposed group

What is a cohort study?

Selection based on EXPOSURE Selection of 2 groups of NON-DISEASED ANIMALS- one group EXPOSED to factor (which may be a risk factor for disease), another group is NOT EXPOSED They are followed over time and their change in disease status (outcome) is recorded (prospective) can be retrospective as well- use of population with pre-existing, historical data of exposure, evaluate what happened to unexposed vs. exposed populations over time.

What are the two types of bias?

Selection bias- refers to the procedures used to select units that are included in the study Misclassification bias- refers to the measurements of outcome or exposure after units were selected. (can be differential or non-differential)

What is a case-control (observational analytic) study?

Selection of a group with the disease and a group without the disease (OUTCOME) - selection based on OUTCOME comparison of the frequency of the EXPOSURE factors in the cases with that in the controls usually conducted RETROSPECTIVELY, but can be conducted PROSPECTIVELY (in which cases are enrolled as they occur) eg. brain cancer and cell phone use - cases and controls were found- people with and without brain cancer - they were then quizzed about their cell phone usage in the past. subjects are selected FIRSt based on their OUTCOME status. Cases represent an unknown proportion of all cases from the source population- therefore no prevalence or relative risk can be calculated.

What does low temperature?

Shock or very ill; anaesthesia young animals in a cool environment

What does a dry mucous membrane mean?

Sign of severe dehydration

What is the standard deviation?

Square root of the sample variance it estimates the average variation of the n values from the mean tells us how much VARIABILITY can be expected among individuals

What are measures of STRENGTH? what are measures of IMPACT?

Strength- risk ratio, odds ratio Impact- risk difference, attribution fraction, number needed to treat

What are the patient vital signs?

TPR (Temperature, Pulse/heart rate, Respiration- pattern and rate) Mucous membrane characteristics (colour, moisture, capillary refill time) Pain?

What is a causal diagram?

Take component causes and display them as a pathway diagram.

What is RD (Risk difference)?

The % treatment success for the new treatment less than that for the previous treatment. Risk of positive outcome - risk of negative outcome

Describe how a chi‐square test is derived on data from a 2*2 table

The chi-square test uses these differences between observed and expected numbers to calculate the chi-square statistic. 1. The expected number of dogs in each of the four cells is calculated under the null hypothesis that both treatments have the same % treatment failures. eg. 75% of each group are cured 2. The chi-square test then uses these differences between observed and expected numbers to calculate the chi-square statistic. Chi square = sum of all cells [(observed-expected)2/expected] eg. Chi-square statistic = = (28-26.25)2/26.25 + (32-33.75)2/33.75 + (7-8.75)2/8.75 + (13-11.25)2/11.25 = 0.830 * Note that with greater differences in the observed treatment successes between groups, the chi-square statistic increases. Assuming the null hypothesis is true, the probability of observing a chi-square 0.09 statistic as large or larger than 0.830 is 0.362 or 36.2% (p-value- found with technology) As the p-value is not low (0.362), the observed data could easily have occurred under the null hypothesis and we can not reject the null hypothesis. In other words, this particular study provides no justification for concluding that the % treatment successes are higher amongst dogs treated with the new product.

What does it mean if the p value is smaller than alpha?

The null hypothesis is excluded and the difference is 'statistically significant'

What does it mean if the p value is large?

The null hypothesis is probably true

What does it mean if the p value is small?

The null hypothesis is unlikely to be true

What is the number needed to treat?

The number needed to be treated before one additional case of the disease is cured/prevented.

Interpret a p value of 0.06 when alpha is 0.05

The probability of obtaining a t-statistic of 2.98 or larger, if thenull hypothesis is true, is 0.006 The probability of getting this result is only 0.6% if the null hypothesis is true. Hence we reject HO and say the difference in means is 'statistically significant'.

What is risk?

The probability that an outcome will happen - can be the probability of cure if treated with a certain drug - can be the risk/probability of disease if exposed to a risk factor

What is the confidence interval and how do you find and interpret it?

The range in which we are confident that the true mean will fall within. What would be the variability of the mean in other samples? - use the 95% confidence interval to work this out. - it should fall in the interval we find (we are 95% sure it will) - "the interval from ... to ... has a 95% chance to contain the true population mean. - "Given repeated sampling and calculation of 95% confidence intervals for each sample estimate, 95% of them will include the true population mean." CI= mean +/- t x SEM end up with a range from ... to ... t- critical vale, depends on the number of samples and is obtained from software eg. for sample of 10, "t" is 2.26

What is the odds ratio? (OR)

The ratio between the odds of the outcome in the exposed group to the odds of the outcome in the unexposed group

What is odds?

The ratio of one outcome to that of an alternative outcome - could be the odds of disease if vaccinated compared to odds of disease if not vaccinated

What is the risk ratio?

The ratio of the risk of an outcome in the exposed group to the risk of the outcome in the unexposed group

What is the 'dual model' of decision making?

The use of type 1 and type 2 decision making type 1- INTUITION (gut instinct) what we see obviously, simple cognition type 2- ANALYTIC we apply our scientific knowledge to look a scientific facts etc. (apply CLINICAL REASONING TOOLS- apply the knowledge you have learnt)

What does a number needed to be treated of 1 mean? What does a low NNT mean?

This is ideal, it means that only one case needs to be treated to cure/prevent one additional case of disease. reciprocal of RD NNT= 1/RD low NNT- it is an effective treatment

What do we use SEM (Standard error of the mean) for?

To calculate the confidence interval

How do you check hydration status?

Touch MMS Tent skin Look at eyes (bright or dull? sunken?)

What do we do with the problem list?

Use it to make a list of differential diagnoses

What are the requirements of a chi test?

Used to COMPARE PROPORTIONS Outcome and exposure/intervention are BINARY Observations must be independent of each other

What is an ANOVA test? (Analysis of variance)

Used to test equality of the medians of two or more populations eg. comparison of three diets for control of dental calculus Requirements - outcome continuous and data normally distributed - Two or more comparison groups - groups must have equal variance (homoscedasticity)

What are two-sample t tests?

Used to test equality of the medians of two populations Requirements - outcome continuous and data normally distributed - two comparison groups

What is Wilcoxon's rank sum test*?

Used to test equality of the medians of two samples Requirements - Outcome continuous; two comparison groups - data doesnt need to be normally distributed - Non-parametric equivalent of two sample t test

What is a paired t test?

Used to test equality of the medians of two samples/populations when the observations arise as paired samples eg. blood glucose in high -fibre vs. low fibre diet, same dogs for each test (PAIRED SAMPLES) Requirements - outcome must be continuous and differences between the pairs must be normally distributed - two comparison groups- paired

What is Wilcoxon's signed rank test (2)?

Used to test the difference between two samples using matching pairs Requirements - Outcome continuous - Data does NOT NEED TO be normally distributed - Non-parametric equivalent to paired t-test

What is the Kruskil Wallis test?

Used to test the equality of the medians of two or more samples requirements - outcome continuous; two or more comparison groups - Data does not need to be normally distributed

what is Wilcoxon's signed rank test? (1)

Used where it is required to test whether the median of a sample or population is different from a particular value Requirements - Outcome continuous, one group - data DOESNT NEED TO BE NORMALLY distributed - non-parametric equivalent to one sample t-test

What are one-sample t tests?

Used where it is required to test whether the median of a sample or population is different from a particular value eg. grower pigs expected to grow a mean of 607g/day - check if they are?- if mean meets this value Requirements - Outcome continuous and data normally distributed - one group

What is the 95% confidence interval covers 0?

We conclude that it there is no significant difference between the two groups, because the difference between them could be 0 • A confidence interval that includes the value of no difference indicates that groups do not significantly differ from each other

What do we do in a diagnostic procedure after we have collected signalment, history, and physical examination?

We create a problem list

What do we do with the problem list and list of differential diagnoses?

We make a diagnostic plan, which we use to make a treatment plan.

What is the ethical matrix?

Wellbeing, autonomy, and justice of treated organism, producers, consumers, biota

What is virtue ethics?

What would a GOOD VET do?- Compassion, discernment, Trustworthiness, integrity, conscientiousness Values: • Personal development • Practical wisdom • Questioning of personal values & motives • How to translate intentions into actions • Mentoring, role modelling

What is negative confounding?

When the confounder makes the relationship look less significant or like it doesn't exist at all when it does.

What is a cross-sectional study?

When the sample of the population is taken at one point in time and is examined for the presence of an outcome (e.g. disease) and the exposure status regarding to risk factors. used to destermine PREVALENCE (and RISK FACTORS) can calculate risk ratio and odds ratio

What tests do not require data to be normally distributed?

Wilcoxon rank test, Kruskal-wallis

What are the tests used to COMPARE GROUPS that are CONTINUOUS and NOT normally distributed?

Wilcoxon tests (Used where it is required to test whether the median of a sample or population is different from a particular value, or if 2 means are different to each other, or if 2 paired means are different to each other) - like the t-test but for tests that are NOT normally distributed! Kruskal-Wallis Test like ANOVA - used where is it is required to test if 2 or more medians differ

What software can you use to calculate p-values and confidence intervals?

WinPei

When would you use each?

You would use experimental if the intervention is easily applied, and you would use observational if it is not easily applied.

Whats are YOPIs?

Young Old Pregnant Immunocompromised

What kind of confidence interval does a large sample size have?

a small confidence interval

What does selection of the appropriate statistical tests depend on?

a. the data type (eg. the chi-test is appropriate for binary outcome data) b. assumptions required for various tests

What is heart rate/pulse measured in? how do you measure it?

beats per minute (bpm) PULSE- gently roll index and middle finger over the femoral artery, feeling for pulse count the number of pulses in 15 seconds multiply by 4 OR HEART RATE (auscultation/palpation)- place hand or stethoscope over heart, listening for heart beat count the number of beats in 15 seconds multiply by 4 to get bp

what does a prolonged CRT mean?

blood not circulating adequately, shock, heart disease, other illnesses

What are the kinds of observational analytic study designs?

case-control study, cross-sectional study, cohort study

What is a random error in a study?

chance variation

What are the common statistical tests for comparing groups?

chi square test, WinPepi test, one-sample t test, two-sample t test, paired t test, ANOVA (Analysis of variance), Wilcoxon's signed Rank test (1), Wilcoxon's signed rank test*, wilcoxon's signed rank test 2, Kruskal-Wallis test, multivariable analysis, linear regression, logistic regression, Kaplan-Meier curve with Log-Rank test, Cox Proportional Hazards Regression

What can you look for in the eyes?

conjunctivitis,

What is the test used to COMPARE PROPORTIONS (dichotomous outcome and exposure) ?

dichotomous outcome and exposure CHI SQUARE TEST

What are the types of quanitative variables?

dichotomous, nominal and ordinal

What are the types of quantitative vairables?

discrete and continuous

What can crackles in lung sounds mean?

distal lung disease (fluid= oedema/pus)

How do you calculate the 95% confidence interval for the mean?

for a sample of 10 "t" is 2.26- use software to give t (depends on the sample size) = 6.0 ± 2.26*0.40 = 6.0 ± 0.90 • The 95% confidence interval (CI) is from 6.0 ‐ 0.90 = 5.10 to 6.0 + 0.90 = 6.90

What are some other things you should look for when listening to the heart?

heart MURMURS (abnormal heart sounds caused by turbulent blood flow in the heart) and ARRHYTHMIA ( irregular or abnormal heart rhythm- can be fast, slow, or variable)

When should you seek immediate veterinary help when an animal is seizuring?

if it lasts for more than 3-5 minutes if the seizures are clustering (more than 2 in a 24 hour period) Otherwise, seek veterinarian attention later

What can the relationship between OR and RR tell us about the occurrence of disease?

if they are close together, the disease is RARE If they differ markedly, it is a common disease

What does a high heart rate mean?

ill, painful, scared

What does a high respiration rate mean?

ill, painful, scared

What is history broke up into? How can you get a good knowledge of the history of an animal?

immediate general You can get a good history by asking careful and detailed questions eg. My dog has diarrhoea Questions- about duration, frequency, description of diarrhoea - has the animal been wormed recently? when was the last time? etc.

What are the kinds of experimental analytic study designs?

laboratory trials, clinical trials

what is the test used for continuous outcome multivariable data?

linear regression

What does a blue or purple mucous membrane mean?

low oxygen content in blood, severe heart of respiratory problem

What does a pale mucous membrane mean?

low red blood cell count (anaemia) or peripheral vasoconstriction (blood vessels constrict)

What are the palpable lymph nodes?

mandibular lymph nodes, prescapular lymph nodes, accessory axillary lymph node, inguinal lymph node, popliteal lymph node

What can you look for in the ears?

otitis externa

What to look at in the abdomen?

palpation, lumbar vertebrae, umbilicus, mammary glands, kidneys, intestines, bladder, perineum/genitals

What can friction, or 'velcro-'like sounds in the lungs mean?

pleural disease

How do causes operate? (in different ways)

predispose: increase level of susceptibility (e.g. age, sex, previous illness) - enable: facilitate manifestation of the disease (e.g. poor nutrition, bad housing) - precipitate: disease onset is associated with a specific disease agent (e.g. infectious agent) - reinforce: repeated exposure (e.g. to the same agent without immune response) may aggravate the presence of a disease - interact: the effect of two or more causes acting together is often greater than summing the individual effects

What ae measures of spread?

range, percentiles, variance, standard deviation

What else do you look at when looking at respiration?

respiration effort, how deeply does the animal breath? Listen for NOISES: wheezes, coughs, gasps

What does a low respiration rate?

shock or very ill, anaesthesia or sedation

What does a slightly tacky mucous membrane mean?

sign of mild hydration, could also be due to panting

What should you look for when you examine the limbs and tail?

skin, joints, pads, nails.

What should you look for on the neck?

skin, jugular veins, larynx, trachea, lymph nodes

What is large bowel diarrhoea like?

straining, mucous, and frank blood (recent blood, fresh)

What are the tests used to COMPARE GROUPS that are CONTINUOUS and NORMALLY DISTRIBUTED?

t-tests ( 1 population comparing mean to a certain value, 2 populations comparing means, 2 paired populations comparing means 1 and 2) ANOVA (Used to test equality of the means of two or more populations)

What is positive confounding?

when the confounder makes the positive relationship between exposure and outcome look STRONGER (or exist to begin with)

What is the p value? ***

• A p‐value is calculated AFTER the statistical test has been performed • A p‐value is always related to the hypothesis test • The p value is the probability of observing a test statistic as large or larger than that observed, if the null hypothesis is true (basically indicating the probability of the 'data occurring' if the null hypothesis was true) • If p is large, then the data are consistent will the null hypothesis - We conclude that there is NO strong evidence that effect tested really exists • If p is very small, it is unlikely the null hypothesis is true - If the p‐value is less than alpha, the null hypothesis is rejected - We say the difference is 'statistically significant' IF P VALUE IS LARGE, THEN DATA IS CONSISTENT WITH NULL HYPOTHESIS because there is a large probability that the null hypothesis is true IF P VALUE IS SMALL , IT IS UNLIKELY THE NULL HYPOTHESIS IS TRUE IF THE P VALUE IS SMALLER THAN ALPHA, the NULL HYPOTHESIS IS REJECTED- we say the difference is 'statistically significant'

What are the weaknesses of using a p-value to assess if a relationship exists?

• As level of significance of 0.05 is a completely arbitrary cut‐off - A dichotomous interpretation of p‐values (i.e. 0.05 or less = 'significant' and above 0.05 = non‐significant) might be inappropriate - A p‐value of 0.04 means little different from a p‐value of 0.07 • A p‐value provides no information about the likely size of effect - We are typically interested in the magnitude of an effect, not just whether an effect is likely to exist or not (but confidence intervals do provide information on the magnitude of an effect) • Statistical significance does not equate to biological, clinical or economic importance - A statistically significant result may be of little importance

What values in confidence intervals show no difference between groups?

• For Means, Risk difference - 95% CI contains "0", difference of 0= no difference • For Odds Ratios and Risk ratios - 95% CI contains "1" - ratio could be 1:1, if 1 is in the boundaries, which points to NO DIFFERENCE Conclusion: Observed difference is not significant at p = 0.05