Term
|
Definition
| results of a test should not vary based on who does the test or where it is done |
|
|
Term
|
Definition
| test must be able to distinguish between those with disease and those without disease |
|
|
Term
|
Definition
| test result should have a reasonable of changing decisions |
|
|
Term
|
Definition
| it may not be worth doing if it is too expensive, painful, or difficult to do |
|
|
Term
|
Definition
| diagnostic tests we want to evaluate - available when gold standard is not practical (too expensive, invasive, culture, time, impractical etc) |
|
|
Term
|
Definition
| definitive test for the disease (reference method) |
|
|
Term
|
Definition
| ability of a test to accurately classify a group of patients known to have a disease |
|
|
Term
|
Definition
| ability of a test to accurately classify a group of patients to be free of a disease |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| T/F? You can improve both sensitivity and specificity. |
|
Definition
|
|
Term
| If increase sensitivity, we will __________ secificity. |
|
Definition
|
|
Term
| If we increase specificity, we will _________ sensitivity. |
|
Definition
|
|
Term
| Receiver Operator Characteristic Curve (ROC curve) |
|
Definition
| plot of sensitivity on the y axis against (1-specificity) on the x axis for varying values of the threshold/cut off point |
|
|
Term
|
Definition
| identify systematic errors (bias) that can affect study conclusions |
|
|
Term
| Three examples of internal validity |
|
Definition
| spectrum bias, verification bias (referral/work-up bias), observational bias (expectation, information bias) |
|
|
Term
|
Definition
| how the study relates to your situation or patient |
|
|
Term
| 2 examples of external validity |
|
Definition
| is the test potentially relevant to my practice, was the spectrum of patients representative of the patients who will receive the test? |
|
|
Term
|
Definition
| when either sensitivity or specificity is overestimated |
|
|
Term
| When a diseased group is skewed toward highter severity that in practice - sickest of the sick |
|
Definition
| sensitivity is overestimated |
|
|
Term
| when non-diseased troup is skewed toward greater health - wellest of the well |
|
Definition
| specificity is overstimulated |
|
|
Term
| Verification (referral/work up bias) |
|
Definition
| use of new test to decide whether to apply gold standard |
|
|
Term
| Verification bias __________ sensitivity and _________ specificity |
|
Definition
| overestimates, underestimates |
|
|
Term
|
Definition
| subconsciously influenced by knowledge of the cases, the more subjective the test - the more susceptible test results are to observational bias |
|
|
Term
| Diagnosis is an __________ process |
|
Definition
|
|
Term
| when distinguishing abnormal from normal you will see biological variation ___________ animals and __________ animals |
|
Definition
| within animals, between animals |
|
|
Term
|
Definition
|
|
Term
| abnormality is associated with |
|
Definition
|
|
Term
| Normal levels for a lot of measurements |
|
Definition
| overlap for some individuals |
|
|
Term
|
Definition
| clinical signs observed are compared to profiles or descriptions of diseases we hold in our memory *not the best choice! |
|
|
Term
|
Definition
| prevalence of disease for individuals with specific risk factors, frequency of occurrence of clinical signs observed within those diseases |
|
|
Term
| pathophysiological reasoning |
|
Definition
| system and lesion are indentified using knowledge of disease mechanisms. If a problem what is related to that problem ? |
|
|
Term
|
Definition
| clinical experience, patient's risk factors, history and physical exam, previous testing - always check the database! |
|
|
Term
|
Definition
| anything that will predict the presence of disease with a greater probability than chance alone |
|
|
Term
| Every time you run the same sample there should be |
|
Definition
|
|
Term
| criteria for evaluating diagnostic tests |
|
Definition
| reliability-precision, accuracy, usefulness, and value |
|
|
Term
|
Definition
| degree of agreement among repeated observations |
|
|
Term
|
Definition
| degree of agreement between the observed value and the true value |
|
|
Term
| High precision and low accuracy |
|
Definition
| far from value, but test has high precision |
|
|
Term
|
Definition
| compares results when the test is administered repeatedly by the same observer (Often assumed random differences) |
|
|
Term
|
Definition
| compares the results when measurements are made by different observers (assumed random and systematic differences) |
|
|
Term
| Reliability-Precison and Accuracy |
|
Definition
| 'test specific' intrinsic to the test |
|
|
Term
|
Definition
| 'context specific' dependent on test and clinical scenario |
|
|
Term
|
Definition
| what you though before, based on history and clinical examination |
|
|
Term
| Post-test disease probability |
|
Definition
| what you thought + what you think now due to new information and test results |
|
|
Term
|
Definition
| ability of a test to accurately classify a patient whose disease status is unknown |
|
|
Term
| positive predictive value (PPV) |
|
Definition
| probability that a patient whose test result is positive has the disease |
|
|
Term
| negative predictive value (NPV) |
|
Definition
| probability that a patient whose test result is negative does not have the disease |
|
|
Term
| Look at case #1 screening test |
|
Definition
|
|
Term
| 5 different way how you can estimate the pre-test probability of disease (some just factors, not nec. ways to estimate) |
|
Definition
| literature, local database, clinical judgement(history and clinical signs), does not have to be precise, may use range (sensitivity analysis) |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| Test with high sensitivity have _______ false negative results and ______ NPV |
|
Definition
|
|
Term
| When should you use a sensitive test? |
|
Definition
When its advantageous to 'rule out' a diagnosis in early stages of diagnostic workup to reduce number of possible diseases in list
a false negative is dangerous/penalty for missing disease (FN animal entering country - could create serious consequences) and final phases of eradication of infectious disease |
|
|
Term
| Tests with high specificity have _____ false positive and ______ PPV |
|
Definition
|
|
Term
| When to use a specific test |
|
Definition
when advantageous to 'rule in' diagnosis made based on other
False positive is dangerous: if test and slaughter is measure, cost of too many FP's would be great
highly toxic treatment (cancer dignosis before chemotherapy |
|
|
Term
|
Definition
| when a test has a high sensitivity (Sn) a negative test rules out disease |
|
|
Term
|
Definition
| when a test has a high specificity (Sp) a positive test rules in disease |
|
|
Term
|
Definition
| predicting the progess or outcome of the disease |
|
|
Term
| 2 phases of natural history of a disease |
|
Definition
| preclinical phase, clinical phase |
|
|
Term
|
Definition
| disease onset, pathologic evidence of disease, signs and symptoms |
|
|
Term
|
Definition
| signs and symptoms, medical care sought, diagnosis, tx, outcome |
|
|
Term
| Natural history of a disease |
|
Definition
| disease process that unfolds over time - natural history is the sequence of developments |
|
|
Term
| name 4 outcomes of disease |
|
Definition
| death, cure, remission, recurrence |
|
|
Term
|
Definition
| decrease or disappearance of signs and symptoms of disease |
|
|
Term
|
Definition
|
|
Term
|
Definition
| disease phase at which disease is diagnosed |
|
|
Term
|
Definition
| proportion of patients with a disease who die of it |
|
|
Term
|
Definition
| proportion of patients who are alive one years after diagnosis |
|
|
Term
| Any measures used to quantify prognosis must be.... |
|
Definition
| case based (denominator is the number of indiv. with the specified disease) |
|
|
Term
| Ways of expressing prognosis |
|
Definition
|
|
Term
| expressing prognosis as time |
|
Definition
|
|
Term
|
Definition
| length of time that half of the study population survives |
|
|
Term
| other prognosis measures (3) |
|
Definition
| response, remission, recurrence |
|
|
Term
|
Definition
| proportion of patients showing some evidence of improvement following an intervention |
|
|
Term
|
Definition
| proportion of patients entering a phase in which disease is no longer detectable |
|
|
Term
|
Definition
| proportion of patients who experience a return of disease after a disease-free interval |
|
|
Term
|
Definition
| factor able to give information on the clinical outcome of each patient; identify groups of patients with the same disease who have different prognosis |
|
|
Term
|
Definition
| the disease phase at which disease is diagnosed |
|
|
Term
|
Definition
| factor able to give information on the clinical outcome of each patient |
|
|
Term
|
Definition
| demographic (age), disease-specific (tumor size/metastasis at time of diagnosis), comorbidities (other patient's conditions and diseases) |
|
|
Term
|
Definition
| one group of individuals with the target disorder are followed over time, the occurrence of the outcome of interest (death, recurrence, remission) is monitored |
|
|
Term
|
Definition
| time measure on a subject who does not have the outcome/event under study |
|
|
Term
| 2 reasons for censored observations |
|
Definition
| animal drops out of study before having the event (incomplete follow-up, withdrawl), animal makes it to the end of the study without having the event |
|
|
Term
| Look at the "follow up graphs" |
|
Definition
| demonstrates prognosis etc. |
|
|
Term
|
Definition
| estamate the survival at each point in time when each death occurred; survival probabilities are not calculated at censoring times |
|
|
Term
| Kaplan-Meier Method equation |
|
Definition
| S(t) = {(N(t)-E(t)/(N(t)}*S(Previous Event Time) |
|
|
Term
|
Definition
| proportion surviving to the time t who survived beyond time t |
|
|
Term
|
Definition
| proportion of original sample making it to time t |
|
|
Term
| Critical appraisal of studies that report prognosis |
|
Definition
| inception cohort, referral patterns, follow-up completeness |
|
|
Term
|
Definition
| animals are followed from a well-defined location along the course of disease (ex: onset of symptoms, time of diagnosis, beginning of tx) should be the same for all patients |
|
|
Term
|
Definition
| Avoid missing animals that died or recovered early on the disease course |
|
|
Term
|
Definition
|
|
Term
| Example of how referral patterns affect external validity: |
|
Definition
| referral centers for the disease are likely to see cases with worse prognosis |
|
|
Term
|
Definition
| long enough to assure every patient is followed until disease outcomes/recover |
|
|
Term
| follow-up study complete: |
|
Definition
| if there is a large number of loss-to-follow up (20%) prognosis becomes inaccurate. If reasons are related to follow-up risk of bias |
|
|
Term
| Determining if a follow-up study of patients was sufficiently complete |
|
Definition
| perform sensitivity analysis |
|
|
Term
|
Definition
| makes efficient use of all available data in cohort to describe prognosis |
|
|
Term
|
Definition
| help us to provide case specific prognosis information |
|
|
Term
| studies reporting prognosis |
|
Definition
| should be clearly state the zero time, have long and complete follow up times, and account for known prognosis factors |
|
|
Term
|
Definition
probability of having the disease P(TP);
probability of NOT having the disease P(FP) |
|
|
Term
| Negative (Non-focal scan) |
|
Definition
probability of having the disease P(FN);
probability of NOT having the disease P(TN) |
|
|
Term
Post-test disease probability
(equation) |
|
Definition
| pre-test disease probability + negative results = post-test disease probability |
|
|
Term
| Do the post-test probability questions in the PPT (causation 4th) |
|
Definition
|
|
Term
|
Definition
| factor that changes the risk (probability) of developing an outcome (disease) in the future |
|
|
Term
|
Definition
cholesterol and heart disease in humans
castration and prostatic cancer in dogs
tv watching and childhood obesity
diet and metabolic diseases in dairy cows |
|
|
Term
|
Definition
| statistical relationship between two or more events, characteristics or other variables |
|
|
Term
|
Definition
| change in one variable is responsible, directly or indirectly, for an observed change in another variable |
|
|
Term
| Do epidemiologic studies prove causation |
|
Definition
| NO. Epidemiological studies cannot prove that causation occurs, but they may be useful in supporting a causal association |
|
|
Term
|
Definition
| always the possibility that na observed association is due to chance alone |
|
|
Term
| An association can be due to (4 things) |
|
Definition
| chance, bias, confounding, causation! |
|
|
Term
|
Definition
| association results form errors in the study, design, implementation, or analysis |
|
|
Term
|
Definition
| relationship distorted by add'l variables (confounder), which is associated to the factor under study and the disease |
|
|
Term
| Algorithm for associations |
|
Definition
| rule out random error, rule out bias, valid associton (often using criteria ex: Hill's) |
|
|
Term
|
Definition
|
|
Term
|
Definition
| confounding and causation |
|
|
Term
| hierarchy of study design to establish cause-effect associations |
|
Definition
from - strength of evidence to +
case report, case series, case control/cross sectional, cohort, randomized clinical trial |
|
|
Term
|
Definition
| studies designed to describe population characteristics =, such as occurrence of disease by time and place |
|
|
Term
|
Definition
| studies designed to examine etiology and casual associations |
|
|
Term
| Descriptive studies do not have |
|
Definition
|
|
Term
| uses of descriptive studies |
|
Definition
| trend analysis, surveillance, health-care planning, and hypothesis generation |
|
|
Term
| Is there a risk in overstepping the data in a descriptive study? |
|
Definition
|
|
Term
|
Definition
| detailed presentation of a single case, least publishable unit in med lit, asks the W questions, generally not representative of the general course of disease, successful outcome (unusual tx) |
|
|
Term
|
Definition
| Collection of cases, useful to describe clinical characteristics, clinical course of disease, and natural history of disease; surveillance: first signs of new disease; hypothesis generation |
|
|
Term
|
Definition
| study associations, comparison group; exposure to outcome |
|
|
Term
| if exposure is applied by researcher it is |
|
Definition
| an experimental study (randomized controlled trial) |
|
|
Term
|
Definition
| classified based on the temporal direction of the trial |
|
|
Term
|
Definition
| defined population --> gather data on exposure and diseaes --> exposure/disease factors |
|
|
Term
|
Definition
| defined population, cases vs controls |
|
|
Term
|
Definition
| defined population with choice or circumstance --> exposed vs not exposed over time --> outcome (disease/no disease) |
|
|
Term
|
Definition
| from defined population to new data collection and hypothesized outcome of disease in the future; forcasting disease! |
|
|
Term
| retrospective cohort study |
|
Definition
| use of existent datasets to track past disease over time |
|
|
Term
| 2 Criteria to establish cause and effect associations |
|
Definition
| ability to establish temporal sequence of events; risk of bias |
|
|
Term
|
Definition
| error in the study that produces results that depart systematically from the truth |
|
|
Term
|
Definition
| selection bias, observational (information) bias |
|
|
Term
|
Definition
| stems from an absence of comparability between groups being studied (unbalanced groups) |
|
|
Term
| observational (information) bias |
|
Definition
| when information on exposure, outcome, and/or covariates of interest is collected differently between groups |
|
|
Term
|
Definition
|
|
Term
|
Definition
| longitudinal study = prostpective study |
|
|
Term
| prospective cohort study = |
|
Definition
| concurrent cohort study = concurrent prospective study |
|
|
Term
| retrospective cohort study = |
|
Definition
| historical cohort study = non-concurrent prospective study |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| When do we use information on incidience? Used in studies to: |
|
Definition
| predict the risk to develop disease, associate risk factors to disease, predict prognosis, and evaluate new therapies |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| measures of risk at the individual level |
|
Definition
| relative risk, attributable risk |
|
|
Term
|
Definition
| ex: 'how many times MORE LIKELY' are exposed individuals to become more diseased, RELATIVE to non-exposed indiv?' |
|
|
Term
|
Definition
|
|
Term
| LOOK AT RELATIVE RISK OF DISEASE INTERPRETATION SLIDE - causality iii april 16 |
|
Definition
|
|
Term
|
Definition
| what is the risk of disease attributable to the exposure |
|
|
Term
|
Definition
|
|
Term
| Measures of risk at the population level (2) |
|
Definition
| population attributable risk and population attributable fraction |
|
|
Term
| population attributable risk |
|
Definition
| predicts the reduction in risk achievable if a risk factor is removed from the population |
|
|
Term
| population attributable risk eqn |
|
Definition
|
|
Term
| population attributable proportion |
|
Definition
| measure of what proportion of disease in a population is attributable to the risk factors |
|
|
Term
| Population attributable proportion |
|
Definition
| AFp = AR/(Pexp * IE + (1-Pexp) * INE) |
|
|
Term
|
Definition
Organism present in every case
Isolate from case and grow in pure culture
Organism causes disease when inoculated into susceptible animal
Organism can be recovered from animal and identified |
|
|
Term
|
Definition
| separation of strong and weak criteria |
|
|
Term
| Temporality (VERY important) |
|
Definition
| cause must precede effect in time, criterion is inarguable, study designs based on temporal sequence |
|
|
Term
|
Definition
| the stronger the relationship between the independent variable (risk factor) and the dependent variable (disease), less likely that relationship is due to something else or by chance |
|
|
Term
|
Definition
| does increased exposure result in more of the outcome |
|
|
Term
|
Definition
| has effect been seen by others |
|
|
Term
| Cats, as DH play role in trasmission of T. gondii. A study was conducted to determine the seroprevalence and risk factors for T. gondii infection in domestic casts. Serum of cats was tested. To ID risk factors for infection a questionnarie was administered to the cat owners when they visit the vet to draw blood. the study is best described as |
|
Definition
|
|
Term
| study investigated the effect of lameness on repro performance on several commercial dairies. existent data collected during routine herd health and production control program was used o address the question. two groups were generated from th original dataset (cows with and without lameness). this is best described as |
|
Definition
| retrospective cohort study |
|
|
Term
| during fall and early winter vets reported an unusual cluster of neurologic illness. horses had ataxia primarily on rear limb and muscle fasciculations or trembling. Many had acute onset. This was the first reported WNV outbreak among horses in USA |
|
Definition
|
|
Term
| To compare incidence of and breed-related risk factors for gastric dilatation-volvulus (GDV) among several dog breeds, owners of dogs that did not have a history of GDV were recruited at dog shows. Information concerning the dogs' medical history, genetic background, personality, and diet was obtained from owners, and owners were contacted by mail and telephone at approximately 1-year intervals to determine whether dogs had developed GDV or died. This study is best described as: |
|
Definition
|
|
Term
| A study aimed to identify risk factors for colic caused by simple colonic obstruction and distension in horses. Colic horses were recruited from 2 veterinary school clinics. Healthy horses were chosen by selecting randomly an owner from those who had brought a horse to the same Veterinary School in the previous calendar year. A questionnaire was design to identify risk factors. This study is best described as: |
|
Definition
|
|
Term
| 3 steps of decision making-therapy |
|
Definition
| Identify objective of treatment, select treatment, specify treatment target |
|
|
Term
| 3 steps in how to decide which treatment to prescribe |
|
Definition
| induction, deduction, faith |
|
|
Term
|
Definition
| no control, regression to the mean, cure and remission are outcomes of disease, misdiagnosis, 'seems to work' |
|
|
Term
|
Definition
| all participants have the same chance of being assigned to each of the study groups |
|
|
Term
|
Definition
| ensures groups in clinical trials are comparable, reduces risk of selection bias, most powerful method of eliminating known and unknown confounding variables |
|
|
Term
|
Definition
| sequence based on a random-number |
|
|
Term
|
Definition
| clinicians should be UNAWARE of which treatment the next patient is receiving |
|
|
Term
|
Definition
| no intervention, observation, placebo tx, standard drug tx |
|
|
Term
|
Definition
| effect (usually positive or beneficial) of being included in a scientific study (can be a response of the animal owner) |
|
|
Term
|
Definition
| an inactive substance, means 'I do nothing' in Latin |
|
|
Term
|
Definition
| beneficial and attributable to the expectation that the regimen will have an effect; not acceptable in life-threatening conditions |
|
|
Term
|
Definition
| keeping individuals involved in the study unaware of the assigned intervention |
|
|
Term
|
Definition
| Single (owner), double (owner + vet), triple (owner + vet + person analyzing data) |
|
|
Term
|
Definition
| minimize observation bias, improve patient compliance and retention, reduce co-interventions |
|
|
Term
| Cats, as definitive hosts, play an important role in the transmission of Toxoplasma gondii. A study was conducted to determine the seroprevalence and risk factors for T. gondii infection in domestic cats. Serum samples of cats were tested for T. gondii antibodies. To identify the risk factors for infection with T. gondii in cats, a questionnaire was administered to the cat owners when they visit the veterinary to draw the blood sample. This study is best described as |
|
Definition
|
|
Term
| A study investigated the effect of lameness on reproductive performance on several commercial dairy farms. The existent data collected during the routine herd health and production control program was used to address the question. Two groups were generated from the original dataset (cows with and without lameness). This study is best described as |
|
Definition
| retrospective cohort study |
|
|
Term
| During fall and early winter 1999, veterinarians with the U.S. Department of Agriculture and the New York Department of Agriculture and Markets reported an unusual cluster of neurologic illness in horses on Long Island. The investigators described twenty cases of equine neurologic illness. Horses had ataxia (95.7%), primarily rear limb (90.5%,) and muscle fasciculations or trembling (55%). Many had acute onset (90.5%). This was the first reported West Nile Virus outbreak among horses in USA. This study is best described as |
|
Definition
|
|
Term
| To compare incidence of and breed-related risk factors for gastric dilatation-volvulus (GDV) among several dog breeds, owners of dogs that did not have a history of GDV were recruited at dog shows. Information concerning the dogs' medical history, genetic background, personality, and diet was obtained from owners, and owners were contacted by mail and telephone at approximately 1-year intervals to determine whether dogs had developed GDV or died. This study is best described as: |
|
Definition
|
|
Term
| A study aimed to identify risk factors for colic caused by simple colonic obstruction and distension in horses. Colic horses were recruited from 2 veterinary school clinics. Healthy horses were chosen by selecting randomly an owner from those who had brought a horse to the same Veterinary School in the previous calendar year. A questionnaire was design to identify risk factors. This study is best described as: |
|
Definition
|
|
Term
| Gold standard for evaluating the efficacy of therapeutic and preventative interventions |
|
Definition
| randomized controlled trial |
|
|
Term
| 2 important features of randomized controlled trials |
|
Definition
| randomization and blinding |
|
|
Term
| Intention to treat analysis (2) rules |
|
Definition
| use every subject who was randomized according to randomized treatment assignment; ignore compliance protocol deviations, withdrawal, and anything that happens after randomization |
|
|
Term
| Compliance and Intention to treat eqn |
|
Definition
| look at slides in therapy II ppt |
|
|
Term
|
Definition
| probability of obtaining the results observed, if there are no difference between the two groups to be compared (due to chance) |
|
|
Term
|
Definition
| no real difference betwee nthe two groups to be compared |
|
|
Term
| Type I error (alpha): False Positive |
|
Definition
| risk of concluding that there is a difference in outcome among groups when, in fact, there is not |
|
|
Term
| Type II error (beta): False negative |
|
Definition
| concluding that tx does not work when in fact it does |
|
|
Term
|
Definition
| the probability to identify an effect of tx, if one exists; analogous to the sensitivity of diagnostic test |
|
|
Term
|
Definition
| the chance of obtaining the same results as the experiemtn is high |
|
|
Term
|
Definition
| small sample sized; studies with statistically non- significant results and low power are inconclusive; common problem |
|
|
Term
| the research plan should be formally stated where? |
|
Definition
|
|
Term
| protocol for clinical trial (4) |
|
Definition
| procedures for handling problem cases, procedure for informed consent, sample size calculations and procedure for analyzing results, appendices: forms |
|
|
Term
|
Definition
|
|
Term
| Look through owner's consent obligations |
|
Definition
|
|
Term
|
Definition
| refers to the greater likelihood that studies with positive results will be published |
|
|
Term
| publication bias - does it matter? (5) |
|
Definition
| distorts the scientific record; hides truth;influences clinician's decision making; misleads policy makers; causes harm to patients |
|
|
