Exam II Flashcards
(110 cards)
1
Q
What are design selections based on?
A
- perspective of question (hypothesis)
- ability/desire to force group allocation (randomization)
- ethics of methodology
- efficiency and practicality (time and resource devotion)
- cost
- validity of acquired information
2
Q
What are the different hypotheses researchers formulate before the study takes place?
A
null hypothesis (Ho)- a research perspective which states there will be no (true) difference between the groups being compared
-conservative and commonly utilized
-perspectives
alternative hypothesis (H1): perspective which states there will be a (true) difference between the groups being compared
3
Q
What are the two blanket designs for human designs?
A
observational: study design considered “natural”
- researchers observe subject elements occurring naturally or selected by individual (naturally or freely)
- typically not able to prove causation
- NO RESEARCHER FORCED GROUP ALLOCATION
- useful for forced interventions (not ethical)
Interventional: study designs considered experimental
- investigator selects interventions (exposure)
- there is researcher forced group allocation -> randomization
4
Q
What the studies that can be found under the two blanket designs?
A
Observational:
- cases (reports/seres)
- ecological
- cross sectional (analytical)
- case control (analytical)
- cohort (analytical)
Interventional:
- pre-clinical
- phase I
- phase II
- phase III
- phase IV
5
Q
What are the two types of error and what do they mean?
A
Type I error: false positive
Type II error: false negative
6
Q
What three perspectives are taken for a hypothesis?
A
- superiority
- noninferiority
- equivalency
7
Q
What is the difference between a population and sample?
A
population: all individuals making up a common group, from which a sample can be obtained, if desired
- not study population which is group of individuals selected for the study
sample: a subset or portion of the full complete population -> representatives
- random processes to draw sample
- when complete population is not feasible
8
Q
How is a study population selected?
A
-research hypothesis/question
-inclusion and exclusion selection criteria and case and control group or exposed vs nonexposed
+desired vs logical vs plausible
+impact generalizability -> external validity
-ethics (principles of bioethics must be met)
+some don’t agree with the use of placebo when treatment is available
-equipoise (genuine confidence that an intervention may be worthwhile in order to use on humans)
9
Q
What are the 4 key principles of bioethics?
A
- autonomy: self rule/self determination. Participants must…
- have full and complete understanding of the risks and benefits
- decide for oneself without outside influences
- beneficence: benefit or do good for the patient (NOT SOCIETY)
- justice: equal and fair treatment regardless of patient characteristics
- nonmaleficence: DO NO HARM! Researchers must not…
- withhold information
- provide false information
- exhibit professional incompetence
10
Q
What is the Belmont Report? What are the three guiding principles?
A
-document that determines ethicality of a study.
- respect for persons: research voluntary, subjects autonomous
- beneficence: research risks are justified by potential benefits
- justice: risks and benefits are equally distributed
11
Q
What is consent vs assent?
A
consent: agreement to participate, based on being fully and completely informed (given by mentally capable individuals of legal consenting age)
assent: agreement to participate, based on being fully and completely informed, given by mentally capable individuals NOT able to give legal consent (children and mentally handicapped)
- requires assent of subject and consent of guardian
12
Q
Who determines the ethicality of a study?
A
Institutional Review Board (IRB) or the Ethics Committee
- protect human subjects from undue risk BEFORE study
- IRB regulated by Federal statues developed by Department of Health and Human Services -> follow Common Federal Rules
13
Q
What are the different levels of IRB review?
A
Full Board: used for ALL interventional trials with more than minimal/no risk to patients -> medication related studies
Expedited: minimum risk and/or no patient identification
Exempt: no patient identifiers, low/no risk, data set analysis, environmental studies, use of existing data/specimen
14
Q
Who determines safety AFTER a study has begun?
A
-Data Safety and Monitoring Board (DSMB)
-semi-independent committee not involved with the conduct of the study but charged with reviewing study data as study progresses, to assess for undue risk or benefit
+pre-determined review period
+can end study early for overly positive or negative effects
15
Q
What occurs during the pre-clinical phase of an interventional study?
A
- occurs prior to human research
- bench and animal research
16
Q
Whats occurs during Phase I of an interventional study?
A
SAFETY
-small N (20-80), usually healthy volunteers to assess safety/toxicity, dosing and pharmacokinetics in population of interest, short duration
*doesn’t always include healthy subjects
17
Q
What occurs during Phase II of an interventional study?
A
safety and EFFICACY
-larger N (100-300), commonly utilize patients with condition of interest and nothing else, used to expand on purpose of phase I study(safety) but also begin assessing efficacy in diseased population, short-medium duration
*narrower inclusion criteria
18
Q
What occurs during phase III of an interventional study?
A
EFFICACY
-larger N(1000-3000), used patients with condition of interest to continue determination of safety, primary purpose is to assess efficacy, longer duration
+superiority v noninferiority v equality format
+FDA mandates phase 3 before drug approval, must beat placebo and exhibit equipoise
+companies only have to show efficacy in 3 out of 5 studies
19
Q
What occurs during phase IV of an interventional study?
A
SAFETY
-assesses long term effects (risks and benefits) in a large population of diseased patients, drug CAN die after phase 4 due to long term effects, always named drug/device
20
Q
What are the advantages and disadvantages of interventional trials?
A
advantages:
- cause precedes effect -> shows causation
- only design used by FDA approval process
disadvantages:
-cost
-complexity/time (development/approval/conductance)
-ethical considerations**this is good***
-generalizability (external validity)
+study pop may not be equivalent to general pop
21
Q
What is the difference between an explanatory and pragmatic study?
A
explanatory: trying to prove causation or explain why something is happening, set methods and they can’t change to suit patient
pragmatic: clinical setting (adjust dosage, etc to treat patients), allow people with comorbidies and multiple meds
+bad -> no control or regiment (adds extra confounding)
22
Q
Differentiate between simple and factorial studies.
A
simple: randomizes(divides) subjects into 2 groups, tests a single hypothesis, randomizes only once
factorial: randomizes multiple times! Tests multiple hypotheses, but must increase sample size
23
Q
What are some positives and negatives of the factorial study?
A
- improves efficiency for answering clinical qs
- increases study pop size
- increases complexity
- increase risk of drop out (due to complexity)
- may restrict generalizability of results
24
Q
What is the difference between a parallel and cross-over study?
A
parallel: group simultaneously and exclusively managed
-NO SWITCHING of intervention groups after randomization
+simple and factorial are parallel
cross-over (Self control): groups serve as their own control by crossing over from one intervention to another during the study
- allows for smaller N
- control for the same person taking the same drugs (identical people)
- requires wash out period
25
What is a wash out period?
- required in a cross over study
- controls for lingering effects of previous drug before switching groups
- gets drug out of system of patient before taking another
26
What are the different research studies? Put them in order from increasing strength.
- meta-analyses (most evidence)
- systemic reviews
- pragmatic trials, randomized, double blind controlled trials
- cohort
- case controlled
- cross sectional
- ecological
- case series
- case reports
- animal research
- in vitro research
27
What is a run-in/lead in phase?
-study subjects are blindly given one or more placebos for initial therapy (defined time period) to determine a "new" base-line of disease (standardization)
+can assess protocol compliance (determine how much of the meds were taken and keep or ditch patient based on how well they follow the protocols)
+can be a "wash out" for existing meds -> reduces one common exclusion criteria
+determines amount of placebo effect
+assesses placebo and Hawthorne effects
28
What are some disadvantages to a cross over study?
- only suitable for long term conditions which are not curable or which treatment provides short-term relief
- duration of study for each subject is longer
- carry over effects during cross over (wash out required -> prolongs study)
- treatment by period interaction -> disease that get better or worse during different times of the year (ex. depression or allergies)
- smaller N requirement only applicable if within subjects variation less than between subjects variation
- complexity in data analysis
29
What are the different outcomes that can be talked about?
primary- MOST IMPORTANT, key outcome
+main research question used for developing the study
secondary/tertiary- lesser importance, but still valuable and possible for future hypothesis generation
composite- combines multiple endpoints into a single outcome
30
What is the difference between a patient oriented endpoint and a surrogate marker?
patient oriented endpoint: most clinically relevant outcome, more important in a study, can be increased or decreased
ex. death, stroke or MI, hospitalization, preventing need for dialysis
surrogate marker: elements used in place of evaluating patient oriented/direct endpoints, risk factors for disease
ex. blood pressure (risk of stroke), cholesterol (risk of MI), change in SCr worsening renal function)
31
What is the difference between non-random and random selection and group allocation?
Non-random: subjects don't have an equal probability of being selected or assigned to each intervention group, not good protocol
ex. morning vs evening clinic attendance
+patients attending clinic on odd days vs even days
+first 100 patients admitted to hospital
Random: most commonly utilized, subject do have an equal probability of being assigned to each intervention group
32
What is the purpose of randomization?
to make groups as equal as possible based on known and unknown factors (confounders)
+attempts to reduce systematic bias between groups which impacts results
+Table 1 customarily used to show group characteristics
+equality is NOT guaranteed
33
How is the effectiveness of the randomization process documented?
- p values shown in table format
- p values not shown but text-statement given in key of table
- p value not shown but given in article
34
What are the three different types of randomization?
simple- equal probability for allocation into any study group
blocked- ensures balance within each intervention group -> researchers want to assure that all groups are equal in size
stratified- ensures balance within known confounding variables, makes characteristics of patients equal reducing confounding
+gender, age, disease, severity/duration, comordodities
35
What are three types of blinding?
single-blinding- subjects are not informed which intervention they are receiving (researchers do)
double-blinding- neither researchers or subjects know which group they have been allocated into
open-label- everyone knows which intervention each subject subject is receiving
36
How can one determine the effectiveness of blinding?
A post hoc study: ask patients and researchers, depending on blindness, which group they think they were apart of. If a high percentage guesses correctly, then the blinding wasn't very effective
37
What is a placebo?
- also know as a dummy
- inert treatments made to look identical in all aspects to the active treatments (dosage form, dosing frequency, monitoring, therapy requirements, etc)
38
What is a double dummy?
more than one placebo used...such as an inhaler and injection
39
What is the placebo effect?
- improvement in condition by power of suggestion and due to the care being provided
- improvement can be as large as 30-50%
40
What is the Hawthorne effect?
-desire of the study subject to please investigators by reporting positive results regardless of treatment allocation, want a positive outcome for study
41
What is a post hoc group analysis and why are these generally frowned upon in a study?
- not accepted as appropriate when not prospectively planned -> "data dredging or fishing"
- if there is no difference between the two studied groups then the researcher starts comparing groups in different ways until they find a difference
42
What should be considered when determining an appropriate sample size?
-take into consideration the number of drop outs
43
How can a researcher control for drop outs in a study?
-intent to treat (most conservative): keeping the dropped out patients's data and utilizing in the study
+last observation carried forward: data from the last known data collection and carry that forward throughout the study
+no benefit: convert all subsequent yet missed assessments for a subject to a null effect
-per-protocol/efficacy analysis: compliance is predefined, only accept subjects with 80-90% compliance of study protocol
as treated: ignores group assignments, allows subjects to switch groups and be evaluated in group they moved to, end up in, or spent the most time in
44
Why is intent to treat beneficial to a study aka why is it good to keep the data?
- preserves randomization process
- preserves baseline characteristics and group balance at baseline which controls for known and unknown confounders
- maintains statistical power (original sample size)
45
What are the benefits of using a per-protocol study?
biases estimates of effect (commonly over estimates effects)
| -reduces generalizability
46
What are some general methods to assess and improve compliance?
```
Assess:
-drug levels (multiple useful sites)
-pill counts at each visit
-bottle counter tops
Improve:
-frequent follow up visits/communication
-treatment alarms/notifications
-medication blister packs or dosage containers
```
47
What are case control studies?
-observational, analytical studies allowing researcher to be passive observer of natural events occurring in individuals with the disease/condition of interest (cases) who are compared yo people who do not have the condition of interest (controls)
-based on disease status
useful when studying a rare disease or investigating an outbreak
-odds ratio
48
What are case control studies useful for?
rare diseases
49
Why would one utilize a case study design?
- unable to randomize (unethical, illegal, not feasible)
- limited resources (time, money, subjects)
- disease of interest is rare and little is known about its associations/causes
- prospective exposure data, derived from prospective cohort study, is difficult/expensive to obtain and/or very time inappropriate
50
Are case studies retrospective or prospective?
retrospective
51
What are some strengths of case studies?
- good for assessing multiple exposures of one outcome
- disease is rare
- useful in calculating odds and ORs
- less expensive then interventional trials and prospective cohorts
- useful when ethical issues limit interventional studies
- useful for dynamic populations
- useful when disease has a long induction/latent period
52
How are cases selected for a case study?
-defined by investigator using accurate, medically-reliable, and efficient data sources (applied to all participants)
-objectively, consistently, accurately, and with validity
-clinically supportable/definable criteria are best
+from published, professionally-recognized and accepted diagnostic criteria and/or from multiple sources of data
53
What is the danger of the selection process?
misclassification
54
How are the controls of a case study selected?
-control selection
-goal: to assess for the presence of an association between exposure and known condition of interest by selecting non-disease individuals from the sample population which produced the cases
+controls should represent the baseline risk in the general or reference population
-the way controls are selected is a major determinant in whether any conclusion is valid (internal validity)
55
What are the criteria for selecting a control group?
-make the groups as close as possible except the presence of the disease of interest
+if exposure has no effect, then the odds will be exactly the same for both groups and OR will be 1.0
-group can come from several sources -> population, but the closer the better
+institutional/organizational/provider
+spouse/relatives/friends
+participated in same event
56
How can one be a control and a case in a study?
- can be associated with an outbreak investigation with multiple exposures
- in a situation of a brief (acute) change in risk of the outcome in interest (hazard period)
57
What is a case-crossover design?
- subjects are their own controls during the other times they don't have the acute change in risk
- the only case-control study design able to adequately attempt to address the issue of temporality
58
What is a nested case-control study?
-studies conducted after or out of a prospective cohort study
-subjects in cohort ultimately developing disease are defined as cases
+diseased used in a new (different) study design
+used to evaluate other exposures
59
How are samples taken in a nest study?
- survivor sampling: sample of non-diseased individuals (survivors) at end of study period
- base sampling: sample of non-diseased individuals at start of study period
- risk-set sampling: sample of non-disease individuals during study period at same time when case was diagnosed
60
What are the two biggest biases encountered in a case study?
selection bias: related to the way subjects are chosen for study
+any factor in the design/execution of a study that causes study groups to be different leads to spurious association between variables
+less of, or not, a significant issue during case-crossover study designs
recall bias: the amount /specificity that cases or controls recall past events different
+usually cases are more likely to recall past exposures and levels of exposure
61
What are the two different types of matching schemes utilized in case studies?
individual matching:
- matches individuals based on specific patient-based characteristics
- used when each case has unique and important characteristics
group matching:
- proportion of cases and proportion of controls with identical characteristics are matched
- requires cases to be selected first
- don't match on anything that might be a risk factor
62
What in general is a cohort study?
-observational, analytical studies allowing researcher to be a passive observer of natural events occurring in naturally exposed and unexposed groups
+group allocation based oin exposure status or group membership
63
What are cohort studies useful for?
rare exposure
64
What are the other names for cohort studies?
longitudinal, incidence, follow-up
65
What does a cohort study typically generate?
risk ratio
66
What is a cohort study trying to figure out?
Trying to figure out which groups have the disease based on knowing which groups were exposed
67
What are some reasons one would choose a cohort study?
- unable to randomize unethical, illegal, not feasible)
- limited resources (time/money/subjects)
- the exposure of interest is rare and little is known about associations/outcomes
- more interested in incidence rates/predictors of risks for outcome of interest (more than effects)
68
Timewise, how can a cohort study be conducted?
retrospective, prospective, ambidirectional
69
What is a prospective cohort study?
-exposure group is selected on the basis of a pastor current exposure and both groups followed into the future to assess for outcome(s) of interest (which has yet to occur) and then compared
70
What is a retrospective cohort study?
- at the start of the study, both the exposure and outcome of interest have occurred
- retrospectively start at time of exposure and follow forward to the point of outcome occurrence (known), in the present
- exposure still has yet to occur before outcome of interest and group allocation is based on exposure status, not disease status
71
What is an ambidirectional cohort study?
-uses retrospective design to assess past differences but adds all data collected on additional outcomes prospectively from start of study (looking for outcomes in the past and into the future)
72
What are the different types of cohorts? And famous examples?
-birth cohort: individuals assembled based on being born in a geographic region in a given time period
-inception cohort: individuals assembled at a given point based on some common factor (where they live, work, etc)
+ex. useful for single-group non-comparisons for incidence rate determination, Framingham study, nurse study
-exposure cohort: individuals assembled based on some common exposure (connected to environmental or other one time events)
+9/11 firefighters
73
What are the different sizes of cohort studies?
- fixed cohort: cohort which can't gain members, but CAN have loss to follow ups (ex. 9/11)
- closed cohort: a fixed cohort with no loss to follow ups
- open/dynamic cohort: a cohort with new additions and some loss to follow ups (can increase or decrease over time -> Framingham study)
74
How do you select individuals for the exposed vs non-exposed study?
exposed:
- easier part
- allocate subject based on pre-defined criteria of exposure
non-exposed:
- make groups as close as possible (coming from same cohort/population but not exposed)
- if exposure has no effect, then risk will be exactly the same for both groups and RR will be 1
75
What are the three sources that non-exposed individuals may be selected from?
internal (BEST):
-patients from same cohort, yet who are unexposed (most similar
-if there are levels of exposure, you may have to use the lowest exposure group as a comparator
general population:
-used as a second choice when the internal is not realistically possible (ex. everyone is exposed or exposure subjects taken from population)
comparison cohort (WORST):
-simply attempt to match groups as close as possible on numerous personal characteristics (can't control for other potentially harmful exposures in comparison cohort, also causing disease)
76
What are the general strengths of a cohort study?
- good for assessing multiple outcomes of one exposure (hard to control for other exposures if more than one plausible for being associated with an outcome)
- useful when exposures are rare
- useful in calculating risk and RR
- less expensive than interventional trials
- good when ethical issues limit use of interventional
- good for long induction/latent periods (retrospective)
- able to represent temporality (prospective)
77
What are the strengths and weaknesses of a prospective study?
Advantages:
- can obtain a greater amount of study important info from patients (more control over specific data collection)
- follow up/tracking of patients may be easier
- better at giving answer to temporality
- may look at multiple outcomes from a single exposure
- can calculate incidence and incidence rate
Disadvantages:
- time, expense, and lost to follow ups
- not efficient fro rare disease -> use case control
- not suited for long induction/latency conditions
- exposure (or amount) may change over time
78
What are the strengths and weaknesses of a retrograde cohort study?
Advantages:
- best for long induction/latency conditions
- able to study rare exposures
- useful if the data already exists
- saves time and money compared to prospective studies
Disadvantaged:
- requires access to charts, databases, employment records (may not be complete/thorough enough for the study)
- information may not factor in or control for other exposures to harmful elements
- patients may not be available fore interview if contact necessary for missing or incomplete data
- exposure (or amount) may have changed over time
79
What kind of error does lost to follow up create?
type 2
80
What affects outcome of occurrence in groups?
- level of exposure
- induction
- latency period
81
What are the major biases to consider in a cohort study?
- healthy worker effect
| - selection bias (how exposure status is defined/determined)
82
What is a cross-sectional study?
-observational, descriptive/analytical studies that examine relationships of health/disease to other variables of interest at the same time
-PREVALENCE study
-entire population or a subset is selected for the study
-what occurs at a specific point in time or time frame across a large population (a snap shot in time)
++++ANY STUDY THAT SAYS NATIONAL IS CROSS SECTIONAL+++
83
What does a cross sectional study focus on?
-simultaneously on disease and population characteristics, including exposures, health status, health care utilization, etc
+seeks associations not causation
+generates and test hypotheses
+by repetition in different time periods, can be used to measure change/trends
84
How are most cross sectional studies taken?
-surveys or databases related to different aspects of US pop
+different perspectives (inpatient vs outpatient)
85
What are the important factors to consider in a cross sectional study?
- person (with variables of interest)
- place (defined boundaries)
- time
86
What are some advantages to a cross sectional study?
-quick and easy to perform for the researcher using the data
+data already collected and deidentified
-useful for
+determining prevalence and risk factors across population
+measuring current health status and planning for health services across the population
+evaluating differences in sub groups within the population
87
What are the disadvantages of a cross sectional study?
-prevalent cases may represent survivors
+may be difficult to sort out factors associated with risk of disease from factors associated with survival (treatment and severity)
-difficult to study diseases of low frequency (prevalence is proportional to the incidence of the disease times its duration)
-problems in determining temporal relationship of a presumed cause and effect
+due to fact that fact exposure and disease histories are taken at the same time
88
What are the different methods of collecting data for a cross sectional?
1. collect data on each member of the population
+frequently utilized in city/state level evaluations -> ongoing collection
2. Take a sample of the population and draw inferences to the remainder
89
What are the different types of probability sampling schemes?
-probability samples (most common)
+every element in the population has a known (non-zero) probability of being included in sample -> multi-probability, multi-stage selection
1. simple random samples: obtain list of pop names, assign numbers to names and use random generator to select samples
2. stratified random samples: mutually exclusive strata- age or socioeconomic groups, divide pop into subgroups (take simple random from each subgroup), some studies oversample certain subgroups such as minorities
90
What are the other types of sampling schemes?
-systemic or convenience samples (not completely random or probable)
+decide on what fraction of pop is to be sampled and how they will be sampled
Ex. all persons whose last name begins with M-Z, persons in clinic every M/W/F, random starting point and take every 20th person
***introduces selection bias****
91
How is new data collected?
- questionnaires/surveys
| - physical assessment
92
What are some issues with how new data is collected?
-not likely to be the same individuals year to year
93
What are the actual surveys utilized in cross sectional studies?
- National Health and Nutrition Examination Survey (NHANES)
- National Health Interview Survey (NHIS)
- National Ambulatory Medical Care Survey (NAMCS)
- National Hospice Care Survey (NHCS)
- Behavioral Risk Factor Surveillance System (BRFSS)
94
What does NHANES ask?
-assesses the health and nutritional status of adults and children
-combines interviews and physical examinations
+interviews include demographic, socioeconomic, dietary, adn health related qs
+exam consists of medical, dental, physiological measurements adn lab tests
-survey sample is selected to represent the US pop of all ages
+oversamples persons over 60, blacks and hispanics
95
What does NHIS examine?
-principal source of info on health of the civilian, non-institutionalized pop
+central role in NSFG and NAMCS/NHCS
+represent US pop of all ages
-data are collected through personal household interview
-consists of a set of core qs that remain largely unchanged and a set supplements used to respond to public health data needs as they arise
96
What does NAMCS examine?
- a national survey designed to meet the need for objective, reliable information about the provision and use of ambulatory medical care services in the US
- based on a sample of visits to non-federal, office-based physicians primarily engaged in direct patient care
97
What does NHCS assess?
-a combined national survey designed to describe national patters of healthcare delivery in non-federal hospital based settings, including:
+discharges from inpatient departments and institutions, and visits to EDs, outpatient departments and ambulatory surgery centers
-integrates NHDS, NHAMCS, and DAWN
98
What does BRFSS assess?
-a state based system of telephone health surveys that collects information on health risk behaviors, preventative health practices, and health care access primarily related to chronic disease and injury
+monthly in all 50 states, DC, Puerto Rico, US Virgin Islands, and Guam
+500,000 adults are interviewed by telephone -> largest landline phone interview in the world
+youth BRFSS conducted by questionnaire in schools
99
What are the four different screening outcomes?
- true positive
- true negative
- false positive
- false negative
100
What is sensitivity?
- how well a test can detect presence of disease when the disease is present -> positivity of test
- proportion of the time that a test is positive in a patient that does have the disease
- highly sensitive test has a low false negative rate
101
What is specificity?
- how well a test can detect the absence of disease when the disease is absent -> negativity of test
- proportion of time that a test is negative in a patient that does not have the disease
- highly specific test has a low false positive rate
102
What is positive predictive value?
- how accurately a positive test predicts the presence of disease
- percentage of true positive's with positive test (correct prediction)
- aka predictive value-positive (PVP)
103
What is negative predictive value?
- how accurately a negative test predicts the absence of disease
- percentage of true negative's in patients with a negative test (correct prediction)
- aka predictive value-negative
104
What is a diagnostic accuracy/diagnostic precision?
-proportion of time that a patient is correctly identified as having either a disease or not having a disease with a positive or negative test
105
What are likelihood ratios?
- ratio of the probability of a given test result (positive or negative) for a person with the disease/ probability of the same test result (positive or negative) for a person without the disease
- can either positive or negative
106
What values do the different likelihood ratios need to be in order to be considered beneficial?
- LR+ should be greater than 10 to demonstrate the test is most beneficial
- LR- should less than 0.1 to demonstrate the test is most beneficial
107
Multiple Cutoff values?
- a test that reveals that theree isn't only a positive or negative result. There are numerical values to determine a cutoff of acceptance and abnormal physiological behavior.
ex. blood sugar, vitamins
108
What is validity? What are the different kinds of validity?
-ability to accurately discern between those that do and those that do not have disease -> "telling the truth"
+internal validity: extent to which results accurately reflect what was being assessed (true situation of study population)
+external validity: extent to which results are applicable to other populations (not included in the original study, aka generalizability)
109
What is a ROC curve?
- receiver operator curve
| - a graph comparing sensitivity (y) and 1-specificity (x)
110
What is reliability?
-ability of a test to give the same result on repeated uses
+reproducibility/consistency
-a valid test is always reliable, but a reliable test is not always valid
