Week 2 Flashcards
BSE: Goal is to understand a population based on:
Goal is to understand a population based on a sample
BSE: Descriptive statistics have 3 branches including
- Distribution
- Measures of Central Tendency: mean, median, mode
- Measures of variability: how far do things vary from the mean
BSE: What is the difference between a population and sample?
- Population: The collection of units to which we want to generalize a set of findings or a statistical model (the whole)
- Sample: A smaller (but hopefully representative) collection of units from a population used to determine truths about that population
BSE: Mean and Standard Deviation are obtained from a sample but are used to:
Mean and Standard Deviation are obtained from a sample but are used to estimate the mean and SD of the population
BSE: What does Measuring the ‘Fit’ of the model mean?
- The mean is a model of what happens in the real world: the typical score
- It is not a perfect representation of the data
- How can we assess how well the mean represents reality?
BSE: If have a large standard deviation, describe what will happen to the distribution
- Large standard deviation = wide distribution
BSE: If have a small standard deviation describe what happens to the distribution
- If have a small standard deviation the distribution is narrow (and tall?)
- ideal, want distribution to be close to the measure of central tendency
BSE: The normal distribution ideally has a _______________ mean, median, mode
The normal distribution ideally has the same mean, median, mode
BSE: What is a Gaussian distribution?
- Another name for normal distribution
BSE: What percentage of a sample lies within 1 standard deviation from the mean/median/mode?
68.3% should lie -1 SD and + 1 SD from the mean/median/mode
BSE: What percentage of the sample lies between the 2 SD from mean/median/mode?
95% for -2 SD & + 2 SD
BSE: What percentage of the sample lies between the 3 SD from mean/median/mode? F
~ 99% for - 3 SD & + 3 SD
BSE: T/F can have more than 1 mode for a sample
True, can have a singular mode, bimodal, mulitmodal
BSE: T/F: Normal distribution has a skew
- False, normally mean, median, mode
- Skew is symmetry of distribution
BSE: What is skew? What is + or negative skew?
- Skew is asymmetry of distribution
- Positive skew: scored bunched at low values with the tail pointing to high values
- Negative skew: scored bunched at high values with the tail pointing to low values
BSE: What is highest in a positive skew? Why?
Mean > median > mode
B/c the outliers are pulling/increasing the value of the mean
BSE: What is normally highest in a negative skew?
Mode > Median > Mean
BSE: What is Kurtosis?
A distribution with an abnormal distribution
BSE: What does bimodal distribution suggest?
- Suggests 2 populations within a sample
- i.e. “students who understand neuro vs students who do not understand neuro”
BSE: Level of Measurement-Give examples of Ratio
- Height
- Weight
- BP
- Kelvin
BSE: Levels of Measurement-give examples of nominal
- Variables that have two or more categories with no intrinsic order
- I.e. gender, blood type, types of housing
BSE: What is the best central tendency measure for nominal data?
Mode
BSE: What is the best central tendency measure for skewed or ordinal data?
Median
BSE: What is ordinal data?
- Variables with two or more categories
- “How much do you like my sweater?” “Not very much,” “it’s okay,”Yes, a lot”
- But not able to place a definitive value on
BSE: When is mean the best central tendency measure?
- When using data distribution that is continuous and symmetrical, such as when your data is normally distributed
BSE: What is the most appropriate measure of central tendency when the data has outliers?
Median
BSE: What is the best central tendency for ratio data?
Mean
BSE: Levels of Measurement-what is the difference between Interval and Ratio?
- Interval: variables for which their central characteristics is that they can be measured along a continuum and they have a numerical value
- Ratio: Interval variables, but with the added condition that ZERO of the measurement indicates that there is none of that variable. I.e. 0 BP, 0 Ht, 0 Wt
BSE: In a box plot, what does the box represent?
Median
CPR: What portion of the medulla controls quiet inspiration?
- Dorsal respiratory group stimulates the diaphragm via the phrenic nerve for quiet inspiration
CPR: T/F-the dorsal respiratory group controls quiet inspiration while the ventral respiratory group controls quiet expiration
False
DRG controls quiet inspriation
VRG controls active expiration and inspiration, only
- Quiet expiration does not require any muscles or stimulation
CPR: At what PaO2 will ________________ chemoreceptors start to fire?
- Peripheral
- When PaO2 is less than 70 mmHg will start to increase ventilation
- Strongest activity when PaO2 is less than 60 mmHg
CPR: Normally, what is the main driving force for respiration?
- Driven by CO2
- O2 drives some, but not as powerful for influencing respiration
CPR: The central chemoreceptors for neural control of respiration are considered neurons. Describe the implications of this.
- Neurons can be blocked such that in general anesthesia uses extremely high levels of CO2/drugs to override and inhibit central chemoreceptors
- The inhibition of central chemoreceptors will not be able to stimulate the DRG
CPR: T/F-changes in PaCO2 and arterial pH have the same influence of respiration due to their relationship in the bicarbonate buffer
- False, arterial CO2 levels are more powerful in controlling respiration rate
CPR: What is the PaCO2 in metabolic acidosis?
- PaCO2 level is lower than normal b/c there is hyperventilation to reduce the H+ content
CPR: Describe the PaCO2 influences in Respiration rate during the following conditions:
1. Metabolic acidosis
2. Sleep
3. Narcotics
4. Anesthesia
- PaCO2 sensitivity is high so strong effect on RR
- PaCO2 sensitivity is lower, so can have higher PaCO2 levels before respiration rate changes
- PaCO2 sensitivity is increasingly lower, so body allows higher PaCO2 levels
- PaCO2 sensitivity is lowest so high PaCO2 allowed with minimal changes to respiration rate
CPR: What is the pneumotaxic center?
- Decreases inhibition of the DRG (dorsal respiration center)
CPR: What does the Apneustic center do?
Stimulates DRG to increase inspiration
CPR: When levels of O2 are normal in the blood, describe how it influences respiration
- When O2 levels are normal there is no influence to respriation
- Only when PaO2 < 60 mmHg does O2 levels of influence respiration
CPR: What are J receptors?
- Located in the interstitial space of the lung
- Respond to irritation of the lungs and reduce the depth of breathing
- Induce shallow, rapid breaths of patients with inflammation or edema
CPR: How does body temperature influence respiration?
- Think increased body heat, increased metabolic work which produces CO2
- The body will then need to remove the increased levels of CO2 which is done by influencing and increasing respiration
CPR: What is hypercapnia?
- High level of CO2 in the blood
CPR: In COPD exacerbation, why don’t treat with pure O2?
- In obstructive diseases can have hypercapnia and hypoxemia
- The hypercapnia needs to be reduced by increasing expiration
- When CO2 levels are high and O2 levels are sufficiently low (< 60 mmHg) then O2 will be the main driving force for respiration
- If administer pure O2 too fast then the body will think the levels are okay and return respiration back to normal which will disallow removal of CO2/H+ = Acidosis
CPR: Describe why PaCO2 does not drive ventilation during exercise
- During exercise, the ventilation will go up correspondingly with the PaCO2 production
- Getting rid of PaCO2 at same rate as production so PaCO2 level is stable and thus cannot influence RR
CPR: T/F-Increased ventilation during exercise is the driven by PaCO2 levels rather than PaO2
- False, the exact mechanism of increased ventilation is not influenced by O2 or CO2!
CPR: Describe the changes in PaCO2, PaO2, pH, and O2-Hb curve during exercise
- PaCO2 & PaO2 not changed due to increased ventilation matching production and needs of body
- pH likely doesn’t change but can lower due to lactic acid production
- O2 affinity for Hb will lower to be able release O2 too tissues & Curve will shift right
CPR: What is Cheyne-Stoke breathing?
- Occurs in patient with stroke or heart failure & is an abnormal response to increase in PaCO2
- Takes longer for the brain to respond to increased levels of PaCO2
BSE: What term corresponds to this definition:
It also includes facilitating the
informed consent process and recommending mutually agreed-upon
diagnostic and/or therapeutic steps, or health promotion and disease
prevention strategies
Informed Consent
BSE: What term matches this definition:
- KNOWLEDGE OF OSTEOPATHIC PRINCIPLES, PRACTICE, AND OMT
- Compare and contrast the relative value, advantages, and disadvantages of different OMT techniques.
Competency Domain 1
BSE: What is the goal of competency domain 3?
- Have knowledge and be able to explain principles or health, disease, and diagnostic and tx options to patients
BSE: What is the purpose of competency domain 4?
Communicate best clinical evidence, including osteopathic principles and practice, to patients and colleagues
BSE: What is the purpose for section 6.1 of informed consent?
- Explain and apply the ethical principles of autonomy
- Identify the patient’s ability to participate in shared decision-making
BSE: What is the goal of section 6.2 of informed consent?
- Exhibit respect and compassion for the patient’s autonomy
BSE: What is the purpose of Informed Consent section 6.3?
- Respect patient autonomy and the right of the patient to be fully involved in decisions about care
BSE: What is the goal of Informed consent section 6.4?
- Properly establish the physician-patient relationship by examining, diagnosing, and treating in a consensual manner, and conscientiously maintaining the relationship consistent with the proper legal and ethical standard
BSE: What is informed consent competency domain section 7?
- Work to improve and promate care and culture of patient savety
- Maximize their resources to benefit the community at large
- Only one of the 7 competencies that doesn’t explicitly refer to informed consent or closely related concepts
BSE: What is Torts?
- Compensation of victim
- Plaintiff vs. Defendant
- Civil level
BSE: What is the difference between battery and assault ?
- Battery: harmful or offensive contact
- Assault: an act intended to cause, apprehension of harmful or offensive contact
BSE: List 2 occurences requiring informed consent not including medical treatment, surgery, or examination
- Participation in medical research
- Participation in medical teaching experiences
BSE: What are 4 key concepts to obtaining informed consent in Kentucky?
- Reasonable physican standards
- Reasonable patient standards
- Linguistic barriers
- Cultural barriers
BSE: What is a confidence interval?
- Using the sample mean to estimate the true mean
- Related to inferential statistics
BSE: What is a confidence interval?
- Using the sample mean to estimate the true mean
- Related to inferential statistics
BSE: ______ of values are found within 2 standard deviations from the mean
95% of values would be found within 2 SD from the mean which helps explain confidence interval
BSE: 95% of values would be found within ____ SD
95% of values would be found within 2 standard deviations from the mean
BSE: What is the central limit theorem?
- If your sample size is large enough, the distribution of means will approximate the normal distribution even if the population is not normal (Gaussian)
BSE: How do you calculate standard error?
SE = σ / sqr rt (n)
where σ = sample standard deviations (95%)
n = population
BSE: What is the difference between a 99% and 95% confidence interval?
- 99% confidence interval is wider and less precision but more accurate because true mean more likely to be present
- 95% confidence interval is more precise because narrower. But less accurate because true mean could lie outside
BSE: How many standard deviations are within 99% of the confidence interval?
3 standard deviations
BSE: 2.5 Standard deviations lie within _____% of a confidence interval
99
BSE: When do you use standard error?
- Sample > 30
- Sampling distribution is normal with a mean = to a population mean
- SD equal to the standard error of the mean
- Standard error is the SD of the same means
BSE: What is standard error?
- Standard error is the standard deviation of sample means
- It is a measure of how representative a sample is likely to be of the population
BSE: What is a better way to say “95% confidence interval contains the true mean with probability 95%”
“ 5% change the interval does not contain the true mean”
BSE: In stats, if there are overlaping confidence intervals, what can you determine
- If the confidence intervals overlap tells you the mean could come from the same population
- Can suggest there is no effect, of per say a drug
BSE: In stats, what can you deduce if confidence intervals do not overlap?
- Suggests the samples come from two different populations such that if a medicine was being observed, it might suggest the drug produces a different effect
- The result would be statistically significant if P < 0.05
BSE: What can be said about the SD if the sample size increases?
The SD will decrease since the denominator increases
BSE: Central tendency and variability is an example of __________ statistics
descriptive
BSE: Confidence intervals are a type of _____________ statistics
Inferential stats
BSE: Give an example of independent probability & define
- If flip coin once & get heads.
- Will the next flip be dependent on the previous flip? No
- Occurrence of one thing tells you nothing about the occurrence of another
BSE: When calculating independent probability, what do you do to the chances
Multiply
Chance of red hair = 0.01
Chance of having syphilis = 0.05
Chance of meeting a redhead with syphilis = 0.01 * 0.05 = 0.0005 or 0.05% chance
BSE: What is nonindependent probability and give an example?
- One probability relies on the previous
- I.e. have a bag of orange and blue balls. If randomly draw 1 orange ball, the next draw will be affected by the first.
BSE: How do you calculate nonindependent probability concepts
- Multiply but will likely have fractions, i.e have a bag of 10 balls, 5 orange, 5 blue
- Draw 1 orange ball is 5/10 chance
- Next draw, 1 orange ball 4/9 chance
- Portability of drawing 2 orange balls was 5/10 * 4/9 =0.222 or 22% chance
BSE: What is a type I error?
- False positive
- The probability of getting a Type I error is the α-level (usually 0.05)
- Assuming no effect in the population, if we repeated an experiment 100 times, we would expect on five occasions to obtain a test statistic large enough to make us think there is a genuine effect
BSE: Independent events cannot be ____________ exclusive
Independent events cannot be mutually exclusive
BSE: Independent events cannot be ____________ exclusive
Independent events cannot be mutually exclusive
BSE: What is a type II error?
- False negative - Occurs when we believe that there is no effect in the population when, in reality, there is - The probability of getting a Type 2 error is the β-level (often 0.2) - If we replicated an experiment 100 times in which an effect exists, we would fail to detect the effect in 20 of the experiments
BSE: As the probability of making a Type I error _________, the probability of making a Type II error __________
As the probability of making a type I error decreases, the probability of making a type II error increases
BSE: What is a null hypothesis
- Says there is no difference between 2 things
BSE: How do can you change statistical power?
- Variation
- Sample size (preferred large)
- Significance level chosen
- Size of effect
- Little scatter (smaller SD)
BSE: When comparing variables, what common statistical tests are used for numerical variables?
- Comparing 2 groups being evaluated? T-Test
- Comparing 3 or more groups being evaled? ANOVA
BSE: When comparing variables, what type of statistical tests is used for categorical or proportional variables?
- Small sample size? Fisher exact test
- Large sample size? Chi-squared test
BSE: How can you discern when to use a paired T test?
- When what’s being evaluated have shared characteristics, i.e. husband and wives being compared-they live together
- Same sample being evaluated, like pre-test vs post-tests for a class, same group is being evaluated
BSE: What can be inferred about the measures of central tendency in a normal distribution?
Mean, median, mode are identical
BSE: How will an increase in prevalence change positive predictive value?
An increase in prevalence will increase positive predictive value
BSE: Describe the relationship of prevalence and incidence for short term diseases and give an example
- Prevalence = incidence for short duration diseases
- Ex. common cold prevalence = incidence since it goes away on its own
BSE: How do you calculate incidence? What is important to remember when calculating this?
- Incidence = new cases / population that can get disease
- Need to remove people who have the disease at the start of observation from the population studied, i.e. population who get lung cancer in a 10 yr period, if 2500 have lung cancer at the start of the study they don’t count toward the denominator
BSE: If something is deemed very precise, what can you infer about its statistical power?
High precision = high statistical power
BSE: How to calculate sensitivity?
Sensitivity = A / (A+C)
BSE: How to calculate Specificity?
Specificity = D / (D+B)
BSE: How to calculate negative predictive value?
NPP = D / (C+D)
BSE: What does negative predictive value mean?
- If you test negative for a disease, what is the likelihood you don’t have the disease
BSE: What does positive predictive value mean?
How do you calculate it?
- PPV = A / (A+B)
- If you test positive for a disease, what is the likelihood you have the disease
BSE: How to calculate Sensitivity
Sensitivity = True positive / True positive + False negative
BSE: How to calculate specificity?
Specificity = True Negative / False positive + True negative
BSE: How to calculate PPV?
Positive predictive value = True positive / True positive + False positive
BSE: How to calculate NPV?
Negative predictive value = True negative / False negative - True negative
BSE: How to calculate prevalence?
Prevalence = True Positive + False negative / population
BSE: Assume 100% sensitivity and 100% specificity of a test. If the sensitive test is negative, what can you infer?
- SN-N-OUT: Sensitive test (if negative) will rule out the disease
- If a pap smear is negative, you can rule out cervical cancer
BSE: Assume 100% sensitivity and 100% specificity of a test. If the specific test is positive , what can you infer?
- SP-P-IN
- If specific test is positive will rule in disease
- If colposcopy is positive, person has cervical cancer
- A specific test is used for ruling in a disease, as it rarely misclassifies those without a disease as being sick
BSE: A perfectly sensitive test will have _______ True positive and ______ true negative
A perfectly sensitive test will have 100% true positive and 100% true negative
BSE: Why do we use sensitive tests for screenings? What kind of test is used after screening? Why?
- With sensitivity, goal is to have mostly true positive with some false positive
- Use specific test after screening
BSE: With increasing disease prevalence, you get:
With increasing disease prevalence, you get a higher PPV & Lower NPV
BSE: If prevalence is low, what can you infer?
If prevalence is high, what can you infer?
- If prevalence is low, false positive is high
- If prevalence is high, false negative is high
CPR: What is atelectasis?
a partial or complete collapse of the lung that occurs when the air sacs within the lung, called alveoli, lose air
CPR: Is Bronchiectasis a restrictive or obstructive lung disease?
Describe how this would impact FEV1/FVC ratio
Obstructive
- FEV1 is decreased b/c harder to expel air since increased elastin
- FVC is decreased but FEV1 is decreased more
- Ratio reduced overall
CPR: Idiopathic pulmonary fibrosis & ankylosing spondylitis are restrictive or obstructive lung diseases?
Describe how this would impact FEV1/ FVC ratio
- Idiopathic pulmonary fibrosis & ankylosing spondylitis are restrictive lung diseases
- FEV1 is decreased due to decrease compliance
- FVC is decreased
- both decrease proportionally, entire ratio goes down
CPR: How to calculate Minute ventilation?
Vm = Tidal volume * RR
CPR: _________ agonists increase smooth muscle contraction of the bladder and arteriole
alpha 1 agonists increase smooth muscle contraction of the bladder and arteriole
CPR: From where do the greater splanchnic nerve arise?
The greater splanchnic nerve arises from the sympathetic trunk at levels T5-T9
CPR: What is the drug of choice for pulmonary artery hypertension? By what mechanism does it act? This drug is known to be expensive, what is a cheaper alternative
- Bosentan
- Endothelin-1 receptor antagonist
- Sildenafil/Viagra
OPP: When using AP XR for identifying structures that pass through the diaphragm, what is the pneumonic and what does it mean?
- ” I 8 10 eggs at 12
- IVC = T8
- Esophagus = T10
- Aorta = T12
OPP: How can you tell if a chest XR is captured during inspiration
- Should show 10th-11th posterior ribs
OPP: Where is the costophrenic angle and what is significant about it in XR?
- Located laterally and distally on the ribs proximal to the hemidiaphragm
- If there is blunting at this angle can indicate pleural effusion
OPP: On chest XR, what is the order of structures from superior to inferior on patient L side of body?
- Highest is aortic arch
- Second is pulmonary artery
- Lower is atrial appendage
OPP: On chest XR, what is the order of structures from superior to inferior on patient R side of body?
- Superior vena cava
- Ascending Aorta
- IVC
CPR: Losartan is a:
explain how it effects the heart and lungs
- Losartan is a nonselective beta-adrenergic antagonist
- Heart: B1 would increase HR & contractility normally, and blocker would decrease this
- Lungs: B2 would normally bronchodilator but if blocked, bronchoconstriction
BSE: If you increase the cutoff value for + vs - tests, what will occur?
- If increase the cutoff value for a + or - test, will increase false negative values and decrease false positive
BSE: What happens if you lower the cutoff value for + or - test results?
Lower the cutoff value for + or - test results will increase false positives and decrease false negatives
BSE: What do you want to be high in all screening tests?
- Want high sensitivity in screening tests
BSE: What is an ROC curve and what does it compare?
- ROC - receiving operating characteristic curve
- Compares sensitivity and false positive rate
CPR: What does a R shift on hemoglobin-oxygen disassociation curve mean?
A “right shift” on the hemoglobin-oxygen dissociation curve means that hemoglobin has a decreased affinity for oxygen, causing it to release oxygen more readily to tissues
CPR: What does a L shift on the Oxygen-Hb disassociation curve represent?
A “left shift” on the hemoglobin oxygen dissociation curve means that hemoglobin has an increased affinity for oxygen, making it more likely to bind to oxygen and less likely to release it to the tissues, essentially indicating that the blood is holding onto oxygen more readily than usual; this can occur in situations like alkalosis (high pH) or hypothermia
BSE: Define “likelihood”
- Likelihood that a given test result would be expected in a patient with the target disorder compared to the likelihood that the same result would be expected in a patient without the target disorder
BSE: What does LR + ratio mean?
LR + = probability of a positive test in an individual with a disease / probability of a positive test in an individual without a disease
BSE: What does LR - ratio mean?
LR - = probability of a negative test in an individual with a disease / probability of a negative test in an individual without a disease
BSE: What does 1-specificity = ?
1 - specificity = False positive rate
BSE: What does 1 - Sensitivity = ?
1 - sensitivity = False negative rate
BSE: What is a “good” LR + value?
What is a preferred LR- value?
- LR+ want to be greater than 10
- LR - want to be less than 0.1
BSE: What does an LR+ > 10 indicate?
LR + > 10 indicates a highly specific test
BSE: What does an LR- < 0.1 indicate?
LR- < 0.1 indicates a highly sensitive test
BSE: How do you calculate risk?
Risk = new cases/ persons at risk
BSE: For epidemiology, in case control studies, what measure of association might you use?
Odds ratio
BSE: For epidemiology, for cohort studies, what measure of association might you use?
Use risk
BSE: Define odds ratio & when to use it?
- Used in case-control studies
- Represents the odds of exposure among cases (disease) vs odds of exposure among controls (not disease)
BSE: Define relative risk and when to use it
- Used in cohort studies
- Risk of developing disease in the exposed group divided by risk in the unexposed group
BSE: For rare diseases what can you infer about odds ratio and relative risk?
For rare diseases, you can approximate odds ratio approximates relative risk
BSE: When have an odds ratio or relative risk and performing confidence interval, what must you not do
Cannot have 1 in the confidence interval or it is not statistically significant
BSE: What does number needed to harm mean and how to calculate it?
- NNH: How many people need to give a drug to cause harm
- NHH = 1/ Absolute Risk
BSE: Why do you want a low NNT as 1?
Why do you want a high NNH?
- Low NNT of 1 means, give a drug and it treats everyone, lower number = better treatment
- Want to have to give a lot of the drug to people in order to harm/ number of patients who need to be exposed to a risk factor for 1 patient to be harmed
BSE: How do you calculate attack rate?
AR = ill / (ill + well) X 100 (during a time period)
BSE: How to calculate odds ratio?
OR = ad/bc
BSE: How to calculate Relative risk?
RR = [a/ (a+b) ] / [ c/ (c+d) ]
CPR: _____ decreases O2 release from HbO2 at the tissue level
CO, Carbon monoxide decreases O2 from HbO2 at the tissue level
CPR: In normal lungs, where is V/Q ratio the highest? Why?
- At the apex of the lung
- There is lowest at the apex b/c there are fewer alveoli
- There is less perfusion due to gravity
CPR: What does Alpha 1 Antitrypsin do?
- AAT is a protien that forms in the liver and moves to the lungs
- Tunrs off neutrophil elastase which is an enzyme that helps fight lung infections but can destroy healthy tissue
- AAT is inactive while lungs are healthy
CPR: What is CaO2?
- CaO2 is the total arterial oxygen content in blood
- Depends on both Hb O2 saturation and PaO2 in the blood
CPR: In exercise, why is there no change in PaCO2?
B/c it should be carried and bound to Hb for exhalation & removal
- Increased production of PaCO2 but it is removed and does not build up
BSE: What does positive predictive value you tell you?
The likelihood that a person who has a positive result does have the disease, condition, biomarker, or mutation (change) in the gene being tested
BSE: Positive predictive value is a way of measuring:
PPV is way of measuring how accurate a specific test is
BSE: What does NPV tell you?
NPV is the likelihood that a person who has a negative test result indeed does not have the disease, condition, biomarker, or mutation (change in the gene being tested)
BSE: The negative predictive value is a way of measuring:
NPV is a way of measuring how accurate a specific test is
BSE: How does PPV change with increasing disease prevalence?
- Since PPV is the eval of true positives compared to all the exisiting positives (including false positive)
- The higher the disease prevalence, the higher the PPV b/c there will be more true positives
BSE: How does NPV change with increasing disease prevalence?
- Since NPV is comparing the true negatives against all the existing negatives (including false negatives), the higher the disease prevalence the lower the NPV
BSE: How to calculate prevalence?
True Positive + False Negative / Total population ( True positive + False negative + False positive + True negative)
BSE: What is dichotomous data?
A type of categorical data that has only 2 caterogies/values
i.e. Boy/Girl, Yes/No, Today/Tomorrow
BSE: When do you use a t-test? Give an example?
- T test: checks difference b/t means of 2 groups
- Ex. Comparing mean BP b/t men and women
BSE: When do you use an ANOVA test? Give an example
- ANOVA: Checks differences between the means of 3 or more groups
- Comparing the mean BP b/t members of 3 different ethnic groups
BSE: When to use a chi-square test? Give an example
- Use a chi square test to check differences between two or more percentages of proportions of categorical outcomes
- I.e. Comparing the percentage of members of 3 different ethnic groups who have essential HTN
- Note here the HTN is not a numerical data
BSE: When variables being compared are categorical, what common statistical tests would you use?
- Chi square for small sample size
- Fisher exact for large sample size (but not too large)
BSE: If the variables being compared are numerical, what common statistical tests are used?
- T test for 2 groups
- ANOVA for 3+ groups
BSE: What are 3 types descriptive studies?
- Cohort studies
- Case series
- Case reports
BSE: (Kingery) what are Cohort studies?
Defined populations which are followed in an attempt to determine distinguishing subgroup characteristics
BSE: What are 3 reasons for development of Evidence based medicine?
- Cope with information overload
- Cost control
- Public impatience for best diagnostics and tx
BSE: What are the 5 rules of Title II in HIPPA?
- Privacy rule
- Transactions & codes
- Severity rule
- Unique identifiers
- Enforcement
BSE: What set of data has the greatest strength of evidence?
Evidence based medicine
BSE: What is the PICO tool?
- Population
- Intervention
- Comparison
- Outcome
BSE: In PICO tool, what does “comparison” refer to?
- The alternative or standard treatment that is being compared to the intervention in the clinical question
- i.e. standard blood glucose control for the clinical question “in adult patients with diabetes, does intensive blood glucose control compared to standard blood glucose control result in reduction in HbA1C studies?”
BSE: What is a confounding bias?
Occurring when the observed association between exposure and outcome is influenced by a third variable that is associated with both the exposure and outcome
BSE: What are 6 steps to assessing strength of evidence?
- Identify research question
2 Identify the available evidence - Evaluate the quality of the evidence
- Synthesize evidence
- Draw conclusions
- Communication strength of evidence
BSE: What are 5 components to evaluate generalizability?
- Study population
- Study setting
- Intervention characteristics: is intervention or Tx similar to other interventions
- Study design: minimizes bias
- External validation: have results been replicated in other studies
BSE: What is the purpose of Title II of HIPPA?
Rules to standardize the electronic exchange of patient-identifiable, health related information, based on electronic data interchange
BSE: In Title II of HIPPA, what does privacy rule protect?
Protects health information
BSE: Per Title II of HIPPA, what is security rule?
Specifically dealing with electronic protected health information
BSE: Who enforces penalty if there is violation of HIPPA?
Department of Health & Human services office of civil rights imposes penalty
BSE: Why is prevalence almost equivalent to incidence in short duration diseases?
- People are cured as fast as prevalence
- i.e. common cold
CPR: How to calculate incidence?
Incidence = # of new cases/ # of people at risk
BSE: How does increased survival time change incidence and prevalence?
- Increased survival times does not change incidence
- Increased survival increases prevalence
BSE: How does increased mortality change prevalence and incidence?
- Increased mortality decreases prevalence and does not change incidence
BSE: How does fast recovery time change incidence and prevalence?
- Fast recovery time decreases prevalence does not change incidence
BSE: How does extensive vaccine administration affect prevalence and incidence?
- Vaccine administration decreases both incidence and prevalence
BSE: How do decreased risk factors change prevalence and incidence?
Decreased risk factors decrease prevalence and incidence
BSE: How to calculate prevalence?
P = Case # / Total population
BSE: How does increased precision change standard deviation and statistical power
Increased precision decreases standard deviation
Increased prevision increases statistical power (1 - β)
BSE: How does systematic error change accuracy?
Systematic error decreases accuracy of a test
BSE: How does random error change precision?
Increased random error decreases precision
BSE: What is the difference between systematic error and random error?
- Random error: a mistake - Systematic error: i.e. caused by calibration mistake on a device
BSE: If a test is sensitive, if the result is negative what can you determine?
SN N OUT
- Negative will rule out disease
BSE: If you have a specific test, with a positive result, what can you infer?
Specific test with + result will rule disease in
BSE: What test is normally the screening test? What about follow up testing
Screening use sensitive test
Secondary test, i.e. breast biopsy = use specificity tests