Public Health Flashcards
Systematic errorcan only be decreased by ______.
Eliminating the source ofbias
Whilerandom error (random errorsin data) can be decreased by repeating measurements and/or increasing the sample size.
Random error (random errorsin data) can be decreased by ___________.
Repeating measurements and/or increasing the sample size.
A test with a highprecisionwill have________.
Minimalrandom error
Precisionimproves with _______.
Decreasedstandard deviationand increasedpowerof a statistical test.
Interrater Reliability
The test yields the same results when performed by different researchers.
Parallel-test Reliability
The reliability of a new test is compared with an established test.
The new test determines the reliability of a test in comparison to another test, the reliability of which has already been established.
Similar statistical results imply a similar degree of reliability.
Test-retest Reliability
The test yields the same results when repeated on the same subjects.
A test with high validity will have _________.
Minimal systematic error and bias.
Internal Validity
Extent to which a study is free of error (most often in the form ofbias) and the results are therefore true for the study sample.
Highinternal validitycan be achieved by matching study groups according to age, sex, and other characteristics, and observing measures to reduce systemic errors (bias) to a minimum
Highinternal validitycan be achieved by ________.
Matching study groups according to age, sex, and other characteristics, and observing measures to reduce systemic errors (bias) to a minimum
External Validity
Refers to whether study results can be extrapolated from a sample population to the general population (generalizability).
A test has high external validity if results in the sample group reflect the actual figures (eg, prevalence) in the general population.
A study with highexternal validityresults can be reproduced in different sample groups and has highinternal validity
A test has high external validity if __________.
Results in the sample group reflect the actual figures (eg, prevalence) in the general population.
A study with _____external validityresults can be reproduced in different sample groups and has ____internal validity
high; high
Berkson Bias
Individuals in sample groups drawn from ahospitalpopulation are more likely to be ill than individuals in the general population.
Attrition Bias
Participants lost to follow up have a different prognosis than those who complete the study.
Most commonly seen in prospective studies.
Risk that the remaining participants differ significantly from those lost to follow up.
Susceptibility Bias
One disease predisposes affected individuals to another disease, and the treatment for the first disease is mistakenly interpreted as a predisposing factor for thesecond disease.
Survival biasalso known as__________.
Prevalence-incidencebiasandNeyman bias
Survival bias (also known asprevalence-incidencebiasandNeyman bias)
When observed subjects have more or less severe manifestations than the standard exposed individual.
If individuals with severe disease die before the moment of observation, those with less severe disease are more likely to be observed.
If individuals with less severe disease have a resolution of their disease before the moment of observation, those with more severe disease are more likely to be observed.
Most commonly occurs incase-controlandcross-sectionalstudies.
Sampling bias also known as __________.
Ascertainment bias
Sampling bias (ascertainment bias)
Occurs when certain individuals are more likely to be selected for a study group, resulting in a nonrandomized sample.
This can lead to incorrect conclusionsbeing drawn about the relationship between exposures and outcomes.
Limitsgeneralizability.
Types ofsampling bias:
- Nonresponse bias → Nonresponder characteristics differ significantly from responder characteristics because nonrespondersdo not returninformation during a study (e.g., subjects do not return a call).
- Healthy worker effect → The working population is healthier on average than the general population. Severely ill individuals do not usually work, so any sample consisting of only subjects that work is not representative of the general population.
- Volunteer bias → Individuals who volunteer to participate in a study have different characteristics than the general population.
Types ofSampling Bias (ascertainment bias)
- Nonresponse bias → Nonresponder characteristics differ significantly from responder characteristics because nonrespondersdo not returninformation during a study (e.g., subjects do not return a call).
- Healthy worker effect → The working population is healthier on average than the general population. Severely ill individuals do not usually work, so any sample consisting of only subjects that work is not representative of the general population.
- Volunteer bias → Individuals who volunteer to participate in a study have different characteristics than the general population.
Nonresponse Bias
Nonresponder characteristics differ significantly from responder characteristics because nonrespondersdo not returninformation during a study (e.g., subjects do not return a call).
Healthy Worker Effect
The working population is healthier on average than the general population.
Severely ill individuals do not usually work, so any sample consisting of only subjects that work is not representative of the general population.
Volunteer Bias
Individuals who volunteer to participate in a study have different characteristics than the general population.
Solutions to Selection Bias (Ascertainment Bias)
- Randomization
- Subjects are randomly assigned to the exposure and control groupsto ensure that both groups are roughly equal inbaseline characteristics(often displayed in a table, e.g., inrandomized controlled trials).
- Controls for both known and unknownconfounders
- Successful if possibleconfoundingcharacteristics (e.g., socioeconomic demographics,family history) are approximately equally distributed between the exposure and control groups - Ensure the sample is representative of the population of interest (e.g., incase-control studies).
- Ensure the correct reference group is chosen for comparison.
- Collect as much data on the characteristics of the participants as possible.
- Nonresponder characteristics should not be assumed. Instead, undisclosed characteristics of nonresponders should be recorded as unknown.
Referral Bias
Results when patients are sampled from specialized medical centers and therefore they do not represent the general population.
Recall Bias
Awareness of a conditionby subjectschangestheir recall of relatedrisk factors(recall a certain exposure).
Common inretrospective studies.
Solution → reducing timeto follow upin retrospective studies (e.g.,retrospective cohort studiesorcase-control studies)
Recall Bias Solutions
Reducing timeto follow upin retrospective studies (e.g.,retrospective cohort studiesorcase-control studies)
Allocation Bias
Systematic difference in the way that participants are assigned to treatment and control groups.
Solution → randomization
Allocation Bias Solutions
Randomization
Late-look Bias
Sending a survey out to people diagnosed with a fatal illness five years after diagnosis will preferentially sample those with a low grade disease (or few comorbidities).
Types of Selection Bias
- Berkson bias
- Attrition bias
- Susceptibility bias
- Survival bias (also known asprevalence-incidencebiasand Neyman bias)
- Sampling bias (ascertainment bias)
Types of Information Bias
- Measurementbias
- Reporting bias
- Interviewer bias
Information Bias
Incorrect data collection, measurement, or interpretation that leads to misclassification of groups or exposure.
Information is gathered differently between the treatment and control groups.
Insufficient information about exposure and disease frequency among subjects.
Solution → standardize data collection.
Types ofinformation bias:
1. Measurementbias → anysystematic errorthat occurs when measuring the outcome
- Reporting bias: a distortion of the information from research due to the selective disclosure or suppression of information by the individuals involved in the study. Can involve the study, design, analysis, and/or findings. Results in underreporting or overreporting of exposure or outcome
- Interviewer bias: Different interviewing approaches prompt different responses by interviewees, which results in researchers finding differences between groups when there are none.
Information Bias Solutions
Standardize data collection
MeasurementBias
Anysystematic errorthat occurs when measuring the outcome
Reporting Bias
A distortion of the information from research due to the selective disclosure or suppression of information by the individuals involved in the study.
Can involve the study, design, analysis, and/or findings. Results in underreporting or overreporting of exposure or outcome
Interviewer Bias
Different interviewing approaches prompt different responses by interviewees, which results in researchers finding differences between groups when there are none.
Misclassification Bias
Results from an incorrect categorization of subjects regarding their exposure status, outcome status, or both.
Cognitive Bias
Thepersonal beliefsof the subjects and/or investigators.
Solution → Use placebo, blinding and prolong the time of the observation to monitor long term effects.
Types of cognitive bias:
- Response bias → Study participants do not respond truthfully or accurately because of the manner in which questions are phrased (e.g., leading questions) and/or because subjects interpret certain answer options to be more socially acceptable than others.
- Observer bias(experimenter-expectancyeffect orPygmalion effect) → The measurement of a variable or classification of subjects is influenced by the researcher’s knowledge or expectations.
- Confirmation bias → The researcher includes only those results that support their hypothesis and ignores other results.
- Placebo and nocebo effects → A placebo or nocebo affects subjects’ preconceptions/beliefs about the outcome.
Cognitive Bias Solutions
Use placebo, blinding and prolong the time of the observation to monitor long term effects.
Types of Cognitive Bias
- Response bias
- Observer bias(experimenter-expectancyeffect orPygmalion effect)
- Confirmation bias
- Placebo and nocebo effects
Response Bias
Study participants do not respond truthfully or accurately because of the manner in which questions are phrased (e.g., leading questions) and/or because subjects interpret certain answer options to be more socially acceptable than others.
Observer Bias is also known as ___________.
Experimenter-expectancyeffect orPygmalion effect
Observer Bias(experimenter-expectancyeffect orPygmalion effect)
The measurement of a variable or classification of subjects is influenced by the researcher’s knowledge or expectations.
Confirmation Bias
The researcher includes only those results that support their hypothesis and ignores other results.
Placebo and Nocebo Effects
A placebo or nocebo affects subjects’ preconceptions/beliefs about the outcome.
Performance Bias
Researchers provide different levels of attention/care to different groups, or subjects change their responses when they become aware of which group they are allocated to.
For example:
1. Procedure bias → When patients or investigators decide on the assignment of treatment and this affects the findings. The investigator may consciously or subconsciously assign particular treatments to specific types of patients (e.g., one group receives a higher quality of treatment; patients in treatment group spend more time in highly specialized hospital units ). Solution → blinding
- Hawthorne effect → subjectschange their behavioronce they areawarethat they are being observed. Especially relevant in psychiatric research. This type ofbiasis difficult to eliminate. Solution → blinding
Procedure Bias
When patients or investigators decide on the assignment of treatment and this affects the findings.
The investigator may consciously or subconsciously assign particular treatments to specific types of patients (e.g., one group receives a higher quality of treatment; patients in treatment group spend more time in highly specialized hospital units).
Solution → blinding
Procedure Bias Solution
Blinding
Hawthorne Effect
Subjectschange their behavioronce they areawarethat they are being observed.
Especially relevant in psychiatric research.
This type ofbiasis difficult to eliminate.
Solution → blinding
Hawthorne Effect Solution
Blinding
Cofounding Bias
Anythird variablethat has not been considered in the study but thatcorrelates with the exposure and the outcome. Aconfoundercan be responsible for the observed relationship between the dependent andindependent variables.
Solutions to cofounding:
- Performmultiple studieswith different populations.
- Randomization(randomized controlled trials)
- Crossover studydesign
- Restriction (epidemiology) → a study design in which only individuals who meet certain criteria are included in the study sample (e.g., only male individuals with a particular disease are included in a study to avoid the influence ofgenderon the exposure and outcome). Disadvantages → limit generalizability and makes obtaining a large sample group difficult
- Matching (epidemiology) → A study design in which study participants aregrouped intopairswith similar attributes. Commonlyused incase-control studiesto minimizeconfounding. Disadvantages →Does not completely eliminateconfounding, can introduceconfoundingif the effect of the matching factor on disease occurrence cannot be studied anymore, can introducebias
- Stratified analysis
Solutions to Cofounding
- Performmultiple studieswith different populations.
- Randomization(randomized controlled trials)
- Crossover studydesign
- Restriction (epidemiology) → a study design in which only individuals who meet certain criteria are included in the study sample (e.g., only male individuals with a particular disease are included in a study to avoid the influence ofgenderon the exposure and outcome). Disadvantages → limit generalizability and makes obtaining a large sample group difficult
- Matching (epidemiology) → A study design in which study participants aregrouped intopairswith similar attributes. Commonlyused incase-control studiesto minimizeconfounding. Disadvantages →Does not completely eliminateconfounding, can introduceconfoundingif the effect of the matching factor on disease occurrence cannot be studied anymore, can introducebias
- Stratified analysis
Lead Time Bias
Early detection of disease ismisinterpretedasincreased survival (Lead time is the average length of time between thedetectionof a disease and the expectedoutcome).
Often discussed in the context of cancer screening.
Lead-timebiasoccurs whensurvival timesare chosen as an endpoint ofscreening tests.
Solution → Measure theback-endsurvival by adjusting for the severity of the disease at the time of diagnosis. The gold standard forscreening testeffectiveness is to usemortality ratesinstead of survival times.
Lead Time Bias Solutions
Measure theback-endsurvival by adjusting for the severity of the disease at the time of diagnosis.
The gold standard forscreening testeffectiveness is to use mortality ratesinstead of survival times.
Length Time Bias
Apparent improvementin the duration of survival for aterminal diseasewith along clinical course (e.g.,slow-growingtumor).
Often discussed in the context of cancer screening.
Solution → Arrange patients according to theseverityof the disease. Use arandomized controlled trialto allocate subjects into control and treatment groups.
Length Time Bias Solutions
Arrange patients according to theseverityof the disease.
Use arandomized controlled trialto allocate subjects into control and treatment groups.
Surveillance Bias
Outcome is diagnosed more frequently in a sample group than in the general population because of increased testing and monitoring.
Results in misleadingly high incidenceand prevalence rates.
Solution → Compare the treatment group to an unexposed control groupwith a similar likelihood of screening. Select an outcome that is possible in both the exposed and unexposed groups.
Surveillance Bias Solutions
Compare the treatment group to an unexposedcontrol groupwith a similar likelihood of screening.
Select an outcome that is possible in both the exposed and unexposed groups.
Effect Modification
A third variableinfluences the effect of an exposure on an outcome.
Occurs when the exposure has adifferent effecton the control and treatment groups.
Not considered a type ofbiasin itself, but rather a biological phenomenon (i.e., the exposure has a different impact in different circumstances).
Example:A certain drug works in children, but does not have any effect on adults.
Solution →stratified analysis. Stratifyingparticipants into subgroups according to the third variable results in a stronger relationship in one subgroup. Stratified analysishelps to differentiateeffect modificationfromconfounding. When the population isstratifiedaccording to a factor, different results will be seen depending on whether it is aconfounderor aneffect modifier.
- Well-known examples of effect modification include:
1. Effect of estrogens on the risk of venous thrombosis (modified by smoking)
2. Risk of lung cancer in people exposed to asbestos (modified by smoking)
Effect Modification Solutions
- Stratified analysis.
- Stratifyingparticipants into subgroups according to the third variable results in a stronger relationship in one subgroup.
- Stratified analysishelps to differentiateeffect modification fromconfounding.
- When the population isstratifiedaccording to a factor, different results will be seen depending on whether it is aconfounderor aneffect modifier.
Cognitive biases can lead to diagnostic errors for example:
- Confirmation bias →tendency of an investigator to favor results that support his or her hypothesis or preconceived notions, ignoring other results.
- Anchoring bias → tendency to inappropriately rely on initial perception or information, which hinders laterjudgmentwhen new information becomes available (e.g., favoring a diagnosis proposed earlier despite new evidence)
- Availability bias → tendency to make judgments based on the availability of information in an individual’smemory (e.g., when a physician makes a premature diagnosis that comes to mind easily and quickly due to having seen several patients with a similar clinical presentation)
- Framing bias → the tendency to be influenced by how information is presented (e.g., the order of symptoms and/or emphasis placed on specific findings).
- Visceral bias → a tendency for clinical decisions to be influenced by positive or negative feelings towards the patient (e.g., doubts regarding symptoms when described by a patient withsubstance use disorder)
Confirmation Bias
Tendency of an investigator to favor results that support his or her hypothesis or preconceived notions, ignoring other results.
Anchoring Bias
Tendency to inappropriately rely on initial perception or information, which hinders laterjudgmentwhen new information becomes available (e.g., favoring a diagnosis proposed earlier despite new evidence)
Availability Bias
Tendency to make judgments based on the availability of information in an individual’smemory (e.g., when a physician makes a premature diagnosis that comes to mind easily and quickly due to having seen several patients with a similar clinical presentation)
Framing Bias
Tendency to be influenced by how information is presented (e.g., the order of symptoms and/or emphasis placed on specific findings).
Visceral Bias
Tendency for clinical decisions to be influenced by positive or negative feelings towards the patient (e.g., doubts regarding symptoms when described by a patient withsubstance use disorder)
Intention-to-Teat analysis
All patients who initially enrolled in the study, includingdrop-outs, are included in the analysis of study data.
Advantage:
- Allows the investigator or reader of the study to draw accurate conclusions regarding the effectiveness of an intervention
- Helps to reduceselection bias → Participants who arerandomizedare included in the analysis and analyzed according to the group they were originally assigned to.
- Preservesrandomization(a large dropout rate may be influenced by the treatment itself.Intention-to-treatanalysis preservesrandomizationby limiting theconfounding effects of dropout due to treatment)
Disadvantages
- Noncompliance can lead to a conservative estimate of the treatment effect.
- Heterogeneity might be introduced (e.g., noncompliant,drop-out, and compliant subjects are analyzed together).
- Susceptible totype II error
Per-Protocol Analysis
Treatment and control groups are compared using data from only those study participants who adhered to the study protocol.
Advantages:
- Improves the estimate of the real effect of treatment under optimal conditions
- Can be used innoninferiority trials,phase 1 trials, andphase 2 trials
Disadvantages:
- Loss ofrandomization(increasedselection bias)
- Overestimates the effects of the tested treatments (because it does not takedrop-outsinto account)
- Possibly a significant reduction in sample size
- Increases risk ofbias
- Excludes participants that did not adhere to the protocol or were lost to follow up
Stratified Analysis
Stratifyingparticipants into subgroups according to a third variable to eliminateconfounding(e.g., by age,gender, race).
Evaluation of the association between exposure and disease is performed within each stratum (e.g., crudeodds ratio).
Crudeodds ratioorrelative risk, as well as theirconfidence intervals, are used to measures the strength of the association (how different they are).
Eliminatesconfounding.
Example: evaluating the association betweenobesityand cardiovascular disease afterstratifyingby age.
Evidence Based Medicine
Practice of medicine in which the physician uses clinicaldecision-makingmethods based on the best available current research frompeer-reviewedclinical and epidemiological studies with the aim of producing the most favorable outcome for the patient.
Levels of Evidence
Method used inevidence-based medicineto determine the reliability of findings from a clinical and/or epidemiological study.
Level I –> findings from at least onehigh-qualityrandomized controlled study.
Level III –> Expert opinions
(Level I –> most reliable; Level III –> least reliable).
Level I of Evidence
Findings from at least onehigh-quality randomized controlled study.
(Level I –> most reliable; Level III –> least reliable).
Level III of Evidence
Expert opinions
(Level I –> most reliable; Level III –> least reliable).
Grades of Clinical Recommendation (according to Evidence Based Medicine Guidelines)
System developed by medical societies, healthcare regulatory entities, and governments to rate clinical evidence and create guidelines for clinical practice based on medical evidence.
From A-D (A highly recommended; D not recommended)
I –> evidence is insufficience
Prevalence / (1 - Prevalence) =
Incidence rate (IR) × Average duration of the disease (T)
Incidence Rate (IR) =
[Prevalence / (1 - Prevalence)] / average duration of the disease (T)
Birth rate
Number oflive birthswithin a specific time interval
Fertility Rate
Rate oflive birthsamongwomen of childbearing age(15–44 years) in a population within a specific time interval
Morbidity
Number of individuals in a population with a disease at a specific point in time (i.e., thedisease burdenin a population)
Mortality
Number of deaths in a population within a specific time interval
Standard Deviation
How much variability exists in a set of values, around the mean of these values.
Standard Error
An estimate of how much variability exists in a (theoretical) set of sample means around the true population mean.
Affected by extreme outliers
Mean
Most resistant to outliers. Can be used to describe a qualitative data.
Mode
Positively skewed distribution (right-skeweddistribution)
Data set has a peak on the left side and along tail on the right(positive direction).
Mean > median > mode.
Negatively skewed distribution (left-skeweddistribution)
Data set has a peak on the right side and along tail on the left(negative direction).
Mean < median < mode.
In normal distribution (bell curve, Gaussian distribution) recommended measures of central tendency are _________; recommended measure of spread is __________.
Mean, median and mode; standard deviation.
In skewed (asymmetrical) distribution recommended measure of central tendency is the __________; recommended measure of spread is _________.
Median; range or interquartile range
(Standard deviationshould be avoided because it requires themean. Themeanshould be avoided as ameasure of central tendencyin skewed data sets)
Null hypothesis(H0)
The assumption that there isno relationship between two measured variables(e.g., the exposure and the outcome)or no significant difference between two studied populations.
Alternative hypothesis(H1)
The assumption that there is arelationship between two measured variables (e.g., the exposure and the outcome)or a significant difference between two studied populations. This hypothesis is formulated as a counterpart to thenull hypothesis.
- Directionalalternative hypothesis(one-tailed) → specifies the direction of a tested relationship. It states that one variable is predicted to be larger or smaller than null value.
- Non-directionalalternative hypothesis(two-tailed) → only states that a difference exists in a tested relationship (does not specify the direction)
Directionalalternative hypothesis(one-tailed)
Specifies the direction of a tested relationship. It states that one variable is predicted to be larger or smaller than null value.
Non-directionalalternative hypothesis(two-tailed)
Only states that a difference exists in a tested relationship (does not specify the direction)
P-value
Probabilitythat a statistical test leads to the false conclusion that there is a relationship between two measured variables (e.g., the exposure and the outcome) or that there is a significant difference between two studied populations.
Probability of obtaining test results at least as extreme as those observed during the test, assuming that H0 is correct.
Probability that the study results occurred by chance alone, given that the null hypothesis is true.
- If thep-valueis equal to or less than a predetermined significance level(usually setat 0.05), the association is consideredstatistically significant(i.e., the probabilitythat the result was obtained by chance is< 5%).Thenull hypothesisis rejected and thealternative hypothesisis accepted. If p value > 0.05, null hypothesis is not rejected and the results are not significant.
- It is not possible to prove H1is true, but having ap-value that is lower than thesignificance levelindicates that it is very unlikely that the H0is correct.
If thep-valueis equal to or less than a predetermined significance level(usually setat 0.05), the association is considered________.
statistically significant(i.e., theprobabilitythat the result was obtained by chance is< 5%).
Thenull hypothesisis rejected and thealternative hypothesisis accepted.
If p value > 0.05, null hypothesis is ________.
not rejected and the results are not significant
It is not possible to prove H1is true, but having ap-valuethat is _______ than thesignificance levelindicates that it is very unlikely that the H0is correct.
lower
Type 1 error (α)
Thenull hypothesisis rejectedwhen it is actually trueand, consequently, thealternative hypothesisis accepted, although the observed effect is actually due to chance(false positiveerror).
Stating that there is an effect or difference when none exists (H0 incorrectly rejected in favor of H1).
Significance level (type 1 errorrate) → theprobabilityof atype 1 error(denoted with “α”)
Significance level (type 1 errorrate)
Probabilityof atype 1 error(denoted with “α”)
- Thesignificance levelis determined by the principal investigatorbefore the study is conducted.The higher thesignificance level, the higher the likelihood that a difference between two groups is the result of chance.
- For medical/epidemiological studies, thesignificance level αis usuallyset to 0.05
- If p < α, then assuming H0 is true, the probability of obtaining the test results would be less than the probability of making a type I error. H0 is therefore rejected as false.
The higher thesignificance level, the _____ the likelihood that a difference between two groups is the result of chance.
higher
If p < α, then assuming H0 is true, the probability of obtaining the test results would be _____ than the probability of making a type I error. H0 is therefore ________.
less; rejected as false.
Type 2 error (β)
Thenull hypothesisis acceptedwhen it is actually falseand, consequently,thealternative hypothesisis rejected even though an observed effect did not occur due to chance(false negativeerror).
An existing effect is overlooked.
A small sample size increases the likelihood of atype 2 error.
Stating that there is not an effect or difference when one exists (H0 is not rejected when it is in fact false).
- Type 2 errorrate →theprobabilityof atype 2 error(denoted by “β”)
A small sample size _______ the likelihood of atype 2 error.
increases
Stating that there is not an effect or difference when one exists (H0 is not rejected when it is in fact false).
Statistical power (1-β)
Theprobabilityof correctly rejecting thenull hypothesis (of making type II error), i.e.,the ability to detect a difference between two groups when there truly is a difference
Complementary to thetype 2 errorrate
Positively correlates with the sample sizeand the magnitude of the association of interest(e.g.,increasing the sample sizeof a study wouldincrease itsstatistical power)
Positively correlates with measurementaccuracy
By convention, most studies aim to achieve 80%statistical power.
Increase power and decrease β by increasing sample size, expected effect size and precision of measurement
Statistical power (1-β) positively correlates with ________.
Sample size
Magnitude of the association of interest(e.g.,increasing the sample sizeof a study wouldincrease itsstatistical power)
Measurementaccuracy
Increase power and decrease β by _________.
increasing sample size, expected effect size and precision of measurement
Standard deviation (σ)
Describes the variability or dispersion of data in relation to itsmean.
It is calculated by first calculating themean. Themeanis subtracted from each population data set value. Each difference is squared and added together. The total sum is divided by the total number of data setvalues - 1. The square root of this value is thestandard deviation(σ).
The standard deviation is the square root of the variance.
σ = √variance
Variance
SDˆ2
Standard Error of the Mean
Estimate of how much variability exists in a (theoretical) set of sample means around the true population mean.
Demonstrates how accurate the estimate of themeanis likely to be.
Influenced by thestandard deviation(e.g., a greaterSDincreases the chance of error) and the sample size (a smaller sample size will increase the chance of error)
SEM=standard deviation/√(sample size)
Standard error of the mean is influenced by ______.
thestandard deviation(e.g., a greaterSDincreases the chance of error) and the sample size (a smaller sample size will increase the chance of error)
SEM=standard deviation/√(sample size)
Standard normal value(Z-score,Z-value, standard normalized score)
Enables the comparison of populations with different means and standard deviations
Standard normalvalue =(value -populationmean) divided bystandard deviation
Determines how many standard deviations an observation is above or below themean
Z-scorefor a 90%confidence interval= 1.64
Z-scorefor a 95%confidence interval= 1.96
Z-scorefor a 97.5%confidence interval= 2.24
Z-scorefor a 99%confidence interval= 2.58
Confidence Interval
Provide a way to determine a population measurement or a value that is subject to change from a sample measurement.
Is therange of values that are highly likely to contain the true sample measurement
Confidence interval for sample mean = mean +/-Z(standard error of the mean) =mean+/- Z (standard deviation/√samplesize)
A lower confidence level means that a greater degree of error is possible and the measurement is less reliable.
Smallstandard deviation→narrowconfidence intervals
A lower confidence level means that a ______ degree of error is possible and the measurement is _____ reliable.
greater; less
Smallstandard deviation→____ confidence intervals
narrow
_____ standard deviation→narrowconfidence intervals
small
_________ confidence intervalsbetween two groups may indicate that there isnostatistically significantdifference.
Overlapping
__________confidence intervalsbetween two groups signify that there is astatistically significantdifference.
Nonoverlapping
If theconfidence intervalincludes thenull hypothesis, the result is ________ and thenull hypothesis_______ rejected.
not significant; cannot be
If the 95%confidence intervalofrelative riskorodds ratioincludes ____, the result is not significant and thenull hypothesis cannot be rejected.
1.0
If the 95%confidence intervalofrelative riskorodds ratioincludes 1.0, the result is _________ and thenull hypothesis ________ rejected.
not significant; cannot be
If the 95%confidence intervalof a difference between the means of two variablesincludes ___, the result is not significant and thenull hypothesiscannot be rejected.
0
If the 95%confidence intervalof a difference between the means of two variablesincludes 0, the result is ______ and thenull hypothesis______ rejected.
not significant; cannot be
A 95%confidence intervalthat does not include thenull hypothesiscorresponds to ap-valueof ____.
0.05
A 95%confidence intervalthat __________corresponds to ap-valueof 0.05.
does not include thenull hypothesis
A 99%confidence intervalthat does not include thenull hypothesiscorresponds to ap-valueof ______.
0.01
A 99%confidence intervalthat __________corresponds to ap-valueof 0.01
does not include thenull hypothesis
The null hypothesis (H0) is rejected (and results are significant) when…
95% CI for mean difference excludes 0
95% CI OR or RR excludes 1
CIs between two groups do not overlap
The null hypothesis (H0) is accepted (and results are significant) when…
95% CI for mean difference includes 0
95% CI OR or RR includes 1
CIs between two groups do overlap
Clinical Significance
Describes an important change in a patient’s clinical condition, which may or may not be due to an intervention introduced during a clinical study
Vs. statistical significance → describes a true statistical outcome (i.e., that is determined by statistical tests) that has not occurred by chance
Statistical and clinical significance do not necessarily correlate (Ex. A certain drug may cause a small drop in blood pressure that isstatistically significant. Whilestatistically significant, this small change in blood pressure may not benefit the patient and be considered clinically insignificant)
Pearson correlation coefficient (r coefficient)
Statistical measure of the strength and direction of a linear relationship between two variables expressed on a scatter plot.
It ranges anywhere from -1 to +1, where -1 represents a perfectly negative linear relationship and +1 represents a perfectly positive linear relationship.
Spearman correlation coefficient
Nonparametric method to measure the correlation between two ranked variables (i.e., compares ordinal scale variables, not nominal scale variables). Due to this method, the Spearman correlation coefficient is less susceptible to the effect of outliers but is relatively imprecise, as it does not use all information from a data set.
Coefficient of determination
rˆ2
Amount of variance in one variable that can be explained by variance in another variable
Parametric Tests
Used to evaluatestatistically significantdifferences between groups when the study sample has anormal distributionand the sample size is large (eg, ANOVA, t-test, Pearson correlation coefficient)
T-test
Calculates the difference between the means of two groups or between a sample and population or a value subject to change; especially when samples are small and/or the population or a value subject to change distribution is not known.
One-sample t-test
Thet-valuecan be classified according a table that listst-valuesand their corresponding quantiles based on the number of degrees of freedom (df) and thesignificance level(α value).
Calculates whether a samplemeandiffers from the populationmean(μ0)
Prerequisite → normal distribution
Two-sample t-test
Calculates whether the means of two groups differ from one another
Involves converting two sets of values from two different groups into a single set of values by calculating the difference between the two sets of observations. This single set of differences can then be analyzed in a fashion similar toone sample t-tests.
Prerequisites
- Both sample groups are drawn from the same population and have the same (but unknown)variance.
- The difference between the observations in the two groups approximately follows anormal distribution.
Unpaired t-test(independent samplest-test)
- Two different groups are sampled at the same time
Paired t-test(dependent samplest-test)
- The same group is sampled at two different times
Unpaired t-test(independent samplest-test)
Two different groups are sampled at the same time
Paired t-test(dependent samplest-test)
The same group is sampled at two different times
Analysis of variance(ANOVA)
Calculates thestatistically significantdifference between≥ 3 independentgroups by comparing their means(an extension of thet-test)
- One-wayanalysis of variance assesses1 variable(e.g., themeanheight of women inclinics A, B,and Cat a given point in time, the variable is height).
- Two-wayanalysis of variance assesses2 variables(e.g., themean height of women and themeanheight of men inclinics A, B,and Cat a point in time, the variables aregenderand height)
Nonparametric Tests
Used to evaluate thestatistically significantdifference between groups when the sample has nonnormal distribution and the sample size is small (eg, Spearman correlation coefficient, Mann-Whitney U test, Wilcoxon test, Binomial test)
Binomial Test
Examines whether the observed frequency of an event with binary outcomes (e.g., heads/tails, dead/alive) is statistically probable or not
Categorical Tests
Used to evaluate thestatistically significantdifference between groups withcategorical variables(nomean values) (eg, chi-square test, fisher exact test)
Chi-square Test(Xˆ2test)
Calculates thedifference between 2 or more percentages or proportions of categorical outcomes.
Aims to determine how likely outcomes are to occur due to chance(used incross-sectional studies)
Fisher Exact Test
Calculates thedifference between 2 or more percentages or proportions of categorical outcomesbut, unlike aChi-square test, is used when the study sample is small. Also aims to determine how likely it was the outcomes occurred due to chance.
Histogram
Similar to abar graphbut displays data on ametric scale. The data is grouped into intervals that are plotted on thex-axis.
Useful for continuous data
Box Plot
Quartilesandmedianare used to display numerical data in the form of a box.
Useful for continuous data
Shows the following important characteristics of data:
- Minimum and maximum values
- First and thirdquartiles
- Interquartile range
- Median
- Easily showsmeasures of central tendency, range, symmetry, and outliersat a glance
Scatter Plot
Graph used to display values for (typically) two variables of data, plotted on the horizontal (x-axis) and vertical (y-axis) axes using cartesian coordinates, which represent individual data values
Helps to establish correlations between dependent andindependent variables
Helps to determine whether a relationship between data sets is linear or nonlinear
Informed Consent
A process (not just a document/signature) that requires:
Disclosure: discussion of pertinent
information (using medical interpreter, if
needed)
- Understanding: ability to comprehend (avoid jargon, individualize and relate to the patient)
- Capacity: ability to reason and make one’s
own decisions (distinct from competence, a
legal determination)
- Voluntariness: freedom from coercion and
manipulation
Patients must have a comprehensive understanding of their diagnosis and the risks/benefits of proposed treatment and alternative options, including no treatment, complications, morbidity, risk of death, types and risk of anesthesia.
Patient must be informed of their right to revoke written consent at any time, even orally.
Competence
Legal assessment of a patient’s ability to make decisions
Assessed by a court of law (physicians do not have the legal power to pronounce patients legally incompetent)
Questions of legal competence arise in the presence of reduced mental capacity (e.g., severe mental illness, intoxication, impulsive/constantly changing decisions, or decisions that are inconsistent with the patient’s values)
Exceptions to Informed Consent
Waiver—patient explicitly relinquishes the right of informed consent
Legally Incompetent—patient lacks decision-making capacity (obtain consent from legal surrogate) or no surrogate available
Therapeutic Privilege—withholding information when disclosure would severely harm the patient or undermine informed decision-making capacity
Emergency situation—implied consent may apply
Examination, treatment or quarantine to prevent epidemics
If patient decision to refuse treatment posses a safety risk to their own wellbeing and/or welfare of others (eg, in the event of severe psychosis, patients with active tuberculosis)
Consent for Minors
A minor is generally any person < 18 years old (except Mississippi < 21 years; Nebraska and Alabama < 19 years). Parental consent laws in relation to healthcare vary by state. In general, parental consent should be obtained, but exceptions exist for emergency treatment (eg, blood transfusions, trauma, suicidal ideation) or if minor is legally emancipated (eg, married, self-supporting, or in the military).
A parent cannot refuse an emergency life saving intervention for a minor, not even for religious reasons.
If a parent continue to refuse treatment for a non emergent but fatal medical condition seek a court order
Situations in which parental consent is usually not required
Sex (contraception, STIs, prenatal care— usually not abortion)
Drugs (substance use disorder treatment)
Emergency/trauma
Minor serving a sentence of confinement
If the parents of the patients are themselves minors the grandparents give consent
General principles for exceptions to confidentiality
- Potential physical harm to self or others is serious and imminent
- Alternative means to warn or protect those at risk is not possible
- Steps can be taken to prevent harm
Examples of exceptions to patient confidentiality (many are state specific)
-Patients with Suicidal/homicidal ideation
-Abuse (children, elderly, and/or prisoners)
-Duty to protect—state-specific laws that sometimes allow physician to inform or somehow protect potential victim from harm
-Patients with Epilepsy and other impaired automobile drivers
- Reportable Diseases (eg, STIs, hepatitis, food poisoning); physicians may have a duty to warn
public officials, who will then notify people at risk. Dangerous communicable diseases, such as TB or Ebola, may require involuntary treatment.
-Patients request physician to share information with another party
-Patient has suffered penetrating trauma from assault (gunshot wound, stab wound). In this case, law enforcement must be notified
-Patient is a minor and care does not involve sexual or addiction medicine
Decision Making Capacity
Note that decisions made with capacity cannot be revoked simply if the patient later loses capacity.
Intellectual disabilities and mental illnesses are not exclusion criteria for informed decision-making unless their condition presently impairs their ability to make healthcare decisions.
Capacity is determined by a physician for a specific healthcare-related decision (eg, to refuse medical care).
Components (assessing capacity is of MASSIVE importance):
- Decision is not a result of Mental illness exacerbation
- Patient is ≥ 18 years of Age or otherwise legally emancipated
- Decision is not a result of altered mental Status (eg, delirium, intoxication)
- Decision remains Stable over time
- Patient is Informed and understands
- Decision is consistent with patient’s Values and goals
- Patient Expresses preferences
Shared Decision Making
A model in which patients and physicians decide on the best treatment option together
Empowers patient, as it is based on the patient’s personal values, cultural beliefs, and preferences
Results in better health outcomes and increases patient satisfaction
Surrogate Decision Maker
May be appointed by patients (e.g., medical power of attorney), legally appointed (e.g., court-ordered guardian), or next of kin (if no advance directive exists).
The exact hierarchy of decision-making varies from state to state. However, it generally follows the following order:
1. A mentally competent patient capable of expressing his/her own decision
- Advance healthcare directive
- Durable medical power of attorney: a document through which an individual designates a surrogate health care decision-maker in the event that he/she lacks decision-making capacity or competence
- Living will: should the durable medical POA and the living will be in conflict, the POA can override the living will only if such a decision is in line with the patient’s most recently expressed wishes. - Next of kin–> Spouse, Adult child, Parent, Adult sibling, in about half of states: “close friend”
- Ethics committee or legal consult
Ethics committee or legal consult for a decision making when
- When no surrogate is available or there is ongoing dispute (i.e., between equal priority surrogates) regarding who takes precedence as surrogate.
- Legal action may be necessary to appoint a legal guardian if there is no surrogate available or there is ongoing dispute between equal priority surrogates.
- When there is ongoing disagreement (i.e., between equal priority surrogates) about treatment decisions that cannot be resolved
- When a physician determines that a surrogate’s decision may go against the patient’s best interests or preferred decision
- When a physician determines that a surrogate’s decision is made to benefit a third party, rather than in the patient’s best interests
Directive Counseling
Ethically appropriate when only one treatment option is medically reasonable and has clearly superior evidence based support
Assisted Decision Making
Occurs when a family member or a care giver helps the patient in making a medical decision (but does not make the decision for the patient)
Substituted Judgement
Occurs when a surrogate decision maker makes a health care decision for an incapacitated patient based on the surrogate’s knowledge of the patient wishes and values.
Therapeutic Privilege
A physician determines thatfull disclosurewould cause severe harm to the patient’s severe psychological harm (e.g., following an unfavorable prognosis)
Elements of an optimal error disclosure
- Clearly admit an error has occured
- State the course of events leading to and during the error, avoiding jargon
- Explain the consequences of the error, both immediate and long term (if necessary)
- Describe corrective steps and future preventative steps
- Express personal regret and apology
- Allow ample time for questions and continued dialogue
Physician-Assisted Suicide
When a physician supplies a patient with the means to end their own life (e.g., a physician provides a patient with alethal doseofmorphinethat the patient then self-injects).
Illegal in most state.
Euthanasia
Active termination of a terminally ill patient’s life by a physician to end suffering. (e.g., a physician injects alethal doseofmorphine).
Euthansia is illegal in the United States.
Terminal Sedation
It is legal to adjust medical therapy accordingly to provide relief frompainand suffering in a patient with terminal illness, despite hastening the patient’s dying process (e.g., increasing doses ofmorphinein a patient with metastatic cancer).
Legal and distinct fromeuthanasiain so far as the intent must be to relievepainrather than bring about death, even though it may hasten the dying process.
Not an appropriate means of addressing primarily existential suffering, e.g., death anxiety.
Principle of Double Effect
Ethical principle that legitimizes an act of good intent despite causing serious harm (e.g. self defense homicide orterminal sedation).
Withdrawal of Care
Patients with capacity (or their surrogatedecision-makers) have the right to refuse any form of treatment at any time, even if that would result in that patient’s death.
Physicians should make an effort to understand the reasons behind the patient’s decision for refusing treatment.
There is no ethical difference between withholding care and withdrawing care at a later time. The patient may refuse treatment before or request to stop treatment after its commencement.
Treatment can be considered futile if:
- There is no evidence for the effectiveness of treatment
- If the intervention has previously failed
- Iflast-linetherapy is failing
- If treatment will not fulfill the goals of care
Hospitals must decline organs that are considered unsuitable:
- Sepsis
- HIV
- Poor organ function (e.g., patient died ofacute renal failure, sokidneys would be refused)
- Hypothermia
- Patient> 80 years old
- Prolonged organischemia(e.g., patient found deceased at home)
Death Criteria
Death can be diagnosed if a patient meets criteria forbrain deathor cardiopulmonary death. Only one of these conditions is required, although they may coexist.
- Brain death is is an irreversible, complete loss of function of the entire brain (including thebrainstem), even if cardiopulmonary functions can be upheld by artificial life support
- Cardiopulmonary death is the absence of a spontaneous heartbeat in anasystolicpatient.
A hypothermic patient must be warmed to normal body temperature before death can be diagnosed.
Notifiable Diseases
HIV
Sexually transmitted infections(e.g.,gonorrhea,chlamydia)
Hepatitis
Tuberculosis
Rabies
Meningococcal meningitis
Some forms offood poisoning(e.g.,salmonellosis)
Primordial Prevention
Actions that address environmental, socioeconomic, and behavioralrisk factorsthat affect a population as a whole to prevent potential disease or injury.
Target → entire population.
Health promotion beginning in childhood to encourage positive and discourage negative lifestyle habits, mass education, legislation.
Examples → programs on food safety and nutrition guidelines, campaigns discouraging tobacco and drug use (e.g.,smoke-freeair laws in public buildings), building bicycle and sidewalks to promote physical activity.
Primordial prevention aims to preventrisk factorsfrom developing in the first place, whereasprimary preventiontargets existingrisk factorsto prevent the onset of a disease.
Strategies to encourage adherence
- Identify potential barriers(e.g.,disease-relatedknowledge)
- Apatient-centered approach
- Communication about treatment options
- Provide patient education (written instructions regarding medication, discuss medication side effects)
- Reminder system (regularfollow-upvisits. email reminder programs, electronic alerts, phone call reminders)
Accountable Care Organization
- Coordinatednetwork of doctors, specialists, and hospitalsthat are voluntarily enrolled
- Encourage health care providers to work together to achieve a higher quality of care at a lower cost.
- Patient and provider participation in anaccountable care organization is completely voluntary.
- Cover medicare patients
- Enrollment of specialists on a voluntary basis
- Member costs varies (expected to reduce overall costs)
Exclusive Provider Organization
- Network of doctors, specialists, and hospitals
- No primary care physician needed
- Nocoverage forout-of-networkproviders
- Specialists can be seen without a referral from a primary care physician.
- Low member costs
Health Maintenance Organization (HMO)
- Least flexible but also lowest costs for members
- Network of doctors, specialists, and hospitals
- Primary care physician is the first contact person.
- Nocoverage forout-of-networkproviders, except emergency visitsthat are covered atin-networkrates
- Referral needed from primary care physicianto see a specialist
- Women have direct access to obstetric and gynecological care.
- Low member costs
Preferred Provider Organization
- Network of doctors, specialists, and hospitals
- No primary care physician needed
- Coverage forout-of-networkproviders
- Highout-of-pocketpayments
- Specialistscan be seen without a referral from a primary care physician.
- High member costs
Point of Service (POS)
- Combination ofHMOandPPO. Patients still have to go through a primary care physician (point-of-service)
- Network of doctors, specialists, and hospitals
- Primary care physician is the first contact person.
- Coverage forout-of-networkproviders
- Highout-of-pocketpayments
- Referral needed from primary care physician to see a specialist
- Moderate member costs
Medicare Eligibility
- Individuals≥ 65 yearsoldwho have worked (or their spouse) and paidMedicaretaxes for at least10 yearsand who are US citizens or have a permanent legal residence
- Patients withend-stage renal failureoramyotrophic lateral sclerosis
- Individuals with permanent disabilities irrespective of age
Medicare Parts
- Part A → hospital care,hospice carefor terminal patients, skilled nursing facility care (if services are needed daily after a minimum3-daystay in a hospital)
- Part B → doctor’s fees, emergency department visits, diagnostic tests, rehabilitation
- Part C(MedicareAdvantage Plan) → all services covered byparts Aand B, plus a private insurance plan. “All in one” plan that allows people to enroll in aprivate health insuranceplan approved byMedicare. Medicarepays other organizations, such as insurance companies, hospital systems, or managed care organizations, to provide care.
- Part D → prescription drugs(individuals can addpart Dto either of the programs)
The two main coverage options are OriginalMedicare (part Aandpart B) andMedicareAdvantage (part C). Individuals also have the option of addingpart Dto their main coverage.
Medicaid
Jointly funded by the state and federal governments
Eligibility →individuals with low income
Coverage → hospital care, laboratory tests, diagnostic tests (such asx-rays), doctors’ visits, skilled nursing care, vaccinations, home health care
Children’s Health Insurance Program(CHIP)
Eligibility → uninsured children of families with low income, but not low enough to qualify forMedicaid
Fee-for-Service
- Health care providers are compensated for each individual service provided(e.g., individual laboratory tests, imaging studies, procedures).
- Incentivizes health care providers to overtreat patients becausecompensationis based on the number of services provided
- Associated with high overall health care costs
- A fixed payment schedule with discounted prices can be negotiated between health care providers and payers (i.e.,discounted fee-for-service).
- A health care organization is compensated with a fixed amount for all services provided for aclinically-defined episode of care (e.g.,hip replacement,cholecystectomy). Payment is then distributed to the health care providers.
- Incentivizes health care providers to deliver efficient care (e.g., health care providers avoid unnecessary procedures)
- Carries the risk that patients will be undertreated becausecompensationdoes not rely on the quantity or quality of services provided for eachclinically-defined episode of care (e.g., physicians may be pressured to prematurely discharge patients from the hospital because thecompensationreceived for each admission is unaffected by an early discharge.
Global Payment
- Health care providers are compensated with a single payment for all the services included in a single episode of care.
- Often used for nonurgent surgery, with the coverage extended to all the pre- and postoperative visits
Capitation
- Health care providers are compensated a fixed amount per patient during each payment period, regardless of the actual amount of health care utilized by the patient.
- Incentivizes health care providers to deliver efficient care (e.g.,cost-effectivepreventive health care to avoid larger downstream costs)
- Carries the risk that patients will be undertreated becausecompensationis not based on the quantity or quality of services provided
- Often used byhealth maintenance organizations(HMOs)
Per Diem Payment
- A health care organization is compensated a fixed amount per patient per day for a specific care service provided, regardless of the actual costs involved in providing services for any particular patient.
- Often used for reimbursement of inpatient service → A hospital and payer can negotiate a fixed per diem rate for most routine inpatient services (e.g., common medical and surgical conditions) based on the average daily cost of admission.
Eligibility for Hospice Care
- Estimated life expectancy< 6 months
- Patients are usually onMedicare,Medicaid, or private insurance plans.
- The patient (and family) has made the decision to stop curative orlife-preservingtreatment in order to maximize quality of life.
Not all treatment should be withdrawn. Antibiotics, for example, can still be given if the patient develops an infection.
Adverse Reaction
Unintended, unexpected, or undesirable consequence of a correct and justified action
Adverse Event
Any harm resulting from medical management rather than an underlying disease.
Anadverse eventmay or may not be the result of amedical error.
Can be preventable or nonpreventable
Near Miss(Close Call)
Medical errorthat could have resulted in anadverse eventbut did not, either incidentally or due to a timely intervention (e.g., a nurse identifies that a doctor’s order is incorrect)
Medical Error
Accidental failureto perform an action as intended (e.g., misdosing the appropriate drug) or accidentally performing the incorrect action (e.g., prescribing the wrong drug), which could potentially result in harm to the patient
- Individual error → medical error resulting from the failure of a single healthcare professional due to negligence
- Systems error → medical errorresulting from a series of actions and/or factors in treatment or diagnosis, from flaws in technical and organizational design and/ordecision-making, and from failure to recognize and mitigate hazards and risks in the healthcare setting
Medical Malpractice
Negligent conduct on the part of a healthcare provider or performance of a medical task with an unreasonable lack of skill
- Negligence → failure to perform an action with the skill, care, and knowledge expected of a healthcare provider under reasonable circumstances. This includes medical ignorance (i.e., failing to have the necessary current medical knowledge to diagnose and treat a specific condition).
Safety Culture
Culture that promotessafety awareness, develops and implements measures for the maintenance of asafe work environment, and ensures that individuals canopenly express safety concerns
Improves the identification of errors
Event reporting systemsfacilitate internal and external monitoring through the collection of data on errors.
Sentinel Event
Wholly unexpected or unacceptableadverse eventthat results in serious injury or death of a patient
The term “sentinel event” was developed by The Joint Commission, while the term “never event” was developed by the Agency for Healthcare Research and Quality. There is some overlap in the definitions of sentinel events and never events.
Never Event
Seriousadverse eventthat is clearly identifiable, causes serious injury or death, and is considered sufficiently preventable that it should never occur
Examples include injury, disability, or death due to the following:
- Wrong-sitesurgery
- Wrong-patientsurgery
- Post-procedureretention of a foreign object in a patient
- Suicide,suicide attempt, orself-harmwithin ahealth care facility
- Using contaminated devices or medications
- Using medical devices for purposes other than the intended function
- Administering the wrong medication
The term “sentinel event” was developed by The Joint Commission, while the term “never event” was developed by the Agency for Healthcare Research and Quality. There is some overlap in the definitions of sentinel events and never events.
AmeliorableAdverse Event
Unintended consequence of medical management that is unavoidable but could have been less severe if a different protocol were followed.
Latent Error
Typically caused by a failure of design and/or of the larger system in question (e.g., the hospital).
Domains of health care quality (STEEEP)
- Safe → Avoid or minimize risks and hazards that may lead to harm.
- Timely → Reduce delays in administering healthcare that may lead to harm.
- Effective → Provide evidenced-based healthcare and avoid services or treatments of doubtful benefit.
- Efficient → Minimize or avoid wasting medical equipment, time, and energy.
- Equitable → Provide equal care to all irrespective of gender, ethnicity, sexuality, and socioeconomic status.
- Patient-centered → Individualize treatment with respect for patient preferences, values, and needs.
Structural Quality Measurement
Measures of the resources available to ahealthcare facility(e.g., equipment, facilities).
For example, physician-patientratio, number of beds, number of nutritionists available for diabetic patients
Process Quality Measurement
Measures of the performance of a health care system as it was planned.
For example, percentage of individuals who receive a particular preventive service (e.g., immunizations, cancer screening,HbA1cmeasurement) over a period of time
Outcome Quality Measurement
Measures of the final impact of service provided by a healthcare facility, including mortality and morbidity.
For example, averageHbA1cof patients over a period of time, maternal mortality rates, rates ofnosocomial infections
Balancing Quality Measurement
Measures of the impact of one system on another.
Given that resources are limited, it is important to consider whether the resources allocated to one system might be better utilized in another system.
For example, cost-benefitanalysis (e.g., usingnumber needed to treat) of hiring more nutritionists to educate diabetic patients
Pareto Chart
Bar graphwith bars representing the frequency or cost of problems.
Used to identify the significance of individual problems.
Helps analyze how certain factors contribute to a variety of issues
Shewhart Chart(Control Chart)
Graphic representation in which data is plotted over time to determine if a perceived improvement in quality over time isstatistically significant
Run Chart(Time Plot)
Line graph that plots data in a time sequence to analyze trends.
Used in healthcarequality improvement, for example, to analyze the impact of an intervention over time and to help determine whether the improvement is a random or a true trend.
Use minimal mathematical complexity (Unlike aShewhart chart).
Convenience Sampling
Study population that is drawn from a population that is easy to reach.
Since the study population is not chosen at random, this type of study can lead tobiasand may not be as effective as arandomizedcontrol trial.
SMART Criteria
Used to improve patient safety.
Goals should beSpecific,Measurable,Achievable,Realistic, andTimely.
Plan, Do, Study, Act Cycle(PDSA)
- Plan → Define an area that needs improvement and plan potential interventions or innovationsto bring about a corrective change.
- Do → Test the new intervention.
- Study → Assess the impact of the intervention on quality of healthcare.
- Act → Implement the new interventionif the previous test showed a positive impact on healthcare quality or restart the cycle if the results were negative or had no significant impact.
Lean Process Improvement
Process of continually evaluating the methods used to identify and eliminate factors that use time, energy, and resources without improving patient outcome (e.g., avoiding unnecessary investigations)
Burnout
Overall exhaustion due to an excess of stress over an extended period of time
- Lack of motivation and interest
- Feelings of failure and helplessness
- Cynical and detached work attitude
- Impaired immune function
- Increased risk ofmedical errorsdue to decreased concern, reduced professional efficacy
Fatigue
- Chronic sleep deprivation
- Decreased energy and motivation
- Impaired cognitive function
- Increased risk ofmedical errorsdue to impairment in intellectual function
Active Error
Error at thedirect level of contactbetween healthcare personnel and patients. Has animmediate impacton the patient.
For example, surgery on the incorrect site, wrong route of drug administration
Latent Error
Error that could contribute to patient harm. Latent error(s) in conjunction withactive error(s) can lead to anadverse event.
For example, flaws in hospital organization, implementation of new equipment without adequate staff training
Communication Error
Error in communication between the healthcare personnel and the patient as well as among healthcare personnel.
For example, errors in → history taking; explaining planned medical procedures to the patient; written communication (e.g., poor handwriting); verbal communication (e.g., using nonstandard terminology or jargon)
Diagnostic Error
Errors or delays in diagnosis.
For example, not ordering the required investigations, use of outdated tests, failure to adequately monitor clinical signs orlaboratory studies
Treatment Error
Errors or delays in treatment.
For example, inappropriate medical procedures, incorrect administration of treatment, incorrect drug dosage or method of use, failure to provide treatment or respond to diagnoses in a timely manner
Preventive Error
Errors in prophylaxis.
For example, failure to implement appropriate prophylaxis, failure to provide adequate monitoring orfollow-uptreatment, failure in equipment and system maintenance
Medication Error
Errors in prescribing; errors in transcribing; errors in dispensing; errors in administration.
For example, failure to correctly transcribe drug names, dosages, routes of administration (e.g, misinterpretation of a trailing zero), incorrect drug dispensing due to an error related to medications that look or sound alike
Domains of Health Care Quality (STEEEP)
- Safe → Avoid or minimize risks and hazards that may lead to harm.
- Timely → Reduce delays in administering healthcare that may lead to harm.
- Effective → Provide evidenced-based healthcare and avoid services or treatments of doubtful benefit.
- Efficient → Minimize or avoid wasting medical equipment, time, and energy.
- Equitable → Provide equal care to all irrespective of gender, ethnicity, sexuality, and socioeconomic status.
- Patient-centered → Individualize treatment with respect for patient preferences, values, and needs.
Structural Quality
Measures of the resources available to ahealthcare facility(e.g., equipment, facilities).
For example, physician-patientratio, number of beds, number of nutritionists available for diabetic patients
Process Quality
Measures of the performance of a health care system as it was planned.
For example, percentage of individuals who receive a particular preventive service (e.g., immunizations, cancer screening,HbA1cmeasurement) over a period of time
Outcome Quality
Measures of the final impact of service provided by a healthcare facility, including mortality and morbidity.
For example, averageHbA1cof patients over a period of time, maternal mortality rates, rates ofnosocomial infections
Balancing Quality
Measures of the impact of one system on another. Given that resources are limited, it is important to consider whether the resources allocated to one system might be better utilized in another system.
For example, cost-benefitanalysis (e.g., usingnumber needed to treat) of hiring more nutritionists to educate diabetic patients
Pareto Chart
Abar graphwith bars representing the frequency or cost of problems.
Used to identify the significance of individual problems.
Helps analyze how certain factors contribute to a variety of issues
Shewhart Chart(Control Chart)
Graphic representation in which data is plotted over time to determine if a perceived improvement in quality over time isstatistically significant
Run Chart(Time Plot)
A line graph that plots data in a time sequence to analyze trends.
Used in healthcarequality improvement, for example, to analyze the impact of an intervention over time and to help determine whether the improvement is a random or a true trend.
Use minimal mathematical complexity (Unlike aShewhart chart).
Convenience Sampling
A study population that is drawn from a population that is easy to reach.
Since the study population is not chosen at random, this type of study can lead tobiasand may not be as effective as arandomizedcontrol trial.
SMART Criteria
Used to improve patient safety.
Goals should beSpecific,Measurable,Achievable,Realistic, andTimely.
Plan, Do, Study, Act Cycle(PDSA)
- Plan → Define an area that needs improvement and plan potential interventions or innovationsto bring about a corrective change.
- Do → Test the new intervention.
- Study → Assess the impact of the intervention on quality of healthcare.
- Act → Implement the new interventionif the previous test showed a positive impact on healthcare quality or restart the cycle if the results were negative or had no significant impact.
Lean Process Improvement
Process of continually evaluating the methods used to identify and eliminate factors that use time, energy, and resources without improving patient outcome (e.g., avoiding unnecessary investigations)
Burnout
Overall exhaustion due to an excess of stress over an extended period of time
- Lack of motivation and interest
- Feelings of failure and helplessness
- Cynical and detached work attitude
- Impaired immune function
- Increased risk ofmedical errorsdue to decreased concern
Fatigue
- Chronic sleep deprivation
- Decreased energy and motivation
- Impaired cognitive function
- Increased risk ofmedical errorsdue to impairment in intellectual function
Alert Fatigue
Tendency of health care providers to become desensitized to and subsequently ignore alerts prompted byclinical decision support systemsdue to the excessive number or limited clinical relevance of the alerts
Active Error
Error at thedirect level of contactbetween healthcare personnel and patients.
Has animmediate impacton the patient.
For example, surgery on the incorrect site, wrong route of drug administration
Latent Error
Error that could contribute to patient harm. Latent error(s) in conjunction withactive error(s) can lead to anadverse event. For example, flaws in hospital organization, implementation of new equipment without adequate staff training
Communication Error
Error in communication between the healthcare personnel and the patient as well as among healthcare personnel.
For example, errors in → History taking; explaining planned medical procedures to the patient; written communication (e.g., poor handwriting); verbal communication (e.g., using nonstandard terminology or jargon)
Diagnostic Error
Errors or delays in diagnosis.
For example, not ordering the required investigations, use of outdated tests, failure to adequately monitor clinical signs orlaboratory studies
Treatment Error
Errors or delays in treatment.
For example, inappropriate medical procedures, incorrect administration of treatment, incorrect drug dosage or method of use, failure to provide treatment or respond to diagnoses in a timely manner
Preventive Error
Errors in prophylaxis.
For example, failure to implement appropriate prophylaxis, failure to provide adequate monitoring orfollow-uptreatment, failure in equipment and system maintenance
Medication Error
Errors in prescribing; errors in transcribing; errors in dispensing; errors in administration.
For example, failure to correctly transcribe drug names, dosages, routes of administration (e.g, misinterpretation of a trailing zero), incorrect drug dispensing due to an error related to medications that look or sound alike
Several elements of an optimal error disclosure
- Clearlyadmitan error has occurred
- State the course of events leading to and during the error, avoiding jargon
- Explain the consequences of the error, both immediate and long term (if necessary)
- Express personal regret andapology
- Describecorrective stepsand future preventative steps
- Allow ample time for questions andcontinued dialogue
Human Factors Design
Forcing functions (those that prevent undesirable actions [eg, connecting feeding syringe to IV tubing]) are the most effective. Equipment, process, method, or system design features that prevent error by forcing the best option by default. Most effective technique for minimizingadverse eventsbecause they inhibit a chain of action that causes or perpetuates error
Standardization improves process reliability (eg, clinical pathways, guidelines, checklists), efficiency, communication, and safety
- Standardizedterminology →helps preventcommunication errors (e.g., use of the chemical name of a drug rather than its generic name or brand name) - Standardized abbreviations and writing styles → help prevent communication and transcribing errors. Tall man lettering → a strategy used to prevent dispensing errors related to medications that look or sound alike (e.g., buPROPion vs. busPIRone) - Standardizedmedication reconciliation → performed during transitions of care (e.g., admission, transfer, discharge) in which new medication orders are compared to past medication orders. Used to avoidmedication errorssuch as duplication, omission, dosing errors, and drug interactions - Standardizedhandoff → exchange of predefined items of information between health care providers when passing on responsibility for the care of patients (e.g., at the end of a shift) to reduce the risk ofmedical errors. Should include the patient's diagnoses, code status, a summary of the hospital course, current medications, recent changes to management, explicit suggestions for anticipated concerns, tasks to complete - Time out → a perioperative pause conducted by the surgical team to confirm the patient's identity, surgical site, and planned procedure. Intended to prevent harm from a wrong procedure at the wrong site or patient. - Simplification reduces wasteful activities (eg, consolidating electronic medical records). Reduction of complexity of equipment, systems, and processes to increase efficiency and reduce the risk of error
Effective communication
- Closed-loop communication → a communication method in which the receiver of information repeats that information back to the sender, upon which the sender confirms that the information has been received as intended (used to avoidcommunication errors) - Patient involvement → prior to administering drugs or inducing anesthesia, the patient is asked for verification of identity and any further relevant information (e.g., site of surgery, planned procedure, consent) - Computerized order entry(CPOE) → a system in which electronically placed orders for medications, tests, procedures, and consults are directly transferred to the recipient (used to prevent errors due to poor handwriting or ambiguous abbreviations)
Morbidity and mortality reviews (M&M) → a standardized process wherein departmental meetings are held to review complications and deaths that occur in hospitals to improve safe clinical practice and avoid preventable errors.
Closed-Loop Communication
Communication method in which the receiver of information repeats that information back to the sender, upon which the sender confirms that the information has been received as intended (used to avoidcommunication errors)
Root Cause Analysis
Retrospective analysisperformed after amedical errorhas occurred in order to identify the (root) causes of the error and implement measures toprevent recurrence.
Procedure:
- Identify themedical error. (“What happened?”)
- Retrospective analysis of all possible factors that could have led to the errorwith the support ofpatient documentation, the equipment/drug used, and the environment the patient was in. If feasible, the patient should also be interviewed.
- Examples include → slippery floor, infectious ward, nonquarantined patient, management protocol used - Identify what could have prevented the error from occurring. (“Why did it happen?”)
- Tool used →Fishbone diagram, also calledIshikawa diagramorcause and effect diagram. Used to break down and display all potential events that led to the error - Implement preventive measures. (“What can be done to prevent recurrence?”). Examples include → updating technology, employing double checks, using checklists, staff education on new policies
Failure Mode and Effects Analysis(FMEA)
Prospective analysisto anticipate potential risks and hazards and proactively implement measures toprevent amedical error from occurring.
Procedure:
- Identify what could go wrong and what impact that could have (failure mode).
- Identify why it could go wrong (failure causes).
- Identify the outcomes of potential failures (failure effects).
- Prioritize the hypothetical failures by their chance of occurrence and severity of outcome.
- Proactive implementation of corrective measures
Causal criteria(Bradford-Hill criteria)
List of criteria that helps to establish causality in epidemiological studies
- Strength of association(effect size) → a quantitative measure of the degree of relationship between two variables. The stronger association, the more evidence for causation.
- Dose-response relationship(biological gradient) → tests whether greater exposure usually leads to a higher occurrence of the outcome (e.g., the greater the exposure toionizing radiation, the higher the risk ofmalignancy)
- Temporality → tests whether the outcome occurs after the effect (e.g.,surgical site infectionoccurs afterincisionof theskin)
- Reproducibility(consistency) → Thestrength of associationfound between two factors increases when similar findings are observed in different studies (e.g., in different places, with different sample sizes). Repeated observations of the findings in multiple distinct samples.
- Specificity → It is more likely that a causal relationship exists between a factor and an effect when a specific disease occurs in a specific population at a specific time (e.g., there is an increase in the number ofleukemiacases in a small town after a chemical factory is built nearby).
- Biologic plausibility → The relationship between an exposure and an outcome is usually consistent with current biological and medical knowledge (e.g.,carcinogensin cigarettes causelung cancer, and water molecules do not).
- Coherence (epidemiology) → tests whether new evidence is in agreement with previously established evidence. If there is a contradiction with previously established results, this reflects negatively on the likelihood of a causal relationship between a given set of factors (e.g., an epidemiological study that finds acorrelationbetween higher rates oflung cancerin men while there is no available biological data to support it).
- Experimental evidence → data drawn from provisional experimentation (e.g., a distance of at least2 metersbetween people is correlated with a decrease in the number ofCOVID-19cases). Empirical evidence supporting the presumed cause and effect
- Analogy (epidemiology) → When there is strong evidence of a causal relationship between an exposure and an outcome, there is a greater likelihood of a causal relationship between another similar exposure and outcome (e.g., when one class of medication is known to produce an effect, it is likely that another agent of that class produces a similar effect).
Reverse Causality
Association between exposure and outcome that is different than common presumption (e.g., people assume that low socioeconomic status causesschizophrenia, but in fact,schizophreniacauses a decline in socioeconomic status over time)
Patient-Centered Interviewing Techniques
- Introduction
- Agenda setting
- Reflection (actively listen and synthesize information offered by the patient, particularly with respect to primary concern(s))
- Validation
- Recapitulation
- Facilitation (encourage the patient to speak freely without guiding responses or leading questions. Allow the patient to ask questions throughout the encounter)
Expressing Empathy
- Partnership → reassure the patient that you will work together through difficult times, and offer appropriate resources.
- Empathy → acknowledge the emotions displayed and demonstrate understanding of why the patient is feeling that way.
- Apology → take personal responsibility when appropriate, or offer condolences for the patient’s situation
- Respect → commend the patient for coming in to discuss a problem, pushing through challenging circumstances, keeping a positive attitude, or other constructive behaviors.
- Legitimization → assure the patient that emotional responses are understandable or common.
- Support → offer to help the patient through difficult times.
Delivering Bad News
- Setting → Offer in advance for the patient to bring support. Eliminate distractions, ensure privacy, and sit down with the patient to talk.
- Perception → Determine the patient’s understanding and expectations of the situation
- Invitation → Obtain the patient’s permission to disclose the news and what level of detail is desired.
- Knowledge → Share the information in small pieces without medical jargon, allowing time to process. Assess the patient’s understanding.
- Emotions → Acknowledge the patient’s emotions, and provide opportunity to express them. Listen and offer empathetic responses.
- Strategy → If the patient feels ready, discuss treatment options and goals of care. Offer an agenda for the next appointment.
Hand Disinfection
Decreases transient skin flora more effectively and dehydrates the skin to a lesser extent compared to handwashing.
Although the lipids on the stratum corneum are emulsified by alcoholic preparations and are thereby released from their structure, they remain on the skin as long as the hands are not subsequently rinsed.
Disinfection → minimum contact time: 30–60 seconds
Indication:
Before and after contact with each patient
Before work, before and after breaks, as well as before (self-protection) and after going to the bathroom.
Before handling medication, syringes, and infusions
After removing contaminated gloves
Contact Precautions
Used for the care of patients with drug-resistant pathogens (e.g., MRSA, VRE), enteric infections (e.g., Clostridium difficile, Escherichia coli O157:H7), scabies, impetigo, and draining abscesses
Includes performing hand hygiene and wearing gloves and gowns when getting into the patient’s room (even when direct contact with the patient or infected material is not expected).
Patients should be kept in isolation or in cohort (keeping those with similar illness in one room)
Medical equipment should be dedicated to a single patient. If not possible, disinfect before reuse.
Droplet Precautions
Used for the care of patients with suspected or confirmed infection with pathogens that spread with droplets, such as Neisseria meningitidis, Bordetella pertussis, Influenza, Parainfluenza, Adenovirus, Haemophilus influenzae type b, Mycoplasma pneumoniae, and Rubella
Patients should be kept in isolation or in cohort.
Includes wearing masks within a distance of 3– 6 feet from the patient and masking patients during transport
Implement hand hygiene after contact with respiratory secretions.
Airborne precautions
Used for the care of patients with suspected or confirmed tuberculosis, measles, varicella, smallpox, and severe acute respiratory syndrome (SARS) infections
Patients should be kept in a private room with negative air pressure (the door of the isolated room must remain closed).
Individuals must wear a respirator when entering the room.
Minimize transport of patients and mask them if it is mandatory.
Implement hand hygiene after contact with respiratory secretions.
Prevention of catheter-associated urinary tract infections
Avoiding unnecessary catheterization
Using sterile technique during catheter placement
Cleaning the catheter surrounding area with soap and water suffices for maintenance.
Prompt removal when the catheter is no longer needed
Using clean intermittent catheterization in patients with neurogenic bladder (in this technique, the catheter is immediately removed after bladder drainage and gets either discarded (single-use catheter) or cleaned (reusable catheter))
Use of antibiotic-coated catheters or prophylactic antibiotics should be avoided as it might lead to the development of drug-resistant pathogens.
Prevention of intravascular catheter-related infections (e.g., central venous line infection)
Implementing hand hygiene and strict aseptic technique during insertion
Using a cap, mask, long-sleeved sterile gown, sterile gloves, and a sterile full-body drape
Preparing skin with chlorhexidine and alcohol before inserting the catheter
Systemic anticoagulation and antibiotics may be considered in oncology patients who require long-term central venous access.
Changing dressings regularly
Prevention of ventilator-associated infections
Follow ventilator bundle protocol (a set of 4 standardized practices to prevent ventilator-associated infection)
Elevate head of the bed to 30–45°.
Assess readiness of extubation daily.
Interrupt sedation daily (sedation vacation).
Administer peptic ulcer and deep vein thrombosis prophylaxis.
Use noninvasive ventilation where possible.
Place tubes through the mouth rather than nose, if possible.
Maintain proper patient oral hygiene (e.g., use of oral chlorhexidine).
Consider antibiotic prophylaxis in patients that underwent emergency intubation.
Prevention of surgical site infections
Delay elective procedures until all infections, even those remote from the surgical site, are treated.
In patients with planned surgery
- Ask the patient to cease smoking at least one month before surgery.
- Ask the patient to bathe or shower the night before the operation.
Skin preparation routine in the operating room should include an alcohol-based agent.
Parenteral periprocedural antimicrobial prophylaxis
- Should not be given routinely, except after cesarean deliveries
- If given, antimicrobial prophylaxis should to be timed to ensure appropriate concentrations at the time of incision.
- In clean and clean-contaminated procedures, no additional antimicrobial prophylaxis should be given after the incision has been closed.
No topical antimicrobial agents should be applied to the incision.
Blood glucose should be monitored; target levels during surgery are < 200 mg/dL.
Ensure normothermia.
Range
The difference between the largest and smallest value in a data set. For example, in the data set “27, 3, 4, 9,” the range is 24 (i.e., 27-3).
Interquartile range
The range from the second to the third quartile. Calculated by establishing the difference between the 75th and 25th percentile.
Less influenced by extreme data values (outliers).
