Study guide Flashcards
What is a ratio, and an example of cases and noncases ?
Ratio = Something/Something (Case/Non-Case)
What is proportion, an Example ?
Proportion = Something/Collective (Case/Pop.)
What is Rate in terms of people?
Rate = Something/Time or Person (Case/X people or X time)
What is incidence in cases? Are they acute or chronic ? What is Attack rate, and who is a risk ? What prevention are there to stop the incidence ?
INCIDENCE New Cases! Incidence = (New Cases)/(Number at risk) Usually ACUTE Infections Attack Rate = Incidence in those with PARTICULAR RISK PRIMARY PREVENTION stops incidence! (prevent infection)
What is Prevalence ? What is an example in terms of cases ? What two types of prevalence are there ? Is it for Chronic or acute dieases ? What form of prevention stops prevalence in terms to cure ? What is tertiary prevention ? and what does it improve ? What are some examples ?,
PREVALENCE Current Cases! ( P) present! Prevalence = (ALL CASES)/(Number at risk) 2 types = POINT (At one time) and PERIOD (Span) Prevalence Usually CHRONIC Diseases SECONDARY PREVENTION stops prevalence! (Cures) Note – TERTIARY PREVENTION improves quality of life (example: drugs that extend life, treat, but not cure)
How is prevalence/incidence affected when… …A cure is found?
Prevalence down, no change to incidence. since cure happens after infection,
A vaccine is found?
Incidence down, prevalence down in time. Since happens before infection
therapy extends life?
Prevalence up, no change to incidence. there less pople dying from the current pouplation / people at risk
Why are Crude, Specific And Standard Rates used for ?
Important for Population Statistics
What is crude rate, and what is it for ?
Crude Rate = All Cases/Total Population
What is Specific rate, and what is it for ?
Specific Rate = Cases in Group/Total of Group
What is case fatality rate ?
Case Fatality Rate = Death From Cause/Population with Cause
What is proportionate mortality rate ?
Proportionate Mortality Rate = Death Due to Cause/All Deaths
What is standardized rate for ?
Standardized (Adjusted) Rate: Calculated by adjusting for demographic variable (age, sex, race, etc.)
What is Epidemiology to an MD?
Distribution of diseases in a Population. Focus on GROUPS, not INDIVIDUALS Everything is in numbers!! Look at RATIOS and RATES:
What is primary care, secondary, and tertiary prevention care ?
Primary prevention aims to prevent disease or injury before it ever occurs. Secondary prevention aims to reduce the impact of a disease or injury that has already occurred. Tertiary prevention aims to soften the impact of an ongoing illness or injury that has lasting effects.elping people manage long-term, often-complex health problems and injuries
.Draw the actual sick and diagnosic sick on the diagram.
.What are the four total outcomes of the diagram ?
.what is the difference between specificity and sensitivity.
.Can you calculate the specificity and specialty of this example ?
.What is ppv vs npv ?
.What is the postive and negetive predicted value ?
.What is an example of accuracy in the diagram, and how do you calculate it ?
.What is overall accuracy, and how do you calculate it ?
What does true and false, and positive and negative mean ?
True= meets both the same result,
flase- not the same
postive= was shown before to have it
negetive = was not shown before ( diagnosed)
ppv and npv depend on the dignognised horizontal
super/ super+ part
sensity and specifilty depend on healthy and non healthy, vertial
¡If we have more cases of the disease, what happens to…
¡…the sensitivity?
NO CHANGE
¡the specificity?
NO CHANGE
¡the PPV?
INCREASES
¡the NPV?
DECREASES
Out of the two study types, explain observational.
¡Observational:What is happening in the natural world?
Of the second study type, explain experimental ?
¡Experimental
What does this intervention cause?
After observational studies done.
(Clinical Trials)
What does case report ? what type of study , is it indivudual or group ? what type of study is it ?
¡Case-Report: N=1,descriptive study, one individual.
Most likely if it starts with case, it is observational
.Out of three observational studies, case series is a— study, that looks into …
¡Case-Series: N>1, several case-reports combined.
In observation studies, case series looks into how many people and what happens to the report ?.
What is cross-sectional study based on ? what does it show ? When does it happen, what does it record , does it do associations and causation ? Is there a sign of incidence, and what is it dependent on ?
¡Case-Series: N>1, several case-reports combined.
¡Cross-Sectional Study: Prevalence Based.
- Shows who has what disease.
- Happens at fix time.
- Records Risk Factor during this time.
- Only Associations, not causation!
- No sign of Incidence.
- Chi squared dependent!
Of observational studies, case control study looks at ? It compares ?, what does it asses , and does not ? What is it used for ?.
do studies that look back at time for period of time lead to causation or association ?
¡CASE-CONTROL Study: DISEASE vs NON-DISEASE
- Compare risk factors in sick versus not sick
- Retrospective,
- Assess Causality, not Prevalence/incidence
- Used for Rarer Diseases, Cheap, Quick.
causation
Cohor study is a observational study, what does it look at ? is it prospective or retrospective or both ? What does it asses and not, what are the group size ?
¡COHORT Study: EXPOSED vs NON-EXPOSED
- Compare diseases in those with/without risk factors.
- Prospective, can be retrospective
- Assess Incident and causality, not Prevalence.
- Large Sample, Expensive, Time-Consuming.
.What happens when the postivie and false negitive get bigger ?
Explain what is a ROC curve?
What are some study types ?
How do you calculate for case-control ?. What do you use ?
.How do you calculate for cohort study ?.
.
How to do calculate a Cohor study
answer how do you get OR, RR, and AR ? with a table 90 40
10 60
There is another cohort calculation, what is attributable risk ?
¡Attributable Risk:
Percent of Cases Due to Exposure
AR% =(RR – 1)/RR
Number needed to treat (NNT) other cohort calculations . What is it ? What type of questions do they aks ?
¡Number Needed To Treat (NNT):
How many people are treated to prevent one case.
NNT = 1/AR
(Also, NNH: how many harmed?)
.Calculate AR, and NNT
What are some observation studies- clinical trials ? What is phase 1 -4
.
What is baseline will be comparison group, subject randomly assigned
double-blind design, and can be crossover design
What are two types of error ?
which one is unknown influences that unpredictable vs jus not using the right tools and using it wrong .
.What are the three things to consider when selection bias ?
What are sampling bias? What is non-respondent bias, Ascertainment bias, late-look Bias, and solution-biased.
This has to do with how study population compares to whole population.
¡Non-Respondent Bias: Volunteers may be different from those that do not volunteer. Sample Population doesn’t represent non-volunteers
¡Ascertainment Bias: Sicker people more likely to enter into treatment and volunteer for studies. Sample population doesn’t represent less sick.
¡Late-Look Bias: Sicker people who are too sick to participate in study are not included in study. Sample population does not represent the more sick.
¡
¡Solutions – Randomize elegible population, weigh the data during analyses (Specific and standardized rates)
What are sampling bias? What is non-respondent bias, Ascertainment bias, late-look Bias, and solution-biased?
What is Measuring bias and of measuring bias, what is Leading Questions ?
¡Leading Questions: How terrible are you? Vs. How would you rank yourself?
What is Measuring bias and of measuring bias, what is Hawthorne Effect ?
¡Observer knows they are in a study, behavior changes. Watched = Stress
What is Measuring bias and of measuring bias, what is Recall Bias?
¡Recall Bias: Responses incorrect due to poor memory, making things up, distorted view on things.
What is Measuring bias and of measuring bias, what is Observer Bias?
¡Observer Bias: Researcher’s past experiences bias study implementation (ie: Psychiatrists think we are all mentally ill.)
¡
What is a solution to deal with measuring bias ?
¡Solutions – Make questions easier, use a control group, equally measurements between groups (if skewed, it’s skewed in both groups.)
What is Expectancy BIAS?
This is a bias caused by researcher unintentionally.
What are two expectancy bias, and what is the solution ?
This is a bias caused by researcher unintentionally.
¡Demand Characteristics: Behavioral difference by smiling or nodding at certain questions, so to get a specific response.
¡Pygmalion Effect: Researcher believes his hypothesis is correct. Treats subjects in a way to get a specific results. IE: If I tell my subject that they will do bad on a test, they will score worse.
¡
¡Solutions – Double Blind studies, properly trained researchers.
What is Lead-Time BIAS ? What is the solution ?
¡Earlier Detection is confused with Living Longer!
¡It can happen when new screening method that catch disease earlier, but fail to prevent death.
¡Solution – examine life expectancy to ascertain any bias.
What is Proficiency BIAS ?
¡Poorly Implemented Interventions or Treatment.
¡Example – Surgical treatment, different surgeons have different outcomes for the same procedure.
¡Solution – Randomize treatment provider.
What is Confounding
¡Hidden causes…
¡Remember the example of coffee and lung cancer?
Strong Association between coffee and lung cancer found.
When the analysis was redone separating smokers from non-smokers, those who smoked and drank coffee had an association to lung cancer, while the non-smoking coffee drinkers did not…
Solution – multiple studies, multiple analyses, and a really good study design.
