# Behavioral Sciences Flashcards

What is a cross-sectional study?

collects data from a group of people to assess frequency of disease at a particular point in time

What is the primary limitation of a cross-sectional study?

while it can demonstrate risk factors associated with a disease, it does not establish causality

What is a case-control study?

one that compares a group of people with disease to a group without disease and looks for prior exposure or risk factors

What is a cohort study?

one that compares a group with a given exposure or risk to a group without such an exposure to see if exposure affects the likelihood of disease

A case-control study can be used to calculate the ___ while a cohort study can be used to calculate the ___.

- case-control: odds ratio

- cohort study: relative risk

What elements of a clinical trial can be used to improve the quality of such a study?

- randomization
- controls
- double- or triple-blinding

What is a triple-blind study?

one in which the treatment recipients, treatment providers, and data analyzer are all “blind”

What are the four phases of a clinical trial? Describe the group of subjects used for each and what is being assessed?

- phase I: small number of healthy volunteers used to assess safety, toxicity, pharmacokinetics and pharmacodynamics
- phase II: small number of patients with disease used to assess whether it works (efficacy, optimal dosing, adverse effects)
- phase III: randomized study in which patients are assigned to either the treatment under investigation or to the best available treatment/placebo used to determine if it is good or better than the current standard of care
- phase IV: surveillance of patients after the treatment has been approved used to detect rare or long-term adverse effects

How are PPV and NPV calculated?

PPV = TP/(TP + FP) = proportion of positive test results that are true positives NPV = TN/(TN + FN) = proportion of negative test results that are true negatives

Describe PPV in words.

the probability that a person who has a positive test result actually has the disease

Describe NPV in words.

the probability that a person with a negative test result actually does not have the disease

What is pretest probability?

the baseline risk of the population from which the individual is pulled (i.e. prevalence within the population)

How does pretest probability relate to PPV and NPV?

- PPV varies directly with pretest probability

- NPV varies inversely with pretest probability

PPV and NPV vary according to what other measure?

pretest probability

Why do we typically use sensitivity and specificity rather than PPV and NPV?

because sensitivity and specificity are fixed properties of a test whereas PPV and NPV vary based on pretest probability

Describe sensitivity in words.

the probability that when the disease is present, the test is positive

Describe specificity in words.

The probability that when the disease is absent, the test is negative

How are sensitivity and specificity calculated?

sensitivity = TP/(TP + FN) = proportion of people with the disease who test positive specificity = TN/(TN + FP) = proportion of people without the disease who test negative

How do sensitivity and specificity mathematically relate to the false-negative and false-positive rate, respectively?

sensitivity = 1 - false negative rate specificity = 1 - false positive rate

What is the difference between incidence and prevalence?

- incidence looks at new cases in the given time frame

- prevalence looks at all cases that exist at a specific point in time

For what sorts of diseases does prevalence nearly equal incidence?

those with a short duration of disease

How is the odds ratio calculated?

ad/bc (cross multiple)

Odds ratios are typically reserved for what kind of study?

case-control studies

How is relative risk calculated?

RR = [a/(a+b)]/[c/(c+d)] = risk of developing disease in the exposed group divided by risk in the unexposed group

Relative risk is typically reserved for what kind of studies?

cohort studies

Give an example of relative risk calculation.

21% of smokers get lung cancer and 1% of non-smokers do, so RR = 21/1 = 21

How is attributable risk calculated?

a/(a+b) - c/(c+d)

Give an example of attributable risk.

21% of smokers get lung cancer and 1% of non-smokers do, so smoking has an AR of 20%

How is relative risk reduction calculated?

1 - RR

Give an example of relative risk reduction.

if 2% of vaccinated individuals get the flu and 8% of unvaccinated individuals do, then RR= 2/8 and RRR = .75

How is absolute risk reduction calculated?

c/(c+d) - a/(a+b)

Give an example of absolute risk reduction.

if 2% of vaccinated individuals get the flu and 8% of unvaccinated individuals do, then ARR is 6%

How is NNT calculated?

1/ARR

How is NNH calculated?

1/AR

What is the difference between a test’s precision and accuracy?

- precision (aka reliability): consistency and reproducibility
- accuracy (aka validity): absence of systematic error or bias

Which kinds of error decrease the precision of a test? Which kind decreases the accuracy of a test?

Random error decreases the precision

Systematic error decreases the accuracy

What two effects does increasing precision of a test have on analysis of the results?

- reduces standard deviation

- increases statistical power (1-B)

The most common type of selection bias is what?

sampling bias

How can selection bias be reduced?

randomization