Causality
If an adverse drug reaction occurs after either a placebo or new drug are given, is it likely that the adverse drug reaction is caused by the new drug? (phrase as answer, not question)
No
The sensitivity of a claims-based disease algorithm if 100 people truly have the disease based on a gold standard, but the claims-based algorithm only identifies 30 of these 100 people as having the disease. (phrase answer as question)
What is a sensitivity of 30%?
A clinical trial phase during which a new drug is administered to a very small group of healthy volunteers to establish the basic dosing, pharmacokinetics, and whether the new drug is safe. (phrase answer as a question)
What is a Phase I clinical trial?
Would a switch from a brand version of a medication to a generic version of a medication be captured in prescription medication claims in most encounter databases? (phrase as answer, not question)
Yes. Note that OTC medication use would not, nor would medications administered during a hospital or SNF stay. Medications purchased fully out of pocket by the patient, such as through the Walmart $4 generic program also would not be captured.
An observational study design that involves sampling individuals who had the outcome event, then identifying control individuals who did not have the event, and then "looking back" in time before the outcome status to assess if a medication exposure was used. (phrase answer as a question)
What is a case-control study?
The bias that is most likely to occur when a patient has a history of the outcome and this history also makes them more likely to receive the treatment. (phrase answer as question)
What is confounding bias?
The specificity of a claims-based disease algorithm if 100 people truly do *not* have the disease based on a gold standard, but a claims-based algorithm only identifies 80 of these 100 people as *not* having the disease. (phrase answer as question)
What is a specificity of 80%?
Is it true that "real world" patients are generally healthier than those enrolled in clinical trials and that is why we need to do post-marketing safety studies? (phrase as answer, not a question)
No, patients excluded from participation in clinical trials are generally sicker and more medically complex.
Data that are generated through billable interactions between a patient and the healthcare system, that are primarily used for reimbursement / payment of claims, and that often clearly identify what medications the patient picked up from the pharmacy. (phrase answer as a question)
What are encounter data? OR, what are health insurance claims data?
An observational study design that involves first identifying whether people were exposed to a medication or not, and then "looking forward" in time after exposure status to assess if an outcome occurred (phrase answer as a question)
What is a cohort study? Note that if data already existed, this is a retrospective cohort.
A covariate that influences both the probability of receiving a medication exposure and the probability of having an outcome event. (phrase answer as question)
What is a potential confounder?
The proportion of people with positive tests who truly have the disease. (phrase answer as question)
What is the positive predictive value?
Absorption, distribution, metabolism, and elimination. (phrase answer as a question)
What are the four components (or phrases or parameters) of pharmacokinetics?
Data that are generated at the point of care as providers record and track the health information of a patient, and that often include laboratory values, vital signs, and free-text notes. (phrase answer as a question)
What are electronic health record data?
The total days covered by medication use divided by the total number of days in the observation period. (phrase answer as question)
What is the proportion of days covered? Note that you can use the dispensing dates from the prescription drug claims and the days supplied to figure this out.
You observe that 1) the adverse drug reaction occurred after the drug was administered, 2) the adverse drug reaction worsened when the dose of the drug was increased, and 3) an adverse drug reaction ceased when the drug was discontinued. (phrase answer as question)
What are findings or critical pieces of information that support the causal role of a drug in an adverse drug reaction?
The bias that occurs when patients who gave birth to children with a birth defect (cases) are more likely than those who gave birth to children without a birth defect (controls) to recall medication exposures that they had while they were pregnant. (phrase answer as question)
What is a differential exposure misclassification?
Clinical trials are underpowered to detect very rare adverse drug events. (phrase answer as question)
Three types of information that both encounter and electronic health record databases contain. (phrase answer as question)
What are demographics, diagnoses, and medications? (Other answers are possible)
Initiation, implementation, and persistence. (phrase answer as question)
What are the three stages of medication adherence?
On a case report, information about the patient's characteristics (e.g., demographics), the adverse drug reaction or event that occurred, test and procedure results, a narrative case summary, and specific product identification. (phrase answer as question)
What are the components necessary to assess causality in an ideal case report?
The bias that occurs when health care professionals know the person's medication exposure status and are more likely to monitor for and document/measure outcome events among those who received the medication (versus those who did not). (phrase answer as question)
What is a differential outcome misclassification?
Drugs being catalyzed by enzymes, often in the liver. (phrase answer as question)
What is metabolism?
If a patient sees multiple providers that use different electronic health record systems, information in any given database will be incomplete regarding their total care received. (phrase answer as question)
What is a key limitation of electronic health record data?
Patient, systematic and medication-related. (phrase answer as question)
What are types of factors that influence medication adherence?