Statistics Flashcards
Formula for NNT
1/(absolute risk reduction)
Number of patients to treat to prevent a bad outcome
Number needed to treat (NNT)
Meta-analysis can help to (inc/dec) power (detecting whether or not a difference is significant between two groups) of a study
Increase
Analysis looking at correlation between one or more independent variables (e.g., exposures, risk factors), which can be quantitative or qualitative, and a quantitative dependent variable (e.g., lab values)
Regression analysis
Analysis that quantitatively combines and analyzes the results of several studies to conduct an analysis with higher statistical power than that of individual studies
Meta-analysis
Measure that shows how frequently a specific event (i.e., initiation of drug abuse) occurs in one group relative to how frequently it occurs in another group over time
Hazards ratio
Measures strength of association between 2 categorical variables. For example, to determine the association between a risk factor (e.g., poor socioeconomic status) with an outcome (e.g., early start of taking drugs.
Relative Risk
Needed to harm equation
1/ Attributable Risk =
1/ [(risk of event with new factor like med) - (risk of event in control group)
Researcher’s subconscious influence on the subjects of a research study
Expectancy bias
Selection bias that can occur by recruiting hospitalized patients as the control group
Berkson’s bias
Measure used for case control studies
Odds ratio
Case control studies are almost always prospective/retrospective
retrospective
Increasing power of a study allows for increased possibility to reject a _____ if it is truly false
Null hypothesis
Lifetime prevalence of mental disorders in adolescents as found in the National Comorbidity Survey – Adolescent Supplement is around __%
50%
The occurrence of new cases over a specified period of time in a population
Incidence
The extent to which one diagnosis is replicated by multiple physicians on one patient is best described by ______
Precision
Rate that refers to the proportion of people with a particular condition who end up dying from that condition.
Case fatality rate
2% of pts develop arrthymia, 40% die from the arrythmia. 40% is the case fatality rate
ID the rate type.
Proportion of people dying from a particular disease in the general population.
Mortality Rate
The degree of relationship between two constructs that is theoretically present, and is found to be related in the study (using two measures for anxiety produce similar results, for example)
Convergent validity
A type of observer-expectancy bias that occurs when the expectations of the investigators affect the outcome of the study
Pygmalion effect
A statistical method that includes all patients, including those who dropped out of the study and analyzes them based on their assigned group
Intention to treat analyses
A major concern in studies that experience unequal loss of participants from different groups being compared
Attrition bias
Bias related to the timing of diagnosis rather than the effectiveness of an intervention
Lead time bias
a concern in retrospective studies where participants must recall past behaviors or events
Recall bias
