HS 326 flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Allocation concealment

Procedure that hides the upcoming treatment assignment from the person enrolling participants, preventing selection bias and preserving the integrity of randomization.

Intention-to-treat analysis

Analysis that includes all randomized participants in their assigned groups regardless of adherence, preserving randomization and estimating real-world effectiveness.

Cluster randomized trial

Trial in which groups rather than individuals are randomized; analysis must account for intracluster correlation, which inflates required sample size.

Confounding variable

A factor associated with both the exposure and the outcome, not on the causal pathway, which can distort the observed exposure-outcome relationship.

Crossover design

Design where participants receive each intervention in sequence with a washout period; assumes no carryover effect and controls for between-person variability.

GRADE certainty rating

Framework rating evidence as high, moderate, low, or very low, downgrading for risk of bias, inconsistency, indirectness, imprecision, and publication bias.

Funnel plot asymmetry

Scatterplot pattern in meta-analysis where smaller studies show larger effects; suggests publication bias or small-study effects, though heterogeneity can also cause it.

Absolute risk reduction

Difference in event rates between control and treatment groups; its reciprocal gives number needed to treat and conveys clinical magnitude better than relative measures.

Surrogate endpoint

Laboratory or intermediate measure used as a substitute for a patient-important outcome; valid only when evidence links it reliably to clinical benefit.

RoB 2 tool

Cochrane risk-of-bias instrument for randomized trials, assessing randomization, deviations, missing data, outcome measurement, and selective reporting.

Sensitivity and specificity

Proportions of diseased and nondiseased people correctly classified by a test; they are intrinsic test properties largely independent of prevalence.

Positive predictive value

Proportion of positive test results that are true positives; increases as disease prevalence rises, unlike sensitivity and specificity.

Likelihood ratio positive

Sensitivity divided by one minus specificity; values above 10 substantially raise post-test probability when combined with pre-test odds.

Number needed to treat

Reciprocal of absolute risk reduction; the number of patients treated to prevent one additional adverse outcome over the study period.

Statistical power

Probability of detecting a true effect when one exists, equal to 1 minus beta; conventionally set at 0.80 during sample size planning.

Standard error of mean

Standard deviation divided by the square root of sample size; quantifies precision of the sample mean and shrinks as sample size grows.

Incidence versus prevalence

Incidence counts new cases per population at risk per time; prevalence counts all existing cases and approximates incidence multiplied by duration.

Case fatality rate

Number of deaths from a specified disease divided by number of confirmed cases, usually expressed as a percentage over a defined period.

Lead-time bias

Apparent survival improvement caused by earlier detection through screening, without any actual postponement of death.

Length-time bias

Screening preferentially detects slower-progressing disease with longer preclinical duration, making screened groups appear to survive longer.

Herd immunity threshold

Proportion of a population needing immunity to halt transmission, approximated as one minus one divided by the basic reproduction number.

Age-adjusted rates

Rates standardized to a reference population age structure, allowing valid comparison of disease burden across populations with differing age distributions.

Pre-test to post-test probability

Convert pre-test probability to odds, multiply by the likelihood ratio, then convert resulting odds back to probability.

Therapeutic threshold

Probability of disease at which expected benefit of treatment exceeds expected harm, guiding the decision to treat empirically.

Shared decision-making

Collaborative process combining best evidence, clinician expertise, and patient values and preferences, often supported by decision aids.

Minimal clinically important difference

Smallest change in an outcome that patients perceive as beneficial, used to judge whether statistically significant findings matter in practice.

Respect for persons

Belmont principle requiring informed consent, voluntariness, and extra protection for those with diminished autonomy in research participation.

Research misconduct

Fabrication, falsification, or plagiarism in proposing, performing, or reporting research; excludes honest error and legitimate disagreement.

Authorship criteria

Requires substantial contribution, drafting or critical revision, final approval, and accountability for the work; gift and ghost authorship violate integrity.

Institutional review board

Committee that reviews and approves human-subjects research, ensuring risks are minimized and informed consent and equitable selection are documented.

Teach-back method

Communication technique asking patients to restate information in their own words, confirming comprehension and revealing gaps in understanding.

SBAR handoff

Structured communication format covering Situation, Background, Assessment, and Recommendation, used to convey clinical information clearly across teams.