BIOEPI “ACCURACY”

Accuracy 

  • refers to the closeness of a measurement or an estimate to the true value. 

Validity 

  • refers to the extent to which a test or measurement accurately represents what it intends to measure.

Construct Validity 

  • evaluates how well a test or measurement reflects the theoretical concept it is intended to measure.

Internal Validity 

  • assesses whether the observed effects in a study are truly due to the independent variable and not due to other confounding factors.

External Validity 

  • determines the extent to which study results can be generalized to other populations, settings, or times.

Bias

  • any systematic error in the design, conduct, or analysis of a study that results in a mistaken estimate of an exposure’s effect on the risk of disease.

Differential Bias 

  • occurs when measurement errors differ between study groups, leading to systematic misclassification and distorted results

Nondifferential Bias

  • affects all study groups equally and generally biases results toward the null hypothesis, making it harder to detect a true association.

Selection Bias

  • arising from how participants are chosen

  • occurs when study participants are not representative of the target population, leading to incorrect associations between exposure and outcomes. 

Information Bias 

  • from errors in measurement or reporting

  • arises from errors in data collection, measurement, or recall.

Recall Bias 

  • differences in memory accuracy between study groups

Interviewer Bias 

  • when researchers unintentionally influence responses

Minimizing bias 

  • is essential for maintaining the validity of a study. 

Confounding Bias 

  • from unaccounted third variables affecting the relationship between exposure and outcome

Restriction

  • Excluding participants with a known confounder. 

Matching

  • Pairing study participants with similar characteristics. 

Randomization

  • Randomly assigning participants to different groups to distribute confounders equally. 

Multivariate Analysis

  • Using statistical models (e.g., regression) to adjust for confounding effects.

Stratification

  • involves dividing a study population into subgroups (strata) based on a potential confounder (e.g., age groups, gender).

Adjustment

  • Is a statistical technique used to control for confounding.

Standardization

  • Adjusting rates to a standard population.

Multivariable regression models

  • Using statistical methods (e.g., logistic regression) to control for multiple confounders simultaneously.

Precision 

  • refers to the consistency and reproducibility of a measurement

Reliability 

  • is the extent to which a measurement tool consistently produces the same results under the same conditions.

Test-retest reliability

  • Measures consistency over time. 

Inter-rater reliability

  • Assesses consistency between different observers.

Internal consistency

  • Examines how well different parts of a test measure the same concept.

Statistical significance

  • determines whether an observed effect is likely due to chance.

P-values

  • quantifies the probability that the observed data (or more extreme results) would occur under the null hypothesis

Sample size

  • refers to the number of subjects in a study.