Glossary - Research Basics and Terminology

• Classification scale; each person is assigned to one “type”
• Type of data: categorical, qualitative, discrete
• Example: blood type (A, B, 0), type of breath sound, type of arthritis, etc.

• Bell-shaped
• Mean, mode, and median all have the same value

• Data that is symmetrically distributed around a mean value
• Description of distribution via mean and standard deviation

• Statistical hypothesis
• The population parameter is equal to the claimed value

• The number of patients that need to be treated before one person would experience the desired outcome
• Number needed to treat to benefit: number of patients that need to be treated before one person would experience a beneficial outcome
• Example: NNT= 20: this means that 20 people need to be given a stroke prevention drug before one stroke can be prevented
• Number needed to treat to harm: number of patients that need to be treated before one person would experience a harmful outcome
• Example: NNT= 40: this means that 40 people can be treated with cardiac surgery before one person dies 

• Data collection method which will always come to the same result
• Example:

• Chance of an event occurring; this is calculated by taking the number of individuals in the sample who experience the event divided by the number of individuals for whom the event did not occur
• Mainly used in case-control studies
• Small risk → Odds Ratio ~ Relative Risk
• Ranges:

1. 1. OR> 1.0 – exposure may increase the odds of a disease
   2. OR> 1.0 – exposure does not affect the odds of outcome
   3. OR< 1.0 – exposure may decrease the odds of a disease

• Example: OR= 3.7 for likelihood to die from a new antihypertensive drug as compared to an existing drug

→ Patients who received the new antihypertensive drug die 3.7 times more often than patients who received an existing drug.

→ The odds to die with the new antihypertensive drug is 3.7 times higher than with the existing drug.

Further information: Explaining Odds Ratios

• Ranking scale; no known interval
• Type of data: categorical, semi-quantitative (= quantitative but open to individual interpretation), discrete
• Example: pain levels, joint laxity grades, Manual Muscle Testing grades, level of assistance

• Value that a certain percentage of data falls below
• Example: 

P=       Population

I=        Intervention (or diagnosis, prognosis) being evaluated

C=       Comparison (usually to gold standard or no treatment)

O=       Outcome

Example:

Is physical activity at least twice a week for more than ten minutes as effective as the antihypertensive X in preventing high blood pressure in adults over 18 years?

P=       adults over 18 years

I=        physical activity at least twice a week for more than ten minutes

C=       antihypertensive X

O=       prevention of high blood pressure