Module 3-Slides Flashcards
Population
(Target Population)
ENTIRE group of interest to whom the study findings may be generalized
Accessible Population
SUBSETS of the entire / target population who are AVAILABLE for researchers to study
Sample
SUB-GROUP of the accessible population; serves as reference group to estimate characteristics of and draw conclusions
*Final Population being studied
Inclusion Criteria for Selection
Primary traits of the target and accessible populations that will make someone ELIGIBLE to be a participant (characteristics of interest)
Exclusion Criteria for Selection
Factors that would preclude someone from being a subject (INELIGIBLE)
EX: undesirable attributes - confounding factors/variables
Confounding/ Extraneous Variables
VARIABLES that may CONFOUND the results or interfere with interpretation of the findings
Probability Samples
(Selecting approach)
created via a process of RANDOM sampling or selection; to generate LESS BIASED AND CLOSER REPRESENTATIVE of the population
Random sampling
equal chance/probability for all in population to be selected
Nonprobability Samples
(Selecting approach)
made by non-random methods; more frequently used in CLINICAL STUDIES
Types of Probability Sampling
- Simple random
- Systematic
- Stratified random
- Cluster
- Disproportional
Types of Nonprobability Sampling
- Convenience
- Purposive
- Quota
- Snowball
Convenience Sampling
(nonprobability)
use of available participants
“accidental sampling”
MOST FREQUENTLY USED SAMPLING METHOD
Purposive Sampling
(nonprobability)
used when specific expertise / experience of participants is needed
handpicked /invited bc of known characteristics
used in qualitative studies
Quota Sampling
(nonprobability)
incorporate a stratification but lack of a randomization
subjects recruited to represent various levels/classes
Snowball Sampling
(nonprobability)
used when it is difficult to find subjects
recruit via communication of “word-of-mouth”
usually when dealing with difficult/sensitive topics
Examples of Convenience Sampling
Subjects found in a clinic
Volunteer subjects through ads/flyers
Sample Size
Direct impact on validity of statistical conclusion; power analysis to determine the adequate sample size
LARGER samples tend to be
MORE REPRESENTATIVE of the populations than smaller samples
Decrease sampling error
Sampling Error
difference between an observed STATISTIC from a sample and the population PARAMETER
Types of Clinical Trials
- Therapeutic trials: effect of intervention
- Diagnostic trials: accuracy of diagnostic procedures
- Preventive trials: evaluation of whether a procedure or agent reduces risk of developing a disease or disorder
Randomized Controlled Trials (RCTs)
Gold standard for experimental research
Random assignment
Two or more groups (experimental + control groups)
Measurements: Pre- / Post - interventions
Assignment of Subjects / Participants
Can be done either NONRANDOMLY or RANDOMLY, determined by the purpose and nature of the study
-When COMPARING patients with age- and gender-matched healthy control subjects
Balanced group
Achieved when there is NO LACK OF RANDOMIZATION
= # subjects/ = distribution of clinical/ + confounding characteristic
Random assignment is preferred due to what?
validity issues
Independent Variable
(IV)
Intervention
active variable can be MANIPULATED by researcher and assigned to levels of IV (=# of groups)
Attribute variable cannot be manipulated (age/gender, disease duration/staging)
Random Assignment
process of assigning subjects to groups (=chance of being assigned to any group)
minimizes BIAS by creating groups that are similar at the start of trial
Is random assignment the same as random selection?
NO
Selection
process by which subjects are drawn or selected from the population
RECRUITMENT/ENROLLMENT
Assignment (or Allocation)
process by which subjects are placed into or allocated to a group within the study
MANAGEMENT
Random Assignment Strategies
- Simple random
- Block random
- Stratified random
- Cluster random
- Random consent design
- Assignment by patient preference
- Run-in period
Concealed Allocation (Assignment)
ensures group assignment is done w/out knowledge of those involved in the experimental process
minimizes BIAS in group formation
Control Groups
inactive controls: placebo, sham, attention control group
wait list control: delayed start
active controls: (practical, feasible /ethical reasons)
-standard care or others with known efficacy
Blinding (Masking)
minimizes observation BIAS by ensuring those involved in the study are unaware of a subject’s group assignment
-participant/subject/pt
Double Blind
at least 2 of the 3 parties are unaware of group assignment
Single Blind
at least 1 of the 3 parties is unaware of group assignment
Explanatory Trial
(RCT, aka True experimental)
ideal situations to control for confounding
-strict and broad exclusion criteria
-standardized treatment protocol
Pragmatic Trial
(RCT)
more reflective of practice situations
-diverse pt populations w/fewer or min exclusion criteria
-active controls (standard/other common care)
-Tx reflects practice + less standardized
Phases of Clinical Trials /
New drug, surgical procedure
Phase I: Is the Tx safe?
Phase II: Does the Tx work?
Phase III: How does this Tx compare w/ standard care?
Phase IV: What else do we need to know?
