Evidence Based Practice (chapter 5) Flashcards
Step 1 of EBP
Formulating the Clinical or Research Q
Step 2 of EBP
Searching for the evidence
Step 3 of EBP
Evaluating the evidence
Part A of Step 3
Determine the level of evidence
Part B of Step 3
Rate the risk of bias of the evidence
Part C of Step 3
Grade the cliical recommendation under consideration
Step 4 of EBP
Integrating Scientific Evidence,
Clinical Expertise, and
Patient Preferences into Clinical Decision-Making
Step 5
Evaluating the Process
PICO stands for
Population
Intervention
Comparison (intervention)
Outcome
PICO
a helpful procedure in framing research questions
Most rigorous class level of evidence
Class I
Least rigorous class level of evidence
Class VI
Two types of variables in an experiment
independent & dependent
Independent variables
manipulated by the experimenter who is trying to test a hypothesis
Dependent variable
The one that is measured
Treatment/intervention
Refer to applications or procedures that health professionals administer to lessen the effects of, or cure, a disease or a disorder.
Placebo
inactive medications or procedures that are administered the same exact way as the experimental treatment and serve as a comparison
Class I
Well-designed systematic reviews with meta-analysis of more than one double-blinded, randomized, controlled clinical trial
Class I research methods
May involve either primary measurement methods or secondary measurement methods
Primary measurement methods
involve investigators collecting and analyzing their own data for a single experiment
Secondary measurement methods
Involve investigators experimenters or clinicians considering or analyzing data collected in one or more studies conducted by other investigators
Example of secondary measurement methods
Conducting a systematic review of the evidence assessing the effectiveness of a particular diagnostic procedure or treatment of hearing impairment
Systematic reviews
Can be a qualitative review of the evidence or quantitative review
quantitative review
Involves a meta-analysis, or the use of special statistical procedures that combine the results of more than one investigation in order to measure the magnitude of a treatment effect.
systematic review with meta-analysis
Permits the the combining of the results of several studies together to increase the magnitude of the effect of the study
Effect size
a special metric that determines the degree of impact of significant findings
Why are systematic reviews with meta-analysis important?
They are needed to validate research because causality between treatment and positive patient outcomes cannot be established on the basis of a single investigation.
5 reasons for systematic reviews with meta-analysis
1) Get to the “bottom line” of all studies involving a particular treatment or intervention
2) To estimate precisely the amount of benefit derived from a treatment or intervention
3) To increase the number of patients in clinically relevant subgroups
4) Helps solve discrepancies among the findings of studies; can help sort conflicting results
5) can help plan future studies
Class II
Double-Blinded, Prospective Randomized, Controlled Clinical Trials
DBPRCT
Acronym for Double-Blinded, Prospective Randomized, Controlled Clinical Trials
Double-blinding
means that neither the experimenter nor the participant has knowledge of administration of the treatment/intervention or placebo
Prospective randomized
describes studies in which patients are
1) enrolled as participants prior to conduction of the experiment
2) randomly assigned to groups receiving either the treatment or a placebo
3) Measured after some period of time
4) Compared for any statistically significant differences attributable to the treatment or intervention
Advantage of DBPRCTs
Neither the experimenter nor the participant knows whether the treatment or placebo is being administered, eliminating both experimenter and participant bias
Experimenter bias
anything that the experimenter does that has measurable influence on the dependent variable that is not attributable to the independent variable
Participant bias
anything a participant does that has a measurable influence on the dependent variable that is not attributable to the independent variable
Outcome measures
data collected to determine the benefit of treatment
Another advantage of DBPRCTs
participants are randomly assigned to either a control or treatment group
Random assignment
implies that each participant has a 50-50 chance of being assigned to receive the experimental or placebo treatment
benefit of random assignment
reduces sampling bias
Sampling bias
Measurable influence that assignment of participants to treatment or control groups has that is directly not attributable to the independent variable. (Assignment of participants must be based on chance alone)
DBPRCTs
gold standard for scientific evidence
disadvantage of DBPRCTs
difficult to design, execute, & complete
another disadvantage of DBPRCTs
describes outcomes for treatment variables in highly-controlled situations, no those found in the “real world”
another disadvantage of DBPRCTs
ethical dilemma due to random assignment of participants to treatment and control groups
another disadvantage of DBPRCTs
Requires a great deal of time in validating treatment regimens, delaying their availability to the public
fifth disadvantage of DBPRCTs
optimal experimental circumstances necessary for clinical trials research may have limited external validity in the real-world clinical practice involving all types of patients across a wide variety of afflictions, behaviors, and service delivery sites
Class III
Non randomized Intervention Studies
Non Randomized Intervention Studies
also called quasi-experimental studies
Nonrandomized intervention studies
use designs that lack characteristics of DBPRCTs (Lacks random assignment of participants into experimental and control groups)
Nonrandomized intervention studies/quasi-experimental studies
most common in Communication Sciences and Disorders because they are more conducive to the realities of a busy speech & hearing clinic
Nonrandomized intervention studies/quasi-experimental studies
Involves grouping of participants by convenience
Potential weakness of Quasi Experimental studies
The experimenter cannot be sure that the groups are equivalent at baseline or at the beginning of the study; the groups should be matched for critical variables such as age, gender, and so on to strengthen the study
Class IV
Nonintervention studies:
Cohort studies,
Case-controlled studies,
Cross-Sectional Surveys
Non intervention studies
Can be prospective or Retrospective
Prospective studies
Recruit participants prior to the collection of data
Retrospective studies
Data are collected after some sort of event or occurrence
case-controlled studies
participants are recruited after they have developed some type of pathology along with a disease-free control group; are measured on some type of predictor variable; and then statistically compared for significant difference on (a) predictor variable(s) suggesting possible cause(s) or associative variable(s) for developing NIHL.
Case-controlled studies
are retrospective studies
Cohort studies
can be prospective or retrospective
prospective cohort studies
Those in which participants are recruited in the present, before an outbreak of a disease or disorder to assess who does and does not develop a particular condition in teh future
2 purposes of prospective cohort studies
1) to determine the incidence of certain outcomes over time (i.e. descriptive)
2) to analyze associations b/w and/or among risk factors and those outcomes
Incidence
the number of new cases of a disease or disorder that develop over a specified period of time per the number of individuals at risk
Associations
Connections made b/w risk factors and certain diseases
Retrospective cohort study
identifies a sample of individuals who already have a disorder and then measures predictor variables
Prevalence studies
Cross-sectional studies
Prevalence studies/cross sectional studies
Determine the number of individuals that have a particular disorder or disease at one point in time per the number of people at risk
Prevalence
Number of persons per some number of the population that have some sort of condition