EBP Midterm

• Evidence-Based Practice (EBP): Combines the best scientific research, clinical expertise, and patient's values to inform patient care and optimize benefits
• Goal of EBP: To ensure patient care is informed by the best available evidence

Key Vocabulary in EBP:

• Scientific Research: Empirical evidence from the systematic testing of a hypothesis, including both clinical research (human subjects) and nonclinical research (animals, cell specimens)
• Clinical Expertise: Knowledge from years of patient care and improvement efforts, shared through education, mentorship, and reflective practice
• Patient Values and Circumstances: Patient's beliefs, preferences, expectations, cultural identification, medical history, access to services, and environment
• Shared Informed Decision: A shared choice between therapist and patient, informed by the best evidence

The EBP Process:

• Identify a question
• Search for evidence
• Appraise the evidence
• Integrate the evidence
• Evaluate the outcomes

Barriers to EBP:

• Time constraints for reviewing research
• Lack of generalizability of research to individual patients
• Deficient research skills of clinicians
• Intimidating statistical concepts
• Difficulties in searching and appraising research
• Accessibility issues for research information
• Inconsistent EBP culture among therapists
• Importance of EBP: Enhances confidence and assists patients in choosing care options based on systematic assessment, shifting from habit-based practice and leading to patient demand for evidenced based carre

Applying EBP:

• Best evidence may not suit every patient due to circumstances and barriers
• Acknowledge the scientific rationale, advantages, and disadvantages when evidence lacks
• Clinicians should make decisions based on available data
• Clinical expertise requires quality evaluation, like research evidence
• Patient input is crucial in decision-making

Five Steps of the EBP Model:

• Step 1: Identify the need for information and create a searchable clinical question by gathering patient info
• Step 2: Search for available research evidence
• Step 3: Assess research validity and applicability to the clinical question
• Step 4: Integrate research evidence with clinical expertise and patient values
• Step 5: Assess effectiveness and find avenues of improvement at both patient and practice levels

Three Pillars of Evidence for Optimal Outcomes:

• Scientific Research
• Clinical Expertise, inclusive of own therapist values/preferences
• Patient Values and Circumstances

EBP Evidence Sources:

• Work together to determine the optimal treatment
• Evidence Pyramid Top Level: Systematic reviews (SRs) combining studies, the highest level of research evidence
• Below SRs: Individual studies like randomized controlled trials (RCTs), cohort studies
• Evidence Summaries: Clinical practice guidelines (CPGs) considered "secondary" studies and higher. Combine research, expertise, and patient input for recommendations

Key Concepts & Vocabulary:

• Searchable Clinical Question: A query that allows efficient access to the best research evidence
• Patient characteristics: Patient management: Outcome: The elements of a searchable clinical, clinical outcome
• Background Questions: Seek general condition
• Foreground Questions: Specific clinical questions related to a particular patient, that can be answered with the aid of clinical guides or prior research results
• Boolean Search: search refinements through AND, OR, NOT
  • AND narrows, OR broadens
• Keywords: Searchable Terms
• MeSH (Medical Subject Headings): A controlled vocabulary indexing MEDLINE articles, improving information retrieval to find the best search term for the topic
• Search Engine: user database interface
• MEDLINE: Database of evidence
• Clinical Queries: PubMed tool uses pre-defined filters to target clinical articles
• My NCBI: tool that helps saves searches, create filter alerts, and enable personal filters
• Cited By: feature by google scholar that shows the publications which have cited a source
• PICO: used to form questions on specific interventions, Patient/Population, Intervention, Comparison, Outcome

Databases & Search Engines

• Google Scholar: Searches a range of resources providing helpful search tools and cited by features.
• PubMed: Comprehensive health care research, can be refined with techniques like MeSH
• TRIP Database: Finds research and bias
• PEDro (Physiotherapy Evidence Database): physical therapy topics
• APTA (American Physical Therapy Association): Guidelines and Summaries
• AHRQ (Agency for Healthcare Research and Quality): For clinical practice guidelines
• Cochrane Library: Collection of SRs (systematic reviews)

Types of Research

• Systematic Review: information from several studies.
• Randomized Controlled Trial (RCT): participants are randomly assigned.
• Clinical Practice Guidelines (CPGs): Recommendations based on evidence and expert opinion.
• Descriptive Research: Describes a situation in observational designs.
• Predictive Research: Predicts an outcome given measures.
• Causal Research: Determines changes in an intervention.

• Efficiency: strategies locating relevant material quickly.
• Full Text: entire research material
• Filters: narrowing factors such as time, languge, type, et
• Synonyms: Alternate search names
• Impairments: functional or structure problems

Steps for Searching

Identify Information Development, List keywords, Refine from Mesh

List keywords search "OR",Narrow search and filter results while saving links and strategies.

• Applicability: Evaluating can trials benefit patients addressing a clinical question.
• "Close Enough": Perfect matches unattainable, applicable enough.
• Quality: Reliability and validity inform treatment.
• Evidence-Based Practice (EBP): Applying findings

Study Design and Bias

• Randomization: Reduces initial trials
• Stratification: categorizing outcomes, trial runs.
• Blinding: Minimizes bias from all groups.
• Sham Group: Fake tests
• Placebo Effect: Enhances treatment.
• Intention-to-Treat Analysis (ITT): Analyzing dropout assigned values, bigo
• Participant Attrition: Track the process
• Preservation of Original Groups: Maintenance
• Control Group: Standard/Not treatment (to comparitively measure)
• Comparison Group: Alternative treatment to comparitively measure by

Study Factors

• Intervention: Should be measured reality trials.
• Outcome Measures: Patient outcome requirements.
• Surrogate Outcomes: Cholesterol tests,
• Clinically Meaningful Outcome Variables:: outcomes of relevance and importance
• Follow-up Duration: Short term impact
• Realistic Intervention: Treatment that works or is real.

Sampling-Consecutive/Selective, Sampling and power

• Sequential: test all the related clinics results.
• Selective: Recruiting participants with others not chosen randomly.
• Sample Size: Results from the same size
• Power: The potential and impact from study.

Measurement

• Construct: Measure of interest
• Measure: measure of outcome or construct
• Subjective vs. Objective Measures:.
• Patient-Reported Measures: the patients rating and impacts of a condition
• Observer-Rated Measures: the rates of observations.

Terms and definitions on measure of the Bias and Conflict of Interest.

Type Of Bias and how to remedy it

• Selection: Random assign, concealed assignment
• Performance: Treatments, intensity equal measure
• Detection: Blind the evaluators.
• Attrition: Intention to "treat" and include start/finish
• Reporting: Disclose incentives

More Remedy to Bias:

• consistent measures, sampling across the groups using sham and sufficient sample size that includes testing.
• Randomization is a key method, if sample size is large enough.
• Blinding reduces risk of effect.
• Intention-to-treat analysis maintains study groups integrity
• Studies quality of conflict high

• Nominal Data: Represents categories or groups. Examples: treatment groups, sex
• Ordinal Data: Represents categorical data with ranking, but no uniform intervals. Examples: Likert scales
• Discrete Data: Data that can only take on specific, separate values. Examples: number of times someone has had disease
• Continuous Data: Data that can take on any value within a range. Examples: ratio scales with meaningful zero, interval scales with arbitrary zero

Descriptive Statistics

• Summarize/describe main dataset features.
• Central tendency describe the average of a dataset, mean, median and mode being the different measures.
• Variability: spread of measure,
• Inferential Statistics*

Interpret probability differences between groups

• Null Hypothesis*: no differences
• P Value: alpha Level*

The likely and acceptable that these results are due to the original alpha set for values of change between group is significant

Measures between groups should be analyzed with reliability

• Paired Samples T Test/Variance* show a variance outcome.

• Measuring results can involve Cronbach's Alpha, 0.05 Threshold and other metrics for treatment and comparison groups.
• Recruiting can be challenging, which means should report how the sample recruited,
• study should also list inclusion/exclusion criteria.

• Nominal data: Data representing groups
• Continuous Data is also scalpe, which means there is a unit of measurement
• Ordinal Data Data with a specific order
• Descriptive : describe data
• Inferential : interpret result reliable people, measures can be different Alpha: signiance set at 0, this means it will reject the null hypothesis Measures can be between groups set tests etc.

What's A Outcome measure?

• A measure of patient progress through therapy
• Two main results of Question/Performance
• Can be an ICF of health and participation
• Intrinsic Quality can include : Reliability, Validity and clinical impact

Validity: all tests should be a similar quality

• Internal Consistency
• Multiple tests
• When one rater is doing something the results multiple times and comparing to a new measure

Key Concepts Related to Change

Clinicians should be aware and review all the components with good understanding.

Criterion, content and constructs should be taken into deep review.

• Asses the patients status
• What makes a measures different each other.
• Test/Retest, with time line set in place
• Validity, that has content