Loading...

PPT – The Basics: Evidence-Based Practice for Physical Therapists and Physical Therapist Assistants, Online Training Module PowerPoint presentation | free to download - id: 3b778f-Y2U5Y

The Adobe Flash plugin is needed to view this content

The Basics Evidence-Based Practice for Physical

Therapists and Physical Therapist

Assistants,Online Training Module

- Jessica Lambeth, MPT
- jessicamlambeth_at_yahoo.com

Module Purpose The purpose of this online

training module is to share the basics of

evidence-based practice (EBP). This module

focuses on general concepts of EBP, with clinical

scenarios related to school-based physical

therapy.

Module Purpose, continued The main emphasis

of this module is not to promote certain

treatments or tests and measures or tell you

what the recent literature reveals about

pediatric physical therapy practices. As will

be discussed in the coming slides, the research

answer alone is not the correct answer. In

addition, EBP emphasizes the necessity of

learning to perform searches and evaluate the

material independently, related to clinicians

and patients current circumstances. While

initially time consuming, the results are

worthwhile.

Module Purpose, continued For those without a

recent EBP background or training, this module

will not result in immediate efficiency in

literature searches, statistical interpretation,

etc. Hopefully it will, however, encourage more

questions and motivation for EPB. It will also

enable the NC DPI to focus educational sessions

on meeting needs in the area of EBP and to gauge

the current knowledge of school-based PTs and

PTAs. Your feedback and comments will help to

plan future learning opportunities and

resources.

Pretest Please stop here and complete the

pretest if you have not already done so. Please

also keep track of your time, as directed in the

instructions. Also, I dont know answers

have been included to ensure that we receive

appropriate feedback. If you really do not know

the answer, in the pretest or the posttest,

please indicate that in your response. This will

greatly assist in providing necessary learning

opportunities in the future, as well as help us

to evaluate the effectiveness of the online

module format and material.

References The content of this EBP module is

largely based on curriculum from courses within

the transitional Doctor of Physical Therapy

program at Rocky Mountain University of Health

Professions in Provo, Utah. (web address

www.rmuohp.edu)

References, continued Much of the material

within this online training module can be found

in the following texts Guyatt G, Rennie D.

Users' Guides to the Medical Literature-

Essentials of Evidence-Based Clinical Practice.

Chicago AMA Press 2002. Jaeschke R, Singer J,

Guyatt GH. Measurement of health status

ascertaining the minimal clinically important

difference. Control Clin Trials

198910407-15. Portney LG, Watkins MP.

Foundations of Clinical Research Applications to

Practice. 2nd ed. Upper Saddle River, NJ

Prentice-Hall Inc, 2000. Rothstein JM.

Autonomous Practice or Autonomous Ignorance?

(Editorial) Physical Therapy 81(10), October

2001.

References, continued Much of the material

within this online training module can be found

in the following texts Straus SE, Richardson

WS, Glasziou P, Haynes RB. Evidence-based

Medicine How to Practice and Teach EBM. 3rd ed.

Edinburgh Churchill Livingstone 2005. All

clipart came from Microsoft Office http//office.

microsoft.com/en-us/clipart/default.aspx.

Accessed 05/09/2009.

Module Outline 1) What is EBP? Slides 9-22

2) Statistics Review Basic Research, Slides

23-115 3) How to Search for EBP, Slides

116-152 4) How to Interpret Research Related to

EBP, Slides 153-174

- Section 1 Outline
- What is Evidence-Based Practice?
- Evidence-Based Practice General Introduction,

10-12 - Survey Results (NC school-based PTs view of

EBP),13-18 - Guiding Steps to Practice EBP,19-20
- Two Fundamental Principles of EBP, 21
- Best Research Evidence, 22

- Evidence-Based Practice
- General Introduction
- EBP is the integration of the best research

evidence, clinical expertise, and the patients

values and circumstances - Best Research Evidence valid and clinically

relevant research with a focus on

patient-centered clinical research - Clinical Expertise use of clinical skills and

experiences - Patients Values and Circumstances the

patients unique preferences, concerns, and

expectations in his or her setting - (Straus et al, 2005)

- Evidence-Based Practice
- General Introduction
- EBP is Not
- Focused only on research studies
- Only to be used or understood by professionals

who routinely participate in research studies - A discouragement from trying new treatment
- ? There may be little or no research on a

particular topic, or studies with small sample

sizes may have lacked the power to demonstrate

statistical significance (as later explained in

the statistics section)

Evidence-Based Practice General

Introduction Because RCTs are so difficult, we

will always have areas that lack evidence, we

will need to find other credible research

approaches to supply evidence. Keep in mind that

an absence of evidence is different from negative

evidence. An absence of evidence is not an

excuse to ignore the growing body of data

available to guide practice. (RCT randomized

clinical or controlled trials) (Rothstein

2001)

- Survey Results
- Survey responses from North Carolina School-Based

PTs/PTAs about EBP - Survey distributed at the NC Exceptional

Childrens Conference, Nov 2008 - 41 Respondents (www.med.unc.edu/ahs/physical/sch

oolbasedpt for detailed results) - 73 had participated in a conference or course

on the use of EBP in the last 5 years - Of those that participated in a course,

respondent data suggests the course changed their

view of EBP, but their use and practice - of EBP was changed to a lesser degree.

- Survey Results, Continued
- Highest frequency response as to why we should

use EBP ? positive impact on our clinical

practice - 39 of 41 respondents agreed that EBP is

relevant and necessary for PTs in the school

system (2 left that question blank) - ? Highest frequency responses as to why it is

relevant and necessary A focus on EBP results

in improved clinical practice and provides

validation and justification for our role as

school-based PTs/PTAs

- Survey Results, Continued
- When respondents were asked if they were

comfortable searching for and using EBP - Yes 17 (41),
- No 18 (44),
- No Response 6 (15)
- Internet was the most frequently
- used source for searching EBP,
- but continuing education courses
- ranked highest for preference.

- Survey Results, Continued
- A majority of the respondents were comfortable
- ? Determining the level of evidence and

interpreting the authors conclusions - A majority of the respondents were not

comfortable - ? Interpreting statistics, even though

statistical knowledge is often helpful to

evaluate conclusions drawn by the author

- Survey Results, Continued
- The primary barriers to searching and using EBP,

as listed by NC school PTs/PTAs - Time
- Access
- Uncertain how to search for EBP
- Factors that enhance the search and use of EBP,
- as listed by NC school PTs/PTAs listed
- Additional time
- Education on the appropriate use of EBP
- Education in how to access EBP resources

- Survey Results, Continued
- Questions generated from the survey
- How to efficiently and effectively increase the

knowledge and practice of EBP by NC school

PTs/PTAs? - How to address barriers to using and searching

for EBP? - How to enhance use and access of EBP in

school-based practice?

- Guiding Steps to Practice EBP
- Analyze what we know and what we do not know, in

relation to improving our clinical practice.

Form answerable questions to address any gaps in

our knowledge. - Search for and find the best research evidence to

address our questions. - Critically appraise the information, based on its

validity, impact or size of effect, and

applicability. - (Straus et al, 2005)

- Guiding Steps to Practice EBP
- Integrate information gathered from the best

research evidence with clinical expertise and the

patients values and circumstances - Evaluate the effectiveness of any intervention

taken based on steps 1-4, and the effectiveness

and efficiency of the process - (Straus et al, 2005)

- Two Fundamental Principles of EBP
- Evidence alone is never sufficient to make a

clinical decision (page 8) - Consider risks and benefits, costs,

inconvenience, alternative treatment strategies,

patient preferences/values and circumstances. - EBM posits a hierarchy of evidence to guide

clinical decision making (page 8) - Not all research is equal in terms of relevance

and statistical support, however, that does not

mean lower level evidence is not worthwhile. - (Guyatt and Rennie use the term Evidence-Based
- Medicine, EBM)
- (Guyatt and Rennie, 2002)

- Best Research Evidence
- The three sections that follow will focus on

the best research evidence branch of Straus

three components of EBM. (Straus et al, 2005) - Best research evidence is not more important

than the other two branches it is prominent in

this module because knowledge concerning clinical

expertise and patient values and expectations

will vary from situation to situation. - Section 2 (Statistics Review, Basic Research)

will provide the background information necessary

to perform effective searches and interpret the

best research evidence (Sections 3 and 4).

- Section 2 Outline
- Statistics Review, Basic Research
- Types of Research, 25-33
- Hierarchy of Evidence, 34-35
- Variables, 36-44
- Measurement Validity
- Types, 45-48
- Statistics, 49-67
- Sensitivity and Specificity
- Positive and Negative Predictive Value
- Positive and Negative Likelihood
- Receiver Operating Characteristic
- (ROC) Curve
- Responsiveness to Change
- Effect Size versus MCID

- Section 2 Outline, continued
- Statistics Review, Basic Research, continued
- Measurement Reliability, 68-72
- Descriptive Statistics,73-83
- Frequency and Shape of Distribution
- Central Tendency Measures
- Measures of Variability
- Inferential Statistics, 84-115
- Probability
- Sampling Error
- Confidence Intervals
- Hypothesis Testing
- Power

- Types of Research
- Nonexperimental (Observational)
- Descriptive or Exploratory
- No control or manipulation of variables
- Examines populations and relationships
- Experimental
- Researcher manipulates/controls variable(s)
- Comparison of interventions or groups, examines

cause and effect - (Portney and Watkins, 2000)

Types of Research Portney and Watkins suggest

viewing various designs as a continuum of

research with a descriptive, exploratory, or

experimental purpose. Certain designs may

include elements of more than one classification.

(Portney and Watkins,2000)

- Descriptive Research
- Examples
- Case Study Description of one or more

patients, may document unusual conditions or

response to intervention - Developmental Research Examines patterns of

growth and change, or documents natural history

of a disease or condition - Normative Research Establishes typical values

for specific variables - (Portney and Watkins, 2000)

- Descriptive Research, continued
- Examples
- Qualitative Research Collection of

subjective, narrative information (rather than

quantitative, numerical data) in an effort to

better understand experiences - Evaluation Research Assessment of a program

or policy, often by the collection of descriptive

information through surveys or questionnaires - (Portney and Watkins, 2000)

- Exploratory Research
- Examples
- Correlational Methods Examines relationships

between areas of interest, may be used to predict

or suggest, but cannot offer cause and effect - Cohort and Case-Control Studies Used often in

epidemiological research to describe and predict

risks for certain conditions - Methodological Studies Used to evaluate the

validity and reliability of measuring instruments - Historical Research Use of archives or other

records to reconstruct the past to generate

questions or suggest relationships of historical

interest - (Portney and Watkins, 2000)

- Experimental Research
- Example
- Randomized Clinical or Controlled Trial (RCT)

In general, a clinical treatment, or experimental

condition, is compared to a control condition,

often a placebo but in some cases an alternative

treatment, where subjects are randomly assigned

to a group. - (Portney and Watkins, 2000)

- Experimental Research, continued
- Examples
- Single-Subject Design Variation of RCT, study

of an individual over time with repeated

measurement and determined design phases (Portney

and Watkins, 2000) - In an N1 RCT, a single individual receives

alternating treatment and placebo or alternative

treatment, with the patient and the assessor

blinded to intervention allocation. Objective or

subjective measures are then recorded during the

allocation periods. (Guyatt and Rennie, 2002)

- Experimental Research, continued
- Examples
- Sequential Clinical Trial Variation of RCT,

technique that allows for the continuous analysis

of data as it becomes available, does not require

a fixed sample - Quasi-Experimental Research Comparative

research in which subjects cannot be randomly

assigned to a group, or control groups cannot be

used. Lower level of evidence than RCTs. - (Portney and Watkins 2000)

- Experimental Research, continued
- Examples
- Systematic Review Combination of several

studies with the same or similar variables, in

which the studies are summarized and analyzed

(Guyatt and Rennie, 2002) - Meta-analysis Statistical combination of the

data from several studies with the same or

similar variables, to determine an overall

outcome (Portney and Watkins, 2000 Guyatt and

Rennie, 2002)

- Hierarchy of Evidence for Treatment Decisions
- Greatest (Top) to Least (Bottom)
- N of 1 randomized controlled trial
- Systematic review of randomized trials
- Single randomized trial
- Systematic review of observational studies

addressing patient-important outcomes - Single observational study addressing

patient-important outcomes - Physiological studies (studies of blood

pressure, cardiac output, exercise capacity, bone

density, and so forth) - Unsystematic clinical observations
- A meta-analysis is often considered higher than

a - systematic review

(Guyatt and Rennie, 2002)

- Hierarchy of Evidence
- Ideally, evidence from individual studies would

be compiled or synthesized into systematic

reviews, with that information succinctly

consolidated into easily and quickly read

synopses. All relevant information would be

integrated and linked to a specific patients

circumstance. The medical search literature is

still far from this, but working towards that

goal. Efforts include clinical prediction

guidelines and APTAs emphasis on EBP. - (Straus et al, 2005)

- Variables
- Variables Characteristic that can be manipulated

or observed - Types of Variables
- Independent or Dependent
- Measurement Scales/Levels
- ?Classification is useful for communication, so

that readers are aware of the authors hypothesis

of what situation or intervention (independent

variable) will predict or cause a given outcome

(dependent variable) - (Portney and Watkins, 2000)

- Variables Independent or Dependent
- Independent Variable A variable that is

manipulated or controlled by the researcher,

presumed to cause or determine another

(dependent) variable - Dependent Variable A response variable that

is assumed to depend on or be caused by another

(independent) variable - (Portney and Watkins, 2000)

- Variables Measurement Scales
- Useful to convey information to the reader

about the type of variables observed - Necessary to determine what statistical

analysis approach should be used to examine

relationships between variables - From lowest to highest level of measurement,

the scales are nominal, ordinal, interval, and

ratio - (Portney and Watkins, 2000)

- Variables Measurement Scales
- Nominal Scales (Classification Scale)
- Data, with no quantitative value, are organized

into categories - Categorizes are based on some criterion
- Categories are mutually exclusive and

exhaustive (each piece of data will be assigned

to only one category) - Only permissible mathematical operation is

counting (such as the number of items within each

category) - Examples Gender, Blood Type, Side of

Hemiplegic Involvement - (Portney and Watkins, 2000)

- Variables Measurement Scales
- Ordinal Scales
- Data are organized into categories, which are

rank-ordered on the basis of a defined

characteristic or property - Categories exhibit a greater than-less than

relationship with each other and intervals

between categories may not be consistent and may

not be known - (Portney and Watkins, 2000)

- Variables Measurement Scales
- Ordinal Scales, continued
- If categories are labeled with a numerical

value, the number does not represent a quantity,

but only a relative position within a

distribution (for example, manual muscle test

grades of 0-5) - Not appropriate to use arithmetic operations
- Examples Pain Scales, Reported Sensation,

Military Rank, Amount of Assistance Required

(Independent, Minimal) - (Portney and Watkins, 2000)

- Variables Measurement Scales
- Interval Scales
- Data are organized into categories, which are

rank-ordered with known and equal intervals

between units of measurement - Not related to a true zero
- Data can be added or subtracted, but actual

quantities and ratios cannot be interpreted, due

to lack of a true zero - Examples Intelligence testing scores,

temperature in degrees centigrade or Fahrenheit,

calendar years in AD or BC - (Portney and Watkins, 2000)

- Variables Measurement Scales
- Ratio Scales
- Interval score with an absolute zero point (so

negative numbers are not possible) - All mathematical and statistical operations are

permissible - Examples time, distance, age, weight
- (Portney and Watkins, 2000)

Variables Clinical Example A study

investigates how a strengthening program impacts

a childs ability to independently walk. In this

case, the strengthening program is the

independent variable and the ability to

independently walk is the dependent variable.

Amount of assistance required (if ranked maximal,

moderate, minimal, independently, not based on

weight put on a crutch or other quantitative

testing) would be an example of ordinal

data. Studies often have more than one

independent or dependent variable

- Measurement Validity
- Measurement Validity examines the extent to

which an instrument measures what it is intended

to measure (Portney and Watkins, 2000) - For example, how accurate
- is a test or instrument at
- discriminating, evaluating,
- or predicting certain items?

- Measurement Validity
- Validity of Diagnostic Tests
- Based on the ability for a test to accurately

determine the presence or absence of a condition - Compare the tests results to known results,

such as a gold standard. - For example, a test determining balance

difficulties likely to result in falls could be

compared against the number of falls an

individual actually experiences within a certain

time frame. A clinical test for a torn ACL could

be compared against an MRI. - (Portney and Watkins, 2000)

- Measurement Validity Types
- Face Validity Examines if an instrument

appears to measure what it is supposed to measure

(weakest form of measurement validity) - Content Validity Examines if the items within

an instrument adequately comprise the entire

content of a given domain reported to be measured

by the instrument - Construct Validity Examines if an instrument

can measure an abstract concept - (Portney and Watkins, 2000)

- Measurement Validity Types
- Criterion-related Validity Examines if the

outcomes of the instrument can be used as a

substitute measure for an established gold

standard test. - ? Concurrent Validity Examination of

Criterion-related Validity, when the instrument

being examined and the gold standard are compared

at the same time - ? Predictive Validity Examination of

Criterion-related Validity, when the outcome of

the instrument being examined can be used to

predict a future outcome determined by a gold

standard - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Ways to Evaluate Usefulness of Clinical Screening

or Diagnostic Tools - Sensitivity and Specificity
- Positive and Negative Predictive Value
- Positive and Negative Likelihood Ratios
- Receiver Operating Characteristic (ROC) Curve
- The above mentioned statistical procedures are

often used when researchers are introducing (and

validating) the test. Hopefully the values from

these operations can be found tools testing

manual or in articles evaluating the tools

validity within certain populations or settings.

Measurement Validity Statistics Diagnostic

Reference Table Condition Present

Absent Test Result Positive Negative

(Guyatt and Rennie, 2002 Portney and Watkins,

2000 Straus et al, 2005)

- Measurement Validity Statistics
- Table Labels
- (a) True Positive The subject matter has the

condition, and the test accurately identifies the

presence of the condition - (d) True Negative The subject matter does not

have the condition, and the test accurately

identifies the absence of the condition - (b) False Positive The subject matter does not

have the condition, and the test incorrectly

identifies the presence of the condition - (c) False Negative The subject matter has the

condition, and the test incorrectly identifies

the - absence of the condition
- (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Positive test result the test identifies the

condition as being present - Negative result the test identifies the

condition as being absent - (This may or may not be accurate when compared to

the gold standard). - The tests sensitivity and specificity provide

information about the accuracy of the test. - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Sensitivity
- The ability of the test to obtain a positive

test when the condition is present the ability

to detect a true positive (a) - a/(a c) The proportion that test positive out

of those with the condition - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Specificity
- The ability of the test to obtain a negative

result when the condition is absent, the ability

to detect a true negative (d) - d/(b d) The proportion that test negative out

of those without the condition - Sensitivity and Specificity are often provided in

terms of percents, from 0 to 100 (low to high) - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- ?Helpful Hints to remember
- Sensitivity and Specificity?
- Sensitivity SnNout When a test has a high

sensitivity (Sn), a negative result (N), rules

out (out) the diagnosis. - Specificity SpPin When a test has a high

specificity (Sp), a positive result (P), rules in

(in) the diagnosis - (Straus et al, 2005)

Measurement Validity Statistics Clinical

Example Example Youre choosing between two

tests that screen participation in school based

on physical abilities. A positive result means

that the students physical abilities impact his

or her participation.

Measurement Validity Statistics Clinical

Example One test has a high sensitivity, but a

low specificity. A high sensitivity means that a

negative test will effectively rule out students

whose physical abilities do not impact

participation. However, with a low

specificity, there may be many false positives,

meaning students may test positive for the

condition when, in fact, their abilities do not

impact participation.

Measurement Validity Statistics Clinical

Example Example Youre choosing between two

tests that evaluate participation in school based

on physical abilities. A positive result

means that the students physical abilities

impact his or her participation.

Measurement Validity Statistics Clinical

Example The other test has a low sensitivity,

but a high specificity. A high specificity means

that a positive result will effectively rule in

the condition. However, with a low

sensitivity, there may be many false negatives,

meaning that students may test negative for the

condition, when, in fact, their abilities do

impact their participation.

- Measurement Validity Statistics
- Predictive Values
- Provided in terms of percentages, 0 to 100,

low to high - Positive Predictive Value (PV)
- Probability that a person with a positive test

actually has the condition - a/(a b)
- High PV desired for screening tools, to

prevent excessive unnecessary future testing - Negative Predictive Value (PV-)
- Probability that a person with a negative
- test does not have the condition
- d/(c d)
- (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Likelihood Ratios
- Calculated from the Diagnostic Reference Table
- Requires prior calculation of the pretest

probability of the condition in question - Easy to use when familiar with the concept, but

requires the use of a probability guide chart or

a nomogram (chart that contains pretest

probability and likelihood ratios, with a ruler

connecting those two points to determine posttest

probabilities) - (Guyatt and Rennie, 2002)

- Measurement Validity Statistics
- Likelihood Ratios, continued
- Positive and negative likelihood ratios are

calculated - Determines the usefulness of a diagnostic test.

If a positive or negative result will change the

posttest probability of having a condition to

alter the clinician and patients course of

action, it will be useful. If the likelihood

ratios of the test do not result in a substantial

change of knowledge, the test most likely will

not be useful. - (Guyatt and Rennie, 2002)

- Measurement Validity Statistics
- Receiver Operating Characteristic (ROC) Curve
- Uses sensitivity and specificity information to

find the probability of correctly choosing

between presence or absence of the condition - For example, a test with an area under the ROC

curve of 0.80, would result in the correct

determination of presence or absence of a

condition 80 of the time. - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Responsiveness to chance statistics evaluate a

measurement tools ability to detect change over

time - For example, will administration of the test

pre and post intervention reflect a change in

status, if one actually occurred? - Evaluated by examining the change in scores in

a pretest-posttest design, or using effect size - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Effect Size
- Effect size (ES) is a measure of the amount of

difference. - For example, experimental group A increased their

score on a coordination measure by an average of

15 points, while experimental group B increased

their score an average of 8 points. The ES

between groups would be 7 points, considering the

groups were homogeneous at the start. - (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Effect Size, continued
- An effect size index is a converted effect

size, a standardized measure of change, so that

change scores across different outcome measures

can be compared. - ES is often displayed as a correlation

coefficient, r - Portney and Watkins note that considerations of

ES vary based on the interpreting clinician, but

review Cohens suggestions of scores lt0.4 as

small (treatment had a small effect), 0.5 as

moderate, - and gt0.8 as large
- (Portney and Watkins, 2000)

- Measurement Validity Statistics
- Effect Size versus Minimal Clinically Important

Difference - In addition to numerical data revealed and

statistical significance, the clinician should

also consider what amount of change is clinically

meaningful, such as, how great a gain in strength

or endurance will result in a change in function?

This is often referred to as the minimal

clinically important difference (MCID). - (Jaeschke et al 1989)

- Measurement Reliability Statistics
- Reliability examines a measurements consistency

and freedom from error - Can be thought of as reproducibility or

dependability - Estimate of how observed scores vary from the

actual scores - (Portney and Watkins, 2000)

- Measurement Reliability Statistics
- Reliability Coefficient
- Ratio of reliability (many different types with

various symbol representation) - Range between 0.00 to 1.00
- 0.00 no reliability
- 1.00 perfect reliability
- Reflection of variance, a measure of the

differences among scores within a sample - (Portney and Watkins, 2000)

- Measurement Reliability Statistics
- Correlation
- Comparison of the degree of association between

two variables or sets of data - Used as a basis for many reliability

coefficients - (Portney and Watkins, 2000)

- Measurement Reliability Statistics
- Test-Retest Reliability
- Examines the consistency of the results of

repeated test administration - Traditional Analysis
- Pearson product-moment coefficient of

correlation (for interval or ratio data) - Spearman rho (for ordinal data)
- Current, sometimes considered preferred,

Analysis - Intraclass correlation coefficient
- (Portney and Watkins, 2000)

- Measurement Reliability Statistics
- Rater Reliability
- Intrarater reliability
- Reliability of data collection from one

individual over two or more trials - Interrater reliability
- Reliability of data collection between two or

more raters - (Portney and Watkins, 2000)

- Descriptive Statistics
- Descriptive statistics are used to describe

sample characteristics. - A sample is a subset of a population chosen for

study. Since data often cannot be collected from

an entire population, the data chosen from a

selected sample is intended to be representative

or an estimate of the population data. - Distribution Total set of scores (from a

sample) for a particular variable, given the

symbol, n - (Portney and Watkins, 2000)

- Descriptive Statistics
- Frequency and Shapes of Distribution
- Frequency distribution The number of times

each value, from the variable data, occurred - Drawing a graph of frequency distributions can

result in shapes that characterize the

distributions - Some graphs are asymmetrical, others are

symmetrical - A symmetrical graph with a bell-shaped

distribution is referred to as a normal

distribution. - A skewed distribution presents asymmetrically
- (Portney and Watkins, 2000)

Descriptive Statistics Normal Distributions,

when graphed according to frequency, present in

the shape of a bell with the majority of scores

falling in the middle and progressively fewer

scores at either end. It has special properties

in statistics. (Portney and Watkins,

2000)

- Descriptive Statistics
- Central Tendency Measures
- Used to quantitatively summarize a groups

characteristics. - Mode The score that occurs most frequently
- Median The middle score in numerically ordered

group of data. If there are an even number of

scores, the median is the midpoint between the

two middle scores - Mean The sum of a set of scores divided by the

number of scores, n. Often referred to as the

average of a data set. - (Portney and Watkins, 2000)

- Descriptive Statistics
- Measures of Variability
- Variability is the dispersion of scores.
- Variability is affected (qualified) by five

characteristics - Range
- Percentiles
- Variance
- Standard deviation
- Coefficient of variation
- (Portney and Watkins, 2000)

Descriptive Statistics Measures of

Variability Variability, continued Range

Difference between the highest and lowest scores

in a distribution Percentiles Used to describe

a scores position within a distribution,

distribution data is often divided into

quartiles, or four equal parts Variance (Too

in-depth to describe the statistical background

for this module purpose). Reflects variation

within a set of scores, in square units.

Symbolized in sample data by s (Portney and

Watkins, 2000)

- Descriptive Statistics
- Measures of Variability
- Variability, continued
- Standard Deviation Representative of the

variability of scores surrounding the mean, in

original units of measurement. Square root of

variance. - The larger the standard deviation, the more

spread out the variables scores are around a

mean. - For example, data set (A) 8,9,10,11,12 and the

data set (B) 4,5,10,15,16 both have a mean of 10.

However the standard deviation for set A is

1.58 while the standard deviation of set B is

5.52. - Symbolized in sample data by s2.
- (Portney and Watkins, 2000)

Descriptive Statistics Measures of

Variability Variability, continued Coefficient

of Variation The ratio of the standard deviation

to the mean. (Portney and Watkins,

2000)

- Descriptive Statistics
- Distributions
- Normal Versus Skewed (Non-normal) Distribution
- These theoretical shapes of distribution help

determine what statistical formulas or measures

should be used - The characteristics of normally distributed

data are constant and predictable. For

statistical purposes, the normally distributed

curve is often divided into proportional areas,

each equal to one standard deviation. - Data should be examined for goodness-of-fit

to see if the sample approximates the normal

distribution. - (Portney and Watkins, 2000)

- Descriptive Statistics
- Distributions
- Normal Distribution Statistics
- 1st standard deviation on either side of the

average contains 34.13 of the data - total of 68.62 of the data will be between 1

and -1 standard deviation - Between 1st and 2nd standard deviation contains

13.59 of the data - total of 95.45 of the data will be between 2

and -2 standard deviations - (13.59 times 2) (34.13 times 2)
- (Portney and Watkins, 2000)

- Descriptive Statistics
- Distributions
- Normal Distribution Statistics, continued
- Between 2nd and 3rd standard deviation contains

2.14 of the data - total of 99.73 of the data will be between 3

and -3 standard deviations - (13.59 times 2) (34.13 times 2) (2.14 times

2) - (Portney and Watkins, 2000)

- Inferential Statistics
- Estimate population characteristics from sample

data - Used often when testing theories about the

effects of experimental treatments - Requires that assumptions are made about how

well the sample represents the larger population - Assumptions are based on the statistical

concepts of probability and sampling error - It is important the sample be representative of

the population, so that the results of

interventions on samples can be applied to the

entire population of individuals with those

characteristics. - (Portney and Watkins, 2000)

- Inferential Statistics
- Probability
- Probability
- The likelihood that an event will occur, given

all the possible outcomes. Used often in

prediction. Given the symbol p - Probability may range from p 1 (certain the

event will occur, 100 probability) to p 0

(certain that the event will not occur, 0

probability) - (Portney and Watkins, 2000)

Inferential Statistics Probability Probability,

continued Reflective of should happen in the

long run, not necessarily what will happen on a

given trial. For example, if a treatment has a

60 chance of success, then 60 of people will

likely be successfully treated. That does not

mean the treatment will be 60 successful in an

individual, likely it will either be a

unsuccessful or successful for that

individual. (Portney and Watkins,

2000)

- Inferential Statistics Probability
- Clinical Example
- Probability statistics can be applied to the

distribution of scores - Example, for a normally distributed data set
- Average long jump for a certain group of

children is 35 inches with a standard deviation

of 4 inches. Suppose you want to know the

probability that a child will jump within one

standard deviation (from 31 to 39 inches)? You

know that within one standard deviation on either

side of the mean is 68.2, so that is the

probability that a child will jump within one

standard deviation of the mean.

- Inferential Statistics Probability
- Clinical Example
- Example, continued
- If you wanted to know the probability that a

child will jump more than one standard deviation

above the mean (greater than 39 inches), you can

refer to the data and calculate 15.86. - Charts and graphs are available to calculate the

data in between the standard deviations.

- Inferential Statistics
- Sampling Error
- Sampling Error
- The difference between sample values and

population values - The lower the sampling error, the greater the

confidence that the sample values are similar to

the population values. - To estimate sampling error, the standard error

of the mean statistic is used (too complex to

explain statistical basis in this format) - The larger the sample size, n, the smaller the

standard error of the mean - (Portney and Watkins, 2000)

- Inferential Statistics
- Confidence Intervals
- Confidence Interval (CI)
- Range of scores with specific boundaries, or

confidence limits, that should contain the

population mean - The boundaries of CIs are based on the sample

mean and its standard error - Degree of confidence is expressed as a

percentage - Often, researchers use 95 as a boundary, which

is just slightly less than 2 standard deviations

on either side of the mean - (Portney and Watkins, 2000)

Inferential Statistics, Confidence Interval

Clinical Example A physical therapy treatment

program resulted in the ability for 40 children

with a certain disorder to walk an additional 8

independent steps, on average, within a certain

set of parameters. The 95 CI for this data

was 2 steps. Therefore, we can be 95 certain

that the population mean, or average for all

children with this disorder, is between 6 and 10

extra independent steps. Said another way, if

there were an additional 100 children with the

same condition, 95 of them would likely have an

average that was between an additional 6 to 10

additional independent steps following the

physical therapy treatment.

Inferential Statistics Hypothesis

Testing Hypothesis Testing Used to decide if

effects from an intervention are due to chance or

the result of the intervention Results in one

of two decisions To accept or reject the null

hypothesis (Portney and Watkins,

2000)

Inferential Statistics Hypothesis

Testing Hypothesis Testing, continued

Statistical Hypothesis (also known as the null

hypothesis) Any observed difference (as in

pretreatment to post-treatment or treatment

compared to a placebo) is due to chance. When

the null hypothesis is rejected, the researcher

concludes that the effect of treatment is not

likely due to chance. (Portney and Watkins,

2000)

Inferential Statistics Hypothesis

Testing Hypothesis Testing, continued Alternati

ve Hypothesis Any observed difference is not due

to chance. Often the researcher is trying to

support the alternative hypothesis, as when

trying to prove that one particular treatment is

better. Sometimes, however, the researcher may

be trying to prove that certain interventions are

equal. (Portney and Watkins 2000)

- Inferential Statistics
- Hypothesis Testing, Errors
- Errors in Hypothesis Testing
- Decision to accept or reject the null

hypothesis is based on the results of the

statistical procedures on collected data from

samples. - Decisions are based on sample data, so it is

possible that the results obtained are not

accurate of population data. - There is a chance for error, that the

researcher may incorrectly accept or reject the

null hypothesis - (Portney and Watkins 2000)

- Inferential Statistics
- Hypothesis Testing, Errors
- Type I error (a) Rejecting the null hypothesis

when it is true (for example, deciding that the

difference seen between a treatment group and a

control group is due to the effect of the

treatment, when, in fact, the difference is due

to chance). - A commonly used standard is a 0.05, or the

researchers accept a 5 chance of making a Type I

error - Statistical tests completed with the sample data

are used to calculate p, the probability that an

observed difference did occur by chance. - (Portney and Watkins, 2000)

Inferential Statistics Hypothesis Testing, p and

a Hypothesis Testing Relationship between p

and a If p is greater than the chosen a, then

the researchers chose not to reject the null

hypothesis For example, in a placebo versus

treatment study, the researchers cannot conclude

that the experimental treatment had a different

effect then the placebo. (Portney and

Watkins, 2000)

Inferential Statistics Hypothesis Testing, p and

a Hypothesis Testing Relationship between p

and a, continued If p is less than the chosen

a, then the researchers chose to reject the null

hypothesis For example, in a placebo versus

treatment study, the researchers conclude that

the experimental treatment had a different effect

then the placebo. (Portney and Watkins,

2000)

Inferential Statistics Hypothesis Testing, p and

a Hypothesis Testing Relationship between p

and a, continued Confidence intervals

surrounding the p value can be calculated,

hopefully these are included in data analysis

section of the research study. The CIs between

two groups should not (rare exceptions) overlap,

if a statistically significant difference is

found. (Portney and

Watkins, 2000)

Inferential Statistics Hypothesis, CIs versus

MCID Hypothesis Testing CIs (slide 90) versus

MCID (slide 67) When the null hypothesis is

rejected, the researchers conclude that the

experimental treatment (versus a placebo, for

example) had a statistically significant effect.

The clinician should examine the effect size

and the MCID to ensure that the change is

clinically and functionally relevant. (Guyatt

and Rennie, 2002)

- Inferential Statistics
- Hypothesis, CIs versus MCID
- When a null hypothesis is not rejected, the

researchers may conclude that the experimental

treatment did not have an effect. - However, the researchers should pay close

attention to the confidence intervals. - If the CI does not include the MCID, then the

trial is most likely negative. - If the CI includes the MCID, then the

possibility that the experimental treatment may

have a positive effect cannot be ruled out. The

researchers may wish to run a power analysis,

explained in later slides. - (Guyatt and Rennie, 2002)

- Inferential Statistics, Hypothesis Testing
- Clinical Example
- Children with similar abilities/diagnosis

(homogenous sample) are randomly assigned to two

different groups - Group A receives a physical therapy designed to

improve gross motor skills, and - Group B completes typical daily activities (but

dont worry, due to ethical concerns children in

Group B will receive the same treatment as those

in Group A at the end of the study period).

Inferential Statistics, Hypothesis

Testing Clinical Example 1, continued The

outcome measure will be a tool that tests gross

motor activities with a final, numerical outcome

score. The authors hypothesis that the

experimental group will show statistically

significant gains (experimental hypothesis) and

that the null hypothesis (there is no difference

between groups at the end of the intervention

period) will be rejected.

Inferential Statistics, Hypothesis

Testing Clinical Example 1, continued Initially

, group A and B have similar average pre-test

scores (if not, that can be statistically

corrected). Now suppose that Group A

(experimental group, receiving additional PT

services) increases an average of 9 points from

pretest to posttest, with a CI of 1, 8 to 10.

Group B increases an average of 0.5 points,

with a CI of 2, -1.5 to 3.5.

Inferential Statistics, Hypothesis

Testing Clinical Example 1, continued These

confidence intervals do not overlap, which

corresponds with the statistical analysis

performed that plt0.05 (the predetermined a), and

there is a statistically significant difference

between groups. The null hypothesis is

rejected.

Inferential Statistics, Hypothesis

Testing Clinical Example 1, continued Prior to

the experiment, the authors agreed that the MCID

was 6. Both the statistically determined

(plt0.05 ) and clinically determined criteria in

support of the experimental groups treatment

were met (ES of experimental group,9, was greater

than the MCID, 6).

Inferential Statistics, Hypothesis Testing

Clinical Example 2 Initially, group A and B

have similar average pre-test scores (if not,

that can be statistically corrected). Now,

change the scenario from previous slides.

Suppose that Group A (experimental group,

receiving additional PT services) increases an

average of 5 points from pretest to posttest,

with a CI of 2, 3 to 7. Group B increases an

average of 0.5 point,s CI 2, -1.5 to 3.5.

Inferential Statistics, Hypothesis Testing

Clinical Example 2, continued These

confidence intervals overlap, and the statistical

analysis performed revealed that pgt0.05 (the

predetermined a). There is not a statistically

significant difference between groups. The

null hypothesis is not rejected.

Inferential Statistics, Hypothesis Testing

Clinical Example 2, continued However, prior

to the experiment, the authors agreed that the

MCID was 6. Since the CI range of Group A

includes that MCID, the beneficial effects of the

treatment from Group A over Group B cannot be

ruled out. The researchers may want to run a

statistical power analysis, to determine if the

study was underpowered, and therefore unlikely to

show a difference of the treatment.

Inferential Statistics Hypothesis Testing, Errors

Errors in Hypothesis Testing, Type II Type II

error (ß) Not rejecting the null hypothesis

when it is false (for example, determining that

differences are due to chance, when they are, in

fact, due to the experimental treatment) (P

ortney and Watkins, 2000)

- Inferential Statistics
- Hypothesis Testing, Errors
- Errors in Hypothesis Testing, Type II
- The complement of Type II error is the

statistical power of the test (1-ß) - Power the probability that a test will lead to

rejection of the null hypothesis, or the

probability of obtaining statistical significance

if the differences are not due to chance - Many researchers use a standard of ß0.20, or a

power of 80, as reasonable protection against

Type II error - (Portney and Watkins,2000)

Inferential Statistics Power Determinants of

Statistical Power Even though the researchers

may not reject the null hypothesis, it does not

always mean that an experimental treatment is not

effective. The power of the study may have

been too low or small to detect a significant

difference. (Guyatt and Rennie 2002,

Portney and Watkins, 2000)

- Inferential Statistics Power
- Determinants of Statistical Power
- Levels set of a and ß
- Variance
- Sample Size (n)
- Effect Size (Difference between two treatments

or variables, Treatments with large changes or

correlations are more likely to produce

significant outcomes) - Increases of effect size, sample size, and alpha

all increase power, while decreases in variance

increases power. - (Portney and Watkins, 2000)

- Inferential Statistics Tests
- Parametric Vs. Nonparametric Tests
- Parametric statistics are statistics used to

estimate population parameters - Primary assumptions of parametric tests
- Samples randomly drawn from populations with

normal distributions - Variances in samples are roughly equal
- Data are measured on interval or ratio scales
- (Portney and Watkins, 2000)

Inferential Statistics Tests Parametric Vs.

Nonparametric Tests Nonparametric tests are

generally not as powerful, and researchers may

choose to use parametric tests, despite not

meeting all generally held assumptions (such as

use of parametric test in a study with ordinal

data) (Portney and Watkins,

2000)

- Section 3 Outline
- How to Search for EBP
- Formation of Clinical Questions used to search

for EBP, 117-121 - Evidence Search Sources of Information, 122-138
- Internet/World Wide Web
- Textbooks
- Specific Journal Subscriptions
- Internet Sources for Medical Information
- The Guide
- Search Strategies, 139-152

- Formation of Clinical Questions
- Used to Search for EBP
- Background Questions general knowledge about a

condition/area of interest - Foreground Questions specific knowledge used to

inform clinical decisions or actions. Often what

researches use when investigating a particular

tr