Childhood Language Disorders Midterm

Front 1

Evidence Based Practice EBP

Back 1

Current,
high-quality research evidence is integrated with practitioner expertise and client
preferences and values into the process of making clinical decisions

Front 2

EBP includes

Back 2

Recognize and integrate the needs of those served -- along with best current research evidence and their clinical expertise in making clinical decisions;
• acquire and maintain the knowledge and skills necessary to provide high quality professional services,
• evaluate prevention, screening, and diagnostic procedures, protocols, and measures to identify maximally informative and cost-effective diagnostic and screening tools, using recognized appraisal criteria described in the evidence-based practice literature;
• evaluate the efficacy, effectiveness, and efficiency of clinical protocols for prevention, treatment, and enhancement using criteria recognized in the evidence-based practice literature;
• evaluate the quality of evidence appearing in any source or format, including journal articles, textbooks, continuing education offerings, newsletters, advertising, and Web-based products, prior to incorporating such evidence into clinical decision making; and
• monitor and incorporate new and high quality research evidence having implications for clinical practice

Front 3

Steps to EBP

Back 3

Form clinical question
Use internal evidence
Find best current external evidence
Critical appraisal
Decide whether evidence is strong enough
Integrate external evidence w/ "intangibles"
Update

Front 4

Internal Evidence

Back 4

Intangibles
Like, your clinical experience

Front 5

External evidence

Back 5

Explicit criteria to judge evidence quality
Relevance (to clinical question)
Level of evidence based on design and quality
Direction, strength, and consistency of the observed outcomes

Front 6

Levels of External Evidence
Ia

Back 6

Systematic meta-analysis
(well-designed, randomized controlled study)

Front 7

Levels of External Evidence
Ib

Back 7

Well-conducted single randomized controlled trial w/narrow confidence interval

Front 8

Levels of External Evidence
IIa

Back 8

Systematic review of non-randomized quasi-experimental trial or single subject that documents consistent study outcomes

Front 9

Levels of External Evidence
IIIb

Back 9

High quality quasi-experimental trial or a lower quality RCT or single subject experiment with consistent outcomes, across replications

Front 10

Levels of External Evidence
Bottom 3

Back 10

III - Observational studies w/controls (retro, interrupted, case-control, cohort w/controls)

IV - Observational studies w/o controls

V Expert opinions

Front 11

Experimental design

Back 11

patients enrolled prospectively, randomized to group/condition, and some variable is controlled

Front 12

Quasi-experimental design

Back 12

prospective but not randomized or controlled

Front 13

Non-experimental design

Back 13

patients are identified retrospectively, no manipulation of variables

Front 14

Randomized controlled trial (RCT)

Back 14

Designed BEFORE patients enrolled
Patients are assigned at RANDOM to one group to be compared

Front 15

RCT - pros

Back 15

If randomization is accomplished correctly, best odds that groups do not differ systematically on some variable (other than one of interest)

Front 16

RCT - cons

Back 16

Costly, time-consuming, in some cases may raise ethical issues, potential for volunteer bias

Front 17

Quasi-experimental: prospective w/o randomization

Back 17

E.g., Cohort study: patients w/ and w/o variable are id'd and followed forward in time to compare their outcomes
PROS - ability to study low-incidence and/or late-emerging problems efficiently
CONS - exposure may be linked to a hidden confounder; blinding is difficult

Front 18

Non-experimental studies 1 of 2

Back 18

Cross-sectional/correlation: group is observed at single point; exposure and outcome are determined simultaneously
PROS - cheap, ethically safe
CONS - causality cannot be established, susceptible to recall bias, confounders may not be equally distributed

Front 19

Non-experimental studies 2 of 2

Back 19

Case-control

Front 20

Weakest non-experimental designs

Back 20

Case series: no control group

Case report: N = 1

Front 21

Purposes of Assessment

Back 21

Identify skills person has, and does not have in particular area of communication

Guide intervention design

Monitor communicative growth and performance over time

To qualify a person for special services

Front 22

Screening - why?

Back 22

Answer yes/no question

To identify children at risk

Front 23

Language Assessment - what to assess

Back 23

Receptive: comprehension vocabulary, morphosyntax

Production: expressive vocabulary and morphosyntax

Domains of language

Front 24

Collateral areas to assess

Back 24

Hearing
Speech-motor assessment
Non-verbal cognition
Social functioning

Front 25

Validity

Back 25

Extent to which a test MEASURES WHAT IT purports to measure

Front 26

Construct validity

Back 26

Measures theoretical construct designed to measure

Front 27

Face validity

Back 27

Common-sense match between intended purpose and actual content

Front 28

Content validity

Back 28

Items should be representative of content domain sampled

Evaluated by experts

Front 29

Criterion-related validity

Back 29

Concurrent: test agrees with other valid instruments in categorizing children as normal or not

Predictive: how well child will perform on other valid test

Front 30

Reliability

Back 30

Consistency of measurement in regard to measuring particular skill, behavior

Good reliability: correlation r - .90

Front 31

Test-retest reliability

Back 31

Test given at 2 different times, same/stable scoring

Front 32

Inter-Rater reliability

Back 32

Given by 2 different examiners, stable scoring

Front 33

Internal consistency reliability

Back 33

Subtests of the test rank subject similarly
Parts of the test are measuring something similar to what the whole test is measuring

Front 34

Split-half reliability

Back 34

Consistency among halves of the test

Front 35

Equivalent forms reliability

Back 35

If test has 2 fours, there is consistency among forms

Front 36

Diagnostic accuracy

Back 36

How accurately the test classifies children

Goals - Good accuracy 90% or better

Front 37

Sensitivity

Back 37

Proportion of child WITH DISORDER accurately id'd by test

Front 38

Specificity

Back 38

Proportion of child WITHOUT DISORDER accurately id'd by test

Front 39

Test bias - "Technical" bias

Back 39

Equating performance with ability (may lead to misclassification and underestimation)

Predictive validity is base don correlations between group means, not individual scores

Front 40

Test bias: def'n

Back 40

Invalidity or systematic error in how a test measures for members of a PARTICULAR GROUP

e.g., slow stopwatch used for one group

Front 41

Adverse impact of biased and non-biased tests

Back 41

2ndary effect of biased or non-biased tests

Group differences in test performance that result in disproportionality in selection

Front 42

Criterion referenced assessment

Back 42

Used to determine an individual's level of achievement or skill in a particular area of communication

Front 43

Performance based assessment

Back 43

Describes an individual's skills or behaviors with authentic contexts of use (like home, workplace)

Rationale: Communicative abilities are highly influenced by context - thus, traditional models are of limited use in PLANNING treatments

Front 44

Dynamic assessment

Back 44

Analyzes how much and what types of support or assistance needed to bring the individual's communicative performance to higher level