Preliminary Study of Treatment Effectiveness
• Purpose: To assess the effectiveness of Northwestern University’s Adult Stuttering Treatment Group (ASG)
– A “whole-disorder” treatment program in use since 1970, and trained internationally – One of the frequently “recommended,” but seldom researched treatment approaches
• Subject Pool: All clients enrolled in the ASG since 1975
Treatment Effectiveness
• Efficacy: The extent to which treatment can be shown to be beneficial under optimal (or ideal) conditions • Effectiveness: The extent to which treatment is shown to be beneficial under typical (or realworld) conditions
• Sources: Agency for Health Care Policy and
Research (AHCPR, 1994); Congressional Office of Technology Assessment (1978)
Goals of Treatment
• Clients can achieve fluency when they want to (using modification techniques) • Clients will experience increased level of unmodified fluency (as modifications become more automatic) • Clients accept remaining stuttering (without anxiety, fear, struggle, avoidance, etc.)
– As with other disorders that Patrick reviewed, “recovery” allows some residual stuttering
Schedule of Treatment
• Group and individual sessions with structured generalization tasks • Extensive treatment model
– 2 to 3 times per week for 2 academic quarters (18 weeks total)
– On-going monthly maintenance and problem-solving in the “Continuation Group” following dismissal from ASG
Principles of Treatment
• Combines elements of both “speak more fluently” and “stutter more fluently” approaches to treatment with extensive counseling • Gives client a “toolbox” of several modification techniques they can call upon to increase fluency and decrease sensitivity as necessary
Modification Techniques
• “Speak more fluently” methods
– ERA-SM (Easy Relaxed Approach—Smooth Movement) – Delayed response (pausing before utterances) – Phrasing (pausing within utterances)
• “Stutter more fluently” methods
– – – – – Relaxation Negative practice of tension and tension reduction Voluntary Disfluency/Voluntary Stuttering Cancellation Pull-out
Evaluating the Clinical Records
• Data extracted from clinical records of clients who had enrolled in ASG
– Observable characteristics of stuttering – Use of modification techniques – Situational factors affecting fluency – Cognitive / affective aspects of clients’ recovery (attitudes, feelings, etc.)
• Data collected at diagnostic, before treatment, during treatment, and at dismissal
Observable Characteristics
• Assessed via Systematic Disfluency
Analysis (SDA, Campbell & Hill, 1987, 1994)
– Examines a variety of more typical and less typical disfluency types in language context – Measures frequency, type, duration, number of iterations, and clustering, plus qualitative features (tension, pitch changes, rhythm...) – Five different in-clinic speaking tasks
• Monologue, dialogue, reading, pressure, phone
• Follow-up questionnaire sent to all clients assessing:
– Self-reported level of fluency – Use of modification techniques – Speech attitudes / comfort with speaking – Avoidance of sounds, words, situations – Occurrence of and reaction to relapse
Follow-up Questionnaire
• Asked about client’s success before treatment, immediately after treatment, and at present
Caveats
• Concerns re retrospective studies
– Reliability of measurement – Accuracy of clinical files – Use of currently relevant measures • If such issues are addressed, and results are interpreted appropriately, such studies can provide a meaningful adjunct to other studies of treatment effectiveness
Measurement Reliability
• Reliability data for the SDA have not yet been published, however:
– Students participate in detailed training re identification disfluencies and use of SDA
(e.g., Campbell, Hill, Yaruss, & Gregory, 1996 ).
– Each SDA was reviewed by one of the authors of the SDA technique (Campbell & Hill) – Two preliminary analyses reveal good agreement on counts (Yaruss, in press; Yaruss et al., submitted)
• Pearson Correlations: r .90 (p < .001) • Mean Differences: 0.11% (SD 1.5%)
Accuracy of Clinical Files
• Clinical files are notorious for their inaccuracy (particularly student files) • However, the NU clinic has a rigorous review policy for all clinical reports
– Reports are reviewed by the original supervisor and by a second supervisor who “approves” all reports before they are included in the clinical files
Preliminary Results: 4 Findings
• Changes in client’s speech fluency
– Average Data – Example of Individual Data
• Use of modification techniques • Cognitive and affective changes • Self-reported long-term changes
Finding 1a: Observable characteristics — Group Data (N = 15)
12 10 8 6 4 2 0
Less Typical Disfluencies (t = 5.34; p < .001) More Typical Disfluencies (t = 3.42; p < .004)
Frequency of Disfluencies
Pre-Treatment
Post-Treatment
Finding 1b: Observable characteristics — Individual Data (Subject #1)
Frequency of Disfluencies
8 7 6 5 4 3 2 1 0 Less Typical Disfluencies More Typical Disfluencies Diagnostic PreEvaluation Treatment No Treatment Treatment No Treatment: Treatment: 5 weeks 9 weeks 9 weeks
Before Break
After Break
PostTreatment
Finding 2: Use of modifications at end of treatment (N = 13)
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%
% of Subjects .
ERASM
Del Resp
Vol Disf
Canc
Pullout
Relax
Finding 3: Cognitive /affective changes at the end of treatment
• 67% of clinical records reported that clients achieved some improvement in cognitive / affective aspects
– reduced fear and anxiety leading to increased ability to enter speaking situations – improved attitudes, acceptance leading to increased self-esteem and self-confidence
• But, no specific measures were utilized!
– Judgments based only on clinician’s “feelings”
Finding #4a: Self-rated Level of Fluency at Follow-up (N = 15)
Good Poor
5 4 3 2 1
Before Treatment
Immediately After Treatment
Currently
Finding #4b: Self-rated Speech Attitudes at Follow-up (N = 15)
Good Poor
5 4 3 2 1
Before Treatment
Immediately After Treatment
Currently
Finding #4c: Self-rated Avoidance at Follow-up (N = 15)
Always Never
5 4 3 2 1
Before Treatment
Immediately After Treatment
Currently
Finding #4d: Use of Modification Techniques at Follow-up (N = 15)
Always Sometimes Never
5 4 3 2 1
ERA- Relax Vol. Neg. Del. Canc. PullSM Disf. Pract. Resp. out
Implications
• All clients reported some benefits presumably associated with treatment
– Increased speech fluency (impairment) – Increased ability to approach situations and function at home and work (disability) – Increased participation in society (handicap)
• Many clients reported improvements, even
though they did NOT continue to consistently use the modification techniques
Future Research
• Based on these retrospective results we can begin planning prospective studies:
– Descriptive and experimental group designs to:
• Apply more rigorous assessment of measures throughout the entire treatment process • Gain understanding of time required to establish modifications (to support development of SS study)
– Single-subject designs, e.g.,
• Multiple baseline across subjects to establish internal reliability for assessing treatment effects • Crossover design and component analyses to directly evaluate different aspects of treatment
Conclusions
• Rather than determining that “whole-disorder” treatments should not be used because they have not yet been researched, it seems reasonable to begin to study them in a scientific fashion
– If they prove to be worthless after such study, then by all means, they should not be used – If they prove to be efficacious (whatever that means), then they can be another acceptable means of treatment
• Retrospective studies of treatment effectiveness can help pave the way by:
– providing preliminary assessment of presumed benefits – operationalizating treatment variables Review this lecture