A Critical Review of Protocols for the Diagnostic Assessment of Dysfluency - Coursework Example

Introduction It is generally believed that it is beneficial to treat stuttering in the early stages. In order to help the speech-language pathologist differentiate between early stuttering and normal dysfluency, numerous diagnostic protocols have been developed. While the older diagnostic approaches (primary, secondary, phase III) focused on the full range of degrees of stuttering and not just on beginning stuttering, the newer protocols have one important difference in that they try to identify the child who is just beginning to stutter, or focuses on identifying a chronically stuttering child or someone who might stutter later (Gordon & Luper, 1992). The rationale for the use of a differential diagnostic protocol is based on the fact that there is a high rate of recovery among those who were stutterers at one point of time; about 40-80% of stutterers can have a spontaneous recovery (Bloodstein, 1981). This essay is a critical review of protocols for the diagnostic assessment of dysfluency. Review of literature The checklist by Van Riper, 1971 & 1982 basically includes 26 behavioral characteristics (which require direct observation of the child’s speech) that are divided into seven general categories. The Protocol for Differentiating the Incipient Stutterer provides a systematic method for making specific observations about speech fluency (Pindzola & Dorenda, 1986.) The clinician tries to identify and categorize eight types of auditory behaviors (type of disfluency, size of speech unit affected, frequencies of disfluencies, duration of disfluencies, audible effort, rhythm/tempo/speed of disfluencies, intrusion of schwa vowel during repetitions, and audible learned behaviors) and records it on the Protocol (Pindzola & Dorenda, 1986). With the help of a grid, the clinician can then automatically determine whether the behavior is normal, questionable, or indicative of stuttering (Pindzola & Dorenda, 1986). Although failure in any particular category is not necessarily an indication of a fluency problem, the presence of many questionable behaviors is suggestive of it or indicates the need to monitor the childs fluency (Pindzola & Dorenda, 1986). While the diagnosis is made on the basis of the eight auditory behaviors, supportive information is obtained from the remaining sections of the protocol (i.e., visual evidence and historical/psychological indicators) (Pindzola & Dorenda, 1986). The greatest indicators of incipient stuttering are probably provided by the first five of the eight behaviors. The format of the Protocol has 3 advantages: it guides the speech language pathologist in clinical observations of the client; the speech language pathologist knows exactly which behaviors to count or quantify, and areas in need of data collection can be delineated; and finally, the Protocol provides interpretive guidelines for normal behaviors and clinical signs, and helps the clinician differentiate normal disfluencies from incipient stuttering) (Pindzola & Dorenda, 1986.) Pindzola, 1987, used an experimental version of the Protocol for Differentiating the Incipient Stutterer to 43 children having normal language, fluency, voice, articulation, and hearing to establish normative reference points. The minimum obtainable score was 14 and maximum score 42. The normal group had scores ranging from 14–21. The Protocol was used to measure the construct validity in 7 children seen for clinical evaluation of disfluencies and 7 normal children. The mean total score for the children who stutter was 27.86 (range of 22–40), while for the normal children, it was 17.71 (range of 15–20). Therefore, there was wide difference in the group means without any overlap between the two groups. Adams, 1977, based his criteria mainly on studies published in the 1950s and one from the 1960s comparing disfluencies of children who stutter and that of normally speaking children. The protocol of Adams, 1980, includes five behavioral characteristics to differentiate between the normally disfluent child and the incipient stutterer. This protocol can be completed just by speech evaluation of the child. The number of incipient stuttering characteristics, and their magnitude or strength helps to make a decision whether the child is an incipient stutterer. The Chronicity Prediction Checklist (CPC) of Cooper & Cooper, 1985, helps to differentiate between children expected to recover without intervention and those who would not. The checklist contains 27 questions, in a yes or no format. The historical indicators include: family history of stuttering, increase in severity of the stuttering, types of early stuttering, and the parent’s presence or absence of concern about stuttering. Attitudinal indicators include the child’s self-perception as a person who stutters and how they react to the problem. There are 17 behavioral indicators (presence or absence of a characteristic). The probability of recovery from stuttering is estimated from the number of yes responses (0–6 = predictive of recovery, 7–15 = requiring vigilance, and 16–27 = predictive of chronicity). The Stuttering Prediction Instrument for Young Children (SPI) by Riley, 1984, provides reliability and validity information. Two experienced clinicians independently measured the interexaminer reliability of SPI by administering it to eight children; a rank order correlation of .95 on total SPI scores was found. The part-whole reliability was determined by computing the correlations between the subsections and the total score. Section II (reactions to the child’s disfluencies) had the highest correlation with total score (.68). The number of iterations of a repeated vowel (.20) and duration of vowel prolongations (.21) were least correlated with total score. The obtained part-whole correlations for phonatory arrest, articulatory posturing, and severity (number) of iterations of part-word repetitions were all between .57 and .60. The subtests with the highest part-whole correlations were given greater weight in the final form of the SPI. Curlee, 1980, developed a protocol to identify chronic rather than incipient stuttering. Seven criteria in the protocol help to identify those who will not recover without treatment. The examiner first observes the child’s speech during unstructured play with the parents, following which, direct questions are asked or specific information is requested. The communication stress level is then gradually increased further by interrupting the child or by engaging in play activities. The observations are then compared with the reports obtained earlier from the parents. Critical review Van Riper, 1971 & 1982, does not give any information about how the observations necessary for making the differential diagnosis were to be made. Only four of the 26 behavioral characteristics involve quantification. A considerable subjective judgment on the part of the examiner is required for the Van Riper protocol. Cooper & Cooper, 1985, does not provide any specific guidelines for the size of speech sample, and only makes a general statement about the prior conditions necessary before the protocol is administered. Among the 27 items in the Cooper Chronicity Prediction Checklist, three behavioral indicators or 11% of the items are based on quantifiable data. This is a relatively low degree of reliance on quantitative measures. Like the Van Riper protocol, a considerable subjective judgment on the part of the examiner is required. Curlee, 1980, does not specify the length of the samples nor provide any information on how the criteria are computed, weighted, or utilized in the diagnostic determination, but the protocol describes the techniques for evaluating the child’s fluency in different degrees of communicative stress. Three of the seven items involve quantitative information. The protocol includes qualitative judgments (e.g., vocal tension and increases in loudness) within the first two criteria. The overall case selection strategy utilizes qualitative and quantitative judgments equally. Three protocols (Adams, 1977, 1980; Pindzola, 1987; Riley, 1984) stipulated a specific size database for making differential diagnoses between incipient stuttering and normal disfluency. The size of the database recommended was different in all protocols. In the Stuttering Prediction Instrument (Riley, 1984), some individual items rely on quantified information. Subjective and quantitative criteria are utilized equally. The Adams protocol, 1980, based the greatest proportion of their diagnostic criteria on quantifiable characteristics. In Pindzola’s Protocol for Differentiating the Incipient Stutterer, 5 out of the 14 criteria (36%) are based on frequency of occurrence, duration, or number of reiterations. The greatest proportion of the diagnostic criteria is based on quantifiable characteristics. After reviewing the 6 protocols, Gordon & Luper, 1992, observed that some degree of subjective evaluation as well as objective quantification were present in all protocols. While some protocols are designed only to identify chronic stuttering, others are designed to identify incipient stuttering. A wide variation is present in the extent of documentation provided for the specific criteria and for the validity and reliability of the protocols. There are differences in the in recommended data collection procedures, and the relative amount of reliance upon objective, quantitative data among the protocols. The number and type of speech and nonspeech criteria utilized to make the final diagnostic determination, varies in the different protocols. The frequency and/or percentage criteria were utilized by all protocols to differentiate normally disfluent speech from stuttering (Gordon & Luper, 1992). However, there is a great variation among the different protocols in the weightage given to frequency/ percentage criteria and the specific behaviors that are counted for the frequency/percentage criteria (Gordon & Luper, 1992). There is no uniform agreement on which behaviors are crucial and the amount of disfluency required to categorize as stuttering. All these variations can make the clinician undecided on which protocol to use (Gordon & Luper, 1992). There is a possibility of making a misdiagnosis when using a diagnostic protocol; there are two kinds of errors, false positive error, and false negative error (Gordon & Luper, 1992). A false positive error occurs when a diagnosis is made that a person has a problem, when they really do not (Gordon & Luper, 1992). A false negative error occurs when a diagnosis is made that a person does not have a problem, when they really do have a problem (Gordon & Luper, 1992). These kinds of errors in diagnosis result in errors in treatment (Gordon & Luper, 1992). Making the diagnostic categories more distinct can reduce false positive errors. As an example, the Protocol for Differentiating the Incipient Stutterer differentiates between normal disfluency and stuttering on the basis of five percentage points instead of one percentage point by some other protocols (Gordon & Luper, 1992). Making the criteria for the diagnosis of stuttering stricter can also avoid false positive errors. For example, the protocols by Cooper and Cooper, 1985 and Pindzola, 1987 result in a total score, and in the Adams, 1980 protocol, children who meet or surpass at least four of the five criteria are considered to be incipient stutterers (Gordon & Luper, 1992). Conclusion A critical review of the six protocols shows that all protocols differ in more than one aspect and have some drawbacks. In order to obtain valid diagnostic results, a representative sample is an absolute necessity; however, none of the current differential diagnostic protocols have focused on this aspect. Only the protocols of Riley, 1984 and Pindzola, 1987, attempts to assess the reliability of protocols and presented any formal validation; although neither of the two protocols measured the instrument’s predictive validity. Only three protocols (Adams, 1977, 1980; Pindzola, 1987; and Riley, 1984) indicated a specific size database for differentiating between normal disfluency and stuttering; although there was no agreement in any of the 3 protocols on the size and type of speech sample that is required to make a diagnosis. There is no uniform agreement in any protocol on which behaviors are crucial and the amount of disfluency required to categorize as stuttering. The use of the Protocol for Differentiating the Incipient Stutterer, Cooper & Cooper, 1985 and Adams, 1980 may reduce false positive errors. It is recommended that diagnostic protocols focus on having a representative sample and standardize the procedures for eliciting the speech sample. The size and type of speech sample that is required to make a diagnosis should be evaluated. Further studies on the validity of the protocols should be conducted. Studies should be conducted to analyze the extent to which factors like objective/subjective criteria or speech/nonspeech criteria affect the reliability and validity of the protocols. More research is needed on identifying, which behaviors are crucial and the amount of disfluency that is required to categorize as stuttering. References Adams, MR (1977). A clinical strategy for differentiating the normally nonfluent child and the incipient stutterer. Journal of Fluency Disorders, 2, 141–148. Adams, MR (1980). The young stutterer: Diagnosis, treatment and assessment of progress. Seminars in Speech, Language, and Hearing, 1, 289–299. Bloodstein, O (1981). A handbook on stuttering. Chicago: The National Easter Seal Society. Cooper, EB, Cooper, CS (1985). Cooper personalizedfluency control therapy (rev. ed.). Allen, TX: DLM Teaching Resources. Curlee, RF (1980). A case selection strategy for young disfluent children. Seminars in Speech, Language, and Hearing, 1, 277–287. Gordon, PA, Luper, HL (1992). The Early Identification of Beginning Stuttering I: Protocols. AJSLP. Gordon, PA, Luper, HL (1992). The Early Identification of Beginning Stuttering II: Problems. AJSLP. Pindzola, RH (1987). Stuttering intervention program. Tulsa,OK: Modern Education Corporation. Pindzola, RH, Dorenda, TW (1986). A Protocol for Differentiating the Incipient Stutterer. American Speech-Language-Hearing Association:2-10 Riley, GD (1984). Stuttering Prediction Instrument for Young Children (rev. ed.). Austin, TX: Pro-Ed. Van Riper, C (1971). The nature of stuttering. Englewood Cliffs, NJ: Prentice-Hall. Van Riper, C (1982). The nature of stuttering (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall. Read More