Methodology for Perceptual Assessment of Speech in Patients With Cleft Palate: A Critical Review of the Literature ANETTE LOHMANDER, S.L.P., PH.D. MARIA OLSSON, S.L.P., M.SC. Objective and Design: This review of 88 articles in three international journals was undertaken for the purpose of investigating the methodology for perceptual speech assessment in patients with cleft palate. The articles were published between 1980 and 2000 in the Cleft Palate–Craniofacial Journal, the International Journal of Language and Communication Disorders, and Folia Phoniatrica et Logopaedica. Results and Conclusions: The majority of articles (76) were published in the Cleft Palate–Craniofacial Journal, with an increase in articles during the 1990s and 2000. Information about measures or variables was clearly given in all articles. However, the review raises several major concerns regarding method for collection and documentation of data and method for measurement. The most distressing findings were the use of a cross-sectional design in studies of few patients with large age ranges and different types of clefts, the use of highly variable speech samples, and the lack of information about listeners and on reliability. It is hoped that ongoing national and international collaborative efforts to standardize procedures for collection and analysis of perceptual data will help to eliminate such concerns and thus make comparison of published results possible in the future. KEY WORDS: cleft palate, literature review, methodology, perceptual speech assessment

Speech is an important, although controversial, aspect of the management of cleft lip and palate around the world. This is due in part to the special problems related to clinical research methodologies used for the evaluation of cleft lip and palate treatment such as multidimensionality of outcome, length of follow-up, sample sizes, and reproducibility and validity of outcomes measures (Roberts et al., 1991). Furthermore, the diversity of approaches to surgical management renders evidence regarding speech outcome inconclusive. A recent survey revealed that 201 different European teams used 194 protocols for one cleft subtype, making comparison of outcomes impossible (Shaw et al., 2000). Another problem pertains to the fact that speech outcomes were routinely reported by professionals other than speech pathologists (Jackson et al., 1983), thus calling the validity of the results into question. The situation has changed over the past decades, and speech-language pathologists are now valued members of the interdisciplinary team

(Grunwell and Sell, 2001). However, procedures for speech assessment continue to vary considerably, and the validity of results can still be questioned. Perceptual speech evaluation is the basis of the speech assessment. The final decision regarding whether an individual has nasality or other speech problems is based on the listener’s subjective measurement (Moll, 1964). In a State-of-the-Art article, Kuehn and Moller (2000) presented a thorough review of methods used for description of speech characteristics related to cleft palate during the last 50 years. Although no information about speech material and method of assessment was provided in that article, the authors stated that ‘‘we must strive for a more standardized protocol for describing articulation . . . ’’ (p. 348–345) regarding the different aspects of speech in clefts. In other words, more consistent reporting in the literature and comparison of speech findings across centers is needed. Previously, this need was stressed by Morris (1973), highlighted by Dalston et al. (1988) in their article on minimal standards for results of surgery on patients with cleft lip and palate, and later substantiated by Grunwell and Sell (2001). Each of these authors identified a number of methodological flaws regarding perceptual evaluation of speech in cleft lip and palate that make evaluation and comparison of data invalid. The flaws include lack of information regarding selection of patients/subjects and specified ages at surgery and assessment, the use of different professionals as raters, and the lack of inter- and intrarater reliability. Methodological shortcomings

Dr. Lohmander is Head of the Department and Speech-Language Pathologist. Ms. Olsson is Speech-Language Pathologist and Researcher at the Cleft Palate Centre, Sahlgrenska University Hospital, and the Department of Logopedics, Sahlgrenska Academy at Go¨teborg University, Go¨teborg, Sweden. This work was in part presented at the 9th International Congress on Cleft Palate and Related Craniofacial Anomalies; Go¨teborg, Sweden; June 2001. Submitted October 2002; Accepted April 2003. Address correspondence to: Dr. Anette Lohmander, Department of Logopedics, Sahlgrenska Academy at Go¨teborg University, Box 452, SE 405 30 Go¨teborg, Sweden. E-mail [email protected]. 64

Lohmander and Olsson, SPEECH ASSESSMENT IN CLEFT PALATE

in reporting speech results in patients with cleft palate were also highlighted by Lohmander-Agerskov (1998). In this report, a critical review of published articles on speech and delayed hard palate closure revealed that information was either lacking or differed in terms of the inclusion criteria for dependent variables such as cleft type, age at surgery, and age at speech assessment. Despite these failings, these studies now constitute the basis for opinion against delaying the timing of hard palate repair. Comparisons between studies are also difficult to make because of different formats for collecting speech samples and analyzing data, the use of different stimuli to elicit responses, and different criteria to score responses (Grunwell and Sell, 2001). Several efforts have been made to provide clinicians and researchers with methodological guidelines for assessment of speech in patients with cleft lip and palate. Peterson-Falzone et al. (2001) summarize the different methods used in the United States for assessment of articulation and resonance and make recommendations on speech material, testing and analyzing procedures, and documentation to enable the clinician to make adequate treatment plans. Recommended parameters for the evaluation and treatment of patients with cleft lip and palate or other craniofacial anomalies were also published by the American Cleft Palate–Craniofacial Association in 1993 and updated in 2000. Grunwell et al. (1993) stated that speech problems ‘‘consequent upon the cleft condition can be predicted and need to be investigated by routine speech assessment procedure.’’ (p.11). As a consequence, the Great Ormond Street screening assessment is now accepted as the national procedure for speech assessment in patients with cleft palate in the United Kingdom (Sell et al., 1994, 1999). This procedure includes both recommended speech samples and instructions for documentation and analysis. Because the procedure was considered too detailed for a national clinical audit, the Cleft Audit Protocol for Speech was developed specifically for that purpose (Harding et al., 1997). Because the two procedures have common speech material, the results can be compared more easily (Sell and Grunwell, 2001). Two examples of minimal standards for speech evaluation in patients with cleft palate have been presented with the intention for international use. Hirschberg and Van Demark (1997) made a proposal for standardization of speech and hearing evaluations to assess velopharyngeal function. Recently the Eurocleft Biomed Project published minimum standards for cleft care in Europe that included not only speech variables to be assessed but also recommendations for speech material and ages for assessment (Shaw et al., 2000). Thus, there is growing interest in coordination of procedures for speech assessment in patients with cleft lip and palate, primarily between centers and within languages. Ideally it should also be possible across languages, which would enhance evaluation of speech outcome related to treatment. The need for standardized routines for speech assessment across languages in cleft was stated in an international task force report at the 8th International Congress on Cleft Palate and Related Craniofacial Anomalies, Singapore, in 1997 (Henningsson and Hutters, 1997). In that pre-

65

sentation, the authors focused on the assessment of consonant articulation and highlighted the need to eliminate the language influence in comparisons of speech outcomes. In an attempt to identify similarities in the speech assessment of patients with cleft lip and palate in different parts of the world, four current procedures were presented and discussed at a symposium at the 9th International Congress on Cleft Palate and Related Craniofacial Anomalies in 2001. The purpose of the symposium was to set guidelines for future discussion on an international standardized procedure for perceptual assessment. The four procedures presented were 1) the Great Ormond Street Speech Assessment (GOS.SP.ASS) used in the United Kingdom (Sell et al., 1994, 1999), 2) the Japanese system for assessing cleft palate speech, 3) the perceptual system for evaluation of speech in cleft lip and palate used in the United States (American Cleft Palate–Craniofacial Association, 1993), and 4) Cross Linguistic Outcome Comparison (CLOC), a system for perceptual assessment of cleft palate speech, with special reference to cross-linguistic speech outcome comparisons. The presentation revealed that common areas of assessment were hypernasality, nasal emission, and pressure consonants. Otherwise, differences in method for assessment were detected, both regarding speech sample, method for eliciting, documentation, and rating procedure. For example, isolated sustained vowels, high vowels in words, and vowels in counting or spontaneous speech were used for rating of hypernasality on scales with different numbers of scale points. The assessment of pressure consonants varied from narrow transcription of target sounds in words or sentences to rating of amount of different compensatory errors on pressure consonants. The impression from the literature and the symposium is that there is a lack of reported information and large differences in ways of collecting and analyzing data concerning perceptual assessment of speech in patients with cleft palate. The symposium concluded that an international agreement of minimal standards for speech assessment is needed so that published results can be compared. Such an agreement must also include minimal standards for reporting data. A survey of the situation was thought to be of great value for the continuing work. Therefore, the aim of this investigation was to conduct a review of published articles regarding the methodology for perceptual assessment of speech in patients with cleft lip and palate. The intent was to find out to what extent basic data on material and procedure are being reported and the variability in speech material and method for assessment of speech in cleft palate. METHOD

AND

MATERIAL

Three journals (Cleft Palate–Craniofacial Journal [CPCJ], European Journal of Disorders of Communication/International Journal of Language and Communication Disorders, and Folia Phoniatrica et Logopaedica) were examined for articles published from 1980 to 2000 that contained perceptual studies of speech in patients with cleft palate. The journals were available at the library of the Department of Logopedics and Pho-

66

Cleft Palate–Craniofacial Journal, January 2004, Vol. 41 No. 1

TABLE 1 Distribution of Reviewed Articles by Year and Journal* No. of Articles Year

CPCJ

FPL/IJLCD

1980–1984 1985–1990 1991–1994 1995–1999 2000 Total

10 16 27 20 3 76

2 1 3 6 0 12

TABLE 2 Number of Subjects Used in Reviewed Articles (n 5 88) No. of Subjects

No. of Studies

% of Studies

,5 5–20 21–99 100–199 .200

7 32 21 22 6

8 36 24 25 7

* CPCJ 5 Cleft Palate–Craniofacial Journal; FPL 5 Folia Phoniatrica et Logopaedica; IJLCD 5 International Journal of Language and Communication Disorders.

RESULTS Distribution by Year (Table 1)

niatrics, Sahlgrenska University Hospital/Go¨teborg University (Sweden). For practical reasons, this review did not include all journals including articles on speech in cleft palate. We found, as did Whitehill (2002), that the greatest number of articles on cleft palate speech appeared in the CPCJ. To find articles from Europe that were not published in the CPCJ, we also included two European and international journals, which yielded 12 articles. The database PubMed was used to find the articles of interest. The key words for the search for each journal in the Journal Browser was ‘‘speech and cleft’’. Printed copies of the journals from 1990 to 2000 were also reviewed to make sure that all relevant articles were included. A total number of 93 articles were found in the three journals. The procedure was repeated by a second investigator a few months later to ensure that all articles were identified and deemed relevant. The second review resulted in the exclusion of five articles. Of the remaining 88 articles, 76 articles were published in the CPCJ, three were in the European Journal of Disorders of Communication/International Journal of Language and Communication Disorders, and nine were in Folia Phoniatrica et Logopaedica. The articles were examined for the following categories: • Number of subjects. • Control subjects (included or not). • Age of the subjects (if range was not included, mean of age was noted). • Number and type of speech sample. • Tape recording (performed or not). • Number of listeners for perceptual assessment. • Type of listener. • Perceptually assessed variables. • Method for perceptual judgments. • Inter- and intrareliability measures (performed or not). • Instrumental method (included or not). • Type of instrumental method (if included). All data were pooled into groups within each category, and the frequency of occurrence for each group within each category was calculated.

Of the 76 articles in the CPCJ that included a perceptual speech evaluation in patients with cleft lip and palate, 26 were published during the 1980s and twice as many (50) during the 1990s and 2000. Of the 12 articles published in the two European and international journals, three were published in the 1980s and nine in the 1990s. The survey did not manage to examine the research design in all articles because of unclear or missing information but clearly a retrospective description of speech after surgical intervention dominated (case series reports). Very few seemed to have dealt with evaluation of speech intervention (only four were found), although the objective was sometimes difficult to find out. Subjects The number of subjects in the reviewed articles is shown in Table 2. Thirty-two of the 88 articles (36%) used 5 to 20 subjects, which was the most common sample size. A control group was included in 17 of the 88 articles (19%). Three of the articles with control subjects were published during the 1980s and 14 during the 1990s and 2000. The age of the subjects is presented in Table 3. More that half of the articles included subjects with a very wide age range. In 31 of the 88 articles (35%), the ages of the subjects were between 0 and 18 years. An even wider age range was found in 30 of the articles (34%), which included subjects older than 18 years. Mean ages without range values were given in three articles. Each of these articles was included in that age group in the table that had a range value that covered the actual mean value. Information about the age of the subjects was missing in 10 of the 88 articles (11.5%; Fig. 1). TABLE 3 Age of Subjects Used in Reviewed Articles (n 5 88) Age of Subjects

No. of Studies

% of Studies

,3 ,6 ,18 ,Adult Adults Missing

6 10 31 30 1 10

7 11.5 39 34 1 11.5

Lohmander and Olsson, SPEECH ASSESSMENT IN CLEFT PALATE

67

TABLE 5 Number of Listeners in Reviewed Articles (n 5 88) Number of Listeners

No. of Studies

% of Studies

28 19 12 12 8 9

32 22 14 14 9.5 10.5

One listener Two listeners Three listeners 4–10 listeners .10 listeners Missing

FIGURE 1 Percentage of articles with missing data in reviewed articles (n 5 88).

Number and Type of Speech Samples for Perceptual Judgment Information regarding speech samples is provided in Table 4. In 33 of the 88 articles (38%), only one type of speech sample had been used. Most common was ‘‘spontaneous speech,’’ which was found in 13 articles. ‘‘Sentences’’ were used in eight articles, ‘‘words’’ in six articles, ‘‘text’’ in four articles, and ‘‘single phonemes’’ in two articles. Two different types of speech material had been used in 18 of the 88 articles (20%), a combination of single words and spontaneous speech (nine articles) was the most common combination. Information about speech sample was totally missing in 10 of the 88 articles (11%; Fig. 1). Documentation of Speech In half of the articles, audiotape or videotape recordings had been used for documentation and evaluation. The information was missing in 13 of the 88 articles (15%; Fig. 1). Listeners The number of listeners used in the articles is presented in Table 5. In 28 (32%) of the 88 articles, only one listener had been used for speech evaluation. A panel of 4 to 10 listeners or more than 10 listeners had been used in 20 (23%) of the TABLE 4 Number of Speech Samples for Perceptual Judgment in Reviewed Articles (n 5 88) Speech Sample

No. of Studies

% of Studies

One type* Two types† 3–4 types Several/mixed Missing

33 18 12 15 10

38 20 14 17 11

* Spontaneous speech (13), sentences (8), words (6), text (4), and phonemes (2). † Sentences and text (3), words and spontaneous speech (9), syllables and sentences (3), sentences and spontaneous speech (1), counting and spontaneous speech (1), and syllables and text (1).

88 articles. The type of listeners used in the articles is shown in Table 6. In 70 of the 88 articles (80%), a speech-language pathologist participated in the speech assessment, and in nine of these, naı¨ve listeners or other professionals (surgeons, phoniatricians, dentists, phoneticians) also participated. Only four articles (4%) used other professionals (surgeons, phoneticians, and clinical linguists) or naı¨ve listeners exclusively. Information about the number of listeners was missing in 9 of the 88 articles (10%). Information about the judge’s profession was missing in 14 of the 88 articles (16%; Fig. 1). Number and Type of Assessed Speech Variables A summary of the number of speech variables assessed is shown in Table 7. In 43 of the 88 articles (49%), only one variable was assessed. The most common single variables were ‘‘resonance’’ (used in 18 of the articles), and ‘‘articulation’’ (used in 18 articles). The second most common number of speech variables assessed was three used in 22 articles (25%). The types of variables assessed are shown in Table 8. In 62 of the 88 articles, ‘‘resonance’’ was included together with other variables. ‘‘Articulation’’ was included together with other variables in 56 of the 88 articles. Method for Perceptual Judgment Table 9 shows the methods by which perceptual speech characteristics were judged. Interval rating scales with 3 to 10 points were the most common method for assessment used in 65 of the 88 articles (78%). In nine of these, another method for assessment of articulation was used as well (for example, percent correct consonants, frequency and type of consonant errors, or a description of errors). Transcription only was used in eight articles (8%), and percentage of consonants correct, consonant inventory, place and manner description, or a description and frequency of errors were reported. Other methods for assessing speech were used in four articles (4%): percentage correct (1), visual analog scale (1), description and freTABLE 6 Type of Listeners in Reviewed Articles (n 5 88) Type of Listeners

No. of Studies

% of Studies

SLP* SLP and others or naive Others or naive Missing

61 9 4 14

69.5 10 4 16.5

* SLP 5 speech-language pathologist.

68

Cleft Palate–Craniofacial Journal, January 2004, Vol. 41 No. 1

TABLE 7 Number of Assessed Speech Variables in Reviewed Articles (n 5 88) Variables

One Two Three Four to five

No. of Studies

43 11 22 12

TABLE 8 Type of Assessed Speech Variables in Reviewed Articles (n 5 88) Variables

Included in No. of Studies

Resonance Articulation Perceived airflow Intelligibility Voice quality Velopharyngeal function Grimace

62 56 29 11 11 9 2

% of Studies

49 13 25 13

quency of errors (2). Information about method for assessment was missing in 4 of the 88 articles (4%) (Fig. 1). Reliability Measures About half of the articles (45) included reliability measures and 23 of these both inter- and intrajudge reliability. Seventeen included interjudge reliability only and five intrajudge reliability only. Information about reliability was missing in 43 of the 88 articles (49%; Fig. 1). Instrumental Use In 52 of the 88 articles (59%), instrumental analysis had been used as a complement to the perceptual evaluation. In 17 (33%) of the 52 articles including instrumental analysis, visual imaging (videofluoroscopy, nasoendoscopy, cephalometry) was used, whereas acoustic measurements (Nasometer [Kay Elemetrics, Lincoln Park, NJ], NORAM [Royal Institute of Technology, Stockholm, Sweden], Spectrography [Kay Elemetrics]) were used in 15 articles (28%). Several methods were used in nine articles (17%). In eight articles (15%), aerodynamic instrumentation had been used. Other types of instrumental equipment (laryngography, electromyography, electropalatography) were used in three articles. DISCUSSION The results of this review support the impression that speech-language pathologists are increasingly active in reporting speech outcomes as a result of treatment in patients with cleft lip and palate. This is reasonable because both decisions about treatment and judgments as to whether one surgical technique is superior to another are based on speech evaluations. Thus, speech is one of the key outcomes of cleft surgery. A lot of work has been done (particularly in the United States and the United Kingdom) to develop valid measures for perceptual evaluation of cleft lip and palate speech. Articulation tests permitting developmental comparison with published norms as well as showing the interaction between sound development and any velopharyngeal problems have been used in the United States for many years (e.g., the Templin-Darley Tests of Articulation [Templin and Darley, 1969]). Special articulations tests for patients with cleft palate have also been developed (e.g., the Iowa Pressure Articulation Test [Morris et al., 1961] and the GOS.SP.ASS [Sell et al., 1999]). Methods for describing and assessing patterns of cleft palate misarti-

culations have been described by Trost-Cardamone (1990) and Harding and Grunwell (1993) and recommended by the ACPA (1993, updated 2000). In the present literature review, it was also noted that information about measures or variables as well as the number of patients included was clearly given in all articles. However, when it comes to method for collection and documentation of data and method for measurement, this review raises several major concerns. Data Collection The age at assessment of patients is extremely important information when the outcome measure is speech. Even so, information about age of assessment was missing in 10 of the articles (Fig. 1). Most articles reported on speech in patients up to 18 years of age. There was often a wide age range, even though the study design was cross-sectional. It was rather difficult to establish the study design in many of the articles, but it seems that the vast majority were retrospective cross-sectional studies. Given that the most common sample size in the articles was only 5 to 20 patients in a study, a cross-sectional design with several ages included gives very few patients in each age group. With mixed cleft types, the groups get even smaller. The difficulty of getting large samples in the research of cleft lip and palate was one of the special problems in clinical research listed by Roberts et al. (1991). With an incidence of about one birth per 500, few centers are able to collect adequate samples for hypothesis testing within a reasonable time period (i.e., when important dependent variables can be controlled). The literature is therefore flawed by small samples (Roberts et al., 1991). Existing longitudinal data and retrospective intercenter studies, however, provide excellent bases to study predictors of speech outcome and for prospective trials (Roberts et al., 1991). TABLE 9 Method of Judgment in Reviewed Articles (n 5 88) Method

No. of Studies

% of Studies

Interval scale* Absent/present Transcriptions Other Missing

65 7 8 4 4

74 8 9 4.5 4.5

* In combination with other method for quantification of consonant articulation in nine studies.

Lohmander and Olsson, SPEECH ASSESSMENT IN CLEFT PALATE

It is highly important that the speech material is chosen according to the purpose of the study, yet information about speech material was missing in 13 of the reviewed articles (Fig. 1). For clinical purposes, Peterson-Falzone et al. (2001) recommended a variety of sampling procedures. They emphasized the importance of a sample of the patient’s conversational speech (i.e., connected speech). Such a sample provides information about the consistency of the patient’s errors and impact of context. Peterson-Falzone et al. (2001) further recommended articulation testing, the inclusion of syllable and sentence repetition, and audiotape recording the performance for documentation. Most clinicians probably use some or all of these sampling procedures. Spontaneous speech, single words, or both were the most common speech material used in the reviewed articles. Documentation of the performance on audiotape was done in only half of the articles. Otherwise, protocol documentation was used. This is not usable for research purposes, however, in which the speech samples should be randomized and repeated for reliability measures (see Data Analysis below). For many aspects of speech assessment, low reproducibility exists because of variation in speech pattern associated with linguistic differences (Roberts et al., 1991). For comparison of results, the speech material from which the assessments have been made must be phonetically similar. Otherwise, it will be unclear whether it is the treatment or the speech material that makes the difference. The similarity of the speech material is even more important when comparing speech outcome in patients with different languages (Henningsson and Hutters, 1997). Some languages (and speech material) are more vulnerable to cleft palate than others, depending on the number of consonants that require a high intraoral pressure. An evaluation of speech using ‘‘vulnerable’’ speech material could therefore result in worse speech results, compared with an evaluation using less vulnerable material. In intercenter studies across languages, this influence must be eliminated. Only speech materials (consonants) that are similar can be compared, which means that these should be similar with regard to the potential influence of the cleft condition on consonant productions (Henningsson and Hutters, 1997). Data Analysis Information about listeners and type of listeners was missing in many articles (Fig. 1), and often only one listener had been used. This is in agreement with Whitehill’s (2002) findings in her critical review of articles about perceptual analysis of intelligibility. According to Kuehn and Moller (2000), there is a consensus that multiple-listener judgments are preferred to single-listener judgments. Unfortunately, this practice has yet to be fully adopted. As Whitehill (2002) points out, interrater reliability depends on more than one rater. Accordingly, this parameter was not even reported in half of the articles, either in the review by Whitehill or in the present one. Furthermore, in articles that

69

reported reliability, the method was not fully specified. For example, in one article: ‘‘Intrajudge and interjudge reliability were assessed utilizing videotaped examples of speech samples seen at another institution.’’ (Warren el al., 1994, p. 259). Tape recordings seem to have been used in only half of the studies. The actual figure could be different because information was missing in 13 articles. Thus, a large number of studies have been conducted using live ratings and constitute retrospective descriptions of speech outcome after surgical treatment. There has been some variation among results from different listening conditions. Stephens and Daniloff (1977) claim that articulation judgments from audio recordings are likely to be less reliable than are live ratings, whereas Moller and Starr (1984) found that different listening conditions such as live, audio, and audiovisual, result in similar judgments of nasality, articulation, and intelligibility. In any case, tape recordings (audio or video) should be used in research for better reliability and to be available for future reference. Unfortunately, tape recordings are not always properly used. For example, in one article, ‘‘tape recordings . . . were assessed by the same three speech pathologists, who had worked full time for more than 10 years in the cleft palate department. The interjudge reliability was accordingly very high but was not measured’’ (Brønsted et al., 1984, p. 172). For the basic purpose of describing speech in a center, the minimum standards for evaluation of speech should be fulfilled (i.e., speech data should be randomized and repeated, the speech-language pathologists involved in the treatment should not be used as raters, and both inter- and intrarater reliability should be reported [Sell et al., 2001]). Obviously, this requires tape recordings. The choice of audio or video recordings should be made according to the purpose of the study. Only one variable was assessed in most articles (either a resonance or an articulation variable) and most commonly on an interval rating scale. There is an ongoing discussion about whether interval scaling is a valid measure of different speech variables. For intelligibility, it is not (Kent et al., 1994), and there is recent evidence that interval scaling might not be optimal for resonance variables either (Zraick and Liss, 2000; Whitehill et al., 2002). Information of measures was given in all articles, but method of assessment was missing in four articles. In more than half of the articles, instrumental analysis had been used in addition to perceptual analysis, assuming an acceptable relationship between the instrumental and perceptual assessment. In some articles, a test of correlation was performed between the instrumental measurement and the perceptual judgment. As Kuehn and Moller (2000) point out, there is currently no instrumental technique for measuring nasality that demonstrates a sufficient relationship to perceived nasality. SUMMARY

AND

CONCLUSION

From the present review of the literature, it can be concluded that the number of articles during the 1990s including a perceptual assessment of speech in patients with cleft lip and

70

Cleft Palate–Craniofacial Journal, January 2004, Vol. 41 No. 1

palate has doubled. Unfortunately, the minimum standards for reporting speech data that are valid and reliable, as proposed by Dalston et al. (1988), have not been sufficiently addressed in many of the articles, particularly regarding data collection and use of reliability measures. It is gratifying that speech and language pathologists and therapists in the United Kingdom and Japan have agreed on national procedures for data collection and analysis of speech in patients with cleft lip and palate. Recommendations of the same kind were published in the United States almost 10 years ago and in Europe in the beginning of this millennium. From an international point of view, the Eurocleft Group study was a pioneering work for crosslinguistic comparison (Grunwell et al., 2000). Since then, a proposal by Henningsson and Hutters (1997) for cross-linguistic analysis of speech, with disregard for language-related characteristics, has been suggested and applied to the Scandcleft project, which deals with five different languages. In that project, detailed routines for data collection have also been established (Scandcleft work in progress). Ongoing national and international work will hopefully support the use and further development of standardized procedures for data collection, analysis, and report of speech in patients with cleft lip and palate. Acknowledgment. Special thanks to Christina Persson, Med. Lic. Sc., for help with the repeated review.

REFERENCES American Cleft Palate–Craniofacial Association. Parameters for evaluation and treatment of patients with cleft lip/palate or other craniofacial anomalies. Cleft Palate Craniofac J. 1993;30(suppl). Brønsted K, Liisberg W, Ørsted A˚, Prytz S, Fogh-Andersen P. Surgical and speech results following palatopharyngoplasty operations in Denmark 1959– 1977. Cleft Palate J. 1984;21:170–179. Dalston RM, Marsh JL, Vig KW, Witzel MA, Bumsted RM. Minimal standards for reporting the results of surgery on patients with cleft lip, cleft palate, or both: a proposal. Cleft Palate J. 1988;25:3–7. Grunwell P, Brønsted K, Henningsson G, Jansonius K, Karling J, Meijer M, Ording U, Wyatt RM, Vermeij-Zieverink E, Sell D. A six-centre international study of the outcome of treatment in patients with clefts of the lip and palate: the results of a cross-linguistic investigation of cleft palate speech. Scand J Plast Reconstr Hand Surg. 2000;34:219–229. Grunwell P, Sell DA. Speech and cleft palate/velopharyngeal anomalies. In: Watson ACH, Sell DA, Grunwell P, eds. Management of Cleft Lip and Palate. London: Whurr; 2001. Grunwell P, Sell D, Harding A. Describing cleft palate speech. In: Grunwell P, ed. Analysing Cleft Palate Speech. London: Whurr; 1993. Harding A, Grunwell P. Relationship between speech and timing of hard palate repair. In: Grunwell P, ed. Analysing Cleft Palate Speech. London: Whurr; 1993. Harding A, Harland K, Razzel R. Cleft Audit Protocol for Speech (CAPS). Broomfield, Chelmsford, Essex, UK: Speech/Language Therapy Department, St. Andrews Plastic Surgery Centre; 1997. Henningsson G, Hutters B. Perceptual assessment of cleft palate speech, with special reference to minimum standards for inter-centre comparisons of

speech outcome. International Task Force Report. In: Transactions of the 8th International Congress on Cleft Palate and Related Craniofacial Anomalies. Singapore: Stamford Press; 1997:xxxiii-xxxvii. Hirschberg J, Van Demark DR. A proposal for standardization of speech and hearing evaluations to assess velopharyngeal function. Folia Phoniatrica Logop. 1997;49:158–167. Jackson I, McLellan G, Scheker LR. Primary veloplasty or primary palatoplasty: some preliminary findings. Plast Reconstr Surg. 1983;72:153–157. Kent RD, Miolo G, Bloedel S. The intelligibility of children’s speech: a review of evaluation procedures. Am J Speech Lang Path. 1994;3:81–95. Kuehn DP, Moller KT. Speech and language issues in the cleft palate population: the state of the art. Cleft Palate Craniofac J. 2000;37:1–35. Lohmander-Agerskov A. Speech outcome after cleft palate surgery with the Go¨teborg regimen including delayed hard palate closure. Scand J Plast Reconstr Surg Hand Surg. 1998;32:63–80. Moll KL. ‘Objective’ measures of nasality. Cleft Palate J. 1964;1:371–374. Moller KT, Starr CD. The effects of listening conditions on speech ratings obtained in a clinical setting. Cleft Palate J. 1984;21:65–69. Morris HL. Velopharyngeal competence and primary cleft palate surgery, 1960–1971: a critical review. Cleft Palate J. 1973;10:62–71. Morris HL, Spriestersbach DC, Darley FL. An articulation test for assessing competency of velopharyngeal closure. J Speech Hear Res. 1961;4:48–55. Peterson-Falzone SJ, Hardin-Jones MA, Karnell MP. Assessment of speechlanguage problems. In: Peterson-Falzone SJ, Hardin-Jones MA, Karnell MP, eds. Cleft Palate Speech. 3rd ed. Philadelphia: Mosby; 2001. Roberts CT, Semb G, Shaw WC. Strategies for the advancement of surgical methods in cleft lip and palate. Cleft Palate Craniofac J. 1991;28:141–149. Sell D, Grunwell PA. Speech assessment and therapy. In: Watson ACH, Sell DA, Grunwell P, eds. Management of Cleft Lip and Palate. London: Whurr; 2001. Sell D, Grunwell P, Mildinhall S, Murphy T, Cornish TC, Williams A, Bearn D, Shaw WC, Murray J, Sandy J. Cleft lip and palate care in the United Kingdom (UK)—the Clinical Standards Advisory Group (CSAG) Study. Part 3—speech outcomes. Cleft Palate Craniofac J. 2001;38:30–37. Sell D, Harding A, Grunwell PA. A screening assessment of cleft palate speech ‘‘GOS.SP.ASS’’ (Great Ormond Street Speech Assessment). Eur J Disord Commun. 1994;29:1–15. Sell D, Harding A, Grunwell PA. GOS.SP.ASS’98: an assessment for speech disorders associated with cleft palate and/or velopharyngeal dysfunction (revised). Int J Lang Commun Dis. 1999;34:17–33. Shaw WC, Semb G, Nelson P, Brattstro¨m V, Mo¨lsted K, Prahl-Andersen B. The Eurocleft Project 1996–2000. Standards of Care for Cleft Lip and Palate in Europe. European Commission Biochemical and Health Research. Amsterdam: IOS Press; 2000. Stephens MI, Daniloff R. A methodological study of factors affecting the judgement of misarticulated /s/. J Commun Dis. 1977;10:207–220. Templin MC, Darley FL. The Templin-Darley Tests of Articulation. 2nd ed. Iowa City: Bureau of Educational Research and Service, University of Iowa; 1969. Trost-Cardamone JE. The development of speech: assessing cleft palate misarticulations. In: Kernahan DE, Rosenstein SW, eds. Cleft Lip and Palate: A System of Management. Baltimore: Williams and Wilkins; 1990. Warren DW, Dalston RM, Mayo R. Hypernasality and velopharyngeal impairment. Cleft Palate Craniofac J. 1994;4:257–270. Whitehill T. Assessing intelligibility in speakers with cleft palate: a critical review of the literature. Cleft Palate Craniofac J. 2002;39:50–58. Whitehill T, Lee ASY, Chun JC. Direct magnitude estimation and interval scaling of hypernasality. J Speech Lang Hear Res. 2002;45:80–88. Zraick RI, Liss JM. A comparison of equal-appearing interval scaling and direct magnitude estimation of nasal voice quality. J Speech Lang Hear Res. 2000; 43:979–988.