2003 IFA Congress: Montreal, Canada

The Dutch Stuttering Primary Prevention Project

Caroline Nater-Berkeljon

INSO, Wamveldseweg 47, 7204138, Zutphen, the Netherlands.


A parental history of stuttering makes children more susceptible to stuttering. Little is known about the effect of primary intervention. The goal of the project was to use primary intervention to reduce the incidence and prevalence of stuttering in young fluently speaking children with a parental history of stuttering. The prevention program was undertaken by 99 fluent speaking children. To date, 93 children have finished the one year program and 65 children have been followed for more than one year. The results obtained so far show that the incidence of stuttering is comparable to that of non-high risk children. Primary prevention is successful to reduce stuttering.


  1. Introduction
Stuttering is most likely to occur between the age of 2 and 5. The incidence of stuttering appears to be 5% (Andrews, et al., 1983). Half the risk has disappeared at the age of 4, three quarter by the age of six, and virtually all by the age of 12 (Andrews, 1984). The prevalence of stuttering in prepuberal children is about 1% (Bloodstein, 1995). Stuttering has a higher incidence and prevalence for males. Although the male-female ratio for young children (age 2) (Yairi, 1983)

is balanced, ratios of 4:1 and 5:1 were found in older children and adults (Bloodstein, 1995). The

difference between the incidence and prevalence indicates that the majority of children recover from a period of stuttering, a process known as natural remission (Bloodstein, 1995). The chance of remission decreases as the length of time of stuttering increases (Andrews, 1984) and if the onset has been before the age of 3 years (Johannsen, 2000). In the etiology of developmental stuttering genetic and environmental factors take an interactionist position. The hereditary component is responsible for the relatively higher incidence and prevalence of stuttering in youngsters with a parental history of stuttering (Kidd et al., 1978). Having a first degree relative who stutters increases the incidence to 28 % and the prevalence to 7,8 % (Kloth et al., 1995). Males have a higher risk to be affected, and more chance of persistent stuttering (Kloth et al., 1993). In the environmental component the negative and demanding attitudes that parents may display with regard to their child’s communicative behavior induces negative emotion that makes their offspring more vulnerable to speech disorganization. Thereby many early stutter intervention programs involve modification of parent-child verbal interactions (e. g Smith & Kelly, 1997). Specially parents are advised parents to use shorter and less complex syntax, and slow their rate of speech and turn-taking (Couture, 1990). Most early intervention programs focus on the prevention of persistent stuttering (secondary prevention), or on stuttering becoming a handicap (tertiary prevention). Little is known about the effect of intervention before the onset of stuttering (primary intervention). This study addresses the question of influencing the onset and course of stuttering in high-risk pre-school children with a parental history of stuttering by Primary Intervention. The specific objectives are: 1) to decrease the incidence from 28 % as reported by Kloth et al. (1995) to 5.6 % and the prevalence from 7.8% to 1.6 %, and 2) to determine the effect on the onset and Course of stuttering by modifying the parent-child verbal interaction before the onset of stuttering.

  1. Methods

The Dutch Primary Prevention Project is a multi-year prospective study (1995-2009) based on fieldwork designed to obtain knowledge about the effect of primary prevention among a group of high-risk pre-school fluently speaking children with a parental history of stuttering. The project is based on consultation of the parents without direct therapy for the child. A general aspect of consultation is that the parents’ viewpoints, observations, and descriptions of the child’s problem are considered most important. This implicates a high confidence in the parents’capacity to estimate and describe the childs’ status. The (stuttering) parents applied to the project by responding to advertisements in common Dutch magazines, local papers, information bulletins in health-care insurance companies or mini posters. Relevant selection criteria of this study were: the childs’ age (around two years of age), a parental history of stuttering, monolingual fluently speaking Dutch, no history of stuttering therapy for the child or for the family in the first degree (parents, sisters and brothers) for less than a year prior to the start of the project, normal development and no syndromes. To carry out this study a stuttering prevention program was written. The program was undertaken by 17 therapists, all stuttering specialists with a broad experience in stuttering therapy and working in a specialized stuttering centre or practice, spread over ten cities in the Netherlands. The children were randomly spread over two approaches: direct and indirect (see below). The different approaches made it possible to determine which way of intervention was effective. To meet late onset of stuttering each child will be followed up starting from one year after the start of the prevention program until the child reaches the age of eight.

Prevention program

The efficiency of the prevention program was based on four elements (Bosman & Hosman, 1992): spreading the sessions over a long period of time, using different kinds of media, using written hand-outs outlining the main parts of each session, and giving information in two ways: individual and in a group. The main base of the prevention program was consultation, where the parents contribute their knowledge of their child and their stuttering experience, and the speech- language therapist contributes professional knowledge of stuttering. Two approaches were used: a direct approach in which the parents are directed in time pressure elements, such as speaking rate and turn taking, and an indirect approach in where the parents were not directed in their speech- behaviour but confirmed in their spontaneous speech interactions. The prevention program included seven sessions (3 individual sessions and 4 group sessions) spread over a year. During the prevention program information was given verbally, written on paper and recorded on video (the video included instructions on the verbal parent-child interaction and was used only for the direct approach). A broad outline of the prevention program is shown in Table 1.


Table l. A broad outline of the prevention program for each approach


106 subjects (65 boys and 41 girls) were included in this study. Seven of them dropped-out during the prevention program. These subjects were eliminated from the study. Thus, ninety nine (99) subjects took part in the prevention program (61 boys and 38 girls; ratio l.6:l.0). All subjects were fluent monolingual native Dutch speaking pre-school children with a parental history of stuttering (29 stuttering mother, 59 stuttering father and 5 stuttering father and mother). To date, 93 subjects (58 boys and 35 girls; ratio 1.6 : 1.0) completed the prevention program and are the subject of this primary publication. The children were aged between 9 and 32 months (mean 19, SD = 4.0) at the time they started the program and were aged between 18 and 40 months (mean=29, SD 5.0) when they finished the program.


Before attending the prevention program the childs’ fluency was assessed with an, in the Netherlands, common used screening questionnaire of stuttering called the SLS (Screening Lijst Stotteren). After determining speech fluency the subject started the project by admission to the prevention program for approximately one year. The stuttering therapist followed a strict protocol and registered data for each subject and session. Each group session lasted two hours and each individual session lasted one hour. In the individual sessions numbered 1, 5 and 6 the parents were accompanied by the child. All individual sessions with the company of the child were video-taped. At the last session of the prevention program the fluency of the child was once more assessed using the SLS. From one year after completing the prevention program until the age of eight, the parents were asked to report annually on their childs’ fluency by filling out the Disfluency Questionnaire (Kloth, et al, 1989) and an SLS form. If a child persisted in stuttering, the parents were free to choose any kind of intervention, however those children were discharged from the project. During the project different points of measurements were used: t0: start prevention program, tl :end prevention program and t2-t7: the yearly fo1low-up measurements.


Fluent speech was determined by an SLS score of 6 or 7. SLS scores above 7 marked two categories of disfluent speech: score 8-11 indicated borderline disfluencies and a score above 11 indicated stuttering. For subjects with a SLS score above 7, the information contained in the Disfluency Questionnaire (Kloth, et al., 1989) was analysed. Fluency failures were categorized as follows: normal disfluencies, defined as revisions, interjections, non-tense silent pause, phrase repetitions, or slow whole word repetitions; borderline disfluencies, defined as slow, non-tense sound or syllable repetitions; and stuttering disfluencies, defined as rapid sounds or syllable repetitions, and tense silent or oral prolongations. A subject was classified as a stutterer if the parents indicated on the Disfluency Questionnaire (Kloth et al, 1989) that they often or very often had observed stuttering types of disfluencies and regarded their child as a stutterer.

  1. Results
The number of subjects spread over the two approaches, their age at the start and the end of the project, gender and stuttering parent are shown in Table 2.


Table 2. Number of  subjects( N ), age, gender and stuttering parent for each approach.

At the initial session (t0), all the 93 subjects had a SLS-score of 6 and none of them were reported to display dysfluencies. These subjects were submitted to the prevention program for approximately one year. An overview of data in timelines is given in Table 3.


Table 3. Overview in timelines ( to, t1 ,t2, t3 ) for each approach, specified in number of subjects( N ), the mean SLS score, minimum and maximum SLS score and the number of stuttering children (N(st)).

After submission to the prevention program (tl) 16 subjects had a SLS score above 7, see Figure 1.


Figure 1. SLS scores t]. For all subjects with a score above 7.

Eleven out of these 16 subj_ects had an SLS score of 8 or 9, which indicated bordeline dysfluencies but none of them were regarded as stutterer by their parents and professional. Five

subjects, (3 boys, 2 girls) had a SLS score of 12, 14 or 18 which indicated stuttering dysfluencies and all were regarded as stutterer by their parents and the professional. Four of these 5 subjects started an intervention program and were discharged from the project.

The age at onset of stuttering was, with the exception of one male subject, below 3 years (mean 2;4 years, SD 0.9). The age at onset was slightly higher for boys (mean 2;10, SD 0.3) than for girls (mean 2;5, SD 0.85). 

The SLS scores of the subjects in the indirect approach were lower than the direct approach.

Although 7 subjects in the indirect approach showed a SLS score above 7, none of them were regarded as stutterer neither by their parents or the professional.


Figure 2. SLS scores t2. For all subjects with score above 7.

At the one-year-follow-up measurement (t2) 89 subjects (48 direct approach, 41 indirect approach) remained in the project. For this moment at t2, data were available for 66 subjects (43 direct approach, 23 indirect approach). Sixteen subjects (13 boys, 3 girls) had an SLS score above 7, see figure 2.

Even at t2, for the indirect approach the SLS scores were lower and no subject was regarded as stuttering by their parents. a

However, 5 boys submitted to the direct approach, had a SLS score above 11, three of them (mean age: 3;9 year, SD 0.7) were classified as stutter by their parents. None of them started therapy and all remained in the project.

For this moment at the two-year-follow-up measurement (t3) data of 51 subjects (direct approach 38, indirect approach 13) out of 89 were available. Ten subjects, all submitted to the direct approach, had a SLS score above 7. Four of these (all boys) had an SLS score above 11 and the parents of two boys of them were regarded their child as stutterer (age at onset 3;9 and 7;6 years of age). One of them started an intervention program and was discharged from the project.

Still, the SLS scores for subjects submitted to the indirect approach were lower and none of these subjects was regarded as stutter by their parents. From measurement point t4 and further, too few measurements were available for analysis yet. However, a very small group subjects (12; 7 boys and 5 girls) is nearly 8 years of age and close to the end of the project, so their speech can be viewed within a couple of years. All these subjects were submitted to the direct approach. All the girls had an SLS score of 6 or 7 during the whole project and remained fluent. An overview of the SLS scores for the boys is given in Figure 3.


Figure 3. SLS scores for 7 boys over the period of the project.

Although two boys had an SLS score of 6 over the complete project, the other five had periods with scores above 7, and 2 of them (subject 6 and 7) were regarded as stutterer for a short period of time (age at onset 7;6 and 4;4 years), and one boy (subject 4) has been regarded as stutterer since measurement point tl (age at onset 2;6 years).

  1. Discussion and conclusion

Ninety-nine high-risk children with a parental history of stuttering participated in the prevention program and were subjects of this prospective long term project. Initially all subjects were viewed as non-stutter, at the end of the prevention program five subjects were regarded as stutter. With these findings the incidence of stuttering is 5.4%. The findings do not provide evidence for the high incidence of stuttering in children with a parental history of stuttering (Kloth et al., 1995; Van Praag & Jansen, 1980) but relate to the incidence of stuttering in non-high-risk children (Andrews, et al , 1983). Of those children who came to be seen as a stutterer, 3 were male and 2 were female. This yielded a male to female ratio of l.5:1 which is comparable to the ratio found by Kloth et al. (1995). In the first year of the project the subjects who attended this project were younger than those in the study of Kloth et al. (1995). Follow-up a year later, where the subjects of both studies are of the same age, revealed that the incidence of stuttering is 4.6 % with all male stutterers. The prevalence is 1.5%. The incidence and prevalence is much lower than that reported by Kloth et al. (1995). The incidence decreased even more in the follow-up two years later (2.5%). In this follow-up the children were between 4 and 5 years of age, and the decline of the incidence is according to reports of Andrews (1984). What is remarkable is that all stutterers are male. The risk of persistent stuttering may be predicted by persistently high SLS scores after completion of the prevention program and one-year after follow up. However the present analysis focussed only on the small amount of data currently available.

For both approaches, the support and information given as early as possible will benefit from an understanding and competent environment from the moment of onset. Of interest is the different effect of the approaches. Although some of the children submitted to the indirect approach seem to be at risk for stuttering none of them were regarded as a stutterer. This may be due to the fact that parents are not directed in their speech behavoir which probably eliminates a component of stress. Stress can be seen as one of the possible triggers of stuttering (Bertens & Weeda, 1998).

The main goal of this study was to decrease the incidence of stuttering from the 28% reported by Kloth et al. (1995) to 5.6% in children who are genetically vulnerable to stuttering. Based on preliminary results it seems that the objective can be met and the best way to do it seems to be primary prevention in which the parents are supported in their spontaneous verbal child interactions.

This project was financially supported by he prevention foundation. The author wish to thank Ad Bertens, Henny Hermans, and Maarten Heijzelendoorn for their unforgettable support.

Andrews, G. (1984) The epidemiology of stuttering. In R.F. Curlee & W.H. Perkins (Eds.), Nature and treatment of stuttering. San Diego, CA: College-Hill.

Andrews, G., Graig, A., Feyer,A., Hoddinott, S., Howie,P., & Neilson, M. (1983), Stuttering A review of research findings and theories circa 1982. Journal of Speech and Hearing disorders, 48, 226-246.

Bloodstein, O. (1995). A handbook on stuttering. (5‘*‘ edition). San Diego, CA: Singular Publishing group, Inc.;

Conture, E. (1990). Stuttering, (2nd ed.), Englewood Cliffs, NJ: Pentice-Hall.

Johannsen, H.S. (2000). Design of the longitudinal study and influence of symptomatology; heredity; sex ratio and lateral dominance on the further development of stuttering, Fluency disorders: theory, research, treatment and self-help, Proceedings of the third world congress of fluency disorders in Nyborg, Denmark. Section 3. Intervention 107

Kidd, K., Kidd, J ., & Records, M. (1978). The possible causes of the sex ratio in stuttering and its implications, Journal of Fluency Disorders, 3, 13-23;

Kloth, S.A.M., Janssen, P., & Kraaimaat, F.W. (1993). The genetic component in stuttering: a family study. Stem-, spraak-, en Taalpathologie, 1993, Vol. 2/4, 247-256.

Kloth, S.A.M., J anssen, P., & Kraaimaat, F.W. (1989). Dysfluency Questionnaire. Academic Hosptal Department of Phoniatrics, Utrect, Holland.

Kloth, S.A.M., Kraaimaat, F.W., Janssen, P., & Brutten, G.J., (1995). Speech-motor and linguistic skills of young stutterers prior to onset, Journal of Fluency Disorders, 20, 157-170.

Ryan, B.P. (2000). Prediction of spontaneous recovery in preschool children who stutter. Fluency disorders: theory, research, treatment and self-help, proceedings of the third world congress of fluency disorders in Nyborg, Denmark.

Smith, A., &, Kelly, E., (1997) Stuttering: A dynamic, multifactorial model. In R.F. Curlee & G.M. Siege] (Eds.) Nature and treatment of stuttering. New directions. (2nd ed.) Needham Heights, MA; Allyn & Bacon.

Wall, M.J., & Myers, F.L. (1984). Clinical management of childhood stuttering. Baltimore, MD: University Park Press.

Yairi, E. (1983) The onset of stuttering in two- and three-year-old children: A preliminary report. Journal of Speech and Hearing Disorders, 48, 171-177.

join button


The IFA implemented Japanese translations of some pages on the site for the 2018 Joint World Congress. Choosing Japanese below to see these translations.

Not all pages are translated, but you can use Google translate to see a machine translation using the switch below

Google Translate


Journal of Fluency DisordersBrowse the current issue

The official journal of the International Fluency Association
IFA Members receive online access to JFD as a member benefit.

Read more: JFD

Twitter Feed @official_IFA

IFA on Facebook