Stuttering is a speech fluency disorder that presents with repetitions, prolongations of sounds and syllables, and speech blocks. Verbal or situational avoidance behavior and involuntary movements may develop over time in patients diagnosed with stuttering (Mulligan et al., 2001; American Psychiatric Association, 2013). Stuttering that persists into adulthood can lead to significant restrictions in quality of life and social and professional development (McAllister et al., 2012).

The most frequent form of stuttering, childhood onset speech fluency disorder, occurs in at least 5% of all children and typically presents between the ages of 3 and 6 years (Bloodstein and Ratner, 2008; Reilly et al., 2013). Recovery frequently occurs within the first years, particularly in girls. Stuttering persists after puberty in approximately 1% of the general population, with a male to female ratio of 4 to 1 (Yairi and Ambrose, 2013). Currently, access to treatment is limited by regional availability (Donaghy and Smith, 2016).

Since 1979, the German Stuttering Association has aimed to “counteract the development of stuttering and to improve the living situation of people who stutter” (Bundesvereinigung Stottern & Selbsthilfe e.V., 2020), an effort that has included performing critical reviews of available therapies in terms of accessibility, evidence, and efficiency.

In dialogue with therapists, and during the creation of the guidelines for speech fluency disorders (Neumann et al., 2017), the scarcity of data regarding the current state of stuttering therapy in Germany was emphasized (Radtke, 2019; Waltersbacher, 2019). Questions regarding (1) the frequency with which stuttering is diagnosed in Germany and (2) the form, intensity, and duration of current stuttering therapies in Germany were raised. Another unanswered question was (3) whether the intensity of speech therapy (intensive therapy lasting several weeks compared with one or two weekly sessions for several months) had any relevant impacts on the success or the duration of therapy.

Answering these questions will provide insight into current treatment realities, will help identify treatment traditions and patterns, and might encourage a debate on optimized use of treatment resources for stuttering, cluttering, and developmental disorders of speech in Germany and other countries.

To understand how timely the diagnosis is made, and how timely treatment is initiated, we also assessed the proportion of newly diagnosed patients in the year 2017, and the proportion of treatment allocation in the first year of diagnosis.

Co-existing disorders may influence the long-term response to treatment (Iverach et al., 2009) and might shed light on potential underlying disease mechanisms. A range of disorders has been reported to occur more frequently among individuals who stutter. We used this large database to verify or refute these reports, and contrasted it with the comorbidities of cluttering as well as developmental disorders of speech and language.

We also assessed data on developmental disorders of speech and language, as well as on cluttering. Speech and language abilities are key factors for successful schooling and career development (McAllister et al., 2012; Dubois et al., 2020), and they have received increasing attention in Germany after the relatively poor performance of German pupils during the early runs of the Program for International Student Assessment (PISA)¹ (Turner and Adams, 2007). Cluttering, on the other hand, is a rare disorder characterized by a speech rate that is perceived to be abnormally rapid, with some overlap with stuttering (Myers et al., 2012; Bona, 2019). Specific developmental disorders refer to disorders in which development is delayed in one specific area, such as speech and language, which can present with a broad range of clinical characteristics (Neumann et al., 2009) without affecting other areas of development. These disorders provide a useful background and context for the data on stuttering.

This study was performed as a retrospective analysis of secondary data, conducted using the research database of Wissenschaftliches Institut der AOK (Allgemeine Ortskrankenkassen) (WIdO, Research Institute of the Local Health Care Funds, Berlin, Germany). AOK is the largest sickness fund group within Germany’s statutory health insurance system, able to provide access to the medical details of approximately 32% of the total German population (Nimptsch et al., 2014; Busse et al., 2017; Karagiannidis et al., 2020). 87.7% of citizens in Germany have statutory health insurance (Federal Ministry of Health, 2020), and membership is open to everyone, regardless of factors such as profession, income, age, or comorbidities (Busse et al., 2017; Karagiannidis et al., 2020). Available data were anonymous at the patient level but included patient characteristics, such as age, sex, diagnosis, admissions as inpatients, practitioner consultations, medications used, and other items associated with the use of healthcare services. In Germany, physicians’ claims must be submitted at the end of each quarter, generating four time units for each year in the dataset, with each unit representing a 3-month period. In total, 16 quarters were available for the consecutive insurance years of 2015–2018. The study was approved by the ethics committee of the University of Frankfurt. No funding sources were obtained for this study. STROSA guidelines (Standardized Reporting Of Secondary data Analyses) were followed (Swart et al., 2016).

This study represents the first nationwide study to use German health insurance data to analyze the incident and prevalent diagnoses of a disorder of speech and language, including stuttering (F98.5), cluttering (F98.6), and developmental disorders of speech and language (F80), and to evaluate the speech therapy treatment patterns.

Recently, an increasing proportion of children have received treatment for language development disturbances, which has been the focus of lively debate. Our data are consistent with this reported nationwide trend. Table 1 shows that an increasing percentage of the total insured population has been diagnosed with an ICD-10 diagnosis of developmental disorders of speech and language, stuttering, or cluttering. This increase can largely be attributed to developmental disorders of speech and language, whereas the proportion of stuttering individuals among insurance members has remained relatively stable, at approximately 0.3% of the total population (Supplementary Table 1).

The age distribution observed in this study is consistent with the known epidemiology of speech and language disorders, which peak at approximately 4 to 5 years of age, which is a phase of active language development. For stuttering, this peak is earlier, at 30–36 months of age (Mansson, 2000; Bloodstein and Ratner, 2008). Our conservative condition of two preceding years without the diagnosis might have shifted incidence peaks slightly toward older ages.

The annual crude prevalence of stuttering peaked in our sample among patients 5 years of age, which represented 0.65% of the total population. This finding is consistent with the literature summarized in Chapter 3 of Bloodstein and Ratner (2008). In contrast with annual prevalence, as analyzed in this study, the frequent statement that “five percent of all children stutter” (Walsh et al., 2015) refers to the cumulative lifetime prevalence of stuttering. However, we did not analyze longitudinal data; therefore, we cannot provide a lifetime prevalence for our cohort.

In our dataset, the gender ratio for all speech and language disorders favors girls, who are less affected, and remained stable across all the years included in the database under study. This is consistent with the literature (McLeod and McKinnon, 2007; Arrhenius et al., 2018). With regard to stuttering, a yet unresolved question is whether the gender imbalance increases over the years, which would be expected because spontaneous recovery is observed more frequently in girls than in boys (Yairi and Ambrose, 2013; Kefalianos et al., 2017); however, this assumption is not universally supported by population-based data (National Institute on Deafness and Other Communication Disorders, 2010). Unfortunately, such longitudinal insights cannot be derived from our cross-sectional dataset.

The allocation of treatment resources is another debated topic (Wissenschaftliches Institut der AOK, 2012). Our population-based study yielded novel insights regarding treatment prescription behavior. Some affected individuals did not receive therapy, although the diagnosis-based coding system used by ICD-10 does not reflect disease severity or the need for therapy. Therefore, a substantial proportion of affected individuals may not seek therapy because they are only mildly or briefly affected. Indeed, spontaneous recovery occurs early, typically within the first months of speech dysfluency, and most patients recover within the first 2 years after disease onset (Lattermann, 2011), although the extent of early recovery is debated (Reilly et al., 2013).

In our sample, approximately 45% of individuals diagnosed with stuttering received dedicated speech therapy within the year of receiving a coded ICD-10 diagnosis. In Germany, mandatory child screening examinations performed by pediatricians are associated with a high attendance rate, rendering the possibility of underdiagnosis unlikely (Schmidtke et al., 2018; Santos et al., 2020), at least for moderate or severe cases (Winters and Byrd, 2020). In addition, the diagnostic tools available to evaluate preschoolers have improved (Neumann et al., 2014a, b). In our sample, a striking peak in treatment allocation was observed starting at 6 years of age, which coincides with the beginning of regular schooling in Germany, a milestone that may trigger increased demand for treatment. This is difficult to reconcile with the current recommendations for a maximum wait-and-see delay of 1 year (Neumann et al., 2017). Because earlier therapy is likely more able to induce lasting recovery (Jones et al., 2005; Neumann et al., 2017), the current peak of therapy at the time of school entrance may indicate the possibility of earlier treatment. In Germany, speech language treatment usually does not take place in schools. Hence, frequency and spacing of therapy are not necessarily determined by traditions of school scheduling.

The prevalence of treatment in our study was higher with 45% during the first year of diagnosis than in the Australian Early Language in Victoria Study (Kefalianos et al., 2017). Of those parents who reported about seeking treatment by the age of 7 years, only 16.7% of children with persistent stuttering had received intervention for stuttering at some point during the preschool years. In those who had recovered from stuttering only 13.4% received stuttering treatment. At the age of 7 years, 39% of parents of children with persistent stuttering reported that they had sought help or advice for their child’s stuttering at some point during the preschool years, whereas 28% of parents of children who had recovered from stuttering reported seeking help or advice (Kefalianos et al., 2017).

The yearly mean number of therapy sessions for stuttering was slightly lower than 20, reflecting the receipt of two of the usual prescriptions for ten treatment sessions. Treatment frequency is usually once per week. We cannot infer the total treatment duration because we analyzed yearly, partially independent, cross-sectional samples rather than longitudinal data. Therefore, total treatment duration or intensification of therapy over time cannot be inferred. By comparison, the mean number of treatments for developmental disorders of speech and language was comparable, at 21 sessions per year. Of note, there is no formal upper limit of sessions per case per year. In practice, the limit is the number of prescriptions for a single case issued and signed by the pediatricians.

Although the neurological background of stuttering is increasingly understood (Neef et al., 2015), the factors that influence the evolution toward recovery or persistency remain elusive. Comorbidities, such as anxiety or other psychiatric disorders, increase the likelihood of relapse following treatment (Iverach et al., 2009; Menzies et al., 2014). A higher prevalence of comorbidities among the cohorts with stuttering, cluttering, or developmental disorders of speech and language could be confirmed in this study. In addition, atopic disorders, such as hay fever, have been shown to be associated with stuttering persistency (Strom and Silverberg, 2016; Ajdacic-Gross et al., 2020). In our sample, we could substantiate this evolving matter by showing increased odds ratios for atopic disorders. A putative interaction between the two groups of disorders remains to be elucidated and is beyond the scope of this explorative study.

The significance of this study arises from the analysis of health insurance data for a sample population that represents 32% of the German population. This study represents the first time that such an analysis has been performed for a disorder of speech and language. Prevalence and gender ratio data were consistent with the international literature. The crude prevalence of developmental disorders of speech and language increased from 2015 to 2018, whereas the crude prevalence for stuttering remained stable. For cluttering, the numbers were too low to draw reliable conclusions. Proportional treatment allocation for stuttering peaked at 6 years of age, which is the school entrance year, and is later than the prevalence peak of stuttering. Future analyses should explore whether new approaches to treatment could improve outcomes for severely affected patients. Follow-up longitudinal studies will allow an even better characterization of treatment intensity and duration.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

MS and ASt developed the idea for this study and conceived the manuscript. AW collected the data. MS, AW, and ASt performed the statistical analysis. ASt created the tables and figures. All authors wrote the manuscript, discussed the results, contributed to the final manuscript, and approved the final manuscript for publication.

MS serves as chairman of the German Stuttering Association. He reports personal fees and grants from Deutsche Forschungsgemeinschaft, Primate Cognition (Leibniz-WissenschaftsCampus), Scientific Organizations (EFCN, UCL, DGKN, and IVS), and pharmaceutical companies (Novartis, GlaxoSmithKline, UCB, and Medtronic), all outside the submitted work. AW, Projektleiterin AOK-Heilmittel-Informations-System; HS, Stellvertretender Geschäftsführer WIdO; and ASc, research associate, are employees of Wissenschaftliches Institut der AOK. ASt reports personal fees and grants from Arvelle Therapeutics, Desitin Arzneimittel, Eisai, GW Pharmaceuticals, LivaNova, Marinus Pharmaceuticals, Medtronic, UCB Pharma, and Zogenix, all outside the submitted work.