Stuttering is a developmental disorder of communication that affects approximately 5% of children born in the United States and Western Europe. Children are at highest risk for beginning to stutter between their second and fourth birthdays. The risk decreases gradually thereafter, with few onsets occurring after 9 or 10 years of age (Andrews and Harris, 1964). The percentage of older children, adolescents, and adults who stutter is much lower, about 0.5%–1.0% (Andrews, 1984; Bloodstein, 1995), and the discrepancy between the percentage of children affected (i.e., incidence) and the percentage of older children and adults who stutter (i.e., prevalence) indicates that 75%–90% of the children who begin to stutter stop. Complete, untreated remissions of stuttering are most likely to occur within 2 years of onset (Andrews and Harris, 1964; Yairi and Ambrose, 1999; Mansson, 2000), with decreasing frequency after that.

Most of the data on the epidemiology of stuttering have been obtained from cross-sectional surveys that asked informants if they or family members currently stutter or had ever stuttered. The credibility of such data is compromised by a number of methodological weaknesses (Yairi, Ambrose, and Cox, 1996). Prospective, longitudinal studies employing trained examiners have been completed in England (Andrews and Harris, 1964) and Denmark (Mansson, 2000). The incidence (5.0%) and remission rate (>75%) reported by both studies were remarkably similar, despite substantial differences in their designs and the populations studied.

The incidence, prevalence, and remission or persistence of stuttering are affected by sex and family histories of stuttering. More than two-thirds of the children who stutter have first-, second-, or third-degree relatives who currently or once stuttered (Ambrose, Yairi, and Cox, 1993). Like most speech-language disorders, stuttering affects more males than females, with about twice as many young male preschoolers affected as females, a ratio that increases to four or more males to every female among adults (Ambrose, Yairi, and Cox, 1993). Similar ratios have been reported in other countries and cultures (Bloodstein, 1995; Ambrose, Cox, and Yairi, 1997; Mansson, 2000). The increase in male-female ratio with age reflects, in part, higher rates of remission among females (Ambrose, Cox, and Yairi, 1997), whereas family histories of remission and persistence are linked, respectively, to untreated remissions of stuttering within 2 years of onset or its persistence for 3 or more years (Ambrose, Cox, and Yairi, 1997).

Findings from family pedigree studies are consistent with the vertical transmission (i.e., generation to generation) of a genetic susceptibility or predisposition to stutter but are inconsistent with autosomal dominant, recessive, or sex-linked transmissions (Kidd, Heimbuch, and Records, 1981; Yairi, Ambrose, and Cox, 1996). Twin studies have found that stuttering occurs in both members of monozygotic twin pairs much more often than in same-sex, dizygotic twins (e.g., Howie, 1981); however, the lack of concordance of stuttering in some monozygotic twin pairs indicates that both genetic and environmental factors are involved in some, if not all, cases. Segregation analyses suggest that a single major locus is a primary contributor to stuttering phenotypes but that other genes are involved in determining whether or not stuttering persists (Ambrose, Cox, and Yairi, 1997). Research centers in the United States and Europe are currently engaged in linkage analyses designed to identify the specific genes involved.

Current etiological theories reflect diverse beliefs about the nature of stuttering, its origins, and the levels of description that will provide the most useful scientific explanation. However, no theory of origin has achieved general acceptance in the field. There are, for example, cognitive theories, such as Bloodstein's (1997) anticipatory struggle theory, which hypothesizes that a child's belief that speech is difficult elicits tension that causes stuttering when he or she tries to speak; psycholinguistic theories (e.g., Ratner, 1997), which propose that the linguistic processes responsible for everyday speech errors are also responsible for developmental stuttering; motor control theories, which link stuttering to sensorimotor or speech motor processes (e.g., Neilson and Neilson, 1987); and multifactor theories, which attribute stuttering to an interplay of the cognitive, linguistic, motor, and affective processes involved in spoken language (e.g., Perkins, Kent, and Curlee, 1991; Smith and Kelly, 1997).

Adult stuttering typically involves various behaviors and affective-cognitive reactions that affect, in varying degrees, interpersonal communication, vocational opportunities, and personal-social adjustment. Frequent repetitions and prolongations of sounds and syllables and blockages of speech occur that cannot be readily controlled (Perkins, 1990). These speech disruptions are often accompanied by facial grimaces and tremors, disrhythmic phonation, and extraneous bodily movements that seem to involve muscle tension and excessive effort.

Stuttering does not occur randomly, but as if constrained by various linguistic variables (Ratner, 1997). For example, it occurs more often than chance would suggest on initial words of phrases, clauses, and sentences, on stressed syllables, and on longer, less familiar words. With the exception of the sentence-initial position, these are the same loci that attract the speech errors of nonstuttering speakers (Fromkin, 1993), suggesting that the same linguistic variables affect both groups' speech.

The frequency and severity of adults' stuttering may vary substantially across time, social settings, and contexts, but it often occurs on specific words (e.g., the person's name) and in situations where stuttering has frequently occurred in the past (Bloodstein, 1995). Thus, adults' prior stuttering experiences may lead them to avoid specific words and speaking situations and to develop attitudes and beliefs about speaking, stuttering, and themselves that can be more disabling or handicapping than their stuttered speech.

Stuttering is substantially reduced, sometimes eliminated, in a number of verbal activities (Bloodstein, 1995). For example, most adults stutter little or not at all when singing, speaking while alone or with pets or infants, reading or reciting in unison with others, pacing speech with a rhythmical stimulus, and reading or speaking with auditory masking. Stuttering reappears, however, as soon as such activities end.

Much of what is known about adults who stutter has been obtained by studies that compared stuttering and nonstuttering speakers' motor, sensory, perceptual, and cognitive abilities, as well as the two groups' affective and personality characteristics. A variety of differences have been found, but usually with substantial overlaps in the data of the two groups. In addition, it is seldom clear if or how such differences might be functionally related to stuttering. A comprehensive review of this work led Bloodstein (1995) to conclude that the two groups are similar, for the most part, except when they speak. Recent advances in brain imaging technology, however, have allowed investigators to compare the brain activity patterns of the two groups while the subjects read aloud, and the brain activity of stuttering speakers during stuttered and fluent speech.

There are no reliable differences in cerebral blood flow of stuttering and nonstuttering adult men when they are not speaking (Ingham et al., 1996; Braun et al., 1997). A series of H₂¹⁵O positron emission tomography (PET) of the two speaker groups during solo and choral reading conditions found greater right than left hemisphere activity in the supplementary motor and premotor areas (BA 6), anterior insula, and cerebellum and reduced activity in primary auditory areas (BA 41/42) of stuttering speakers in the solo condition, but exactly the opposite pattern of activity in nonstuttering speakers (Ingham, 2001). These differences decreased, however, when fluent speech was induced in stuttering speakers by having them read aloud in unison (i.e., choral reading) with a recording.

A follow-up PET study of subsets of the same two groups was conducted while subjects imagined they were reading aloud (Ingham et al., 2000). Stuttering subjects were instructed to imagine they were stuttering in the solo condition and fluent in the choral condition. The patterns of activity that had occurred when each group read aloud were similar to those observed when speakers merely imagined they were reading. As such studies continue, a much better understanding of the brain-behavior substrates of stuttered and stutter-free or normal speech may be achieved.

Adults who have been stuttering most of their lives often develop various situational fears, social anxieties, lowered expectations, diminished self-esteem, and an array of escape and avoidance behaviors. Prior to initiating treatment, clinicians should obtain a thorough history of an adult's stuttering and prior treatment, including major current concerns, treatment expectations, and goals, and should assess attitudes, affective reactions and behaviors, and self-concepts that may require treatment. Analyses of samples recorded in various speaking situations document the type, frequency, duration, and overall severity of stuttering. Such information allows clinicians to select appropriate treatment strategies, track progress, and determine when treatment objectives have been achieved. Current treatment strategies focus either on modifying adults' affective, cognitive, and behavioral reactions to stuttering (e.g., Prins, 1993; Manning, 1996) or on learning speech production techniques (i.e., fluency training) to reduce or eliminate stuttering (e.g., Neilson and Andrews, 1993; Onslow, 1996). Most clinicians apparently prefer a combined strategy to manage the constellation of speech, affective, and cognitive symptoms commonly presented by adults who stutter. No well-controlled clinical trials of stuttering treatments have been reported, but some relapse in treatment gains is common. It is generally agreed, therefore, that complete, permanent recovery from chronic stuttering is rare when stuttering persists into adult life, regardless of the treatment employed. Consequently, local self-help groups of the National Stuttering Association, which promote sharing of experience, information, and support among members, have become an increasingly important component to successful, long-term management of stuttering in adults in the United States.

See also speech disfluency and stuttering in children.