Martin A. Young
D epartment of Speech Pathology and Audiology, Illinois State University
A critical review of research and opinion concerning age of onset, prevalence, a nd
recovery from stuttering indicates some inconsistency among findings: the reported
recovery rates may be too high, although all data sources hadlimitations.
Recent research reported by Sheehan and Martyn (1970) and Cooper (1972)
suggests that a large proportion of individuals (80%) who once considered
themselves stutterers no longer do so, and recovery from stuttering may occur
without the intervention of speech therapy. At best, these data have been
interpreted to indicate that speech therapy for school-age stutterers had no
effecton eventual recovery. At worst, portions of the data revealed that public
school speech therapy was negatively associated with recovery from stuttering.
High spontaneous recovery rates for stuttering, if valid, clearly suggest a
reevaluation of current therapy approaches. In terms of stuttering theory,
high recovery rates would be cause for reconsideration of both medical and
behavioralmodels as possible explanations for stuttering behavior.
Since recovery rates for stuttering are usually based on verbal reports of past
events rather than on direct observation, it is i m p o r t a n t that these data be
verified by comparison with other information. Prevalence and onset data
offer an o p p o r t u n i t y for such comparisons. T o place the literature review that
makes up themajor portion of this paper into a helpful perspective, some
conclusions from that literature may be summarized as follows: (1) onset of
stuttering occurs in childhood with essentially no new onsets after nine years
of age, (2) prevalencO of stuttering is 0.7% in a school-age population,
(3) recovery from stuttering occurs in 80% of young adults who stutter, and
(4) prevalence of stuttering is0.7% in a young adult population. Presumably
one or more or all of these four estimates must contain substantial error.
aPrevalence is usually defined as the number of cases of a disease existing in a population
at any given time. Incidence is generally defined as the number of cases of a disease a ppearing
per unit of population within a time interval. Clearly, incidence data for the problemo f
stuttering are not available.
Downloaded from jshd.asha.org on September 21, 2010
JOURNAL OF SPEECH A N D HEARING DISORDERS
I f 0.7% of school children are said to stutter, 80% are said to recover from
stuttering, and no new cases appear after nine years of age to replace those
who recover, then a prevalence rate of 0.7% in young adults is too high.
Ifthe 80% recovery rate is correct, then the prevalence of stuttering in young
adults should be something like 0.15%. If the 0.7% prevalence rate in young
adults and the 80% recovery rate are accurate, then the prevalence of stuttering in school children should be closer to 3.5%, which is much higher than
any estimate reported in the literature. Finally, with a prevalence estimate
of 0.7% forboth children and adults, the 80% recovery rate might be considered to represent some phenomena having little to do with recovery from
the problem of stuttering as it is generally known.
T h e rest of this paper will be concerned with a more detailed evaluation
of data sources and a search for possible biases that might reduce the apparent
discrepancies. My bias should be made clear: the recoveryrates are too high.
A critical examination of each data source, however, reveals much possibility
for error, and for this reason no opinion should be fixed at this point.
THE AGE OF ONSET
A re there large numbers of new onsets of stuttering after age nine to replace
those children who are said to have recovered from stuttering? Despite excellent
agreement to the...