(Stroke. 1999;30:1370-1379.)
© 1999 American Heart Association, Inc.
Original Contributions |
Improving Outcomes for Persons With Aphasia in Advanced Community-Based Treatment Programs
From LingraphiCARE America, Inc, Language Care Center, Palo Alto, Calif (L.B.A., R.D.S.); Department of Rehabilitation Medicine, Mills-Peninsula Hospital, San Mateo, Calif (L.B.A.); Language Care Center of Kansas City (Kan) (J.S.A.); and Kanza Group, Kansas City, Mo (J.S.A.).
Correspondence to Richard D. Steele, PhD, LingraphiCARE America, Inc, 3600 W Bayshore Rd, Suite 202, Palo Alto, CA 94303-4229. E-mail steele@cdr.stanford.edu or rsteele{at}aphasia.com
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose—Studies have yet to document that community-based aphasia treatment programs routinely produce results comparable or superior to published research protocols. We explore this issue here in an outcome study of individuals with aphasia enrolled in 2 community-based, comparably managed and equipped therapy programs, which use a specially designed computer-based tool that is employed therapeutically in adherence to an extensive, detailed, and formally trained patient care algorithm.
Methods—Patients (n=60) were assessed before and after treatment with standardized instruments at both the impairment and the disability levels. Pretreatment and posttreatment means were calculated and compared, with statistical significance of differences established with the use of 1-tailed matched t tests. One-way ANOVAs were used to analyze the comparability of patient performance changes among various subgroups, eg, patients in acute versus chronic stages of aphasia, patients by aphasia diagnostic type at start of care, patients by severity level at start of care, and patients by treatment location.
Results—Analysis shows that patients spanned a wide range of aphasia diagnostic types, impairment severity levels at start of care, and times after onset. Patients' mean performance scores improved significantly in response to treatment in all measures assessed at both the impairment level and the functional communication level. Mean overall improvements ranged from 6.6% to 19.8%, with statistical significance ranging from P=0.0006 to P<0.0001. ANOVAs revealed no significant differences between improvements in patients in the acute versus chronic stages of aphasia, between those at different impairment severity levels at start of care, between those treated at different locations, or, at the functional level, between those with different diagnostic types of aphasia at start of care.
Conclusions—–Measures of both language impairment and functional communication can be broadly, positively, and significantly influenced by therapy services that are delivered to persons with aphasia in these community-based programs. The significant improvements are shown to be available to individuals with chronic as well as acute aphasia and independent of diagnostic type of aphasia, impairment severity at start of care, or geographic program location.
Key Words: aphasia • rehabilitation • therapy, computer-assisted • treatment outcomes
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Approximately one third of stroke patients will suffer the disruptive, often devastating consequences of aphasia.1 The efficacy of aphasia treatment, including computer-based interventions, has been widely evaluated by individual studies,2 3 4 5 6 7 8 9 10 11 12 13 by expert opinion after literature review,14 15 and by meta-analysis.16 17 The general consensus is that aphasia therapy is helpful for improving specific measures of language function if delivered over a sufficient period of time with adequate intensity. Holland et al,14 for example, concluded in 1996 after thorough literature review that "considering this evidence collectively in its most conservative form, the conclusion can be drawn that people who become aphasic following a single, left-hemisphere thromboembolic stroke and who receive at least 3 hours of treatment each week for at least 5 months, regardless of the time post-onset of stroke, make significantly more improvement than people with aphasia who are not treated."
Such conclusions derive from results achieved primarily in the context of academic research; however, it is primarily in the community that individuals with aphasia must be identified, reached, and treated. There, given current healthcare constraints, the establishment and maintenance of programs for effective aphasia remediation are posing myriad new challenges. Nonetheless, the long-term viability of aphasia therapy depends on its ability to promote and improve functional outcomes in real-world settings of constraints and limitations. In light of this, and building on work reported elsewhere,18 19 20 we present relevant experiences from community-based aphasia therapy programs that use both a consistent, structured treatment methodology21 and standard pretreatment and posttreatment testing at both the impairment (speech-language test performance) and the disability (functional communication) levels.22
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Treatment Programs
The 2 community-based treatment programs participating in this study are located in different parts of the country and provide speech-language therapy services to adults for reimbursement. One of these is a freestanding, for-profit speech-language clinic in Palo Alto, Calif (PA site), and the other is an outpatient facility owned and operated by a for-profit hospital in Kansas City, Kan (KC site). Operationally, both programs are similarly organized, equipped, and managed. Both sites offer comparable, formally structured speech-language therapy.
Descriptions of these programs, organizationally and operationally,
appear elsewhere.18 20 21 They are distinguished by the
presence of several key components, including the following: (1) an
extensive and detailed patient care algorithm, which helps treating
clinicians specify therapeutic clinical pathways; (2) an online
database that captures, holds, and reports patient demographic,
diagnostic, assessment, treatment, and
response-to-treatment data; (3) a proprietary treatment technology
called the Lingraphica System, providing access to, among other things,
an extensive toolbox of specially designed, interactive multimodal
materials for use with and by patients; and (4) a formal training
program for the speech-language pathologists who must competently draw
on and integrate use of the preceding 3 components. Figure 1
shows representative
appearances and behaviors of selected icons from the treatment
technology and illustrates icon use in one type of therapy exercise,
namely, "icon plus spelling."
|
Subjects
Subjects in this study were drawn from individuals diagnosed with
aphasia and treated in one of the community-based treatment programs.
Referrals for treatment in these programs came from physicians,
hospitals, other community-based programs, speech-language
pathologists, friends, family, and self-referral. To qualify for
inclusion in this study, subjects had to meet 3 criteria: (1)
assignment to 1 of 8 aphasia diagnostic categories through
administration of the Western Aphasia Battery (WAB)23 ; (2)
completion of at least 1 month of therapy in the community-based
program; and (3) receipt before and after treatment of the language
subtests from the WAB. We did not require subjects to be either
right-handed or primarily English speaking.
Demographic, diagnostic, and treatment characteristics of
the subject sample are presented in Table 1. Subject language diagnoses spanned the
full spectrum of the 8 aphasia types from the WAB, with cases of
global, Wernicke's, Broca's, and anomic aphasia combined accounting
for 53 of the 60 cases (88.3%). Forty-six of the 60 subjects (76.7%)
were 
6 months after onset, placing them beyond the presumed period of
spontaneous recovery and into the period of chronic aphasia; the
remaining 14 subjects (23.3%) were still in the period of acute
aphasia (<6 months after onset) at start of care.24 25 26
For 53 of these 60 subjects (88.3%), Center-based treatment
represented resumption of therapy after discharge from 1 or
more previous courses of speech-language therapy elsewhere.
|
The subject sample reported on here accrued from the referral streams
of the 2 programs. Analysis of the fates of referrals to these
programs reveals a stepwise winnowing process, with patient numbers
diminishing at each step. For example, during the accrual period at the
PA site, a total of 258 patients were referred as potential candidates
for benefit. After speech-language evaluation, 185 of the 258 were
given a treatment diagnosis of one or another type of aphasia (as
opposed to other treatment diagnoses, such as voice disorders or
apraxia of speech). After resolution of any scheduling, transportation,
and support issues, 170 were enrolled in treatment. Of the enrollees,
143 received treatment for 30 days. Of those, 105 were assessed
before treatment with the WAB (others, particularly during the earlier
period of this program, received instead the Boston
Diagnostic Aphasia Examination,27 whose
results were reported in an earlier article).18 Of these
latter, 64 received posttreatment assessment with the WAB, with 52
receiving the relatively complete posttreatment assessment that allows
for the calculation of the Aphasia Quotient (AQ) of the WAB. The
current PA sample comprises the first 30 of this latter group
chronologically. Of the 12 cases without the AQ-required administration
of all first 4 WAB subtests (and who are not described in this
article), analysis shows that the Spontaneous Speech subtest
was omitted in 7 cases, the Auditory Verbal Comprehension subtest was
omitted in 8 cases, the Repetition subtest was omitted in 3 cases, and
the Naming subtest was omitted in 7 cases, in various overlapping
patterns. Two subtests not involved in the calculation of the AQ were
also omitted in some of these administrations: the Reading subtest in
13 instances and the Writing subtest in 26 cases.
Treatment
Patients participated in therapy with their treating clinicians in
individual, hour-long sessions. Table 1
presents
quantitative information on treatment for subjects in this sample. The
overall mean number of treatment sessions per patient was 41.7 (SD
24.1; range, 10 to 132).
In clinical sessions, therapists typically employed stimulus-response strategies in treatment activities, using stimuli from specified materials loaded on the treatment technology. For any particular patient, of a given aphasia diagnostic type, at a given level of severity, the patient care algorithm suggests clinical pathways through a sequence of therapeutic exercises that have been found in our experience to be beneficial to patients of the given type. During treatment, focus was invariably on improving patients' functional communication outside the clinic, as opposed to training for higher scores on discharge retesting. When patient responses in therapy sessions so indicate, batteries of exercises are loaded onto the patient's system as prescribed home practice. At home, patients are to complete the prescribed clinical exercises, and additionally they may pursue materials of their own choosing, explore semantic domains to review lexical items within, or find other activities that engage their interest. Analysis has shown that patients typically engage in such self-directed activities approximately 2 hours per day.18 Patients were discharged from treatment when any of the following occurred: (1) progress in functional communication reached a plateau as determined by the clinician; (2) funding became unavailable for continued therapy; or (3) intercurrent medical or other problems required discharge. Most patients were discharged because of the first condition, reaching a plateau in their functional communication.
Tests
Treatment program procedures specify administration of the
WAB23 and, more recently, the Communicative Effectiveness
Index (CETI)28 to persons with aphasia at start of care
and at discharge. The former provides assessment at the impairment
level, and the latter provides assessment at the disability
(functional) level.22
The WAB, in addition to assessing speech-language impairment overall, assigns patients to 1 of 8 aphasia diagnostic categories and also provides an overall quantitative metric of aphasic severity: the AQ, which ranges from 0 to 100. The WAB has been psychometrically characterized and found valid and reliable.29 Its 6 language subtests were administered in their entirety to this sample, except for 9 subjects in whom the Reading subtest was not completed (1 in the PA program, 8 in the KC program) and 23 subjects in whom the Writing subtest was not completed (8 in the PA program, 15 in the KC program). The reason most frequently given by clinicians for failure to complete all WAB subtests was lack of time.
The CETI was administered at start of care and at discharge to 13 PA patients and 16 KC patients of more recent enrollment. It provides ratings of functional performance of patients in important communicative activities of everyday life, as assessed by persons with opportunities for observing them frequently in relevant situations. The CETI consists of a 16 visual analog scale items assessing areas of functional communication in the patient's living environment. It is designed to be administered by caretakers and has been shown to be sensitive to change in communication behaviors. The CETI has been psychometrically characterized and found valid and reliable.28
All testing and rating were accomplished in standard ways, assessing subjects' unaided, natural-language performance, without the specialized treatment technology. Specifically, the WAB was given to subjects by a trained and licensed speech-language pathologist familiar with its contents and practiced in its administration and scoring. The CETI, in turn, was rated by a person who was close to the subject and who also was familiar with that person's communicative strengths and weaknesses in activities of normal everyday living. This rater was most commonly a spouse, sibling, or adult child of the subject, less commonly a close friend, neighbor, or caregiver. As a rule, patients had no access to test materials between start of care and discharge (on average >20 weeks apart) to minimize the likelihood of practice effects accounting for patient performance improvements.
Statistical Analysis
Data were analyzed with the use of Data Desk 6.0 software
on a Macintosh Quadra 840AV. Using raw WAB and CETI scores, we
calculated pretreatment and posttreatment means and compared them with
1-tailed matched t tests.30 WAB AQs were
also calculated for subjects and analyzed. When significant
differences were found with the use of matched t tests,
1-way ANOVAs were conducted to explore a possible further dependence on
additional parameters, such as aphasia
diagnostic category at start of care, impairment severity
of aphasia at start of care, program location, or patient assignment to
acute versus chronic aphasia. When the 1-way ANOVA revealed a
significant difference, post hoc analysis was conducted with
the Bonferroni test to identify underlying factors.31
Finally, the 
2 test was used to probe the
significance of the distribution of aphasia diagnostic
types among chronic patients after treatment compared with their
pretreatment distribution.30 Throughout, the level for
rejection of the null hypothesis was set at P=0.05. Where
achieved, statistical significance is denoted below by an asterisk
(*).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Table 2
shows pretreatment
and posttreatment mean scores (raw) for all assessed areas, with
associated observed t values and P values for the
differences of means. For the WAB language subtests, absolute percent
mean improvements were as follows: Spontaneous Speech, +12.0%*
(P<0.0001); Auditory Verbal Comprehension, +6.8%*
(P<0.0001); Repetition, +6.6%* (P<0.0001);
Naming, +8.7%* (P<0.0001); Reading, +7.4%*
(P=0.0004); Writing, +8.8%* (P=0.0006); and AQ,
+9.1%* (P<0.0001). For the functional communication
items from the CETI, absolute percent improvement overall was +19.8%*
(P<0.0001).
|
Figure 2 graphically displays the AQ
changes of the 60 subjects versus the times after onset. Of the 60
subjects in the study, 55 (91.7%) showed a higher AQ score after
treatment, while 5 (8.3%) showed a lower AQ score. Figure 2
shows that these improvements are found across severity levels at start
of care and throughout the range of times spanned after onset in this
sample.
|
Figure 3 graphically displays the CETI
overall changes available for 29 of the patients versus the times after
onset when the changes occurred. Analysis of CETI score changes
before and after treatment showed improvement in all 29 patients
tested. Figure 3 reveals that all patients improved, regardless
either of the initial level of severity or of the time after onset when
they received the treatment.
|
Fourteen patients were still in the acute stage of recovery (<6 months
after onset) at start of care, while the remaining 46 were in the stage
of chronic aphasia (6 months after onset). Mean AQ improvement in
patients in the acute stage was +8.0* points (SD 10.9,
P=0.017), while in patients in the chronic stage mean AQ
improvement was +9.4* points (SD 8.2, P<0.0001). One-way
ANOVA showed no significant difference between these groups
(F1,58=0.28, P=0.597) with respect to
AQ improvement.
Analysis of mean pretreatment and posttreatment CETI overall scores revealed significant improvement in both the acute and chronic groups. Among patients in the acute group (n=10), the CETI improvement was +23.3* (SD 11.2, P=0.0001), while in patients in the chronic group (n=19), the improvement was +17.9* (SD 13.1, P<0.0001). One-way ANOVA reveals no significant difference between these groups (F1,27=1.24, P=0.275) with respect to CETI overall improvement.
Figure 4 shows AQ changes in the sample
as a function of AQ severity at start of care. Twenty-one subjects fell
into the lowest quarter (AQ <25), and in response to treatment their
mean AQ score improved +6.7* (SD 6.1, P=0.0001). Eighteen
subjects fell in the low-mid quarter (AQ 25 to 50), and in response to
treatment their mean AQ score improved +12.6* (SD 12.1,
P=0.0004). Nine subjects fell into the high-mid quarter (AQ
50 to 75), and in response to treatment their mean AQ score improved
+12.2* (SD 7.5, P=0.0013). The remaining 12 subjects fell
into the highest quarter (AQ >75), and in response to treatment their
mean AQ score improved +5.8* (SD 5.6, P=0.0041). One-way
ANOVA revealed a trend toward significance in overall AQ responses to
treatment among these groups (F3,56=2.59,
P=0.062). Figure 4 suggests that the trend toward
significance may reflect floor and ceiling effects apparently
manifesting in the lowest and highest quarters, in contrast to the
middle 2 quarters.
|
When we compare the programs in different geographic locations, the 30 patients treated at the PA site showed a mean AQ improvement after treatment of +9.6* (SD 10.1, P=0.0001), while the 30 patients treated at the KC site posted a mean AQ improvement after treatment of +8.5* (SD 7.5, P=0.0001). One-way ANOVA of these AQ changes showed no significant difference between improvements at the 2 locations (F1,58=0.26, P=0.61). In functional communication, the 13 CETI-assessed patients at the PA site showed a mean overall improvement after treatment of +24.3* (SD 13.6, P=0.0001), while the 16 CETI-assessed patients at the KC site showed overall improvement of +16.1* (SD 10.5, P=0.0001). One-way ANOVA of these CETI score changes revealed a trend toward significance at the 2 locations (F1,27=3.41, P=0.076).
One-way ANOVAs showed no significant dependence of treatment parameters on aphasia diagnostic category at start of care. In particular, 1-way ANOVA of duration of treatment by aphasia diagnostic category (F7,52=1.88, P=0.09), frequency of treatment by type of aphasia (F7,52=1.19, P=0.33), and total number of treatment sessions by aphasia diagnosis (F7,52=1.33, P=0.26) revealed no significant differences among the various aphasia categories at start of care.
Figure 5 graphically depicts, in dot plot
format, AQ responses to treatment of all 60 subjects, grouped by their
aphasia diagnostic categories at start of care. All
categories showed overall improvement in response to treatment, and in
the majority of categories in which n is >1, the improvements were
significant. Thus, patients at start of care with Broca's aphasia
(n=21) showed a mean AQ improvement of +13.9* points
(SD=10.0, P<0.0001), those with anomic aphasia
(n=13) improved +5.8* points (SD=6.4, P=0.007),
those with global aphasia (n=11) improved +6.2* points
(SD=3.9, P=0.0004), those with Wernicke's aphasia
(n=8) improved +3.6 points (SD=10.1, P=0.347),
those with conduction aphasia (n=3) improved +16.0* points
(SD=5.5, P=0.038), and those with transcortical motor
aphasia (n=2) improved +13.1 points (SD=2.7,
P=0.092). One-way ANOVA of AQ changes by aphasia type at
start of care revealed the presence of significant differences over the
spectrum of aphasia categories (F7,52=2.58,
P=0.023). Post hoc analysis using the Bonferroni
test showed that the overall significance resulted from trends toward
significance in 2 pairwise category comparisons: Broca's (+13.9*)
versus Wernicke's (+3.6), for which P=0.053, and Broca's
(+13.9*) versus anomic (+5.8*), for which P=0.096. At the
functional level as assessed by the CETI overall category, there were
no significant differences among the aphasia categories available for
analysis ( F6,22=0.41,
P=0.87).
|
Table 3 shows the evolution in aphasia
diagnostic categories in response to treatment among
patients in the chronic stage (n=46). It is based on patients'
WAB assignments to aphasia categories before and after treatment;
however, to eliminate borderline cases, we required that reassignment
to a different diagnostic category posttreatment be
accompanied by a concurrent AQ shift, upward or downward, of 5.0
points from the pretreatment AQ score. By these criteria, 29 of the 46
patients in the chronic phase showed no change of
diagnostic category after treatment, while 17 showed a
change of aphasia type to a less severe diagnostic category
(eg, from global to Broca's aphasia) accompanied by a rise in AQ
score. No patients in the chronic phase showed a change to a more
severe diagnostic category that was accompanied by an AQ
drop of 5.0 points. This pattern is fundamentally different from that
documented in the 1977 report of Kertesz and McCabe,32 in
which they followed 22 untreated persons in chronic aphasia
longitudinally using the WAB. In that study, patients on the whole did
not change AQ score significantly in a year's time, and only 1 patient
changed diagnostic category (to a more severe category:
from Broca's to global aphasia). A 2 test for
changes among more severe, same, or less severe categories, with the
published data of Kertesz and McCabe32 as
representative of the expected pattern, shows the
present changes among 46 patients in the chronic stage to less
severe diagnostic categories to be significant
(22=55.6,
P<0.001*).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
There is an extensive, often contradictory, literature on aphasia recovery and the impact of therapy. Since study methodologies vary, it can be difficult to compare individual studies and draw meaningful conclusions regarding the relative merits of any particular treatment approach. Depending on which study is cited, improvement may or may not occur in a given population. Wertz et al6 elegantly demonstrated Porch Index of Communicative Abilities score improvements of up to 18.6 percentile points in intensively treated (8 to 10 h/wk) acute stroke patients, whereas Lincoln et al4 were not able to show treatment effects for a group of 104 patients treated with less intensive (2 h/wk) therapy. Additionally, community-based studies whose main focus was not aphasia but that retrospectively inspected this parameter (and possibly more closely reflected the natural course of individuals with aphasia in contexts in which speech-language therapy is not always a primary focus) have reported little benefit from speech therapy in an aphasic stroke population.1
Many such studies are open to criticism on grounds of flawed design and/or execution.33 34 Even when this is not the case, aphasia therapies are idiosyncratic and frequently highly individualized by the treating clinician to conform to a particular patient's unique combination of deficits and residual capabilities.16 17 Because of such treatment variability, few therapy approaches are sufficiently defined and codified to lend themselves to critical evaluation. In this regard, the positive assessment recently accorded to melodic intonation therapy represents a laudable if instructive exception.35 There is general agreement, however, that, on the condition that treatment intensity and duration exceed certain minimal thresholds, beneficial therapeutic effects can be demonstrated.14 16 17
In this context, we believed it important to investigate the effects of a community-based aphasia treatment program that provided therapy of known intensity using a structured and consistent therapeutic model and standardized measurement tools, thus diminishing effects attributable to varying treatment criteria, therapeutic approaches, and evaluation methods. We also believed it important to demonstrate our results in the context of the real-world financial constraints of current healthcare settings.
The data presented here corroborate and extend the pattern reported earlier of significant improvements in persons with aphasia treated through these programs, across a broad range of aphasia types, severities, and times after onset.18 19 20 21 Moreover, with a sample size here considerably >30, the means and SDs herein reported represent, by the central limit theorem, fair approximations to the underlying population means for outcomes from these treatment programs.36 ANOVA, furthermore, shows that the improvements are replicable across geographic sites. Of note, significance is herewith established for every language subtest of the WAB, for the WAB AQ, and for functional communication overall as assessed by the CETI. Lomas et al28 found a mean gain of +11.4% in CETIs administered 6 weeks apart in recovering persons with aphasia undergoing speech/language therapy. This diverges from the current report of +19.8% gain, but neither the patient populations from which the samples were drawn nor the test-retest periods are comparable. In particular, the present patient sample was, in its majority, well into the chronic stage of aphasia and comprised mainly patients discharged from previous courses of speech-language therapy elsewhere.
Overall recovery from aphasia has been related to a variety of
variables, including lesion size, aphasia type, and, perhaps most
importantly, initial severity.37 Recovery furthermore does
not follow a random pattern but tends to move along more or less
predictable paths. Kertesz and McCabe32 noted that persons
with global aphasia have the poorest prognosis, while anomic,
transcortical, and conduction aphasics have the best. This formulation
is consonant with the present findings. For example, Figure 5 shows that the present sample of persons with global
aphasia at start of care had, on discharge, the poorest AQ profile of
any diagnostic category, whether measured by lowest AQ
score or highest AQ score; those with anomic, transcortical, and
conduction aphasia had the highest AQ profile for these same
measures. However, it is worth noting that the analogous
statements hold for these same groups compared at start of care as
well: those with global aphasia are ranked lowest, while those with
anomic, transcortical, and conduction aphasia are ranked
highest. What is most striking about Figure 5 in this
regard is, in fact, the stability of pretreatment versus posttreatment
order rankings of AQ means across diagnostic types, in the
presence of such widespread and significant patient improvements.
Patients emerge from these treatment programs with improvements spread
relatively evenhandedly and equably across the diagnostic
spectrum rather than concentrated in particular diagnostic
categories. Interestingly, this is the improvement pattern that one
would expect when the mechanisms underlying the remediation processes
are of a general cerebral character rather than tied to more specific
capabilities—psycholinguistic or other—that may be differentially
compromised in the various aphasia diagnostic
categories.
In addition, a relatively consistent pattern within aphasia
diagnostic types is also discernible when data from Figure 5 are analyzed. This pattern includes the following, in
comparing posttreatment scores with pretreatment scores of the same
category: (1) a relatively small rise in the lowest AQ scores, with
mean low score improvements in Figure 5 of +0.98; (2) a
much larger rise in the highest AQ scores, with mean high score
improvements in Figure 5 of +11.23; and (3) a notable extension
of range between the highest and lowest scores, with a mean range
extension in Figure 5 of 34.3% (from mean range=32.70 before
treatment to mean range=43.93 after treatment). Such a pattern suggests
a situation in which, within each diagnostic category,
almost all patients improve somewhat, with the least impaired patients
within categories more advantageously poised to derive additional
benefits. In this regard, the ANOVAs conducted earlier in this article
serve to underscore the fact that many of the traditional prognostic
indicators, such as time after onset, severity at start of care, or
aphasia diagnostic category, are in fact not those that
differentiate between patients showing greater and lesser improvements
after this treatment program. If this is borne out by further studies,
the question of the relevant prognostic indicators for attainment of
benefit from this treatment program must be considered.
To complicate matters, whether and how the "canonical" prognostic indicators correlate with response to therapy is even less well understood in chronic aphasia than in acute aphasia: fewer studies have been conducted, fewer subjects involved, and fewer hypotheses tested. In our chronic population, we were able to show a significant evolution to less severe aphasia diagnostic categories after treatment. Tantalizingly, these upward paths broadly recapitulate, diagnostically, natural courses of recovery from aphasia during the period of spontaneous recovery. Persons with global aphasia in our sample who improved all evolved to Broca's aphasia; persons with Broca's aphasia evolved to either conduction or transcortical aphasia; transcortical aphasia became anomic aphasia; and, when such evolution took place, persons with anomic aphasia at start of care performed within normal limits after treatment. These changes cannot be fully understood in terms of the expected underlying structural lesions of classic language areas38 ; rather, they support the notion of more widely distributed language mechanism underlying latent or residual language capacity.15 39 40
Patient sample bias, incomplete test administration, and choice of test instruments are potential sources of error in these results. Patients arrived at these programs through referrals, which means that this is not a random sample of persons with aphasia but rather one of patients who, in the judgment of the referral sources, would be able to participate in therapy and make significant functional gains. Such a sample will not include those individuals with aphasia—too low functioning, too high functioning, or other—who were deemed not treatable or not worth the effort of treating by referrers. In a study of 335 acute stroke patients with aphasia, Pedersen et al1 reported 52% as severe, 16% as moderate, and 32% as mild on the basis of a 3-point scale. Twenty-one of the 60 patients (35%) in our study had AQs <25, and 9 of the 60 (15%) had scores in the top quartile, broadly consistent with the findings of Pederson et al. Although the 2 groups are not entirely comparable because our sample included persons with both acute and chronic aphasia, this suggests that our patients were fairly representative of the natural distribution of aphasia severity and therefore mirrored the typical caseload of community aphasia programs.
Some uncertainty in interpreting these results also arises from the fact that not all tests were administered to all patients. Absence of subtest scores for whatever reasons—perception of no improvement by treating clinician, patient at ceiling on pretest, lack of time for testing, or others—introduces distortions of unknown type, direction, and magnitude. In addition, our results are viewed through a specific set of test instruments. We elected to use one language measure exclusively, the WAB, in the interest of consistency and comparability. The WAB can clearly differentiate between normal and aphasic language and has demonstrated good test-retest reliability23 29 but may not have been sensitive to small changes in language performance, which could have been detected by other more specific or specialized instruments. Similarly, a number of functional assessment instruments exist, each with a differing perspective on functional communication. We chose the CETI because of its demonstrated sensitivity to change and because it investigates the caregiver/family perspective of communicative function, which may not always coincide with the clinician's opinion.41
A large majority of patients treated in these specially equipped community-based aphasia treatment programs showed significant improvements in both language impairment and communicative function, regardless of time after onset, severity at start of care, or type of aphasia. In one sense, these findings should not surprise: indeed, they are in accord with the emerging view that patients in both acute and chronic aphasia may be candidates for significant improvement through resumption of treatment.14 16 17 The corroborating evidence that is offered here comes specifically from an outcome study, the proper topics of which are the existence, direction, magnitude, and statistical significance of changes between 2 points in time. With these particular focuses, outcome studies represent a variety of research that is assuming ever-increasing importance in the environment of managed care.42 This is because superior clinical outcomes, when available at competitive costs through replicable programs, are attractive to payers. Outcome studies also have their limitations. They do not, for example, allow the determination of conclusions regarding either absolute or comparative efficacy or for any attribution of causality. Answers to these latter questions come from properly designed, prospective, randomized, scientifically controlled research. It is hoped, in fact, that the present report will spur such research activity to investigate these latter questions. At present, for example, whether one specific element of the program or a synergy of components contributes more to these improved outcomes remains an open issue.
If the present findings warrant special interest, it would have to center on questions of why and how such changes—widespread, beneficial, and large—are taking place. These improvements strike not by ability to be documented but by their ubiquity, magnitude, generalization, and robustness. Aphasiologists have sometimes speculated about latent capacities for additional speech-language improvements that could be stimulated in late aphasia.43 In preceding decades, such hypotheses could only remain speculation. More recently, brain imaging techniques have shown value in revealing significant relationships between language performance and brain functioning.44 It is submitted that such newer study techniques must complement the more traditional, behaviorally based study methodologies, such as efficacy and effectiveness trials, to fully understand these outcomes, if we are to learn how best to proceed to improve the tools, materials, and methods for aphasia rehabilitation in community-based clinical practice.
|
Acknowledgments |
|---|
This study was supported by LingraphiCARE America, Inc, Bethany Medical Center of Kansas City, Kan, and Mills-Peninsula Hospital of San Mateo, Calif. The authors would like to acknowledge the key contributions of clinicians at the community-based treatment programs whose enthusiasm, commitment, and professional expertise have helped the patients in this study so significantly. We gratefully acknowledge the valuable comments and helpful suggestions of our reviewers. We would also like to especially thank the patients and their families for sharing with us their experiences, insights, and hopes.
|
Footnotes |
|---|
Drs Aftonomos and Appelbaum are part-time Medical Directors at Language Care Center Programs owned or managed by LingraphiCARE America, Inc, and Dr Steele is an employee of LingraphiCARE America, Inc.
Received February 9, 1999; revision received April 7, 1999; accepted April 7, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
- Pedersen PM, Jørgensen HS, Nakayama H, Raaschou HO, Olsen TS. Aphasia in acute stroke: incidence, determinants, and recovery. Ann Neurol. 1995;38:659–666.[Medline] [Order article via Infotrieve]
- Basso A, Capitani E, Vignolo LA. Influence of rehabilitation on language skills in aphasic patients: a controlled study. Arch Neurol. 1979;36:190–196.[Abstract]
-
Aten J, Caliguiri M, Holland AL. The efficacy of
functional communication therapy for chronic aphasic patients. J
Speech Hear Disord. 1982;47:93–96.
[Abstract/Free Full Text] - Lincoln NB, Mulley GP, Jones AC, McGuirk E, Lendrem W, Mitchell JRA. Effectiveness of speech therapy for aphasic stroke patients: a randomised controlled trial. Lancet. 1984;324:1197–1200.
- Loverso FL, Prescott TE, Selinger M, Wheeler KM, Smith R. The application of microcomputers for the treatment of aphasic adults. In: Brookshire RH, ed. Clinical Aphasiology. Vol 15. Minneapolis, Minn: BRK Publishers; 1985:189–195.
- Wertz RT, Weiss DG, Aten JL, Brookshire RH, Garcia-Buñuel L, Holland AL, Krutzke JF, LaPointe LL, Milianti FJ, Brannegan R, Greenbaum H, Marshall RC, Vogel D, Carter J, Barnes NS, Goodman R. Comparison of clinic, home, and deferred language treatment for aphasia: a Veterans Administration Cooperative Study. Arch Neurol. 1986;43:653–658.[Abstract]
- Katz RC. Efficacy of aphasia treatment using microcomputers: clinical forum. Aphasiology. 1987;1:141–175.
- Steele RD, Weinrich M, Kleczewska MK, Carlson GS, Wertz RT. Evaluating performance of severely aphasic patients on a computer-aided visual communication system. In: Brookshire RH, ed. Clinical Aphasiology. Vol 17. Minneapolis, Minn: BRK Publishers; 1987:46–53.
- Steele RD, Weinrich M, Wertz RT, Kleczewska MK, Carlson GS. Computer-based visual communication in aphasia. Neuropsychologia. 1989;27:409–426.[Medline] [Order article via Infotrieve]
- Weinrich M, McCall D, Weber D, Thomas K, Thornburg L. Training on an iconic communication system for severe aphasia can improve natural language production. Aphasiology. 1995;9:343–364.
- Brookshire RH. Introduction to Neurogenic Communication Disorders. 5th ed. St Louis, Mo: Mosby; 1997.
-
Katz RC, Wertz RT. The efficacy of computer-provided
reading treatment for chronic aphasia adults. J Speech Lang Hear
Res. 1997;40:493–507.
[Abstract/Free Full Text] -
Naeser MA, Baker EH, Palumbo CL, Nicholas M, Alexander
MP, Samaraweera R, Prete MN, Hodge SM, Weissman T. Lesion site
patterns in severe, nonverbal aphasia to predict outcome with a
computer-assisted treatment program. Arch Neurol. 1998;55:1438–1448.
[Abstract/Free Full Text] - Holland AL, Fromm DS, DeRuyter F, Stein M. Treatment efficacy: aphasia. J Speech Hear Res. 1996;39:S27–S36.
-
Albert ML. Treatment of aphasia. Arch
Neurol. 1998;55:1417–1419.
[Abstract/Free Full Text] - Robey RR. The efficacy of treatment for aphasic persons: a meta-analysis. Brain Lang. 1994;47:582–608.[Medline] [Order article via Infotrieve]
-
Robey RR. A meta-analysis of clinical outcomes
in the treatment of aphasia. J Speech Lang Hear Res. 1998;41:172–187.
[Abstract/Free Full Text] - Aftonomos LB, Steele RD, Wertz RT. Promoting recovery in chronic aphasia with an interactive technology. Arch Phys Med Rehabil. 1997;78:841–846.[Medline] [Order article via Infotrieve]
- Appelbaum JS, Steele RD, Aftonomos LB, Harris VM. Improving outcomes for aphasic patients in advanced, Center-based treatment programs. Paper presented at: 1998 Clinical Aphasiology Conference; June 14–17, 1998; Asheville, NC.
- Appelbaum JS, Harris VM, Chadwell AL. Maximizing functional outcomes in communication for persons with aphasia. Paper presented at: Tenth Annual Stroke Rehabilitation Conference; October 15–16, 1998; Boston, Mass.
- Steele RD, Aftonomos LB, Appelbaum JS, Harris VM. Language Care CenterSM Treatment Programs for promoting recovery in aphasia: tools, materials, methods, and outcomes. Aphasiology. In press.
- World Health Organization. International Classification of Impairments, Disabilities, and Handicaps. Geneva, Switzerland: World Health Organization; 1980.
- Kertesz A. Western Aphasia Battery. New York, NY: Harcourt Brace Jovanovich; 1982.
- Butfield E, Zangwill O. Re-education in aphasia: a review of 70 cases. J Neurol Neurosurg Psychiatry. 1946;9:75–79.
- Godfrey CM, Douglass E. The recovery process in aphasia. Can Med Assoc J. 1959;80:618–624.
- Luria AR. Restoration of Function After Brain Injury. New York, NY: Macmillan; 1963.
- Goodglass H, Kaplan E. The Assessment of Aphasia and Related Disorders. 2nd ed. Philadelphia, Pa: Lea & Febiger; 1983.
-
Lomas J, Pickard L, Bester S, Elbard H, Finlayson A.
Zoghaib C. The Communicative Effectiveness Index: development and
psychometric evaluation of a functional measure for adult
aphasia. J Speech Hear Disord. 1989;54:113–124.
[Abstract/Free Full Text] - Shewan CM, Kertesz A. Reliability and validity characteristics of the Western Aphasia Battery (WAB). J Speech Hear Disord. 1984;45:308–324.
- Hatch E, Farhady H. Research Design and Statistics for Applied Linguistics. Rowley, Mass: Newbury House Publishers; 1982.
- Velleman PF. DataDesk Version 6.0. Ithaca, NY: Data Description, Inc; 1996.
-
Kertesz A, McCabe P. Recovery patterns and prognosis in
aphasia. Brain. 1977;100:1–18.
[Free Full Text] - Wertz RT, Deal JL, Holland AL, Kurtzke JF, Weiss DG. Comments on an uncontrolled aphasia no treatment trial. ASHA. 1986;38:31. Letter.
- Wertz RT. Aphasia in acute stroke: incidence, determinants, and recovery. Ann Neurol. 1996;40:129–130.[Medline] [Order article via Infotrieve]
- Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Assessment: melodic intonation therapy. Neurology. 1994;44:566–568.
- Downie NM, Heath RW. Basic Statistical Methods. 3rd ed. New York, NY: Harper & Row; 1970.
- Kertesz A. What do we learn from recovery from aphasia? In: Waxman SG, ed. Advances in Neurology, Vol. 47: Functional Recovery in Neurological Disease. New York, NY: Raven Press; 1988:277–292.
- Damasio AR. Aphasia: review article, medical progress. N Engl J Med. 1992;326:531–539.[Medline] [Order article via Infotrieve]
- Weiller C, Isensee C, Rijntjes M, Huber W, Müller S, Bier D, Dutschka K, Woods RP, Noth J, Diener HC. Recovery from Wernicke's aphasia: a positron emission tomographic study. Ann Neurol. 1995;37:723–732.[Medline] [Order article via Infotrieve]
- Damasio H, Grabowski TJ, Tranel D, Hichwa RD, Damasio AR. A neural basis for lexical retrieval. Nature. 1996;380:499–505.[Medline] [Order article via Infotrieve]
- Lomas J, Pickard L, Mohide A. Patient versus clinician item generation for quality-of-life measures: the case of language-disabled adults. Med Care. 1987;25:764–769.[Medline] [Order article via Infotrieve]
- Ellwood PM. Outcome management: a technology of patient experience. N Engl J Med. 1988;318:1549–1556.[Medline] [Order article via Infotrieve]
- Darley FL. Aphasia. Philadelphia, Pa: Saunders; 1982.
-
Desmond JE, Sum JM, Wanger AD, Demb JB, Shear PK,
Glover GH, Gabriele JDE, Morrell MJ. Functional MRI measurement of
language lateralization in Wada-tested patients. Brain. 1995;118:1411–1419.
[Abstract/Free Full Text]
This article has been cited by other articles:
 |
|
 |
|
  S. S. Bullain, L. S. Chriki, and T. A. Stern Aphasia: Associated Disturbances in Affect, Behavior, and Cognition in the Setting of Speech and Language Difficulties Psychosomatics, June 1, 2007; 48(3): 258 - 264. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Meinzer, D. Djundja, G. Barthel, T. Elbert, and B. Rockstroh Long-Term Stability of Improved Language Functions in Chronic Aphasia After Constraint-Induced Aphasia Therapy Stroke, July 1, 2005; 36(7): 1462 - 1466. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Pulvermuller, B. Neininger, T. Elbert, B. Mohr, B. Rockstroh, P. Koebbel, and E. Taub Constraint-Induced Therapy of Chronic Aphasia After Stroke Stroke, July 1, 2001; 32(7): 1621 - 1626. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||