Dysphagia

DOI 10.1007/s00455-012-9409-x

ORIGINAL ARTICLE

Validation and Application of the M.D. Anderson Dysphagia

Inventory in Patients Treated for Head and Neck Cancer in Brazil
Renata Lı́gia Vieira Guedes • Elisabete Carrara-de Angelis •

Amy Y. Chen • Luiz Paulo Kowalski •

José Guilherme Vartanian

Received: 2 August 2011 / Accepted: 3 May 2012

Ó Springer Science+Business Media, LLC 2012

Abstract Analysis of quality of life (QOL) has revealed overall QOL. In conclusion, the present study validates the
that preservation of swallowing, speech, and breathing adaptation of the MDADI to the Brazilian Portuguese
functions has a direct impact on QOL and that these language and provides another tool to evaluate the impact
functions are important patient-reported outcomes. The of dysphagia on the QOL of head and neck cancer patients.
purposes of this study were to adapt and culturally validate
the M.D. Anderson Dysphagia Inventory (MDADI) to the Keywords Quality of life
Deglutition
Deglutition
Brazilian Portuguese language and to evaluate QOL related disorders
Head and neck neoplasms
Questionnaires
to dysphagia in patients treated for head and neck cancer.
This was a cross-sectional study that included 72 adult
patients with a mean age of 63 years who were treated for Dysphagia is a common complication of head and neck
head and neck cancer. Construct validity and reliability squamous cell carcinomas that causes functional and social
analyses were performed through the comparison of the limitations, nutritional deficiencies, mood disorders, and
MDADI with three other health-related QOL question- worsening quality of life (QOL) in patients affected by and
naires administered at the time of enrollment and MDADI treated for the disease. Characterization of this symptom is
application 2 weeks thereafter, respectively. Reliability important in caring for these patients because more than
was established by assuring both internal consistency half the patients with head and neck cancer have some
(Cronbach’s a) and test–retest reliability (intraclass corre- degree of dysphagia during treatment [1–3].
lation coefficient, ICC). Test–retest reliability for the total Classically, outcome measures for cancer treatment
score in the MDADI had an ICC greater than 0.795 were based on medical eradication of the disease and dis-
(p \ 0.001). The MDADI had significant statistical corre- ease-free survival rates. Survival rates have improved as
lations with the other questionnaires. Patients treated for more aggressive and effective treatments have become
head and neck cancer had a mean total score of 83 on the available; therefore, global QOL in patients with cancer
MDADI, which is indicative of minimal limitation in has been incorporated as an important outcome measure
[1–3]. Analysis of QOL has revealed that the preservation
of swallowing, speech, and breathing functions has a direct
R. L. V. Guedes
E. C. Angelis
Voice, Speech and Swallowing Rehabilitation Department, impact on QOL and that these functions are important
A. C. Camargo Hospital, São Paulo, Brazil patient-oriented outcomes [3, 4].
The M.D. Anderson Dysphagia Inventory (MDADI) is a
A. Y. Chen
self-administered and validated inventory developed spe-
Otolaryngology Head and Neck Surgery Department,
Emory University Hospital, Atlanta, GA, USA cifically to evaluate the impact of dysphagia on the QOL of
English-speaking patients who undergo treatment for head
L. P. Kowalski
J. G. Vartanian (&) and neck cancer [5]. The MDADI is composed of 20
Head and Neck Surgery and Otorhinolaryngology Department,
questions divided in four domains: global, physical, func-
A. C. Camargo Hospital, Rua Professor Antonio Prudente, 211,
Liberdade, São Paulo, SP 01509-900, Brazil tional, and emotional. The MDADI is scored from 0 to 100,
e-mail: jgvartanian@uol.com.br with lower scores indicating a greater impact of dysphagia

123
 R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

on the patient’s QOL. The MDADI has been compared to and all participants signed a consent form approved by the
the swallowing-related QOL (SWAL-QOL) questionnaire, Institutional Ethics Committee.
which is a general set of measurements of QOL, but the
MDADI is a more specific and easily applied tool for Data Collection
evaluating QOL in head and neck cancer patients. Fur-
thermore, it was shown to be less distressing and, therefore, All participants were asked to complete a packet of self-
facilitated patient adherence and accurate self-administra- administered questionnaires during the routine outpatient
tion [6]. clinic visit; they also received another MDADI questionnaire
In Brazil, the incidence of head and neck cancer is and were told to return it within 15 days by mail. The 15-day
estimated to be nearly 14,000 new cases/year [7], which is interval was chosen to measure test–retest reliability because
a heavy burden on the country’s health-care system. A this interval was thought to be sufficient time to prevent
specific tool designed to evaluate QOL related to dysphagia patients from remembering their responses to the first
in head and neck cancer is not available in Portuguese, administration of the scale but not enough time to allow
which is the language spoken in Brazil. Therefore, the clinically meaningful change to occur. The packet included
purposes of the present study were to adapt and validate the following questionnaires: the Brazilian Portuguese ver-
culturally the MDADI to Brazilian Portuguese and to sion of the MDADI, the Brazilian Portuguese validated form
evaluate QOL related to dysphagia in patients treated for of the University of Washington Quality of Life Question-
head and neck cancer at A. C. Camargo Hospital, which is naire (UW-QOL) [11], the Brazilian Portuguese version of
a major referral center for cancer treatment in Brazil. the Swallowing-related Quality of Life Questionnaire
(SWAL-QOL) [12–15], and the Hospital Anxiety and
Depression Scale (HAD) [16]. We decided to use these
Materials and Methods questionnaires, different from the original MDADI study,
because the PSS-HN is not validated in the Brazilian Por-
Translation and Cultural Adaptation tuguese language, and the UW-QOL questionnaire was
designed specifically for head and neck cancer patients,
The first step in this study was translation and cultural which could provide a more significant construct validity as
adaptation of the MDADI questionnaire to the Brazilian opposed to the generic survey instrument SF-36. The charts
Portuguese language. We followed internationally accepted of enrolled subjects were reviewed and demographic, tumor,
guidelines [8–10]. Two bicultural experts translated the and treatment data were collected.
original English version of the MDADI to Portuguese. A
third bicultural person compared the two versions, and an Survey Instruments Scoring
iterative consensus was reached. The consensus version of
the Brazilian Portuguese translation was sent to two addi- The MDADI consists of 20 questions, subdivided into one
tional bicultural experts, who performed a similar back- global question that assesses overall QOL aspects related to
translation process (from Brazilian Portuguese to English). swallowing, and three subscales (domains) over which the
This back-translated version was subsequently compared to other 19 items are distributed: emotional (E), physical (P),
the original English version to ensure that the translations and functional (F). Five responses are possible to each
were accurate. Discrepancies between the original and question (strongly agree, agree, no opinion, disagree,
back-translated versions were resolved by repeating the strongly disagree), and each domain item is scored from 1
process as needed. to 5. All items, except F2, are scored with one point for
strongly agree and five points for strongly disagree. The F2
Psychometric Validation item on the functional domain is scored with five points for
strongly agree and one point for strongly disagree. The
The next step was psychometric validation. We tested the global question was scored individually, and the mean
translated version on a consecutive series of patients seen score of each subscale (emotional, physical, and func-
at the head and neck outpatient clinic of A. C. Camargo tional) was multiplied by 20 to obtain a total score with a
Hospital between October 2008 and October 2009. Inclu- range from zero (extremely low functioning) to 100 (high
sion criteria required adult Portuguese-speaking patients functioning). A higher MDADI score is indicative of better
treated for squamous cell carcinoma of the upper aerodi- day-to-day functioning and QOL.
gestive tract who had at least 1 year of disease-free sur- The University of Washington Quality of Life
vival. Patients with recurrent and/or metastatic disease, as (UW-QOL) questionnaire is a well-validated, concise, and
well as illiterate patients, were excluded from the study. easy-to-complete and easy-to-interpret disease-specific
Eligible patients were invited to participate in the study, QOL questionnaire. Each domain item on the UW-QOL

123
R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

scale is scored from 0 to 100, with the composite score analysis of the influence of the clinical and therapeutic data
being the mean of the 12 domains. Also, there are three on the results obtained from the MDADI.
general questions about overall quality of life, scored
independently from the 12 domains. A higher score is Statistical Analysis
indicative of better QOL [11, 17].
The HAD scale is a questionnaire with 14 questions used Pearson’s correlation coefficient and Spearman’s rho were
to evaluate the presence of anxiety and depression in patients used to evaluate the correlations between continuous and
with physical diseases. The 14 questions are divided into two ordinal variables, respectively. The nonparametric Mann–
subgroups (seven questions to evaluate anxiety and seven for Whitney and Kruskal–Wallis tests were used to compare
depression); the answers to each question range from 0 to 3. means between the groups. The statistical analysis was
The sum of each subgroup denotes the category of the performed using version 12.0 of the SPSS statistical
patients: 0–7, noncases; 8–10, doubtful cases; and 11–21, program for Windows (SPSS, Inc., Chicago, IL, USA).
definite cases. In this study we divided the HAD scores into
two categories: patients without depression (‘‘noncases’’)
and those who may or definitely have depression (‘‘doubtful’’ Results
and ‘‘definite’’ cases).
The SWAL-QOL has 44 questions and 11 domains: Translation and Cultural Adaptation
burden, eating desire, eating duration, symptoms, food
selection, communication, fear, mental health, social Translation of the MDADI from English to Brazilian Por-
function, sleep, and fatigue. The patient is asked to tuguese was performed with slight modifications to maintain
describe the frequency of each symptom (always, often, semantic equivalence such that some words were substituted
sometimes, hardly ever, or never). The overall score ranges for others with the same meaning in order to achieve better
from 0 to 100, and a higher value indicates a better QOL communication. The negative sentences were grammatically
related to swallowing. changed to affirmative in order to avoid confusing double-
negative sentences, e.g., ‘‘I do not feel self-conscious when I
Reliability eat’’ was changed to ‘‘I feel self-conscious when I eat.’’ The
research team did not find it necessary to remove any of the
Reliability was established by assuring both internal con- 20 items from the original English version; therefore, the
sistency (Cronbach’s a) and test–retest reliability (intra- Portuguese version was assembled with 20 questions arran-
class correlation coefficient, ICC) at 2 weeks in the ged in four domains: global (one question), emotional (six
absence of interim treatment. Internal consistency is con- questions), physical (eight questions), and functional (five
sidered good if a approximates 0.70 but does not exceed questions) (Appendix).
0.90, as anything over this value implies the presence of
potential redundant items [18]. Test–retest reliability was Sample Characteristics
measured with the ICC [19].
Seventy-two consecutive patients, mostly male patients
Construct Validity with a mean age of 63 years, were included in the test of
this version. The larynx was the most frequent primary
The three forms of validity are content, criterion, and con- tumor site, and most patients presented with stage T1–T2
struct. Content validity was established with a rigorous neoplasms. Most were treated by surgery exclusively or by
approach to item development in the original form and surgery followed by radiotherapy (Table 1).
is maintained by a rigorous process of translation and back-
translation. The criterion validity is determined by com- Reliability and Construct Validity
parison with a ‘‘gold-standard’’ instrument, which is difficult
in this scenario because there is no ‘‘gold-standard’’ instru- The Cronbach’s a obtained for each domain of the MDADI
ment to evaluate QOL related to dysphagia in head and neck (global, emotional, functional, and physical) was within the
cancer patients. Construct validity is present if the scale optimum range (Table 2). The test–retest reliability for the
behaves according to hypothesized relationships. We total score for the MDADI had an ICC greater than 0.795
hypothesized that the global score of the MDADI should (p \ 0.001) (Fig. 1), which is considered good.
correlate with the global question about overall QOL of the Construct validity was evaluated according to the
UW-QOL. We also hypothesized that higher depression hypothesis that patients with higher scores on the HAD-D
scores and larger tumors would correlate with decreased scale should score lower on the MDADI. The mean score on
MDADI scores. Additionally, we performed a thorough the MDADI for the doubtful and definite cases of depression

123
 R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

Table 1 Demographic and clinical characteristics (N = 72)

Variable Category N (%)

Gender Male 60 (83.3)

Female 12 (16.7)
Age (years) Range 33–77
Median 62.3
Mean ± SD 63 ± 9.537
Primary tumor site Oral cavity 18 (25.0)
Oropharynx 15 (20.8)
Larynx 34 (47.2)
Hypopharynx 5 (6.9)
Clinical T stage T1–2 40 (55.6)
T3–4 32 (44.4)
Clinical N stage N0 43 (59.7)
Fig. 1 MDADI test–retest data (ICC = 0.795); ICC intraclass cor-
N1 10 (13.9)
relation coefficient
N2 9 (12.5)
N3 3 (4.2)
Table 3 Correlation between UW-QOL global quality-of-life ques-
Nx 7 (9.7) tionnaire and mean MDADI total score (p \ 0.05)
Clinical M stage M0 72 (100)
UW-QOL global N MDADI T (mean ± SD)
Treatment Surgery 23 (31.9)
RT 11 (15.3) MDADI T 1 (Excellent) 12 93 ± 5.427
Surgery ? RT 17 (23.6) 2 (Very good) 23 85.13 ± 11.178
Surgery ? RT ? CHT 5 (6.9) 3 (Good) 19 82.42 ± 13.426
RT ? CHT 16 (22.2) 4 (Mean) 14 76.64 ± 15.320
SD standard deviation; x ignored; RT radiotherapy; CHT 5 (Bad) 3 71.33 ± 2.309
chemotherapy 6 (Very bad) 1 58
QOL = quality of life; MDADI T = MDADI total score;
Table 2 Cronbach’s a coefficients for total score on MDADI N = number of patients; SD = standard deviation
subscales
Instrument No. of items Cronbach’s a
MDADI Results
MDADI global 1 0.697
MDADI emotional 6 0.824
Patients treated for head and neck cancer had a mean total
MDADI functional 5 0.706
score of 83.13 on the validated Brazilian Portuguese ver-
MDADI physical 8 0.785
sion of the MDADI, which is indicative of minimal limi-
MDADI total score 19 0.812
tation in overall QOL [20]. Evaluation of each individual
domain yielded mean scores of 87.19 for emotional, 88.69
for functional, and 75.23 for physical well-being, indicat-
was 69.9, whereas patients without depression presented ing minimal limitation in the emotional and functional
with a mean score of 85, thereby resulting in a statistically domains and moderate limitation in the physical domain
significant difference (Kruskal–Wallis test, p = 0.01). Fur- (data not shown). Women scored worse on the total
thermore, a comparison between total scores from the MDADI, which was greatly influenced by differences in
MDADI and the UW-QOL questionnaire revealed statisti- the physical domain (Table 5), but we did not observe
cally significant results on the Kruskal–Wallis test and differences in the overall test or individual domains when
Spearman’s correlation (p \ 0.01, rs = 0.704). There was patients older than 65 years were compared with younger
also significant correlation between global QOL and total patients. Although we did not observe a statistically sig-
means obtained on the MDADI (p = 0.01, rs = 0.703) nificant difference in individual MDADI domains between
(Table 3). Comparisons of the total MDADI with each patients with early-stage and advanced tumors, there was a
domain of the SWAL-QOL revealed strong, moderate, and trend toward worse total MDADI QOL in patients with
some weak Pearson’s correlations, all of which were statis- more advanced tumors. In contrast, the presence of
tically significant (Table 4). metastasis significantly correlated with decreased QOL in

123
R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

Table 4 Correlation between total MDADI score and SWAL-QOL [11, 14, 21, 22]. Evaluation of QOL in head and neck
items (p \ 0.001) cancer patients is an important outcome measure and
SWAL-QOL items MDADI T p Correlation should include domains that reflect the disease and treat-
ments’ impact on feeding, swallowing, speech, communi-
Burden 0.595 \0.001 Moderate
cation, and appearance. To address these criteria
Eating desire 0.605 \0.001 Moderate adequately, it is imperative to use specific questionnaires
Eating duration 0.454 \0.001 Weak rather than encompass the activities of daily living and
Symptoms 0.665 \0.001 Moderate social life in addition to the questionnaires being available
Food selection 0.679 \0.001 Moderate and validated in the patient’s native language [23].
Communication 0.418 \0.001 Weak Adequate evaluation of the reliability of a specific
Fear 0.484 \0.001 Weak questionnaire protocol involves analysis to determine
Mental health 0.509 \0.001 Moderate internal consistency and test–retest reliability (intraclass
Social 0.719 \0.001 Strong correlation coefficient, ICC) for all of the psychometric
Sleep 0.285 0.015 Without correlation measures. The internal consistency of a specific question-
Fatigue 0.461 \0.001 Weak naire is evaluated according to each domain, assuming that
individual questions in each domain correspond to the
same topic. It is known that the internal consistency
all single domains and global scores of the MDADI. coefficient increases with an increasing number of ques-
Additionally, patients who underwent either cervical lymph tions for a specific domain [24], and the internal consis-
node dissection or combined surgery plus radiotherapy also tency coefficient is best evaluated with Cronbach’s a
had worse total MDADI scores (Table 5). coefficient. An optimum Cronbach’s a coefficient should
Global and single domains for the MDADI were also range between 0.70 and 0.80. At this study, this coefficient
evaluated according to primary tumor sites, including the was 0.69 for the global domain, 0.82 for the emotional
mouth, oropharynx, larynx, and hypopharynx. Patients with domain, 0.70 for the functional domain, and 0.78 for the
primary hypopharynx tumors had decreased scores on the physical domain, which are results that can be considered
total MDADI (75.2) and on the global (74.6), functional satisfactory.
(79), and physical (69.7) domains, and patients with pri- The reliability of a questionnaire for a foreign language
mary mouth tumors had decreased scores in the emotional is determined using the test–retest method in which the
domain (64), although these differences were not statisti- same individual is asked to answer the same questionnaire
cally significant (data not shown). with a specific time frame between each assignment. The
Only two patients from this cohort were being feed by test–retest validation is considered adequate when the ICC
nasoenteral tubes at the time of completing the question- is above 0.70 [25]. Although the value obtained in this
naire, and those patients reported lower total MDADI validation study (0.795) was below the original value
scores than patients fed by mouth (p \ 0.05). obtained for the English version for the MDADI (0.960)
[26], the ICC is still in the optimum range and validates use
of the Portuguese language version of the MDADI.
Discussion The results of the Portuguese version of the MDADI
were further correlated with the UW-QOL, SWAL-QOL,
Assessment of QOL related to dysphagia is important to and HAD-D scales for psychometric validation. The
objectively score patient-oriented outcomes, and it has depression score on the HAD scale was inversely corre-
been increasingly used as a follow-up tool in head and neck lated with the results on the Portuguese version of the
cancer clinics. We have successfully adapted and validated MDADI, i.e., the higher the QOL indicated by the inven-
a Portuguese adaptation of the original version of the tory, the lower the chances of a patient’s presenting with
MDADI, and we subsequently used this tool to demon- symptoms of depression. This finding is in accordance with
strate in a cohort of patients treated for head and neck previous studies that suggest that emotional, social, and
cancer that QOL related to dysphagia is worse in females, familiar well-being have a direct influence on the devel-
in those with metastasis, and in patients who underwent opment of depression in patients treated for head and neck
either neck lymph node dissection or combined surgery tumors [15, 27].
plus radiotherapy treatment. Evaluation of QOL related to There was also a strong correlation between the overall
dysphagia enables objective scoring of patients’ daily score on the Portuguese version of the MDADI and
needs and feeding difficulties and, therefore, is vital in the swallowing domain of the UW-QOL questionnaire
guiding the rehabilitation team in assessing and following (p \ 0.01), and also with the general question about overall
patients’ progress in a longitudinal reintegration program quality of life of the UW-QOL, which corroborates the

123
 R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

Table 5 Mean MDADI scores according to variables

Variable MDADI global MDADI emotional MDADI functional MDADI physical MDADI total

Gender
Male (N = 60) 82.67 ± 27.176 88.33 ± 14.071 89.90 ± 13.686 77.07 ± 15.498 84.48 ± 12.937
Female (N = 12) 65.00 ± 34.245 81.50 ± 13.833 82.67 ± 19.095 66.04 ± 12.498 76.33 ± 12.608
P 0.103 0.077 0.224 0.024 0.031
Age
\65 years (N = 43) 78.60 ± 30.674 86.09 ± 15.559 88.14 ± 15.284 74.66 ± 16.653 82.33 ± 14.011
[65 years (N = 29) 81.38 ± 26.689 88.83 ± 11.887 89.52 ± 14.319 76.07 ± 13.919 84.31 ± 11.914
P 0.723 0.12 0.598 0.747 0.674
T stage
T1–T2 (N = 40) 83.00 ± 24.620 88.45 ± 15.018 90.40 ± 14.459 77.14 ± 16.532 84.85 ± 14.092
T3–T4 (N = 32) 75.62 ± 33.595 85.62 ± 13.095 86.56 ± 15.208 72.84 ± 14.049 80.97 ± 11.743
P 0.412 0.136 0.120 0.203 0.060
N stage
N0 (N = 43) 88.84 ± 22.383 90.98 ± 11.342 92.19 ± 11.278 80.26 ± 14.568 87.19 ± 10.985
N? (N = 22) 70.91 ± 31.306 81.09 ± 17.454 84.82 ± 16.661 67.89 ± 13.849 77.45 ± 14.355
P 0.007 0.007 0.041 0.001 0.003
Education
Elementary (N = 29) 80.69 ± 30.930 86.69 ± 15.769 89.52 ± 15.507 73.67 ± 15.878 82.79 ± 14.226
High school (N = 20) 85.00 ± 21.398 89.00 ± 12.773 88.90 ± 13.210 79.03 ± 13.819 85.10 ± 12.152
Superior (N = 22) 75.45 ± 32.031 87.18 ± 13.218 87.27 ± 16.125 74.86 ± 16.242 82.36 ± 13.000
P 0.527 0.891 0.724 0.498 0.737
Treatment
Surgery (N = 23) 80.87 ± 27.947 87.13 ± 13.589 91.65 ± 10.849 76.67 ± 16.688 84.52 ± 12.406
Radiotherapy (N = 11) 89.09 ± 18.684 92.73 ± 11.841 91.64 ± 10.652 85.00 ± 14.304 89.18 ± 11.071
Combined (N = 38) 76.32 ± 31.829 85.63 ± 15.030 86.05 ± 17.476 71.53 ± 14.059 80.53 ± 13.727
P 0.612 0.151 0.438 0.027 0.042
Neck dissection
Yes (N = 40) 72.50 ± 32.875 84.85 ± 13.250 88.40 ± 14.684 72.23 ± 15.616 81.23 ± 12.288
No (N = 32) 88.75 ± 20.280 90.13 ± 14.932 89.06 ± 15.206 78.98 ± 14.792 85.50 ± 13.998
P 0.037 0.017 0.700 0.068 0.044
Values are mean ± SD (standard deviation). Variables are gender (N = 72), age (N = 72), T stage (N = 72), N stage (N = 65), education
(N = 71), treatment (N = 72), and cervical neck dissection (N = 72)

sensitivity of both instruments to measure the negative speech, aesthetics, and emotional state [11, 21, 22, 28]. We
impact of dysphagia on the patient’s quality of life. believe that correlations between MDADI subscales and
Comparisons between this version of the MDADI and SWAL-QOL could be stronger for some domains; how-
the SWAL-QOL resulted in distinct correlation values, ever, for the purpose of this validation study, such corre-
which ranged from weak (four domains) to moderate (five lations were not necessary.
domains) and strong (one domain); all correlations were Patients with more advanced disease, lymph node
statistically significant (p \ 0.05). The social domain of metastasis, and the need for combined surgery and radio-
the SWAL-QOL was strongly correlated with the overall therapy tend to have lower scores on QOL questionnaires
MDADI results, which is expected because the MDADI is [5, 29–31]. This result was seen in our cohort (Table 5).
known to be a more sensitive and reliable tool for evalu- Furthermore, our patients with compromised lymph nodes
ating the influence of dysphagia on emotional and psy- scored significantly worse on the overall and emotional
chological aspects of health. These different correlations domains of the MDADI, in accordance with previous
might be explained by the inclusion of patients with vari- publications that suggest that the severity of swallowing-
able tumor sites among the different studies, as it is known related dysphagia is directly correlated with tumor size,
that tumor site greatly influences chewing, swallowing, type, resection size, and treatment effects [32–34].

123
R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

With respect to the location of the primary tumor, Minha capacidade de deglutição limita minhas
patients with hypopharynx neoplasms had lower scores on atividades diárias
all MDADI domains, although the difference in scores of ( )Concordo totalmente ( )Concordo ( )Sem opinião
those with neoplasms at other locations was not statistically ( )Discordo ( )Discordo totalmente
significant. Similar results were found in a retrospective E2. Eu tenho vergonha dos meus hábitos alimentares
study of 73 patients treated for hypopharyngeal carcinoma ( )Concordo totalmente ( )Concordo ( )Sem opinião
in which dysphagia was the most severe acute and long- ( )Discordo ( )Discordo totalmente
term adverse event that was reported (50 % of patients), F1. As pessoas têm dificuldade de cozinhar para mim
with a score of 34 on the Quality of Life—Head and Neck ( )Concordo totalmente ( )Concordo ( )Sem opinião
35 Questionnaire (EORTC QLQ H&N35) [35]. Another ( )Discordo ( )Discordo totalmente
study of 110 laryngectomy patients proposed to evaluate P2. É mais difı́cil engolir no fim do dia
whether dysphagia affects laryngectomees’ QOL func- ( )Concordo totalmente ( )Concordo ( )Sem opinião
tioning and psychological well being. The results showed ( )Discordo ( )Discordo totalmente
that laryngectomees with dysphagia had lower scores on E7. Sinto-me inseguro quando me alimento
the questions that address overall QOL (UW-QOL) over ( )Concordo totalmente ( )Concordo ( )Sem opinião
the past 7 days from the questionnaire application [36]. ( )Discordo ( )Discordo totalmente
Although only two patients in our cohort were being fed E4. Eu estou triste pelo meu problema de deglutição
nasoenterally, both had much lower scores on the overall ( )Concordo totalmente ( )Concordo ( )Sem opinião
MDADI (mean = 63) compared to the rest of the cohort. ( )Discordo ( )Discordo totalmente
This difference is indicative of the major influence that the P6. Deglutir é um grande esforço
presence of a feeding tube has on QOL, mainly due to ( )Concordo totalmente ( )Concordo ( )Sem opinião
social exclusion and meal-related events. Although we ( )Discordo ( )Discordo totalmente
have not used an outcome measure of diet for the patients’ E5. Deixo de sair de casa por causa do meu problema
analysis, it is expected that some of them have a restricted de deglutição
diet with respect to consistencies, as head and neck cancer ( )Concordo totalmente ( )Concordo ( )Sem opinião
treatment is associated with the decreased scores in nor- ( )Discordo( )Discordo totalmente
malcy of diet [37, 38]. F5. Meu problema de deglutição tem me causado
In conclusion, the present study validates the Brazilian– perda de rendimentos financeiros
Portuguese language version of the MDADI. This provides ( )Concordo totalmente ( )Concordo ( )Sem opinião
another tool to evaluate the impact of dysphagia on the ( )Discordo ( )Discordo totalmente
QOL of head and neck cancer patients. P7. Eu levo mais tempo pra comer por causa do meu
problema de deglutição
Acknowledgments R. L. V. Guedes thanks CAPES (Coordenação ( )Concordo totalmente ( )Concordo ( )Sem opinião
de Aperfeiçoamento de Pessoal de Nı́vel Superior) for financial
support.
( )Discordo ( )Discordo totalmente
P3. As pessoas me perguntam, ‘‘Porque você não
Conflict of interest The authors declare that they have no conflicts pode comer isto?’’
of interest or financial ties to disclose. ( )Concordo totalmente ( )Concordo ( )Sem opinião
( )Discordo ( )Discordo totalmente
E3. Outras pessoas se irritam por causa do meu
problema de deglutição
Appendix ( )Concordo totalmente ( )Concordo ( )Sem opinião
( )Discordo ( )Discordo totalmente
Questionário de Disfagia M. D. Anderson (MDADI) P8. Eu tenho tosse quando eu tento beber lı́quidos
Nome: _______________________________________ ( )Concordo totalmente ( )Concordo ( )Sem opinião
Data: ________________ ( )Discordo ( )Discordo totalmente
Este questionário pergunta sobre sua habilidade de engolir F3. Meus problemas de deglutição atrapalham minha
(deglutir). Estas informações irão nos auxiliar a entender vida pessoal e social
como você se sente em relação à sua deglutição. As ( )Concordo totalmente ( )Concordo ( )Sem opinião
questões que seguem foram preparadas por pessoas que ( )Discordo ( )Discordo totalmente
têm problema com sua deglutição. Alguns dos itens podem F2. Eu me sinto à vontade para sair pra comer com
ser relevantes para você. meus amigos, vizinhos e parentes
Por favor, leia cada questão e marque a resposta que ( )Concordo totalmente ( )Concordo ( )Sem opinião
melhor reflete sua experiência na última semana. ( )Discordo ( )Discordo totalmente

123
 R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

P5. Eu limito minha alimentação por causa da minha 10. Peters M, Passchier J. Translating instruments for cross-cultural
dificuldade de deglutição studies in headache research. Headache. 2006;46:82–91.
11. Vartanian JG, Carvalho AL, Yueh B, Furia CL, Toyota J,
( )Concordo totalmente ( )Concordo ( )Sem opinião McDowell JA, Weymuller EA Jr, Kowalski LP. Brazilian–Por-
( )Discordo ( )Discordo totalmente tuguese validation of the University of washington quality of life
P1. Perco peso devido ao meu problema de questionnaire for patients with head and neck cancer. Head Neck.
deglutição 2006;28:1115–21.
12. McHorney CA, Bricker DE, Kramer AE, Rosenbek JC, Robbins
( )Concordo totalmente ( )Concordo ( )Sem opinião J, Chignell KA, Logemann JA, Clarke C. The SWAL-QOL
( )Discordo ( )Discordo totalmente outcomes tool for oropharyngeal dysphagia in adults: I—con-
E6. Eu tenho baixa auto-estima por causa do meu ceptual foundation and item development. Dysphagia. 2000;15:
problema de deglutição 115–21.
13. McHorney CA, Bricker DE, Robins J, Kramer AE, Rosenbek JC,
( )Concordo totalmente ( )Concordo ( )Sem opinião Chignel K. The SWAL-QOL outcomes tool for oropharyngeal
( )Discordo ( )Discordo totalmente dysphagia in adults: II—item reduction and preliminary scaling.
P4. Eu sinto que estou conseguindo deglutir uma Dysphagia. 2000;15:122–33.
grande quantidade de alimentos 14. McHorney CA, Bricker DE, Robins J, Kramer AE, Rosenbek JC,
Chignel K. The SWAL-QOL and SWAL-CARE outcomes tool
( )Concordo totalmente ( )Concordo ( )Sem opinião for oropharyngeal dysphagia in adults: III—documentation of
( )Discordo ( )Discordo totalmente reliability and validity. Dysphagia. 2002;17:97–114.
F4. Eu me sinto isolado por causa dos meus hábitos 15. Portas JG. Validação para a lı́ngua portuguesa–brasileira dos
de alimentação questionários: qualidade de vida em disfagia (SWAL-QOL) e
satisfação do paciente e qualidade do cuidado no tratamento da
( )Concordo totalmente ( )Concordo ( )Sem opinião disfagia (SWALCARE). São Paulo: Dissertação de Mestrado-
( )Discordo ( )Discordo totalmente Fundação Antônio Prudente; 2009.
Obrigado por completar este questionário! 16. Botega NJ, Bio MR, Zomignani MA, Garcia C Jr, Pereira WAB.
Mood disorders among medical in-patients: a validation study of
the hospital anxiety and depression scale (HAD). Rev Saúde
Pública. 1995;29:355–63.
17. Hassan SJ, Weymuller EA Jr. Assessment of quality of life in
head and neck cancer patients. Head Neck. 1993;15(6):485–96.
References 18. Streiner DL, Norman GR. Health measurement scales: a practical
guide to their development and use. 2nd ed. Oxford: Oxford
1. Vartanian JG, Carvalho AL, Yueh B, Priante AV, de Melo RL, University Press; 1989.
Correia LM, Köhler HF, Toyota J, Kowalski IS, Kowalski LP. 19. Deyo RA, Diehr P, Patrick DL. Reproducibility and responsive-
Long-term quality-of-life evaluation after head and neck cancer ness of health status measures. Statistics and strategies for eval-
treatment in a developing country. Arch Otolaryngol Head Neck uation. Control Clin Trials. 1991;12(suppl):142S–58S.
Surg. 2004;130:1209–13. 20. Chen PH, Golub JS, Hapner ER, Johns MM III. Prevalence of
2. Gillespie MB, Brodsky MB, Day TA, Sharma AK, Lee FS, perceived dysphagia and quality-of-life impairment in a geriatric
Martin-Harris B. Laryngeal penetration and aspiration during population. Dysphagia. 2009;24:1–6.
swallowing after the treatment of advanced oropharyngeal can- 21. Rogers SN, Laher H, Overend L, Lowe D. Importance-rating
cer. Arch Otolaryngol Head Neck Surg. 2005;131:615–9. using the University of Washington quality-of-life questionnaire
3. Thomas L, Jones TM, Tandon S, Katre C, Lowe D, Rogers SN. in patients treated by primary surgery for oral and oropharyngeal
An evaluation of the University of Washington quality of life cancer. J Craniomaxillofac Surg. 2002;30:125–32.
swallowing domain following oropharyngeal cancer. Eur Arch 22. Bandeira AKC, Azevedo EH, Vartanian JG, Nishimoto IN, Ko-
Otorhinolaryngol. 2008;265(Suppl 1):S29–37. walski LP, Carrara-de Angelis E. Quality of life related to
4. Gillespie MB, Brodsky MB, Day TA, Lee F, Martin-Harris B. swallowing after tongue cancer treatment. Dysphagia. 2008;
Swallowing-related quality of life after head and neck cancer 23(2):183–92.
treatment. Laryngoscope. 2004;114:1362–7. 23. Gliklich RE, Goldsmith TA, Funk GF. Are head and neck specific
5. Chen AY, Frankowski R, Bishop-Leone J, Hebert T, Leyk S, quality of life measures necessary? Head Neck. 1997;19:474–80.
Lewin J, Goepfert H. The development and validation of a dys- 24. Bland JM, Altman DG. Cronbach’s alpha. Br Med J. 1997;
phagia-specific quality-of-life questionnaire for patients with 314:572.
head and neck cancer: the M. D. Anderson dysphagia inventory. 25. Menezes PR, Nascimento AF. Validade e confiabilidade das
Arch Otolaryngol Head Neck Surg. 2001;127:870–6. escalas de avaliação em psiquiatria. In: Gorenstein C, Andrade
6. Kazi R, Prasad V, Venkitaraman R, Nutting CM, Clarke P, Rhys- LHSG, Zuardi AW (organizadores). Escalas de avaliação clı́nica
Evans P, Harrington KJ. Questionnaire analysis of swallowing- em Psiquiatria e psicofarmacologia. São Paulo: Lemos Editorial;
related outcomes following glossectomy. ORL J Otorhinolaryn- 2000. p. 23–8.
gol Relat Spec. 2008;70:151–5. 26. Schindler A, Borghi E, Tiddia C, Ginocchio D, Felisati G, Ot-
7. Ministério da Saúde. Instituto Nacional de Câncer. Estimativa/ taviani F. Adaptation and validation of the Italian MD Anderson
2010 incidência de câncer no Brasil. Rio de Janeiro: INCA; 2010. dysphagia inventory (MDADI). Rev Laryngol Otol Rhinol.
8. Guillemin F, Bombardier C, Beaton D. Cross-cultural adaptation 2008;129:97–100.
of health-related quality of life measures: literature review and 27. Braz DSA, Ribas MM, Dedivitis RA, Nishimoto IN, Barros APB.
proposed guidelines. J Clin Epidemiol. 1993;46:1417–32. Quality of life and depression in patients undergoing total and
9. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines partial laryngectomy. Clinics. 2005;60:135–42.
for the process of cross-cultural adaptation of self-reported 28. Goguen LA, Posner MR, Norris CM, Tishler RB, Wirth LJ,
measure. Spine. 2000;25:3186–91. Annino DJ, Gagne A, Sullivan CA, Sammartino DE, Haddad RI.

123
R. L. V. Guedes et al.: Validation of the M.D. Anderson Dysphagia Inventory

Dysphagia after sequential chemoradiation therapy for advanced regarding dysphagia-related quality of life in cancer of the oro-
head and neck cancer. Otolaryngol Head Neck Surg. 2006;134: pharynx and nasopharynx. Int J Radiat Oncol Biol Phys.
916–22. 2008;72:1119–27.
29. Gaziano JE. Evaluation and management of oropharyngeal dys- 35. Keereweer S, Kerrebijn JD, Al-Mamgani A, Sewnaik A, de Jong
phagia in head and neck cancer. Cancer Control. 2002;9:400–9. RJ, van Meerten E. Chemoradiation for advanced hypopharyn-
30. Nguyen NP, Frank C, Moltz CC, Vos P, Smith HJ, Karlsson U, geal carcinoma: a retrospective study on efficacy, morbidity and
Dutta S, Midyett A, Barloon J, Sallah S. Impact of dysphagia on quality of life. Eur Arch Otorhinolaryngol. 2012;269(3):939–46.
quality of life after treatment of head and neck cancer. Int J 36. Maclean J, Cotton S, Perry A. Dysphagia following a total lar-
Radiat Oncol Biol Phys. 2005;61:772–8. yngectomy: the effect on quality of life, functioning, and psy-
31. Nguyen NP, Frank C, Moltz CC, Vos P, Smith HJ, Nguyen PD, chological well-being. Dysphagia. 2009;24:314–21.
Martinez T, Karlsson U, Dutta S, Lemanski C, Nguyen LM, 37. Pillon J, Gonçalves MI, De Biase NG. Changes in eating habits
Sallah S. Analysis of factors influencing aspiration risk following following total and frontolateral laryngectomy. Sao Paulo Med J.
chemoradiation for oropharyngeal cancer. Br J Radiol. 2009;82: 2004;122:195–9.
675–80. 38. Campbell BH, Spinelli K, Marbella AM, Myers KB, Kuhn JC,
32. Carrara-de Angelis E, Mourão LF, Furia CLB. Avaliação e Layde PM. Aspiration, weight loss, and quality of life in head and
tratamento das disfagias após o tratamento do câncer de cabeça e neck cancer survivors. Arch Otolaryngol Head Neck Surg.
pescoço. In: Carrara-de Angelis E, Furia CLB, Mourão LF, Ko- 2004;130:1100–3.
walski LP, editors. A atuação da fonoaudiologia no câncer de
cabeça e pescoço. São Paulo: Editora Lovise, 2000; p. 155–62.
33. Logemann JA, Rademaker AW, Pauloski BR, Lazarus CL, Mittal Renata Lı́gia Vieira Guedes SLP
BB, Brockstein B, MacCracken E, Haraf DJ, Vokes EE, Newman
Elisabete Carrara-de Angelis SLP, PhD
LA, Liu D. Site of disease and treatment protocol as correlates of
swallowing function in patients with head and neck cancer treated Amy Y. Chen MD
with chemoradiation. Head Neck. 2006;28:64–73.
34. Teguh DN, Levendang PC, Noever I, van Rooij P, Voet P, van Luiz Paulo Kowalski MD, PhD
der Est H, Sipkema D, Sewnaik A, de Jong RJB, de la Bije D, José Guilherme Vartanian MD, PhD
Schmitz PI. Treatment techniques and site considerations

123