e66(1)

C OPYRIGHT
2018 BY T HE J OURNAL OF B ONE AND J OINT S URGERY, I NCORPORATED

HSS Dysphagia and Dysphonia Inventory (HSS-DDI)

Following Anterior Cervical Fusion
Patient-Derived, Validated, Condition-Specific Patient-Reported Outcome
Measure Outperforms Existing Indices
Alexander P. Hughes, MD, Stephan N. Salzmann, MD, Okezie K. Aguwa, MD, Courtney Ortiz Miller, BS,
Roland Duculan, MD, Jennifer Shue, MS, Frank P. Cammisa, MD, Andrew A. Sama, MD,
Federico P. Girardi, MD, Ashutosh Kacker, MD, and Carol A. Mancuso, MD

Investigation performed at the Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, NY

Background: Dysphagia and dysphonia are common complications after anterior cervical spine surgery; however, re-
ported prevalences vary greatly due to a lack of reliable clinical standards for measuring postoperative swallowing and
speech dysfunction. The Hospital for Special Surgery Dysphagia and Dysphonia Inventory (HSS-DDI) was developed as a
patient-derived, patient-reported instrument to measure dysphagia and dysphonia more accurately after anterior cervical
spine surgery than existing indices.
Methods: This multiphase survey-development study implemented a mixed-methods approach. Phase 1 involved quali-
tative assessment of postoperative patient-reported swallowing or speaking deficiencies to assemble a draft survey.
Phase 2 established test-retest reliability and finalized the 31-item HSS-DDI. Phase 3 compared the HSS-DDI with the
Swallowing-Quality of Life (SWAL-QOL) questionnaire and the M.D. Anderson Dysphagia Inventory (MDADI) for validity and
responsiveness.
Results: Phase 1, performed to formulate the draft survey, included 25 patients who were asked about speech and
swallowing dysfunction after anterior cervical spine surgery involving at least 3 vertebral levels. Phase 2 included 49
patients who completed the draft survey twice. The mean scores (and standard deviation) for each administration of the
HSS-DDI were 67 ± 24 and 75 ± 22, the Cronbach alpha coefficients were both 0.97, and the intraclass correlation
coefficient was 0.80. The 31-item HSS-DDI was finalized with all but 2 items having weighted kappa values of ‡0.40.
Phase 3 included 127 patients and established external validity, with most correlation coefficients between the HSS-DDI
and the SWAL-QOL and MDADI ranging from 0.5 to 0.7. Internal validity was established by identifying worsening HSS-DDI
scores with increases in the number of vertebral levels involved (p = 0.02) and in the Surgical Invasiveness Index (p =
0.006). HSS-DDI responsiveness ascertained by effect size (0.73) was better than that of the SWAL-QOL and MDADI. The
average administration time for the HSS-DDI was 2 minutes and 25 seconds.
Conclusions: The HSS-DDI is efficient, valid, and more responsive to change after anterior cervical spine surgery than
existing surveys.
Clinical Relevance: The HSS-DDI fills a gap in postoperative assessment by providing a reliable, more clinically sen-
sitive, patient and condition-specific evaluation of dysphagia and dysphonia prospectively and longitudinally.

A
nterior cervical spine surgery has been increasingly used
to treat various cervical pathological conditions1,2. It is
safe and effective, but dysphagia and dysphonia are
commonly reported complications3-6. The published preva-
lences of dysphagia and dysphonia after anterior cervical spine
surgery are highly variable, ranging from <1% to 71% and
<1% to 60%, respectively5-14. Possible reasons for these wide
ranges include different definitions of dysphagia and dysphonia

Disclosure: Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of
Health under Award Number UL1TR000457. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the
article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arena outside the
submitted work (http://links.lww.com/JBJS/E672).

J Bone Joint Surg Am. 2018;100:e66(1-10) d http://dx.doi.org/10.2106/JBJS.17.01001

 e66(2)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
and variations in measurement methods, surgical techniques,
and reporting consistency6,11. Surgeon underreporting of these
complications has also been described15. Dysphagia and dys-
phonia can be major determinants of patient outcomes and are
associated with increased morbidity, length of hospital stay, and
hospital-related costs16.
Patient-derived outcome instruments have been in-
creasingly utilized across medical and surgical specialties. There
is consensus that the most valid, reliable, and sensitive way
to measure clinical outcomes across a broad spectrum of
metrics is with patient-derived and validated instruments6,17. To
date, there is no universally accepted outcome measurement of
dysphagia and dysphonia following anterior cervical spine
surgery, making comparisons among studies difficult. Current
instruments are either oversimplified, nonvalidated, not patient-
derived, or non-specific to patients treated with anterior cervical
spine surgery6,17,18. A unique feature of this patient population is
the anticipated improvement of postoperative dysphagia and
dysphonia with time. A scale that is responsive to these lon-
gitudinal changes is needed to counsel and manage patients
clinically. For instance, patients treated with anterior cervical
spine surgery have different psychosocial concerns and natural
histories compared with those with chronic neurodegenerative
disease (the population for which the Swallowing-Quality of
Life [SWAL-QOL] questionnaire was developed) and those
with oropharyngeal cancer (the patient population assessed
with the M.D. Anderson Dysphagia Inventory [MDADI])19,20.
Dysphagia and dysphonia are also subjective conditions for
which patient-reported instruments seem to be more clinically
relevant and effective in assessing degrees of dysfunction
compared with clinician-administered surveys17.
The purpose of this study was to create a valid, reli-
able, patient-derived, patient-reported survey instrument
(the Hospital for Special Surgery Dysphagia and Dysphonia
Inventory [HSS-DDI]) to measure dysphagia and dysphonia
after anterior cervical spine surgery. Additionally, we com-
pared the performance of this novel instrument with that of
preexisting, widely used surveys. The study design is similar
to previously validated methodologies used for patient-derived
instruments21-24.
Materials and Methods
I n this multiphase study, we used a mixed-methods approach
to address dysphagia and dysphonia during recovery from
anterior cervical spine surgery. The study was approved by our
institutional review board, and all patients provided written
informed consent.
Phase 1: Identifying Items for Draft Survey
Patients were eligible for phase 1 if they were ‡18 years old,
spoke English, were presenting for their first postoperative visit
following anterior cervical spine surgery, and had had the
surgery approximately 4 weeks prior to that visit. To increase
the capture of patients with relevant symptoms, only those who
had undergone fusion of ‡3 levels were included. Patients were
asked open-ended questions about their experiences with
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N
swallowing and speech since the surgery and whether these
experiences affected their daily routines, interactions with others,
and mood. They also were asked whether they made accom-
modations for eating and speaking. Patients were encouraged
to freely volunteer comments, and their responses were re-
corded verbatim.
Their responses were assessed with standard qualitative
techniques using grounded theory, a process by which responses
are reviewed to identify unique concepts and then grouped into
larger categories through an iterative process. Three investi-
gators independently conducted the analysis and, through
consensus, arrived at the final set of categories. Patients were
enrolled until data saturation, when no new concepts were
volunteered.
Phase 2: Assembling Draft Survey, Assessing Test-Retest
Reliability, Selecting Final Items, and Scoring
A draft survey was assembled using categories from phase 1 as
survey items. Dysphagia and dysphonia items were grouped
separately. Items and ranked response options were phrased
using patients’ terminology.
The draft survey was tested on additional patients who
were enrolled in a manner similar to that used in phase 1. To
establish test-retest reliability, patients completed the ques-
tionnaire twice, 3 to 5 days apart. The questionnaire was first
administered in an in-person interview at enrollment and then
again during a telephone follow-up. Only patients without
newly prescribed interventions for dysphagia or dysphonia
were enrolled in phase 2. Furthermore, the total enrollment for
phase 2 was meant to reflect rigorous criteria for repeatability
testing by including 35 to 50 patients25.
The weighted kappa statistic was used to measure agree-
ment, above that due to chance, between the first and second
administrations for each item. Weighted kappa values range
from 21 (perfect disagreement) to 0 (agreement no better
than chance) to 11 (perfect agreement)26. An item was re-
tained for the final survey if the kappa value was ‡0.40. This is
lower than the customary threshold because the courses of
postoperative dysphagia and dysphonia are expected to change
with time. Thus, perfect agreement reflecting an unchanged
status is not a realistic target. A system was developed to gen-
erate an overall score (described below), and the intraclass
correlation coefficient (ICC) was calculated to measure within-
patient agreement in scores. For both kappa and the ICC, a
value of <0.4 was considered slight to fair agreement; ‡0.4 to
0.6, moderate agreement; and >0.6, good to excellent agree-
ment26. Internal consistency was assessed with the Cronbach
alpha coefficient for the overall score.
Phase 3: Validation and Responsiveness
Validity and responsiveness were ascertained by comparing the
HSS-DDI with 2 existing validated surveys developed to mea-
sure dysphagia due to other diagnoses: the SWAL-QOL and
MDADI. Additional patients were enrolled and completed the
HSS-DDI, SWAL-QOL, and MDADI during their 4 and 8-week
postoperative visits.
 e66(3)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N

TABLE I Demographic and Clinical Characteristics

Variables Phase 1 (25 Patients)* Phase 2 (49 Patients) Phase 3 (127 Patients†)

Mean age (range) (yr) 64 (48-79) 58 (19-84) 56 (19-90)

Women (%) 64 74 60
Education (%)
Grade school NA 0 0
Middle school NA 2 1
High school NA 14 17
Some college NA 14 18
College graduate NA 37 34
Graduate school NA 31 29
Unknown/not reported 100 2 1
Marital status (%)
Never married 12 10 11
Married 72 78 74
Divorced/separated 12 8 10
Widowed 4 2 4
Unknown/not reported 0 2 1
Work status (%)
Full time NA 35 37
Part time NA 8 8
On sick leave/disability NA 29 32
Unemployed NA 2 3
Retired NA 20 17
Homemaker NA 0 1
Student NA 2 2
Unknown/not reported 100 4 1
Total no. of vertebral levels fused (%)
1 0 24 24
2 0 31 36
3 52 37 28
4 40 8 10
5 8 0 2
Mean SSI‡ (range) 13 (9-18) 10 (6-15) 10 (5-18)
SSI distribution (%)
5-8 0 24 24
9-10 0 24 32
11-12 52 41 31
13-15 40 10 13
16-20 8 0 0
21-25 0 0 0
>25 0 0 0

*NA = not available. †Forty-eight patients also participated in phase 2. ‡The possible range for the SSI (Surgical Invasiveness Index) is 0 to 48,
with a higher score indicating more complex surgery.

The SWAL-QOL is a 44-item survey that assesses de- mental health, social function, fear of choking, eating duration,
creased quality of life (QOL) in 10 psychosocial areas due to eating desire, difficulty communicating, sleep disturbance, and
neurologic or mechanical oropharyngeal dysphagia. These fatigue. The MDADI is a 20-item survey that assesses the
areas, divided into scales, address food selection, burden, impact of head and neck cancer treatment on QOL. The
 e66(4)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N

TABLE II Sample Responses to Open-Ended Questions About Effects of Surgery on Swallowing and Speech

Throat pain
Dependence on liquids
Modified eating habits
Fewer food options
Difficulty taking pills
Regurgitation
Limits social interactions
Emotional effects
Quality of speech
Interactions with others
Change in speaking
“It was sore. If I ate something toasty it made my throat scratchy. It was horrible, a nightmare!”
“I have pain; it feels like a very sore throat. Like someone is stepping on my neck.”
“Lots of water helps. The water is soothing. I drink 10-12 bottles of water a day.”
“I have to have a glass of water with everything; otherwise I would spit it right back up.”
“I get coughing spells when the portion is too large, the food isn't soft, or it is too hot or cold.”
“I chew more and chew longer.”
“I eat slower and smaller portions. I drink my soup through a straw.”
“Bread and meat don’t work at all. Applesauce and yogurt I can eat without complications.”
“I would choke especially with calcium pills that are big. My husband would crush them up and put them in
applesauce or sweet potatoes so I can eat them.”
“Pills are getting stuck and it takes minutes to get them down.”
“I have massive difficulties swallowing. I will cough up food or pills that I thought went down hours after swallowing
them.”
“Well I try not eating in other people’s company. I cough, I regurgitate. Sometimes I have to spit into a napkin…
that’s gross.”
“It was frustrating on a social level; you don't want to go out and it has an impact on what you want to eat.”
“I can't go out anymore. I have to eat with dozens of napkins around me so I can spit up food.”
“The nutrition person tried feeding me an apple. I felt like I couldn't breathe anymore. I get really anxious and scared
when I choke on food. I feel like I can't breathe.”
“I was hoarse because I felt like I had strep throat; you know how people sound, like that kind of hoarse.”
“I can't speak loud anymore at all. I can't yell.”
“I have to repeat words sometimes because my wife doesn't understand me. No particular words, just in general.”
“People have definitely noticed. It sounds real different.”
“I have short telephone calls with friends and family.”
“I would not speak for as long as I usually would. Talking felt like an effort.”
“I don't speak so loud and don't speak long sentences.”
“I hold my throat…when I talk.”

MDADI is divided into 4 scales: global impact of dysphagia and
impact of dysphagia on emotions, psychosocial function, and
eating. For both surveys, each scale is scored separately with a
possible range of 0 to 100; higher scores reflect better status.
Additionally, each scale is considered independently; an overall
score is not calculated.
At the second postoperative visit, patients were also asked
to rate their swallowing and speech compared with the first
postoperative visit. The response options were much worse,
worse, the same, better, and much better.
Scores for all 3 surveys at each postoperative visit were
calculated, and the percentages of patients scoring the mini-
mum and maximum values were ascertained. HSS-DDI ex-
ternal validity was established by comparison with each of the
SWAL-QOL and MDADI scales using Spearman correlation
coefficients with 95% confidence intervals (CIs). HSS-DDI
internal validity was verified by demonstrating that more ex-
tensive surgery (described below) was associated with worse
survey scores.
Responsiveness was established in several ways. First, the
within-patient change in score (postoperative visit-2 score minus
postoperative visit-1 score) for the HSS-DDI was compared with
that of the SWAL-QOL and that of the MDADI using Spearman
correlation coefficients with 95% CIs. Second, responsiveness
was ascertained by comparing the effect sizes of the 3 scales. The
effect size reflects the change in score as a function of the
variability of the score and is calculated as the mean change in
score divided by the standard deviation (SD) of that change.
Comparison of effect sizes among different scales reveals
which scale is more sensitive or responsive to change in a
sample. By convention, an effect size of 0.2 is small, 0.5 is
moderate, and ‡0.8 is large. Finally, responsiveness was as-
certained by comparing the change in score with the global
transition ratings of change in swallowing and speaking,
which were captured at the second postoperative visit.
Responses and score changes from the first to the second
postoperative visits were compared using repeated-measures
analysis of variance (ANOVA).
Clinical and surgical data, including preoperative
diagnosis and number of vertebral levels involved, were ob-
tained for patients at all phases by chart review. The mag-
nitude of surgery was assigned based on the spine-specific
Surgical Invasiveness Index (SII), a compendium of surgical
features accounting for the number of vertebrae involved,
 e66(5)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N

TABLE III Proportions of Patients Who Had Each Dysphagia and Dysphonia Item and Weighted Kappa Values

Dysphagia and Dysphonia Items Frequency (%) Weighted Kappa Value

Had throat pain when swallowing:

Saliva 73 0.52
Liquids 71 0.46
Moist solid foods 75 0.41
Dry solid foods 73 0.40
Had throat tightness when swallowing 82 0.32
Food got stuck in throat 67 0.49
Gagged on food 43 0.55
Gagged on pills 55 0.46
Regurgitated food or pills 39 0.50
Changed diet or types of food eaten 61 0.46
Took smaller bites 90 0.44
Chewed more than usual 90 0.52
Needed more time to eat 67 0.63
Drank more fluids than usual 63 0.40
Modified the way took pills 49 0.43
Coughed more 61 0.46
Was anxious about breathing and swallowing 57 0.39
Lost weight 55 0.48
Limited social interactions (avoided restaurants or eating with others) 55 0.47
Overall, was bothered by swallowing ability 75 0.53
Had throat pain when speaking 55 0.42
Voice not loud enough (could not project) 59 0.62
Voice too hoarse or raspy 59 0.55
Voice too deep 35 0.49
Difficulty being understood during conversations (had to repeat words) 31 0.71
Difficulty being understood during telephone conversations 43 0.61
Spoke less than wanted to 41 0.58
Others commented that speech was different 41 0.67
Limited social or professional interactions because of speech 33 0.44
Was frustrated by speaking ability 29 0.65
Overall, was bothered by speaking ability 29 0.67

decompression, fusion, instrumentation, and approach27. Points
are assigned for each feature for each vertebral level, and a total
score is calculated as the sum of points from all levels. For the
cervical spine, the possible score range is 0 to 48, with a higher
score indicating greater invasiveness. In our study, scores were
grouped according to quartiles.
Results
Phase 1: Identifying Items for Draft Survey
saturation was achieved. The major categories that were
identified were throat pain, dependence on liquids, modified
eating habits, fewer food options, difficulty taking pills,
regurgitation, limitations in social interactions, emotional
effects, quality of speech, interactions with others, and change in
speaking (Table II). Categories were then transformed into 20
dysphagia-related and 11 dysphonia-related items.
Phase 2: Assembling Draft Survey, Assessing Test-Retest

T wenty-five consecutive patients (Table I), all with in-

volvement of ‡3 vertebral levels, were enrolled at a mean
of 29 days (range, 14 to 41 days) after surgery until data
Reliability, Selecting Final Items, and Scoring
A draft survey was formatted so that the dysphagia and
dysphonia items were clustered separately. The items were
 e66(6)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
Fig. 1
Hospital for Special Surgery Dysphagia and Dysphonia Inventory (HSS-DDI).
2017 Hospital for Special Surgery. All rights reserved.
prefaced with the statement “During the past 2 weeks how
often did you experience the following?” Response options
ranged from “none of the time” (4 points) to “all the time”
(0 points).
Another group of 49 patients (Table I), with involve-
ment of up to 4 vertebral levels, were enrolled at a mean of
29 days (range, 14 to 38 days) after surgery. These patients
completed the survey at enrollment and again during a
telephone interview at a mean of 5 days (range, 3 to 11 days)
later. The dysphagia items were endorsed more often (39%
to 90%) than the dysphonia items (29% to 59%) (Table III).
All but 2 items had weighted kappa values of ‡0.40. These 2
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N
items (throat tightness when swallowing [kappa = 0.32] and
anxiety about breathing and swallowing [kappa = 0.39])
were frequently endorsed (82% and 57%, respectively)
and therefore were not eliminated. Thus, all items were re-
tained to form the 31-item HSS-DDI (Fig. 1). An overall
score is calculated as the sum of all responses (raw score)
divided by the maximum possible score (4 · 31 = 124)
multiplied by 100, and can range from 0 to 100. Higher
scores indicate better function, or less dysphagia and/or
dysphonia (Fig. 2).
The mean scores (and SDs) for the first and second
administrations of the HSS-DDI to the patients in phase 2 were
 e66(7)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N

Fig. 2
Scoring of Hospital for Special Surgery Dysphagia and Dysphonia Inventory (HSS-DDI).
2017 Hospital for Special Surgery. All rights
reserved.

TABLE IV Questionnaire Scores at First and Second Postoperative Visits and Effect Sizes

Difference Between First

First Postoperative Visit Second Postoperative Visit and Second Visits
% with % with % with % with Within-Patient
Mean Min. Max. Mean Min. Max. Change Effect
Survey Score* ± SD Score Score Score* ± SD Score Score in Score† P Value Size‡

HSS-DDI 71 ± 24 1 6 82 ± 20 0 14 11 ± 15 <0.0001 0.73

SWAL-QOL
Food selection 77 ± 30 6 47 88 ± 23 1 69 11 ± 23 <0.0001 0.48
Burden 67 ± 31 8 25 83 ± 28 5 61 16 ± 27 <0.0001 0.59
Mental health 80 ± 26 2 43 85 ± 26 2 63 5 ± 19 0.002 0.26
Social functioning 83 ± 25 2 54 92 ± 19 1 77 8 ± 18 <0.0001 0.44
Fear of choking 83 ± 21 1 39 88 ± 19 0 57 6 ± 15 <0.0001 0.40
Eating duration 72 ± 33 7 42 82 ± 28 2 58 9 ± 24 <0.0001 0.38
Eating desire 83 ± 29 6 57 91 ± 20 1 72 8 ± 19 <0.0001 0.42
Communication 87 ± 21 1 63 91 ± 20 0 77 4 ± 16 0.01 0.25
Sleep 49 ± 28 9 9 62 ± 31 6 22 13 ± 25 <0.0001 0.52
Fatigue 51 ± 26 4 6 61 ± 26 3 9 11 ± 22 <0.0001 0.50
MDADI
Global 84 ± 27 5 61 89 ± 23 2 76 6 ± 19 0.001 0.32
Emotional 84 ± 20 0 44 87 ± 18 0 49 2 ± 14 0.08 0.14
Functional 86 ± 19 0 50 89 ± 17 0 57 3 ± 13 0.01 0.23
Physical 76 ± 25 0 30 82 ± 23 0 43 6 ± 16 <0.0001 0.38

*The possible range is 0 to 100, with a higher score indicating better status. †Score from second postoperative visit minus score from first
postoperative visit. ‡Change in score divided by SD of change.
 e66(8)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N

67 ± 24 and 75 ± 22, the Cronbach alpha coefficients were both
0.97, and the ICC was 0.80.
Phase 3: Validation and Responsiveness
One hundred and twenty-seven patients (Table I), with
involvement of up to 5 vertebral levels, were enrolled in phase
3 during the first postoperative visit, at a mean of 29 days
(range, 9 to 44 days) after surgery. Patients completed the
HSS-DDI, SWAL-QOL, and MDADI during this visit and
again at the second postoperative visit at a mean of 31 days
(range, 8 to 89 days) later. The mean score on the HSS-DDI at
the first visit was 71, with few patients having minimum or
maximum values (Table IV). The mean scores for the dif-
ferent scales of the SWAL-QOL ranged from 49 to 87, and
more than a third collectively had the maximum score. This
scenario was more pronounced for the MDADI, with almost
half of the patients having the maximum score. The scores on
all surveys were higher at the second postoperative visit,
indicating clinical improvement.
The external validity of the HSS-DDI and each scale of
the SWAL-QOL and the MDADI was established with the
Spearman correlation coefficient to account for the distribu-
tions of the latter scales (Table V). Most coefficients were in the
0.5 to 0.7 range with narrow 95% CIs, and the HSS-DDI was
consistently correlated with the SWAL-QOL and MDADI.
Internal validity was established by comparison with the extent
of surgery. The HSS-DDI scores worsened (decreased) as the
number of vertebral levels involved increased (80 for 1 level, 72
for 2 levels, 61 for 3 levels, and 68 for ‡4 levels; p = 0.02).
Similarly, the HSS-DDI scores worsened as the SII increased
(80 for an SII of 5 to 8, 74 for 9 or 10, 61 for 11 or 12, and 67 for
13 to 15; p = 0.006).
When responsiveness was ascertained by calculating the
difference in HSS-DDI scores between the first and second
postoperative visits (Table IV) and comparing that value with
the differences in the SWAL-QOL and MDADI scores (Table V),
the Spearman correlation coefficients demonstrated a consis-
tent association for most comparisons. Also, the effect size of
the HSS-DDI (0.73) was markedly larger than those of the
SWAL-QOL and MDADI (range, 0.14 to 0.59), indicating that
the HSS-DDI captured more change as patients improved
clinically. Finally, in the analysis of responsiveness according to
the patients’ rating of the overall change in swallowing and
speech between postoperative visits, 5% rated swallowing to be
worse or much worse, 38% rated it as the same, and 57% rated
it as better or much better and the corresponding changes in
the HSS-DDI scores were 22, 7, and 15 (p = 0.002). Similarly,
2% rated their speech as worse or much worse, 63% rated it as
the same, and 35% rated it as better or much better and the
corresponding changes in the HSS-DDI scores were 23, 9,
and 16 (p = 0.007). The average administration times for the
HSS-DDI, SWAL-QOL, and MDADI were 2 minutes and
25 seconds, 5 minutes and 13 seconds, and 2 minutes and 13
seconds, respectively.

TABLE V Correlations Between HSS-DDI and the SWAL-QOL and MDADI

Scores from First Scores from Second

Postoperative Visit Postoperative Visit Change in Scores*
Correlation Correlation Correlation
Comparison Surveys Coefficient 95% CI Coefficient 95% CI Coefficient 95% CI

SWAL-QOL
Food selection 0.59† 0.46-0.69 0.71† 0.61-0.79 0.39† 0.23-0.53
Burden 0.68† 0.57-0.76 0.79† 0.71-0.85 0.50† 0.36-0.62
Mental health 0.71† 0.61-0.78 0.80† 0.72-0.85 0.42† 0.27-0.56
Social functioning 0.63† 0.51-0.72 0.63† 0.51-0.72 0.38† 0.22-0.52
Fear of choking 0.69† 0.58-0.77 0.77† 0.68-0.83 0.44† 0.29-0.57
Eating duration 0.72† 0.63-0.80 0.80† 0.72-0.85 0.55† 0.42-0.66
Eating desire 0.58† 0.45-0.68 0.66† 0.55-0.75 0.31‡ 0.15-0.46
Communication 0.63† 0.51-0.72 0.63† 0.51-0.72 0.51† 0.37-0.63
Sleep 0.42† 0.26-0.55 0.44† 0.28-0.57 0.29‡ 0.12-0.44
Fatigue 0.49† 0.35-0.61 0.51† 0.37-0.63 0.23§ 0.05-0.39
MDADI
Global 0.68† 0.58-0.77 0.63† 0.51-0.72 0.41† 0.25-0.54
Emotional 0.69† 0.58-0.77 0.69† 0.58-0.77 0.30‡ 0.13-0.45
Functional 0.70† 0.60-0.78 0.57† 0.44-0.68 0.32‡ 0.15-0.47
Physical 0.84† 0.78-0.88 0.85† 0.79-0.89 0.55† 0.41-0.66

*Score from second postoperative visit minus score from first postoperative visit. †P £ 0.0001. ‡P < 0.001, > 0.0001. §P = 0.01.
 e66(9)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
Discussion
I n this study, we developed a survey, using patient-derived
variables based on patient interviews and confirmed by a
spine surgeon panel, to specifically address dysphagia and
dysphonia outcomes after anterior cervical spine surgery. The
survey was reliable on the basis of repeated measurements. An
overall score reflects the amount of both dysphagia and dys-
phonia experienced. The HSS-DDI is unique in that it was
validated for measuring dysphagia and dysphonia after anterior
cervical spine surgery and it is more responsive to change after
cervical spine surgery than existing surveys. Thus, it offers a
standardized way to more accurately understand the prevalence
and natural history of this specific postoperative condition.
Furthermore, it should aid in better understanding mitigating
factors and future strategies to reduce the prevalence of these
complications.
The evaluation of dysphagia and dysphonia has been
widely studied but, in the anterior cervical spine surgery
population, this has been done either in a subjective manner
utilizing nonvalidated instruments or with instruments
designed to reflect chronic disease. An example is the widely
used but nonvalidated surgeon-derived Bazaz Dysphagia
Score, with which 4 categories of severity scores are calcu-
lated on the basis of symptom interpretation by clini-
cians6,10,28. In a study of 470 patients, Nagoshi et al. reported
the risk factors and prevalence of dysphagia based solely
on patient-reported swallowing difficulty and qualitative
assessment of symptoms29. The HSS-DDI presents a more
comprehensive evaluation of dysphagia and dysphonia
symptoms by minimizing potential confounders and sub-
jective interpretation.
Currently, dysphagia and dysphonia following anterior
cervical spine surgery are often assessed with instruments
derived from populations with chronic disease, which is
distinctly different from an acute postoperative state. The
SWAL-QOL is a validated, 2-arm survey with 44 questions that
describes dysphagia in a wide range of patients with oro-
pharyngeal dysphagia, including those with neurodegener-
ative disease, malignancy, traumatic injury, and vascular and
obstructive etiologies19. Consequently, some questions that
deal with chronic disease, such as utilization of a feeding tube,
generally do not apply to the spine surgery population. The
MDADI, published in a 2001 report by Chen et al.20, is a vali-
dated dysphagia-specific QOL questionnaire derived from
assessments of patients with oropharyngeal cancer. The impact
of a cancer diagnosis or progressive neurodegenerative disease
may present psychological factors that affect health-related
QOL30. The HSS-DDI was developed using the responses of
patients who had been selected according to narrow inclusion
criteria, creating an evaluation tool that was more specific for
the outcomes of anterior cervical spine surgery. Specific items
include throat tightness, additional fluid requirements, and
issues with taking pills, which can affect QOL and recovery
from anterior cervical spine surgery. These and other impor-
tant items are unique to the HSS-DDI and are not addressed by
existing surveys.
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N
The major strength of this study is that it is derived
from a narrowly defined patient population and therefore
better able to capture clinical changes in dysphagia and dys-
phonia after anterior cervical spine surgery. The patient-
derived approach permitted the inclusion of diverse dysphagia
and dysphonia items that clinicians may not spontaneously
select with internal consistency. An expert spine surgeon
panel confirmed our survey, which provided surgeon-specific
content validity. Throughout the study, the HSS-DDI scores
correlated with the SWAL-QOL and MDADI scores, provid-
ing external validity that was not subject to ceiling effects. The
survey is reliable on the basis of repeated measurement of
individual survey items and overall scores. The HSS-DDI
demonstrates internal validity, indicated by worsening of
scores as the number of cervical levels increased. The HSS-
DDI is time-efficient, with a shorter administration time
compared with the SWAL-QOL. The survey is also easy to
self-administer and well-received by patients. Over time, the
HSS-DDI captured more symptom change as patients
improved clinically and was more responsive than the SWAL-
QOL and MDADI.
This study had several limitations. Only English-speaking
patients were enrolled. The survey was administered to pa-
tients from a tertiary-care orthopaedic referral hospital, where
expectations and previous evaluations could influence expec-
tations regarding dysphagia and dysphonia. Patients were
recruited from the practices of 4 surgeons, which may not
reflect other practice settings. Additionally, in phase 2 of the
study, the first survey was conducted during an in-person
interview and the second was conducted over the telephone;
thus, the modality of survey administration potentially
affected patient responses. To minimize response and scor-
ing burden, the HSS-DDI focuses on types and frequency of
symptoms rather than ranking or individually weighting
symptoms. Finally, since the authors of this study are the
developers of the survey, an additional independent valida-
tion in an external cohort should be performed. Until then,
recommendations regarding utility and application should
be considered provisional.
In conclusion, we developed a valid and reliable patient-
derived survey instrument to assess dysphagia and dysphonia
following anterior cervical spine surgery. This novel survey
is condition-specific and more responsive over time than
other instruments commonly used to evaluate these com-
plications. The HSS-DDI is clinically relevant and useful for
prospective and longitudinal assessment of dysphagia and
dysphonia in patients treated with anterior cervical spine
surgery. n
NOTE: The authors thank Zachary S. Gulergun, Janina Kueper, Lukas P. Lampe, Mohamed Moawad,
and Oliver C. Sax for their assistance with data collection in this study.
Alexander P. Hughes, MD1
Stephan N. Salzmann, MD1
Okezie K. Aguwa, MD1
Courtney Ortiz Miller, BS1
 e66(10)
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 1 00-A N U M B E R 10 M AY 16, 2 018
d d
Roland Duculan, MD1
Jennifer Shue, MS1
Frank P. Cammisa, MD1
Andrew A. Sama, MD1
Federico P. Girardi, MD1
Ashutosh Kacker, MD2
Carol A. Mancuso, MD1
References
1. Marawar S, Girardi FP, Sama AA, Ma Y, Gaber-Baylis LK, Besculides MC, Mem-
tsoudis SG. National trends in anterior cervical fusion procedures. Spine (Phila Pa
1976). 2010 Jul 1;35(15):1454-9.
2. Marquez-Lara A, Nandyala SV, Fineberg SJ, Singh K. Current trends in demo-
graphics, practice, and in-hospital outcomes in cervical spine surgery: a national
database analysis between 2002 and 2011. Spine (Phila Pa 1976). 2014 Mar
15;39(6):476-81.
3. Papadopoulos EC, Huang RC, Girardi FP, Synnott K, Cammisa FP Jr. Three-level
anterior cervical discectomy and fusion with plate fixation: radiographic and clinical
results. Spine (Phila Pa 1976). 2006 Apr 15;31(8):897-902.
4. Daniels AH, Riew KD, Yoo JU, Ching A, Birchard KR, Kranenburg AJ, Hart RA.
Adverse events associated with anterior cervical spine surgery. J Am Acad Orthop
Surg. 2008 Dec;16(12):729-38.
5. Danto J, DiCapua J, Nardi D, Pekmezaris R, Moise G, Lesser M, Dimarzio P.
Multiple cervical levels: increased risk of dysphagia and dysphonia during anterior
cervical discectomy. J Neurosurg Anesthesiol. 2012 Oct;24(4):350-5.
6. Rosenthal BD, Nair R, Hsu WK, Patel AA, Savage JW. Dysphagia and dysphonia
assessment tools after anterior cervical spine surgery. Clin Spine Surg. 2016
Nov;29(9):363-7.
7. Smith-Hammond CA, New KC, Pietrobon R, Curtis DJ, Scharver CH, Turner DA.
Prospective analysis of incidence and risk factors of dysphagia in spine surgery
patients: comparison of anterior cervical, posterior cervical, and lumbar procedures.
Spine (Phila Pa 1976). 2004 Jul 1;29(13):1441-6.
8. Rihn JA, Kane J, Albert TJ, Vaccaro AR, Hilibrand AS. What is the incidence and
severity of dysphagia after anterior cervical surgery? Clin Orthop Relat Res. 2011
Mar;469(3):658-65.
9. Fengbin Y, Xinwei W, Haisong Y, Yu C, Xiaowei L, Deyu C. Dysphagia after anterior
cervical discectomy and fusion: a prospective study comparing two anterior surgical
approaches. Eur Spine J. 2013 May;22(5):1147-51. Epub 2013 Jan 1.
10. Bazaz R, Lee MJ, Yoo JU. Incidence of dysphagia after anterior cervical
spine surgery: a prospective study. Spine (Phila Pa 1976). 2002 Nov 15;27
(22):2453-8.
11. Yu S, Chen Z, Yan N, Hou T, He S. Incidence and factors predictive of dysphagia
and dysphonia after anterior operation with multilevel cervical spondylotic myelop-
athy. Clin Spine Surg. 2017 Nov;30(9):E1274-8.
12. Mehra S, Heineman TE, Cammisa FP Jr, Girardi FP, Sama AA, Kutler DI. Factors
predictive of voice and swallowing outcomes after anterior approaches to the cer-
vical spine. Otolaryngol Head Neck Surg. 2014 Feb;150(2):259-65. Epub 2013 Dec
23.
13. Winslow CP, Winslow TJ, Wax MK. Dysphonia and dysphagia following the
anterior approach to the cervical spine. Arch Otolaryngol Head Neck Surg. 2001
Jan;127(1):51-5.
14. Tervonen H, Niemelä M, Lauri ER, Back L, Juvas A, Räsänen P, Roine RP,
Sintonen H, Salmi T, Vilkman SE, Aaltonen LM. Dysphonia and dysphagia after
anterior cervical decompression. J Neurosurg Spine. 2007 Aug;7(2):124-30.
15. Edwards CC 2nd, Karpitskaya Y, Cha C, Heller JG, Lauryssen C, Yoon ST, Riew
KD. Accurate identification of adverse outcomes after cervical spine surgery. J Bone
Joint Surg Am. 2004 Feb;86(2):251-6.
HS S D Y S P H A G I A A N D D Y S P H O N I A I N V E N T O RY ( HS S - DDI )
FO L LOW IN G ANT E R IO R CERVI C A L FUS IO N
1Departmentof Orthopaedic Surgery, Hospital for Special Surgery,
New York, NY
2Department of Otolaryngology – Head and Neck Surgery, Weill Cornell
Medicine, New York, NY
ORCID iD for A.P. Hughes: 0000-0001-7293-9672
16. Starmer HM, Riley LH 3rd, Hillel AT, Akst LM, Best SRA, Gourin CG. Dysphagia,
short-term outcomes, and cost of care after anterior cervical disc surgery. Dyspha-
gia. 2014 Feb;29(1):68-77. Epub 2013 Aug 14.
17. Anderson KK, Arnold PM. Oropharyngeal dysphagia after anterior cervical spine
surgery: a review. Global Spine J. 2013 Dec;3(4):273-86. Epub 2013 Aug 30.
18. Riley LH 3rd, Vaccaro AR, Dettori JR, Hashimoto R. Postoperative dysphagia in
anterior cervical spine surgery. Spine (Phila Pa 1976). 2010 Apr 20;35(9)(Suppl):
S76-85.
19. McHorney CA, Bricker DE, Robbins J, Kramer AE, Rosenbek JC, Chignell KA. The
SWAL-QOL outcomes tool for oropharyngeal dysphagia in adults: II. Item reduction
and preliminary scaling. Dysphagia. 2000 Summer;15(3):122-33.
20. Chen AY, Frankowski R, Bishop-Leone J, Hebert T, Leyk S, Lewin J, Goepfert H.
The development and validation of a dysphagia-specific quality-of-life questionnaire
for patients with head and neck cancer: the M. D. Anderson Dysphagia Inventory.
Arch Otolaryngol Head Neck Surg. 2001 Jul;127(7):870-6.
21. Mancuso CA, Cammisa FP, Sama AA, Hughes AP, Ghomrawi HMK, Girardi FP.
Development and testing of an expectations survey for patients undergoing lumbar
spine surgery. J Bone Joint Surg Am. 2013 Oct 2;95(19):1793-800.
22. Mancuso CA, Wentzel CH, Ghomrawi HMK, Kelly BT. Hip preservation surgery
expectations survey: a new method to measure patients’ preoperative expectations.
Arthroscopy. 2017 May;33(5):959-68. Epub 2016 Dec 31.
23. Mancuso CA, Cammisa FP, Sama AA, Hughes AP, Girardi FP. Development of an
expectations survey for patients undergoing cervical spine surgery. Spine (Phila Pa
1976). 2013 Apr 20;38(9):718-25.
24. Cody EA, Mancuso CA, MacMahon A, Marinescu A, Burket JC, Drakos MC,
Roberts MM, Ellis SJ. Development of an expectations survey for patients under-
going foot and ankle surgery. Foot Ankle Int. 2016 Dec;37(12):1277-84. Epub 2016
Sep 20.
25. Donner A, Eliasziw M. Sample size requirements for reliability studies. Stat
Med. 1987 Jun;6(4):441-8.
26. Kramer MS, Feinstein AR. Clinical biostatistics. LIV. The biostatistics of con-
cordance. Clin Pharmacol Ther. 1981 Jan;29(1):111-23.
27. Mirza SK, Deyo RA, Heagerty PJ, Konodi MA, Lee LA, Turner JA, Goodkin R.
Development of an index to characterize the “invasiveness” of spine surgery: vali-
dation by comparison to blood loss and operative time. Spine (Phila Pa 1976). 2008
Nov 15;33(24):2651-61; discussion 2662.
28. Koreckij TD, Davidson AA, Baker KC, Park DK. Retropharyngeal steroids and
dysphagia following multilevel anterior cervical surgery. Spine (Phila Pa 1976).
2016;41(9):E530-4.
29. Nagoshi N, Tetreault L, Nakashima H, Arnold PM, Barbagallo G, Kopjar B,
Fehlings MG. Risk factors for and clinical outcomes of dysphagia after anterior
cervical surgery for degenerative cervical myelopathy: results from the AOSpine
International and North America studies. J Bone Joint Surg Am. 2017 Jul 5;99
(13):1069-77.
30. Skeppholm M, Ingebro C, Engström T, Olerud C. The Dysphagia Short Ques-
tionnaire: an instrument for evaluation of dysphagia: a validation study with
12 months’ follow-up after anterior cervical spine surgery. Spine (Phila Pa 1976).
2012 May 15;37(11):996-1002.