Acta Pædiatrica ISSN 0803–5253

REGULAR ARTICLE

The reliability of the Neonatal Oral-Motor Assessment Scale

Saakje P da Costa ([email protected])1 , Cees P van der Schans2
1.School for Health Care Studies, Hanze University Groningen, University for Applied Sciences, Eyssoniusplein 18, 9714 CE Groningen, the Netherlands
2.Centre for Applied Research and Innovation in Health Care and Nursing, Hanze University Groningen, Groningen, the Netherlands

Keywords Abstract
Oral feeding performances, Oral motor problems,
Objectives: Sucking problems in preterm infants can be specified by means of visual observation. The
Preterm infants, Sucking skills, Swallowing
Neonatal Oral-Motor Assessment Scale (NOMAS) is the visual observation method most commonly
Correspondence
S. P. da Costa, School for Health Care Studies, used to assess the non-nutritive sucking (NNS) and nutritive sucking (NS) skills of infants up to
Hanze University Groningen, University for approximately 8 weeks postterm. During the first 2 min of a regular feeding the infant’s sucking skill is
Applied Sciences, Eyssoniusplein 18, assessed, either immediately or on video. Although NOMAS has been used since 1993, little is
NL-9714 CE Groningen, the Netherlands.
Tel./Fax: +31 (0)50 526 8776 | known about the method’s reliability. The aim of our study was to determine the test-retest and
Email: [email protected] inter-rater reliability of NOMAS.
Received Methods: The 75 infants included in this study were born at 26–36 weeks postmenstrual age (PMA).
16 February 2007; revised 26 April 2007; accepted Four observers participated in the study. They were trained and certified to administer NOMAS in the
15 October 2007.
Netherlands by M.M. Palmer between 2000 and 2002.
DOI:10.1111/j.1651-2227.2007.00577.x
Results: We found the test-retest agreement of NOMAS to be ‘fair’ to ‘almost perfect’ (Cohen’s kappa
[] between 0.33 and 0.94), whereas the inter-rater agreement with respect to the diagnosis was
‘moderate’ to ‘substantial’ (Cohen’s , between 0.40 and 0.65). As a diagnostic tool, however, the
current version of NOMAS cannot be used for both full-term and preterm infants. For a measuring
instrument such as NOMAS, one should aim at reliability coefficients for inter-rater and test-retest
agreement of at least 0.8. A Cohen’s  of 0.6 or less we find unacceptable. Nonetheless, by
observing sucking and swallowing according to a protocol much useful information can be gathered
about the development of an infant’s sucking skills. For instance, whether the infant is able to
co-ordinate sucking and swallowing, whether the infant can maintain sucking, swallowing and
breathing during the continuous phase and whether the infant is able to suck rhythmically with
equally long bursts. In addition, NOMAS offers useful aids for intervention.
Conclusions: NOMAS should be re-adjusted in order to improve inter-rater agreement, and at the same time
current insights into the development of sucking and swallowing should be incorporated in the method.

In case of low birthweight in addition to prematurity these

INTRODUCTION
problems are often more serious and longer lasting, particu-
Feeding problems occur frequently in preterm infants dur-
larly in the case of gastrointestinal disorders and if medical
ing their first year of life (1), particularly in infants of ges-
interventions such as artificial ventilation had been neces-
tational age (GA) of 32 weeks or less (2–4). However, the
sary (3,4). Difficulties during feeding may also lead to insuf-
exact prevalence of feeding problems in preterm infants is
ficient intake. Insufficient intake, especially in the case of a
unknown. In the case of preterm infants, feeding difficul-
newborn that is ill, may lead to tension on the part of the
ties usually have a medical cause (gastrointestinal, neuro-
caregiver. And tense interactions between the infant and his
logical or pulmonary) due to immaturity and diseases of
environment could be a breeding ground for behavioural and
one or more organ systems and often painful, but medi-
parent-related feeding problems in the long run. For these
cally necessary interventions in the infant’s face, mouth and
reasons it is important to intervene as quickly as possible
throat region related to these problems. Infants born prior to
and to find out whether feeding problems persist over time,
34-week GA suffer more gastrointestinal and oral sensory
or recover.
problems, such as abnormal oral reflex activity (1,3,4).
Sucking and swallowing movements of newborns can
Most feeding difficulties in preterm infants are caused by
be assessed in different ways. In the case of direct assess-
immature or inadequate coordination of the sucking, swal-
ment, sucking and swallowing can be described by means of
lowing and breathing sequence. In cases of impaired coordi-
various measures such as measuring saturation, heart rate,
nation, liquid may be aspirated into the trachea and so into
pharyngeal pressure, breathing pattern, and the duration of
the lungs. Aspiration may occur with no observable signs.
inhaling and exhaling (2,6–9). On the basis of these assess-
In some cases infants may choke, be short of breath or have
ments conclusions may be drawn regarding the coordination
disorders of the respiratory tract, and a decrease in oxygen
of breathing and swallowing, sucking pressure, efficiency,
saturation, apnoea and bradycardia may occur (5).
frequency and duration, and the respiratory phase in which


C 2007 The Author(s)/Journal Compilation 
C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 21–26 21
NOMAS reliability Costa and Schans

swallowing occurs. A drawback of these invasive measuring swallows followed by a 3–5 sec pause. Therefore, sucking
techniques is the impact they have on the ill newborn, such becomes less stable and more difficult to assess. In the case
as tubes down the infant’s throat to measure pressure, and of preterm infants (approximately until full-term age), the
the complex measuring and analyzing instruments necessary continuous phase only lasts about 30 sec, influenced by
to generate the data. neurologic function and cardiorespiratory control (2).
Problems with sucking and swallowing can also be spec- During observation by using NOMAS the researcher does
ified by means of indirect observation. We can distinguish not touch the infant nor is the infant attached to any measur-
between clinical feeding assessment and swallowing assess- ing apparatus. If the infant is too sleepy or does not want to
ment (9). The standardized assessment methods available drink for another reason (such as stomach cramps or distrac-
to assess nutritive sucking (NS) or oral feeding skills are tions in its surroundings), the attempt is postponed to a next
presented in Table 1 (11–19). Most of these methods can be feeding time. The number of sucking movements during one
used either for observing bottle-fed infants (14) or for breast- sucking burst is counted and the duration of the pauses be-
feeding (11–13,15–17). Five methods, including the Neonatal tween bouts of sucking is noted. Jaw and tongue movements,
Oral-Motor Assessment Scale (NOMAS) and the analysis of such as the degree and rhythm of jaw lowering and tongue
feeding behaviour with direct linear transformation (DLT) cupping, are analyzed on the basis of 28 items and entered
can be used for observing both breastfeeding and bottle- on the NOMAS form (Table S1). Even though NOMAS may
feeding (11,16–19) infants. The fact that the markers on the be used during breastfeeding as well as bottle-feeding, it may
infant’s face have to be placed very carefully and the fact be more difficult to administer during breastfeeding because
that a DLT procedure is used, are probably the main reasons of the flow: infants adjust their way of swallowing to the flow
why the latter method is still little used. of their mother’s milk (11,20). This results in jaw movements
NOMAS (11,20), a visual observation method, is a much of varying speed and magnitude. As a consequence, our clin-
used, noninvasive instrument to assess the NS and NNS ical observation was that the infants’ jaw movements could
skills of infants up to the age of about 8 weeks postterm erroneously be scored as disorganized.
(Table S1). NOMAS allows infant sucking to be divided into
three categories on the basis of the 28 items on the scale. METHOD
In 2004 we started a study on the development of swallow-
• A normal sucking pattern is displayed by infants who can ing in preterm infants. Seventy-five infants were included in
coordinate sucking, swallowing and breathing properly the study: 15 were at risk for bronchopulmonary dysplasia,
during both NNS and NS. 17 were extremely low birthweight preterms and 20 were
• A disorganized sucking pattern can be observed in in- healthy preterms. The control group comprised 23 healthy
fants who are unable to coordinate sucking, swallowing full-term infants. The preterm infants were born at 26–36
and breathing. This pattern is displayed by newborns who week GA. We excluded infants from the study who suffered
suffer from breathing problems, infants with a heart con- severe multiple congenital disorders, severe predispositional
dition or infants with gastrointestinal problems. Before cerebral disorders and periventricular echo densities with
reaching term, preterm infants usually display immature cysts. In addition, infants of drug-addicted mothers were
sucking patterns that match their age. If this sucking pat- also excluded. We examined each infant 10–12 times: once
tern is seen after term it is considered abnormal. There- a week between the ages of 34 and 40 weeks postmenstrual
fore, the infant’s age is an important element to take into age (PMA) and once a fortnight between 40 and 50 weeks
account before diagnosing a sucking pattern as disorga- PMA. The reliability study was part of the first phase of a
nized. research project on the development of sucking patterns in
• A dysfunctional sucking pattern is displayed by infants preterm infants and its relationship with neurodevelopmen-
whose motor reactions and jaw and tongue movements tal outcome at 2 and 5 years of age.
are abnormal and therefore inadequate, as is the case in Four NOMAS observers participated in our study. They
infants with neurological (or anatomical) disorders. had been trained and certified by M.M. Palmer in the Nether-
lands between 2000 (observers A and B) and 2002 (ob-
The infant’s sucking skill is assessed during NNS and dur- servers C and D). In order to qualify for a certificate the
ing the first 2 min of a regular feeding, either immediately or assessor is required to correctly assess all three diagnoses on
recorded on video for assessment later on. five NOMAS video recordings (i.e. a 100% correct classifi-
Many authors (2,7,8,11,21) indicate that full-term infants cation into the categories normal, disorganized or dysfunc-
have a continuous sucking phase during the first 2–3 min. In tional), and to obtain 80% agreement on all 28 items per
this phase the oral reflex activity is present most strongly and recording (22). Due to practical reasons (illness or pressure
the sucking bursts are most stable (the sucking-swallowing- of work), the four observers were unable to perform the same
breathing rhythm). After 2 min, due to gastrointestinal number of assessments. Although A observed 54 recordings
influences—the stomach filling up so the infant feels less hun- and B 126, they observed 50 of the same recordings together.
gry and a reduction of the oral reflex activity—the contin- Observer C observed 71 recordings and D 42, and they ob-
uous sucking phase is replaced by the intermittent phase. served 20 recordings together. The four observers together
This phase is characterized by bursts of sucking and a few assessed a total of 293 recordings.

22 
C 2007 The Author(s)/Journal Compilation 
C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 21–26
Costa and Schans NOMAS reliability

Table 1 Assessments of infant oral-sensorimotor function for feeding

Assessment Description

1. Neonatal Oral-Motor Assessment Scale (NOMAS); Palmer, Crawley Checklists of behaviours in categories of normal, disorganized, and
and Blanco, 1993 (10) dysfunctional tongue and jaw movements. From birth up to 8 weeks’
corrected age.
2. Systematic assessment of the infant at the breast (SAIB); Shrago and Observations related to alignment, areolar grasp, areolar compression and
Bocar, 1990 (11) audible swallow.
3. Preterm Infant Breastfeeding Behaviour Scale (PIBBS); Nyqvist et al., Diary kept by mother: rooting, amount of breast in mouth, latching, sucking,
1996 (12) sucking bursts, swallowing state, letdown and time.
4. Breastfeeding evaluation for term infants; Tobin, 1996 (13) Purpose: to identify when a mother would benefit from lactation support. List
of expectations for feedings. Full-term infants in the neonatal intensive care
unit.
5. Bottle-feeding flow sheet; Van den Berg, 1990 (14) Observations of state, respiratory rate, heart rate, nipple, form of nutrition,
position, coordination, support quantity and duration changes over time.
6. Infant feeding evaluation; Swigert, 1998 (15) Nonstandardized evaluation: means of documenting a variety of observations,
including infants’ responses to attempted interventions. Devised for birth
to 4 months, components for preterm or ill infants not specified.
7. Semi-demand feeding method for healthy preterm infants; McCain, The method combines the use of non-nutritive sucking to promote waking
2003 (16) behaviour for feeding, use of behavioural assessment to identify readiness
for feeding, and systematic observation of and response to infant
behavioural cues to regulate frequency, length and volume of oral feedings.
8. Early feeding skills assessment for preterm infants (EFS); Thoyre, A checklist for assessing infant readiness for and tolerance of feeding and for
Shaker and Pridham, 2005 (17) profiling the infant’s developmental stage regarding specific feeding skills.
9. Analysis of feeding behaviour with direct linear transformation; Mizuno By placing markers on the lateral angle of the eye, tip of the jaw and throat
et al., 2005 (18) during sucking while the face of the infant is recorded in profile, the jaw
and throat movements are calculated using the direct linear transformation
(DLT) procedure.

Adapted from Rogers and Arvedson, 2005 (10).

Following Palmer’s method, a video recording was made Table 2 Interpretation of Cohen’s kappa () values between 0 and 1 (26)
of the infants at different ages during the first 2 min of NS. Value of  Strength of agreement
We stored the recordings on a digital videodisc and two
NOMAS assessors assessed each recording. Subsequently, 0.00–0.20 Slight
we determined the test-retest and inter-observer reliability. 0.21–0.40 Fair
In contrast to Palmer, we determined the reliability of the 0.41–0.60 Moderate
0.61–0.80 Substantial
diagnoses and not that of the items. On average, the four
0.81–1.00 Almost perfect
assessors assessed 70 recordings twice with an interval of
3 months between assessments. The data of the first assess-
ment were not available to them on the occasion of the sec-
ond assessment.
RESULTS
For test-retest agreement (Table 3) there was a considerable
difference between assessor A with the highest score ( =
Statistical analysis 0.948) and D with the lowest score ( = 0.331). Thus intra-
Assessor agreement is defined by Popping as ‘sameness of rater agreement ranged from ‘fair’ to ‘almost perfect’. With
classification’ (24). According to Popping, Cohen’s kappa average reliability coefficients of 0.67, the test-retest reliabil-
(), that is, ‘the proportion of agreement after chance agree- ity of assessors B and C was ‘substantial’.
ment is removed from consideration’ (24), is the best mea- We were curious to know whether there was a differ-
sure to determine agreement between assessors in case of ence in reliability between the assessments of recordings
the a posteriori method of coding nominal data. As shown of preterm infants as compared to those of full-term in-
in Table 2, a reliability coefficient of 0.60 is considered the fants. The reason being that it is perhaps easier to assess
minimum for acceptable assessor agreement, whereas  = a mature sucking pattern than it is to assess an imma-
0.80 or higher is considered ‘almost perfect’ or ‘satisfactory’ ture sucking pattern (see Table 3). Although the number
(24–26). Although no absolute definitions are possible, the of the observations was incomplete, making it impossible
following guidelines should help: Cohen’s  is determined to do a comparison based on figures, we found no indi-
between two observers and between two viewings of the cation that there was a difference between the intra-rater
same recording by each assessor. agreement of the preterm infants and that of the full-term


C 2007 The Author(s)/Journal Compilation 
C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 21–26 23
NOMAS reliability Costa and Schans

Table 3 A comparison of the intrarater agreement between recordings of preterm and full-term infants (number of observations)

Assessors Total Preterm infants Full-term infants

Kappa Number of observations Kappa Number of observations Kappa Number of observations

A 0.948 54 1.00 31 0.841 23

B 0.694 126 0.685 77 0.718 49
C 0.659 71 0.752 37 0.630 34
D 0.331 42 na 13 0.410 29

na = not available.

Table 4 A comparison of the inter-rater agreement between recordings of preterm and full-term infants (number of observations)

Total Preterm infants Full-term infants

Kappa Number of observations Kappa Number of observations Kappa Number of observations

A vs. B 0.406 50 0.484 26 0.385 24

C vs. D 0.652 20 0.714 16 na 4

na = not available.

infants. In the case of inter-rater agreement (Table 4), as- sucks per burst continues to vary throughout the duration
sessors C and D had assessed less than half of the record- of sucking. There may also be intraburst variability as the
ings together due to the practical reasons mentioned above. sucking-swallowing-breathing ratio changes (20, p. 74).
Our results in Table 3 show that assessors A and B agreed In case of a segment of sucking counting less than 10
with each other less often than did C and D. The interpreta- sucking-swallowing-breathing movements, it is classified as
tion of the reliability coefficients ranged from ‘moderate’ to ‘arrhythmic jaw movement’ also if it occurs towards the
‘substantial’. end of the 2-min observation segment. In the meantime it
has become clear, however, that in the case of preterm in-
DISCUSSION fants it is not realistic to take a 2-min observation segment
We found the test-retest agreement of NOMAS with respect as point of departure before they have reached term age
to the diagnosis to be ‘fair’ to ‘almost perfect’ (Table 4), because a continuous phase in these infants only lasts 30 sec.
whereas the inter-rater agreement with respect to the di- Some assessors diagnose such situations as normal because
agnosis was ‘moderate’ to ‘substantial’ (Table 4). The reason the overall impression of sucking is normal.
for the ‘moderate’ inter-rater reliability possibly lay in the One of our concerns about using NOMAS as a diagnostic
lack of agreement in scoring the separate items and/or in tool is that since NOMAS was developed in 1993 many stud-
the interpretation of some items belonging to the diagnosis ies have been published that describe the nutritive and non-
‘disorganization’. It is remarkable that the items that score nutritive aspects of sucking. We compared Palmer’s findings
lowest in Palmer’s study are the same items that caused con- as set out in NOMAS with recent studies on sucking and the
fusion and disagreement in our study. What struck us was development of sucking, swallowing and sucking patterns.
that one assessor would attach a different diagnosis to the Four questions arose regarding several aspects of NOMAS.
same score than would the other assessor. In particular, this (1) Palmer indicates that NOMAS ought to be adminis-
was the case for the items ‘inconsistent jaw degree’ and ‘ar- tered for at least 2 min (11). More recently she suggested
rhythmic jaw/tongue movements’: that NOMAS be administered for at most 2 min because the
(1) ‘Inconsistent jaw degree’: The degree of jaw opening continuous phase of sucking lasts 2 min (23). Mizuno et al.
that occurs during the suction component can be noted to found a continuous phase of 30 sec in preterm infants (2).
vary each time, causing jaw excursions to be of unequal size Does this imply that for the assessment of sucking pattern
(20, p. 74). in preterm infants NOMAS should only be administered for
During different courses Palmer issued different state- 30 sec?
ments on this point. During the course she offered in the (2) Palmer mentions ‘10–30 suck/swallows per burst’ as
Netherlands in May 2006, she stated that the diagnosis ‘dis- being part of a mature sucking pattern (11, p. 28). She states
organization’ might not be given in the presence of this item that:
alone (pers. comm.).
(2) ‘Arrhythmic jaw movements’: During a 2-min timed • The interburst variation should be stable’.
segment of sucking, the jaw movements that occur are jerky, • ‘Ten or more sucks per burst means a mature sucking pat-
inconsistent, irregular and do not flow in a coordinated way. tern, less than 10 sucks per burst is abnormal and is not
Sucking bursts are of unequal length, and the number of part of a mature sucking pattern’.

24 
C 2007 The Author(s)/Journal Compilation 
C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 21–26
Costa and Schans NOMAS reliability

Palmer does not mention a development in the number reliability of the method with regard to the diagnosis, as
of sucking movements per sequence, nor does she specify was our aim. The inter-rater agreement of all 26 items, that
whether there is a quantitative difference between the num- is, 13 items dealing with the functioning of the tongue and
ber of movements an infant shows in its sucking pattern. 13 items dealing with the functioning of the jaw, is expressed
Recently, Qureshi et al. spoke of an average of 10 sucking in percentage agreement and ranges between 63% and 100%.
movements per sequence at term and of 20 sucking move- The score on 17 of the 26 items is 80% or higher. Subse-
ments per sequence at 1 month postterm (7). It seems ad- quently, Palmer revised NOMAS. She added one item to
visable to consider the results of the study by Qureshi et al. category ‘dysfunction’, she subdivided two items into three
when using NOMAS. subitems each, she transferred one item from category ‘dis-
(3) Palmer only speaks of a 1:1:1 rhythm when consid- organized’ to ‘dysfunctional’ and she redefined one item
ering bottle-feeding and indicates a nonrhythmic intraburst (Tables S1 and S2). The reliability of the revised version
as abnormal and one that should be scored as disorga- was not investigated. The large range in agreement be-
nized. In breastfeeding, rhythm depends on the flow and tween the assessors made it impossible to say anything about
a nonrhythmic intraburst (e.g. suck-swallow-breath suck- the reliability of the classifications by the instrument as a
suck-swallow-breath) is not abnormal and should not be whole. Moreover, Palmer’s study did not take into account
diagnosed as ‘disorganization’ (11,20). agreement based on chance as determined by, for instance,
Qureshi et al. concluded that during the first month of life, Cohen’s .
infants develop from a 1:1:1 suck-swallow-breath rhythm to In conclusion, the following issues need to be addressed:
a 2:1:1 or 3:1:1 rhythm, thus displaying their increased skill NOMAS requires adjustment as far as the instructions
to collect a larger amount of food in the valleculae that is about the interpretation of the items is concerned. At present
swallowed at once (7). the interpretation and/or classification of the items (espe-
Palmer does not mention the infant’s ability to collect food cially with regard to the diagnosis ‘disorganization’) is not
from a number of sucking movements as part of the matura- consistent. In addition, a clear distinction should be made
tion process. It signifies the first step towards developing a between the interpretation in the case of bottle-feeding and
new way of feeding. We advise noting the number of sucking breastfeeding.
movements per burst when using NOMAS (11,20). And, in As far as the diagnosis ‘disorganization’ is concerned,
accordance with Qureshi, we advise not to regard a rhythm the emphasis should lie on the fact that breathing is not
different from 1:1:1 as abnormal. coordinated with sucking and swallowing. Taking into ac-
(4) In her publications Palmer points out that NOMAS in- count the extent to which sucking behaviour is diagnosed
forms us about the jaw and tongue movements during suck- as disorganized seems meaningful when assessing preterms.
ing, about the coordination of sucking-swallowing-breathing In so doing it is possible during follow-up to better assess
and about the difference between nutritive and non-nutritive the development of sucking behaviour and the necessity of
sucking. She also suggests noting the bolus volume the infant intervention.
ingests during the 2 min of NOMAS administration (23). Ac- The length of the time segment to be measured, either
cording to Qureshi, during the first month of life, the amount preterm or postterm, should be determined on the basis of
of cm3 per swallow doubles and the number of swallowing Mizuno’s recent data on the continuous phase prior to term
movements increases to 46–50 per minute (7). We recom- age (2).
mend counting the number of swallowing movements per According to Qureshi, NOMAS should be extended with
minute as a measure of swallowing efficiency. the fact that at term an infant should be able to do 10 sucking-
Palmer states that NOMAS has predictive value (22). She swallowing-breathing movements per burst and at 4 weeks
bases this statement on the finding that 9 out of 34 infants of age this should have increased to approximately 20 (7). If
who had a dysfunctional sucking pattern in infancy had de- an infant is unable to do this, this fact should be incorporated
veloped abnormally when they were re-examined at 2 years in the diagnosis. The number of swallowing movements per
of age. The follow-up study included only 18 of the origi- minute should count as a measure for increased efficiency
nal 34 children, and the result does not specify the degree of sucking and swallowing.
of abnormal functioning at the age of 2. In our opinion, to Moreover, Qureshi recommends that the diagnosis ‘disor-
say that NOMAS has predictive value on the basis of this ganization’ should not be based on intraburst arrhythmic-
evidence, is insufficient. Nevertheless, it appears that prac- ity. In the case of this diagnosis, care should be taken with
titioners set great store by the value that the diagnosis ‘dys- interburst arrhythmicity (7). Until such adjustments come
function’ may have regarding expectations of neurodevelop- into effect, NOMAS can be used for detailed observation of
mental outcome at a later age. an infant’s sucking pattern for purposes of intervention but
NOMAS is used mainly for full-term infants with suck- not for diagnoses because especially in the case of preterm
ing and swallowing difficulties. Even though it has been in infants, the differentiation into three diagnoses is not suffi-
use since 1993, little is known about the instrument’s in- ciently reliable if the assessment is performed by different
trarater and test-retest agreement. Palmer (11) studied inter- observers. We recommend testing the intraobserver reliabil-
rater agreement of each NOMAS item in 35 infants aged ity of NOMAS observers. In addition, we advise against in-
35–49 weeks and weighing more than 1900 g at the time volving more than one assessor in the longitudinal follow-up
of assessment (23–42 week GA). Palmer did not study the of one and the same infant.


C 2007 The Author(s)/Journal Compilation 
C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 21–26 25
NOMAS reliability Costa and Schans

In case NOMAS is used as a means to measure neurode- 15. Vandenberg KA. Behaviorally supportive care for the
velopmental outcome for research purposes, we recommend extremely premature infant. In:Gunderson L, Kenner C,
editors. Care of the 24–25 week gestational age infant (small
baby protocol). San Francisco, CA: Neonatal Network, 1990;
• that each measurement be assessed by two reliable asses- 129–57.
sors, and 16. Swigert N. The Source® for Pediatric Dysphagia. 1998; San
• to reach a consensus in case of absence of agreement. Diego, CA: Singular.
17. McGain GC, Gartside PS. Behavioral responses of preterm
Because the test-retest agreement is not sufficient for ev- infants to a standard-care and semi-demand feeding protocol.
eryone, the individual observer should be tested regularly Newborn Infant Nurs Rev 2002; 2: 187–93.
18. Thoyre SM, Shaker CS, Pridham KF. The early feeding skills
and receive extra training if need be. We expect the inter-
assessment for preterm infants. Neonat Netw 2005; 24.3: 7–16.
rater agreement to improve when the test-retest agreement 19. Mizuno K, Aizawa M, Saito S, Kani K, Tanaka S, Kawamura
increases, and when the instrument is adjusted. Our point H, et al. Analysis of feeding behavior with direct linear
of departure is that the test-retest and inter-rater agreement transformation. Early Hum Dev 2006; 82: 199–204.
of measures such as NOMAS should have a Cohen’s  of at 20. Palmer MM. Identification and management of the
least 0.8. As far as we are concerned a Cohen’s  of 0.6 or transitional suck pattern in premature infants. J Perinat
lower is unacceptable. Neonatal Nurs 1993; 7: 66–75.
21. Lau C, Alagugurusamy R, Schanler RJ, Smith EO, Shulman RJ.
Characterization of the developmental stages of sucking in
References preterm infants during bottle-feeding. Acta Paediatr 2000; 89:
846–52.
1. Hawdon JM, Beauregard N, Slattery J, Kennedy G.
22. Palmer MM. Developmental Outcome for neonates with
Identification of neonates at risk of developing feeding
dysfunctional and disorganized sucking patterns: preliminary
problems in infancy. Dev Med Child Neurol 2000; 42: 235–9.
findings. Transdiscip J 1999; 3: 299–308.
2. Mizuno K, Ueda A. The maturation and coordination of
23. Palmer MM. Personal communication. First Dutch
sucking, swallowing and respiration in preterm infants.
Certification Course. WKZ/UMCU Utrecht 2000.
J Paediatr 2003; 1: 36–40.
24. Popping R. Overeenstemmingsmaten voor nominale data.
3. Rommel N. Diagnosis of oropharyngeal disorders in young
Thesis, Rijksuniversiteit Groningen, 1983.
children. Thesis, University of Leuven, 2002.
25. Cohen J. A coefficient of agreement for nominal scales. Educ
4. Engel-Hoed Van Den J. Eet-en drinkproblemen bij jonge
Psychol Meas 1960; 20: 37–46.
kinderen. Acco 2006: 24–26.
26. Landis JR, Koch GG. The Measurement of observer agreement
5. Garg BP. Dysphagia in children: an overview. Semin Pediatr
for categorical data. Biometrics 1977; 33: 159–75.
Neurol 2003; 10: 252–4.
6. Mathew OP. Science of bottle feeding. J Pediatr 1991; 119:
511–19.
7. Qureshi MA, Vice FL, Taciak VL, Bosma JF, Gewolb IH.
Changes in rhythmic suckle feeding patterns in term infants in Supplementary material
the first month of life. Dev Med Child Neurol 2002; 44.1: 34–9. The following supplementary material is available for this
8. Lau C, Kusnierczyk I. Quantitative evaluation of infant’s
article:
non-nutritive and nutritive sucking. Dysphagia 2001; 16.1:
58–67.
Table S1 Neonatal Oral-Motor Assessment Scale (NO-
9. Gewolb IH, Vice FL. Maturational changes in the rhythms,
patterning, and coordination of respiration and swallow MAS), 1993 revision (Copyright  C 1990 Marjorie Meyer

during feeding in preterm and term infants. Dev Med Child Palmer)
Neurol 2006; 48: 589–94. Table S2 Neonatal Oral-Motor Assessment Scale (NOMAS)
10. Rogers B, Arvedson J. Assessment of infant oral sensorimotor original 1990 version (Copyright C 1990 Majorie Meyer

and swallowing function. Ment Retard Dev Disabil Res 2005; Palmer)
11.1: 74–82.
11. Palmer MM, Crawley K, Blanco IA. Neonatal Oral-Motor This material is available as part of the online article from:
Assessment Scale: a reliability study. J Perinatol 1993; XIIL: http://www.blackwell-synergy.com/doi/abs/10.1111/j.1651-
28–34.
2227.2007.00577.x
12. Shrago L, Bocar D. The infant’s contribution to breastfeeding.
J Obstet Gynecol Neonatal Nurs 1990; 19: 209–15. (This link will take you to the article abstract).
13. Nyqvist KH, Rubertsson C, Ewald U, Sjoden PO.
Development of the Preterm Infant Breast-feeding Behavior
Please note: Blackwell Publishing is not responsible for the
Scale (PIBBS): a study of nurse-mother agreement. J Hum content or functionality of any supplementary materials sup-
Lact 1996; 12: 221–8. plied by the authors. Any queries (other than missing mate-
14. Tobin DL. A breast-feeding evaluation and education tool. rial) should be directed to the corresponding author for the
J Hum Lact 1996; 12: 47–9. article.

26 
C 2007 The Author(s)/Journal Compilation 
C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 21–26