The development of a neonatal stool colour comparator

The development comparator

of a neonatal stool colour

Ellena M Salariya and Catherine M Robertson

The policies which exist to monitor the passage of neonatal stool vary between and within hospitals. When stool colour is required, the recorded observations are invariably subjective opinions and as such are largely unreliable. The development of a stool colour comparator allowed an objective test to be used by professional staff and mothers. This facilitated data collection for a study which sought to determine, among other factors, the transition time of meconium through to the yellow stool of the milk-fed baby, during the early neonatal period. The transition time of the meconium is indicative of gastrointestinal activity in the newborn and midwives are able to correlate accurately documented changes in stool colour with the baby’s daily weight loss to assess the quality and/or adequacy of early infant feeding particularly breast feeding. The stool colour comparator is used in postnatal wards at Ninewells Hospital, Dundee by mothers and midwives and the numerical coding has replaced the previously used subjective abbreviations Met (Meconium), Ch (Changing) and Y (Yellow).

INTRODUCTION The passage of first stool is an important physiological occurrence in all newborn babies and is usually indicative of gut patency. Meconium may be passed by the full-term healthy baby at the time of birth or later (Hughes 1952; Sherry & Kramer 1955; Clark, 1977; De Carvalho et al,

Ellena M Salariya MPhil, RGN, RM Senior Midwife In-service Training/Research, Department of Midwifery, Ninewells Hospital and Medical School, Dundee DD2 lUB, Scotland. Catherine M Robertson BA, PhD, RGN, RNT Director of the Centre for Nursing Studies, Department of Molecular and Life Sciences, Dundee Institute of Technology, Dundee DDl 1HG. Manuscript accepted 27 July 1992 Requests for offprints to EMS

1985) and then changes from the initial black of meconium through varying shades of green and brown until the baby passes the characteristic yellow stool of the milk-fed baby at four days or later, after birth. The amount of meconium passed by a fullterm healthy baby has been estimated to be 200-600g (Grulee & Bonar, 1926). Analysis of meconium stools indicates that they contain about 1 mg of bilirubin per gram of wet weight and that more than 50% of this is unconjugated (Odell, 1976). There has been some suggestion that meconium retention after birth is associated with the development of neonatal jaundice (De Carvalho et al, 1985) because intestinal bilirubin is potentially available for absorption from the small and large bowel and may add to the 35

36

MIDWIFERY

bilirubin

load for clearance

sen & Herman,

by the liver (Broder-

1963; Gourlay

& Arend,

The policies which exist to monitor

1986).

the pass-

age of stool in the neonate vary between and within hospitals. Whilst it is common policy to record the passage of first meconium, of its complete considered toring

evacuation

to be important.

and

recording

the timing

is not always noted or of

Subsequent meconium

varies widely from a daily verbal enquiry mother)

about

observations and urine,

monipassage (of the

the baby’s bowel movements

by mothers

to

or midwives of all stool

passed by the baby.

cation was often experienced. In the past, differences of opinion were found to exist in the identification

of stool colour

heard (by EMS) to comment ination

of baby records,

different

to achieve

and

their

colour

is recorded

baby’s feeding

type and pattern,

record.

observations

These

Hospital

and

the passage of all stools along

out caretaking

more

in relation

accurate

and

information

for

a

gut transit time of meco-

to the incidence

a method

classification

objectively

to facilitate of

stool

of neonatal identification

colour

and cost effectively,

transition

was required.

are by

when carrying

activities for the baby. The feed

The envisaged (stool colour) comparator

record is held at the baby’s cot during the period

The

of hospitalisation.

During

appeared

colour

from

transition

in

with the

and recordings

(or care-giver)

by the investigator,

in a standard

made initially by midwives and subsequently the baby’s mother

it was

study which, along with other variables,

jaundice, (Ninewells

‘I thought

tal confirmed this. In the absence of an existing tool and in order

nium

Medical School, Dundee)

-

postnatal wards within the study hospi-

would seek to quantify

At the study hospital

mothers

green but the midwife said it was yellow’. Exam-

planned

Background

between

and other observers. When asked about the colour of their baby’s stools, mothers were often

the process

meconium,

of stool

individual

solution

paper

to the

strip

of

three

babies may pass stools in varying shades, depen-

selected

dent on however,

their feeding type. The colours are found to fall into four broad

duced. The observer

categories

of black, green,

number

brown and yellow.

Prior to the development of the neonatal stool midwives and mothers colour comparator, identified and classified stool colours subjectively

above

problem

at first

to be simple and straightforward. or

and numbered

the soiled napkin,

four

columns

A

appropriately would be pro-

would place the strip over

select the matching

and record it. The procedure

colour by was envis-

aged as being similar to using shade cards when selecting

or matching

paint.

for record keeping purposes. Three categories only were identified; ‘Meconium’, ‘Changing’ and ‘Yellow’ and these were recorded

in abbre-

viated form as Met. Ch. and Y. on baby records by midwives and mothers. The abbreviated information was also entered onto the Neonatal Record (SMRll) by midwifery staff and was presumably of interest to paediatricians. Communication with midwives, prior to the development of the stool colour comparator, indicated that although verbal information about the stool colour transition and its recording format was given to the mothers soon after delivery, misinterpretation of abbreviations and/or disagreement about colour classifi-

Colours required The range of colours required for the stool comparator included black, green, brown and yellow. To facilitate appropriate choice of colours, a sample sheet of commercially available Pantone coloured paper by Letraset was inspected. The ‘sample’ sheet revealed many shades of the broad categories of colours required for the stool colour comparator and it became apparent that the required shades would have to be selected by direct comparison with transitional stool colours on the baby’s ‘soiled’ napkins, during the early neonatal period.

MIDWIFERY

comparator

METHOD Soiled

disposable

between

napkins

X am and 9am

placed in individual

were

collected

daily by midwives

disposal

bags. These

and were

labelled giving the baby’s name, date and time of birth,

method

of feeding

and time of napkin

change when the sample was acquired. The napkins were examined as soon as these became

available

to the investigator,

this did not ensure ‘freshness’ as napkins

although

of the stool sample

are not necessarily

changed

at the

time of stool passage. Similar delay would apply to the eventual study data collection

and this was

not considered to be a problem. Whether or not neonatal stool colour changes on making contact with the napkin or after ‘ageing’ would apply to the data collection

and although

acknowledged,

was not examined further. The sample stool, as it lay on the napkin,

was

matched with the coloured samples sheet of the commercially available paper. Sample matching was carried

out in daylight

and under artificial

lighting, and no differences between colour selections were found. Examination of samples was made over the full time range necessary the stools to change tions were made,

colour.

Eighty-five

including

several

for

observagroups

of

the same baby, over a period

of

samples

from

four

five days. Equal sample sizes were from breast and bottle-fed babies. The

to

obtained

colour matt black was selected initial

meconium,

found

to

‘changing’

be

three

regularly

as comparable

other

colours

comparable

to

the

colours of the stool in both breast and

were examined apparent

that

more

minimal

closely,

however,

differences

only

when these were compared

the three recurring

it was were

with two of

colours. Four Pantone

Paper

colours by Letraset were therefore selected-matt black and three others coded 450, 457 and 110. Sheets of the paper were duly ordered from the suppliers for use in the construction of the comparator.

The comparator types Although

it had been anticipated

the estimated

was prohibitive comparators

as only around would

and produced

initial cost of this 4,000

be required

disposable

for

the pro-

posed study, i.e. 150 babies X 6 napkin changes/ day x 5 days in hospital. There would be no benefits regarding subsequent larger orders, as far as could be envisaged at that time. The development

and production

of an alternative

non-disposable

comparator

had also to be pur-

sued to enable

the selected

colours

for inter-observer

to be tested

reliability prior to production

of a final comparator. In addition

to the

above, a non-disposable to be sterilised

requirements comparator

repeatedly.

Plastic covered cards,

like those issued by banks, such establishments,

described would need

libraries

and other

was the model initially con-

sidered. However, examination of these laminated cards showed that the number of different colours did not exceed three, eg blue and red on white or yellow or any other colour combination. The stool colour comparator

would have to have

five colours,

i.e. a white background

f-our selected

‘stool’ colours.

The study

Department

of Medical

plus the

Physics,

at the

hospital,

development

was approached about the and production of a non-dispos-

able comparator discussion

as envisaged

a prototype

by EMS, and after

was produced.

to

were

bottle-fed babies and two further colours were seen to compare infrequently. When the latter found

would be disposable

professionally,

37

that the final

The prototype This model resembled a plastic credit card in size and shape and measured 100 mm x 60 mm (Fig. colours:

black,

brown, green and yellow, each measuring

1). Four

circles

of the selected

20 mm

diameter, were pasted across the upper area of a light-weight white card. Four circles of the same dimensions were cut in the central area of the card directly opposite the coloured circles to facilitate viewing and comparison of stool colour, when the comparator was held above the stool sample. Numbers were placed in the lower area beneath the selected colours, from left to right 1 = black, 2 = brown, 3 = green and 4 = yellow. Difficulties were encountered when the comparator was enveloped in soft plastic and sealed by a process of heat application to enable

38

MIDWIFERY 2000 kg. The comparator

was completed

when

the edges were dressed and sealed with perspex adhesive. Each colour block was engraved with its reference green

number,

and

measured

4

=

1 = black, 2 = brown, 3 = yellow.

The

X 38mm

140mm

comparator

and the estimated

production

cost was &l per comparator.

comparator

was immersed

for 30minutes period

at three

of three

in Hycolin

hourly

The

solution

intervals

over a

days to test its suitability

for

repetitive sterilisation and was found to be water-tight. There was no deterioration of BROWN Fig. 1 Prototype

colour

GREEN

of comparator

quality

when

this was compared

tool

apertures and necessitated the comparator and/or the non-disposable

comparator

liquid disinfectant

between

sealings around perfected

to be sterilised in

observer/users.

The

the viewing areas could not be

to allow sterilisation

with

controls. The comparator however, proved to be unsuitable, as light reflected from the viewing

by this method

much adjustment of stool sample being

viewed, before comparisons could be made. It was believed that this variability of positioning of stool sample and/or comparator liable

and

this

version

would be unre-

was rejected.

In

an

and the model was rejected.

attempt to reduce glare in the viewing apertures

Version 1

surface

the coloured This version of the comparator the Medical

Physics

was also made at

Department

at the study

hospital and was a laminar construction ing of 1.5mm thick polycarbonate*,

comprisdouble

block area was masked and the top

Vacu-blasted

However,

to provide

the modification

a matt finish.

altered

the colour

appearance of the stool sample being viewed and this version was finally rejected.

sided adhesive tape+ and 4 colour blocks* (Fig. 2). The

four colours

blocks were mounted

side of the adhesive

were cut in the centre The tape provided coloured

to one

tape and viewing apertures of each coloured

a natural background

blocks and viewing apertures,

block. for the

Version 2

It had been noted that the coloured paper, selected for use in the comparator, was also produced

in photographic

paper and was guar-

as well as

anteed

an adhesive surface for jointing. The tape and coloured blocks were then presented to the

colour

reproduction.

tration

at the study hospital

inside face of a pre-shaped

feasibility

top frame of polycar-

by the

manufacturers

to effect

The Department

of photographing

agreed

* Polycarbonate is manufactured by Rohm Plastics under the trade-mark ‘MAKROLON’. + Double-side adhesive tape is a 3M product. * Pantone coloured papers, by Letraset.

BROWN

Fig. 2 Version 1 of comparator

to test the

the four

bonate and pressed together. The backing strip of adhesive tape was removed from the assembly and the bottom pre-shaped frame was applied to the exposed adhesive and pressed home. To improve the seal, the assembly was placed in a hydraulic press and squeezed with a load of approximately

GREEN

tool

exact

of Illusselected

MIDWIFERY

study hospital

from

delivery

of

mothers

were

instructions

given

about

and one neonatal BROWN

2-28

February

a full-term

baby.

written

Each sample was compared

tool

coded, using the comparator,

Fig. 3 Version 2 of comparator

the mother

Three

were obtained.

colours,

in correct

and

sequence,

for ultimate use as

The photographs

printed

on

matt

were developed

paper

and

measured

125 mm x 90 mm each. The photographs

were

cut into 15 mm strips and each strip had to be numerically

coded 1,2,3

and 4 manually (Fig. 3).

To test the comparator the stool

sample

identified.

of the baby producing

and

in either

the appropriate natural

The problem

the colours

manually

was

or artificial

light as

of having to number

after printing

been solved by placing appropriate

could have numbers

on

the colours prior to photographing. However, because the comparator was disposable, 250300 strips would be required inter-observer

to test the selected

colours

for

mothers

and midwives. This version cost 7 p per

strip and was rejected

because

possibility of colour blindness.

mother was asleep comparisons were made by two staff members only. A total of 909 observations were made, observers observers.

and

599 between

3 10 between

mother/midwife midwife/midwife

reliability

for economic

between

reasons.

FINDINGS The

findings

were measured

according

This

between pairs of observers,

using the Weighted

Kappa

scale of agreement

1968). The degree of agreement (Cohen, between mother/midwife observers was 0.79 and between midwife/midwife observers 0.87 which is approximately demonstrated

79% and 87% respectively high

degrees

strips

3 of

the

was constructed selected

coloured

by hand

when

papers

were

pasted onto white light-weight card lengthwise and cut transversely into 1Omm strips. The colours were numbered individually 1 = black, 2 = brown, 3 = green and 4 = yellow. The cost of materials for this version was very small and sixty sets of 5 comparators were produced ness for testing (Fig. 4).

time

of meconium

term’ babies, during

to determine passage

the

in ‘full

the first five days of life in

both breast and bottle-fed

babies.

Version 4 Professional printers were asked to produce the final comparator (Fig. 5) for use in the proposed study.

in readi-

Testing the comparator The comparator was tested for inter-observer reliability by 50 mothers who agreed to enter the study on admission to the postnatal wards at the

and

of inter-observer

reliability. The findings enabled an objective system to be transition

version

to the

degree of agreement

used by midwives and mothers Version

of the

When the

colour

of light or glare

and the colours had been exactly reproduced guaranteed.

the sample

of staff available at

the strip was held over

No reflection

experienced

and

by three observers,

the time. Males were excluded comparators.

and

hundred

independently

and any two female members greater

and

information

stool samples

GREEN

1987 after

Midwives

the testing.

39

BROWN

GREEN

Fig. 4 Version 3 of comparator

tool

40

MIDWIFERY

may be unaware nium

and

transition

of the initial colour

its subsequent

normal

determine

GREEN

Fig. 5 Version 4 of comparator

especially

tool

through

its transition

to the yellow stool of the milk fed baby, could be expected to be affected by baby feeding types and patterns. The assessment of total evacuation however,

is at present

and somewhat inaccurate a laboratory

subjective

unless it is measured

(De Carvalho

in

transit time of total meconium can be measured cost effectively when an objective test system is change

to identify

classify

stool

colour

in the clinical situation.

The neonatal

stool colour

used as a teaching strate

and

the

transition

process

from

mothers

and midwives

stool colour replaced the

comparator

aid by midwives

change

meconium

of

to

is now

to demonstool

yellow

to identify

colour and

by

and classify

coding numerically. subjective assessment

This has of stool

colour change at the study hospital previously recorded as Met., Ch. and Y. It is suggested that midwives

and others

ance of assessing

during

-

the first five

to effective

data collection

in research

(or demand)

this type of

Acknowledgements We wish to thank mothers in wards, 33.34 and 36, and all Midwifery Personnel, whocontributed to thecarryingout of the study, the Medical Physics and Medical Illustration Departments for advice, Miss Fleming, Director of Midwifery Services and Mrs. A. Gosling for typing assistance.

et al, 1985).

The development and subsequent use of the stool colour comparator demonstrates that gut

applied

of early baby feeding

feeding

studies which include information.

DISCUSSION

of meconium,

the quality breast

stool colour criteria, to

days of life. The recording of stool colour transition by this method of classification could contribute

The passage of meconium

of

to yellow stool. Midwives in postnatal

wards at the study hospital correlate change timing, along with other BROWN

of mecopattern

acknowledge

total evacuation

the importtime of meco-

nium objectively during the early neonatal period and that neonatal stool colour comparators be made available to all mothers of newborn babies to facilitate this. The stool colour comparator is an ideal teaching aid for mothers who

References Brodersen R, Hermann L L 1963 Intestinal reabsorption of unconjugated bilirubin: A possible contributing factor in neonatal jaundice. Lancet 1: 1242 Clark D 1977 Times of the first void and stool in 500 newborns. Paediatrics 60: 457 Cohen J 1968 Weighted kappa nominal scale agreement with provision for scaled disagreement or partial credit. Phycological Bulletin 70: 2 13-220 De Carvalho M, Robertson S, Klaus M 1985 Fecal bilirubin excretion and serum bilirubin concentrations in breast-fed and bottle-fed infants. Journal of Pediatrics 107: 786790 Gourley G R, Arend RA 1986 Beta-glucuronidase and hyperbilirubinemia in breast-fed and formula-fed babies Lancet 1: 644-646 Grulee G C, Bonar B E 1926 The newborn: physiology and care, Appleton-Century Crofts, New York Hughes J 1952 Pediatrics in general practice. McGrawHill Book Co Inc, New York Ode11 G B 1976 Neonatal jaundice, In Popper H, Schaffner F (eds) Progress in liver disease V. Grune 8c Stratton Inc, New York Sherry S N, Kramer I 1955 The time of passage of first stool and first urine by the newborn infant. Journal of Pediatrics 46: 158-I 59.