Tài liệu Consumer and competition policy directorate doc

1.1 Aims of this research 22.2.3 Classifying the type of mouthing behaviour 6 3.1 Children's mouthing characteristics and child care details 7 4.2 Estimates of daily mouthing times accor

Research into the mouthing behaviour of children up

to 5 years old.

Research commissioned by the Consumer and Competition Policy Directorate, DTI.

Department of Trade and Industry

1 Victoria StreetLondon

SW1H 0ET

Carried out under contract by:

Dr Beverley Norris Stuart SmithInstitute for Occupational ErgonomicsSchool of Mechanical, Materials, Manufacturing Engineering and ManagementUniversity of Nottingham

University Park, Nottingham, NG7 2RDTel: 0115 9514039 Fax: 0115 9514000 July 2002

URN 02/748

1.1 Aims of this research 2

2.2.3 Classifying the type of mouthing behaviour 6

3.1 Children's mouthing characteristics and child care details 7

4.2 Estimates of daily mouthing times according to what was mouthed and how 13

4.3.1 Categories of toys and other objects mouthed 23 4.3.2 Mouthed toys and other objects by material 25

the study, specifically:

i) Information regarding the methodology and protocols of the study

ii) Background data on the sample: Details of children’s mouthing habits, the child’s family, theobservers and their care and work circumstances and child care arrangements

iii) Estimated data on time spent each day mouthing on dummy/soothers, fingers, toys, and otherobjects by how the items were mouthed, for each age group

iv) Details of the validation study

v) Discussion of the findings of the work

vi) Details of the items mouthed in the study

vii) Literature review of chemical and mechanical hazards from children’s toys and other plastic items

viii) The forms used in data collection

Information presented in this report should be used in conjunction with the summary report described above

1.1 AIMS OF THIS RESEARCH

The research described in this report was designed to expand upon the knowledge of themouthing behaviour of children The ages of children to be observed in this study ranged from

1 month to 5 years, to investigate the extent of exposure to products at this older age (given thefindings of Warren et al (2000) that over 20% of the children in their study carried a non-nutritivesucking habit past the age of 36 months) The overall aim of the project was to produce a figurefor the total time that children within this age range are expected to mouth items per day Thiscould then be used to assess potential mechanical hazards posed by items on which childrenwere found to mouth

1.2 DATA COLLECTION

The behaviour of young children may be greatly affected by all types of stimuli In studies such

as this where we are attempting to gain information on normal and natural behaviour of children

it is imperative that the environment in which the data are gathered is as unaffected by theobservation as possible This means that the child should be surrounded by things that they areused to and know For this reason, the research by Groot et al (1998) and Juberg et al (2001) bothrely on data being recorded by the parent(s) of each child using a diary format The child isobviously used to the presence of their parent and so natural behaviour may be assumed Theuse of a structured formal diary recording system serves to control the type of information thatthe parent is recording and therefore provides a level of validity and reliability to the research

Greater levels of reliability would possibly be obtained through the use of trained researchersobserving the child in their home environment, as the observer is trained to record certain types

of behaviour consistently, and so providing good repeatability of observation The same is truefor the use of video recording equipment, where very short behaviours can be reliably recorded.However, it is likely that a child will be very aware of a video camera and being filmed, and thiscould lead to a change in behaviour Likewise, the presence of a stranger in the home is likely toalter the behaviour of the child being observed to a greater or lesser extent, so while thereliability of the data may be high, the circumstances are less "realistic" and lower the validity ofthe data Another problem with video recording is that young children are often on the move,which would make capturing every behaviour extremely difficult For these reasons this researchused parents/carers as observers, to ensure children’s mouthing behaviour, and behaviourgenerally, were as natural as possible to ensure high data validity

Observations were carried out at home rather than nurseries or childcare establishments as thehome environment has a vast range of items that are accessible to children, not all of which areintended for use by them This is a worst case scenario but also realistic compared to more sterileenvironments such as playgroups and nurseries where the contents of rooms are very carefullyand rigidly controlled

Because of the factors mentioned above it was felt that an examination of the reliability ofparents as observers should be carried out This has not been investigated by prior studieswhich have also used parents as observers A comparison of parent and trained observer datacollection was carried out on a sample of 25 children Information on this validation part of theresearch may be found in Section 5 of this report

Page 4

fifteen minute observation sessions A period of two weeks was allowed for the observations to

be completed and the data returned These observation sessions were to be spread out overdifferent times of the day and the week as follows:

i) Weekdays:

• 4 fifteen minute sessions between the child waking and 11am

• 4 fifteen minute sessions between 11am and 2 pm

• 4 fifteen minute sessions between 2pm and 6pm

• 4 fifteen minute sessions between 6pm and the child going to bed for the night

ii) Weekend:

• 1 fifteen minute session between the child waking and 11am

• 1 fifteen minute session between 11am and 2 pm

• 1 fifteen minute session between 2pm and 6pm

• 1 fifteen minute session between 6pm and the child going to bed for the night

This gives in total 5 hours of observation time on each child Observations could be done at the discretion of the parent at any time within these zones However, no more than twoobservation sessions were allowed within the same time zone on any one day, and at least 30minutes was required to be left between observation sessions These provisions were included

in order to ensure data was gathered spread evenly across the days and week to gather arepresentative picture of the child’s behaviour, and also to reduce potential errors from fatiguecaused by observation

In each 15 minute observation session the following information was recorded:

• the duration of each mouthing behaviour

• what types of items were mouthed

• the type of mouthing behaviour for each item mouthed

• what was mouthed

• which room of the house the observation was undertaken in

• other people present in the room

• whether the child was feeling unusually ill or tired during the observation

• how long it had been since the child last ate and slept

• what the child was doing during the observation

Information gathered over the whole study for each participant also included:

• details of the child’s physical and psychological development

• details of the child’s family (e.g size, type of house lived in, social class)

• details of the time the child spent outside of the family home in a care environment (e.g a nursery, playgroup)

• details of toys mouthed by each child, part of toy mouthed and what each part mouthed was made of

• details of what other objects were mouthed and what they were made of

• details of the child’s daily routine for the first and last days that they were observed i.e what time they got up and went to bed, and time spent eating and sleeping during the day

The questionnaires and forms used to gather these data are presented in Appendices A, B, C and D

Page 6

The use of a diary type observation recording system requires that the forms for recordingbehaviour are simple and intuitive to complete For this reason, a pre-pilot stage was undertaken.

Several alternative observation recording forms were produced and trialed in a local day nursery

by members of the Product Safety and Testing Group (PSTG) trained in observation protocols Forthe purposes of this stage, a single child was observed by both members of the PSTG for a

15 minute period, with the type and duration of each mouthing behaviour being recorded Thisprocess was repeated for each of the sample forms in turn The results were then analysed forthe similarity of findings between the two members in terms of duration and type of behaviours

Ease of use of each form was also assessed One observation recording form was chosen to beused in the pilot trials

2.1.2 Pilot trials

The materials developed for use in the pilot trials were as follows:

• Consent form

• Set of written instructions

• Observation recording form (as trialed in the pre-pilot stage)

• Questionnaire

• Daily routine form

• Toy and object details forms

Also included in the observation pack were a stop-watch, a pen and a freepost labelledenvelope for returning the completed observations

The pilot trials followed the same format that was proposed for the main part of the study

Participants were given the folder containing all the required materials (as listed above) Theywere given only brief instructions about what they were being asked to do, as it was anticipatedthat the only contact with participants taking part in the main study would be on the telephone,and so the main instructions included in the observation pack would provide all of the necessaryinformation Participants were given one week to complete the five hours observation time

Seven parents were recruited to undertake the observations of their child for the purposes ofpilot testing the protocols and materials developed for the research Two were unable tocomplete the study for personal reasons, and results from one participant did not arrive back atthe PSTG until after the deadline for analysis, providing data from four participants for analysis

The ages of the children observed were 4 months (female), 10 months (male), 27 months(female) and 55 months (female) Feedback was obtained from each participant on the ease ofuse of the materials contained within the observation pack which resulted in changes beingmade to the documents The final versions that were then used in the main study are presented

in Appendices A to D

2.2 OBSERVATION PROTOCOLS

2.2.1 Timing of mouthing behaviours

Participants in the research were each provided with a stopwatch for timing the duration ofmouthing events They were not required to start, stop and reset the watch each time an eventoccurred Instead, the stopwatch was started at the beginning of the session and not stoppeduntil the 15 minutes was over When the behaviour began, i.e when the item first entered themouth, the time on the stopwatch was recorded on the observation form (to the nearest second),and when that particular behaviour finished the time was again noted This enabled a quickerand more accurate time map of the child’s behaviour to be produced than by starting andstopping the watch for each behaviour, especially for very short time behaviours

2.2.2 Classifying what was mouthed

Products/items mouthed were classified into four categories: a dummy/soother, fingers, toys, and other objects The "fingers" category includes any part of the body of the child or ofanother person, and whether an item was classified as a toy or other object was decided upon

by the observer

2.2.3 Classifying the type of mouthing behaviour

Mouthing behaviour was classified into three categories The instructions given to observers onhow to classify behaviour were as follows:

• Licking/lip touching This is where the item is placed to the front of the mouth, without

actually entering into the mouth The child may be actually licking an object, or may be touching the object to their lips The time recorded should be from when the object first touches the child’s mouth to when the object is no longer in contact with the mouth If a child is licking an object it may be difficult to record because of the short time period involved In this case note the time when the object is put close to the mouth to be licked, and then taken away from the mouth

• Sucking/trying to bite The item is put directly into the mouth of the child The child may

be sucking, holding the object in their mouth or trying to bite (gumming) the object

• Biting or chewing The item is directly in the mouth of the child It is clear that the child is

biting or chewing on it

It can be difficult to tell exactly what a young child is doing with an item if it is in or near theirmouth Observers were told that unless they could see that actual biting or chewing washappening they should tick the sucking/trying to bite category If the child did more than oneaction then they were asked to tick both, e.g if they lick and then bite an item to put a tick underboth headings Photographs were included illustrating the licking and sucking behaviours, asshown in Appendix E, but not biting/chewing; this is difficult to illustrate as the item is shown asjust being at/inside the mouth without the movement which means chewing or biting is occurring

Page 8

Table 1: Number of children observed in each age group

The upper age limit for each age group was the day before the final month age is reached, e.g 1-3 months includes children from one month to two months three weeks and six days

3.1 CHILDREN'S MOUTHING CHARACTERISTICS AND CHILDCARE DETAILS

Background information was gathered on each child and their families, covering the child’sphysical development, mouthing habits of the child, and information on the child’s family

The information collected is detailed below

i) The mouthing habits of the child:

• 29% of children leave bite marks in toys

• 25% of children damage items by biting

• 28% of children use a dummy/soother

• 69% of children were/are breast-fed

ii) Who did the observations, who cares for the child, do they work:

• 99% of the observations were done by the child’s mother

• 97% of the children were (mainly) cared for by their mother

• 47% of main carers did not undertake paid work, 50% worked part-time, 3% worked full-time

Age group

1-3 months3-6 months6-9 months9-12 months12-15 months15-18 months18-21 months21-24 months

Number of girls

679111195621181210

Total

91415171614161239312924

Page 10

iii) Childcare arrangements:

Childcare arrangements for each child observed ranged from none (i.e always at home with themother or father) to 5 days a week Details are presented in Figure 1 of the type of childcare thatchildren taking part in the study attended For each day of childcare separate values are

presented, for example, day one is not necessarily the same day for all children but instead meansthat this represents a single day (or part of a day) where childcare occurs for that child

none

creche playgroup relative / friend

childminder nursery Day one

none

creche playgroup relative / friend

childminder nursery Day two

none creche playgroup

relative / friend childminder

nursery Day three

Age group

1-3 months3-6 months6-9 months9-12 months12-15 months15-18 months18-21 months21-24 months

Minimum

3:516:207:106:457:206:578:306:507:359:359:40

Maximum

13:3012:4811:5011:2311:0512:4212:1513:3913:2013:5014:10

Figure 1: Childcare arrangements, by day, for children observed in the study.

As the number of observation days increases the number of children going to some form ofchildcare decreases Nursery is the most popular form of childcare

Information was gathered on how much time during a typical normal day the child would haveavailable to spend mouthing, that is, time not spent sleeping or eating

Table 2 presents the mean, minimum and maximum times available to mouth for each age group

none

creche playgroup relative / friend

childminder nursery

Day five

creche playgroup

relative / friend childminder

nursery

The following sections present results for:

4.1 Estimated mouthing time according to what was mouthed

4.2 Estimated mouthing time according to what was mouthed and how it was mouthed

4.3 A breakdown of what was mouthed, including by material (and if it is intended to bemouthed – as far as can be known)

The estimated daily mouthing time is extrapolated from the mouthing behaviour recorded over the five hours observation time This is calculated as follows:

Observed mouthing time Total time observed for

Where:

Observed mouthing time = the amount of mouthing time recorded for each child

Total time observed for = the total amount of time that they were observed for (5 hours)

Time available to mouth = the time available over the whole day for the child to mouth

This calculates the average time spent mouthing per hour, multiplied by the total number

of hours each day that each child has available to mouth i.e the number of hours they are awake during the day but not eating This has been based on the assumption that children arelikely to mouth at the same rate throughout the day A Kruskal-Wallis1 test showed no significant differences in observed mouthing between the different times of the day thatchildren were observed

The presented mean values were calculated by dividing the total estimated mouthing time foreach category of item mouthed within each age group by the total number of subjects in that agegroup For example, for 3-6 months olds, only 4 subjects may have actually mouthed on adummy/soother but the total estimate of daily mouthing on a dummy/soother for that age wasdivided by the total number of subjects in that age group (14) to give the mean value of dailymouthing time that could be expected across that age group generally

Because of the skewed2 nature of the raw data used to produce these estimates care should betaken if using mean estimated daily mouthing values for the purposes of safety This is becausethe mean value may not reflect the true average or middle value of the sample for each agegroup, depending on the degree to which the data are skewed

4.1 ESTIMATED DAILY MOUTHING TIME BY ITEM MOUTHED

Figure 2 presents the mean estimated daily mouthing time for males and females combined foreach item mouthed (in hours:minutes:seconds) Each mean behaviour is the average time that achild in each age group could be expected to mouth a given item over a normal day Mean totaldaily mouthing time is the time that any child could be expected to mouth, on anything, onaverage over a normal day, and is the sum of mean mouthing time on all items for each age group

1

The Kruskal-Wallis test is a non-parametric statistical test used to determine whether values of scores for three or more groups are significantly different, in this case we have four groups, i.e those observed between waking and 11am, between 11am and 2pm, between 2pm and 6pm, and 6pm until bedtime.

nature (i.e only a few seconds), with only a small percentage being observed with duration of between 1 and 15 minutes

Page 12

* time available to mouth per day

The presented maximum values are the highest estimated daily mouthing times of any child ineach age category, for each item It does not necessarily follow that the maximum values in eachage group for each item were exhibited by the same child, that is, one child may have mouthed

a lot on a dummy/soother while another child in the same age group may have mouthed a lot onfingers Also presented is the maximum estimated daily mouthing time for mouthing on all andany items, in each age group, by any single child

Figure 2: Estimated mean daily mouthing time for each item mouthed and for all items (total mouthing) (hours:minutes:seconds)

1

The Mann-Whitney test is a non-parametric statistical test used to determine if there are significant differences between scores or values of two groups with different subjects in each group, i.e males and females.

0:00:00 0:15:00 0:30:00 0:45:00 1:00:00 1:15:00 1:30:00 1:45:00 2:00:00 2:15:00

Toys Other objects Not recorded Total

Table 3: Estimated mean and maximum daily mouthing time for all items mouthed (hours:minutes:seconds).

Table 4: Estimated mean and daily mouthing time on all items mouthed (hours:minutes:seconds).

Not Recorded

3-6 months

6-9 months

9-12 months

12-15 months

15-18 months

18-21 months

21-24 months

2 years

3 years

4 years

5 years

Item mouthed

Total estimated mouthing on all items

Mean

Max

Age Group

1-3 months

3-6 months

6-9 months

9-12 months

12-15 months

15-18 months

18-21 months

21-24 months

2 years

3 years

4 years

5 years

1:11:48 1:57:41 1:35:11 1:35:16 1:36:01 1:15:13 1:58:49 1:43:39 1:39:27 1:50:19 0:50:05 0:59:17

3:31:50 3:36:24 5:16:59 6:53:01 4:17:09 5:14:42 6:52:18 6:35:01 7:41:31 8:30:12 5:28:44 10:01:07

square = 30.050, d.f = 11, p<0.01, 2-tailed).

When estimated daily mean mouthing on just toys and other objects is considered, a much clearerpattern of change with age emerges Figure 3 presents the estimated mean daily mouthing times formouthing on toys and other objects only for males and females combined

Figure 3: Estimated mean daily mouthing time for mouthing on toys and other objects and total mouthing on these two items.(hours: minutes: seconds)

It is apparent that after the age of 6-9 months mouthing on toys and other objects generallydecreases

4.2 ESTIMATES OF DAILY MOUTHING TIMES ACCORDING TO WHAT WAS MOUTHED AND HOW

Mouthing time is presented for mouthing on all items and also separately for each category

of item mouthed (for males and females combined), in the following sections:

lick, suck and bite suck and bite lick and bite lick and suck bite suck lick

0:00:00 0:10:00 0:20:00 0:30:00 0:40:00 0:50:00 1:00:00 1:10:00 1:20:00

Suck Lick

Table 5: Estimated mean and maximum daily mouthing time on dummy/soothers by how they were mouthed.

Lick

Suck

Bite

Lick and Suck

Lick and Bite

Lick, Suck and Bite

Not recorded

Suck Lick

Page 18

Table 6: Estimated mean and maximum daily mouthing time on fingers by how they were mouthed (hours:minutes:seconds).

Lick

Suck

Bite

Lick and Suck

Lick and Bite

Lick, Suck and Bite

Not recorded

Suck Lick

Table 7: Estimated mean and maximum daily mouthing time on toys by how they were mouthed (hours:minutes:seconds).

Lick

Suck

Bite

Lick and Suck

Lick and Bite

Lick, Suck and Bite

Not recorded

Suck Lick

Table 8: Estimated mean and maximum daily mouthing time on other objects by how they were mouthed (hours:minutes:seconds).

Lick

Suck

Bite

Lick and Suck

Lick and Bite

Lick, Suck and Bite

Not recorded

4.2.6 Items not recorded

Table 9 presents the data where what was being mouthed was not recorded by the observer Thesedata are not presented graphically due to the small quantities

Lick and Bite

3-6 months

6-9 months

9-12 months

12-15 months

15-18 months

18-21 months

21-24 months

2 years

3 years

4 years

5 years

Mean

Max

Suck and Bite

Lick, Suck and Bite

Not recorded

The toys and other objects most commonly mouthed across the whole sample are shown in Tables

10 and 11, together with the number of times each of the item categories were mouthed by anychild in the study The miscellaneous categories are made up of items that did not fit into any of the other categories and which were listed as being mouthed only once across the whole sample

A few items are listed both as toys and other objects e.g books, jigsaws The items were classified

as such by the observers and will be the result of their perception, and not the authors’, of whether

or not the items were toys

Rank Toy Category

Shoes/slippers/shoe laces 35 Remote controls (TV and video) 32 Packaging (material unknown) 29 Stop watch inc cord 28 Furniture (solid) 27 Child care articles 26

Babies' bottles 10

Number of times mouthed Object category

Table 11: Categories of other objects mouthed across the whole sample

mouthed types of toy In the other objects categories, clothing and soft furnishings (such as duvets)are two of the largest categories

4.3.2 Mouthed toys and other objects by material

Information was gathered on what the toys and other objects that were mouthed in this researchwere made from The materials were classified as follows:

• Wood pulp products (e.g paper)

Within many of these categories are sub-categories, e.g within fabric are cotton and leather

Figure 9 shows the percentage of the toys and objects mouthed made out of these materials

Figure 9: Percentage composition of toys and other objects mouthed.

Plastics accounted for nearly half of all the toys and other objects mouthed in this research(49.2%), with the next highest being fabrics (24%) No other material accounted for greater than10% of the items mouthed (this excludes dummies/soothers)

0 5 10 15 20 25 30 35 40 45 50

The materials of the toys and other objects mouthed were then analysed by age Figure 10 showsthe percentages by material for each age group.

Figure 10: Percentage of items of different types of material mouthed in each age group.

This illustrates that for all age groups plastic and fabrics are the only items mouthed more than10% of the time at any age Also interesting is the cross-over from ages 3-6 to 6-9 months, whereplastic items become more mouthed than fabric items At age 1-3 months fabrics account for 80%

of all items mouthed, but at age 6-9 months this has dropped to 29%, after which it staysreasonably constant at approximately 20% of all items Plastics account for only 15% of itemsmouthed at age 1-3 months, but by age 6-9 months this has risen to 50%, a level approximatelysustained up to 5 years of age

4.3.3 Items intended to be mouthed

All of the toys and other objects that were mouthed by children in this study were judged as towhether they were items that are intended to be mouthed, for example teethers, rattles, cups,bottles and drinking straws This classification was made on informed opinion only as parents werenot asked to indicate whether or not the items were intended to be mouthed Table 12 shows theitems recorded in the study that were considered as intended to be mouthed, by each age group

Page 28

0 10 20 30 40 50 60 70 80 90 100

Table 12: Items intended to be mouthed which were mouthed in this study.

Bubble pipeCan

Toy spoon

This shows that feeding items such as bottles, cups, beakers and cutlery, and child care itemssuch as teethers and rattles were mouthed by nearly all age groups Of concern here is thatballoons were mouthed by nearly all ages over 6 months Table 13 presents those itemsconsidered as intended to be mouthed (i.e everything including those items in Table 10) as

a percentage of all items mouthed, according to age group

Table 13: Items intended to be mouthed as a percentage of all items mouthed.

These data show that at least an estimated 75% of the items that were mouthed by children

in this study were considered not intended to be mouthed

An analysis was made of how many of the plastic items were considered as intended to

be mouthed, as shown in Table 14

Table 14: Items intended to be mouthed as a percentage of all plastic items mouthed.

Items intended to be mouthed

as a % of all items mouthed

Items intended to be mouthed

as a % of all plastic items mouthed

Within each age group for each subject the total number of toys and other objects mouthed by eachchild was calculated, as a way of assessing the variety and range of items children will put into theirmouths A mean value for each age group was then calculated and these data are presented inFigure 11.

Figure 11 Mean number of toys and other objects mouthed within each age group.

This shows that the average number of items mouthed each day per child peaks at age 6-9 months(26 items), and then gradually decreases as the age of the child increases Children aged 1-3 monthsmouth the fewest number of items on average (3 items)

0 5 10 15 20 25 30

In order to assess the validity and reliability of the observation method used in this study theobservations were repeated with a random selection of 25 of the sample, using parentalobservation, plus two alternative methods; trained observers and video recording For eachperson a single 15-minute observation session of their child was undertaken at their home Themouthing behaviour of the child in this 15 minute session was recorded simultaneously by theparent, a trained observer and also captured on video All protocols were the same as the maindata gathering part of the study The data gathered from the three different observation methodswere then compared.

5.1 ANALYSIS OF MOUTHING TIME

Table 15 shows the mouthing behaviours recorded by a trained observer, by the parent and byvideo for the group of 25 children as a whole

Table 15: Comparison of mouthing summary statistics for the three observation methods All times are in minutes: seconds format.

The number of behaviours and mouthing times observed by the trained observer and the parentwere very similar, with total mouthing time, mean mouthing time and maximum observedmouthing time all being exactly the same, and number of behaviours observed and standarddeviation being very similar

As anticipated, the video analysis captured more incidences of mouthing behaviour and thuspotentially more mouthing time than the direct observation methods Additionally, mean andstandard deviation mouthing times were lower than those found by direct observation Thiswould be expected as video analysis allows for much more accurate observation and timecapture than direct observation, with the lower standard deviation indicating less variance inthe range of mouthing behaviours than observed by trained observer or parent

Page 32

Total number of behaviours observed in 15 minutes for all children

Total observed mouthingtime for all children

Mean observed mouthing time

Standard deviationmouthing time

Maximum observed mouthing time

short motion away from the mouth which would not have been (and was not) picked up bydirect observation

A one way analysis of variance (ANOVA)1 was conducted on the mouthing time data todetermine if there were any statistically significant differences between the data gathered by the three methods No significant difference was found

5.2 ANALYSIS OF WHAT WAS MOUTHED

Observations of what was mouthed (i.e dummy/soother, fingers, toys or other object) were madeand compared across the three groups to determine similarity of observation Figure 12 illustrates thenumbers of observations recorded for each type of item mouthed as recorded by the three methods

Figure 12 Illustration of what was mouthed by the observed children as recorded by trained observer, parent and video analysis.

Three of the observations made by parent and video analysis did not record what was mouthed bythe child It is unknown why the parent did not record what was mouthed, while the video analysiscould not see what was being mouthed due to the child being obscured from view As the videoanalysis observed more behaviours it follows that more behaviours were counted for each type It isinteresting to note that while the two direct observation methods (i.e parent and trained observer)counted the same number of dummy and toy observations, the number of finger and other objectobservation differed by five and four counts respectively This may be because in these cases theother objects mouthed were small items, which made it difficult to make a distinction between thechild mouthing the small object held in their hand or the hand itself

1

An analysis of variance (ANOVA) is a parametric test used to determine whether values or scores for three or more groups of subjects are significantly different.

This is the parametric equivalent of the Kruskal-Wallis test.

0 20 40 60 80 100 120 140 160 180

Trained observer Parent Video

Observation typ

Other object observations Toy observations Finger observations Dummy observations

e

5.3 ANALYSIS OF HOW THE CHILDREN MOUTHED

A comparison was also made of how the mouthing behaviour of the children was classified, i.e.into the categories of licking, sucking or biting Figure 13 graphically shows this comparison

Figure 13 Illustration of how the observed children mouthed as recorded by trained observer, parent and video analysis.

Again, video analysis counted more behaviours and so numbers of types of mouthing behaviourwill be higher overall Three cases of the type of mouthing behaviour were not differentiated byvideo analysis (again because of the impaired view of the child) and 24 cases were not

differentiated by parental observations

This investigation into the validity of the data gathering protocols show that the data gathered in thisstudy are reliable As expected video observations captured more behaviours and longer mouthingtimes than direct observations, but the presence of the video camera also changes the behaviour ofthe child The similarity of the number of mouthing behaviours and the total mouthing timesobserved by the parents and the trained observer mean that the observations of the parents used

to produce the estimates of daily mouthing can be treated with a high degree of confidence

Page 34

0 20 40 60 80 100 120 140 160 180

Trained observer Parent Video

Observation type

Lick and suck behaviours Biting behaviours Sucking behaviours Licking behaviours

6.1 VALIDITY AND RELIABILITY OF THE STUDY

The validation study showed that there was very good agreement between the trained observersand the parents who were used to observe in this study Parental observation at home was chosen

as the least intrusive and least disruptive method of recording children’s natural and normalbehaviour The only limitation to the use of parents (and trained observers for that matter) waswhen compared to video observation Here it was found that through the use of video analysissoftware, very short breaks in mouthing behaviours, when a child may have momentarily moved anitem out of, or away from their mouth, could be recorded These could not be recorded by the normaleye However, no significant differences were found between the three methods of observation

6.2 MOUTHING TIMES

6.2.1 Mean and maximum mouthing times

Overall, mouthing generally shows little relationship with age The results presented earlier showthat for overall mouthing (i.e on all items including fingers) the highest estimated mean dailymouthing time on all items is for the 18-21 month age group (1:58:49), and the lowest at age 4(0:50:05) This is probably due to the wide variety of items mouthed, including clothes, fingers andthumbs Dummy use does not show an obvious relationship with age

When it comes to toys and other objects around the home, children aged 6-9 months mouth morethan any other age group, and mouthing time generally decreases from this age (estimated meandaily mouthing time 1:03:40 hours:minutes:seconds) Children aged 1-3 months mouth the least onthese items (0:05:28) This is to be expected as children at this age are reliant on their parents toprovide them with items to play with (and thus to mouth on) as they are not mobile and able tomove about the house to find new items The majority of "other objects" mouthed by children in this1-3 months age group were their parents’ clothing

It is important however to consider the maximum estimated mouthing times which children may mouth

on items over a whole day, as this represents the highest likely exposure to mouthed hazards, andpresent a slightly different picture to the mean mouthing times The highest times are presented below:

• Highest maximum daily mouthing time on dummy/soother - age group 9-12 months (5:23:45)

• Highest maximum daily mouthing time on fingers - age group 5 years (9:02:45)

• Highest maximum daily mouthing time on toys - age group 6-9 months (3:46:46)

• Highest maximum daily mouthing time on other objects - age group 2 years (2:57:58)

Maximum estimated daily mouthing on a dummy/soother is higher than on fingers for all agesexcept 5 year olds, but there is no real pattern of increase or decrease with age Maximum mouthing

on fingers is approximately the same up to 15-18 months, and then steadily increases to peak at

5 years Again, this increase is probably attributable to the presence of children in the sample who

have a thumb sucking habit Maximum mouthing on toys follows a pattern of increase (to 6-9 months)

- decrease (to 18-21 months) - increase (to age 2) - decrease (to age 5) Finally, maximum mouthing

on other objects stays generally quite level to age 21-24 months, where it then increases to peak atage 2, and then decreases to approximately the same level as before

A possible "worst case" of mouthing on toys and other objects for each age could be taken by addingthe two values together Although the maximum mouthing estimates for mouthing on each itemdoes not necessarily come from the same child this would give an indicator of a value for dailymouthing that could possibly be reached Adding together the estimated maximum mouthing valuesfor toys and other objects for each age group, we find that the highest daily mouthing time is for

2 year olds (5:03:46) and the 6-9 month olds (4:57:09), with the lowest value for children aged 1-3 months (0:29:10)

6.2.2 What is mouthed

The focus of this research is on children’s mouthing time on toys and other objects as they areprobably the most likely to offer risk, both chemical and mechanical, compared to mouthing ondummies/soothers or body parts, the two other categories of mouthed items It has been assumedfor this discussion that dummies are items intended to be mouthed, and as such will be regulated byappropriate standards

The highest mean mouthing time on each category occurs in different age categories, but all aremouthed most by children under 1 year, except dummies/soothers Toys and other objects are bothmouthed most by age group 6-9 months:

• Highest mean daily mouthing time on dummy/soother - age group 18-21 months(1:09:92)

• Highest mean daily mouthing time on fingers - age group 3-6 months (0:49:03)

• Highest mean daily mouthing time on toys - age group 6-9 months (0:39:10)

• Highest mean daily mouthing time on other objects - age group 6-9 months (0:24:30)

Children aged from 3 to 15 months were found to mouth more on toys than other objects, a patternwhich was reversed from age 15-18 months to 5 years (with the exception of children aged 21-24months) Such a pattern of mouthing can be explained by the fact that as children learn to walk and thusbecome more mobile they become less reliant on their parents to provide them with items intended forplay, and are more able to move around the house exploring and finding new items of interest

6.2.3 How items are mouthed

Figure 14 shows the percentage contribution to total mouthing for each age group and each type ofmouthing behaviour

Page 36