Condition category
Date applied
Date assigned
Last edited
Retrospectively registered
Overall trial status
Recruitment status
No longer recruiting
Publication status

Plain English Summary

Not provided at time of registration

Trial website,

Contact information



Primary contact

Dr Semira Manaseki-Holland


Contact details

Public Health
Epidemiology & Biostatistics
University of Birmingham
B15 2TT
United Kingdom

Additional identifiers

EudraCT number number

Protocol/serial number


Study information

Scientific title

Investigation of the effect of swaddling upon lower respiratory infections, respiratory function, crying and sleep, thermal regulation and growth in infants 0 - 7 months in Mongolia: a randomised controlled trial


Study hypothesis

Null Hypothesis: Swaddled children do not have higher rates, or more severe cases of acute lower respiratory infection (ALRI) (compared to non-swaddled children).

Primary outcome study questions:
1. What is the effect of swaddling upon all episodes of pneumonia in Ulaanbaatar infants less than 7 months of age?
2. What is the effect of swaddling upon the severity of pneumonia in Ulaanbaatar infants less than 7 months of age?

Secondary outcome study questions:
1. What is the effect of age upon the first episode of pneumonia and the effect of age upon the relationship between swaddling and ALRI?
2. What are the effects of swaddling on oxygen saturation (SaO2) and respiratory rate (RR) in a sample of asymptomatic infants?
3. What is the effect of swaddling upon infant sleeping and infant fussing/crying?
4. What environmental temperatures are experienced by Mongolian infants in Ulaanbaatar? How do these relate to infant thermoregulation? How do traditional practices (type of housing, swaddling and bed sharing) contribute to effective thermoregulation in infants in harsh environmental conditions? Is there any evidence of a loss of thermal equilibrium in infants in winter in Ulaanbaatar?
5. What is the effect of swaddling upon the weight, height, head circumference of infants in Ulaanbaatar up to the age of 6 months?

Please note that this is the phase I study of the Swaddling and Health Research Project, phase II trial details can be found at:

Ethics approval

1. London School of Hygiene and Tropical Medicine (LSHTM) Ethics Committee approved on the 5th April 2001 (ref: ETH2003-04/107)
2. Ministry of Health (MOH) of Mongolia approved on the 10th July 2001 (ref: 2/1947)

Study design

Randomised controlled trial

Primary study design


Secondary study design

Randomised controlled trial

Trial setting


Trial type

Quality of life

Patient information sheet

Not available in web format, please use the contact details below to request a patient information sheet


Infant respiratory function/thermoregulation/weight and height gain/sleep, fussing and restlessness


All infants were recruited at birth and randomised to swaddling versus non-swaddling groups. Babies were then followed up intensely for 7 months.

Swaddling group:
The babies randomised into the swaddling group were to be swaddled in the common traditional Mongolian method as much as they felt comfortable. Traditional swaddling in Mongolia was as follows:
Swaddling was a tightly wrapping (2 adult fingers pass under the cloth) of a baby in several layers of cloth, covered by one warmer blanket from head to toe in a straight position after birth for about six months. Two or three ropes were used across a baby's body for binding to avoid unwrapping. For the first three months baby was swaddled most of the day and night with changes of the soiled cloth made every three or eight hours. After this age, practice varies with reducing hours of swaddling, but most infants were swaddled during sleep at least until 7 - 12 months. Duration and tightness of wrapping decreased with age. Furthermore as the baby grew, swaddling clothes for indoor may not have covered the head and would enclose the legs, trunk and arms up to the infant's necks, while after 6 months, in some occasions it only enclosed up to the waist. In the swaddling group, there was expected to be some variation in the swaddling period after 2 to 3 months, but it was usually expected to be more than 17 in 24 hours for those living in colder dwellings. Since the non-swaddling clothes had to be provided with alternatives to the readily available swaddling clothes and blankets, the swaddled group were given a blanket and 3 cotton sheets at the time of recruitment.

Swaddling, a cultural norm, was not a difficult infant care practice to achieve in this study. Swaddled babies would often be placed horizontal on a bed or carried, and not positioned vertically as with some cultures.

Non-swaddling group:
The non-swaddling group were instructed not to swaddle at all. The Mongolian stores did not sell the required size and warmth of clothing required to keep a newborn infant warmly clothed during the Siberian winters of Mongolia. Our recruitment pilot, in September 2000, indicated that in order to enable this alternative childcare method to be possible for the families randomised to the non-swaddling group, the project needed to provide some alternatives. Therefore, at the time of recruitment, the project gave each family a sheepskin sleeping bag and hat, a cotton sleeping bag, 8 cotton 0 - 3 months outfits. At 4 and then 12 weeks, the non-swaddling group was given another sheepskin hat and 10 outfits of a bigger size, which would fit the babies until 6 months of age. Half the outfits were made in Mongolia and half were body-suits made in the UK and transported to Mongolia. The sleeping bags were designed in consultation with 'Grobags' company in England, who have worked closely with Professor Fleming (a collaborator with the study at Bristol University, and an international expert in infant temperature regulation and SIDS). The sleeping bags were made in Mongolia but the manufacturers applied specific requirements developed in the UK. The families were encouraged to make or buy their own additional outfits for their infants, but if they could not, the project ones given to them were supplied in a large enough number in order to allow for washing dirty outfits at the same time as wearing up to 3 layers of outfits for the cold winter.

The sleeping bags were to be used at the parent's discretion, but the sheepskin sleeping bag especially was thought to be particularly useful for the cold evenings and when going outside. A leaflet was prepared for each family to explain what outfits they should expect to receive and how to use these to clothe their baby.

In the non-swaddling group, it was expected that some families would not adhere to their instructions of not swaddling because of cultural pressures. To monitor compliance and obtain a measure of exposure, multiple sources of information were used:
1. A retrospective 24-hour history of swaddling/clothing: recorded by the field-workers during each 3-weekly home visit through recording the carer's descriptions of the style of dress/clothing/ swaddling and length of each during the past 24-hour within the 24-hour retrospective diaries (a kind of behaviour diary adapted to record swaddling/clothing pattern)
2. Direct observation of fieldworker during 3-weekly visits: of the dressing/clothing/ swaddling of each child and discreetly recording observations of the type of baby clothing items that each family washed and hung to dry on the day of visit (available for observation during the majority of visits)
3. Analysis of the 4-days prospective mother's diaries used for sleeping and crying/fussing outcome measure: as mother's own reported behaviour during 2 time periods
4. Unscheduled visits from a monitoring fieldworker: in order to estimate unreported non-compliance conducted for all families 1 to 3 times
5. End-study fieldworker impressions' questionnaire: of their impression about compliance in each family and the consistency and extent of accuracy of the reports from each family during each 3 months of child's life

For analysis, the first and third methods of exposure recordings above were quantifiable and were combined to reveal exposure to swaddling per child for use in per-protocol analysis. The other three methods detailed above were triangulated with the data from the diaries and concluded that over 95% of families' compliance was revealed through the 3-weekly 24-hour retrospective diaries and the 4-day prospective diaries.

Intervention type



Not Applicable

Drug names

Primary outcome measure

ALRI/pneumonia, identified by passive surveillance at the hospital and primary care centres. Doctors were re-trained in the relevant WHO’s IMCI diagnostic criteria of ALRI and study protocols. Families were incentivised with free transport and medication to ensure better referral to study doctors in case of illness. Home visits by trained fieldworkers every 3 weeks took history of signs, symptoms and all doctor referrals or admissions and these were correlated with passive surveillance data. Infants with signs of ‘severe’ to very ‘severe’ pneumonia (IMCI defined) could be admitted and receive a CXR and pulse oximetry by trained study radiographers (trained by WHO expert using WHO criteria). X-ray films were assessed blind by 2 radiologist experts in WHO, Geneva, using consensus and WHO diagnostic pro-forma at the study end.

Measured throughout the study after recruitment until the end of the study (infants birth till 7 months of age, starting September 2002 until end of June 2003).

Secondary outcome measures

1. ALRI/pneumonia defined as in primary outcome measure and compared with age by completed months from birth, measured throughout the study after recruitment until the end of the study
2. Oxygen saturation and respiratory rate per minute (SaO2 and RR), measured at 9 - 10 weeks after birth
3. Sleeping and fussing/crying were defined as the overall 24-hour duration and number of bouts of crying and of sleeping and their particular measurements in night and day.
Two measurements of 2 to 24 hours were taken: Four successive typical days when infants were between 4 - 6 and 12 - 14 weeks of age.
4. Infants’ temperatures taken from three different body sites (abdomen, periphery and micro-environment) and compared to room temperature, measured by field workers who made visits according to a fixed, regular schedule.
5. Length or height, head circumference, or Mid Upper Arm Circumference (MUAC) in cm (to the nearest 0.5cm); and weight in grams. Infant weight and length were recorded at birth, and then the mean ages of 3, 16, 22, and 31 weeks after birth. Mid-upper arm circumference (MUAC) and head circumference were measured at 52 weeks after birth.

Overall trial start date


Overall trial end date


Reason abandoned (if study stopped)


Participant inclusion criteria

1. All babies born in the only four maternity hospitals of Ulaanbaatar (greater than 95% births were in these facilities)
2. Within 48 hours of birth if resident of Ulaanbaatar
3. Mother was well enough to discuss consent

Participant type


Age group




Target number of participants


Participant exclusion criteria

1. Refusal to consent
2. Birth weight less than 2500 g
3. Less than 36 weeks gestation
4. Obvious congenital abnormalities (with clear medical consequences)
5. Need for infant intensive care treatment
6. Residence in ‘very warm’ apartments (defined by the mother) since a pilot study indicated such families considered the home to be too hot for the baby to be wrapped

Recruitment start date


Recruitment end date



Countries of recruitment


Trial participating centre

Public Health, Epidemiology & Biostatistics
B15 2TT
United Kingdom

Sponsor information


Individual Sponsor (UK)

Sponsor details

c/o Semira Manaseki-Holland
Public Health
Epidemiology & Biostatistics
University of Birmingham
B15 2TT
United Kingdom

Sponsor type




Funder type


Funder name

The Wellcome Trust (UK) (grant ref: 063468)

Alternative name(s)

Funding Body Type

Funding Body Subtype


Results and Publications

Publication and dissemination plan

Not provided at time of registration

Intention to publish date

Participant level data

Not provided at time of registration

Basic results (scientific)

Publication list

1. 2007 results in
2. 2010 results in

Publication citations

  1. Results

    Narangerel G, Pollock J, Manaseki-Holland S, Henderson J, The effects of swaddling on oxygen saturation and respiratory rate of healthy infants in Mongolia., Acta Paediatr., 2007, 96, 2, 261-265.

  2. Results

    Manaseki-Holland S, Spier E, Bavuusuren B, Bayandorj T, Sprachman S, Marshall T, Effects of traditional swaddling on development: a randomized controlled trial., Pediatrics, 2010, 126, 6, e1485-92, doi: 10.1542/peds.2009-1531.

Additional files

Editorial Notes