Study design, setting and participants
This is a phase III, double-blind, randomised placebo-controlled trial drawing on clinical and community based samples. The trial has been approved by the Royal Children’s Hospital Human Research Ethics Committee (HREC 30111).
Recruitment is from a range of services widely used by and readily accessible to parents seeking medical advice regarding their crying babies in Melbourne, Australia. These comprise: the Royal Children’s Hospital Emergency Department (RCH ED), the RCH Unsettled Babies clinic (a tertiary referral-based clinic for assessment and management of unsettled babies), Tweddle Child and Family Health Centre (a mother-infant parenting centre), two Maternal Child Health centres (universal nurse health checks in the Boroondara and Moonee Valley districts), and paediatricians at the RCH and in private practices. Interested families can also directly contact the study team to be involved.
A total sample of 160 infants less than three months (13.0
weeks) old with infant colic is being recruited between August 2011 and August 2012. The expected total duration of the study from the start of recruitment to the last subject finishing the six month follow-up is 18
months. The treatment period is one month, with a follow-up period of six months.
Each infant must meet all of the following criteria to be enrolled in this study:
1. Infant colic, i.e. crying or fussing
days over seven
days (as defined by the modified Wessel’s criteria) by caregiver’s report at the time of study commencement;
2. Less than three
months (i.e. up to and excluding 13.0
weeks) old at the time of study commencement;
3. Greater than 36
weeks gestation at birth; and
4. Birth weight of more than 2500
Infants meeting any of the following criteria are excluded from the study:
1. Failure to thrive (weight gain
100 grams/week averaged from birth to the last recorded weight);
2. Major medical problems (eg. ill, immunocompromised, major developmental or genetic abnormality);
3. Taking solids, antibiotics or L reuteri and, if breastfeeding, has mother taking L reuteri at the time of study commencement;
4. Cow’s milk protein allergy, defined as resolution of crying after a paediatrician-recommended two week trial of dairy-free diet in the infant, or in the mother if breast-feeding, at the time of study enrolment; and
5. Caregiver has insufficient English to understand informed consent and complete questionnaires.
A block randomisation schedule to maintain balance between treatment arms has been prepared by an independent statistician, not directly involved in the analysis of the study results, from the Clinical Epidemiology and Biostatistics Unit (CEBU) at the RCH and supplied to Pharmacy who dispense the study product. Randomisation is stratified by method of infant feeding (breastfed versus formula-fed) and age at the time of study commencement (≤6
weeks versus >6
weeks, due to the natural crying peak occurring at around six
weeks of age). Infants on a mixture of both breast- and formula-feeding are allocated to the formula-fed stratum. To minimise potential biases, the study is double-blind whereby treatment allocation is concealed from all study investigators and participants.
Treatment arms, dosage and route of administration
The intervention is L reuteri DSM 17938
cfu per drop as five drops per day) in an oil suspension. Under refrigeration, it is stable for 21
months at 2°C to 8°C (as documented by the manufacturer, BioGaia AB, Stockholm, Sweden). The placebo is maltodextrose in the same oil suspension, identically packaged and stored, and has the same appearance, colour and taste as the intervention. Both intervention and placebo are labeled with only the randomisation number, batch number, expiry date, and the statement “For clinical trial use only”.
Carers administer five drops of study product orally to each infant once daily for 28
days. The dose does not need to be given at a fixed time each day, nor does it need to be given with feeds. However, for compliance and ease of administration, we recommend that families give the dose with the same feed each day.
shows the primary and secondary outcome measures at different time points of the study. Table
shows other information collected at different time points, including family demographics, potential confounders, physical assessments, and records of compliance and side effects.
Additional information collected
Study procedure details
show details of study procedures.
Perera diagram showing treatment and follow-up procedures.
Eligible families who present through the RCH ED are identified in three ways: i) by the treating doctor, ii) by the ED social worker, or iii) through the ED electronic database. Eligible families who present through the RCH Unsettled Babies Clinic are identified by their treating doctors. Eligible families who present through the Boroondara and Moonee Valley Maternal and Child Health Nurses and Tweddle Child and Family Services are identified by their nurses or intake worker. Potentially eligible families are referred to study team members, who call families or approach them in the ED or clinic if appropriate.
The study team establishes eligibility criteria. Infants and breastfeeding mothers who are on a paediatrician-recommended dairy-free diet, infants who are taking L reuteri/solids/antibiotics, and breastfeeding mothers who are taking L reuteri at the time of study enrolment are reassessed for eligibility by telephone call after two weeks. Eligible families are approached for consent. Each family can choose to be visited at home or attend a visit at RCH for the first study visit. Study researchers call the families one day before the proposed study visit to confirm the home or RCH visit. During this call, study researchers reconfirm eligibility and consent, remind families who have not completed the consent form and baseline questionnaire to complete them, and instruct caregivers to collect a stool sample from their babies and store it in a freezer. Study researchers also record on the Clinical Record Form whether the babies are breast- or bottle-fed.
During the visit, the researcher reconfirms eligibility, asks the mother to complete the baseline questionnaire, and reviews with the family in detail the study diary, anticipated phone calls, the procedure of administering the study product to the infant, and the anticipated procedures at the end of the study. The researcher weighs the baby, collects the stool sample and transports it back to the RCH on ice in a portable cooler, and then stores it in the laboratory −80°C freezer.
Study period Days 1 – 28
On days 1, 7, 14 and 21 the study team sends a mobile text (SMS) message to families to remind families to fill in the weekly diary. Study researchers call families on day 28 to remind families to a) stop administration of the study product; b) fill in the one
month questionnaire; c) collect a stool sample; and d) make an appointment for the second (final) visit.
During the second visit, the researcher weighs the infant, and collects a) the study diary; b) the one
month questionnaire; c) the study product bottle and d) a stool sample which is transported back to the RCH on ice in a portable cooler, and then stored in the laboratory -80°C freezer.
Six month follow-up
At six months, the families are sent a letter together with a 24-hour study diary, a questionnaire and a reply-paid envelope. Caregivers are asked to complete the study diary and questionnaire, and return both in the enclosed reply-paid envelope.
Compliance with the study product
Compliance is assessed by weighing of study bottles pre- and post- dispensing, collection of faecal samples at the end of the intervention period to assess for presence of L reuteri in each infant’s stool, and weekly diary to record administering of the product.
Estimation of sample size
We have selected a sample size of 160 to provide 80% power to detect a minimum effect size of 0.5 standard deviations, difference in the mean daily crying time between treatment groups with a significance level of p
0.05, allowing for a drop out rate of 20%. If the data are substantially skewed, 80% power would be maintained to detect a minimum effect size of 0.525 [63
]. In Savino’s trial [60
], the percentage reduction in median daily crying time at day 28 compared to baseline was 70% for the probiotic and 26% for simethicone, suggesting substantive differences in this outcome are likely. With a much larger sample size (double that of Savino’s trial) in our study, we are confident we would also be able to detect a smaller difference in daily crying time reduction. Our sample size of 160 infants provides 80% power to detect a minimum effect size of 0.85 standard deviations, difference in the mean daily crying time between treatment groups within either the formula-fed or breastfed babies, assuming 40% of infants in the sample are exclusively breastfed.
We will be conducting an intention-to-treat analysis in which participants are compared according to the group to which they were randomly allocated regardless of participants' compliance, crossover to other treatments or withdrawal from the study. This approach preserves the prognostic balance in the study arms achieved by randomisation.
Baseline characteristics and study outcomes will be described for each treatment group using means and standard deviations for normally distributed continuous outcomes, while medians and inter-quartile ranges will be used for continuous outcomes that are skewed. Proportions for categorical data will be given. All primary and secondary outcomes for the intervention versus control groups will be compared with t tests and non-parametric tests for continuous data, and chi squared tests for dichotomous data. The primary outcome is the crying time/24
hours at day 28. We will additionally consider a dichotomised indicator of treatment response defined as a 50% reduction in crying time. Subgroup analyses are intended a priori to examine treatment differences amongst breastfed babies and those who are formula-fed, and amongst infants with a family history of atopy and those without.
Regression models will be used to estimate treatment effect sizes, odds ratios and 95% confidence interval, adjusting for potential confounding factors identified a priori and measured at baseline. These include infant age (≤6
weeks versus >6
weeks, taking into account the natural course of infant colic symptomatology with the peak of crying at six
weeks), mode of birth delivery, maternal smoking, use of antibiotics or probiotics in infants and breastfeeding mothers, maternal diet if breastfeeding, maternal mental health scores, and known causes of crying (e.g. fever, eczema, vaccination, vomiting, and diarrhoea). Random effects regression models will be used for further longitudinal analysis examining trends in crying time within individual babies.