Data collection and ethics
The questionnaire used in the face-to-face interview was
adapted from the validated Nepal Demographic and Health
Survey and another breastfeeding instrument [14], and
was subsequently pretested on 25 postpartum women
for cultural appropriateness, content validity and understanding;
see Additional file 1. Fifteen female data enumerators
conducted the baseline interview from December
2011 to January 2012 to collect information on sociodemographics
and obstetric characteristics of the pregnant
women. The cohort of participants was then followed up
at 4 weeks, 12 weeks and 22 weeks postpartum through repeated
household visits by the same enumerators. Infant
feeding information, including breastfeeding knowledge
and support, husband’s and mother/mother-in-law’s preferences
and feeding plans, was collected at 4 weeks during
the second visit. Attitude towards breastfeeding was asked
at the third visit, perception of sufficiency of breast milk
was solicited at 4 and 22 weeks, whereas supplementary
feeding was assessed at all three follow-up visits.
The study was approved by the Human Research
Ethics Committee of Curtin University (approval number
HR 130/2011), Ethical Review Board of the Nepal
Health Research Council (approval number 88/2011) and
the District Public Health Office of Kaski. An information
sheet was distributed and read to each participant before
obtaining her signed or thumb-print consent. Confidentiality
of the information provided was maintained throughout
the study.
Statistical analysis
All data entry was performed by the first author, who
further checked for correctness and consistency upon
completion. Descriptive statistics were used to summarise
the outcome variables and other variables of interest
using the SPSS package version 20. Four levels of education
were recorded: none, primary (1-5th grade), secondary
(6-10th grade) and college (after 10th grade). Employment
status was categorised as: employed (full-time salaried job),
semi-employed (wage based labour, small business or
employed abroad), and unemployed (agricultural, housewife
or nothing). ‘Breastfeeding information’ was recorded
as “none” if the woman did not receive such information
from anywhere. Absence of ‘encourage to breastfeed’
meant there was no encouragement or motivation from
anybody else to breastfeed after delivery. Similarly, lack of
‘advice on feeding method’ implied no one had advised
the participant how to feed her newborn. Intended
duration of breastfeeding and target time to introduce
solid foods were initially recorded as continuous variables
but recoded as categorical to facilitate analysis.
Results
Participant profile
Of the 639 women who took part in the second interview
at 4 weeks after delivery, 615 and 515 participants
were eventually followed up at the third (12 weeks) and
fourth (22 weeks) interviews. Attrition was due to lost to
follow up (n = 39) during the period and non-completers
at 5 months (n = 85). Table 1 shows the socio-demographic
characteristics of the cohort. At baseline about half the
women were first time mothers, within the age range 20–
24 years, and belonged to the upper caste. Almost 80% of
them were unemployed while the majority (70.8%) received
secondary or above education. These characteristics were
similar at subsequent follow ups (Table 1). Moreover, no
significant differences in these variables were found between
drop outs and women remaining in the cohort at
the two time points, suggesting that the loss-to-follow-up
was random and no apparent bias was introduced through
attrition.