Recruitment timeline and challenges
The participant recruitment period started end of April 2017 and finished end May
2018. The original timeline for the recruitment was between May 2016 and January 2017.
Delay in trial commencement was due to several unforeseen problems with the export
of the fortified milk from Europe to Brazil, long process for research ethical clearance
and the need to re-select the study site.
The research team developed a tailored fortified milk powder for non-commercial use. Acquisition of raw material, blending, packaging, physical and chemical analysis
of the product (fortified milk and plain milk) were performed in Denmark. Export of
dairy products to Brazil is tightly regulated, particularly and unanticipated for
non-registered and non-commercial products. Given the major delays in the product
export the research team decided to change the research protocol to offer women a
local commercially available milk powder and individual sachets with vitamin D3 and
CAPOLAC® (source of calcium) for domestic fortification. CAPOLAC® and vitamin D3 were blended and stored in single-dose sachets. All women were provided
with two daily doses of semi-skimmed milk powder (20g) to be reconstituted in 200ml
of potable water for each serving and two sachets. Participants in the fortified intervention
group received sachets with calcium and vitamin D and those in the plain milk group
received placebo sachets.
The team also faced some challenges when selecting the study site. Several contacts
with the local centres were made in 2015, while the research team was seeking funding
and developing the study application, to obtain information on the centre prenatal
logistics, monthly uptake, demographic profile, facilities, and provision of dental
care to pregnant women. For safety reasons (especially regarding the security of the
field workers), the involvement of sites in areas at risk of conflict and violence
was excluded. The safety of centres located in slums worsened during this selection
process and became impossible to include them as recruitment sites. Therefore, two
large health centres located in Duque de Caxias, with catchment area including deprived
areas, were selected. Consent was given by the centres directors to conduct the studies
within centres’ premises with minimal interference in their routine practice. However,
the senior management of the centres changed after local elections in 2016 and consent
from one of the centres was withdrawn.
Between April 2017 and May 2018 (53 weeks), participant recruitment was halted several
times due to strikes of health care professionals in different municipalities of Rio
de Janeiro, an episode of armed robbery, local riots, and public holidays. All primary
health care services in Duque de Caxias were closed for a total of 16 weeks between
July and October 2017 because of general strikes. After the re-opening of the health
centre in October 2017, the research team was victim of an armed robbery and had equipment
and personal belongings robbed. Consequently, fieldwork, including recruitment of
new participants, was interrupted for two weeks. The health centre did not book new
prenatal appointments for the week between Christmas and new year (December 2017)
and during the week of the carnival celebration (February 2018). Therefore, recruitment
was interrupted for two additional weeks. Furthermore, primary care centres were closed
during national and local holidays (non-overlapping with strikes, carnival and Christmas
break), resulting in four extra days of recruitment interruption.
During the recruitment stage, Brazil was facing a period of political instability
prior to and after the presidential impeachment in late 2016. The state of Rio de
Janeiro also faced health care and security crises during this period. There were
several riots and dragnets in Duque de Caxias, resulting in closure of local shops
and some public places for safety reasons. Although the health centre was open on
these occasions the overall number of service users attending the centre was greatly
reduced.
Overall, during the 53 weeks of enrolment period, recruitment was interrupted for
approximately 21 weeks.
Recruitment target
The total study recruitment target was of 120 women from two sites over 36 weeks and
the estimated recruitment rate was of 1.7 participants per week/ centre (120/36/2).
However, only one site was involved, and 70 women were recruited (58.3% of the original
target). The actual crude and net recruitment rates were 1.3 women/week/ centre (70/53
week) and 2 women/week/ centre (70/32weeks), respectively.
Exclusions and non-participations
In total, 767 pregnant women were approached by the research team, however 548 women
(71.5%) did not meet the initial eligibility criteria. The largest exclusion category
(58%) was advanced pregnancy beyond week 20 at first prenatal visit (318 out of 548).
This was followed by declined invitation (9.1%) and lack of contact details (8.6%)
to invite women to attend additional assessment visits (e.g. periodontal examination),
presence of high blood pressure (4.2%), use of fixed braces (4.2%), non-appearance
for the HIV/Syphilis test (3.1%) within the study timeframe (before 20 weeks gestation),
presence of kidney stones (2.9%) or syphilis (2.9%). Only (1.8%) reported either milk
allergy, lactose intolerance or daily consumption of vitamin D and supplement. Figure
2 provides more details on reasons for exclusion. Reasons for declined study invitation
were not recorded for all 50 declining women. Some women were limited by childcare
or employment obligations.
Of those invited for the periodontal examination (dental screen, n= 219), 92 were
eligible (42%).
Figure 2. Study enrolment and exclusion flow diagram
Non-participation of eligible women was identified after periodontal examination.
Of the women meeting the eligibility criteria (n=92), 9 (9.8%) did not attend the
baseline blood test, 2 (2.2%) had a miscarriage, 3 (3.2%) presented high blood pressure
and 8 (8.7%) withdrew consent. The reasons for withdrawal and missed appointment varied.
This included transportation barriers, family obligations, moved to another area,
changed prenatal care site, unresponsive to phone calls and disconnected telephones.
In total, 70 women were randomised. One woman asked to leave the study right after
randomisation. The recruitment yields showed that 9.1% (70/767) of all women contacted
at first prenatal visit and 76.1% (70/92) of those screened eligible were enrolled
into the trial.
Barriers and facilitators
The qualitative data regarding factors which could hinder or facilitate recruitment
and enrollment to the study was divided into five themes (Table 2): study design and
intervention; food myths, social support, views on prenatal care services and finance.
Table 2. Factors influencing recruitment and participation in the study.
In general, women did not report concern about potential random allocation into different
groups given that all groups would benefit from the intervention. They also reported
interventions to be acceptable and relevant. However, women reported some resistance
to drinking pure milk and suggested consuming the milk in smoothies and porridges
instead. One woman stated that “If it is mixed with chocolate powder, I can have it. But it is manageable if I add
it in foods, canjica and porridges”.
Women were asked whether they would share the milk with the rest of the family. Most
women reported that they would share the milk with their children ("Yes, because they would want it and I would give the milk to my children”), but some reported that they would try to consume the milk when the children were
not around (“but when it is time to drink the milk, it is not necessary to drink in front of my
son. It can be consumed in the morning when they are sleeping”). One woman reported that she would “drink when the children are at school”.
No major concern regarding safety of milk consumption was observed apart from cases
when the participant had food intolerance. One of the participants commented “if there are recommendations in relation to gestation, I would take it only during
pregnancy and that’s it. I would report if I have any side effect”. Positive attitudes toward consumption of fortified food was observed. One woman
mentioned “the more vitamins the better”.
Regarding food myths and cultural beliefs, they reported that some foods must be avoided
during pregnancy, but milk was not one of them. Women talked positively about milk
consumption during pregnancy. Some believed that consumption of milk and canjica would
increase breast milk production. One of the participants said “My grandma who is old fashion used to say that you should always eat canjica, oranges
with beetroot, and beans. You must have canjica”.
Lack or limitted social support was observed among women. Most women counted on emotional
and informational support (e.g. advice, suggestion and information) from their mothers
(“my husband works, I count on my mum”). Women who had other children often reported childcare difficulties (“I am alone, my mum only takes care of my son when I go to work. That’s all. She is
my family. It is only me and God”).
Regarding their views on prenatal care, women had a positive view on the health care
centre and trusted the doctors (“.. I liked the treatment I received here when I had my daughter 10 years ago... I
trust it”). They reported they chose the centre due to the quality of service and indication
from friends or family (“My friend recommended, she said it is very good”).
Most women were unemployed, and some did not have permanent accommodation and reported
living with extended family or in-laws. Most women had mobile phones, which could
facilitate contact with the health centre, but limited credit to make phone calls.
Cost of transport was cited as one barrier to attend the prenatal care. One woman
said: “Remember to talk about the cost of transport". Another woman complemented the comment by mentioning. “There is no money for the ticket. (…) sometimes there is no money in the house
to go to the health centre.”