Skip to main content
. 2019 Feb 19;2019(2):CD012818. doi: 10.1002/14651858.CD012818.pub2

Krebs 2012a (C).

Methods Study design: multisite, cluster‐randomized controlled trial
 Study duration: 12 months
Start date: July 2008
End date: July 2010
Participants Country and settings:

  1. Democratic Republic of Congo (low‐income country); rural areas

  2. Zambia (lower‐middle‐income country); rural areas

  3. Guatemala (lower‐middle‐income country); rural areas

  4. Pakistan (lower‐middle‐income country); urban/peri‐urban communities


Inclusion criteria:

  1. cluster: stunting rates of at least 20% within cluster. General lack of exposure to micronutrient‐fortified products at time of enrollment

  2. individual families: infant aged 3 to 4 months exclusively or predominantly breastfed with intent to continue breastfeeding through 1 year of age


Exclusion criteria:

  1. any family receiving or likely to receive free or subsidized complementary foods

  2. feeding or intent to feed infant formula or micronutrient‐fortified complementary foods

  3. infants with congenital anomaly, infant of multiple births

  4. neurologic deficit at time of enrollment


Nutritional status: baseline mean LAZ: −1.44 in intervention group, −1.32 in control group
Number: 1236 infants (618 in intervention group, 618 in control group, with 20 clusters in each group)
Age: enrollment at approximately 3 months of age; intervention from 6 to 18 months of age
Sex: 53% female in intervention group, 51% female in control group
Typical diet: a pilot study, Krebs 2011, indicated that less than 25% of infants' diets included meats, increasing to greater than 60% in toddlers. Use of micronutrient supplements (vitamin A and iron) highly variable
Interventions Intervention: cooked, diced, lyophilized (freeze‐dried) beef product; 15 g at enrollment increasing to 22.5 g per day at 12 months of age
 Control: micronutrient (zinc and iron)‐fortified rice‐soy cereal supplement, isocaloric to meat supplement; approximately 70 kilocalories/day in 20 g portion increasing to 30 g at 12 months of age
Both groups received 3 educational messages to encourage proper infant and young child feeding:

  1. feed thickened gruels every day;

  2. feed infant/toddler complementary foods (in addition to breastfeeding) at least 3 times a day;

  3. choose a variety of local foods.
Outcomes Primary:

  1. linear growth velocity

  2. weight gain


Secondary:

  1. biomarkers (hemoglobin, ferritin, transferrin receptor, zinc, vitamin B12)

  2. development (Bayley Scales of Infant Development II)


Secondary outcomes were collected from a convenience sample of ˜300 participants per group (60%) of total participants.
Measurement:

  1. linear growth: recumbent length measured with seca infantometer (model 416), duplicate measurements with third measurement, if needed

  2. weight: naked weights recorded using a 334 infant scale accurate to 5 g, duplicate measurements with third measurement, if needed

  3. biomarkers: 3 mL of blood collected via antecubital venipuncture, 2 hours after eating

  4. development: obtained using a standardized Bayley Scales of Infant Development II test kit, instructions, and evaluation/scoring forms, conducted in consistent location and arranged to provide standardized development


Time points:

  1. 6, 9, 12, and 18 months of age for linear growth

  2. 18 months of age for biomarkers and developmental markers. Hemoglobin concentrations specifically were obtained from 63% to 77% of children at 3 of the 4 sites, and not collected from the site in Pakistan.
Notes Funding: "Supported by grants from Eunice Kennedy Shriver National Institute of Child Health and Human Development [HD040657 (UCD), HD043464 (UAB), HD040607 (Drexel), HD043475 (UNC), HD040636 (RTI)], Office of Dietary Supplements, and National Institute of Diabetes and Digestive and Kidney Diseases 9K24 DK083772. The National Cattlemen’s Beef Association partially supported the analyses of the biomarkers for this project and had no input into the study design, implementation, analysis, or interpretation of the data."
Declared conflict of interest: none declared
Other notes: attrition ˜14%; balanced between groups, both in number (86 in intervention group, 88 in control group) and reason
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence generation (selection bias) Low risk Comment: computer‐generated randomization algorithm, stratified by stunting rates within communities
Allocation concealment (selection bias) Low risk Comment: central randomization of clusters after individual participants were recruited within clusters
Blinding of participants and personnel (performance bias) 
 All outcomes Low risk Comment: unable to blind due to nature of intervention, but geographic distance between clusters minimized risk of contamination of intervention or communication among study participants in different clusters
Blinding of outcome assessment (detection bias) 
 All outcomes Low risk Comment: objective outcome assessment unlikely to have introduced bias
Incomplete outcome data (attrition bias) 
 All outcomes Low risk Comment: attrition roughly equal between intervention (13.9%) and control (14.2%) groups
Selective reporting (reporting bias) Low risk Comment: all prespecified outcomes reported
Other bias Unclear risk Comment: partially supported by National Cattlemen’s Beef Association