Childhood Mortality
Conditions
Keywords
Childhood mortality, Azithromycin, Mass treatment, Vaccine Visit targeted treatment
Brief summary
An estimated 7.7 million pre-school aged children die each year, the majority from infectious diseases. Mass azithromycin distributions for trachoma may have the unintended benefit of reducing childhood mortality. We recently demonstrated the biannual mass azithromycin distribution significantly reduces all-cause child mortality in a cluster randomized trial (MORDOR I) conducted in three diverse regions of Sub-Saharan Africa. Our long-term goal is to more precisely define the role of mass azithromycin treatments as an intervention for reducing childhood morbidity and mortality. We propose a cluster randomized trial designed to repeat the original study to confirm the original results in a different geographic study with similarly high child mortality, and to better understand the mechanism behind any effect of azithromycin on child mortality. We hypothesize that biannual mass azithromycin distribution will reduce child mortality compared to placebo, and that this effect will be primarily driven by a reduction in infectious burden. Objectives: 1. Determine the efficacy of biannual mass azithromycin distribution versus placebo in children aged 1-59 months for reduction in all-cause mortality. 2. Determine the efficacy of targeted azithromycin distribution to infants during an early infant healthcare visit (approximately 5th through 12th week of life) on infant mortality. 3. Determine the mechanism behind the effect of biannual mass azithromycin distribution for reduction in child mortality. The study will be conducted in the Nouna District in northwestern Burkina Faso.
Detailed description
Although child health and mortality are improving worldwide, children in the Sahel and sub-Sahel regions of West Africa have the greatest risks of mortality.Burkina Faso's current under-5 mortality rate is estimated 110 per 1,000 live births. Similar to other countries in the region, the major causes of child mortality in Burkina Faso are malaria, respiratory tract infection, and diarrhea. Malnutrition acts as a major underlying contributor to mortality. Interventions that address these underlying causes may be particularly efficacious for reducing mortality. Younger children at are at a higher risk of mortality. Approximately 2/3rd of under-5 deaths occur during the first year of life. In general, the child mortality rate decreases as age increases. While some improvement has been observed, neonatal mortality is declining at a slower rate than post-neonatal childhood mortality. Many child health interventions are designed specifically for children over 6 months of age, such as vitamin A supplementation, seasonal malaria chemoprevention, and lipid-based nutritional supplementation. Identification of strategies that are safe and effective for the youngest children will be required to address persistently high rates of neonatal and infant mortality. The MORDOR I study demonstrated a significant reduction in all-cause child mortality following biannual mass azithromycin distribution. Across three diverse geographic locations in sub-Saharan Africa (Malawi, Niger, and Tanzania), biannual mass azithromycin distribution over a two-year period led to a 14% decrease in all-cause child mortality. In Niger, 1 in 5-6 deaths were averted. These results are qualitatively similar to those of a previous study of mass azithromycin distribution for trachoma control in Ethiopia, which found reduced odds of all-cause mortality in children in communities receiving mass azithromycin compared to control communities. In MORDOR I, the strongest effect of azithromycin was in the youngest cohort of children. Across all three countries, the strongest effect of azithromycin was consistently in children 1-5 months of age, with an approximately 25% reduction in all-cause mortality. However, MORDOR I was not optimized to target the youngest age groups. Although children as young as 1 month were eligible, biannual distributions might not reach some children until 7 months of age. On average, children were first treated at 4 months. Given that there may be a substantial benefit to treating children at younger ages, azithromycin strategies that are designed to target younger age groups may be even more beneficial for reducing child mortality. Here, we propose a randomized controlled trial designed to evaluate the efficacy of mass and targeted azithromycin strategies for child mortality. In the rural northwestern district of Nouna in Burkina Faso, we propose to randomize villages to biannual mass azithromycin distribution or placebo. This study was designed by CRSN and UCSF partners to confirm the results of MORDOR I, evaluate an alternative health systems distribution point (the vaccine visit) for delivery of azithromycin to young children, and to provide a platform for evaluation of potential mechanisms behind the effect of azithromycin by collecting and processing additional specimens and tests. Objectives: 1. Determine the efficacy of biannual mass azithromycin distribution versus placebo in children aged 1-59 months for reduction in all-cause mortality. 2. Determine the efficacy of targeted azithromycin distribution to infants during an early infant healthcare visit (approximately 5th through 12th week of life) on infant mortality. 3. Determine the mechanism behind the effect of biannual mass azithromycin distribution for reduction in child mortality. Study Design: CRSN and UCSF (hereafter, we) will assess childhood mortality over three years, comparing communities where children aged 1-59 months receive biannual oral azithromycin and/or targeted azithromycin during the 5th-12th week of life in conjunction with the first Expanded Programme on Immunization (EPI) vaccine visit or biannual placebo and targeted placebo. All eligible communities in Nouna District will be randomized (278 communities). A random sample of 48 (12/arm) communities from within the HDSS will be selected to participate in the Mortality Plus study, which will entail an annual morbidity exam among 15 randomly selected children per community to monitor infectious disease morbidity, nutritional status, and macrolide resistance. All communities will contribute to the mortality outcome.
Interventions
DRUGAzithromycin
biannual azithromycin in eligible communities to children 1 to 59 months old Targeted azithromycin to children aged 5 to 8 weeks old at the vaccine visit
DRUGPlacebos
biannual placebo in eligible communities to children 1 to 59 months old Targeted placebo to children aged 5 to 8 weeks old at the vaccine visit
Sponsors
Centre de Recherche en Sante de Nouna, Burkina Faso
Bill and Melinda Gates Foundation
University of California, San Francisco
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
QUADRUPLE (Subject, Caregiver, Investigator, Outcomes Assessor)
Masking description
The trial sites will be masked to outcomes, so the responsibility for monitoring interim analysis will fall on the DSMC
Intervention model description
All eligible communities in Nouna District will be randomized in a 1:1 fashion to biannual azithromycin or placebo. Targeted treatment (vaccine visit) will be randomized 1:1 individually to azithromycin or placebo. Randomization will be conducted by T. Porco. Procedural and algorithmic details are provided in an appendix to the Statistical Analysis Plan.
Eligibility
Sex/Gender
ALL
Age
1 Months to 59 Months
Healthy volunteers
Yes
Inclusion criteria
Communities: Inclusion Criteria: * The community location in target district. * The community leader consents to participation in the trial (this does not obviate the need for individual consent, but without overall leadership consent, the community as a whole cannot be part of the trial). * Eligible communities estimated population of between 200-2,000 people * The community is not in an urban area
Exclusion criteria
\- Refusal of village chief Individuals: Inclusion Criteria * All children in the study communities aged 5 to 12 weeks old at the time of the vaccination visit are eligible to participate * Ability to feed orally * Appropriate consent from at least one caregiver * Family intends to stay within the study area
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| All-cause Mortality Rate in Children Aged 1-59 Months | 36 months | All-cause mortality as determined by biannual census among children aged 1-59 months |
| All-cause Mortality Rate in Individually Randomized Children at 4-12 Weeks of Age | 6 months | All-cause mortality as determined by a follow-up visit for individually randomized children at healthy child visits |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Height-for-age Z-score in Individually Randomized Children at Healthy Child Visits | 6 months | The Height-for-age Z-score (HAZ) is calculated using height (cm) measurements taken at healthy child visits and converted into Z-scores using the WHO Growth Standards. Z-scores represent the number of standard deviations from the median of a reference population. A HAZ of 0 represents the median of the reference population, while negative Z-scores indicate below-average height-for-age, and positive Z-scores indicate above-average height-for-age. A HAZ below -2 is indicative of stunting, reflecting chronic undernutrition or growth failure. |
| Mid-upper Arm Circumference in Individually Randomized Children at Healthy Child Visits | 6 months | — |
| Malaria Parasitemia in Children 1-59 Months at 36 Months | 36 months | Malaria parasitemia as measured by thin and thick smears in a random sample of children at 36 months |
| Weight Gain in Individually Randomized Children | 6 months | Change in weight per day from baseline to 6 months |
| Linear Growth in Individually Randomized Children | 6 months | Change in length per day from baseline to 6 months |
| Weight-for-height Z-score in Individually Randomized Children at Healthy Child Visits | 6 months | The WHZ is calculated using weight (kg) and height (cm) measurements taken at healthy child visits and converted into Z-scores using the WHO Growth Standards. Z-scores represent the number of standard deviations from the median of a reference population. A WHZ of 0 represents the median of the reference population, while negative Z-scores indicate below-average weight-for-height, and positive Z-scores indicate above-average weight-for-height. A WHZ below -2 is indicative of wasting, while a WHZ above +2 is considered overweight. |
Countries
Burkina Faso
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Biannual Mass Oral Azithromycin Bi-annual Mass Azithromycin distribution to all children 1-60 months old in participating communities
Azithromycin: biannual azithromycin in eligible communities to children 1 to 59 months old Targeted azithromycin to children aged 5 to 8 weeks old at the vaccine visit | 22,148 |
| Biannual Mass Oral Placebo Bi-annual Mass Placebo distribution to all children 1-60 months old in participating communities
Azithromycin: biannual azithromycin in eligible communities to children 1 to 59 months old Targeted azithromycin to children aged 5 to 8 weeks old at the vaccine visit
Placebos: biannual placebo in eligible communities to children 1 to 59 months old Targeted placebo to children aged 5 to 8 weeks old at the vaccine visit | 22,639 |
| Targeted Oral Azithromycin Targeted azithromycin to children 5 to 12 weeks old at vaccine visit or other healthy child visit
Azithromycin: biannual azithromycin in eligible communities to children 1 to 59 months old Targeted azithromycin to children aged 5 to 8 weeks old at the vaccine visit | 16,416 |
| Targeted Oral Placebo Targeted placebo to children 5 to 12 weeks old at vaccine visit or other healthy child visit
Placebos: biannual placebo in eligible communities to children 1 to 59 months old Targeted placebo to children aged 5 to 8 weeks old at the vaccine visit | 16,461 |
| Total | 77,664 |
Baseline characteristics
| Characteristic | Total | Biannual Mass Oral Placebo | Biannual Mass Oral Azithromycin | Targeted Oral Azithromycin | Targeted Oral Placebo |
|---|---|---|---|---|---|
| Age, Continuous | 10 month STANDARD_DEVIATION 19.22 | 31 month STANDARD_DEVIATION 16.71 | 31 month STANDARD_DEVIATION 16.71 | 1.51 month STANDARD_DEVIATION 0.51 | 1.55 month STANDARD_DEVIATION 0.51 |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Black or African American | 77664 Participants | 22639 Participants | 22148 Participants | 16416 Participants | 16461 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Region of Enrollment Burkina Faso | 77664 participants | 22639 participants | 22148 participants | 16416 participants | 16461 participants |
| Sex: Female, Male Female | 38313 Participants | 11249 Participants | 10883 Participants | 8045 Participants | 8136 Participants |
| Sex: Female, Male Male | 39351 Participants | 11390 Participants | 11265 Participants | 8371 Participants | 8325 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk | EG003 affected / at risk |
|---|---|---|---|---|
| deaths Total, all-cause mortality | 498 / 34,399 | 588 / 33,847 | 82 / 16,416 | 75 / 16,461 |
| other Total, other adverse events | 77 / 269 | 14 / 199 | 62 / 1,606 | 79 / 1,532 |
| serious Total, serious adverse events | 0 / 269 | 0 / 199 | 5 / 1,606 | 1 / 1,532 |
Outcome results
Primary
All-cause Mortality Rate in Children Aged 1-59 Months
All-cause mortality as determined by biannual census among children aged 1-59 months
Time frame: 36 months
Population: Primary analysis included 34399 children in azithromycin clusters and 33847 children placebo clusters during 6 census periods. Person-time for the denominator was calculated by summing the person-time for each intercensal period. Children who were alive during both censuses contributed the person-time for that intercensal period. A cluster that was not censused, eg, due to security, could reenter the study at a later census.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Biannual Mass Oral Azithromycin | All-cause Mortality Rate in Children Aged 1-59 Months | 8.2 deaths per 1000 person years |
| Biannual Mass Oral Placebo | All-cause Mortality Rate in Children Aged 1-59 Months | 10.0 deaths per 1000 person years |
Primary
All-cause Mortality Rate in Individually Randomized Children at 4-12 Weeks of Age
All-cause mortality as determined by a follow-up visit for individually randomized children at healthy child visits
Time frame: 6 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Biannual Mass Oral Azithromycin | All-cause Mortality Rate in Individually Randomized Children at 4-12 Weeks of Age | 82 Participants |
| Biannual Mass Oral Placebo | All-cause Mortality Rate in Individually Randomized Children at 4-12 Weeks of Age | 75 Participants |
Secondary
Height-for-age Z-score in Individually Randomized Children at Healthy Child Visits
The Height-for-age Z-score (HAZ) is calculated using height (cm) measurements taken at healthy child visits and converted into Z-scores using the WHO Growth Standards. Z-scores represent the number of standard deviations from the median of a reference population. A HAZ of 0 represents the median of the reference population, while negative Z-scores indicate below-average height-for-age, and positive Z-scores indicate above-average height-for-age. A HAZ below -2 is indicative of stunting, reflecting chronic undernutrition or growth failure.
Time frame: 6 months
Population: 13641 children in Targeted oral azithromycin arm and 13657 children in Targeted oral placebo arm had valid height and weight measurement
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Biannual Mass Oral Azithromycin | Height-for-age Z-score in Individually Randomized Children at Healthy Child Visits | -0.31 z score | Standard Deviation 1.3 |
| Biannual Mass Oral Placebo | Height-for-age Z-score in Individually Randomized Children at Healthy Child Visits | -0.31 z score | Standard Deviation 1.3 |
Secondary
Linear Growth in Individually Randomized Children
Change in length per day from baseline to 6 months
Time frame: 6 months
Population: 13641 children in azithromycin group and 13657 children in placebo group had valid length measurement at 6 month
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Biannual Mass Oral Azithromycin | Linear Growth in Individually Randomized Children | 8.0 mm/day | Standard Deviation 2.1 |
| Biannual Mass Oral Placebo | Linear Growth in Individually Randomized Children | 8.0 mm/day | Standard Deviation 2.3 |
Secondary
Malaria Parasitemia in Children 1-59 Months at 36 Months
Malaria parasitemia as measured by thin and thick smears in a random sample of children at 36 months
Time frame: 36 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Biannual Mass Oral Azithromycin | Malaria Parasitemia in Children 1-59 Months at 36 Months | 90 Participants |
| Biannual Mass Oral Placebo | Malaria Parasitemia in Children 1-59 Months at 36 Months | 114 Participants |
Secondary
Mid-upper Arm Circumference in Individually Randomized Children at Healthy Child Visits
Time frame: 6 months
Population: 13641 children in azithromycin group and 13657 children in placebo group had valid MUAC measurement at 6 month
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Biannual Mass Oral Azithromycin | Mid-upper Arm Circumference in Individually Randomized Children at Healthy Child Visits | 13.8 cm | Standard Deviation 1.2 |
| Biannual Mass Oral Placebo | Mid-upper Arm Circumference in Individually Randomized Children at Healthy Child Visits | 13.8 cm | Standard Deviation 1.2 |
Secondary
Weight-for-height Z-score in Individually Randomized Children at Healthy Child Visits
The WHZ is calculated using weight (kg) and height (cm) measurements taken at healthy child visits and converted into Z-scores using the WHO Growth Standards. Z-scores represent the number of standard deviations from the median of a reference population. A WHZ of 0 represents the median of the reference population, while negative Z-scores indicate below-average weight-for-height, and positive Z-scores indicate above-average weight-for-height. A WHZ below -2 is indicative of wasting, while a WHZ above +2 is considered overweight.
Time frame: 6 months
Population: 13641 children in Targeted oral azithromycin arm and 13657 children in Targeted oral placebo arm had valid height and weight measurement
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Biannual Mass Oral Azithromycin | Weight-for-height Z-score in Individually Randomized Children at Healthy Child Visits | -0.64 z-score | Standard Deviation 1.2 |
| Biannual Mass Oral Placebo | Weight-for-height Z-score in Individually Randomized Children at Healthy Child Visits | -0.64 z-score | Standard Deviation 1.2 |
Secondary
Weight Gain in Individually Randomized Children
Change in weight per day from baseline to 6 months
Time frame: 6 months
Population: 13641 children in azithromycin group and 13657 children in placebo group had valid weight measurement at 6 month
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Biannual Mass Oral Azithromycin | Weight Gain in Individually Randomized Children | 18 g/day | Standard Deviation 6.2 |
| Biannual Mass Oral Placebo | Weight Gain in Individually Randomized Children | 18 g/day | Standard Deviation 6.4 |