Embedded Files
CONTEXT
CONTEXT
Globally, 149 million children under five years of age are stunted and 2 out of 5 of these stunted children live in low-income countries. Pakistan stands at a stunting rate of 38% and this statistic has remained fairly constant over the last two decades. Stunting hampers children's cognitive development and leads to poor performance at school and work.
Globally, 149 million children under five years of age are stunted and 2 out of 5 of these stunted children live in low-income countries. Pakistan stands at a stunting rate of 38% and this statistic has remained fairly constant over the last two decades. Stunting hampers children's cognitive development and leads to poor performance at school and work.
Reducing under-nutrition among children remains a critical global public health and policy challenge. Despite recent improvements globally, 1 in 4 children of pre-school age remain stunted. Stunting is responsible for 14.5% of deaths and 12.6% of the total disease burden among under-five children, most of which is concentrated in the low- and middle-income countries in Africa and South Asia. Through pathways such as reduced schooling (‑1.6 years on average), reduced height in adulthood (‑6 cm on average), and lower cognitive skills (‑0.6 standard deviations on a typical test), the impact of chronic nutritional deprivation during childhood on lifetime income can be very high. For South Asia and Africa, the income penalty of stunting has been estimated to be as large as 9-10% of GDP per capita.
In Pakistan, 38% children under 5-years of age are stunted, with marked socioeconomic inequality in its distribution (Pakistan DHS, 2017-18).
Poor caregivers do not receive frequent and timely feedback on their child’s growth trajectory, which constrains their understanding of the relationship between different childcare inputs and growth. This is especially important for children living in pockets of poverty where the majority of children may be short for their age, i.e., the reference child that caregivers have (children from within their community) may also be stunted. Consequently, despite access to resources like information on best practices through public health information campaigns and nutrition counseling, caregivers typically do not translate this information into action by adjusting their inputs. Visits to the doctor are infrequent and irregular since clinics are often far away and ‘cost’ caregivers time and money. According to the Multiple Indicator Cluster Survey 2014 in Pakistan, only 4% of households in the lowest wealth quintile sought care for their children at a clinic (UNICEF & Sindh Bureau of Statistics, 2014).
IDEA
IDEA
We propose high-frequency nutrition counseling and in-home growth monitoring to regularly inform caregivers about their children's growth trajectory and nutritional needs. The counseling and measurement is conducted by community health workers with active assistance from the caregivers. Health workers inform caregivers about the child's growth status and advise them on best practices related to hygiene and nutrition.
We propose high-frequency nutrition counseling and in-home growth monitoring to regularly inform caregivers about their children's growth trajectory and nutritional needs. The counseling and measurement is conducted by community health workers with active assistance from the caregivers. Health workers inform caregivers about the child's growth status and advise them on best practices related to hygiene and nutrition.
Design
Design
NOTE Click on images for more detail.
Growth Chart for boys
Growth Chart for girls
We designed an age-appropriate nutrition counseling manual and an in-home growth monitoring tool (GroMoTo) to track children's height for a period of six-months and counsel caregivers on their children's nutritional needs.
Nutrition Counseling Manual
CHWs conducted monthly home visits to provide information to the caregivers about best practices on infant and young child feeding (IYCF) practices along with other childcare inputs such as optimal WASH practices. The manual for CHW-based nutrition counseling was developed by the research team in consultation with Dr. Salma Badruddin - a nutrition expert based in Karachi. The manual includes modules on exclusive breastfeeding for the first six-months after birth, complementary feeding for children in the age groups of 6 to 8 months, 9 to 11 months, and 12 to 24 months, a module on overall hygiene for the mother and child, and an additional module on severe acute malnutrition. CHWs were trained on each module and were given sample spoons and cups for measurement that they carried with them at each visit.
GroMoTo
The growth monitoring tool (GroMoTo) is motivated by Akram and Mendelsohn 2017. The authors used a simple, visual, paper-and-pencil intervention to enable treatment households to measure diarrhea incidence before and after getting free chlorine tablets. In contrast, the control group simply received free tablets with expert advice on why to use them. This simple intervention dramatically increased chlorine tablet use by 56 percentage points (see: http://bit.ly/2aeuJHm).
GroMoTo is similarly outcome-focused as it requires community health workers to help caregivers measure their children's linear growth, i.e., height, and understand where their children stand in relation to healthy growth norms by recording those measurements on a visually intuitive poster-sized chart installed inside the home. The growth chart and record-keeping are similar to those found in community clinics - except that it is placed in the caregiver’s home to provide a regular visual ‘nudge’ to invest in their children's nutrition.
Theory of Change
Theory of Change
TOC.jpgOur theory of change is based on the premise that caregivers do not receive adequate and frequent information on their children's growth status.
Our intervention requires monthly visits to households by community health workers. At each visit, health workers measure and record the child's anthropometric data. The growth trajectory is discussed in light of healthy growth norms and the caregivers are advised on the steps they can take to ensure their child's proper growth.
Implementation
Implementation
The trial was deployed in an urban slum, Gulshan-e-Sikandarabad, in Karachi, Pakistan. A series of focus groups were moderated by the research team followed by training of health workers on measuring and recording anthropometric data. A baseline survey was conducted in July 2019 and households with at least one child between the age of 3 and 21 months were recruited. The intervention continued for a period of six-months in which CHWs visited households once every month. The endline survey coincided with the Covid-19 pandemic and was administered from September to October 2020.
The trial was deployed in an urban slum, Gulshan-e-Sikandarabad, in Karachi, Pakistan. A series of focus groups were moderated by the research team followed by training of health workers on measuring and recording anthropometric data. A baseline survey was conducted in July 2019 and households with at least one child between the age of 3 and 21 months were recruited. The intervention continued for a period of six-months in which CHWs visited households once every month. The endline survey coincided with the Covid-19 pandemic and was administered from September to October 2020.
Focus Group Discussions
Focus Group Discussions
Multiple focus group discussions were conducted prior to the baseline survey. The aim was to understand current practices in the community regarding meal preparation and caregivers' understanding of children's growth indicators. Another focus group was conducted with caregivers from treatment households during the intervention to take feedback on their interactions with CHWs.
Training of Community Health Workers
Training of Community Health Workers
A team of community health workers was trained on measuring and recording anthropometric data and delivering age-specific nutrition counseling. The health workers (in pairs of two) visited the households and delivered the intervention from August 2019 to February 2020.
GroMoTo in Action
GroMoTo in Action
Karachi, where our study site – Gulshan-e-Sikandarabad – is located, has high estimated stunting, at 48% among the lowest wealth quintile (UNICEF & Sindh Bureau of Statistics, 2014). Our study targets childhood stunting through the following components:
Monitoring: 24-hour recall dietary survey and anthropometric measurement (height/weight)
Counseling: Counseling by health worker on IYCF best-practices
GroMoTo: Using the anthropometric measurement from Monitoring to plot on the poster and discuss level and trend with the caregiver
Cash transfer (labeled): Health worker hands over cash (PKR 400 or $11.91 (PPP) at each monthly visit) to the caregiver in an envelope with verbal/written suggestion on using it for better foods for the child. The amount is approximately 5% of monthly consumption expenditure for a household of 7 living on less than $2 a day which can at least buy 1-month’s supply of eggs along with 0.5-month’s supply of milk for one under-5 child.
Combinations of these components formed our three experimental arms.
Treatment 1 (396 HHs): Monitoring + Counseling
Treatment 2 (396 HHs): Monitoring + Counseling + GroMoTo
Treatment 3 (396 HHs): Monitoring + Counseling + GroMoTo + labelled cash-transfer
A control group of 451 HHs was recruited at endline from the same community and matched with the treatment sample to serve the purpose of a control group.
Impact
Impact
We find positive and significant improvement in children's height-for-age z-score (HAZ) and reduction in severe stunting in treatment arms as compared to the matched control group one year after baseline.
We find positive and significant improvement in children's height-for-age z-score (HAZ) and reduction in severe stunting in treatment arms as compared to the matched control group one year after baseline.
Figure 1: Aggregated Treatment Effect
Figure 1 is a coefficient plot that shows the combined treatment effect on outcome variables. The comparison group is the matched control group that did not receive any treatment. Our intervention has resulted in an increase of 0.28 SD in height-for-age z-score (HAZ) and a 5 percentage point reduction in severe stunting at endline.
Figure 2: Disaggregated Treatment Effect
Figure 2 is a coefficient plot that shows the disaggregated treatment effect on outcome variables. The comparison group is the matched control that did not receive any treatment. We see a gain of 0.42 and 0.23 SD in height-for-age z-score (HAZ) in T1 and T2. Severe stunting has also reduced in these arms by 6 and 5 percentage points respectively.
Qualitative Evidence
Qualitative Evidence
When we see it [Growth Chart] we can find out if our child is healthy or not. Green [zone] means the child’s height is fine according to his age and red [zone] means that the child’s height is not fine. The [pictures of] food is in different boxes which means that we have to feed different types of food to our children. We see the chart to see what food we must buy for our kids.
--FGD response from a participant.
The height and weight are important. Along with weight, height must also grow for a child to be healthy. Before this program, no one told us about height being important, not even the doctor.
--FGD response from a participant.
Counseling was very helpful. Now we feed potatoes in the form of french fries and give eggs to our children. I give Cerelac and salan roti to my child now. But they still drink [processed] juice [from the kiryana shop]. We have also started giving khichdi to our children. We try to give chicken in small quantities.
--FGD response from a participant.
Our children used to fall sick by eating outside food. Now we give them less outside food. Health worker told us to start giving our children sooji so now we give that as well. Now we wash our hands with soap. The children get sick less often.
--FGD response from a participant.
Research Team
Research Team
Principle Investigators
Abu S. Shonchoy, Florida International University (United States)
Agha Ali Akram, Lahore University of Management Sciences (Pakistan)
Akib Khan, Uppsala University (Sweden)
Hina Khalid, Information Technology University (Pakistan)
Takashi Kurosaki, Hitotsubashi University (Japan)
Research Support
Sidra Mazhar, Center for Economic Research in Pakistan
Mahrukh Khan, Center for Economic Research in Pakistan
Funding Partners
Funding Partners
Research Support
Research Support
Page updated
Google Sites
Report abuse