India Webinar Identifies Strengths and Gaps in Data, Suggests Directions for Future Research
The Ki knowledge base includes some 38 studies from India, including approximately 10 million observations of more than 250,000 subjects. Last month, we hosted a webinar with partners from HBGDki-India to introduce this knowledge base and pave the way for future research.
Our goals for the webinar were three-fold:
- We wanted to be clear about the data we have, including where it’s strong and where there are gaps.
- We wanted to present the preliminary results of our in-progress rally on stunting and wasting in India.
- We wanted to further stimulate the development of the HBGDki-India community, especially in advance of the GC-India announcement (see below).
Vishak Subramoney, a member of Ki’s Data Services team, surveyed the India-relevant knowledge base, highlighting the richness and depth of anthropometric data and key gaps that need filling: studies with ultrasound, brain imaging, and MRI imaging, studies from key states in northern India, including Bihar and Uttar Pradesh, and studies of WASH or nutrition interventions.
Then Andrew Mertens of the UC-Berkeley data science team presented preliminary results of an analysis of stunting and wasting. By looking at incidence instead of prevalence, which we can do because of the size of the knowledge base, Andrew and his team concluded that both stunting and wasting incidence are highest from birth to six months of age, even after excluding stunting and wasting at birth. This finding suggests that we need to focus on preventing and treating both conditions among very young children in India. Andrew also saw extreme seasonal variation in wasting incidence, with up to a standard deviation increase during monsoon season. This finding suggests that we need to understand what’s happening in the summer months that leads to spikes in wasting so that we can intervene. Final results from a more comprehensive analysis will be presented later this year.
In the question and answer period following Andrew’s presentation, participants in the webinar suggested a variety of questions they’d like to explore using the dataset. This kind of collaboration on a research agenda is precisely the point of these webinars, and we are excited about the fact that GC-India has just announced a call for proposals on data science to address these and other questions.
Grand Challenges-India Launches MNCH Data Challenge
On July 3, GC-India, a partnership between the foundation and BIRAC (India’s Biotechnology Industry Research Assistance Council), issued a call for proposals on data science to address health challenges. This follows the Grand Challenges Explorations data challenge for Brazil, launched earlier in the year.
Until early August, GC-India will be accepting proposals for projects that take a Ki approach and use India-specific Ki data, although investigators can also bring in data sets they or their collaborators collected in the past or use data from public repositories.
For more information about the call, please go to this website. It includes a primer about data science for research scientists, a primer about maternal, newborn, and child health for data scientists, a summary of the data sets available for analysis, and detailed information about the types of proposals GC-India is looking for.
We believe the India dataset can be analyzed to yield insights on a wide range of issues that affect vulnerable populations, including but not limited to:
- Finding patterns that link individuals with positive health outcomes despite a high number of risk factors
- Converting correlations into causal hypotheses (e.g., establishing the impact of air pollution on fetal growth)
- Combining data focused on improving child survival with data focused on improving early neurodevelopment
- Determining critical periods for intervention during pregnancy and early childhood
- Pinpointing the relative contributions to health outcomes of diet quantity versus quality
In the short-term, we are optimistic that this challenge will generate specific solutions to specific problems, or knowledge that leads directly to specific solutions. In the long-term, we hope that it helps Indian research and data scientists who haven’t even considered working together before build a community that tackles big problems together.
Interesting Finding About Surgical Checklists Points the Way Toward a New Research Agenda
Atul Gawande’s book The Checklist Manifesto took the world by storm. The fact that something as simple as a surgical checklist could save lives transformed the way people thought about not just medicine but every other facet of life. That’s why it was so troubling when, in a study run by Gawande’s team in the Indian state of Uttar Pradesh, the checklist didn’t work.
In the New York Times Magazine, Siddhartha Mukherjee, another leading health writer (author of Emperor of All Maladies: A Biography of Cancer) dug into what happened in Gawande’s India study and what it means. (Gawande himself wrote an excellent piece in the New Yorker on the same subject).
It turns out that despite an intensive peer-coaching program, not enough health workers adopted the behaviors included in the checklist. For example, only 35 percent of birth attendants who received coaching washed their hands before conducting procedures, even though handwashing was one of the items on the checklist. According to Mukherjee, “features unique to labor and delivery in obstetric hospitals in India may have made checklists ineffective”—including the possibility that the checklist competed with a long-established body of local knowledge for precedence.
The larger point is “the extent to which human behavior remains an uncharted frontier for medicine.” This is one of the reasons we’re excited about Ki’s geographically focused work in Brazil and India: it promises to generate context-specific insights that we can compare and contrast to start developing hypotheses related to human behavior as well as human physiology. Ultimately, it is the blend of all these hypotheses that will save the most lives.
Update on Data Sets:
The following studies have completed QC since the last announcement.
Ki1033518 – iLiNs-DYAD-Ghana
Efficacy of small-quantity lipid-based nutrient supplements (SQ-LNS) consumed by women during pregnancy and the first 6 months postpartum, and by their infants from age 6 to 18 months
Ki1135782 – INCAP (update to the data – changes to birthweights and anthropometrics measurements)
Institute of Nutrition of Central America and Panama: nutrition supplementation longitudinal next generation study
Ki1148112 – iLiNs-DYAD-M (update to the data – minor changes to existing datasets including feeding information)
Supplementing maternal and infant diet with high-energy, micronutrient fortified Lipid-based Nutrient Supplements (LNS)
We have also received the raw data for the CHAIN study and are working to map this data to our common data standard. Building the evidence base for care of acutely ill, undernourished children (2–23 months old) in limited resource settings – cohort study. www.CHAINnetwork.org