Hey! We are high school students on a mission to improve the accessibility of micronutrient (MN) screening in global health through low-cost point of care biosensors.
Where we left off…
From our last newsletter, we were looking into cheap point-of-care (POC) biosensors to diagnose zinc deficiency. However, it was way too early to make a decision, and we branched out again to look at the gaps in population MN assessment. We chatted with 8 experts from WHO, CDC, USAID, UNICEF, Micronutrient Forum, the Micronutrient Forum and the Food Fortification Initiative who had expertise on large scale data collection and analysis. We wanted to thoroughly understand:
The magnitude of importance of gathering population assessment of MNs in low-middle income countries (LMICs)
The status quo on the lack of data/recent data
Dig holes into the principle root causes for this lack of data/recent data
Our greatest expert insights!
Data is the most crucial to designing an MN program and maximising its efficiency
Data can demonstrate a need for intervention programs in a region/country. High levels of customization is required since there is no uniform intervention program. Each country has different MNs which are deficient and to different extents. Most countries have had some population MN assessment conducted, but it has only been conducted once.
“Data should be collected every 5 years in lower-resource settings, however for most countries their data is over 10 years old, this need for recent data will only grow as the world changes rapidly.” - Dr. Lisa Rogers, WHO.
The biggest problems with collecting population MN data are the high costs and tediousness
Most LMICs do not have the capacity to analyze blood samples regionally, so they are shipped to labs in the North (usually in Europe). This drives up cost and is extremely tedious.
Cost
We were surprised to find that the primary root cause of high costs are not from the cost of sampling assays. For instance, the VitMin lab in Germany can measure 7 indicators all for $5. (This cost can vary depending on the lab and types of analytes) Depending on country, the high cost typically comes from storing and shipping samples, often involving a cold chain supply, which is difficult to implement in low-resource regions.
“Oftentimes they buy car freezers because there’s no electricity. They might have to provide gas so the car freezers can run 24h. This is costly. Extra -80ºC freezers are also expensive costing ~$15,000 each” - Dr. Maria Jefferds, CDC.
Additionally, there is barely enough funding to support MN data collection. $1M budgets are given for big nutritional surveys in comparison to $90M budgets that go towards data collection for HIV. This is because no one is immediately affected of MN deficiency, but people are directly affected of HIV, thus it is “seen as a greater concern.”
“Funding wise, it’s almost 100% off. It’s never readily available, and there must always be fundraising.” - Dr. Annette Imohe, UNICEF
Tediousness
It could take up to 5 years from initial proposal to get population data. Only 50 samples can be taken a day since it takes ~6 minutes per blood draw. Additionally, since there are only 2-3 labs globally that do MN assessment, there is often a backlog of analyte detection due to the massive demand. To make matters more complex, some countries have bias towards certain labs and some labs are only capable of analyzing certain indicators. Fundraising is also required, and it can take ~6months to 2 years to actually have any data collection rolled out due to the time needed.
“It would take 60 days to collect 3,000 samples alone.” - Dr. Omar Dary, USAID/WHO
It’s not about testing everyone, it’s about choosing the best sample representative
Depending on the region, collecting 300 samples per a population size of the average US state gives enough usable data to implement an intervention program. In contrast, collecting census data shows 20-30% error rates, due to the loss in quality of measuring such a massive number of samples, deeming the collected data unusable.
“A large number isn’t always the goal, but good representative samples are” - Becky Tsang, Food Fortification Initiative
Our biosensor idea is validated!
We wanted to understand MN population assessment better to make sure that we targeted the root causes of the lack of MN data/recent data in LMICs. We are thrilled to say we got the green light on this POC low-cost biosensor from every meeting, and we can finally start to work towards making this real!
This is still the initial concept of using a biosensor for MN population assessment. Ultimately we want to make this a multiplex assay, but we are picking one MN as a start. Our biggest challenge will be accurately measuring MN concentrations in a drop of blood, since there’s a 25% error rate drop blood versus venous blood. We also have to consider the inflammation indicators associated with each micronutrient measured with our biosensor.
We received plenty of input from experts, but we are going to continue our independent research and reach out to more people with further questions.
Our progress and growth
The value of cold outreaching
Assumptions will always stay assumptions until proven otherwise. We had originally thought the high costs associated with lab analysis were of the cost of assays. And had we not asked experts about their experience, we never would have found out about the shipping or storage costs. We also did not consider time or tediousness as a factor contributing to data inaccessibility. Every stat we have quoted here is from talking to experts, we could not have found from papers.
Advice, Resources and Connecting
If you would like to further discuss some of our topics or ideas, feel free to schedule a meeting with us or shoot us an email! Everything we mentioned is a more simplified version of the bigger picture. We would love to hear your thoughts.