Colin Rogerson, M.D., MPH, explains the importance of strengthening maternal-child record linkage to better understand how a mother’s health impacts a child throughout life.
Transcript:
We know that the health of a mother and the medications a mother takes and the chronic illnesses a mother has or acute illnesses has an impact on the baby. It’s difficult, though, to study that relationship because sometimes the health of the mother will affect the baby right at birth, but sometimes it won’t affect the baby for years afterwards. And it’s hard for researchers to have all the information that they need to see what happens to a child over their lifespan.
Probabilistic record linkage is looking for different records for the same person, and what we did, which is a little bit different, is we tried to look at different records for a mother and a child and determine if those records are a mother-child linkage. That relationship isn’t very well captured in our electronic medical records. Some new records do have a linkage that’s established at birth, and so they keep track of that, but many don’t.
And so what we tried to do is use that probabilistic record linkage algorithm approach to try and determine if two records that don’t have that already set link, if we can determine statistically if they are a mother-child link record. The ultimate goal is if we can establish that maternal-child link using just demographic data in an electronic medical record, then we can help this type of maternal-child research go forward.
