Emails, Brian Tsai, public affairs specialist, Kimberly Rodriguez, intern, Office of Public Affairs, National Center for Health Statistics, Centers for Disease Control and Prevention, June 27-29, 2017
9:08 a.m.
June 27, 2017
The CDC’s National Center for Health Statistics (NCHS) does not have state-by-state data of maternal mortality.
The latest official estimates for maternal mortality that CDC/NCHS has are from 2007. This is because of the uneven implementation by the states of the 2003 standard certificate. We determined that the national numbers, particularly if they were tracked over time, would be misleading. One can actually calculate maternal mortality rates by state using our mortality database called CDC WONDER. The data are still available, but not in tabular form.
If you wanted to figure out the ICD 10 (code of death) codes for maternal mortality, they are A34, O00-O95, O98-O99. O96-O97 are late maternal causes and are excluded from the classic maternal mortality measure.
You could also reach out to the CDC’s Division of Reproductive Health regarding information on maternal mortality at 404-639-3286 or media@cdc.gov
Brian (Tsai)
From: Selby, Gardner
Sent: Tuesday, June 27, 2017 10:22 AM
Subject: RE: Seeking maternal mortality rates, latest available year, for the 50 states (state by state)
Could you walk me through precisely how to glean rates from WONDER?
G.
10:52 a.m.
I have included images of the WONDER search table results (using the codes that were given to you previously) to this email for you. This is data from 2015 and is ordered by state as you requested. However, you will see that many columns in the chart say "suppressed" or "unreliable." Please look at the information below regarding this:
About suppressed data and unreliable rates:
Sub-national data representing fewer than ten persons (0-9) are suppressed for year 1989 and later years. See Assurance of Confidentiality for more information.
Rates are marked as "unreliable" when the death count is less than 20.
https://wonder.cdc.gov/wonder/help/ucd.html
https://wonder.cdc.gov/wonder/help/ucd.html#Assurance of Confidentiality
In the future if you ever need help with our WONDER Database, you can email the WONDER Help Desk at cwus@cdc.gov.
Regards,
Kim
Kimberly Rodriguez
Centers for Disease Control and Prevention National Center for Health Statistics | Office of Public Affairs
11:18 a.m.
June 28, 2017
My name is Brian Tsai and I'm a Public Affairs Specialist at the CDC's National Center for Health Statistics.
11:22 a.m.
Kim is an intern with NCHS.
-----Original Message-----
From: Selby, Gardner
Sent: Wednesday, June 28, 2017 12:27 PM
Subject: Three fresh queries
What's meant by "age-adjusted?"
Is there a way to provide non-age-adjusted figures or a good reason not to?
Also, these rates from WONDER seem on the very low side. That is, if I read right, the Texas rate in 2015 was 0.5 per 100,000 live-births, way down from what's otherwise reported as a rate exceeding 30.
?
g.
3:12 p.m.
The age-adjusted rate is a summary measure of mortality risk that controls for variation in age distributions. Here is a report that should be helpful - https://www.cdc.gov/nchs/data/nvsr/nvsr47/nvs47_03.pdf
I am currently waiting to hear back from our mortality statisticians to verify whether the CDC WONDER data is correct. Hopefully, I'll hear something soon. What is your deadline?
While we're waiting, I did find a journal article that mentions maternal mortality data in Texas that you may find more useful.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5001799/
Brian
10:09 a.m.
June 29, 2017
Our mortality statistician has told me that the codes are correct with the late maternal deaths excluded from "maternal deaths." However, the maternal mortality rate is calculated per 100,000 live births not per the population. If it had been per population, the column we produced with crude rates are non-age-adjusted figures. However, neither of these columns should be used.
You would need to calculate per births. As for age-adjusting, we don't normally age-adjust the maternal mortality rate. Since the age distribution women give birth at has changed over time.
The big issue is that not all states have the standard question yet, so they can't all be compared and there is no national estimate yet. AL, CA, and WV would be the ones without something comparable in 2015. WV's numbers don't allow you to see it in the table. If we combined maternal and late maternal, that alleviates the difference in CA's question.
Lastly, we are aware of cases in which recent pregnancy is overreported, particularly at older ages. So, there is a quality issue with the data also.
I suggest using the data from this journal article instead.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5001799/
Brian