On 2017 Oct 06, Susan Altfeld commented:

We thank Dr. Mages for his interest in our work; however, we do not agree with his assessment. While it would have been more precise to talk about “non-Hispanic Black mothers” and “Hispanic mothers”, the authors of the paper we were citing (Colson, Willinger, Rybin et al., 2013) themselves refer to “non-Hispanic Black infants” and “Hispanic infants” in their study (most likely adhering to the NCHS standard of attributing mother’s race-ethnicity to the infant.) We are aware of the difference between race and ethnicity, but are not sure what point Mages was trying to make, as combining the two in demographic categories (e.g. “non-Hispanic black”) is standard practice in reporting epidemiological data (hence the common use of the term “race/ethnicity”.) The focus of our commentary was not racial/ethnic differences in SUID rates, but on the need for a more nuanced, harm reduction approach to safe sleep messaging. Many caregivers, of diverse backgrounds, have failed to adhere to abstinence messages regarding infant sleep and new approaches are needed to impact caregiver knowledge and behavior.

On 2017 Oct 01, David Mage commented:

Altfeld et al., PMID: 28838726, made an excellent study of the recent efforts to reduce sleep-related infant deaths in the U.S. However, they made both an explicit error and an implicit error that needs to be called to the readers’ attention:

Explicit Error: On page 2, they refer to “mothers of non-Hispanic Black infants” and “mothers of Hispanic infants.” [NB: Hispanic is an ethnicity and not a race] Apparently, the authors did not read carefully their references [1, 2, 21] and the Technical Notes cited therein. The CDC and NCHS both explicitly state that their cited racial data are either the mother’s self-identified race if monoracial, or “To provide uniformity and comparability of these data [to census data, per OMB requirement], multiple race [of mother] is imputed to a single race.” (see NCHS Technical Notes). The authors of reference 21 conducted an evening at-the-door survey of parents of recently born infants, and “Participants were asked: ‘Which of the following best describes [your, or] the mother's race or ethnic background?’ They were then read a list but also given the option to name one that was not listed.” The mother answered 84% of the time and 16% of the time another adult person responded for her;

Implicit Error: The authors seem to have forgotten that the father of an infant often plays a role in the determination of the race of the infant. I used the word “often” instead of always because “Tis a wise child that knows its own father.” [Puddn’head Wilson, Mark Twain] {Old Burlesque joke: Enter little redhaired boy. Clown: What lovely red hair you have. Do mommy and daddy have red hair too? Little Boy: No --- But we do have a milkman with red hair. (audience laughs) Exit boy.} Therefore, implicitly, the self-identified race of the mother is the same race as the infant if, and only if, the mother is monoracial, and the father (either known or unknown) is the identical race as the mother.

On 2017 Oct 01, David Mage commented:

Altfeld et al., PMID: 28838726, made an excellent study of the recent efforts to reduce sleep-related infant deaths in the U.S. However, they made both an explicit error and an implicit error that needs to be called to the readers’ attention:

Explicit Error: On page 2, they refer to “mothers of non-Hispanic Black infants” and “mothers of Hispanic infants.” [NB: Hispanic is an ethnicity and not a race] Apparently, the authors did not read carefully their references [1, 2, 21] and the Technical Notes cited therein. The CDC and NCHS both explicitly state that their cited racial data are either the mother’s self-identified race if monoracial, or “To provide uniformity and comparability of these data [to census data, per OMB requirement], multiple race [of mother] is imputed to a single race.” (see NCHS Technical Notes). The authors of reference 21 conducted an evening at-the-door survey of parents of recently born infants, and “Participants were asked: ‘Which of the following best describes [your, or] the mother's race or ethnic background?’ They were then read a list but also given the option to name one that was not listed.” The mother answered 84% of the time and 16% of the time another adult person responded for her;

Implicit Error: The authors seem to have forgotten that the father of an infant often plays a role in the determination of the race of the infant. I used the word “often” instead of always because “Tis a wise child that knows its own father.” [Puddn’head Wilson, Mark Twain] {Old Burlesque joke: Enter little redhaired boy. Clown: What lovely red hair you have. Do mommy and daddy have red hair too? Little Boy: No --- But we do have a milkman with red hair. (audience laughs) Exit boy.} Therefore, implicitly, the self-identified race of the mother is the same race as the infant if, and only if, the mother is monoracial, and the father (either known or unknown) is the identical race as the mother.

On 2017 Oct 06, Susan Altfeld commented:

We thank Dr. Mages for his interest in our work; however, we do not agree with his assessment. While it would have been more precise to talk about “non-Hispanic Black mothers” and “Hispanic mothers”, the authors of the paper we were citing (Colson, Willinger, Rybin et al., 2013) themselves refer to “non-Hispanic Black infants” and “Hispanic infants” in their study (most likely adhering to the NCHS standard of attributing mother’s race-ethnicity to the infant.) We are aware of the difference between race and ethnicity, but are not sure what point Mages was trying to make, as combining the two in demographic categories (e.g. “non-Hispanic black”) is standard practice in reporting epidemiological data (hence the common use of the term “race/ethnicity”.) The focus of our commentary was not racial/ethnic differences in SUID rates, but on the need for a more nuanced, harm reduction approach to safe sleep messaging. Many caregivers, of diverse backgrounds, have failed to adhere to abstinence messages regarding infant sleep and new approaches are needed to impact caregiver knowledge and behavior.