COVID-19 Vaccination in Pregnancy in Massachusetts: What Broad Race Categories Hide

This page summarizes a report by the Massachusetts Department of Public Health and CDC on births from May 2021 to October 2022.

Pregnancy raises the risk of severe COVID-19 illness and death, yet pregnant people have had lower COVID-19 vaccination coverage than others of reproductive age, and national data show lower coverage and higher COVID-19 rates among Hispanic and Black pregnant people. But those data usually come in six broad federal categories (American Indian or Alaska Native, Asian, Black, Native Hawaiian or Other Pacific Islander, White, and Hispanic) that can hide big differences within each group. Massachusetts looked closer, breaking the data into 12 racial and 34 ethnic groups.

How the study worked

The state linked COVID-19 vaccination records from the Massachusetts Immunization Information System to birth certificates for residents who gave birth between May 1, 2021, and October 31, 2022, matching on combinations of name, date of birth and address. Coverage meant at least one dose of any COVID-19 vaccine, and vaccination before or during pregnancy was counted separately from vaccination after delivery. Race and ethnicity came from the pregnant person's own report on the birth certificate, where people could choose or write in every race, ethnicity and tribe they identify with, so someone identifying as both Black and Asian counts in both groups.

Results

Of 102,275 people who gave birth, 41.6% had received at least one dose before or during pregnancy. Coverage rose from 22.6% in May 2021 to 50.6% in April 2022, then dipped to 45.7% by October 2022. Counting vaccination at any time, before, during or after pregnancy, coverage was 56.7%, and it may have been as high as 73.1% after allowing for records that could not be linked.

Every broad group hid wide differences. For example, coverage was 39.7% among people identifying as Asian and Laotian but 50.0% among those identifying as Black and Asian Indian.

What it means

Asian pregnant people had the highest coverage and Hispanic pregnant people the lowest, consistent with earlier national findings. But the within-group spread shows why local, disaggregated data matter: a similar analysis of COVID-19 deaths in Hawaii found large differences among Native Hawaiian, Pacific Islander and Asian subgroups, and stressed highlighting community concerns without further stigmatizing underserved groups. Disaggregated data can help states and communities tailor their response, with culturally relevant messages in people's preferred languages and trusted messengers to address vaccine hesitancy.

The authors emphasize that racial and ethnic categories are social constructs, not biological ones. Health inequities persist in part because of structural and institutional racism, which creates barriers to care, medical mistrust and marginalization, and the lower coverage among American Indian or Alaska Native, Black and Hispanic pregnant people can be seen in that context. The pandemic could worsen existing inequities in maternal illness and death.

Limitations

• Only pregnancies ending in a live birth were included.
• Exact-match record linking probably undercounts vaccination, possibly more so for non-English names.
• Vaccinations given out of state or by federal agencies not required to report are missing.
• The imputed upper estimate may be biased if unlinked records are not random.
• Race and ethnicity categories overlap, so groups cannot be compared directly; counting people in every group they identify with was essential to respect self-identification.
• Some subgroups were small, with unstable estimates, and Massachusetts may not reflect other states.

Sources

• Shephard HM, Manning SE, Nestoridi E, et al. "Inequities in COVID-19 Vaccination Coverage Among Pregnant Persons, by Disaggregated Race and Ethnicity — Massachusetts, May 2021–October 2022." MMWR 72(39). https://www.cdc.gov/mmwr/volumes/72/wr/mm7239a2.htm
• Supplementary table: https://stacks.cdc.gov/view/cdc/133102
• The report's abstract gives Hispanic subgroup coverage as ranging from 10.8% to 61.1%, while its results give 20.8% to 61.1%; this page gives only the upper figure. Its charts are not reproduced because the report was led by a state health department.