Babies must survive before they can thrive: Part 1 of 2 posts on disparities in birth outcomes

Media reports of the harrowing pregnancy, childbirth, and postpartum experiences of Beyoncé and Serena Williams focused public attention on longstanding disparities in maternal and infant health.  Despite their personal wealth and access to quality healthcare, neither of these high-profile, highly successful women avoided the serious medical risks associated with pregnancy among African American mothers and their babies.  While Serena, Beyoncé, and their infants survived, not all families are so fortunate.

 The data

In the United States, African American mothers are 3 to 4 times more likely than white mothers to die from pregnancy-related causes, and according to the Centers for Disease Control and Prevention (CDC),  African American infants are more than twice as likely as white infants to die in the first year of life (11.4 vs. 4.9 deaths per 1,000 live births).  In King County, which ranks #2 in the state for overall health outcomes, the mortality rate for African American infants (8.2 deaths per 1,000 live births) is 2.4 times the rate for white infants and 2.8 times the rate for Asian infants (see chart below).

While the national media have focused primarily on Black-white disparities, in King County the infant mortality risk for American Indians/Alaska Natives is at least as great as for African Americans.  And nationally, American Indians/Alaska Natives were the only race/ethnicity group in the country where infant mortality did not decline between 2005 and 2014.

International comparisons

 While racial disparities in birth outcomes are not unique to the United States, something about American culture appears to produce – possibly even magnify – these differences. Comparisons by the World Bank show that:

  • Infant mortality in the U.S. (6 per 1,000 live births in 2017) is higher than in 34 of 35 countries with an advanced economy (the only exception is Malta, which has the same rate as the U.S.); moreover, the U.S. infant mortality rate is double or triple the rates in 24 of 35 advanced-economy countries.
  • Maternal mortality in the U.S. (14 per 100,000 live births in 2015) is higher than in 32 of 35 advanced-economy countries (the only exceptions are Puerto Rico and Latvia); and U.S. maternal mortality is at least double the rates in 20 of these countries. The U.S. is one of only 13 countries in the world where, between 1990 and 2015, the maternal mortality rate increased.
  • These departures from international norms for the developed world are driven by the substantial race differences in birth outcomes in the U.S. A comparison of international infant mortality data revealed that the death rate for infants born to non-Hispanic white mothers in the U.S. was the same as the average rate for infants born to mothers in countries like Canada and the Slovak Republic (about 5 deaths per 1,000 live births), while mortality for babies born to U.S. African American mothers (12 deaths per 1,000 live births) was the same as in countries like Barbados, Brazil, and Peru.

Perhaps the most striking finding is that mothers born outside the U.S. have better birth outcomes than those born here. This is especially true for Black mothers. In a landmark study using more than 5 million linked birth-to-infant death records, babies born to Black women who had themselves been born in the U.S. experienced higher risks of infant mortality (+33%), low-birthweight (+61%), and preterm birth (+48%), compared to the babies of foreign-born Black mothers.

 What underlies the differences?

Both nationally and locally, the “usual suspects” (education, poverty, age, marital status, access to healthcare) do not adequately explain the magnitude of these disparities.  A recent review concluded, “factors that generally are considered to be protective for pregnant women do not provide the same benefits for black women” and, conversely, “conventional risk factors tend to have a more negative effect on black infant outcomes.”

For example, researchers at the Brookings Institute and Duke University and the Insight Center for Community Economic Development have reported the results of a national study in which infant mortality was higher among African American women with advanced degrees than among white women who didn’t finish high school (see below).  Similarly, while Washington’s Department of Health reported that, “in general, infant mortality is higher among mothers who completed fewer years of formal education,” this was not true for African Americans, “for whom the rate does not change as educational attainment increases.”

Note: In the chart above, IMR = Infant Mortality Rate.  Chart originally published in “Fighting at Birth: Eradicating the Black-white Infant Mortality Gap,” March 2018.

According to Washington’s 2017 Infant Mortality Reduction Report, although the risk for infant mortality is generally greater for unmarried women, this difference is only significant among whites.  As with education, marriage offers no measurable protection for the babies of married African American women, whose risk of dying is at least as high as that of babies born to unmarried white women (8.4 vs. 6.1 deaths per 1,000 in Washington state).  Similarly, for white mothers the risk of infant mortality is highest in the teen years and after age 40, but for African American mothers the risk remains high (>10 deaths per 1,000 live births) throughout their childbearing years.

Full circle

These findings lead back to the paradox of why, even with significant advances in obstetric and perinatal medicine, longstanding racial disparities in birth outcomes have not diminished. Part II of this series will explore growing evidence for the idea that racism in the United States – not just overt discrimination, but the day-to-day experiences of growing up as a female of color in this country – can exact a cumulative physiological toll that for many remains undetected until motherhood.


For more information on infant mortality in King County, see Communities Count.  Public Health-Seattle and King County’s Community Health Indicators website presents data on other indicators showing racial disparities related to birth outcomes, including:





Seattle lauded for leading edge transportation policies

The first installment of The Atlantic’s new CityLab Insights series hails Seattle as “the most successful transit city in the country.” Leading with the news that Seattle became home to America’s 2nd largest fleet of dockless bicycles after replacing 500 bike-sharing stations with close to 9,000 of the free-range variety, the article points to recent success in 3 areas:

  • Increasing the supply of transit by: (a) approving $53.8 billion to double our region’s light rail system; and (b) adding 700,000 rides to the city’s bus network.
  • Reforming parking and land/road use policies to release developers from the obligation to build off-street parking in new, densely populated “urban villages” that have frequent transit service.
  • Breaking new ground in “transportation demand management” (TMD), which works with local employers “to manage their supply of parking and other benefits, and to shape demand through incentives, rewards, and games.”

To accommodate increased demand for safe and convenient bike-parking and support the city’s goal of quadrupling bicycle ridership by 2030, the Seattle City Council has significantly upgraded bicycle parking requirements.  Highlights of the new legislation include:

  • Increasing the amount of required bicycle parking.
  • Requiring office buildings with more than 100,000 square feet to provide shower facilities for both genders (shower facilities are exempted from new buildings’ size limits).
  • Allowing developers to trade 1 car stall for 2 bicycle parking spaces, allowing removal of up to 20% of required car parking.
  • Requiring access to bike parking without the use of stairs.
  • Requiring that bike rooms accommodate family, cargo, and electric bikes.
  • Adding a bike valet provision for major even venues.

To look at commute trends by mode of transportation in King County, go to Communities Count’s Trend by Mode of Transportation,  click “Mode of Transport Trends,” click the mode of interest (drove alone, carpooled, public transportation, walked, biked, or worked at home), then click the cities for which you would like to see trends.  Not surprisingly, densely populated Seattle has shown the greatest increases in biking and walking to work.  Click here to see the effect of light rail on commuter choices in Tukwila.

(Although it’s flattering to be called out as a model for other cities, we have a long way to go before we come close to the “deserted freeway” image depicted by CityLab Insights.)

Learning from data on homeless students

“Homelessness up again in King County” – the recurring headline, steady as a drumbeat, reminds us of the paradox of our region’s economic prosperity:  A flourishing job market increases competition for housing and squeezes out lower- and middle-income households.  Data informing most policy decisions about regional homelessness come almost exclusively from two sources – COUNT US IN (also known as the Point-in-Time or One-Night Count), and the Seattle/King County Homeless Management Information System (HMIS), a secure online database of information about services provided to people experiencing homelessness.

A third dataset –  HOMELESS STUDENTS DATA (an annual report prepared by the Washington Office of the Superintendent of Public Instruction) – offers a wealth of information about family homelessness, but so far has not been used to guide policy for prevention or mitigation of homelessness.

Although COUNT US IN and HOMELESS STUDENT DATA are updated annually, overlap in the individuals they count is limited by differences in method and definition:


  • Each January, COUNT US IN sends out teams of volunteers to provide a one-time “snapshot” of the number of people of all ages who are experiencing homelessness. This includes counts of sheltered and unsheltered individuals plus an in-person survey of a subset of these individuals. The method is inherently conservative, and its report acknowledges undercounting homeless individuals in suburban and rural communities and those in hard-to-reach subpopulations such as unsheltered families and unaccompanied youth.
  • HOMELESS STUDENTS DATA are collected throughout the school year by teachers and other school personnel and provide a count of students (preschool through grade 12) who were known to be homeless at any time during the academic year.


  • While both data sources count as homeless people living unsheltered or sheltered (in emergency shelters, transitional housing, or safe havens), COUNT US IN, following criteria specified by the U.S. Department of Housing and Urban Development (HUD), excludes homeless individuals and families who are “doubled up” with friends, family, or others in homes, hotels/motels, or other arrangements.
  • In contrast, following specifications from the U.S. Department of Education and the Washington State Legislature, HOMELESS STUDENTS DATA includes “doubling up” in their definition of homelessness: about 2/3 of students without a stable home in the most recent (2016-17) count were doubled up.

Differences aside, both counts lead to the same sorry conclusion:  As the fortunes of some King County residents are tracking the region’s recovery from The Great Recession, increasing numbers of our neighbors are becoming homeless.  According to COUNT US IN 2018, 12,112 individuals were homelessness in Seattle/King County on January 26th, up by 4% from 2017 (due to changes in methodology in 2017, comparisons with pre-2017 counts are discouraged).  Similarly, HOMELESS STUDENTS DATA reported that 8,938 King County public school students experienced homelessness during the 2016-17 school year, up by 8% from the previous year and more than double the number in 2010-11. The numbers of unsheltered individuals have also gone up in the most recent counts, by 15% for COUNT US IN – from 5,485 (2017) to 6,320 (2018) – and by 39% for HOMELESS STUDENTS DATA – from 244 (2015-16 SY) to 339 (2016-17SY).

What can we learn by looking at these datasets together?  Striking similarities emerge when we look at subgroups in these datasets. (Complementing the data from HOMELESS STUDENTS DATA, the Seattle Atlas of Student Homelessness, an in-depth analysis of data from Seattle Public Schools by the Institute for Children, Poverty & Homelessness (ICPH) offers examples of how housing instability compounds existing disparities for outcomes such as academic achievement and school suspensions.)

  • RACIAL DISPARITIES: Blacks/African Americans and Hispanics/Latinos are disproportionately represented in both homelessness counts.
  • DISABILITIES: More than half of the COUNT US IN survey respondents said they were living with at least one disabling condition, and 21% of homeless students were in Special Education (about double the rate for students who were not homeless).
  • LIMITED ENGLISH PROFICIENCY: The COUNT US IN survey found that respondents from families with children were 6 times more likely than those without children to encounter language barriers when trying to access services. HOMELESS STUDENTS DATA reported that 22% of homeless students were “English Language Learners” (compared to 11% of students overall).
  • PRIOR HOMELESSNESS: COUNT US IN 2018 reported a 1-year increase of 779 individuals (28%) experiencing chronic homelessness; more than 1 in 5 respondents to the COUNT US IN survey first experienced homelessness when they were children, and almost half had experienced homelessness before age 25. While HOMELESS STUDENTS DATA does not report on chronic homelessness, the Seattle Atlas of Student Homelessness, found that more than half of Seattle Public School students who were homeless in the 2015-16 school year had also been homeless in previous years.
  • DOUBLING UP: More than 1 in 4 respondents to the COUNT US IN survey reported that immediately before becoming homeless they were “doubled up” (living in a home owned or rented by relatives or friends). This suggests that by counting students who are doubled up, HOMELESS STUDENTS DATA identifies students at risk of ending up on the street, in a shelter, or in transitional housing. Because repeated episodes of homelessness are common (see PRIOR HOMELESSNESS above), paying attention to the doubled-up population could eventually help reduce chronic homelessness. Currently, as noted by the ICPH report on Seattle student homelessness, “doubled-up students are not eligible for many of the same housing resources as other homeless students.”

Finally, while COUNT US IN provides homeless counts for 6 broad regions of King County, HOMELESS STUDENTS DATA offers almost a decade of data for 18 school districts across the region, revealing different patterns over time in different communities.  While the rate of homelessness has leveled off in some school districts, it continues to climb steadily in others.  In Tukwila, for example, student homelessness surged from 47 students (1.6% of enrollment) in 2007-08 to 375 (12.7%) in 2016-17 – an 8-fold increase in less than 10 years.  In the same period, the number of Seattle public school students without a stable home grew from 930 (2.0% of enrollment) to 4,280 (7.9%), while a few districts (Mercer Island, Vashon Island, and Skykomish) never had a year in which more than 20 students experienced homelessness.


Broadening our homelessness policy perspective to include individuals and families who are doubled up could help us identify families at risk for homelessness before they have exhausted their last personal resource (the family and friends willing to take them in).  Chicago is already looking at doubled-up families by, for the first time, linking information from their official database of homeless individuals accessing services (HMIS) with data from the public schools.  By combining data sources, they are able to better understand families’ paths to homelessness and to project future needs for services.  Following a similar course in King County could enable us to come up with a more prevention-oriented approach to what has become a chronic problem in our communities.

NOTE: Communities Count has reported on student homelessness for several years, and is about to update that indicator with 2016-17 data. To coordinate with the newly released COUNT US IN report, this blog previews key findings from that update.

Troubling trends for homelessness, physical activity, food security, and drug-related deaths

Part I of our double blog on recent trends in King County focused on some of the good news from King County Hospitals for a Healthier Community’s new Community Health Needs Assessment.  Now we’ll look at some of the more challenging findings.

Part II.  ACROSS KING COUNTY OVERALL, WHAT’S FAILING TO IMPROVE OR GETTING WORSE?  While many indicators showed little or no improvement in the 3 years since the previous report, those with special relevance for healthcare providers were highlighted in the Community Health Needs Assessment.

Homelessness:  In the context of escalating housing prices, student homelessness in King County more than doubled since 2008, reaching 8,411 (2.9% of enrolled students) in the 2015-16 school year. In 2016-17, this increased further, to 9,407 (3.2% of enrollees).

  • In most school districts, more than 40% of homeless students were in elementary school or pre-kindergarten.
  • In addition to student homelessness, the 2017 Count Us In Report identified 11,643 individuals experiencing homelessness, 50% of whom had one or more disabling conditions.

Too little physical activity: Fewer than 1 in 4 King County adults and youth get the recommended amount of exercise. Insufficient physical activity is associated with obesity, which in turn is linked to diabetes and other chronic diseases (including 4 in 10 cancers diagnosed in the United States).

  • This represents no change for adults, and modest but inadequate improvement for 8th, 10th, and 12th graders, given the importance of physical activity to health.
  • The overall obesity rate for King County adults has been flat since 2009 (at more than 1 in 5 adults). Nationally, adult obesity levels rose for decades, stabilized between 2003 and 2012, then rose again slightly for women.
  • At 22%, the 2015 adult obesity rate in King County was significantly lower than the Washington state rate of 26%, and the national rate of 29% (although the 2011-2015 rate in South Region matches the national rate, at 29%).
  • For King County youth, obesity has held steady around 9% since 2004 except in South Region, where it has increased. In comparison, high school students nationally experienced a steady increase in obesity from 1999 to 2013, which appeared to level off at a higher rate -14% in 2015.

Food insecurity:  By 2016, King County participation in the Basic Food program (formerly food stamps) had not returned to pre-recession levels and was increasing for older adults, especially in South Region.

  • A similar pattern was found for visits to King County food banks.
  • Although survey data about food hardship (running out of food without money to buy more) have not been collected since 2013, use of food assistance is often associated with food insecurity.

Feeling depressed: On the mental health front, 30% of youth reported that, every day for 2 or more consecutive weeks, they felt so sad or hopeless that they stopped doing some of their usual activities.

  • This indicator of depressive feelings has gotten worse in King County since 2004, driven by increases among youth in South Region.
  • Among adults, reports of frequent mental distress (poor mental health in at least 14 of the past 30 days) have increased among Hispanics since 2005, but remained stable among other race/ethnicity groups.
  • While the percentage of heterosexual adults reporting serious psychological distress has been flat since 2009, feeling “nervous, hopeless, restless, depressed, worthless, or that everything was an effort” within the past 30 days has increased significantly among lesbian, gay, and bisexual adults.

Drug-related deaths, especially those related to heroin and methamphetamine, increased dramatically between 2010 and 2016.  New data released last week revealed a 10% increase (from 348 to 379) in King County drug-related deaths in 2017.

What’s UP in King County?

In the midst of rapid economic growth, population growth, and increases in the cost of housing, how are the people who live in King County communities faring?  The new Community Health Needs Assessment released by King County Hospitals for a Healthier Community (HHC) offers a data-informed overview of how we’re doing.    

In a pair of blog posts, we look at what’s changing in our increasingly diverse county – in positive and negative directions.  First, the good news:


Although disparities remain, 3 county-wide successes stand out. (To look at change over time in the hyperlinked visualizations below, please click on the “trends” tabs.)

Since implementation of the Affordable Care Act, health insurance coverage has improved dramatically – for all ages, racial/ethnic groups, and cities.

  • But… most communities of color remain disproportionately uninsured and residents of low-income neighborhoods were least likely to receive recommended preventive services such as vaccines, screenings, and dental care.

Cigarette smoking – still the leading preventable cause of death in the United States – has declined for adults and youth across regions, age groups, and racial/ethnic groups.  For youth, the decline in smoking was accompanied by a decline in overall tobacco use (combining cigarettes, chew, cigars, and hookah), exposure to secondary smoke at home, and use of alcohol, marijuana, painkillers, or illicit drugs.

  • But… smoking is inversely linked to income, with adults with in the lowest-income group 4 times more likely to smoke as those in the highest-income group.

Fewer students in 8th, 10th, and 12th grades are drinking non-diet sodas and other sugar-sweetened beverages daily, mirroring a national trend among high school students.

  • But… students in South Region are more likely than students in other regions to drink sugar-sweetened beverages.

All of these improvements occurred in the context of supportive policy changes – at the federal, state, county, city, and/or school levels.  Nevertheless, some improvements are swamped by the magnitude of ongoing disparities. For example, although health insurance for Hispanic adults improved by 11% between 2013 and 2016, 21% of Hispanic adults were still uninsured in 2016, compared to 4% of non-Hispanic white adults.

While we’re making progress, we still have a ways to go.