Perinatal Disparities: Not Just Black and White ~ Part One

 [Editor's note: Today is part one of a two-part series by Darline Turner-Lee, looking at racial disparities in maternal outcomes.  Come back tomorrow to read Darline's discussion on this recent study.]



INTRO: The Financial Burden and Racial Disparities of the US Health Care System

Despite the enormous amount of money the United States spends annually on health care, nearly 17% of the Gross Domestic Product according to www.HealthCare.gov,  Americans overall are less healthy and have higher morbidity and higher mortality than many other citizens around the world. Nowhere is this more apparent than in US maternal mortality rates. Amnesty International’s Deadly Delivery: The Maternal Health Care Crisis in the USA states,


Maternal mortality ratios have increased from 6.6 deaths per 100,000 live births in 1987 to 13.3 deaths per 100,000 live births in 2006. While some of the recorded increase is due to improved data collection, the fact remains that maternal mortality ratios have risen significantly. African-American women are nearly four times more likely to die of pregnancy-related complications than white women. These rates and disparities have not improved in more than 20 years.


Researchers have tried to determine what is causing this huge health care disparity between African American women and white women. To try to identify the causative factors, most researchers have compared pregnancy outcomes of African American women and white women living in the same or similar neighborhoods and have attributed much of the disparity to socioeconomic status (SES); i.e. poverty, lack of education, poor access to care. Yet SES alone has not been able to account for the large gap in outcomes. In The Neighborhood Contribution to Black-White Perinatal Disparities: An Example From Two North Carolina Counties, 1999-2001 Ashley Schempf and her colleagues at the US Department of Health and Human Services sought,


to determine the total contribution of neighborhood inequalities, both observed and unobserved, to the black-white gap in the birth outcomes, low birth weight (LBW), preterm birth (PTB) and small for gestational age (SGA) using hybrid fixed-effects approach to compare only black and white women who lived in the same neighborhoods”


Schempf et al added in the fixed-effects models to better characterize the impact of neighborhood features such as social cohesion and access to goods and services that may vary depending on the racial make up of the neighborhood. Their aim was to compare the results they found with results from conventionally run studies to get a more complete picture of the effect of neighborhood factors on maternal mortality.


Study Population

The study consisted of 31,489 women: 21,221 white women and 10,268 black women. A validated neighborhood deprivation index was used as an indicator of neighborhood SES. The index was constructed from a principle analysis of 8 variables from the 2000 Census of Population and Housing. Individual maternal control factors included age, years of education, marital status, and gravity. Please refer to the tables at the end of this post from the study for demographic data.


Analytic Methods

The researchers employed a complex series of statistical analyses to the birth certificate data obtained. For each outcome analyzed, they used 4 logistic models to determine 1) The total contribution of neighborhood to the racial disparity and 2) To compare the performance of a neighborhood fixed-effects estimator with the simple control for neighborhood SES (the standard analysis method). Model 1 was an unadjusted logistic that measured the difference of race on maternal outcomes. Model 2 was an adjusted logistic that controlled for individual maternal characteristics. Model 3 was a hybrid fixed-effect model that accounts for all between neighborhood variation in risk that is associated with race and provides a within-neighborhood race contrast. This model controlled for the excess odds of an adverse outcome associated with living in a neighborhood with a higher proportion of black births. Finally, Model 4 controlled for neighborhood SES. (Please refer to the actual study for statistical equations).



The authors noted clear differences in sociodemographic characteristics and neighborhood deprivation. Black women were at least twice as likely to deliver LBW, preterm and SGA infants as white women. When data was adjusted for race and neighborhood deprivation, the disparities were reduced. With all confounding variables controlled for by the fixed effect analyses PTB was the only adverse outcome that continued to have higher adverse outcomes. The disparities in PTB persisted even when parceled out for Moderate preterm birth (MPTB) 32-36 weeks gestation and Very preterm birth (VPTB) <32 weeks gestation. The absolute risk of MPTB was nearly twice that of VPTB while the relative risk of VPTB was much greater than that of MPTB. Based on this data, the authors believe that there are other significant yet unobserved neighborhood contributions to disparities for preterm birth. The results are presented in the tables below.


*Click here to jump to part two of this series.







Posted by:  Darline, Turner-Lee, BS, MHS, PA-C

Maternal Mortality, Maternal Mortality Rate, New Research, Research, Uncategorized , , , , , , ,

  1. avatar
    Walker Karraa, MFA, MA, CD
    November 1st, 2011 at 10:25 | #1

    Thank you Darlene, for bringing such an important issue to our attention. I hope we engage in further conversation about racial disparities. I look forward to reading your work tomorrow.

  2. November 1st, 2011 at 10:39 | #2

    @Walker Karraa, MFA, MA, CD
    Thanks for reading. This was quite eye opening to see. Hopefully we can now start brainstorming ideas on how to narrow the gaps.

  3. avatar
    Walker Karraa, MFA, MA, CD
    November 1st, 2011 at 14:18 | #3

    Are you familiar with Dr. Michael C Lu’s work in this area? I can send you articles off line. @Darline Turner-Lee

  4. avatar
    November 1st, 2011 at 14:39 | #4

    Darlene, This is a bit of a shot in the dark, but it has been on my mind for some time… Could this be induction-related? Related to mis-dating pregnancies? Etc.?

    I can’t see the full study, nor Table 1, so I am not sure if this was addressed, but given the historical and continuing issues with provider bias, I can’t help but wonder what part of this is iatrogenic, directly or indirectly. I am just a layperson with a lot of personal interest, and can’t connect the dots on my own, but I keep coming back to a discussion with my friend, who for some time worked with young women (teens to early twenties) who had been exploited in the sex industry. (In more of a life/counseling capacity, not as a HCP.)

    We were discussing birth generally (I am planning a homebirth) and she remarked that literally every single one of the young women she had worked with had been diagnosed with Pre-E when (if) pregnant. This was about 15-20 young women in a short period. Now, I hesitate to claim it is impossible that they all had Pre-E. We are talking about a group with limited resources, coming from a poor urban environment (NYC), though it was always my understanding that the exact causes of Pre-E are unknown.

    I suspect (just tentatively so far) that these women– nearly all young, black and poor– are being diagnosed with Pre-E almost as a matter of course, as part of their being pushed through a big urban underfunded hospital machine. Just as heavier women are assumed to have GD, small women assumed to have problems with CPD, etc. Could it be racial bias, direct or indirect? Could it be racism in combination with heavier women (as many of all races are) needing larger BP cuffs to get accurate readings, and not getting them? Could young, poor, black women be more likely to get white coat hypertension, for obvious reasons?

    Just a few questions that come up for me.

    Pre-E often calls for induction, including pre-term induction, no? Even when necessary, induction is a significant cause of PTB, and in conjunction with misdating, could be a big contributor to iatrogenic LBW. Then that makes me wonder if black women aren’t more likely to have their dates determined and redetermined solely by HCPs, for reasons of direct and indirect bias (belief in the patient’s own dating accuracy, etc.)

    And on and on…

    Even if true that induction rates are higher and inductions earlier among black women, I wouldn’t suggest that Pre-E diagnoses are necessarily the cause of all this, but I would wonder if there weren’t several causes with similar roots. And even if induction rates and types, etc., are no different, I still admit I strongly suspect differential treatment to be at the root of all or most of this issue.

  5. November 2nd, 2011 at 12:07 | #5

    @Walker Karraa, I would love references to Dr. Liu’s work. Please send to Darline@mamasonbedrest.com.

    Dreamy, the very last sentence of your post pretty much sums it up. Researchers have tried to look at the issue of health care delivery disparities from every which angle, yet few have looked at how provider demeanor and patient/provider interaction actually influence care. It’s a touchy area to say the least, yet one I think we’re going to have to tackle if we’re going to fully address the disparities in health care delivery.

    The US has a long and complicated racial history. No one wants to admit that they have any sort of bias, but the truth of the matter is that we all have biases. But how these biases and/or beliefs play out in health care delivery is the question. I don’t know of any studies looking specifically at how provider demeanor and/or patient/provider interaction impacts care. If anyone has some specific references, please share. I know that this is an area receiving more attention, but it tends to flare emotions so much that not a lot of headway has been made. It will be interesting to see what others have in the way of references and comments. Hold on and let’s see what comes up.

  6. avatar
    November 3rd, 2011 at 11:17 | #6

    Well, Darlene, it feels like Occam’s Razor to me, but yes, of course, “people are touchy.” *sigh*

    It would be difficult to measure “demeanor” per se, of course, but just to look at intervention rates, timing, etc., could tell us a lot in a “circumstantial evidence” sort of way.

  7. November 3rd, 2011 at 11:29 | #7

    I adjusted the text to refer to the two tables included in this post. Darline refers to those which you can see (sort of!) not to any others that were left out.
    Hope that helps.

  8. November 6th, 2011 at 16:17 | #8

    Hi there – There is alot of work being done at UCLA by Dr. Christine Dunkell-Schetter and her colleagues regarding the role of stress and preterm labor….they have found a link in valid & reliable & replicable studies, in particular to African -American women…
    take care, Kathy

  9. avatar
    Christine Morton
    November 6th, 2011 at 21:15 | #9

    Lots of references on the Dr-Pt encounter and race concordance — I have more to share…

    Patient Centeredness, Cultural Competence and Healthcare Quality
    Somnath Saha, Mary Catherine Beach, and Lisa A. Cooper
    J Natl Med Assoc. Author manuscript; available in PMC 2010 February 18.
    PMCID: PMC282458 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2824588/?tool=pmcentrez

    Annals Journal Club: Understanding Concordance in Patient-Physician Relationships: Personal and Ethnic Dimensions of Shared Identity
    Richard L. Street, Jr, Kimberly J. O’Malley, Lisa A. Cooper, and Paul Haidet
    Ann Fam Med. 2008 May; 6(3): 198–205. doi: 10.1370/afm.821.
    PMCID: PMC2384992 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2384992/?tool=pmcentrez

    CONCLUSIONS The physician-patient relationship is strengthened when patients see themselves as similar to their physicians in personal beliefs, values, and communication. Perceived personal similarity is associated with higher ratings of trust, satisfaction, and intention to adhere. Race concordance is the primary predictor of perceived ethnic similarity, but several factors affect perceived personal similarity, including physicians’ use of patient-centered communication.

    Transforming Clinical Practice to Eliminate Racial–Ethnic Disparities in Healthcare
    Donna L. Washington, Jacqueline Bowles, Somnath Saha, Carol R. Horowitz, Sandra Moody-Ayers, Arleen F. Brown, Valerie E. Stone, Lisa A. Cooper, and Writing group for the Society of General Internal Medicine, Disparities in Health Task Force
    J Gen Intern Med. 2008 May; 23(5): 685–691. Published online 2008 January 15. doi: 10.1007/s11606-007-0481-0.
    PMCID: PMC2324135 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2324135/?tool=pmcentrez

    Patient Race/Ethnicity and Quality of Patient–Physician Communication During Medical Visits
    Rachel L. Johnson, Debra Roter, Neil R. Powe, and Lisa A. Cooper
    Am J Public Health. 2004 December; 94(12): 2084–2090.
    PMCID: PMC1448596

    Patient-Provider Racial and Ethnic Concordance and Parent Reports of the Primary Care Experiences of Children
    Gregory D. Stevens, Leiyu Shi, and Lisa A. Cooper
    Ann Fam Med. 2003 July; 1(2): 105–112. doi: 10.1370/afm.27.
    PMCID: PMC1466577

  10. avatar
    Christine Morton
    November 6th, 2011 at 21:17 | #10

    And more on racial disparities in maternal health:

    Dominguez, T. Adverse Birth Outcomes in African American Women: The Social Context of Persistent Reproductive disadvantage. Social Work in Public Health 0; 24: 1-24.

    Dominguez, T.. Race, Racism, and Racial Disparities in Adverse Birth Outcomes. Clinical Obstetrics and Gynecology 2008; 51: 360-370.
    Guendelman S, Thornton D, Gould J, Hosang N.. Social Disparities in Maternal Morbidity During Labor and Delivery Between Mexican-Born and US-Born White Californians, 1996-1998. American Journal of Public Health 2005; 95: 2218-2224.

    Harper, M., E. Dugan, et al. (2007). “Why African-American women are at greater risk for pregnancy-related death.” Annals of Epidemiology 17(3): 180-185.

    Tucker, MJ; Berg, CJ; Callaghan, WM; Hsia, J. The Black-White Disparity in Pregnancy-Related Mortality from 5 Conditions: Differences in Prevalence and Case-Fatality Rates. American Journal of Public Health 2007; 97: 247-251.

  11. November 7th, 2011 at 00:03 | #11

    Hi All,
    Thanks for the references for disparities in birth outcomes. I will work through then all and do a follow up post. I have to admit that I am fascinated by this topic and would like to see us make some headway in my lifetime. Great to see all the work being done. Thanks again for the references.

  12. November 7th, 2011 at 10:18 | #12

    Thank you, Walker, Christine and Kathy for this abundance of additional resources on the topic. Having just finished The Immortal Life of Henrietta Lacks this past weekend it is interesting to contemplate how, in so many ways, very little progress has been made in terms of racial disparities and healthcare.

  13. November 27th, 2011 at 17:46 | #13

    I am entrenched in this work on a daily basis and can only say this – “Where there’s a will, there’s a way”. Our 2007 study came back with a 0% disparity for low-birth weight and prematurity among AA and Hispanic women in my busy urban practice. http://www.commonsensechildbirth.org/jj-way and for the study evaluation http://www.commonsensechildbirth.org/outcomes. Although I developed this midwifery model to address racial and class disparities in MCH, any willing practitioner can provide care this way – it is a prenatal care model with a goal of a full-term, healthy baby for ALL women. As a British -trained midwife here in the US for 22 years, I am still reeling from the shock that these disparities are acceptable. Nothing has changed since I got here, nothing. – Jennie Joseph LM, CPM Founder of The JJ Way – MCH care model.

  1. November 2nd, 2011 at 01:03 | #1