Maternal and Child Health Journal

, Volume 15, Issue 7, pp 931–942

Prenatal Care Utilization in Mississippi: Racial Disparities and Implications for Unfavorable Birth Outcomes

Authors

  • Reagan G. Cox
    • Department of Microbiology and ImmunologyVanderbilt University School of Medicine
    • Office of Health Data and ResearchMississippi State Department of Health
  • Marianne E. Zotti
    • ASB, DRH, NCCDPHPCenters for Disease Control and Prevention
  • Juanita Graham
    • Health Services Chief NurseMississippi State Department of Health
Article

DOI: 10.1007/s10995-009-0542-6

Cite this article as:
Cox, R.G., Zhang, L., Zotti, M.E. et al. Matern Child Health J (2011) 15: 931. doi:10.1007/s10995-009-0542-6

Abstract

The objective of the study is to identify racial disparities in prenatal care (PNC) utilization and to examine the relationship between PNC and preterm birth (PTB), low birth weight (LBW) and infant mortality in Mississippi. Retrospective cohort from 1996 to 2003 linked Mississippi birth and infant death files was used. Analysis was limited to live-born singleton infants born to non-Hispanic white and black women (n = 292,776). PNC was classified by Kotelchuck’s Adequacy of Prenatal Care Utilization Index. Factors associated with PTB, LBW and infant death were identified using multiple logistic regression after controlling for maternal age, education, marital status, place of residence, tobacco use and medical risk. About one in five Mississippi women had less than adequate PNC, and racial disparities in PNC utilization were observed. Black women delayed PNC, received too few visits, and were more likely to have either “inadequate PNC” (P < 0.0001) or “no care” (P < 0.0001) compared to white women. Furthermore, among women with medical conditions, black women were twice as likely to receive inadequate PNC compared to white women. Regardless of race, “no care” and “inadequate PNC” were strong risk factors for PTB, LBW and infant death. We provide empirical evidence to support the existence of racial disparities in PNC utilization and infant birth outcomes in Mississippi. Further study is needed to explain racial differences in PNC utilization. However, this study suggests that public health interventions designed to improve PNC utilization among women might reduce unfavorable birth outcomes especially infant mortality.

Keywords

Preterm birthLow birth weightInfant mortalityPrenatal care utilizationRacial disparities

Introduction

Prenatal care (PNC) has long been endorsed as a means to monitor maternal and fetal health and identify women at risk for unfavorable birth outcomes [13]. Early and continuous PNC is encouraged to allow for the delivery of an array of medical, nutritional and educational interventions at different stages of pregnancy [4, 5]. Despite recommendations, a proportion of women continue to receive insufficient or no PNC. Of particular concern is the large percentage of blacks, Hispanics, Hawaiians and Native Americans who receive insufficient PNC, although factors underlying racial differences in PNC utilization remain poorly understood [617].

Insufficient or no PNC is associated with unfavorable birth outcomes. Women who delay or receive no PNC are more likely to deliver a low-birth-weight (LBW) infant or have a preterm birth (PTB) compared to women receiving early care, and inadequate PNC is associated with an increased risk for neonatal death [1722]. Furthermore, evidence suggests that PNC interventions, such as nutritional counseling, psychosocial counseling or those targeting specific risk factors like smoking, can lower risk for LBW and PTB [23, 24].

Mississippi (MS) is burdened with the highest rates of LBW (percentage of 2003 births: MS = 11.4%; U.S. = 7.9%), PTB (Percentage of 2003 births: MS = 17.9%; U.S. = 12.3%) and infant mortality (Infant deaths per 1,000 live births in 2003: MS = 10.74; U.S. = 6.85) in the nation, as well as persisting racial disparities in the occurrence of these unfavorable birth outcomes [6, 25]. Identifying modifiable risk factors for these outcomes could translate into interventions that successfully improve birth outcomes. Delivering interventions during preconception healthcare visits would be preferable, but many Mississippi women have limited/no access to care outside of pregnancy, so PNC visits may be the first opportunity for healthcare providers to identify risk factors and implement interventions aimed at decreasing the likelihood of LBW, PTB and infant mortality. Identifying women who are not receiving adequate PNC is critical for planning efforts to eliminate disparities in PNC utilization as Mississippi’s distinctive demography provides a challenge for public health efforts. Most residents live in rural areas and more than one-fourth live below the federal poverty level; blacks are more than three times as likely as whites to live in poverty [26]. Racial diversity in Mississippi is also unique: 37% of residents are black, 5% report race/ethnicity other than non-Hispanic white or black, and 2% are of Hispanic origin [27]. Moreover, approximately half of the women giving birth in the state are black, providing a unique population in which to assess racial disparities in birth outcomes [6].

We know of no previous studies that have systematically examined PNC among Mississippi women. This study examined PNC utilization by maternal race among women in Mississippi and sought to elucidate sociodemographic characteristics of women with insufficient PNC. We hypothesized that inadequate PNC utilization contributed to elevated rates of unfavorable birth outcomes in Mississippi, especially among black women. The principal goal of our study was to examine by race the independent association between PNC utilization and PTB, LBW and infant mortality in Mississippi after controlling for confounding risk factors: mother’s age, education, marital status, place of residence, tobacco use and medical risk.

Methods

Study Population

We conducted a population-based retrospective cohort study using linked birth and death certificate files provided by the Mississippi State Department of Health (MSDH) for the years 1996–2003. The cohort included all infants born to Mississippi residents (n = 341,780). Analysis was limited to live-born singleton infants (n = 328,393) of non-Hispanic white and black women (n = 300,710; 91.6%). Infants from plural births were excluded because they are at higher risk for poor birth outcomes compared to singleton births. Infants of other races were excluded due to small sample sizes. Infants born weighing less than 500 g or before 20 completed weeks of gestation were excluded due to questionable viability. The sample was limited to infants with a reported gestational age between 20 and 45 weeks. Infants with missing data for PNC, birth weight, gestational age or any selected maternal factor were excluded from all analyses. The final sample consisted of 292,776 infants. This study was approved by the MSDH Institutional Review Board and was conducted in accordance with prevailing ethical principles.

Variables

PNC utilization was defined according to the Adequacy of Prenatal Care Utilization (APNCU) Index, which has been described in detail elsewhere [28]. In Mississippi, prenatal care initiation and number of visits are self-reported by mothers on infant birth certificates. Women who reported no PNC were classified into a unique category and identified in all analyses as “no care.” Expected PNC visits for a specific gestational age were based upon the American College of Obstetricians and Gynecologists (ACOG) guidelines for an uncomplicated pregnancy [29]. Briefly, APNCU was defined as (1) no care, (2) inadequate- care initiated after the fourth month of pregnancy or receiving less than half of recommended PNC visits, (3) intermediate- care begun by the fourth month of pregnancy and 50–79% of recommended PNC visits, (4) adequate- care begun by the fourth month and 80–109% of recommended visits, or (5) intensive- care initiated by the fourth month and at least 110% of recommended visits.

Several potential confounders were considered. The MSDH provides services in nine public health districts (Fig. 1) and “district” was used as a geographic region variable. Self-reported maternal characteristics (from infant birth certificates) including maternal age, race/ethnicity, education level, marital status, tobacco use and medical risk factors during pregnancy were considered in the analyses. Maternal age was grouped into four categories: 11–17, 18–24, 25–34 and 35+ years. Race/ethnicity was classified as non-Hispanic white or non-Hispanic black (hereafter, white or black) and marital status was defined as married or not married. Maternal education was classified as less than high school, high school graduate or at least one year of college. Maternal tobacco use during pregnancy was classified into different dose categories: none, moderate (1–9 cigarettes per day) or heavy use (10+ cigarettes per day). Medical risk factor indicates any of the following conditions: anemia, acute or chronic lung disease, cardiac disease, diabetes, eclampsia, genital herpes, hemoglobinopathy, hypertension, hydramnios/oligohydramnios, incompetent cervix, history of previous large infant or preterm or small-for-gestational-age infant, renal disease, Rh sensitization, uterine bleeding or other specified condition that placed a woman at risk for a complicated pregnancy.
https://static-content.springer.com/image/art%3A10.1007%2Fs10995-009-0542-6/MediaObjects/10995_2009_542_Fig1_HTML.gif
Fig. 1

Mississippi public health districts

The three outcome variables were infant death (died in first 364 days of life), LBW (weighing between 500 and 2,499 g) and PTB (birth between 20 weeks and 36 weeks, 6 days). Gestational age was calculated based upon date of infant birth and date of last menstrual period (LMP) when the LMP estimate was within 1 week of a clinical estimate of gestational age, otherwise and in the absence of LMP a clinical estimate of gestational age was used [30, 31].

Statistical Analysis

We used SAS 9.1 (SAS Institute Inc., Cary, NC, USA, 2002) for all statistical analyses. Chi-square analyses and t-tests were used to assess the statistical significance of observed differences in outcomes by maternal characteristics. Descriptive statistics for the distribution of live births, infant deaths and infant mortality rates by selected factors were calculated. Mantel–Haenszel Chi-square (M-H χ2) analyses were used to assess the relationship between two variables while controlling for a third; reported statistics represent the alternative hypothesis of a nonzero correlation between the two variables of interest. Multiple logistic regression models were tested for the entire sample and within race groups to identify factors independently associated with LBW, PTB, and infant death. Adjusted odds ratios (aOR) and 95% confidence intervals (CIs) are reported from SAS output. An aOR was considered statistically significant if its 95% CI did not include 1.0. All reported P-values are two-sided.

Results

Table 1 presents the distribution of 292,776 singleton live births by selected maternal characteristics, PNC utilization and birth outcome. Most women were 18–34 years old (85.3%), white (52.4%), married (52.6%) and had at least a high school education (74.0%). Nearly 13% reported smoking and 23.3% had at least one medical risk factor during pregnancy. Maternal sociodemographic characteristics differed significantly according to maternal race. Black women tended to be younger than white women (average age: blacks = 23.6 years, whites = 25.8 years; P < 0.0001), less educated (less than high school: blacks = 31.4%, whites = 21.0%; P < 0.0001) and more likely to be unmarried (blacks = 75.6%, whites = 21.8%; P < 0.0001).
Table 1

Infant birth outcomes by selected characteristics: Mississippi, 1996–2003 birth cohort

Characteristic

Live births

Preterm births

Low birth weights

Infant deaths

 

(n = 292,776)

(n = 45,054)

(n = 25,023)

(n = 2,156)

IMRa

n (%)

n (%)

n (%)

n (%)

Rate per 1,000

Maternal race

 Black/non-hispanic

139,471 (47.6)

26,948 (59.8)

16,196 (64.7)

1,366 (63.4)

9.8

 White/non-hispanic

153,305 (52.4)

18,106 (40.2)

8,827 (35.3)

790 (36.6)

5.2

Maternal age (yrs)

 11–17

22,652 (7.7)

4,528 (10.0)

2,639 (10.6)

239 (11.1)

10.6

 18–24

135,270 (46.2)

20,984 (46.6)

12,350 (49.4)

1,102 (51.1)

8.1

 25–34

114,605 (39.1)

16,068 (35.7)

8,098 (32.4)

675 (31.3)

5.9

 35+

20,249 (6.9)

3,474 (7.7)

1,936 (7.7)

140 (6.5)

6.9

Maternal education level

 Less than high school

75,999 (26.0)

13,665 (30.3)

8,329 (33.3)

767 (35.6)

10.1

 High school

96,172 (32.8)

15,278 (33.9)

8,625 (34.5)

761 (35.3)

7.9

 College

120,605 (41.2)

16,111 (35.8)

8,069 (32.2)

628 (29.1)

5.2

Marital status

 Not married

138,901 (47.4)

25,544 (56.7)

15,446 (61.7)

1,341 (62.2)

9.6

 Married

153,875 (52.6)

19,510 (43.3)

9,577 (38.3)

815 (37.8)

5.3

Tobacco use

 10+ cigarettes/day

23,923 (8.2)

3,580 (8.0)

2,910 (11.6)

255 (11.8)

10.6

 1–9 cigarettes/day

12,782 (4.4)

2,028 (4.5)

1,485 (5.9)

127 (5.9)

9.9

 None

256,071 (87.5)

39,446 (87.6)

20,628 (82.4)

1,774 (82.3)

6.9

Maternal medical risk

 Yes

68,136 (23.3)

14,344 (31.8)

9,877 (39.5)

726 (33.7)

10.6

 No

224,640 (76.7)

30,710 (68.2)

15,146 (60.5)

1,430 (66.3)

6.4

Initiation of prenatal careb

 Pregnancy month 1 or 2

190,551 (65.1)

26,944 (59.8)

15,024 (60.0)

1,256 (58.3)

6.6

 Pregnancy month 3 or 4

71,052 (24.3)

11,373 (25.2)

6,596 (26.4)

562 (26.1)

7.9

 Pregnancy month 5 or 6

21,282 (7.3)

4,385 (9.7)

2,095 (8.4)

199 (9.2)

9.4

 Pregnancy month 7, 8 or 9

7,228 (2.5)

1,384 (3.1)

660 (2.6)

57 (2.6)

7.9

 No care

2,663 (0.9)

968 (2.2)

648 (2.6)

82 (3.8)

30.8

Prenatal care receivedb

     

 Adequate plus

108,029 (36.9)

25,483 (56.6)

15,793 (63.1)

1,104 (51.2)

10.2

 Adequate

142,547 (48.7)

13,741 (30.5)

6,327 (25.3)

685 (31.8)

4.8

 Intermediate

34,645 (11.8)

4,048 (9.0)

1,848 (7.4)

232 (10.8)

6.7

 Inadequate

4,892 (1.7)

814 (1.8)

407 (1.6)

53 (2.5)

10.8

 No care

2,663 (0.9)

968 (2.2)

648 (2.6)

82 (3.8)

30.8

APNCU indexb

 Intensive

93,717 (32.0)

22,101 (49.0)

13,940 (55.7)

960 (44.5)

10.2

 Adequate

134,867 (46.1)

12,434 (27.6)

5,808 (23.2)

617 (28.6)

4.6

 Intermediate

29,543 (10.1)

3,225 (7.2)

1,557 (6.2)

200 (9.3)

6.8

 Inadequate

31,986 (10.9)

6,326 (14.0)

3,070 (12.3)

297 (13.8)

9.3

 No care

2,663 (0.9)

968 (2.2)

648 (2.6)

82 (3.8)

30.8

APNCU index = adequacy of prenatal care utilization summary index based upon month of prenatal care initiation and adequacy of received services index

aInfant mortality rate (IMR) = (number of infant deaths/number of live births) × 1,000

bBased upon Kotelchuck adequacy of prenatal care utilization index [28]. Prenatal care received is Kotelchuck’s adequacy of received services index based upon actual number of prenatal care visits compared to expected number of visits, adjusted for gestational age at birth and month of prenatal care initiation

PNC Among Mississippi Women

Most women (65.1%) initiated PNC in the first two months of pregnancy and 85.6% received an adequate number of PNC visits, resulting in most women (78.1%) utilizing either intensive or adequate PNC on the APNCU Index. Table 2 presents PNC utilization differences according to maternal characteristics. Receipt of adequate or intensive care correlated with older age, being married and medical risk. Less education and smoking were associated with no care and less than adequate care. Black women were significantly more likely to receive inadequate PNC and no care during pregnancy compared to white women (P < 0.0001 for both comparisons). We also observed regional differences in PNC utilization by maternal race (Table 3). In nearly all districts, black women were more likely to receive less than adequate care and tended to initiate PNC later than white women (data not shown).
Table 2

Prenatal care utilization by maternal characteristics: Mississippi, 1996–2003 birth cohort

Characteristic

APNCU indexa

Intensive

Adequate

Intermediate

Inadequate

No care

(n = 93,717)

(n = 134,867)

(n = 29,543)

(n = 31,986)

(n = 2,663)

n (%)b

n (%)b

n (%)b

n (%)b

n (%)b

Maternal race

 

 Black/non-Hispanic

42,441 (30.4)

56,678 (40.6)

15,260 (10.9)

22,928 (16.4)

2,164 (1.6)

 White/non-Hispanic

51,276 (33.4)

78,189 (51.0)

14,283 (9.3)

9,058 (5.9)

499 (0.3)

Maternal age (yrs)

 

 11–17

5,775 (25.5)

8,740 (38.6)

2,666 (11.8)

5,127 (22.6)

344 (1.5)

 18–24

42,381 (31.3)

58,727 (43.4)

14,690 (10.9)

18,050 (13.3)

1,422 (1.0)

 25–34

38,567 (33.6)

57,476 (50.2)

10,540 (9.2)

7,309 (6.4)

713 (0.6)

 35+

6,994 (34.5)

9,924 (49.0)

1,647 (8.1)

1,500 (7.4)

184 (0.9)

Maternal education level

 

 Less than high school

21,687 (28.5)

29,999 (39.5)

9,055 (11.9)

14,003 (18.4)

1,255 (1.6)

 High school

31,457 (32.7)

42,368 (44.0)

10,010 (10.4)

11,376 (11.8)

961 (1.0)

 College

40,573 (33.6)

62,500 (51.8)

10,478 (8.7)

6,607 (5.5)

447 (0.4)

Marital status

 

 Not married

40,934 (29.5)

55,394 (39.9)

15,767 (11.4)

24,465 (17.6)

2,341 (1.7)

 Married

52,783 (34.3)

79,473 (51.6)

13,776 (9.0)

7,521 (4.9)

322 (0.2)

Tobacco use

 

 10+ cigarettes/day

7,542 (31.5)

10,381 (43.4)

2,525 (10.6)

3,135 (13.1)

340 (1.4)

 1–9 cigarettes/day

4,125 (32.3)

5,295 (41.4)

1,435 (11.2)

1,751 (13.7)

176 (1.4)

 None

82,050 (32.0)

119,191 (46.6)

25,583 (10.0)

27,100 (10.6)

2,147 (0.8)

Maternal medical risk

 

 Yes

26,661 (39.1)

27,118 (39.8)

6,083 (8.9)

7,746 (11.4)

528 (0.8)

 No

67,056 (29.8)

107,749 (48.0)

23,460 (10.4)

24,240 (10.8)

2,135 (1.0)

APNCU Index = adequacy of prenatal care utilization summary index based upon month of prenatal care initiation and adequacy of received services index

aBased upon Kotelchuck adequacy of prenatal care utilization index [28]

bRow percentage reported

Table 3

Adequacy of prenatal care utilization by mother’s race and public health district: Mississippi, 1996–2003 birth cohort

District

APNCU indexa

Intensive

Adequate

Intermediate

Inadequate

No care

Mother’s raceb

Mother’s raceb

Mother’s raceb

Mother’s raceb

Mother’s raceb

White

Black

White

Black

White

Black

White

Black

White

Black

(%)c

(%)c

(%)c

(%)c

(%)c

(%)c

(%)c

(%)c

(%)c

(%)c

I

35.5

25.6

50.4

25.6

5.9

13.7

7.7

15.5

0.6

1.8

II

26.5

25.4

52.8

44.6

14.7

13.4

5.8

15.4

0.2

1.3

III

25.6

22.3

56.7

42.3

12.1

13.6

5.4

19.6

0.2

1.8

IV

36.8

33.2

41.7

32.6

15.3

14.4

6.0

18.6

0.3

1.2

V

26.2

35.9

62.5

41.2

7.2

6.9

3.8

14.1

0.3

1.8

VI

45.2

39.7

43.2

37.9

6.7

7.7

4.6

13.6

0.3

1.1

VII

38.1

26.8

47.9

42.1

6.4

10.2

7.4

19.9

0.2

1.1

VIII

37.7

30.1

52.1

47.6

5.5

7.3

4.6

14.0

0.2

1.1

IX

36.4

27.6

45.2

32.4

9.3

18.3

8.5

19.8

0.5

2.0

I

35.5

25.6

50.4

25.6

5.9

13.7

7.7

15.5

0.6

1.8

II

26.5

25.4

52.8

44.6

14.7

13.4

5.8

15.4

0.2

1.3

III

25.6

22.3

56.7

42.3

12.1

13.6

5.4

19.6

0.2

1.8

Mean

34.2

29.6

50.3

38.5

9.2

11.7

6.0

16.7

0.3

1.5

Median

36.4

27.6

50.4

41.2

7.2

13.4

5.8

15.5

0.3

1.3

SDd

6.7

5.6

6.7

7.0

3.8

3.9

1.6

2.7

0.1

0.4

aAPNCU index = Kotelchuck’s adequacy of prenatal care utilization summary index [28]

b,cAnalysis limited to non-hispanic white and black mothers. Percentage (%) of mothers with APNCU score

dStandard deviation

Because black women were more likely to report a medical condition (27.0% vs 19.9%; P < 0.0001), yet appeared less likely to receive intensive PNC compared to white women, we performed ad hoc analyses to assess the relationship between medical risk during pregnancy, maternal race and APNCU. Regardless of race, medical risk increased the likelihood that a woman received intensive over adequate PNC; however, the relationship between APNCU and medical risk differed by maternal race (M-H χ2 = 910.9; P < 0.0001). This correlation appeared to be driven by the large percentage of black women who received inadequate care regardless of medical risk (16.7%) compared to inadequate care among whites (5.9%), as trends in APNCU at higher levels of care were similar by race (data not shown). Racial disparities in PNC were also observed; black women with medical conditions (n = 9,839 or 26.4%) were twice as likely to receive less than adequate PNC compared to white women with medical conditions (n = 3,990 or 13.1%).

PNC Utilization and Preterm Birth (PTB)

Regardless of race, intensive, inadequate and no PNC were associated with a significant increased risk for PTB (Tables 4 and 5). Intensive APNCU was associated with about a three-fold increase in risk for PTB, likely reflecting the tendency for high-risk pregnancies to receive intensive PNC. Women receiving inadequate APNCU had two times the odds of PTB compared to women with adequate APNCU. Although a small group of women (n = 968 or 2.2%), no PNC was significantly associated with high odds of PTB (P < 0.0001). Based upon observed regional differences in PNC utilization between blacks and whites, “health district” was included in the logistic models. It was significant in the models but altered neither the aOR nor associated 95% CI for any variable in the models. Thus, the reported models excluded “health district” in order to be more parsimonious.
Table 4

Adjusted odds ratios for preterm birth, low birth weight and infant death among infants by selected characteristics: Mississippi, 1996–2003 birth cohort

Characteristic

Odds ratioa (95% CI)b

Preterm birthc

Low birth weightc

Infant diedc

Maternal race

 Black/non-hispanic

1.6 (1.5–1.6)

2.0 (2.0–2.1)

1.7 (1.5–1.9)

 White/non-hispanic

1.0 (referent)

1.0 (referent)

1.0 (referent)

Maternal age (yrs)

 11–17

1.1 (1.1–1.2)

1.2 (1.2–1.3)

1.2 (1.0–1.4)

 18–24

0.9 (0.9–1.0)

1.1 (1.0–1.1)

1.1 (1.0–1.2)

 25–34

1.0 (referent)

1.0 (referent)

1.0 (referent)

 35+

1.2 (1.2–1.3)

1.3 (1.2–1.4)

1.1 (0.9–1.3)

Maternal education level

 Less than high school

1.2 (1.1–1.2)

1.2 (1.2–1.3)

1.4 (1.2–1.6)

 High school

1.1 (1.0–1.1)

1.1 (1.1–1.1)

1.2 (1.1–1.4)

 College

1.0 (referent)

1.0 (referent)

1.0 (referent)

Marital status

 Not married

1.1 (1.1–1.2)

1.2 (1.2–1.2)

1.1 (1.0–1.3)

 Married

1.0 (referent)

1.0 (referent)

1.0 (referent)

Tobacco use

 10+ cigarettes/day

1.1 (1.0–1.1)

2.1 (2.0–2.2)

1.8 (1.5–2.1)

 1–9 cigarettes/day

1.1 (1.0–1.1)

1.7 (1.6–1.8)

1.5 (1.2–1.8)

 None

1.0 (referent)

1.0 (referent)

1.0 (referent)

Maternal medical risk

 Yes

1.5 (1.4–1.5)

2.0 (2.0–2.1)

1.5 (1.4–1.6)

 No

1.0 (referent)

1.0 (referent)

1.0 (referent)

APNCU indexd

 Intensive

2.9 (2.9–3.0)

3.7 (3.6–3.8)

2.1 (1.9–2.3)

 Adequate

1.0 (referent)

1.0 (referent)

1.0 (referent)

 Intermediate

1.1 (1.1–1.2)

1.1 (1.0–1.2)

1.3 (1.1–1.5)

 Inadequate

2.0 (1.9–2.0)

1.7 (1.6–1.7)

1.5 (1.3–1.7)

 No care

4.4 (4.0–4.8)

4.8 (4.4–5.3)

4.7 (3.7–6.0)

aOdds ratios were adjusted for all selected characteristics

b95% Confidence interval

cPreterm birth = infant gestational age less than 37 weeks. Low birth weight = infant birth weight less than 2,500 g

Infant death = infant died before first birthday

dAPNCU index = Kotelchuck’s adequacy of prenatal care utilization summary index [28]

PNC Utilization and Low Birth Weight (LBW)

Regardless of race, intensive, inadequate and no PNC were all associated with an elevated risk of LBW (Tables 4 and 5). Women who received intensive care were nearly 4 times as likely to have a LBW infant compared to women with adequate APNCU after adjusting for maternal characteristics and the presence of one or more high risk medical conditions (aOR = 3.7; P < 0.0001). Intensive, inadequate APNCU, no PNC, heavy tobacco use and medical risk were all strong risk factors for delivery of a LBW infant (aOR is close to 2.0 or higher) among both white and black women (Table 5).
Table 5

Adjusted odds ratios for preterm birth, low birth weight and infant death among white and black infants by selected characteristics: Mississippi, 1996–2003 birth cohort

Characteristics

Preterm birtha

Low birth weighta

Infant deatha

Mother’s raceb

Mother’s raceb

Mother’s raceb

White

Black

White

Black

White

Black

Odds ratioc

Odds ratioc

Odds ratioc

Odds ratioc

Odds ratioc

Odds ratioc

(95% CI)d

(95% CI)d

(95% CI)d

(95% CI)d

(95% CI)d

(95% CI)d

Maternal age (yrs)

 11–17

1.0 (0.9–1.1)

1.2 (1.1–1.2)

1.2 (1.1–1.3)

1.3 (1.2–1.4)

1.0 (0.7–1.3)

1.3 (1.1–1.6)

 18–24

0.9 (0.9–1.0)

0.9 (0.9–1.0)

1.0 (1.0–1.1)

1.1 (1.0–1.1)

1.1 (1.0–1.4)

1.1 (0.9–1.2)

 25–34

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

 35+

1.2 (1.1–1.3)

1.2 (1.2–1.3)

1.3 (1.2–1.4)

1.2 (1.2–1.3)

1.3 (1.0–1.7)

1.0 (0.8–1.2)

Maternal education level

 Less than high school

1.2 (1.2–1.3)

1.1 (1.1–1.2)

1.4 (1.3–1.5)

1.1 (1.1–1.2)

1.8 (1.5–2.3)

1.2 (1.0–1.4)

 High school

1.1 (1.0–1.1)

1.1 (1.0–1.1)

1.2 (1.1–1.2)

1.0 (1.0–1.1)

1.3 (1.1–1.6)

1.1 (1.0–1.3)

 College

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

Marital status

 Not married

1.1 (1.1–1.2)

1.1 (1.1–1.2)

1.2 (1.1–1.3)

1.2 (1.1–1.2)

1.3 (1.1–1.6)

1.0 (0.9–1.2)

 Married

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

Tobacco use

 10+ cigarettes/day

1.1 (1.0–1.2)

1.1 (1.0–1.2)

2.1 (2.0–2.2)

2.0 (1.8–2.1)

1.6 (1.4–1.9)

1.8 (1.4–2.3)

 1–9 cigarettes/day

1.1 (1.0–1.2)

1.0 (0.9–1.1)

1.8 (1.6–1.9)

1.5 (1.4–1.7)

1.3 (1.0–1.7)

1.5 (1.2–2.0)

 None

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

Maternal medical risk

 Yes

1.6 (1.5–1.6)

1.4 (1.4–1.5)

2.3 (2.2–2.4)

1.9 (1.8–2.0)

1.7 (1.4–2.0)

1.4 (1.3–1.6)

 No

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

APNCU indexe

 Intensive

3.1 (3.0–3.2)

2.8 (2.7–2.9)

3.9 (3.7–4.1)

3.5 (3.4–3.6)

1.8 (1.6–2.1)

2.3 (2.0–2.6)

 Adequate

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

1.0 (referent)

 Intermediate

1.1 (1.0–1.2)

1.1 (1.1–1.2)

1.1 (1.0–1.2)

1.1 (1.0–1.2)

1.1 (0.9–1.5)

1.4 (1.2–1.8)

 Inadequate

1.9 (1.8–2.0)

2.0 (1.9–2.0)

1.9 (1.7–2.1)

1.6 (1.5–1.7)

1.7 (1.3–2.2)

1.5 (1.3–1.8)

 No Care

4.6 (3.7–5.6)

4.3 (3.9–4.7)

5.3 (4.2–6.8)

4.7 (4.2–5.2)

2.9 (1.4–6.0)

5.4 (4.2–7.0)

aPreterm birth = infant gestational age less than 37 weeks. Low birth weight = infant birth weight less than 2,500 g. Infant death = infant died before first birthday

bAnalysis limited to non-hispanic white and black mothers

cOdds ratios were adjusted for all selected characteristics

d95% Confidence interval

eAPNCU index = Kotelchuck’s adequacy of prenatal care utilization summary index [28]

PNC Utilization and Infant Mortality

All selected maternal characteristics except maternal age and marital status were significant predictors of infant death in the multivariate model (Table 4). Regardless of race, PNC utilization was the strongest factor for infant death after adjusting for maternal age, race, education, marital status, tobacco use and medical risk (Table 5). Intensive APNCU and no PNC were associated with a higher risk for infant death than either intermediate or inadequate APNCU when compared to adequate APNCU. This U-shaped association between APNCU and infant mortality (as well as PTB and LBW) is expected as extensive PNC utilization (“intensive PNC”) is often linked to high-risk pregnancies (Fig. 2). Women with high-risk pregnancies are at elevated risk for and experience higher rates of poor birth outcomes than women with uncomplicated pregnancies regardless of utilizing PNC more frequently. No PNC during pregnancy was associated with about threefold increased risk for infant death among white mothers and over a fivefold increased risk for infant death among black mothers (Table 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs10995-009-0542-6/MediaObjects/10995_2009_542_Fig2_HTML.gif
Fig. 2

U-Shaped association between APNCU and odds of PTB, LBW and infant death: Mississippi, 1996–2003 birth cohort

Discussion

We found that APNCU is significantly and independently associated with PTB, LBW and infant death among Mississippi infants, which is consistent with other studies that found PNC is associated with a lower risk for unfavorable birth outcomes [1724]. We observed increased odds of infant death among black women receiving either intensive care or no care compared to white women receiving the same level of care. Because we observed this difference after controlling for age, education, marital status, tobacco use and medical risk, but did not observe elevated odds for PTB and LBW, we postulate that socioeconomic stresses and post-delivery health-related issues contributed to this observation. Our study design did not allow us to address these important variables, but our observation highlights the need to conduct studies designed to gain a deeper understanding of barriers to care that black Mississippi mothers face.

After considering sociodemographic factors, medical risk and PNC utilization, black women in Mississippi were significantly more likely to have a LBW infant, PTB or infant death compared to white women. Race/ethnicity is a social classification reflecting many cultural, psychological and sociological factors that influence health, health beliefs and health practices. We speculate that observed racial disparities in birth outcomes are largely due to variables that are affected by socioeconomic conditions and individual lifestyles, with a critical variable being preconception maternal health. Recent studies have documented the benefits of preconception and interconception healthcare for improving overall maternal health so as to reduce the risk for subsequent poor birth outcomes and facilitating early entry into PNC [3238]. In late 2008, the MSDH initiated two pilot programs to evaluate the impact of interconception care on the reoccurrence of very LBW births, the leading cause of Mississippi infant death. The MSDH anticipates that high-risk case management during the interconception period will improve birth outcomes and help reduce racial disparities in the state. All Mississippi women could benefit from preconception and interconception services, but often lack insurance coverage during the periods before and between pregnancies [39]. Medicaid is the leading payer of pregnancy care in Mississippi (and nationally), but Medicaid only covers pregnancy services from diagnosis until 6 weeks post-partum, leaving many mothers without coverage during the interconception period. Medicaid programs could save millions of healthcare dollars by shifting to strategies that prevent, rather than treat, poor birth outcomes [40]. Development of health policy to establish a Medicaid waiver program that would provide up to two years of primary care and case management services for women at risk for high-risk deliveries could profoundly impact Mississippi birth outcomes.

About one in five Mississippi women are not receiving adequate PNC. Access to PNC is a concern in many rural regions; some mothers may travel over an hour to the nearest provider or over a hundred miles to a specialist. All Mississippi counties have federally designated medically underserved areas and some have neither a practicing physician nor nurse practitioner [41]. A paucity of providers may help explain some women’s late or failed entry into PNC. Regional strategies that increase access to PNC in rural communities or, at minimum, provide supportive services between PNC visits may be helpful to improve PNC utilization among all Mississippi women.

We found that black mothers are more likely than white mothers to receive inadequate PNC. Racial disparities in APNCU were prevalent statewide and geographic residence did not account for racial differences in birth outcomes when tested in our models. Thus, geographic disadvantage appears to contribute to poor birth outcomes among all Mississippi mothers (regardless of race) consistent with a growing body of research [4244]. In Mississippi, geographic disadvantage, e.g. rural residence, limited providers and low income, impacts a women’s ability to initiate and obtain routine PNC during her pregnancy. Transportation is a critical barrier to care (unpublished data from the MSDH Health Services Data Unit). Outside of Jackson, Mississippi, public transportation is not available and statewide Medicaid transportation services have restrictions. One challenge for Mississippi (and other rural areas in the nation) is how to enable transportation to healthcare services. Unique solutions, statewide and at the county-level, will likely be required. For example, (1) expanded Medicaid transportation services for high-risk case management, (2) mobile health units, or (3) bus/van transportation to clinics on a once-a-week basis.

Black mothers with medical risk factors (more commonly than white mothers) are not receiving adequate PNC, suggesting that preconception health influences observed racial disparities in APNCU. Indicators of preconception health were not available in the present study, but Pregnancy Risk Assessment Monitoring System (PRAMS) data were available for selected indicators among Mississippi mothers who gave birth to live-born infants in 2003 [45]. PRAMS is a population-based surveillance system designed to monitor maternal behaviors/experiences around the time of pregnancy by collecting self-reported survey data from mothers in reporting areas [45]. According to the Mississippi PRAMS data, black mothers were more likely to be obese and report either preconception stress or physical abuse compared to white mothers [45]. The PRAMS data coupled with our observation that black women utilize PNC less than white women suggests there are critical unmet medical needs that are more prevalent among black compared to white mothers. Socioeconomic challenges faced by black women likely affect their preconception health, ability to manage medical conditions and health seeking behaviors during pregnancy. In a study seeking to assess the impact of socioeconomic status (SES) on racial differences in unmet medical needs, Wiltshire et al. [13] found that black women with less education reported fewer unmet medical needs, and suggested that lower self-perceived needs may be due to less education and the influence of social vulnerabilities. Perceived medical needs among Mississippi women might be lower among women who experience overwhelming access barriers or consider medical care a last resort or “emergency only” alternative. Racial disparities in perceived needs and health-seeking behaviors among Mississippi mothers need to be explored in depth in order to identify unmet needs that contribute to racial disparities in birth outcomes.

Limitations

Information on birth certificates is limited and numerous variables that could not be measured, e.g. pre-pregnancy health status, obesity, nutrition, stress, physical abuse, poverty, health insurance status and quality of PNC, may be contributing to high outcome rates and racial differences in outcomes. Birth certificates do not include a measure of SES. We controlled for education, but given statewide poverty levels it is not clear that education is an adequate proxy for SES. Data for substance use during pregnancy were not available in our dataset and we specifically excluded alcohol use due to underreporting on birth certificates. Women may not self-report alcohol/substance use on a birth certificate for fear of legal consequences, e.g. custody issues or imprisonment. Alcohol/substance use during pregnancy likely contributes to excess poor birth outcomes in Mississippi; however, we were unable to evaluate the influence of these behaviors.

PNC is multifaceted and differs based upon utilization (assessed in the present study), provider and content which varies according to risk assessment, medical screening and the implementation of specific interventions. The APNCU index does not address PNC content, may differentially place women with short gestations into the intensive category [46], and is based upon guidelines for an uncomplicated pregnancy. PNC content is different for all women and lack of a gold standard limits our ability to control for this variation. Mothers elect into PNC services and four self-selection processes have been previously described: favorable, estrangement, adverse and confidence selection [47]. We observed that APNCU trends differed significantly by medical risk (Table 2 and data not shown), suggesting both adverse and favorable selection into PNC. These self-selection processes introduce opposing selection biases on study associations (adverse selection results in underestimation, while favorable selection results in overestimation) but we were unable to quantify the degree of bias. Missing data resulted in infants being excluded from the study sample. Overall, 1.9% of infants were excluded and they tended to have risk factors for unfavorable birth outcomes (data not shown). We expect women are likely to underreport tobacco use and over-report PNC because they know these behaviors are associated with favorable outcomes [4850]. Differential exclusion of high-risk infants, underreporting of tobacco use and over-reporting PNC should bias study effects towards the null hypothesis—odds ratios equal to one—and result in underestimation of reported study associations.

Conclusions

We provide empirical evidence to support the existence of racial disparities in PNC utilization and infant birth outcomes in Mississippi. Before targeting racial disparities in PNC with more effective public health and clinical interventions, studies designed to elucidate barriers to PNC at a regional level, evaluate PNC content from different providers and develop a deeper understanding of factors that impact PNC utilization among Mississippi women are needed. Other initiatives such as establishing pre-pregnancy Medicaid eligibility and waiver programs for mothers to receive interconception care could limit financial barriers to care as well as allow mothers to remain linked to healthcare providers.

Mississippi mothers appear to differ in their unmet preconception/interconception healthcare needs. Healthcare resources and barriers to care at the community level vary and are impossible to address with a “one size fits all” plan. However, by working within communities with available resources, interventions can be molded to meet the specific needs of mothers. Pilot programs in the Mississippi Delta and Jackson Metropolitan areas have recently been established to identify community unmet needs and barriers to PNC and design intervention plans to reduce the reoccurrence of very LBW infant births. These programs enroll black mothers who have recently delivered a very LBW infant and help them improve their overall health, optimize child spacing, and develop personal reproductive plans. Program initiatives include (1) identifying community-level transportation assistance options, (2) partnering with Federally Qualified Community Health Centers to establish medical homes for women without primary care providers, (3) providing peer support and case management services at local health departments, and (4) utilizing nursing, social services, nutrition and outreach teams to provide multidisciplinary case management. Ultimately, these programs seek to develop strategies to make interconception care available across Mississippi despite the limited resources. Other states could benefit from similar initiatives as providing preconception and interconception care could facilitate early entry into PNC, ensure that high-risk mothers are identified early in pregnancy and reduce high rates of LBW, PTB and infant mortality across the nation.

Acknowledgments

We gratefully acknowledge Dick Johnson for data linkage support and Daniel R. Bender for his guidance and support on this study.

Copyright information

© Springer Science+Business Media, LLC 2009