Abstract
Unintended birth is associated with adverse maternal and infant outcomes. In 2006, US Hispanics had the highest unintended birth rate (45 births/1,000 women) compared to other groups. One-fifth of US Hispanic women reside in Texas, yet unintended birth among Texas Hispanics has not been studied. The goal of this study was to describe the prevalence and characteristics of unintended birth in this population. Using data from Hispanic participants in the Texas Pregnancy Risk Assessment Monitoring System 2009–2010, we studied unintended birth in relation to demographic, lifestyle and partner characteristics. Adjusted prevalence odds ratios (POR) were computed for each characteristic and the analysis was stratified by maternal nativity (US vs foreign born). The weighted proportion of unintended birth was 49.5 % (CI = 45.9–52.6). In adjusted analyses, women aged 12–19 had a higher prevalence of unintended birth compared to ≥20 years (POR = 2.1, CI = 1.3–3.7). Unmarried (POR = 1.5, CI = 1.1–2.1), uninsured (POR = 1.7, CI = 1.2–2.3), and US-born (POR = 1.6, CI = 1.0–2.6) women had higher prevalence compared to married, insured and foreign-born women, respectively. Among US-born Hispanic women, higher prevalence of unintended birth was associated with being young, unmarried and experiencing psychological stressors within 12 months of giving birth; among foreign-born Hispanic women, higher prevalence was associated with lack of insurance. Efforts to reduce unintended birth in Texas might focus on young, single, uninsured and US-born Hispanic women. Analyses of other pre-pregnancy factors and health outcomes among Texas Hispanics could increase understanding of the differences we observed in unintended birth between US and foreign-born Hispanics.
Keywords: Unintended birth, Hispanic, Texas, PRAMS
Introduction
In the US, about half of all pregnancies are unintended and nearly half of these pregnancies end in abortion [1]. Unintended birth is a public health concern because women with an unintended pregnancy are less likely to seek timely prenatal care and are more likely to consume alcohol, tobacco or other drugs during pregnancy than women with an intended pregnancy [2, 3]. Smoking and drinking during pregnancy are associated with low birth weight, pre-term birth, stillbirth, Sudden Infant Death Syndrome and Fetal Alcohol Syndrome [4]. There is also evidence that unintended birth may adversely impact women’s education and career attainment [5–7]. Additionally, children resulting from unintended pregnancies may have poorer academic and health outcomes than children resulting from an intended pregnancy [3, 8, 9].
In the US, the prevalence of unintended pregnancy is highest in young, low-income, and minority women [1]. Among US states, Texas has one of the highest unintended pregnancy rates (62 pregnancies per 1,000 women aged 15–44 in 2006 [10] versus 52 pregnancies per 1,000 women nationally [1] ). Texas also has a higher fertility rate than the US overall (70/1,000 vs. 63/1,000) and has the fifth highest teenage (ages 15–19 years) fertility rate among US states [11]. In 2010, Hispanic women in Texas, who account for 20 % of all Hispanic births in the US, had higher fertility (86.8 births per 1,000 women aged 15–44) compared to non-Hispanic white (62.1 births per 1,000 women) and non-Hispanic black women (64.0 births per 1,000 women in Texas [11, 12].
Studies in Hispanic populations have found that cultural factors, including country of residence and country of birth, are associated with the prevalence of unintended birth [13, 14]. In a nationally representative study, prevalence was 27 % higher among US-born Hispanic women than foreign-born Hispanics [13]. In a study of women who gave birth in the Brownsville-Matamoros communities along the Texas-Mexico border, the proportion of unintended birth was 22 % more prevalent among US residents than among Mexican residents [14]. Furthermore, an analysis of US Hispanic adolescents from the National Longitudinal Survey of Youth found that nativity was strongly associated with contraceptive use and other reproductive health behaviors [15]. Little is known about the role of nativity in unintended birth prevalence among Texas Hispanics, however. Proximity to the US-Mexico border [16] and language [15, 17] are also associated with adverse reproductive health outcomes, such as preterm birth [16] and early sexual initiation [17]. However, neither border residence nor language has been well studied in relation to unintended birth. The current study examined known risk factors for unintended birth, such as age and marital status, as well as cultural factors that may be important in the Texas Hispanic population, including border residency, language and nativity.
Methods
Dataset and Sample
In this analysis of cross-sectional data we used the 2009 and 2010 Texas Pregnancy Risk Assessment Monitoring System (PRAMS) datasets [18]. The PRAMS system, funded in part by the Centers for Disease Control and Prevention (CDC), gathers state-specific, population-based data on maternal experiences and attitudes before, during, and after pregnancy [18]. A stratified sample of women who have had a recent live birth is drawn from each state’s birth certificate file every month and the women are surveyed two to four months after they have given birth. The data are collected by questionnaire or by telephone (for mail non-respondents) and from the birth certificate and are weighted for sample design, nonresponse, and non-coverage [19]. The combined 2009 and 2010 Texas PRAMS surveys, which had response rates of 67 and 65 % respectively, yielded a sample of 1,287 Hispanic women [18]. In 2010, Texas PRAMS included an oversample of births in the 32 counties closest to the US-Mexico border [18], enabling the examination of border region residence in relation to unintended birth. The study protocol was reviewed and approved by the Texas Department of State Health Services and the University of Texas School of Public Health.
Variables
The main outcome of interest, birth intention, was derived from the following PRAMS question: “Thinking back to just before you got pregnant with your new baby, how did you feel about becoming pregnant?” Possible responses were: “(a) I wanted to become pregnant sooner, (b) I wanted to become pregnant later, (c) I wanted to be pregnant then, or (d) I did not want to be pregnant then or at any time in the future”. Women who chose response option ‘b’ (mistimed) or ‘d’ (unwanted) were considered to have an unintended birth and those who chose ‘a’ or ‘c’ were considered to have an intended birth. Covariates used in this analysis included demographic, lifestyle and partner characteristics. We examined nine maternal demographic variables, including age-group (12–19, 20–29 or ≥30 years), educational attainment (<high school, high school, or >high school), household income (<$10,000/ year, $10,000–$24,999/year, or >$25,000/year), parity (1 or ≥2), border county residence (border or non-border), marital status at the time of birth (married or not married), insurance 1 month before pregnancy (private, public or none), questionnaire/interview language (English or Spanish), and nativity (US, Mexico or other); five lifestyle variables, including Body Mass Index (BMI) (underweight, normal, overweight, or obese), smoked 3 months before conception (yes or no), drank alcohol 3 months before conception (yes or no), psychological stressors experienced within 12 months before birth (0, 1–2, 3–5 or 6–13 stressors), and physical abuse experienced within 12 months before pregnancy (yes or no); and three partner characteristics, including ethnicity (Hispanic or non-Hispanic), nativity (US–born or foreign-born) and age relative to maternal age (<maternal age, <5 years older or ≥5 years older).
Border residence was defined as living in one of Texas’ 32 border counties versus one of the 222 non-border counties [12]. Private and military health insurance were classified as private insurance and Medicaid and Texas Health Steps as public insurance. Psychological stressors included the following events: family member ill, divorce, moved to new address, homelessness, partner lost a job, personal loss of job, frequent arguments with partner, partner said he did not want pregnancy, could not pay bills, experienced a physical fight, partner went to jail, someone close had a drug problem, and someone close died. Most variables had fewer than 3 % missing values. For variables with more than 3 % missing values we considered ‘missing’ as a separate analysis category.
In the adjusted analysis, several variables were collapsed into dichotomous variables or analyzed as a continuous variable. Because the proportions of unintended birth were similar among women aged 20–29 years and those aged ≥30 years, these two age categories were combined into a single age group (≥20 years). A history of experiencing psychological stressors within 12 months before birth was treated as a continuous variable in the model in order to observe how each additional stressor affected the women’s prevalence odds ratio for unintended birth. Nativity variables for women and their partners were recoded into dichotomous variables (US-born or foreign-born) due to small numbers of individuals born in countries other than the US or Mexico and similar proportions of unintended birth for Mexican and ‘Other’ nativity classifications.
Data Analysis
We calculated weighted percentages and 95 % CIs for each sample characteristic and determined statistical significance of each characteristic in relation to birth intention using a Chi square p value of <.05. Crude and adjusted prevalence odds ratios (POR) for unintended birth were calculated using multivariable logistic regression techniques. Adjusted models included all variables that were statistically significant in the bivariate analysis. To determine whether these associations differed by maternal nativity, we constructed separate models for US and foreign-born women. We conducted all analyses with SAS 9.3 and used the proc survey procedure to adjust for study design variables.
Results
About half of births to Hispanic women in Texas during 2009 and 2010 were unintended (Table 1). Among the unintended births, 80 % (502/625) resulted from mistimed pregnancies and 20 % (123/625) from unwanted pregnancies. More than one-third of the women had not graduated from high school, reported an annual household income below $10,000 and did not have health insurance prior to pregnancy. Among cultural characteristics, women’s nativity was split evenly between the US and elsewhere, with the majority of foreign-born women (84.5 %, weighted percent) born in Mexico. Nearly half of surveyed Hispanic women (46.1 %) completed the survey in Spanish and less than one-third of the women were border county residents. Experiencing at least one psychological stressor was reported by three of every four women in the 12 months before giving birth. The majority of the women’s partners were Hispanic and among those for whom partner nativity was known, close to half (44.3 %, data not shown) were born in Mexico.
Table 1.
Demographic, lifestyle and partner characteristics of hispanic women giving birth in Texas, PRAMS 2009–2010
| Characteristic | Frequency N = 1,287a | Weighted percent (95 % CI) |
|---|---|---|
| Demographic | ||
| Pregnancy intention | ||
| Intended | 653 | 50.5 (47.0–54.1) |
| Mistimed | 502 | 39.4 (36–42.9) |
| Unwanted | 123 | 10.0 (7.9–12.2) |
| Missing | 9 | |
| Age (years) | ||
| 12–19 | 214 | 15.9 (13.3–18.4) |
| 20–29 | 636 | 51.9 (48.4–55.4) |
| ≥30 | 437 | 32.2 (28.9–35.5) |
| Education | ||
| Less than high school degree | 538 | 40.5 (37.1–44.0) |
| High school degree | 392 | 33.6 (30.3–37.0) |
| More than high school degree | 357 | 25.8 (22.8–28.9) |
| Household income | ||
| < $10,000 | 463 | 36.2 (32.7–39.7) |
| $10,000–$24,999 | 397 | 35.6 (32.1–39.2) |
| > $25,000 | 314 | 28.18 (24.8–31.5) |
| Missing | 113 | |
| Parity | ||
| 1 | 450 | 33.0 (29.7–36.3) |
| ≥2 | 819 | 67.0 (63.7–70.3) |
| Missing | 18 | |
| Marital status at the time of birth | ||
| Married | 647 | 51.0 (47.4–54.5) |
| Not married | 640 | 49.0 (45.5–52.6) |
| Insurance 1 month before pregnancy | ||
| Private | 422 | 33.0 (29.7–36.4) |
| Public | 173 | 13.7 (11.3–16.2) |
| None | 665 | 53.3 (49.7–56.8) |
| Missing | 27 | |
| Nativity | ||
| United States | 667 | 49.3 (45.7–52.8) |
| Mexico | 538 | 42.9 (39.4–46.4) |
| Other | 82 | 7.8 (5.9–9.8) |
| Language used in survey | ||
| English | 712 | 53.9 (50.4–57.4) |
| Spanish | 575 | 46.1 (42.6–49.6) |
| Border county resident | ||
| Yes | 516 | 23.9 (21.4–26.5) |
| No | 771 | 76.1 (73.5–78.6) |
| Lifestyle | ||
| Smoking 3 months before pregnancy | ||
| Yes | 163 | 11.8 (9.5–14.1) |
| No | 1,108 | 88.2 (85.9–90.5) |
| Missing | 16 | |
| Drinking 3 months before pregnancy | ||
| Yes | 449 | 33.3 (29.9–36.6) |
| Characteristic | Frequency N = 1,287a | Weighted percent (95 % CI) |
| No | 823 | 66.8 (63.4–70.1) |
| Missing | 16 | |
| Body mass index (BMI) | ||
| Underweight | 43 | 3.3 (2.0–4.6) |
| Normal | 493 | 45.2 (41.4–49) |
| Overweight | 293 | 25.2 (21.9–28.4) |
| Obese | 281 | 26.3 (23.0–29.7) |
| Missing | 177 | |
| Psychological stress within 12 months before birth | ||
| Yes | 944 | 75.2 (72.2–78.3) |
| 1–2 stressors | 559 | 43.6 (40.1–47.1) |
| 3–5 stressors | 302 | 24.9 (21.8–28.0) |
| 6–13 stressors | 83 | 6.8 (5.0–8.6) |
| No | 322 | 24.8 (21.7–27.8) |
| Missing | 21 | |
| Physical abuse within 12 months before pregnancy | ||
| Yes | 70 | 5.2 (3.6–6.8) |
| No | 1,113 | 94.8 (93.2–96.4) |
| Missing | 104 | |
| Partner characteristics | ||
| Partner ethnicity | ||
| Hispanic | 1,058 | 83.9 (81.4–86.4) |
| Non-hispanic | 105 | 9.9 (7.7–12.1) |
| Missing | 115 | 6.2 (4.7–7.6) |
| Partner nativity | ||
| United States | 541 | 38.8 (35.4–42.2) |
| Mexico | 491 | 41.8 (38.3–45.3) |
| Other | 76 | 6.7 (4.9–8.5) |
| Missing | 179 | 12.8 (10.5–15.1) |
| Partner age in relation to maternal age | ||
| Younger or same age (0 years) | 856 | 68.6 (65.4–71.9) |
| 1–4 years older | 184 | 14.2 (11.7–16.6) |
| ≥5 years older | 60 | 4.4 (3.0–5.8) |
| Missing | 178 | 12.8 (10.5–15.2) |
Weighted frequency = 385,251
Demographic characteristics related to unintended birth included maternal age, marital status, insurance, language and nativity (Table 2). Unintended birth among teenage participants was more than 50 % [(70.7–45 %)/45 %] higher than unintended birth in older age groups. The proportion of unintended birth was nearly 40 % [(57.7–41.5 %)/41.5 %] higher among unmarried women than among married women. Women without health insurance prior to pregnancy and those who had public insurance prior to pregnancy had proportions of unintended birth 29 % [(52.4–40.6 %)/ 40.6 %] and 50 % [(60.7–40.6 %)/40.6 %] higher, respectively, than women who had private health insurance prior to pregnancy. Women who completed the survey in English had a higher proportion of unintended birth (54.8 %) than women who completed the survey in Spanish (43.3 %). Similarly, women who were born in the US had a higher prevalence of unintended births (57.7 %) than women who were born in Mexico or another country (41.6 and 40.9 %, respectively). No statistically significant difference was observed in the proportions of unintended birth between border and non-border residents.
Table 2.
Percent distribution of unintended birth among hispanic women in Texas, according to demographic, lifestyle and partner characteristics, PRAMS 2009–2010
| Characteristic | Unintended Birth Weighted %(CI) N = 625; 173,820 | p value |
|---|---|---|
| Demographic | ||
| Age (years) | ||
| 12–19 | 70.7 (62.9–78.8) | < .0001 |
| 20–29 | 46.9 (42.0–51.8) | |
| ≥30 | 43.1 (37.0–49.2) | |
| Education | ||
| Less than high school degree | 50.7 (45.3–56.2) | .7319 |
| HS degree | 49.7 (43.4–55.9) | |
| More than HS degree | 47.2 (40.3–54.1) | |
| Household income | ||
| < $10,000 | 53.6 (47.6–59.6) | .1288 |
| $10,000–$24,999 | 50.6 (44.4–56.8) | |
| > $25,000 | 42.6 (35.6–49.6) | |
| Missing | 50.0 (37.7–62.4) | |
| Parity | ||
| 1 | 50.4 (443–56.5) | .7494 |
| ≥2 | 49.1 (44.8–53.5) | |
| Marital status | ||
| Married | 41.5 (36.6–46.4) | < .0001 |
| Not married | 57.7– (52.8–62.7) | |
| Insurance 1 month before pregnancy | ||
| Private | 40.6 (34.5–46.7) | .0007 |
| Public | 60.7 (51.3–70.2) | |
| None | 52.4 (47.5–57.2) | |
| Nativity | ||
| United States | 57.7 (52.8–62.6) | < .0001 |
| Mexico | 41.6 (36.3–46.9) | |
| Other | 40.9 (28.3–53.6) | |
| Language used in survey | ||
| English | 54.8 (50.0–59.6) | .0014 |
| Spanish | 43.3 (38.1–48.4) | |
| Border county resident | ||
| Yes | 53.6 (47.3–59.9) | .1593 |
| No | 48.2 (44.0–52.4) | |
| Life style | ||
| Smoking (3 months < pregnancy) | ||
| Yes | 59.0 (49.0–69.1) | .0512 |
| No | 48.2 (44.4–52.0) | |
| Drinking (3 months < pregnancy) | ||
| Yes | 53.2 (47.2–59.3) | .1463 |
| No | 47.7 (43.3–52.0) | |
| Body mass index (BMI) | ||
| Underweight | 65.3 (46.4–84.1) | .1180 |
| Normal | 50.2 (44.5–55.9) | |
| Overweight | 42.1 (34.7–49.5) | |
| Obese | 53.6 (46.2–61.1) | |
| Missing | 48.8 (39.4–58.2) | |
| Psychological stress within 12 months before birth | ||
| 1–2 stressors | 48.7 (43.3–54.0) | .0001 |
| 3–5 stressors | 55.4 (48.3–62.5) | |
| 6–13 stressors | 71.1 (58.8–83.4) | |
| None | 39.2 (32.3–46.2) | |
| Physical abuse within 12 months before pregnancy | ||
| Yes | 54.2 (38.5–70) | .0037 |
| No | 47.4 (43.6–51.2) | |
| Missing | 68.9 (57.5–80.2) | |
| Partner characteristics | ||
| Partner ethnicity | ||
| Hispanic | 48.4 (44.6–52.3) | .0619 |
| Non-hispanic | 48.1 (36.5–59.8) | |
| Missing | 65.5 (53.4–77.7) | |
| Partner nativity | ||
| United States | 53.2 (47.6–58.8) | .0007 |
| Mexico | 43.1 (37.6–48.5) | |
| Other | 40.2 (26.7–53.7) | |
| Missing | 63.8 (54.4–73.1) | |
| Partner age in relation to maternal age | ||
| Younger or same age | 48.2 (43.9–52.5) | .0124 |
| 1–4 years older | 46.0 (36.7–55.3) | |
| ≥5 years older | 38.3 (22.2–54.4) | |
| Missing | 63.8 (54.4–73.1) | |
Participants who reported experiencing 1–2, 3–5 and 6–13 psychological stressors had a higher proportion of unintended birth (48.7, 55.4, and 71.1 %, respectively) than those who reported no psychologically stressful events (39.2 %). Women with missing information for abuse before pregnancy had significantly greater percentages of unintended birth (68.9 %) than women who reported not being abused (47.4 %). Similarly, women with missing partner nativity and women with missing partner age information had higher proportions of unintended birth than women with non-missing information for these variables. Because women with missing partner age information were the same women as those with missing partner nativity (data not shown), we excluded partner’s age difference from further analyses.
When all significant characteristics were considered simultaneously, only maternal age, marital status, insurance, maternal nativity and psychological stressors were significantly related to unintended birth (Table 3). Women who were 12–19 years old had an adjusted POR of 2.0 (CI = 1.2–3.5) compared to women 20 years or older. Unmarried as compared to married women (POR = 1.5, CI = 1.04–2.0) and women without insurance as compared to those with private insurance (POR = 1.8, CI = 1.2–2.5) also had significantly higher odds of unintended birth. US-born women had a POR of 1.6 (CI = 1.01–2.5) compared to foreign-born women. For every additional stressful event women experienced in the year prior to giving birth, the odds of having an unintended birth increased by 15 %.
Table 3.
Prevalence odds ratios (PORs) for unintended birth among hispanic women in Texas, according to important characteristics, PRAMS 2009–2010
| Characteristic | Adjusted PORa | 95 % CI |
|---|---|---|
| Demographic | ||
| Age 12–19 years (≥20 years = ref. group) | 2.0* | 1.2–3.5 |
| Marital status (married = ref. group) | 1.5* | 1.04–2.0 |
| Insurance-public (private = ref. group) | 1.3 | .8–2.2 |
| Insurance-none (private = ref. group)b | 1.8** | 1.2–2.5 |
| Language-English (Spanish = ref. group)c | 1.1 | .7–1.7 |
| Nativity-US-born (Foreign-born = ref. group) | 1.6* | 1.01–2.5 |
| Lifestyle | ||
| Psychological stressors (Continuous)d | 1.15** | 1.1–1.2 |
| Abuse-yes (no = ref. group)e | .8 | .6–2.6 |
| Abuse-missing (no = ref. group) | 1.3 | .4–1.6 |
| Partner | ||
| Partner’s nativity-US-born (Foreign-born = ref. group) | 1.4 | .8–2.4 |
| Partner’s nativity-missing (Foreign- born = ref. group) | 1.1 | .8–1.7 |
POR, prevalence odds ratio
Statistically significant at p <.05
Statistically significant at p < .01
Adjusted for all variables listed in table
Insurance 1 month before pregnancy
Language used in survey or interview
Experienced psychological stressor within 12 months before birth
Physically abused within 12 months before pregnancy
Among foreign-born women only insurance and paternal nativity were significantly related to unintended birth (Table 4). Foreign-born women without insurance had a POR of 2.3 (CI = 1.4–3.9) for unintended birth compared to foreign-born women with private insurance; those with public insurance also had an elevated POR (2.7), but it lacked statistical significance. Women for whom partner nativity was missing had a POR of 2.2 (CI = 1.1–4.4) compared to foreign-born women with a foreign-born partner. By contrast, among US-born women maternal age, marital status, and psychological stressors, were significantly associated with unintended birth. US-born women aged 12–19 years had more than twice the odds of unintended birth compared to women aged 20 years or older (POR = 2.3, CI = 1.1–4.8). Unmarried women had a POR of 1.9 (CI = 1.2–3.2) compared to women who were married at the time of birth. Finally, among US-born women, the odds of having an unintended birth increased 25 % for each psychologically stressful event experienced in the 12 months before birth.
Table 4.
Prevalence odds ratios (PORs) for unintended birth among Foreign-born and US-born, hispanic women in Texas according to important characteristics, PRAMS 2009–2010
| Characteristic | Foreign-born (N = 620)
|
US-born (N = 667)
|
||
|---|---|---|---|---|
| POR | 95 % CI | POR | 95 % CI | |
| Demographic | ||||
| Age 12–19 years (≥20 years = ref. group) | 1.5 | .6–3.6 | 2.3* | 1.1–4.8 |
| Marital status (married = ref. group) | 1.2 | .7–1.8 | 1.9** | 1.2–3.2 |
| Insurance- public (private = ref. group) | 2.7 | 1.0–7.6 | .8 | .4–2.0 |
| Insurance-none (private = ref. group)a | 2.3** | 1.4–3.9 | 1.4 | .8–2.4 |
| Language-English (Spanish = ref. group)b | 1.2 | .7–2.1 | .9 | .4–2.0 |
| Lifestyle | ||||
| Psychological stress (continuous)c | 1.054 | .9–1.2 | 1.25** | 1.1–1.4 |
| Abuse-yes (no = ref. group)d | 1.5 | .4–5.0 | .5 | .2–1.4 |
| Abuse-missing (no = ref. group) | .7 | .2–2.1 | 2.6 | .8–7.9 |
| Partner | ||||
| Partner’s nativity-US-born (Foreign-born = ref. group) | 1.5 | .8–3.0 | .8 | .5–1.4 |
| Partner’s nativity-Missing (Foreign-born = ref. group) | 2.2* | 1.1–4.4 | .9 | .4–1.9 |
POR, prevalence odds ratio
Statistically significant at P < .05
Statistically significant at P < .01
Insurance 1 month before pregnancy
Language used in survey or interview
Experienced psychological stressor within 12 months before birth
Physically abused within 12 months before pregnancy
Discussion
To our knowledge, this is the first study to describe unintended birth in the Texas Hispanic population. This study provides evidence that the likelihood of births resulting from unintended pregnancy among Texas Hispanics (50 %) is higher than in the US Hispanic population (43 %) [20]. (This difference may be underestimated since the national figure for unintended births among Hispanics does not exclude the state of Texas.) After taking known risk factors into account, births to US-born Hispanic women in Texas are 60 % more likely to be unintended than births to foreign-born Hispanic women.
This study’s findings provide information that may be useful for the design of interventions to prevent unintended pregnancy in Texas. Consistent with past studies of unintended pregnancy [1, 20] we found that young, single and uninsured Hispanic women are at greater risk of unintended birth than older, married and privately-insured women. Our finding that women without insurance have a greater prevalence of unintended birth than women with private insurance may indicate unmet need for family planning services in Texas, especially among adolescents, who are not eligible to receive certain family planning services from publicly funded centers without parental consent [21] and foreign-born women. Although the POR for unintended birth among women with public insurance did not reach statistical significance, few foreign-born women reported having public insurance (N = 35). Different from some past findings for US Hispanic and other women [1, 20], our study did not find education or income to be an important indicator of unintended birth in the Texas Hispanic population. However, large proportions of study women had less than a high school education and a household income <$15,000 and this disproportionality may have prevented us from observing differences associated with these two variables.
Consistent with past studies [22, 23], our study found a positive relationship between the number of recent stressful events experienced and unintended birth. Nevertheless, because our study and past studies used data that was collected retrospectively, it is impossible to know whether it is the experience of a psychologically stressful event that increases the odds of having an unintended birth or whether having an unintended birth increases the odds of reporting such an event. It is also possible that the women misreported the number of stressors they experienced. In addition, the 12-month reference period for the question that collects information about stressful events in PRAMS includes the entire period of pregnancy; some of the stressors reported almost certainly occurred after conception rather than before. To the extent that the experience of a stressor may have impacted prevalence of unintended birth in our study, the relationship appears to be limited to unintended birth in US-born mothers. If true, this difference between US-born and foreign-born women suggests that cultural or other factors not measured in this study may influence how women in different nativity groups cope with or report psychological stressors [24, 25].
Extensive literature on Hispanic populations demonstrates that cultural factors, including language and nativity, have important relationships with reproductive health outcomes [17, 26–30]. Since measures for language and nativity are available in PRAMS, we were able to examine these factors in relation to unintended birth. Although maternal nativity, paternal nativity and language were significant covariates in the bivariate analysis, only maternal nativity remained statistically significant when all three were considered simultaneously. These three variables were tested for collinearity using the variable inflation factor (VIF) and none had a VIF>2.0. In other studies [17, 26, 30] of reproductive health, language has remained a significant covariate even after adjustment for cultural factors such as nativity and generation status; however, some of these studies used more complex measures of language than the one used in this study. A nationally representative study conducted with mothers of a cohort of children born in 2001 [13] found that US born Hispanic women had a significantly higher proportion of unintended birth (52.5 %) than foreign-born Hispanic women (41.4 %). The difference between US-born and foreign-born women observed in that study was similar to the difference obtained in our analysis. The higher proportion of unintended birth among the US-born Hispanics in Texas compared to the unintended birth prevalence among US-born Hispanics nationally [13] may reflect state differences in socioeconomic factors, sex education or access to family planning services [31].
Unintended pregnancy or birth may not have the same meaning or implications for all racial/ethnic and cultural groups [24, 32] and this is supported by the variations in birth intention observed between US-born and foreign-born women in our study. Literature reveals that reproductive behaviors among Hispanic women are influenced by a variety of factors including religion [24], family values [33] and gender-roles, such as machismo and marianismo [34]. Some of these factors could influence reporting of birth intention and could explain the variations in unintended birth that we found between US-born and foreign-born Hispanics in Texas. Nativity may be a better proxy measurement for some of these other cultural factors than border proximity, which could explain why we observed no difference in unintended birth prevalence between border county residents and non-border county women. Furthermore, studies of the level of acculturation in Hispanic populations, commonly measured by language, nativity or generation status, have pointed to reproductive behaviors that could explain unintended birth prevalence differences between US-born and foreign-born Hispanic women. For example, findings have shown that while less acculturated Hispanic women are less likely to use contraception [17], they are also more likely to delay sexual initiation [15, 30] and less likely to have multiple sex partners [30] compared to more acculturated Hispanic women. Analyzing additional pre-pregnancy and reproductive behaviors and/ or health outcomes by birth intention in this population could enhance our understanding of the observed differences in unintended birth among US born and foreign-born Hispanics. Additionally, measures of pregnancy intention that include questions about pregnancy planning and a couple’s emotions before and during pregnancy could aid our understanding of the meaning of unintended pregnancy and unintended birth in different nativity groups [24, 35].
Our study findings are subject to some limitations. Because PRAMS women were interviewed 2–4 months after giving birth, the data are subject to recall bias and other threats to internal validity that are inherent in self-reported survey data. These threats are especially true for our main outcome of interest, unintended birth, because women who have had 2–4 months with their infant may be less likely to report that the baby was unwanted or mistimed, resulting in an underestimation of unintended birth and factors associated with it. The PRAMs data are also limited to pregnancies that have ended in the birth of a live-born infant and thus not representative of women with induced or spontaneous abortions or who deliver a stillborn infant. Despite these limitations, PRAMS data are an important resource and can be used to inform a broad range of state maternal and child health programs and policies [36]. Efforts to prevent unintended births and their consequences in Texas, including the incurrence of substantial economic costs [37], may increase efficiency by focusing on Hispanic women who are young, single, uninsured and born in the US.
Acknowledgments
Data were provided by the Texas Pregnancy Risk Assessment and Monitoring System (PRAMS), a project of the Texas Department of State Health Services (DSHS), and the Centers for Disease Control and Prevention (CDC). The authors would like to thank the Texas Department of State Health Services for their technical and statistical support. Additionally, this article is supported in part by the National Cancer Institute (NCI) through a Community Networks Program Center grant U54 CA153505. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC, NIH, or of TX DSHS.
Contributor Information
Denise Vasquez, Email: denisevasquez017@gmail.com, School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA.
Jill A. McDonald, Email: jillmcd@nmsu.edu, Division of Reproductive Health, National Center for Disease Prevention and Health Promotion, El Paso, TX, USA. College of Health and Social Services, New Mexico State University, Las Cruces, NM, USA
Nuria Homedes, Email: nuria.homedes@uth.tmc.edu, School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA.
Louis D. Brown, Email: louis.d.brown@uth.tmc.edu, School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA
References
- 1.Finer LB, Zolna MR. Unintended pregnancy in the United States: incidence and disparities, 2006. Contraception. 2011;84(5):478–485. doi: 10.1016/j.contraception.2011.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Altfeld SA, Handler A, Burton D, Berman L. Wantedness of pregnancy and prenatal health behaviors. Women and Health. 1997;26(4):29–43. doi: 10.1300/j013v26n04_03. [DOI] [PubMed] [Google Scholar]
- 3.Joyce TJ, Kaestner R, Korenman S. The effect of pregnancy intention on child development. Demography. 2000;37(1):83–94. [PubMed] [Google Scholar]
- 4.Center for Disease Control. [14 Sept 2012.];Unintended pregnancy prevention. 2012a http://www.cdc.gov/reproductivehealth/unintendedpregnancy/
- 5.Thomas A. Policy solutions for preventing unplanned pregnancy. Washington, DC: Brookings; 2012. [Google Scholar]
- 6.Bailey MJ. More power to the pill: The impact of contraceptive freedom on women’s life cycle labor supply. Quarterly Journal of Economics. 2006;121(1):289–320. [Google Scholar]
- 7.Goldin C, Katz LF. The power of the pill: Oral contraceptives and women’s career and marriage decisions. Journal of Political Economy. 2002;110(4):730–770. [Google Scholar]
- 8.David HP. Born unwanted, 35 years later: The Prague study. Health Matters. 2006;14(27):181–190. doi: 10.1016/S0968-8080(06)27219-7. [DOI] [PubMed] [Google Scholar]
- 9.Myrhmam A, Olsen P, Rantakallio P, Laara E. Does the wantedness of a pregnancy predict a child’s educational attainment? Family Planning Perspectives. 1995;27(3):116–119. [PubMed] [Google Scholar]
- 10.Finer LB, Kost K. Unintended pregnancies rates at the state level. Perspectives on Sexual and Reproductive Health. 2011;43(2):78–87. doi: 10.1363/4307811. [DOI] [PubMed] [Google Scholar]
- 11.Martin JA, Hamilton BE, Ventura SJ, Osterman MJK, Curtin SC, Matthews TJ. Births: Final data for 2012. National Vital Statistics Report. 2013;62(9):1–87. [PubMed] [Google Scholar]
- 12.Texas Department of State Health Services. [01 June 2013.];Health facts profile: Texas. 2010 http://www.dshs.state.tx.us/chs/datalist.shtm.
- 13.Guzman L, Wildsmith E, Manlove J, Franzetta K. Unintended births: Patterns by race and ethnicity and relationship type. Perspectives on Sexual and Reproductive Health. 2010;42(3):176–185. doi: 10.1363/4217610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gonzalez FGG, Mirchandani GG, McDonald JA, et al. Characteristics of young women who gave birth in the US-Mexico border region, 2005: The Brownsville-Matamoros sister city project for women’s health. Preventing Chronic Disease. 2008;5(4):1–14. [PMC free article] [PubMed] [Google Scholar]
- 15.McDonald JA, Manlove J, Ikramullah EN. Immigration measures and reproductive health among hispanic youth: Findings from the national longitudinal survey of youth, 1997–2003. Journal of Adolescent Health. 2009;44(1):14–24. doi: 10.1016/j.jadohealth.2008.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.McDonald JA, Mojarro Davila O, Sutton PD, Ventura SJ. A binational overview of reproductive health outcomes among US Hispanic and Mexican women in the border region. Preventing Chronic Disease. 2013;10:130019. doi: 10.5888/pcd10.130019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Afable-Munsuz A, Brindis CD. Acculturation and the sexual and reproductive health of Latino youth in the United States: A literature review. Perspectives on Sexual and Reproductive Health. 2006;38(4):208–219. doi: 10.1363/psrh.38.208.06. [DOI] [PubMed] [Google Scholar]
- 18.Texas Department of State Health Services. [01 May 01 2013.];Texas pregnancy risk monitoring system. TX DSHS. 2013 http://www.dshs.state.tx.us/mch/
- 19.Centers for Disease Control. [15 Oct 2012.];Pregnancy risk assessment monitoring system (PRAMS) 2012b http://ww.cdc.gov/prams/
- 20.Mosher WD, Jones J, Abma JC. Intended and unintended births in the United States: 1982–2010. National Vital Statistics Reports. 2012;55:1–28. [PubMed] [Google Scholar]
- 21.Texas Department of State Health Services. Texas Health Steps Provider Information. Texas: DSHS; 2014. [28 Oct 2014.]. http://www.dshs.state.tx.us/thsteps/providers.shtm. [Google Scholar]
- 22.Orr ST, Miller CA. Unintended pregnancy and the psychosocial well-being of pregnant women. Women’s Health Issues. 1997;7(1):38–46. doi: 10.1016/s1049-3867(96)00017-5. [DOI] [PubMed] [Google Scholar]
- 23.Maxson P, Miranda ML. Pregnancy intention, demographic differences, and psychosocial health. Journal of Women’s Health. 2011;20(8):1215–1223. doi: 10.1089/jwh.2010.2379. [DOI] [PubMed] [Google Scholar]
- 24.Hartnett CS. Are hispanic women happier about unintended births? Population Research and Policy Review. 2012;31:683–701. doi: 10.1007/s11113-012-9252-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sternthal MJ, Slopen N, Williams DR. Racial disparities in health: How much does stress really matter? Du Bois Review. 2011;8(1):95–113. doi: 10.1017/S1742058X11000087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ahluwalia IB, D’Angelo D, Morrow B, McDonald JA. Association between acculturation and breastfeeding among hispanic women: Data from the pregnancy risk assessment and monitoring system. Journal of Human Lactation. 2012;28:167. doi: 10.1177/0890334412438403. [DOI] [PubMed] [Google Scholar]
- 27.Kimbro RT, Lynch SM, McLanahan S. The influence of acculturation on breastfeeding initiation and duration for Mexican–Americans. Population Research and Policy Review. 2008;27:183–199. doi: 10.1007/s11113-007-9059-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Byrd TL, Balcazar H, Hummer RA. Acculturation and breastfeeding intention and practice in hispanic women on the U.S.-Mexico border. Ethnicity and Disease. 2001;11(1):72–79. [PubMed] [Google Scholar]
- 29.Robles JL, Lewis KL, Folger SG, Ruiz M, Gossman G, McDonald JA, et al. Prior contraceptive use among women who gave birth in the US-Mexico border region, 2005: The Brownsville-Matamoros sister city project for women’s health. Preventing Chronic Disease. 2008;5(4):1–19. [PMC free article] [PubMed] [Google Scholar]
- 30.Wingo PA, Kulkarni A, Borrud LG, McDonald JA, Villalobos SA, Green DC. Health disparities among Mexican American women aged 15–44 years: National Health and Nutrition Examination Survey, 1999–2004. American Journal of Public Health. 2009;99(7):1300–1307. doi: 10.2105/AJPH.2008.145169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Kaiser Family Foundation. State Health Facts. Kaiser Family Foundation; 2013. [20 June 2013.]. http://kff.org/statedata/ [Google Scholar]
- 32.Santelli J, Rochat R, Hatfield-Timajchy K, Gilbert BC, Curtis K, Cabral R, et al. The measurement and meaning of unintended pregnancy. Perspectives on Sexual and Reproductive Health. 2002;35(2):94–101. doi: 10.1363/3509403. [DOI] [PubMed] [Google Scholar]
- 33.Caal S, Guzman L, Berger A, et al. Because you’re on birth control, it automatically makes you promiscuous or something: Latina women’s perceptions of parental approval to use reproductive health care. Journal of Adolescent Health. 2013;53(5):617–622. doi: 10.1016/j.jadohealth.2013.05.003. [DOI] [PubMed] [Google Scholar]
- 34.Galanti GA. The hispanic family and male-female relationships: An overview. Journal of Transcultural Nursing. 2003;14(3):180–185. doi: 10.1177/1043659603014003004. [DOI] [PubMed] [Google Scholar]
- 35.Klerman LV. The intendedness of pregnancy: A concept in transition. Maternal and Child Health Journal. 2000;4(3):155–162. doi: 10.1023/a:1009534612388. [DOI] [PubMed] [Google Scholar]
- 36.Centers for Disease Control. PRAMS data to action success stories. Centers for Disease Control; 2013. [31 Oct 2013.]. http://www.cdc.gov/prams/DTA-SuccessStories.html. [Google Scholar]
- 37.Sonfield A, Kost K, Gold RB, Finer LB. The public costs of births resulting from unintended pregnancies: National and state-level estimates. Perspectives on Sexual and Reproductive Health. 2011;43(2):94–102. doi: 10.1363/4309411. [DOI] [PubMed] [Google Scholar]