Monthly Vital Statistics
December 1997, No. 31, Vol. 10

Unintended Births in Missouri

Unintended pregnancy continues to be a major problem in the United States both for the nation and for the individuals involved. Unintended pregnancy is associated with higher rates of inadequate prenatal care, low birth weight, and infant mortality as well as decreased life opportunities and heavier demands on public services.1 Unintended pregnancies include those which occur earlier than desired and pregnancies which are neither wanted at the time of occurrence nor at some future time. Respondents in a 1996 survey indicated that approximately 38 percent of all births in Missouri were unintended.

The information for this report was acquired from the August 1995 follow-back survey of mothers having live births. A random sample of 1500 mothers having live births in August 1995 was drawn from the pool of all Missouri resident live births surviving as of January 1996. Surveys were initially mailed in February 1996 with postcard and two additional follow-up survey mailings completed. Also, telephone follow-up was attempted for the non-responders. This effort resulted in a response rate of 64.3 percent (965 of 1,500) overall with corresponding rates of 68.2 and 44.9 percent for white and black/other races respectively. Table 1 shows the highest response rates were among married women, women 25 years of age and older, women with 13 or more years of schooling, whites and St. Louis County residents . The lowest response rates were noted for women with less than 12 years of education, and blacks.

In order to improve the precision of the sample estimators, post-stratification was done using the known population distributions by race (white/unknown versus black/other races) and education (less than 12, 12, and 13 or more years) for mothers having surviving August 1995 live births. This can possibly reduce non-response bias for differences in pregnancy intendedness or the other survey items between the respondents and non-respondents in each race-education stratum.

To determine type of contraception, respondents were asked what method(s) of birth control they were using in the 12 months prior to the conception of the baby born in August 1995. These methods were broadly grouped into over-the-counter (those methods requiring no contact with medical providers, i.e. condom, foam, rhythm, withdrawal, etc.), and physician-based (oral contraceptive, diaphragm, IUD, Norplant, Depo Provera). Respondents were then asked if they had stopped all contraception prior to pregnancy, and if they had stopped, whether they had stopped in order to conceive. Respondents were then asked to indicate their attitude toward pregnancy just prior to their pregnancy - wanted pregnancy at an earlier time as well as at that time, wanted pregnancy at that time, did not want pregnancy at that time but sometime in the future, or did not want to become pregnant at that time or any time in the future. These responses were collapsed into intended (wanted pregnancy at an earlier time as well as at that time plus pregnancy wanted at that time) and unintended (did not want pregnancy at that time but sometime in the future and did not want pregnancy at that time or any time in the future).

Table 2 presents the results of intention status-by selected characteristics. A majority of the births to unmarried women, women under the age of 20, women with less than 12 years of education, women who smoke during pregnancy, women who received Medicaid during their pregnancy and blacks were unintended. These groups of women also had lower response rates to the survey. Although the rate of unintended births is lower among certain groups of women (married women, women 25-34 years of age, women with 13 or more years of education), it is important to note that even among the groups with the lowest rate of unintended births, nearly one-quarter are not intended at the time of conception. The rate of unintended births is also higher among women who give birth to a low birth weight (<5.5 lbs.) infant. The relationship between unintended births and low birth weight is mitigated by controlling for race and Medicaid status.

The next table examines contraception use and intentionality. As Table 3 indicates the majority of women who have unintended births have used contraception in the year prior to the time of conception. The percent of women having unintended births is lower among women using physician-based contraceptives rather than over-the-counter contraceptives. However, the lowest percent of unintended births is among women using no contraception. This apparent contradiction is explained by the next part of the table which looks at intention status for those who stopped using all contraceptive methods. Of those women who indicated that they stopped using all birth control methods, three quarters indicated that they stopped because they wanted to become pregnant. Of these ninety-three percent indicated that they wanted a child at the time they became pregnant or had desired a child at an earlier time as well as at the time of conception. Not too unexpectedly, 83 percent of those who indicated they had not stopped contraception indicated that their births were unintended. The category of those who indicated that they had not stopped using contraception suggests the need for increased educational efforts on consistency in contraception.

The final comparison in Table 3 is between intention-status as measured by the use of vital record information and intention status as measured by the survey questions. There is a high degree of congruity but also some discrepancy. The vital record definition of unintended assumes that all births to females less than 18, out-of-wedlock births to women with less than a college education, and births within a year of a previous birth are unintended. Approximately 70 percent of both the intended and the unintended according to the vital records definition correlate with the corresponding survey results (intended vital records = intended survey). The lack of complete correspondence indicate the reality that all assumptions have exceptions and the difficulty of determining intention-status by a survey after the birth has occurred. The difficulty in attempting to measure intention-status is further indicated by the variance found by different studies such as the Missouri Maternal and Infant Health Survey2 and an analysis of Medicaid risk appraisals.

The Missouri Maternal and Infant Health Survey(MMIHS) covering births from December 1989-March 1991 found that over fifty-one percent of all births were unintended and that for blacks over three-fourths were unintended. The MMIHS and the August 95 follow-back survey used the same question to inquire about intention but the two studies not only were conducted in different time periods but also employed slightly different methodologies. The MMIHS interviewed some of the respondents in the hospital immediately after birth and thus had a higher response rates for blacks and women with less than a high school education. In addition an intensive follow-up was conducted for those subjects contacted by mail. The infant feeding survey was a sample of all births at one point in time with limited follow-up.

The risk appraisal for Medicaid asks of pregnant clients whether the pregnancy was intended or unintended, and if unintended, whether she was using or not using birth control. Seventy-six percent of the Medicaid clients matched to 1995 births indicated that their pregnancies were unintended compared to the sixty percent unintended found among Medicaid respondents to the infant feeding survey. The difference between these two results for a similar population (August 95 births versus all 1995 births for risk appraised mothers receiving Medicaid) indicate some of the problems with determining intention status. Intention was ascertained through different questions and at different times (after the birth versus during the pregnancy).

In this report we examined the data from a 1996 survey of mothers indicating that 38 percent of August 1995 births resulted from unintended pregnancies despite the use of contraception. The data from this study confirmed previous findings identifying those women at highest risk of unintended pregnancy (unmarried women, women under the age of 20, women with less than a high school education, poor women and black women). Nevertheless, even among those groups of women with the lowest risk for unintended pregnancy (married women, women 25-34, college-educated women) nearly a quarter of the pregnancies were unintended. This study hints at the complexity of the problem since the majority of women with unintended pregnancies report using contraception in the year prior to conception although nearly half had stopped using birth control prior to becoming pregnant. Further research is needed to clarify the conditions under which women do not want to become pregnant but do not consistently use contraception.

Data from the 1996 survey was also compared to other sources of information on pregnancy intention. Both the MMHIS and the 1995 Medicaid study of risk appraised women found higher rates of unintended pregnancy (51 percent and 76 percent respectively). As mentioned the MMHIS had a higher response rate from high risk groups and the Medicaid data was acquired during the pregnancy not after the birth. The differences in the studies illustrate a constant problem with health surveys: 1) those individuals with the problems you are interested in studying are the least likely to respond and 2) attitudes toward the event of interest (i.e. pregnancy) may change as the process continues. Despite the differences in overall rates all studies present the same relationships and indicate that unintended pregnancy continues to be an important problem for residents of Missouri.

1 Institute of Medicine. The Best Intentions: Unintended Pregnancy and the Well-Being of Children and Families. Washington, D.C. 1995.

2The Missouri Maternal and Infant Health Survey a joint project between the National Institute of Child Health and Human Development and the Missouri Department of Health to examine factors related to the prevention and subsequent health care needs of very low birth weight infants. See "Preliminary Results from the VLBW Study", Missouri Monthly Vital Statistics, State Center for Health Statistics 28:3 (May, 1994).


Table 1
Response Rate by Selected Characteristics: Missouri August 1995 Live Birth Follow-Back Survey
Response Response
Number PercentRate
Marital Status
Not Married241 25.049.6
Married724 75.071.4
Age
<20106 10.953.5
20-24244 25.355.4
25-34506 52.470.0
35 and older109 11.378.4
Education
Less than 12128 13.246.7
12335 34.762.4
13 or more495 51.373.2
Race
White854 88.568.2
Black100 10.444.4
All Other9 0.945.0
Resident County
St. Louis City72 7.554.6
St. Louis County173 17.969.8
Jackson114 11.858.2
Rest of Missouri606 62.865.6
Total965 100.064.3


Table 2
Intention Status* of Pregnancy by Selected Characteristics: Missouri August 1995 Live Birth Follow-Back Survey
Intended UnintendedUnknown Weighted
Marital StatusPercent 95% CIPercent 95%CIPercent 95%CIN
Not Married25.9 (20.7-31.1)69.2(63.7-74.7) 4.9(2.3-7.4)1841
Married71.9 (68.974.9)24.8(21.9-27.7) 3.3(2.1-4.5)4412
Age
Less than2025.6 (17.8-33.4)67.3(58.9-75.7) 7.2(2.6-11.8)878
20-2449.3 (43.5-55.1)49.5(43.7-55.3) 1.3(0.0-2.6)1636
25-3471.4 (67.8-75.0)25.6(22.1-29.1) 3.1(1.7-4.5)3089
35 and over 63.7 (55.5-71.9)27.3(19.7-34.9) 9.0(4.1-13.9)649
Education
Less than 1234.0 (26.2-41.7)59.6(51.6-67.6) 6.4(2.4-10.4)1189
1253.0 (48.1-57.9)42.3(37.447.2) 4.7(2.6-6.8)2208
13 or more 72.6 (69.0-76.2)25.4(21.9-28.9) 2.0(0.8-3.1)2856
Smoking During Pregnancy
Smoker46.6 (39.553.7)51.2(44.158.3) 2.2(0.14.3)1120
Non Smoker60.9 (57.864.0)34.9(31.937.9) 4.2(2.95.5)5082
Medicaid Status During Pregnancy
Medicaid36.7 (31.6-41.7)59.9(54.8-65.0) 3.3(1.4-5.2)2281
Non Medicaid71.7 (68.5-74.9)24.4(35.5-42.5) 3.9(2.5-5.3)3780
Low Birth Weight Birth
Low Birth Weight50.6 (38.7-62.5)49.4(37.561.3) 0.0415
Not Low Birth Weight58.9 (56.0-61.8)37.0(34.139.9) 4.1(2.9-5.3)5835
Race
Black 33.2 (24.5-41.9)59.5(50.4-68.6) 7.2(2.4-12.0)921
Total62.7 (59.7-65.6)34.1 58.4 (55.5-61.3) 37.9 (35.1-40.7) 3.8 (2.7-4.9) 6253 (32.0-37.8)3.2 (2.1-4.3)5331
*Intended births included births to women who responded that they wanted a child at the time they became pregnant or that they wanted a child at an earlier time as well as at the time they became pregnant; unintended births were not wanted at the time the woman became pregnant or at any time in the future.

Table 3
Intention Status* of Pregnancy by Contraceptive Use: Missouri August 1995 Live Birth Follow-Back Survey
Intended UnintendedUnknown
Type of Contraception

Before Pregnancy

Percent95% CI Percent95% CI Percent95%CI Weighted

N

Over -the counter151.4 (48.5-54.3)45.0 (42.1-47.9) 3.6 (2.5-4.7)2646
Physician-based260.2 (57.4-63.0) 38.8 (36.0-41.6)1.0 (0.4-1.5)2586
None 67.9 (65.2-70.6)28.8 (26.2-31.4)3.3(2.3-4.3) 1609
Contraception Status Before Pregnancy
Stopped contraception75.3 (72.8-77.8)23.7 (21.2-26.2)1.0 (0.4-1.6)4571
Wanted to become pregnant93.5 (91.6-94.6) 0.4 6.1 (4.7-7.5)(0.0-0.8) 3446
Not stopped 12.0 (10.1 -13.9)83.3 (81.1-85.5)4.6 (3.4-5.8)1443
Vital Records Definition of Intention3
Intended 72.3 (69.7-74.9)24.2 (21.7-26.7)3.5(2.4-4.6) 4375
Unintended 25.8 (23.3-28.3)69.7 (67.0-72.4)4.4 (3.2-5.6)1878
Total58.4(55.5-61.3) 37.9(35.1-40.7) 3.8(2.7-4.9) 6253
*Intended births included births to women who responded that they wanted a child at the time they became pregnant

or that they wanted a child at an earlier time as well as at the time they became pregnant; unintended births were not

wanted at the time the woman became pregnant or at any time in the future.

1Multiple methods may have been selected but were predominantly methods which did not require contact with medical

provider such as condoms, foam, rhythm, withdrawal.

2Multiple methods may have been selected but were predominantly methods which required a prescription by a

medical provider (oral contraceptive, diaphragm, IUD, Norplant or Depo Provera).

3Unintended included births to females less than 18, short spacing between this birth and previous birth (12 months or

less), and out-of-wedlock births to women with less than a college education.


Provisional Vital Statistics for October 1997
Live births decreased sharply in October as 5,920 Missouri babies were born compared with 7,858 one year earlier. Irregular reporting is the primary reason for this sharp decrease.

Cumulative births for the 10- and 12-month periods ending with October also show decreases. For January-October, births decreased by 1.4 percent from 63,081 to 62,233.

Deaths decreased slightly in October as 4,385 Missourians died compared with 4,599 in October 1996. However, cumulative deaths for the 10- and 12-month periods ending with October show slight increases.

The Natural increase for Missouri in October was 1,535 (5,920 births minus 4,385 deaths). Cumulative figures for the 10- and 12-month periods ending with October both show substantial reductions in natural increases.

Marriages increased for all three time periods shown below, while dissolutions of marriage increased slightly in October, but decreased for the cumulative time periods ending with October.

Infant deaths increased slightly for all three time periods shown below. The January-October infant death rate increased from 7.4 to 7.8 per 1,000 live births.

PROVISIONAL RESIDENT VITAL STATISTICS FOR THE STATE OF MISSOURI
July
Jan. July cumulative
12 months ending with July
Item
Number
Rate*
Number
Rate*
Number
Rate*
1996 19971996 19971996 19971996 19971996 19971995 19961997
Live Births7,858 5,920 17.313.4 63,081 62,23314.1 13.9 74,549 71,695 13.9 14.0 13.3
Deaths4,599 4,385 10.1 9.9 45,266 45,65010.1 10.2 54,25454,477 10.1 10.2 10.1
Natural increase 3,2591,535 7.2 3.517,815 16,583 4.03.7 20,295 17,218 3.8 3.83.2
Marriages4,209 3,971 9.39.0 38,140 37,0678.5 8.344,421 43,400 8.58.3 8.1
Dissolutions 2,1902,214 4.85.0 21,310 20,7594.8 4.6 25,453 24,887 4.9 4.84.6
Infant deaths 5458 6.99.8 468483 7.4 7.8 564 5817.7 7.68.1
Population base (in thousands) ...... 5,3595,395 ...... 5,3595,395 ...... 5,3125,343 5,389
*Rates for live births, deaths, natural increase, marriages and dissolutions are computed on the number per 1000 estimated population. The infant death rate is based on the number of infant deaths per 1000 live births. Rates are adjusted to account for varying lengths of monthly reporting periods.
AN EQUAL OPPORTUNITY/AFFIRMATIVE ACTION EMPLOYER Services provided on a nondiscriminatory basis.
Alternate forms of this publication for persons with disabilities may be obtained by contacting the Missouri Department of Health, Center for Health Information Management & Epidemiology/Bureau of Health Data Analysis, P.O. Box 570, Jefferson City, MO 65102; phone (573) 751-6278. Hearing impaired citizens telephone 1-800-735-2966.