Estimation of persons income eligible for the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) in 1989

              United States
Department of
Agrlcutture
F^xland
Consumer
Service
Office of
Analysis and
Evaluation
$
Estimation of Persons
Income Eligible for the
Special Supplemental
Nutrition Program for
Women, Infants and
Children (WIC) in 1989
Final Report
(y
Estimation of Persons Income Eligible for the
Special Supplemental Nutrition Program for
Women, Infants, and Children (WIC) in 1989
Prepared for the U.S. Department of Agriculture
Food and Consumer Service
Under Contract Number 53-3198-2-014
January 1996 (S ' Sigma One Corporation
Acknowledgements
The following persons served as Expert Reviewers of the work undertaken by Sigma One
Corporation throughout the Second WIC Eligibility Study. Their careful review and counsel
provided guidance in the development of analytical and statistical methods and in the
interpretation and presentation of the results of the WIC Eligibility Study. We are grateful for
their quick response to our inquiries and most importantly their dedication to relevance and
analytical rigor which helped the staff of Sigma One Corporation accurately estimate the size of
the population eligible for WIC.
Professor Barbara Abrams
Department of Nutrition, University of California at Berkeley
Professor Kathryn Anderson
Department of Economics, Vanderbilt University
Professor Jean Pierre Habicht
Division of Nutritional Sciences, Cornell University
Professor Frank E. Harrell, Jr.
Department of Community and Family Medicine, Division of Biometry
Duke University
Ms. Stefan Harvey
Center on Budget and Policy Priorities
Professor Betty Kirkley
Department of Nutrition, School of Public Health
University of North Carolina at Chapel Hill
Professor Milton Kotelchuck
Department of Maternal and Child Health, School of Public Health
University of North Carolina at Chapel Hill
Mr. Ed Lazere
Center on Budget and Policy Priorities
Ms. Alice Lenihan
North Carolina WIC Director
National Association of WIC Directors
Professor Chirayath Suchindran
Department of Biostatistics, School of Public Health
University of North Carolina at Chapel Hill
C*r
KSSI
Contents
Executive Summary iii
1. Introduction
WIC Eligibility 2
Analysis Objectives 3
Organization of Report 4
2. Estimates of WIC Income-Eligible Population in 1989 7
National Estimates g
State and County-Level Estimates 8
3. Data Profile and Estimation Methods 13
Analysis Overview 13
Data Sources 15
Census of Housing and Population IS
National Maternal and Infant Health Survey 17
Other Data Sources 18
Methods for Estimating the Population Income-Eligible for WIC in 1989 19
Infants and Children 19
Pregnant Women 22
Postpartum Women (Breastfeeding and Non-Breastfeeding) 25
4. Comparison of Income-Eligible Populations in 1979 and 1989 33
Infants and Children 33
Pregnant Women 36
Methodological Differences in Estimating Women's Categorical Groups 36
Economic and Demographic Changes 38
Postpartum Breastfeeding Women 39
Breastfeeding Patterns 39
Methodological Differences in Estimation of Breastfeeding Women 41
5. Estimating the Effects of Adjunct Eligibility 43
Methods of Estimating Adjunct Eligibility 45
Estimation Results 47
Appendices
A. Infant Mortality and Fetal Loss Rates (1988-1990) A-l
B Estimation of the WIC Categorical and Income-Eligible Population B-l
C. Breastfeeding Patterns in the United States in 1989 C-l
D. Effect of Medicaid Adjunct Eligibility at the State Level D-l
Tables
1. National Estimates of the Number of Women, Infants, and Children Who Were Income
Eligible for the WIC Program in 1989 9
2. State-level Estimates of the Average Monthly Number of Women, Infants and Children
Income Eligible for the WIC Program in 1989 11
3. Comparison of Estimates of Pregnant Women and Infants in 1979 and 1989 37
4. Comparison of Breastfeeding Patterns in the U.S. in 1980 and 1990 40
5. Effect of Medicaid Adjunct Eligibility on the Estimates of WIC Eligible Categories
in the U.S. and U.S. Territories (1989) 49
Figures
1. Distribution of WIC Income Eligibles by Category in 1989 10
2. Distribution of WIC Income Eligibles by Race/Ethnicity in 1989 10
3. Data Flow for Estimates of WIC Eligibility 14
4. Data Sources for Estimation of Women, Infants and Children Income Eligible for WIC
in 1989 16
5. Estimation of Infants and Children Income Eligible for the WIC Program in 1989 21
6. Computation of the Number of Pregnant Women Income Eligible for WIC Benefits
in 1989 24
7. Percent of Women in the U.S. Breastfeeding Their Infants in 1989 27
8. Breastfeeding Patterns of Women with Incomes Less Than 185 Percent of Poverty
in 1989 28
9. Computation of the Number of Postpartum Breastfeeding and Non-Breastfeeding
Women Income Eligible for WIC Benefits in 1989 30
10. Postpartum Women in Families with Incomes Less Than 185 Percent of Poverty
in 1989 31
11. Distribution of WIC Income Eligibles by Category in 1979 and 1989 34
12. Income Distribution of WIC Income Eligibles in 1979 and 1989 35
13. Comparison of Income Thresholds for the WIC and Medicaid Programs 45
••u
Executive Summary
This report presents results from the WIC Eligibility Study II, undertaken by Sigma One
Corporation to calculate the number of persons that met income and categorical eligibility
criteria for the WIC Program in 1989. These estimates of eligibility were required at the
national, state, and county level by the Child Nutrition Act, PL 101-147, which mandated that
the 1990 Census of Housing and Population be used to calculate the number of persons that
met income and categorical eligibility criteria for the WIC Program.
The estimation of WIC eligibility is a two-part process. First, the size of each of the five
categorically eligible groups-pregnant women, postpartum non-breastfeeding women,
postpartum breastfeeding women, infants (0 to 1 year of age), and children (1 -4 years) who
also met the income criterion for the WIC program was estimated. Second, the proportion of
WIC income-eligible persons likely to be at nutritional risk and thus eligible for the WIC
program is estimated using health survey data. This report presents the results of the first
phase of the analysis. The results of the second phase of the analysis will be presented in the
companion volume, Nutritional Risk Analysis and Estimation of Eligibility for the Special
Supplemental Nutrition Program for Women, Infants, and Children (WIC) in 1989.
Nationwide, on the basis of data from the 1990 Census, 9.3 million women, infants, and
children were eligible for the WIC program based on their income and categorical status. The
key findings of this analysis were:
• An estimated 8.96 million persons in the 50 states and the District of Columbia were
income eligible for WIC benefits in an average month of 1989. When estimates for
Puerto Rico, the Virgin Islands, and Guam are included, an estimated 9.3 million
persons were income eligible for WIC in 1989.
• The WIC income-eligible population represented 37 percent of the U.S. population of
pregnant, breastfeeding and non-breastfeeding postpartum women, infants, and children
at all income levels in 1989.
• More than one of three infants and children under five years of age in the United
States were income eligible for the WIC program in 1989. Infants and children
represented 80 percent of the WIC-eligible population.
• In 1989, pregnant women and postpartum non-breastfeeding and breastfeeding women
represented 20 percent of the income-eligible population. More than half of these
women were pregnant women.
• The white non-Hispanic group represented almost half of the income-eligible
population in 1989. One of four income-eligible persons was black non-Hispanic and
one of five income-eligible persons in 1989 was Hispanic.
• In 1989, more than half of the WIC income-eligible population lived in families with
incomes below the federal poverty level.
in
The estimates for this analysis were developed from counts in a special extract of the 1990
Census combined with survey data from the 1988 National Maternal and Infant Health
Survey (NMIHS), and natality and mortality data from Vital Statistics of the United States.
For infants and children, the estimates were based on direct counts from the Census data.
Since the Census does not identify pregnant, postpartum breastfeeding and non-breastfeeding
women, these categories were estimated by a methodology that used vital statistics and
NMIHS data in addition to Census data. The estimates of the average monthly number of
pregnant women were calculated by estimating the number of women pregnant for any
portion of 1989 adjusted for the portion of 1989 they were pregnant. The estimates for
postpartum breastfeeding and non-breastfeeding women were derived from counts of women
with their own infants and infants not with their own mothers. Estimates for all women's
categorical groups were adjusted for multiple births and fetal and infant deaths. The 1989
NMIHS data were used to develop estimates of breastfeeding duration by maternal age and
income level.
A comparison of the estimates from the WIC Eligibility Study I using 1979 Census data and
the results of the WIC Eligibility Study II using 1989 Census data indicated the following:
• The estimated number of persons income-eligible for the WIC program grew by 16
percent, from 7.7 million persons estimated in 1979 to 8.96 million persons in the 50
states and the District of Columbia in 1989.
• Although roughly one in five income-eligible persons were women in both 1979 and
1989, there were relatively more income-eligible pregnant women in 1989.
The differences in the estimated WIC income-eligible population from 1979 to 1989 resulted
from a variety of factors. For infants and children, the primary source of change was
population growth and changes in the economy. For women, the change was due in part to
economic and demographic changes and in part to methodological differences between the
1979 and 1989 estimates.
Under the Child Nutrition and WIC Reauthorization Act of 1989, persons eligible for Food
Stamps, Aid to Families with Dependent Children (AFDC), and Medicaid, as well as members
of families in which a pregnant woman or infant receives Medicaid are considered
automatically income eligible for the WIC program. This study also estimated the effects of
Medicaid adjunct eligibility on the WIC-eligible population to identify the number of
additional persons in each categorical group who would have been eligible for WIC in 1989
because they were Medicaid recipients. The results showed that had the adjunct eligibility
legislation been in effect in 1989, the additional pregnant women, infants, and children would
have represented from 1 to 2.7 percent of the WIC income-eligible population, overall,
depending on the state income cut-off values for Medicaid eligibility and the Medicaid
participation rates.
IV
1. Introduction
The Child Nutrition Act, PL 101-147, mandates that the 1990 Census of Housing and
Population be used to calculate the number of persons that met income and categorical
eligibility criteria for the Special Supplemental Nutrition Program for Women, Infants, and
Children (WIC) in 1989. These estimates of eligibility are required at the national, state, and
county levels. This study was carried out by Sigma One Corporation, under contract with the
Food and Consumer Service (FCS) of the U.S. Department of Agriculture (USDA), to meet
this requirement. Its primary purpose was to develop estimates of the number of persons that
were eligible for the WIC program as determined by the 1990 Census of Housing and
Population and other data. This report presents the principal findings and the methodology of
this study, known as the WIC Eligibility Study II.1 Complete national, state, and county-level
counts of persons income eligible for the WIC program can be found in the two volumes
of the Estimates of Persons Income Eligible for the Special Supplemental Food Program for
Women, Infants and Children (WIC) in 1989 published by the Office of Analysis and
Evaluation, USDA/FCS in August 1993.
Since 1974, when it was first authorized by Congress, the WIC program has provided
supplemental foods to women, infants, and children as part of preventive nutrition and health
services for low-income infants, young children, and pregnant and postpartum women.
Participation in the WIC program has greatly increased from the program's first year of
operation when it served about 88,000 participants per month to an average of 6.5 million
participants per month in 1994. The WIC program seeks to improve the health of participants
and prevent health problems during critical periods of growth and development by providing
selected nutritious foods such as milk and eggs, nutrition education, and access to social
services and such health care services as prenatal care. The WIC program is administered by
FCS. FCS distributes federal funds as grants-in-aid to the state health departments or
'This is the second time that the Decennial Census has been used as the basis for estimating the counts of persons
that met categorical and income criteria for the WIC program at the national, state, and county levels. USDA (1987)
published the results ofthe WIC Eligibility Study I in Estimation ofEligibilityfor the WIC Program, July 1987. That
study was also undertaken by Sigma One Corporation under contract with USDA and drew on data from the 1980
Census.
1
comparable agencies in each of the 50 states, the District of Columbia, Puerto Rico, the
Virgin Islands, and Guam. Local WIC agencies recruit participants and deliver program
services.
Unlike the Food Stamp Program, Aid to Families with Dependent Children (AFDC), and
Medicaid, WIC is not an entitlement program. It is a grant program which operates within
funding limits set annually by Congress. WIC program administrators, researchers, and policy
makers use the estimates of persons eligible for WIC to measure program coverage, identify
underserved areas, and assist in allocation of funds among states.
WIC Eligibility
Eligibility for the WIC program is based on three sets of criteria: categorical, income, and
nutritional risk. To participate in the WIC program, a person must meet all three sets of
criteria. The Child Nutrition Act limits participation in the WIC program to pregnant,
breastfeeding, and postpartum women; infants; and children. Participants must also meet
income qualifications and be certified to be nutritionally at risk.
To qualify for the WIC program, a person must fall into one of five categories. The
categorically eligible groups are pregnant women, women up to six months postpartum
who are not breastfeeding, breastfeeding women up to twelve months postpartum,
infants to age one, and children to age 5.
A participant must also be income eligible. The income for the household in which the
person resides must be at or below 185 percent of the poverty income guidelines
published annually by the Department of Health and Human Services (DHHS). These
poverty guidelines vary by family size.2 States may set lower standards corresponding
to the income limits used in their other health delivery programs, but no state may use
less thi.. 100 percent of these poverty guidelines as the income criterion for WIC
eligibility.
2For example, the Annual Poverty Income Guideline in effect between July 1, 1994, and June 30, 1995, for a
family of four was $14,800, and the corresponding WIC income limit (185 percent of poverty level) for this family
size was $27,380.
Individuals who are categorically and income eligible must also be at nutritional risk.
Nutritional risk is certified at local WIC clinics by a health professional who follows
guidelines or standards established by the state agency.
Among the many nutritional risk conditions that qualify a person for WIC eligibility are
anemia, poor weight gain during pregnancy, low birthweight of newborn, history of high-risk
pregnancies, and poor dietary patterns. The number and type of medical conditions identified
as nutritional risk criteria vary among states. In their Annual State Plans of Operation, WIC
state-level agencies specify the nutritional risk criteria.3
Categorically eligible persons receiving Food Stamps, AFDC, and families in which a
pregnant woman or infant receive Medicaid are automatically income eligible for the WIC
program. The automatic or adjunct eligibility status streamlines the WIC enrollment
procedure by making WIC income determinations unnecessary for people participating in
these other programs.
Analysis Objectives
For the WIC Eligibility Study II, Sigma One Corporation developed estimates of the number
of persons eligible to participate in the WIC program based on data from a special extract of
the 1990 Census of Housing and Population.4 The major objectives of the study were:
(1) to measure the size of the population as represented by the 1990 Census that
met WIC categorical and income criteria, and
(2) to estimate the number of these persons eligible for WIC on the basis of
nutritional risk.
'For a detailed analysis of how nutritional risk is certified at the state level, see the companion volume to this
report, Nutritional Risk Analysis and Estimation of Eligibility for the Special Supplemental Nutrition Program for
Women, Infants, and Children (WIC) in 1989 (forthcoming).
'The Bureau of the Census prepared a special extract file of the 1990 Census for FCS for use in the WIC
analysis. This Census Extract was based on the long form sample of the 1990 Decennial Census of Housing and
Population.
National, state, and county level estimates tor the United States (50 states and DC) were
calculated as well as estimates for Puerto Rico, the Virgin Islands, and Guam. Because the
1990 Census collected information on 1989 annual incomes, these estimates were for calendar
year 1989. This report presents the counts of the income-eligible persons in each category.
These estimates represent the average monthly number of persons in the five categorical
groups at or below 185 percent of poverty level in 1989. The estimates were calculated by
race/ethnicity, poverty level, and age.
The estimation of WIC eligibility is a two-part process. First, the size of each of the five
categorically eligible groups who also met the income criterion was estimated from counts in
the 1990 Census, combined with vital statistics and breastfeeding information. In the second
phase, the proportion of WIC income-eligible persons likely to be at nutritional risk and thus
eligible for the WIC program is estimated using health survey data.
This report is the second in a series of three publications which summarize the WIC
Eligibility Study II. In August 1993, FCS published Estimates of Persons Income-Eligible for
WIC in 1989, National, State and County Tables. This publication provided detailed estimates
of persons income-eligible for WIC, including estimates by race and ethnicity and by age.
This second report presents the methodology for developing these estimates, a summary of the
results, and a comparison of the estimates of the WIC income-eligible population in 1989 with
previous estimates developed based on the 1980 Decennial Census. The final report will
present estimates of the number of persons eligible for the WIC program in 1989, as well as
a detailed description of the methodology used for estimating the incidence of nutritional risk
among the WIC income-eligible population.
Organization of the Report
The remainder of this report is divided into three chapters. Chapter 2 presents the results of
the estimations of WIC categorical and income eligibility for 1989 at the national and state
levels. Chapter 3 describes the data sources used in the analysis and presents the methodology
used to estimate categorical and income eligibility. Chapter 4 compares the results of the 1980
WIC Eligibility Study I using data from the 1980 Census with 1990 Census data from the
current study. This chapter highlights the changes in income distribution, fertility patterns,
and breastfeeding patterns that have taken place in the postcensal period, 1979-1989, and
summarizes methodological differences between the two studies. Chapter 5 presents an
analysis of the impact of Medicaid adjunct eligibility on the WIC-eligible population.
2. Estimates of the WIC Income-Eligible Population in 1989
Nationwide, on the basis of data from the 1990 Census, 9.3 million women, infants, and
children were eligible for the WIC program based on their income and categorical status.5
Compared to FCS' 1979 WIC-eligibles estimate (which is based on the 1980 Decennial
Census), the estimated WIC income-eligible population increased by 16 percent.
The estimate of pregnant women, postpartum non-breastfeeding women, postpartum
breastfeeding women, infants (under 1 year of age), and children (1-4 years) at or below 185
percent of poverty represent the average monthly number of persons income eligible for the
WIC program. These estimates were undertaken at the county, state, and national level for
the 50 states and the District of Columbia. Estimates for Puerto Rico, the Virgin Islands, and
Guam are also included. This chapter presents the principal findings on the estimation of
WIC income-eligible persons in 1989. The chapter includes a summary of the number of
persons who were eligible for WIC benefits in an average month in 1989 at the national and
state level.
The key findings of the WIC Eligibility Study II are:
• An estimated 8.96 million persons in the 50 states and the District of Columbia were
income eligible for WIC benefits in an average month of 1989. When estimates for
Puerto Rico, the Virgin Islands, and Guam are included, an estimated 9.3 million
persons were income eligible for WIC in 1989.
• The WIC income-eligible population represented 37 percent of the U.S. population of
pregnant, breastfeeding and non-breastfeeding postpartum women, infants, and children
at all income levels in 1989.
• More than one of three infants and children under five years of age in the United
States were income eligible for the WIC program in 1989. Infants and children
represented 80 percent of the WIC-eligible population.
• In 1989, pregnant women and postpartum non-breastfeeding and breastfeeding women
represented 20 percent of the income-eligible population. More than half of these
women were pregnant women.
5This estimate includes the 50 states, District of Columbia, Puerto Rico, the Virgin Islands, and Guam.
The white non-Hispanic group represented almost half of the income-eligible
population in 1989. One of four income-eligible persons was black non-Hispanic and
one of five income-eligible persons in 1989 was Hispanic.
In 1989, more than half of the WIC income-eligible population lived in families with
incomes below the federal poverty level.
National Estimates
Table 1 presents a detailed enumeration of the 1989 WIC income-eligible population at the
national level by income/poverty level. These estimates are given for women, infants, and
children living in families with incomes below 75 percent of poverty, below 100 percent of
poverty, below 130 percent of poverty, and below 185 percent of poverty. In 1989, more than
half of the income-eligible population in the United States lived in families with incomes
below the federal poverty guideline.
Figure 1 shows that four out of five persons who were income eligible for WIC benefits in
the United States were infants and children under five years of age. Pregnant women
represented more than half of the women's categorical groups. In 1989 at the national level,
white non-Hispanics represented almost half of the WIC income-eligible population. One of
four income-eligible persons was black non-Hispanic and one of five income-eligible persons
was Hispanic. Figure 2 shows the racial/ethnic distribution of the WIC income-eligible
population.
State and County-Level Estimates
Table 2 presents estimates of the WIC income-eligible population in 1989 for the 50 states,
the District of Columbia, Puerto Rico, the Virgin Islands, and Guam. The state-level estimates
were estimated for six racial/ethnic groups and for the women's categorical groups, the
estimates were presented by maternal age. The estimations for each state and U.S. territory by
racial\ethnic group and maternal age are presented in Estimates of Persons Income Eligible
8
BLANK
PAGE
Table 1. Estimates of the Number of Women, Infants and Children Who Were Income Eligible
for the WIC Program in 1989
United States(50 States and D.C. and Puerto Rico, Virgin Islands and Guam)
(estimates in '000 persons)
Women's Categorical Groups
Income Pregnant Postpartum Postpartum
Level Women Non-Breastfeeding Breastfeeding
Infants Children All
Under 1 1 to4 WIC
year years Groups
vO
All Income
Levels 3,106 1353 991 4,020 15327 24,797
Below 75%
of poverty
Below 100%
of poverty
Below 130%
404
539
of poverty 709
Below 185%
of poverty 1,011
230
304
400
570
101
135
179
261
623 2332 3,690
825 3,105 4,908
1,093 4,098 6,749
1,561 5^03 9306
f
Children
1 Yr. 17%
Infants 17%
Children
2 Yr. 16%
Children
3 Yr. 15%
Children
4 Yr. 15%
Pregnant 55%
Breastfeeding 14%
Postpartum
Non Breastfeeding 31%
Figure 1. Distribution of WIC Income Eligibles by Category in 1989
Includes U.S. (50 states and DC) Puerto Rico, Virgin Islands & Guam
White
Non Hispanic
48%
Black
Non Hispanic
24%
•.v.wv
&$$:■
■:■;■;■:■'!■/
•;•;•:•:vi Hispanic
19%
Other
9%
Figure 2. Distribution of WIC Income Eligibles by Race/Ethnicity in 1989
Includes U.S. (50 states and DC), Puerto Rico, Virgin Islands & Guam
10
Table 2. Estimate* of the Number of Woman, Infants and Chlldran Incoma Eligible for the
WIC Program In 1989 [Humbert In Thousand*]
Stata Women Infanta
age:<1
Children
age: 1-4
All WIC
Pregnant Poatpartum Breastfeeding Groups
Women Non-
Breastfeeding
Year Yeara
Alabama 18.4 10.7 4.6 28.1 107.6 169.4
Alaska 3.3 1.9 0.9 5.1 18.5 29.7
Arizona 19.6 10.8 5.1 29.6 108.4 173.5
Arkansas 12.0 6.9 2.9 18.1 70.9 110.8
California 132.4 72.1 35.2 205.7 736.8 1,182.2
Colorado 12.3 7.1 3.3 19.0 75.2 116.9
Connecticut 6.6 3.8 1.8 10.3 40.0 62.5
Delaware 1.8 1.0 0.5 2.9 11.5 17.7
District of Columbia 3.1 1.7 0.8 4.5 13.4 23.5
Florida 47.2 27.0 12.4 74.1 280.2 440.9
Georgia 29.2 17.0 7.2 44.7 164.1 262.2
Hawaii 4.3 2.5 1.2 6.9 24.8 39.7
Idaho 5.0 2.9 1.4 7.8 30.9 48.0
Illinois 40.3 22.6 10.3 60.8 231.3 365.3
Indiana 18.5 11.0 4.9 29.5 116.0 179.9
Iowa 9.0 5.3 2.5 14.5 58.5 89.8
Kansas 9.0 5.3 2.4 14.2 57.1 88.0
Kentucky 16.9 9.7 4.1 25.7 99.2 155.6
Louisiana 25.6 14.2 6.3 37.4 144.6 228.1
Maine 3.4 2.0 0.9 5.7 24.9 36.9
Maryland 11.1 6.5 3.0 18.5 71.2 110.3
Massachusetts 14.1 7.9 3.8 21.6 85.7 133.1
Michigan 38.4 21.6 9.6 57.9 210.6 338.1
Minnesota 12.6 7.2 3.6 20.4 81.1 124.9
Mississippi 16.9 9.5 4.0 24.8 92.0 147.2
Missouri 19.5 11.2 5.0 30.0 118.0 183.7
Montana 3.5 2.0 1.0 5.5 22.8 34.8
Nebraska 5.9 3.4 1.7 9.4 37.2 57.6
Nevada 4.4 2.6 1.2 7.1 25.8 41.1
New Hampshire 2.0 1.2 0.6 3.4 14.6 21.8
New Jersey 16.3 9.3 4.4 25.6 98.8 154.4
New Mexico 10.0 5.5 2.6 15.0 55.3 88.4
New York 64.8 35.8 17.3 97.3 362.4 577.6
North Carolina 25.8 15.3 6.6 40.6 150.0 238.3
North Dakota 2.7 1.6 0.8 4.2 16.7 26.0
Ohio 40.2 22.9 10.2 61.4 238.4 373.1
Oklahoma 14.6 8.3 3.6 22.3 88.1 136.9
Oregon 10.8 6.2 2.9 17.2 65.9 103.0
Pennsylvania 34.7 20.1 9.3 55.2 220.7 340.0
Rhode Island 2.7 1.5 0.7 4.1 16.8 25.8
South Carolina 16.1 9.4 4.1 24.9 92.3 146.8
South Dakota 3.5 1.9 0.9 5.2 21.4 32.9
Tennessee 20.7 12.0 5.1 32.0 120.1 189.9
Texas 91.7 51.5 23.3 137.4 520.3 824.2
Utah 8.9 5.3 2.6 14.3 55.7 86.8
Vermont 1.5 0.9 0.5 2.5 10.7 16.1
Virginia 18.3 10.9 4.9 29.8 113.3 177.2
Washington 16.7 9.7 4.5 27.0 104.3 162.2
West Virginia 7.9 4.4 1.9 11.5 45.6 71.3
Wisconain 16.1 9.3 4.3 25.0 101.5 156.2
Wyoming 1.9 1.1 0.5 2.9 12.0 18.4
United Statee 972.0 561.4 252.7 1.498J 5,683.3 8.958.1
Puerto Rico 37.0 17.8 7.8 59.3 209.5 331.4
Virgin Islands 0.7 0.4 0.2 1.1 4.7 7.1
Guam 0.9 0.4 0-2 1.4 5.9 8.7
US. PR VI4 Guam M* 576.8 2e8J 1.36M 5JW3.4 93*53
11
for the Special Supplemental Nutrition Program for Women, Infants and Children (WIC) in
1989: National and State Tables.6 Estimates of persons income eligible for WIC benefits for
each county in the United States are presented in Estimates of Persons Income Eligible for
the Special Supplemental Food Program for Women, Infants and Children (WIC) in 1989:
County Tables.1 The estimates for each county which are presented in the USDA publication
include racial/ethnic detail for the population at or below 185 percent of poverty level.
Office of Analysis and Evaluation, Food and Consumer Service. USDA, August 1993.
Office of Analysis and Evaluation, Food and Consumer Service. USDA, August 1993.
12
3. Data Profile and Estimation Methods
The WIC Eligibility Study II was a multistep process that began with the estimation of the
number of persons that met categorical and income criteria for the WIC program. Once these
estimates were obtained, estimates of that population who were also at nutritional risk were
developed using nutritional risk criteria presented in the 1992 State Plans of Operation
compiled by the WIC state-level agencies. The study analyzed nutritional risk criteria that
were measurable in national survey data. Data from the 1988 National Maternal and Infant
Health Survey (NMIHS) and the Third National Health and Nutrition Examination Survey
(NHANES III) were used to assess nutritional risk for the WIC population. Specifically, these
data were used to estimate the proportion of persons likely to be at nutritional risk and thus
eligible for the WIC program on nutritional as well as categorical and income bases.
Analysis Overview
While this report primarily focuses on estimation of categorical and income criteria, Figure 3
and the steps below summarize the entire process for estimating the proportion of persons
meeting all WIC eligibility criteria in 1989.
Step 1. The number of categorically and income-eligible persons was determined
using a special extract from the 1990 Census, NMIHS data, and National
Center for Health Statistics vital statistics data. The output of this step was an
estimate of the number of pregnant, postpartum non-breastfeeding, and
breastfeeding women and infants and children who were income eligible for
WIC benefits in an average month in 1989 in each county and state in the
United States and at the national level.
Step 2. The nutritional risk criteria given in the 1992 State Plans were analyzed to
identify a common set of nutritional risk criteria used by the 50 states, DC,
and the territories.8 Sits of risk criteria were developed for each categorical
group, that is, for women, infants, and children.
'The analysis also included an assessment of any changes occurring in State Plans from 1992-1994. These
changes were reported as part of the 1994 WIC Participants Characteristics Study. Thus the nutritional risk criteria
analyzed reflect those risk criteria used to admit persons into the WIC program in 1994.
13
Figure 3. Data Flow for Estimates of WIC Eligibility
Step One
Task: Calculate number of
categorical/income eKgibies
in 1969.
Input 1990 Census Extract,
NMIHS, NCHS vital statistics.
Output Esama*J numb*
ofWIChcomaalgWspreansnt.
cresetfeedtog & pcetoartum
wornan, iwns a ensoran.
T
Step TWo
Task: Identify representative sets
of nutritional risk criteria for women,
infants & children.
Input: Nutritional risk criteria from
1992 Annual State Plans a 1994 WIC
Participants Characteristics Study.
Outoufc Sato of fiuttoonaf risk
ataria tor woman, Wansi a
cWdren.
■
Step Three Step Four
Task: Apply nutritional risk criteria
to survey data.
Input: Output from Stop 2; NMIHS,
NHANES III data.
Outoub Eslrnetod proportion of
VTCkMDR»elgfbtesat nuMonteri*
by age. reoe, ncorne/poveny aval
at Store i county leva* Conference
Dcunos IOT eaamans.
Task: Multiply estimates of WIC
target population By estimates of the
proportion of persons at nutritional risk.
Input: Output of Steps 1 4 3.
>1
Step 3. Using the risk criteria developed in Step 2, the NMIHS and NHANES III
survey data were analyzed to estimate the proportion of WIC income-eligible
persons likely to be at nutritional risk. The output of this step was the
estimated proportion of women, infants, and children in the U.S. population
who were at nutritional risk.
Step 4. The estimates of WIC income-eligible persons at the national and state levels
(Step 1) were multiplied by the estimated proportion of persons at nutritional
risk (Step 3) to calculate the number of persons eligible for WIC at the
national and state levels. The output of this step was the estimated number of
women, infants, and children in each state and at the national level who met
all WIC eligibility criteria in 1989.
Data Sources
Estimation of categorical and income eligibility for the WIC program was based on several
data sources. These sources included the special extract of the 1990 Census of Housing and
Population which was prepared for the FCS by the Bureau of the Census and survey data
from the 1988 National Maternal and Infant Health Survey. Additionally, natality and
mortality data from Vital Statistics of the United States prepared by the National Center for
Health Statistics (NCHS) were used to estimate the women's categorical groups. Figure 4
presents these sources and describes their contents.
Census of Housing and Population
The Bureau of the Census prepared a special extract of the 1990 Census of Housing and
Population for the FCS which contained the following state and county-level counts:
• Infants and children ages 1 to 5 by race/ethnicity and income level.
• Infants and children by race/ethnicity for whom income could not be
determined at the time of the Census.9
9 At the time the special Census Extract was prepared for FCS, income could not be determined for 59,355
infants and 238,712 children out of a total of 3.95 million infants and 15.06 million children.
15
Figure 4. Data Sources for Estimation of Women,
Infants and Children Income Eligible for WIC in 1989
1990
Decennial
Census of Housing
and Population
(Extract File)
Counts of
all women, infants
and children by income
level, race, ethnicity for
each state, county and
U.S. territory
Vital
Statistics
Series of
NCHS
1988
National
Maternal
and Infant
Health Survey
Infant
and fetal
death rates,
multiple
birth rates
Patterns of
breastfeeding
in the U.S.
in 1988
Data Set for Estimating
Persons Income Eligible for WIC
in 1989
16
• Infants not living with their natural mothers because they were adopted, in
foster care, or living with relatives. These counts were by race/ethnicity and
income level.
• Women who lived with their infants in the home by maternal age, race/
ethnicity, and income levels based on family size at the time of the Census.
• Women who lived with their infants in the home by maternal age,
race/ethnicity, and income levels based on family size prior to the birth of the
infant.
In the Census Extract, data were given for six racial/ethnic categories: non-Hispanic White;
non-Hispanic Black; non-Hispanic American Indian, Eskimo or Aleut; Asian or Pacific
Islander; Hispanic; and non-Hispanic Other. Five income levels were enumerated: persons
with incomes below 75 percent of poverty level, persons below 100 percent of poverty level,
persons below 130 percent of poverty level, persons below 185 percent of poverty level, and
persons at all income levels.10
To assist in the estimation of pregnant and postpartum women and to determine the number
of persons eligible for WIC because of adjunct eligibility with Medicaid, the Census Extract
provided two counts of women with infants. The first count was of the number of women
with infants who fell below 185 percent of poverty level at the time of the Census. The
second count of women living in household below a given poverty level was made by
simulating the family size prior to the birth of the infant, (i.e. by subtracting one from the
reported family size at the time of the Census).
National Maternal and Infant Health Survey
Because the Census does not identify pregnant, postpartum breastfeeding, and postpartum non-breastfeeding
women, these categories were estimated by methods that used vital statistics and
health survey data along with Census data. The 1988 National Maternal and Infant Health
"The Census Extract file defines poverty leve! as the average of the Annual Poverty Income guidelines in effect
between July 1, 1988 and June 30, 1989 (the first half of calendar year 1989) and the guidelines in effect between
July 1, 1989 and June 30, 1990 (the second half of calendar year 1989). For a family of four, the average annual
income poverty level for calendar year 1989 was SI 1,875 and the WIC income eligibility cutoff of 185 percent of
poverty was $21,969 per year. The levels for families living in Alaska and Hawaii were slightly higher.
17
Survey is a nationally representative, cross-sectional study of infant births and deaths
conducted by the National Center for Health Statistics. The survey examined factors such as
adequacy of prenatal care and maternal substance abuse that are related to poor pregnancy
outcomes. Participants were selected for the survey on the basis of information from vital
records. The mothers who participated in this survey responded to mailed questionnaires.
Respondents consisted of women who had live births, fetal deaths, or infant deaths. Hospitals
and delivery attendants also completed questionnai-es. The completed study provided social
and demographic information as well as data on maternal and infant health. It also included
information on prenatal care and health habits, delivery of the baby, hospitalizations before
and after delivery, previous and subsequent pregnancies, mother's and father's characteristics,
family income, and baby's health.
The National Maternal and Infant Health Survey was used to analyze patterns of breastfeeding
in the United States in 1989. Life table techniques were used to determine the proportion of
women with infants age 0 to 12 months who were breastfeeding their infants and the
proportion of women with infants aged 0 to 5 months who were not breastfeeding their
infants. The proportion of women who breastfed their infants was calculated on a monthly
basis.
Other Data Sources
Mortality and natality data from The Vital Statistics Series from the National Center for
Health Statistics were used in the estimation of pregnant and postpartum women to adjust
counts of mothers with infants who were living at the time of the Census for infant deaths and
fetal deaths." Mortality data were available by race for each state. They were not recorded
for different income levels. Appendix A describes the method used to match the racial/ethnic
11 See the NCHS, Vital Statistics of the United States, 1989. Vol II -- Mortality, Public Health Service,
Washington DC. U.S. Government Printing Office, 1993. See also the NCHS, 'Advance report of final mortality
statistics, 1990.' Monthly Vital Statistics Report, Vol. 41, No. 7, (Supplement). Hyattsville: MD: Public Hea th
Service. January 7,1993 as well as NCHS, Vital Statistics ofthe United States, 1989. Vol I- Natality, Public Health
Service, Washington DC. U.S. Government Printing Office, 1993.
18
groups presented in the National Center for Health Statistics publications with the racial/ethnic
groups enumerated in the Ctnsus Extract.
Methods for Estimating tk, Population Income Eligible for WIC in 1989
This section presents the methods which were developed to estimate the number of persons
who were income eligible for the WIC program based upon the 1990 Census. The estimation
of the size of the population which would be eligible on categorical and income grounds
alone requires computational and statistical procedures because pregnancy status and
breastfeeding are not directly enumerated in the Census data. This section explains how the
estimates of the income-eligible population were derived for the various WIC categorical
groups. Appendix B provides a more detailed presentation of this analysis.
Infants and Children
The counts of infants under one year of age and children 1 to 5 years of age were directly
enumerated by the Bureau of the Census. The Bureau of the Census provided counts of
infants and children in the United States by racial/ethnic group and income level for each
county, state, and territory in the United States. Approximately 1.5 percent of these infants
and children could not be assigned to an income category because their family income could
not be determined. The majority of these infants and children lived with other relatives or
were in foster care.12 The study assumed that all of these infants and children with
indeterminant incomes were income eligible for WIC. This assumption was made because
infants and children in foster homes and/or children living in group quarters are certified as
having no income at the time of income certification in the WIC clinic.13
l2In the case of infants, 87 percent of the infants whose family income could not be determined lived with their
relatives, with non-relatives, or were in foster care. The remaining 13 percent of these infants lived in group home
facilities.
"The counts of infants and children with indeterminant incomes in each racial/ethnic group were assigned
proportionately to the income category (less than 75 percent of poverty, 75-100 percent poverty, 100-130 percent
of poverty, 130-185 percent of poverty) based on the income distribution of infants and children with incomes for
the particular racial/ethnic group.
19
The Bureau of the Census found some problems with the accuracy of the ages reported in the
1990 Census. These problems were particularly acute for infants. Respondents tended to
report ages and family size as of the date they completed the questionnaires, rather than as of
April 1, 1990. Ages of infants in years may have been rounded up to age 1 to avoid
reporting an age of 0 years. The latter practice would underestimate the number of infants
and overestimate the number of children 1 year of age. The Bureau of the Census only
released adjusted counts of infants to FCS because of the size of the misclassification problem
in the infants counts and the importance of a reliable estimate of infants for the WIC program.
The counts of children were not adjusted by the Bureau of the Census.14
Figure 5 shows the method used to count the number of income-eligible infants and children
who lived in families with income less than 185 percent of the poverty level. The counts of
income-eligible children were computed for each year of age from 1 to 5. This estimation was
undertaken for each racial/ethnic group in each county, state, and territory of the United
States. In order to assess coverage for different income groups, these counts were presented
by income level for each geographic area: less than 75 percent of poverty, less than 100
percent of poverty, less than 130 percent of poverty, and less than 185 percent of poverty.
These estimates of income-eligible infants and children may differ slightly from other
published Census data because they:
(1) use income poverty guidelines established by DHHS rather than Census poverty
thresholds;
(2) use an adjusted count of infants developed by the Bureau of the Census to account for
misclassification of infants less than one year of age; and
(3) include infants and children for whom income could not be determined at the time of
the Census.
"For more information, see the Bureau of the Census, "Age, Sex, Race, and Hispanic Origin Information from
the 1990 Census: A Comparison of Census Results with Results Where Age and Race Have Been Modified," 1990
CPH-L-74, August 1991.
20
Figure 5. Estimation of Infants and Children
Income Eligible for the WIC Program in 1989
Census Counts of Infants
Who Were Members of
Families with Incomes
Less than 185%
of Poverty Levels
(1,439,411 infants)
+ Census Counts of Infants
Whose Income
Could Not Be Determined
(59355 infants)
1«4M»7W
Incorot-Ellfllbie
(0-11 month*)
Census Counts of Children
(Aged 1-5 Years) Who Were
Members of Families with
Incomes Less than 185%
of Poverty Levels
(5,444565 children)
+ Census Counts of Children
(1-5 Years) Whose Income
Could Not Be Determined
(238,712 children)
M83£77
Inootrw CHgfcle
ChNdwi
CMYM»)
21
Pregnant Women
The average number of pregnant women living in families with incomes below the WIC
income limit were calculated by estimating the total number of women who were pregnant in
1989 adjusted for the portion of 1989 they were pregnant. This adjustment was needed to
determine the number of pregnant women in an average month in 1989, rather than the total
number of women who were pregnant at any time in 1989.
Women who were pregnant in 1989 gave birth between January 1989 and September 1990.
The number of births during this period was estimated using Census counts of women living
with their own infants, infants not living with their mothers, and all infants. Separate
estimates were made of the number of births during three segments of this 21-month time
period:
Births from January 1989 to March 1989. Children born during this period were 12-14
months old at the time of the Census, and thus are included in the counts of children.
A good approximation of children 12-14 months of age is one fourth of the number of
one-year-olds. Because of the misclassification of ages in the infant/one-year-old
children categories in the Census, the number of children 12-14 months of age was
estimated as one-fourth the number of income-eligible infants.15 This estimate was
then adjusted for fetal and infant deaths in order to estimate conceptions from live
births, because a solely infant-based estimate would exclude those pregnancies that
resulted in a fetal or infant death. A final adjustment was made to account for multiple
births in order to avoid double counting mothers of twins, triplets, and other multiple
births.16
Births from April 1989 to March 1990. Births to income-eligible women during this
period were estimated as the sum of (1) the number of mothers living with their own
infants who were income eligible when calculated with a family size less than one'
and (2) the number of income-eligible infants not living with their own mothers. The
counts of mothers living with their own infants was adjusted to reflect infant mortality
and fetal deaths. The count of infants was also adjusted for infant and fetal deaths and
multiple births.
"This assumes a uniform distribution of births throughout 1989-1990.
"See Appendix A for additional details on the rates used
"The calculation of income eligibility was made with a family size one less than that at the time of the Census.
This was done in order to reflect family size during pregnancy. This was the customary practice in WIC clinics in
1989.
22
• Births from April 1990 through September 1990. Because children born during this
period were not yet born at the time of the Census, these births were estimated as one
half of the number of income-eligible infants given by the Bureau of the Census. This
estimate was then adjusted for fetal and infant deaths and multiple births to reflect
conceptions.
These estimates obtained by combining the above births to income-eligible women were then
adjusted by the average proportion of the year for which each group of women was pregnant.
This adjustment was required to estimate the number of pregnant women in an average month
in 1989. For mothers who gave birth in the first quarter of 1989 (January through March
1989), the average duration of pregnancy was 6.5 weeks, or 12.4 percent of the year. For the
mothers who gave birth between April 1989 and March 1990, the average number of weeks
that the mother was pregnant in 1989 was 31.1 weeks, or 59.7 percent of the year. For the
mothers who gave birth between April and September of 1990 (after the Census was taken),
the average duration of pregnancy was 13.6 weeks, or 26.1 percent of the year. These
calculations of the portion of a year that women were pregnant were used to estimate the
number of pregnant women in an average month in 1989. Table B2 in Appendix B presents
the derivation of these proportions of 1989 that the women were pregnant.
The number of income-eligible pregnant women was estimated for each racial/ethnic group in
each county, state, and territory of the United States. Estimates of pregnant women in an
average month of 1989 were computed for these age groups: women under 19 years of age,
women aged 19 to 26 years, women 27 to 35 years, and women 36 years and older. For each
age group, the racial/ethnic groups were then added to calculate the number of income-eligible
pregnant women in each county and state for different income levels.
Figure 6 illustrates the computation of the estimates of income-eligible pregnant women for a
given racial/ethnic group at the state level. The figure shows the computation of the number
of income-eligible pregnant women for black, non-Hispanic women in the state of New York.
This example is done at the state level for a specific racial group to emphasize that the
estimates were computed for each racial group within each state. The national estimates were
computed as the sum of the state-level estimates.
23
BLANK
PAGE
Figure 6. Computation of the Number of Pregnant Women
Income Eligible for WIC Benefits in 1989
(Black Non-Hispanic, New York State)
Census Counts of Mothers with Infants
whose Famity Income Was
Less than 185% Poverty
19,922 Mothers
Census Counts of Infants Not Living
with Their Mothers whose Family
Income Was Less than 185% Poverty
3,961 Infants
Adjusted For
Fetal & Infant
Deaths
1.034
Adjusted For
Fetal and Infant
Deaths
& Multiple Births
1.034-.014-1.019
Census Counts of Infants
with Family Income
Below 185% Poverty
28,519 Infants
Estimate of Children
Bom Between
January 1989
& March 1989
7,130 Infants
Estimate of Children
Bom Between
April 1990
& September 1990
14,260 Infants
Adjusted For Fetal & Infant Deaths
A Multiple Births
-1.018
Number of Women Who Gave Birth
Between April 1989 & March 1990
24,635 Women
Multiplied by
Percentage of Year
Women Were Pregnant
59.7%
Number of Women Who
Gave Birth Between
January 1989 & March 1989
7,258 Women
Multiplied by
Percentage of Year
Women Were Pregnant
12.4%
U Number of Women Who
Gave Birth Between
Aprl 1990 & September 1990
14,516 Women
Multiplied by
Percentage of Year
Women Were Pregnant
26.1%
Average Number of Pregnant Women
Who Gave Birth
Between April 1989 & March 1990
14,707 Women
Average Number of
Pregnant Women Who Gave
Birth Between
January 1989 & March1989
900 Women
Average Number of
Pregnant Women Who Gave
Birth Between
April 1990 4 September1990
3,789 Women
19^96 Pregnant Women Who Were
Income Eligible for WIC In 1989
df
It should be noted that the estimates of income-eligible pregnant women that are based on
births from April 1989-March 1990 are based on family size during pregnancy. The estimates
of income-eligible pregnant women that are based on births from January 1989-March 1989
and April 1990-September 1990 are based on family size after birth. This would tend to
increase the estimate of income-eligible pregnant women, since more women will be
considered income eligible with a larger family size. However, the impact of this effect on the
final estimate of income-eligible pregnant women is relatively small. Three-fourths of the
national estimate of income-eligible pregnant women is derived from the estimates of births
from April 1989-March 1990, which did reflect family size during pregnancy.
Postpartum Women (Breastfeeding and Non-Breastfeeding)
In the WIC program mothers with infants are categorically eligible for the WIC benefits for
the six months after the births of their infants whether they are breastfeeding their child or not
and for up to twelve months if they continue to breastfeed their infants during the 12 months.
The number of income-eligible postpartum (breastfeeding and non-breastfeeding) women was
estimated based on (1) counts of income-eligible women with their infants living in the
household and (2) counts of infants that did not live with their mothers. This second group
included infants living with the father or other relative, living with non-relatives, in foster
care, and in group quarters. This latter group is needed to estimate the number of postpartum
women who were not living with their infants. These infants who did not live with their
mothers would not necessarily have similar incomes to their mothers. Therefore, the number
of income-eligible postpartum women not living with their infants was estimated by assuming
that their income distribution resembled the income distribution of postpartum women with
infants at home for each racial/ethnic group within each state and county. In addition, a small
adjustment was made to include those women who experienced a fetal or infant death."
"Women whose pregnancies were terminated due to a fetal death are categorically eligible for WIC benefits as
non-breastfeeding postpartum women. This group can receive WIC benefits for up to six months following
termination of pregnancy. Women whose babies died during infancy are also eligible for WIC benefits. Depending
upon how old the infants were when they died, these mothers can receive benefits for up to twelve months.
25
The estimation of the WIC categorical groups of breastfeeding women and non-breastfeeding
women required that the counts of all postpartum women be adjusted by the proportions of
women who were likely to be breastfeeding for a given duration. Eligibility status of the
postpartum mother can change each month, depending upon the duration of breastfeeding. For
example, a woman who never breastfed her infant would be categorically eligible as a non-breastfeeding
postpartum woman. A woman who breastfed her infant for two months would
be categorically eligible as a breastfeeding woman for two months and as a non-breastfeeding
woman for four additional months. A woman who breastfed her infant for six months or
more would be categorically eligible as a breastfeeding woman until the month that she
stopped breastfeeding or the infant's first birthday. This movement from one categorical group
to another was taken into account in the estimation of the number of breastfeeding women
and postpartum, non-breastfeeding women.
Estimates of the probability that a woman would breastfeed her infant for a given period of
time were needed to measure the two postpartum categorical groups, (1) the breastfeeding
women up to twelve months postpartum, and (2) the non-breastfeeding women up to six
months postpartum. The 1988 National Maternal and Infant Health Survey was analyzed to
estimate the probability that a woman was breastfeeding in a given month. Demographic
analyses were used to estimate the probability that a mother never breastfed, and the
probabilities that she breastfed for one month, two months, three months, and so on, up
through the final month of the first year. These estimates of the probability that a woman
would breastfeed her infant were computed for different age/income groups in order to
account for differences in breastfeeding patterns among different socioeconomic groups of
mothers and for compatibility with the data which were available from the Bureau of the
Census.19
Figure 7 shows that the breastfeeding patterns of mothers in 1989 differed for women in
families with income less than 185 percent of poverty and women above 185 percent of
19 Appendix C contains more details of the analysis of breastfeeding patterns in the United States in 1989. Life
table techniques with censored data were used to estimate the proportion of women breastfeeding in each month after
birth. In this appendix, the breastfeeding patterns are also examined for different racial/ethnic groups.
26
Figure 7. Percent of Women in the U.S. Breastfeeding
Their Infants in 1989
70
60
50
40
30
20
10
63%
40%
Women Below
185 Percent
of Poverty Level
Women Above
185 Percent
of Poverty Level
Legend =
= Initiated Breastfeeding at Birth
= Breastfed for Less than 6 Months
fc^^ = Breastf©d for 6 Months or More
Source: Analyses of National Maternal and Infant Health Survey, 1988.
27
poverty. Women with incomes below 185 percent of poverty were less likely to initiate
breastfeeding and for those who did breastfeed their infant at birth, these women breastfed for
shorter duration than women above 185 percent of poverty. The breastfeeding patterns of
women also varied by maternal age within each income group. Figure 8 shows that for the
women in the WIC income-eligible group, women under 20 years of age were less likely to
breastfeed their infants than women over 27 years of age. These younger mothers also
breastfed for shorter duration than the mothers who were between 20 and 28 years of age.
The estimation of breastfeeding women and non-breastfeeding women who were income
eligible for WIC was undertaken by maternal age for each racial/ethnic group and income
category in each county and state in the United States. As an example, Figure 9 shows how
the estimates of breastfeeding and non-breastfeeding women were computed for white non-
Hispanic women in California. The state-level estimates of the number of breastfeeding and
non-breastfeeding women were formed by summing the estimates for each racial/ethnic group
within the state for each categorical group. The national estimates were computed as the sum
of the state-level estimates for each categorical group of women.
Figure 10 shows how significantly the duration of breastfeeding affects the categorical
eligibility of postpartum women. Of the 1.47 million postpartum women living in families
with incomes below 185 percent of poverty, approximately half were not categorically eligible
for WIC benefits because they were not breastfeeding their infants past six months of birth.
Breastfeeding women represented one-third of WIC income-eligible women and non-breastfeeding
women represented the remaining two-thirds of all income-eligible postpartum
women.
28
Figure 8. Breastfeeding Patterns of Women
with Incomes Less Than 185 Percent of Poverty in 1989
50 " o
40 0
0
30 o
0 o
20 0 o
10
o o 0 o
0
0 i 1 1 1 1 1 1 1 ' 1 1 1 1
1 2 3 4 5 6 7 8 9 10 t1 12
month after birth
■ under 20 years
■ 20-28 years o 27+ years
Data Source: National Maternal and Infant Health Survey, 1988.
29
BLANK
PAGE
Figure 9. Computation of the Number of Postpartum Breastfeeding and
Non-Breastfeeding Women Income Eligible For WIC Benefits in 1989
(White Non-Hispanic, California)
Census Counts of
Mothers with Infants
whose «tm»y Inoomt WM
kM« than 185% poverty
44,563 Mothers
Census Count of AD Infants Not
Living witti Their Mothers
15,968 Infants
Percent of Mothers With
Incomes Below 185% Poverty
14.8%
Census Counts of
Infants (0-1 Yr.)
with tamHy Income
below 185% poverty
49,481 Infanta
J
Census Counts of Infants
Not Living with Their Mothers
whose Mother's family income
was less than 185% poverty
2,366 Infanta
I
49,610 Live Births
Mothers Who Had
Single Births
2,316 Mothers
I
Mothers Who
Had Multiple Births
(twins, triplet*)
25 Mothers
I
46394
Postpartum Women with Living Infants
(23,447 Woman with Infanta 0-5 Months, 23,447 Woman With Infanta 6-11 Months)
Adjust for
Infant Mortality
Rate = 77/1300
Live Births
Multiple Birth
Rate = 212/1000
Births
Average Percentage of Mothers
with Infants 0-11 Months
Who Were
Breastfeeding their Infants*
18-2%
o
Average Percentage of Mothers
with Infants 0-5 Months
Who Were Not
Breastfeeding their Infants*
72.9%
Wte^l^eW
Woman Who Are
Postpartum
Non Breastfeeding Women
with Living Infants
17,105 Woman
I
Adjust for
Fetal Loss
Rate = 6.3/1,000
Live Births
Postpartum
Women Whose
Infant Died
378 Woman
Postpartum
Women Whose
Fetus Died
313 Women
TV98
WoflktMMM #-M
xxseHslte ferine **^**^^***^***- ^Tf*********^*~ew ^rf**f T'^'^T^**'
The percentage of mothers who were breastfeeding (not breastfeeding) was computed for separate age groups (this chart iBustrates the average breastfeeding rate only). £)
Figure 10. Postpartum Women in Families with Income
Less Than 185 Percent of Poverty in 1989
Non-Breastfeeding
Women with Infants
Age 6-12 Months
45.5%
u>
Non-Breastfeeding
Women with Infants
Age 0-5 Months
37.4%
WIC
Eligible
Breastfeeding
Women with Infants
Age 6-12 Months
4.6%
Non-Breastfeeding
Women
69%
Breastfeeding
Women with Infants
Age 0-5 Months
12.5%
Breastfeeding
Women
31%
i/
32
4. Comparisons of the Income-Eligible Population in 1979 and 1989
The estimated number of persons income-eligible for the WIC program grew by 16 percent,
from 7.7 million persons estimated in 1979 to 8.96 million persons in the 50 states and the
District of Columbia in 1989 based on estimates from Census data. Figure 11 shows that in
both 1979 and 1989, four out of five income-eligible persons were infants and children.
Although roughly one in five income-eligible persons were women in both 1979 and 1989,
there were relatively more income-eligible pregnant women in 1989.
The differences in the estimated WIC income-eligible population from 1979 to 1989 are due
to a variety of factors. For infants and children, the large majority of the income-eligible
population, the primary source of change was population growth and changes in the economy.
For women, the change was due in part to economic and demographic changes and in part to
methodological differences between the 1979 and 1989 estimates.
The income distribution of the population income-eligible for WIC changed between 1979 and
1989. Figure 12 shows that in 1989, a greater share of the income-eligible population lived in
families with incomes below the federal poverty level than in 1979, (i.e. 53 percent of
income-eligible persons had incomes below the poverty level compared to 46 percent in
1979).
Infants and Children
The estimated number of income-eligible infants increased by approximately 9.4 percent, from
1.37 million in 1979 to 1.5 million in 1989. Much of this change was due to a 13.6 percent
increase in the total number of infants during that period. This increase in births was offset
somewhat because a lower proportion of infants were below 185 percent of poverty in 1989
than in 1979 (38.0 percent versus 39.4 percent).
The estimated number of income-eligible children increased by approximately 15.2 percent,
from 4.93 million children in 1979 to 5.68 million in 1989. Again, most of this change was
33
Figure 11. Distribution of WIC Income Eligibles by Category in 1979 and 1989*
1979 1989
Children
64%
(4.93 million)
V£
Infants
18%
(137 million)
Children
63%
(5.68 million)
Pregnant
48%
(MmiHon)
Poetpaftum
BrMStfMdhga
Non-BrMSttotOng
Women
52%
(.71mi*on)
infants
17%
(1.50 million)
Women
20%
>w£«~C ££^ Breastfeednga
A£* 45%
-_ ^^ (.aOfrtBon)
Pregnant
55%
(.97mMon)
Poetpaftum
Income eligible persons = 7.7 million Income eligible persons = 8.9 million
•50 states and District of Columbia >i
Figure 12. Income Distribution of WIC Income Eligibles
in 1979 and 1989*
1979 1989
36%
18%
46%
30%
| < 100 percent of poverty level
f§§3 100-130 percent of poverty level
llllll 130-185 percent of poverty level
53%
'50 states and District of Columbia
3>J
due to the increase in the total number of children— 19.1 percent during that period. This
increase was also offset somewhat because a lower proportion of children were below 185
percent of poverty in 1989 than in 1979 (37.8 percent versus 39.0 percent).
Pregnant Women
The estimated number of income-eligible pregnant women increased by approximately 47
percent from 662,000 in 1979 to 972,000 in 1989. The estimated total number of pregnant
women at all income levels increased by nearly 20 percent. The estimated proportion of
pregnant women who were income eligible also rose substantially, from 26 percent to 32
percent. These differences are due to both the methodological differences in the estimation
and economic/dsmographic changes from 1979-1989.
Methodological Differences in the Estimation of Pregnant Women
Table 3 shows that the percentage increase in the number of pregnant women at all income
levels from 1979-1989 is substantially higher than the overall growth rate for infants (19.6
percent compared to 13.6 percent). However, the total number of pregnant women is directly
proportional to the total number of infants, suggesting that the total number of pregnant
women should grow at virtually the same rate as the total number of infants. The difference,
then, between the 1979 and 1989 estimates of the total number of pregnant women at all
income levels is primarily due to changes in the methodology used to estimate pregnant
women.
As described in Chapter 3, the 1989 estimate is based on Census counts of mothers living
with their infants and infants not living with their mothers. In the 1979 analysis, only counts
of mothers living with their infants were used. In the 1989 estimate, the addition of infants
not living with their mothers increases the estimated number of total pregnant women by
approximately five percent.
36
BLANK
PAGE
Table 3. Comparison of Estimates of Pregnant Women and Infants in 1979 and 1989
1989
Infants Pregnant
Percentage Women Percentage of
of Infants Pregnant Women
Number Who Were Number Who Were
'000s Income Eligible '000s Income Eligible
All incomes
< 185%
3,945
1,499
3,056
972
38% 32%
1979
All incomes
< 185%
3,473
1,370
2,555
662
39% 26%
*7
In addition, this methodological change also affects the estimated proportion of pregnant
women who were income eligible. Infants not living with their mothers were substantially
more likely to be below 185 percent of poverty than were infants living with their mothers
(48 percent versus 35 percent). Thus the addition of the infants who were not living with their
mothers into the estimate tended to increase the proportion of pregnant women who were
estimated as income eligible. This change accounts for approximately half of the increase in
the proportion of pregnant women who were estimated to be income eligible. The remaining
half of the increase is due to economic and demographic changes in the U.S. population.
Economic and Demographic Changes
The increase in the proportion of pregnant women who were income eligible in 1989 also
reflects changes in the economy and changing fertility patterns between 1979 and 1989. The
1989 analysis suggests that more women who were income eligible after the birth of their
infant were also income eligible before the birth of the infants. Demographic factors which
contribute to this change include the increase in births to unmarried women and Hispanic
women.
Fertility rates for unmarried women increased from 29.4 births per 1000 women in 1980 to
43.8 births per 1000 women in 1990. For unmarried teenagers the fertility rate increased from
27.6 per 1000 in 1980 to 42.5 per 1000 in 1990. In 1990, births to unmarried women
accounted for nearly 30 percent of all births. Households with unmarried mothers of young
children tend to be poorer than the households of married women with young children.20
Total fertility rates for Hispanic women are higher than the total fertility rates for women of
all racial and ethnic identities: 2900 per 1000 women versus 2626 per 1000 women.21
"According to the Annual Demographic File of the 1991 Current Population Survey of the U.S., 75 percent of
families with a mother of a child under 5 years of age who had never married lived in families below 185 percent
of poverty level compared to 33 percent of families with a mother of a child under 5 years of age who was married
or had been married in the past (i.e. separated, divorced, widowed).
2lThe total fertility rate is the number of births that 1,000 women would have in their lifetime if, at each year
of age, they experienced the birth rates occurring in the specified year.
38
Between 1980 and 1990, the share of Hispanics in the U.S. population rose from 6.4 percent
to 9 percent. This in part accounted for an overall increase in total fertility from a rate of
2266 in 1980 to 2626 in 1990. The poverty rate for Hispanics is twice the national average.
Thus the growth in the Hispanic population and the higher fertility rates of Hispanic women
also contributed to a significant increase in the proportion of income-eligible pregnant women.
Postpartum and Breastfeeding Women
The estimated number of income-eligible postpartum women (breastfeeding and non-breastfeeding)
increased by 13 percent from 1979 to 1989. There was a significant shift in the
proportion of women estimated as breastfeeding. In the 1979 estimate, nearly SO percent of
postpartum women were identified as breastfeeding. In 1989, this share fell to 31 percent.
Differences between the 1979 and 1989 estimates of postpartum breastfeeding and non-breastfeeding
women are due both to changes in breastfeeding practices and methodological
changes.
Breastfeeding Patterns
Table 4 compares the breastfeeding patterns of women between 1980 and 1989. This analysis
shows that as estimated from national survey data, the proportion of women who breastfed at
birth declined between 1980 and 1990 for women of all incomes (58 percent to 53 percent).
For the women below 185 percent of poverty guidelines, the decline in the proportion of
women who breastfed was larger. In 1980, women whose incomes were at or below 185
percent of poverty initiated breastfeeding 57 percent of the time. In 1989, this percentage
declined to 41 percent. This decline is probably attributable to various factors: increase in
mother's employment, increase in teenage pregnancies and unmarried mothers (young mothers
tend to not breastfeed), and increase in births to Hispanics and non-whites (who are also less
likely to breastfeed). Both in 1980 and 1989, the key determinants of the duration of
breastfeeding were maternal age, income, and race.
39
Table 4. Comparison of Breastfeeding patterns in the US between 1980 and 1990.
Initiated Breastfed for Breastfed for
Breastfeeding less than 6 months more than 6 months
at birth
1980:
All Incomes 58% 38% 20%
Women less than
185% poverty 57%
1990:
All Incomes 53% 32% 21%
Women less than
185% poverty 40% 26% 14%
40
Methodological Differences in the Estimation of Breastfeeding Women
In 1979, the WIC Eligibility Study I assumed that a woman was certified as breastfeeding at
the birth of her infant and then again at the six-month visit. This means that any woman who
stopped breastfeeding at any time during the first six months after birth was classified as
breastfeeding in the 1979 analysis. The 1989 analysis, on the other hand, captured actual
breastfeeding patterns rather than WIC certification practices. A woman who stopped
breastfeeding at any month was assigned to the non-breastfeeding woman's category at that
point. Since the proportion of women who breastfeed decreases each month, the number of
breastfeeding women declined and the number of non-breastfeeding women increased by the
same number as the decline. While this change in methods did not affect the total estimate of
postpartum women, it did affect the distribution between breastfeeding and non-breastfeeding
women in the postpartum category.
41
42
5. Estimating the Effects of Adjunct Eligibility
The Child Nutrition and WIC Reauthorization Act of 1989 required persons eligible for Food
Stamps, Aid to Families with Dependent Children (AFDC), and Medicaid, as well as members
of families in which a pregnant woman or infant receives Medicaid be considered
automatically income eligible for the WIC program. This automatic or adjunct eligibility
process streamlines the WIC enrollment procedure by making WIC income determinations
unnecessary for some applicants. As in the case of all income eligibles, adjunct eligible
persons must meet the categorical criteria and also be at nutritional risk as defined by WIC
for admission to the WIC program.
Income-eligibility levels for the Food Stamp program and AFDC are below those in the WIC
program. Infants, children under five years of age, pregnant women and postpartum women
who would be income eligible for AFDC or the Food Stamp program would be certified to be
income eligible for WIC benefits, because AFDC and the Food Stamp program have lower
income limits than the WIC program. These persons have already been accounted for in the
estimation of WIC eligibles.
Medicaid eligibility guidelines vary considerably across states. They also differ somewhat
from income eligibility guidelines in the WIC program and currently are at or above 185
percent of poverty in many states. In addition, the Medicaid income certification process
determines family incomes (as a percentage of poverty) for pregnant women by counting each
pregnant woman as two persons in determining family size for the income guideline
comparison. This is equivalent to determining income eligibility based on the family's income
after the birth of the infant. This process increases the family size of the pregnant woman,
which in turn increases the income limit to determine income eligibility for Medicaid. During
the period covered by the analysis, WIC determined income eligibility based on the family
size during pregnancy}2 This caused some persons to be adjunctly eligible for WIC
"In 1994, PL 103-48 changed the WIC eligibility standards to allow pregnant women who met the income
standard with a family size one larger than the current size to participate in the program. For pregnant women, this
makes WIC practice more compatible with Medicaid. However, this change only applies to pregnant women. Infants
and children living with a pregnant woman who only qualified for WIC at family size plus one would continue to
43
benefits at family incomes higher than the WIC standard of 185 percent of the poverty
income guideline. For example, in 1989 the annual income cutoff level for a family of three
with a pregnant woman in the household to be eligible for WIC was $21,474 whereas the
comparable Medicaid income threshold for the same family would be $25,829." Figure 13
compares the WIC and Medicaid income thresholds for families with pregnant women.
Unlike Food Stamps or AFDC adjunct eligibles, Medicaid-eligible persons are not included in
the estimation of WIC eligibles if their income fell between the WIC income threshold and
the Medicaid income threshold. They are not reflected in the estimates presented in this
report, or in FCS's August 1993 publications Estimates of Persons Income Eligible for WIC in
1989 (National, State and County Tables). Appendix D presents a methodology for estimating
these additional persons who would be eligible for WIC based on Medicaid adjunct eligibility
rules.
Adjunct eligibility procedures for the WIC program were not uniformly implemented until
after May 1990 and were not in effect throughout most of 1989. However, the effect of
Medicaid adjunct eligibility on the WIC-eligible population was measured by estimating the
potential size of the WIC-eligible population had the adjunct eligibility law been in effect in
1989. Because program changes have also taken place in Medicaid, the 1992 Medicaid
eligibility rules were used to simulate the potential size of the Medicaid adjunctly eligible
population. That is, the size of the adjunct-eligible population was estimated by applying the
income thresholds in effect in 1992 to the categorical groups who were eligible in 1989. This
analysis identified how many additional persons in each categorical group would have been
eligible for WIC in 1989 because they were Medicaid recipients.
Two scenarios were computed. The first scenario assumed that all states used 185 percent of
poverty as their Medicaid income limit. The second scenario was based on the 23 states that
be ineligible for WIC directly and could only participate through Medicaid adjunct eligibility. The number of infants
and children affected is small-approximately one to two percent of the total of children under five years of age.
23Assumes state uses 185 percent of poverty for Medicaid threshold.
44
Figure 13. Comparison of Income Thresholds for the WIC and Medicaid Programs
nn
2 4 6
Fami ly Size Cos defined by WiC program}
D WIC Program
+ Medicaid Program
used 185 percent of poverty as their Medicaid income limit in 1992.24
Methods of Estimating Medicaid Adjunct Eligibility
The differences in the definition of family size used by the WIC program and Medicaid
generate a group of pregnant women, infants, and children who live in families who would be
income eligible for Medicaid, but whose incomes are above the WIC income thresholds. The
breastfeeding postpartum women and non-breastfeeding postpartum women are not affected
for two reasons:
24In 1992, 28 states used an income threshold of at or below 155 percent of poverty for eligibility for pregnant
women. Persons at or below these levels would already be income eligible for WIC benefits, and in these states no
additiomtl persons would be admitted with incomes higher than WIC income thresholds. The remaining 23 states
had Medicaid thresholds equal to 185 percent of poverty. For this analysis, we measured the additional persons who
would be admitted into WIC because their incomes were between the WIC income threshold and the Medicaid
threshold.
45
(1) the family size of a postpartum woman is defined similarly for the WIC and
Medicaid programs, and
(2) postpartum women are not in and of themselves a categorical group for
Medicaid.
The WIC categorical groups which are affected by adjunct eligibility due to family size
differences alone are pregnant women and infants and children who live in households with
pregnant women.
The Census Extract contained two counts of women with their own infants:
• Women who lived with their infant in the home by maternal age, race/ethnicity,
and income levels based on family size prior to the birth of the infant.
• Women who lived with their infant in the home by maternal age, race/ethnicity,
and income levels based on family size after the birth of the infant (family size
at the time of the Census).
The number of the pregnant women adjunctly eligible for WIC benefits was estimated on an
incremental basis by comparing the number of women eligible at 185 percent poverty with
family size given as the family size in the Census to the number eligible using family size
during pregnancy (one less family member). The effect of Medicaid adjunct eligibility was
measured by the difference between these counts of women adjusted for participation in the
Medicaid program.
Members of families in which a pregnant woman receives Medicaid are also eligible for
automatic certification in the WIC program. If a pregnant woman receives Medicaid, the
infants and children under age 5 in her home are automatically eligible for WIC benefits. The
number of additional infants and children who were eligible for WIC benefits because their
family incomes fell between the WIC cutoff and the Medicaid cutoff was computed by
adjusting the number of additional pregnant women whose incomes fell between the Medicaid
46
limit and the WIC limit by the proportion of pregnant women who also had a child age 1
month to five years of age.25
Estimation Results
Two scenarios are presented to measure the effect of Medicaid adjunct eligibility. These
scenarios are based on different assumptions concerning the number of states that use 185
percent of poverty as their Medicaid income limit:
Scenario One: This scenario assumes that all states used 185 percent of poverty
as their Medicaid income limit. This scenario represents an upper bound for the
incremental proportion of persons who would be adjunctly eligible for the WIC
program.
Scenario Two: This scenario is based on the 23 states that used 185 percent of
poverty as their Medicaid income limit in 1992.
Both scenarios assume a 100 percent participation rate for Medicaid-eligible persons.
Table 5 presents the estimates of the additional pregnant women, infants, and children who
would have been eligible for WIC benefits in 1989 because their household's income was
above the WIC income limit but below the Medicaid income limit, as a result of counting
pregnant women as two persons in the Medicaid income determination. The results of the
first scenario indicate that if adjunct eligibility rules had been in effect in 1989 and all states
used 185 percent of poverty as their Medicaid income limit for pregnant women, the
incremental number of WIC eligibles would be 138,000 pregnant women, 16,000 infants, and
86,000 children, if all pregnant women who were eligible for Medicaid participated in the
To calculate the number of infants and children who would have become adjunctly eligible for WIC because
their mother was pregnant in 1989, we calculated the proportion of women with two or more children who had
infants or children ages 1 - 4 years. This calculation was made using national vital statistics data for 1990 as
provided by NCHS. Specifically using information about birth intervals for women with parity equal to or greater
than 2, we calculated the proportion of second or greater births that would fall in the birth interval such that the
existing child would have been under five years of age during some or all of the mother's pregnancy. This was
computed from vital statistics which count the number of births which are spaced within 10-20 months of each other,
given that a mother is pregnant for the second, third, etc. time. Eleven percent of births in 1990 with parity equal
to or greater than two were 10 to 20 months apart. These calculations allowed for the possibility of more than one
child in a household becoming adjunctly eligible as a result of his mother's pregnancy. The proportion of pregnant
women that had a child age 12-59 months in the home at any time during her pregnancy is .62 which was derived
from birth order and birth interval data from the vital statistics for 1990. Source: Table 19 of Advance Report on
Final Natality Statistics, 1990, Monthly Vital Statistics Report February 25, 1993.
47
program. The second scenario is based on the income cutoff limits used in 1992 by the
Medicaid program in each state. In this case, 23 states would have an increase in their WIC-eligible
pregnant women, infants and children due to adjunct eligibility. In these states,
74,000 additional pregnant women, 8,000 infants, and 46,000 children would have been
eligible for WIC benefits.
These scenarios are each an outer bound for the additional number of persons income eligible
for WIC because they use 100 percent participation in Medicaid. If persons eligible for
Medicaid do not participate in Medicaid, they would not become adjunctly eligible for WIC
The effect of Medicaid adjunct eligibility is therefore mediated by the participation rate in the
program. Since the WIC participation rate for pregnant women is likely to be higher than the
entire Medicaid population, the choice of a Medicaid participation rate for pregnant
women and their infants and children is a matter of judgment because we do not know
whether this group of pregnant women would be more or less likely to participate in
Medicaid. Appendix D presents alternative scenarios for estimating Medicaid adjunct
eligibility assuming various participation rates and state income limits for Medicaid.
This appendix also provides additional detail on the methodology used in the Medicaid
adjunct eligibility analysis.
In the United States in 1989, had the adjunct eligibility legislation been in effect, the
additional pregnant women, infants, and children would have represented from 1 to 2.7
percent of the WIC income-eligible population, overall, depending on the state income cut-off
values for Medicaid eligibility and the Medicaid participation rates. The outer bound for the
percentage increase in total WIC eligibles arising from all states using 185 percent of poverty
as the Medicaid income criterion and from 100 percent participation in Medicaid would not
exceed 2.7 percent of all income and categorically eligible persons.
48
BLANK
PAGE
Table 5. Effect of Medicaid Adjunct Eligibility on the Estimates of WIC Eligible Categories in the United States and US terrltories(l 989)*
Income-Eligible
Pregnant
Women
in 1989
Alternative Scenarios:
Participation in Medicaid Is 100% and
all states and territories used
185% of poverty for Medicaid Income:
United States (50 states and DC):
United States (50 states and DC) and
Puerto Rico, virgin Islands, and Guam:
Participation in Medicaid Is 100% and
23 slates used 185% of
poverty(1992 eligibility):
United States (50 states and DC):
United States (50 states and DC) and
Puerto Rico. Virgin Islands, and Guam:
972.0
1,010.6
972.0
1,010.6
Additional Income Eligible Persons by Category
Pregnant Women Percent
In Income Threshold Infanta Children Total Of WIC
Additional Income-Eligible
(%) Number Number Number Ellgibles Population in 1989
(•000) COOO) fOOO) C000)
14.2% 138.4 15.6 86.1
13.9% 140.2 15.8 87.3
7.6% 73.9
7.3% 73.9
8.3 46.0
8.3 46.0
240.0
243.2
128.2
128.2
2.7%
2.6%
1.4%
1.4%
Estimates Derived by assuming Alternative State-Level Income Criteria for Medicaid Eligibility and 100 % Medicaid Participation Rates
if
50
i
Appendix A
Infant Mortality and Fetal Loss Rates (1988-1990)
/•/
Appendix A. Infant Mortality and Fetal Loss Rates (1989-1990)
This appendix describes how the available National Center for Health Statistics (NCHS) vital
statistics data on infant mortality and fetal loss were used to produce infant mortality and fetal
loss estimates for the categories delineated in the 1990 Census Extract racial/ethnic categories.
Infant mortality and fetal loss rates were used to compute an estimate of the number of
pregnant and postpartum women in each state for the WIC Eligibility Study II. These rates
were needed at the same level of disaggregation given in the Census Extract on which the
WIC eligibility estimates are based. The Census Extract for the WIC Eligibility Study II is
tabulated into seven racial/ethnic categories: All Origins White, Non-Hispanic Black, Non-
Hispanic, Non-Hispanic American Indian, Eskimo, and Aleut, Non-Hispanic Asian and Pacific
Islander Hispanic and Non-Hispanic Other. The vital statistics published by NCHS are not
tabulated, for infant mortality and fetal loss, with these specific categories on a state-by-state
basis.' They are tabulated for the following racial groups:
• All races
• White
• Black
• Other races.
Prior to matching the vital statistics data with the Census data, several problems had to be
solved:
(1) The NCHS data available did not provide accurate state-level infant mortality
rate estimates for Hispanics as a separate category in seven states.
(2) The NCHS data available did not provide state-level estimates of infant
mortality for the Native American and Asian/Pacific Islander populations for
the years required by WIC Eligibility Study n. The data is available at the
national level.
(3) Fetal loss data for the relevant years were available only for the following
categories: All Origins, White, Other, and Black within Other
1 NCHS, The Vital Statistics of the United States. (Annual publication) Hyattsville, MD; Public Heath Service.
and NCHS's advance publication of final results in Monthly Vital Statistics Report, Hyattsville. MD Public Health
Service.
A-l
Furthermore, infant mortality rates for 1990 were only available at the all origins aggregation
and fetal loss data were not available for 1990.
The first problem, lack of estimates of infant mortality rates for Hispanics in seven states, was
solved by estimating the Hispanic rate by multiplying the state-specific infant mortality rate
for White and Non-Hispanics by the ratio of the national rate for Hispanics relative to the
national rate for the White ethnic category.2 This was deemed an adequate proxy given that
at the national level in 1989, the Hispanic infant mortality rate was only 5 percent higher than
the rate for White Non-Hispanics. This imputation implies that states with a higher than
average mortality rate for White infants would also have a higher than average mortality rate
for Hispanic infants. These estimates were then assigned to the seven states for which NCHS
did not report reliable infant mortality data for Hispanics. For the other 43 states and the
District of Columbia, it was possible to calculate a separate estimate for Hispanics as a whole
and to adjust the other rates accordingly.
The second problem, the aggregation of Asian and Pacific Islanders, Native Americans, and
Others into one residual category, was solved as follows:
(a) In those states where either the Asian/Pacific Islander group or the Native
American (American Indian, Eskimo and Aleut) group were predominant, the
infant mortality rate corresponding to the Other Non-Hispanic aggregate group
was applied to the predominant group and the national average was applied to
the non-predominant group. The rate for Non-Hispanic Other remained
unchanged.
(b) For those states in which no one group was predominant, the Non-Hispanic
Other rate was applied to all three groups.
A heuristic rule was used in the computation of which group was predominant. An ethnic
category was judged predominant within the Non-Hispanic Other aggregate when the 1990
population of women with own infants in this state represented 75 percent or more of the
1 According to the technical note in NCHS, Advanced Report of Final Mortality Statistics, 1989, Monthly Vital
Statistics Report; Vol 40. No. 8, Supplement 2. Hyattsvilk MD: Public Health Service 1992, these seven states for
1989 were Connecticut. Louisiana. Maryland, New Hampshire, Oklahoma, Rhode Island, and Virginia.
A-2
S3
total population of the Non-Hispanic Other Aggregate (Native Americans, Asian/Pacific
Islanders or Other) in the Census Extract.
This set of rules and computations make maximal use of the state-level data that are available
and preserve relationships that can be accurately estimated at the national level. The resulting
calculations are significantly better than imputing national averages in all cases. The use of
population weights within states insures that the state-by-state desegregation into ethnic
categories remains consistent with the state-level infant mortality rate for All Origins. The
eligibility estimates developed from these it.:zs are accurate at the state level and closely
patterned on the racial/ethnic mix within each state.
Fetal loss estimates were needed for 1988, 1989, and 1990 for the seven racial categories
available in the Census Extract. A simplified rule was used for distributing the published rates
into the specific ethnic/racial categories. First, using the population weights for Non-Hispanic
Black, and the sum of the population weights for Non-Hispanic American Indian, Eskimo and
Aleut, and Asian/Pacific Islander), we solved for the fetal loss rate for Non-Black Other. The
following equivalence table was then used to distribute the rates to the specific racial ethnic
groups:
NCHS Vital Statistics Groups WIC Eligibility Study II Census Groups
All Origins to All Origins.
White to Non-Hispanic White and Hispanic.
Black to Non-Hispanic Black.
Other (Non-Black) to Non-Hispanic American Indian, Eskimo, Aleut
and Non-Hispanic Asian Pacific Islander and
Non-Hispanic Other.
At the time the analysis was completed, the NCHS state-level infant mortality rate for 1990
was not available for each racial/ethnic group. The 1990 infant mortality rate for All Origins
was used to calculate a rate of change since 1989 in the overall state-level infant mortality
rate for each state. The racial/ethnic group specific rates for 1990 were estimated by applying
the overall rate of change since 1989 to the 1989 racial/ethnic specific rates. Likewise, in
A-3 M
order to extend the fetal mortality data to 1989, the 1988 fetal infant mortality data for All
Origins was used to calculate a rate of change in the overall state-level infant mortality rate
between 1989 and 1988. This percentage change at the state level was used to estimate the
1989 racial/ethnic specific rates.
Data for the U.S. territories of Guam, Puerto Rico, and the Virgin Islands were available for
1988. Since the Census data for the territories were not separated by racial groups, there was
no need to disaggregate the infant and fetal mortality data. In order to estimate the infant
mortality and fetal mortality rates for 1989, the change in the 1988 -1989 rates for all races
for the United States was applied to the 1988 rates for the territories. The resulting table of
infant mortality and fetal loss rates by racial/ethnic groups for both states and U.S. territories
follows.
A-4 sf
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
UNITED STATES
All Origins 9.8 9.2 7.5 7.5
White, Non-Hispanic 8.0 7.5 6.4 6.4
Black, Non-Hispanic 18.9 17.8 12.9 12.8
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.4 11.7 6.6 6.1
Asian & Pacific Islander, Non-Hispanic 6.2 5.8 6.6 6.1
Hispanic 8.5 7.9 6.4 6.4
Other, Non-Hispanic 4.8 4.5 6.6 6.1
Alabama
All Origins 12.1 10.2 10.4 10.0
White, Non-Hispanic 9.3 7.9 8.1 7.5
Black, Non-Hispanic 16.9 14.3 14.8 14.8
Am.Indian, Eskimo & Aleut, Non-Hispanic 6.1 5.2 9.5 6.8
Asian & Pacific Islander, Non-Hispanic 6.1 5.2 9.5 6.8
Hispanic 21.2 17.9 8.1 7.5
Other, Non-Hispanic 6.1 5.2 9.5 6.8
Alaska
All Origins 9.2 10.5 6.1 4.5
White, Non-Hispanic 6.7 7.6 4.4 3.3
Black, Non-Hispanic 9.5 10.9 9.3 6.6
Am.Indian, Eskimo ft Aleut, Non-Hispanic 15.2 17.4 9.3 6.6
Asian ft Pacific Islander, Non-Hispanic 6.2 7.1 9.3 6.6
Hispanic 3.1 3.6 9.3 3.3
Other, Non-Hispanic 15.2 17.4 9.3 6.6
Arizona
All Origins 9.2 8.8 6.4 5.4
White, Non-Hispanic 8.5 8.1 6.2 5.2
Black, Non-Hispanic 21.1 20.1 10.4 10.5
Am.Indian. Eskimo A Aleut. Non-Hispanic 8.8 8.4 6.8 5.0
Asian & Pacific Islander, Non-Hispanic 6.2 5.9 6.8 5.0
Hispanic 9.5 9.1 6.2 5.2
Other, Non-Hispanic 8.8 8.4 6.8 5.0
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 & 1989" Tables.
A-5
#
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract
Racial/
Ethnic
Origin
Arkansas
All Origins
White, Non-Hispanic
Black, Non-Hispanic
Am.Indian, Eskimo & Aleut, Non-Hispanic
Asian A Pacific Islander, Non-Hispanic
Hispanic
Other, Non-Hispanic
California
All Origins
White, Non-Hispanic
Black, Non-Hispanic
Am.Indian, Eskimo & Aleut, Non-Hispanic
Asian A Pacific Islander, Non-Hispanic
Hispanic
Other, Non-Hispanic
Colorado
All Origins
White, Non-Hispanic
Black, Non-Hispanic
Am.Indian, Eskimo A Aleut, Non-Hispanic
Asian A Pacific Islander, Non-Hispanic
Hispanic
Other, Non-Hispanic
Connecticut
All Origins
White, Non-Hispanic
Black, Non-Hispanic
Am.Indian, Eskimo A Aleut, Non-Hispanic
Asian A Pacific Islander. Non-Hispanic
Hispanic
Other, Non-Hispanic
1989 1990
Infant Infant
Mortality Mortality
Rate Rate
10.2 9.2
8.3 7.5
15.3 13.8
3.4 3.1
3.4 3.1
3.1 2.8
3.4 3.1
8.5 7.9
7.7 7.1
18.8 17.4
12.4 11.5
6.2 5.7
7.6 7.1
6.2 5.7
8.7 8.8
8.0 8.0
16.2 16.3
5.6 5.7
5.6 5.7
10.5 10.6
5.6 5.7
8.8 7.9
7.1 6.4
20.2 18.1
8.8 7.9
8.8 7.9
7.5 6.7
8.8 7.9
1988
Fetal
Loss
Rate
8.4
6.6
14.2
4.1
4.1
6.6
4.1
6.8
6.2
12.1
6.1
6.1
6.2
6.1
7.6
7.2
13.0
9.2
9.2
7.2
9.2
6.8
5.9
12.0
8.9
8.9
5.9
8.9
1989
Fetal
Loss
Rate
7.4
6.2
10.6
15.6
15.6
6.2
15.6
6.7
6.3
11.7
5.9
5.9
6.3
5.9
7.2
6.8
11.6
9.8
9.8
6.8
9.8
7.3
6.6
12.1
6.4
6.4
6.6
6.4
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 A 1989" Tables.
A-6
S7
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin
Delaware
All Origins 11.8 10.1 6.1 7.5
White, Non-Hispanic 8.7 7.4 5.5 6.2
Black, Non-Hispanic 18.3 15.6 8.3 11.5
Am.lndian, Eskimo ft Aleut, Non-Hispanic 6.1 5.2 2.5 6.7
Asian ft Pacific Islander Non-Hispanic 6.1 5.2 2.5 6.7
Hispanic 12.5 10.7 5.5 6.2
Other, Non-Hispanic 6.1 5.2 2.5 6.7
District of Columbia
All Origins 22.9 20.7 12.7 13.5
White, Non-Hispanic 14.4 13.0 10.2 10.8
Black, Non-Hispanic 25.7 23.3 14.2 15.1
Am.lndian, Eskimo ft Aleut, Non-Hispanic 12.4 11.2 13.2 14.0
Asian ft Pacific Islander, NOB -Hispanic 0.0 0.0 13.2 14.0
Hispanic 6.8 6.2 10.2 10.8
Other, Non-Hispanic 0.0 0.0 13.2 14.0
Florida
All Origins 9.8 9.6 8.6 8.3
White, Non-Hispanic 7.9 7.7 6.7 6.7
Black, Non-Hispanic 15.3 15.0 14.6 13.4
Am.lndian, Eskimo ft Aleut, Non-Hispanic 12.4 12.1 8.2 9.1
Asian ft Pacific Islander, Non-Hispanic 5.2 5.1 8.2 9.1
Hispanic 9.9 9.6 6.7 6.7
Other, Non-Hispanic 5.2 5.1 8.2 9.1
Georgia
All Origins 12.3 12.4 11.3 11.0
White, Non-Hispanic 9.1 9.1 8.4 7.5
Black, Non-Hispanic 18.5 18.6 16.6 17.1
Am.lndian, Eskimo ft Aleut, Non-Hispanic 12.4 12.5 8.2 4.5
Asian ft Pacific Islander, Non-Hispanic 6.1 6.2 8.2 4.5
Hispanic 6.7 6.7 8.4 7.5
Other, Non-Hispanic 6.1 6.2 8.2 4.5
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 ft 1989" Tables.
A-7 Si
Table Al. WIC Eligibility Study
Race/Hispanic Origin Specific Infant Mortality and Fetal Mortality Rates
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Hawaii
All Origins 8.3 6.7 9.0 6.9
White, Non-Hispanic 3.0 4.0 10.7 8.2
Black, Non-Hispanic 10.6 8.6 9.0 6.5
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.4 10.1 8.4 6.5
Asian A Pacific Islander, Non-Hispanic 8.9 7.2 8.4 6.5
Hispanic 12.8 10.4 10.7 8.2
Other, Non-Hispanic 8.9 7.2 8.4 6.5
Idaho
All Origins 9.7 8.7 6.2 7.1
White, Non-Hispanic 9.1 8.2 6.2 7.1
Black, Non-Hispanic 20.4 18.3 6.2 7.1
Am.Indian, Eskimo A Aleut, Non-Hispanic 13.0 11.7 6.2 7.1
Asian & Pacific Inlander, Non-Hispanic 13.0 11.7 6.2 7.1
Hispanic 14.4 13.0 6.2 7.1
Other. Non-Hispanic 13.0 11.7 6.2 7.1
Illinois
All Origins 11.7 10.7 7.4 7.6
White, Non-Hispanic 8.6 7.8 6.1 6.2
Black, Non-Hispanic 21.9 20.0 11.8 12.3
Am.Indian, Eskimo A Aleut, Non-Hispanic 12.4 11.3 6.9 5.2
Asian A Pacific blander, Non-Hispanic 4.4 4.0 6.9 5.2
Hispanic 10.1 9.2 6.1 6.2
Other, Non-Hispanic 4.4 4.0 6.9 5.2
Indiana
All Origins 10.2 9.6 7.3 7.5
White, Non-Hispanic 9.0 8.5 6.8 7.0
Black, Non-Hispanic 18.8 17.6 11.8 11.6
Am.Indian, Eskimo A Aleut, Non-Hispanic 12.4 11.6 2.8 4.6
Asian A Pacific Islander, Non-Hispanic 4.4 4.1 2.8 4.6
Hispanic 10.1 9.4 6.8 7.0
Other, Non-Hispanic 4.4 4.1 2.8 4.6
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 A 1989" Tables.
Jf
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Groups
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Iowa
All Origins 8.3 8.1 6.7 6.3
White, Non-Hispanic 7.7 7.6 6.6 5.9
Black, Non-Hispanic 22.1 21.6 6.7 15.6
Am.Indian, Eskimo ft Aleut Non-Hispanic 13.7 13.4 6.7 12.2
Asian ft Pacific Islander, Non-Hispanic 13.7 13.4 6.7 12.2
Hispanic 10.5 10.3 6.7 5.9
Other, Non-Hispanic 13.7 13.4 6.7 12.2
Kansas
All Origins 8.8 8.4 5.8 6.1
White, Non-Hispanic 7.6 7.3 5.5 5.5
Black, Non-Hispanic 15.6 14.9 9.9 12.4
Am.Indian, Eskimo ft Aleut, Non-Hispanic 8.6 8.2 3.4 4.1
Asian ft Pacific Islander, Non-Hispanic 8.6 8.2 3.4 4.1
Hispanic 7.6 7.3 5.5 5.5
Other, Non-Hispanic 8.6 8.2 3.4 4.1
Kentucky
All Origins 9.2 8.5 8.6 7.9
White, Non-Hispanic 8.2 7.5 8.1 7.4
Black, Non-Hispanic 16.2 14.9 12.9 12.7
Am.Indian, Eskimo ft Aleut, Non-Hispanic 12.4 11.4 4.0 6.3
Asian ft Pacific Islander, Non-Hispanic 8.5 7.9 4.0 6.3
Hispanic 7.1 6.5 8.1 7.4
Other, Non-Hispanic 8.5 7.9 4.0 6.3
Louisiana
All Origins 11.4 11.1 8.2 8.3
White, Non-Hispanic 8.5 8.3 6.7 6.2
Black, Non-Hispanic 16.0 15.6 10.5 11.5
Am.Indian, Eskimo ft Aleut, Non-Hispanic 1.7 1.7 3.4 2.0
Asian A Pacific Islander, Non-Hispanic 1.7 1.7 3.4 2.0
Hispanic 8.9 8.7 6.7 6.2
Other, Non-Hispanic 1.7 1.7 3.4 2.0
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 ft 1989" Tables.
A-9 ID
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Maine
All Origins 7.4 6.2 6.8 5.6
White, Non-Hispanic 6.5 5.4 6.1 5.5
Black, Non-Hispanic 0.0 0.0 6.8 5.6
Am.Indian, Eskimo ft Aleut, Non-Hispanic 16.7 14.0 6.8 5.6
Asian ft Pacific Islander, Non-Hispanic 16.7 14.0 6.8 5.6
Hispanic 8.7 7.3 6.8 5.5
Other, Non-Hispanic 16.7 14.0 6.8 5.6
Maryland
All Origins 10.3 9.5 7.4 7.0
White, Non-Hispanic 8.1 7.5 5.5 4.9
Black, Non-Hispanic 15.9 14.7 11.5 11.7
Am.Indian, Eskimo ft Aleut, Non-Hispanic 5.0 4.7 6.1 2.7
Asian ft Pacific Islander, Non-Hispanic 5.0 4.7 6.1 2.7
Hispanic 8.5 7.8 5.5 4.9
Other, Non-Hispanic 5.0 4.7 6.1 2.7
Massachusetts
All Origins 7.7 7.0 6.3 5.8
White, Non-Hispanic 6.7 6.1 5.7 5.3
Black, Non-Hispanic 18.0 16.4 12.7 10.2
Am.Indian, Eskimo ft Aleut, Non-Hispanic 12.4 11.3 5.7 6.1
Asian ft Pacific Islander, Non-Hispanic 4.7 4.2 5.7 6.1
Hispanic 8.4 7.6 5.7 5.3
Other, Non-Hispanic 4.7 4.2 5.7 6.1
Michigan
All Origins 11.1 10.7 5.2 5.7
White, Non-Hispanic 8.2 7.9 4.3 4.9
Black, Non-Hispanic 22.9 22.1 9.3 9.2
Am.Indian. Eskimo A Aleut, Non-Hispanic 9.6 9.3 3.8 3.7
Asian A Pacific Islander, Non-Hispanic 9.6 9.3 3.8 3.7
Hispanic 8.7 8.4 4.3 4.9
Other, Non-Hispanic 9.6 9.3 3.8 3.7
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 ft 1989" Tables.
A-10 I* I
Table Al.WIC Eligibility Study
Race/Hispanic Origin Specific Infant Mortality and Fetal Mortality Rates.
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Minnesota
All Origins 7.1 7.3 6.6 6.1
White, Non-Hispanic 6.7 6.9 6.3 5.7
Black, Non-Hispanic 27.1 28.0 12.3 13.4
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.2 12.6 6.9 6.5
Asian ft Pacific Islander, Non-Hispanic 12.2 12.6 6.9 6.5
Hispanic 12.5 12.9 6.3 5.7
Other, Non-Hispanic 12.2 12.6 6.9 6.5
Mississippi
All Origins 11.6 12.1 10.1 9.5
White, Non-Hispanic 8.4 8.8 7.1 5.7
Black, Non-Hispanic 14.3 14.8 13.4 13.7
Am.Indian, Eskimo ft Aleut, Non-Hispanic 12.1 12.7 9.2 5.3
Asian ft Pacific Islander, Non-Hispanic 12.1 12.7 9.2 5.3
Hispanic 9.0 9.4 7.1 5.7
Other, Non-Hispanic 12.1 12.7 9.2 5.3
Missouri
All Origins 9.9 9.4 6.7 6.4
White, Non-Hispanic 8.5 8.0 6.1 5.7
Black, Non-Hispanic 17.3 16.4 10.1 9.9
Am.Indian, Eskimo ft Aleut, Non-Hispanic 7.2 6.8 6.1 4.9
Asian A Pacific Islander, Non-Hispanic 7.2 6.8 6.1 4.9
Hispanic 4.3 4.1 6.1 5.7
Other, Non-Hispanic 7.2 6.8 6.1 4.9
Montana
All Origins 11.3 8.6 7.6 7.1
White, Non-Hispanic 8.8 6.7 7.1 6.5
Black, Non-Hispanic 0.0 0.0 7.6 7.1
Am.Indian, Eskimo ft Aleut, Non-Hispanic 22.4 17.1 7.6 7.1
Asian ft Pacific Islander, Non-Hispanic 6.2 4.7 7.6 7.1
Hispanic 26.4 20.1 7.6 6.5
Other, Non-Hispanic 22.4 17.1 7.6 7.1
Rates are number of imam deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 ft 1989" Tables.
A-11 bk
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Nebraska
All Origins 7.9 8.3 6.8 6.3
White, Non-Hispanic 6.9 7.2 6.4 5.9
Black, Non-Hispanic 18.9 19.7 14.0 9.6
Am.Indian, Eskimo ft Aleut, Non-Hispanic 16.8 17.6 3.7 9.6
Asian ft Pacific Islander, Non-Hispanic 16.8 17.6 3.7 9.6
Hispanic 12.0 12.5 6.4 5.9
Other, Non-Hispanic 16.8 17.6 3.7 9.6
Nevada
All Origins 8.1 8.4 7.8 8.5
White, Non-Hispanic 7.5 7.8 6.9 7.4
Black, Non-Hispanic 19.2 19.9 16.2 18.3
Am.Indian, Eskimo ft Aleut, Non-Hispanic 3.2 3.3 6.7 7.6
Asian ft Pacific Islander, Non-Hispanic 3.2 3.3 6.7 7.6
HLpanic 5.4 5.6 6.9 7.4
Other, Non-Hispanic 3.2 3.3 6.7 7.6
New Hampshire
All Origins 8.0 7.1 6.0 6.9
White, Non-Hispanic 8.0 7.1 5.9 6.6
Black, Non-Hispanic 0.0 0.0 8.6 6.9
Am.Indian, Eskimo ft Aleut, Non-Hispanic 0.0 0.0 8.6 6.9
Asian ft Pacific Islander, Non-Hispanic 0.0 0.0 8.6 6.9
Hispanic 8.4 7.5 8.6 6.6
Other, Non-Hispanic 0.0 0.0 8.6 6.9
New Jersey
All Origins 9.3 9.0 7.8 7.8
White, Non-Hispanic 6.4 6.2 6.2 6.3
Black, Non-Hispanic 19.5 18.9 13.7 13.6
Am.Indian, Eskimo ft Aleut, Non-Hispanic 12.4 12.0 10.5 7.7
Asian ft Pacific Islander, Non-Hispanic 4.0 3.8 10.5 7.7
Hispanic 9.1 8.8 6.2 6.3
Other, Non-Hispanic 4.0 3.8 10.5 7.7
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 ft 1989" Tables.
A-12
^
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
New Mexico
All Origins 8.5 9.0 5.7 4.2
White, Non-Hispanic 7.1 7.5 5.4 3.9
Black, Non-Hispanic 21.6 22.8 7.3 5.5
Am.Indian, Eskimo & Aleut, Non-Hispanic 11.5 12.2 7.3 5.5
Asian & Pacific Islander, Non-Hispanic 6.2 6.6 7.3 5.5
Hispanic 8.2 8.6 7.3 3.9
Other, Non-Hispanic 11.5 12.2 7.3 5.5
New York
All Origins 10.6 9.6 9.3 10.0
White, Non-Hispanic 7.7 7.0 7.7 8.3
Black, Non-Hispanic 18.3 16.6 15.3 16.4
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.4 11.2 7.2 7.8
Asian & Pacific Islander, Non-Hispanic 4.3 3.9 7.2 7.8
Hispanic 8.8 7.9 7.7 8.3
Other, Non-Hispanic 4.3 3.9 7.2 7.8
North Carolina
All Origins 11.3 10.6 8.6 8.5
White, Non-Hispanic 8.5 7.9 6.7 6.5
Black, Non-Hispanic 17.8 16.7 13.3 13.1
Am.Indian, Eskimo & Aleut, Non-Hispanic 10.6 9.9 6.5 7.5
Asian & Pacific Islander, Non-Hispanic 10.6 9.9 6.5 7.5
Hispanic 8.7 8.1 6.7 6.5
Other, Non-Hispanic 10.6 9.9 6.5 7.5
North Dakota
All Origins 8.0 8.0 6.7 6.1
White, Non-Hispanic 6.7 6.7 5.9 6.5
Black, Non-Hispanic 14.5 14.4 6.7 6.1
Am.Indian, Eskimo & Aleut, Non-Hispanic 14.2 14.1 6.7 6.1
Asian & Pacific Islander, Non-Hispanic 6.2 6.2 6.7 6.1
Hispanic 11.4 11.3 6.7 6.5
Other, Non-Hispanic 14.2 14.1 6.7 6.1
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 & 1989" Tables.
A-13 (eH
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Ohio
All Origins 9.9 9.8 7.0 6.9
White, Non-Hispanic 8.3 8.2 6.4 6.2
Black, Non-Hispanic 18.6 18.4 9.9 10.1
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.4 12.3 6.2 6.4
Asian & Pacific Islander, Non-Hispanic 2.7 2.6 6.2 6.4
Hispanic 12.1 12.0 6.4 6.2
Other, Non-Hispanic 2.7 2.6 6.2 6.4
Oklahoma
All Origins 8.5 9.2 7.3 8.2
White, Non-Hispanic 7.8 8.4 7.0 7.2
Black, Non-Hispanic 13.8 14.9 10.5 14.8
Am.Indian, Eskimo A. Aleut, Non-Hispanic 9.6 10.4 6.8 8.5
Asian & Pacific Islander, Non-Hispanic 9.6 10.4 6.8 8.5
Hispanic 8.2 8.9 7.0 7.2
Other, Non-Hispanic 9.6 10.4 6.8 8.5
Oregon
All Origins 8.9 8.3 6.0 5.3
White, Non-Hispanic 8.4 7.8 6.0 5.3
Black, Non-Hispanic 24.5 22.8 6.0 5.8
Am.Indian, Eskimo & Aleut, Non-Hispanic 8.6 8.0 6.0 5.8
Asian & Pacific Islander, Non-Hispanic 8.6 8.0 6.0 5.8
Hispanic 11.7 10.9 6.0 5.3
Other, Non-Hispanic 8.6 8.0 6.0 5.8
Pennsylvania
All Origins 10.2 9.6 8.6 8.8
White, Non-Hispanic 7.7 7.3 7.5 7.6
Black, Non-Hispanic 23.0 21.7 14.6 14.5
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.4 11.7 8.7 9.4
Asian & Pacific Islander, Non-Hispanic 9.4 8.9 8.7 9.4
Hispanic 12.7 ;2.0 7.5 7.6
Other, Non-Hispanic 9.4 8.9 8.7 9.4
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 & 1989" Tables.
V
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Rhode Island
All Origins 10.2 8.1 6.9 8.4
White, Non-Hispanic 9.8 7.8 6.5 8.1
Black, Non-Hispanic 18.2 14.5 6.9 10.2
Am.Indian, Eskimo & Aleut, Non-Hispanic 9.4 7.5 6.9 10.2
Asian & Pacific Islander Non-Hispanic 9.4 7.5 6.9 10.2
Hispanic 10.3 8.2 6.9 8.1
Other, Non-Hispanic 9.4 7.5 6.9 10.2
South Carolina
All Origins 12.8 11.7 10.7 10.4
White, Non-Hispanic 8.9 8.2 8.5 6.2
Black, Non-Hispanic 18.8 17.2 14.1 16.6
Am.Indian, Eskimo & Aleut, Non-Hispanic 12.4 11.3 14.1 9.4
Asian & Pacific Islander, Non-Hispanic 4.0 3.6 14.1 9.4
Hispanic 9.9 9.0 8.5 6.2
Other, Non-Hispanic 4.0 3.6 14.1 9.4
South Dakota
All Origins 9.6 10.1 6.5 6.4
White, Non-Hispanic 6.9 7.3 6.1 5.6
Black, Non-Hispanic 13.7 14.5 6.5 9.5
Am.Indian, Eskimo & Aleut, Non-Hispanic 21.9 23.1 6.5 9.5
Asian & Pacific blander, Non-Hispanic 6.2 6.6 6.5 9.5
Hispanic 10.8 11.4 6.5 5.6
Other, Non-Hispanic 21.9 23.1 6.5 9.5
Tennessee
All Origins 10.8 10.3 6.5 5.6
White, Non-Hispanic 8.1 7.7 5.1 5.2
Black, Non-Hispanic 14.1 13.5 10.9 6.8
Am.Indian, Eskimo & Aleut, Non-Hispanic 6.3 6.0 8.9 8.8
Asian & Pacific Islander, Non-Hispanic 6.3 6.0 8.9 8.8
Hispanic 12.9 12.3 5.1 5.2
Other, Non-Hispanic 6.3 6.0 8.9 8.8
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rales calculated from NCHS "Vital Statistics of the United Stales 1988 & 1989" Tables.
A-15 W
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates by Racial/Ethnic Group
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Texas
All Origins 9.2 8.1 6.5 6.7
White, Non-Hispanic 7.5 6.6 6.0 6.1
Black, Non-Hispanic 16.7 14.7 10.0 10.3
Am.Indian, Eskimo ft Aleut, Non-Hispanic 12.4 11.0 5.0 6.6
Asian ft Pacific Islander, Non-Hispanic 6.9 6.1 5.0 6.6
Hispanic 8.3 7.4 6.0 6.1
Other, Non-Hispanic 6.9 6.1 5.0 6.6
Utah
All Origins 8.0 7.5 5.0 6.1
White, Non-Hispanic 7.5 7.0 5.1 6.1
Black, Non-Hispanic 23.8 22.3 5.0 6.1
Am.Indian, Eskimo ft Aleut, Non-Hispanic 14.9 13.9 5.0 6.1
Asian & Pacific Islander, Non-Hispanic 14.9 13.9 5.0 6.1
Hispanic 9.0 8.4 5.0 6.1
Other, Non-Hispanic 14.9 13.9 5.0 6.1
Vermont
All Origins 6.9 6.4 5.9 5.0
White, Non-Hispanic 6.7 6.2 6.0 5.1
Black, Non-Hispanic 0.0 0.0 5.9 5.0
Am.Indian, Eskimo ft Aleut, Non-Hispanic 0.0 0.0 5.9 5.0
Asian ft Pacific Islander, Non-Hispanic 0.0 0.0 5.9 5.0
Hispanic 0.0 0.0 5.9 5.1
Other, Non-Hispanic 0.0 0.0 5.9 5.0
Virginia
All Origins 10.0 10.2 10.1 8.9
White, Non-Hispanic 7.3 7.4 8.0 7.6
Black, Non-Hispanic 18.7 19.1 17.3 13.0
Am.Indian, Eskimo ft Aleut, Non-Hispanic 8.0 8.1 5.9 7.3
Asian ft Pacific Islander, Non-Hispanic 8.0 8.1 5.9 7.3
Hispanic 7.7 7.8 8.0 7.6
Other. Non-Hispanic 8.0 8.1 5.9 7.3
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 ft 1989" Tables.
A-16 t7
Table Al. WIC Eligibility Study
Infant Mortality and Fetal Mortality Rates
Census
Extract 1989 1990 1988 1989
Racial/ Infant Infant Fetal Fetal
Ethnic Mortality Mortality Loss Loss
Origin Rate Rate Rate Rate
Washington
All Origins 9.2 7.8 5.2 5.1
White, Non-Hispanic 8.3 7.1 5.0 5.0
Black, Non-Hispanic 19.8 16.8 9.2 9.7
Am.Indian, Eskimo & Aleut, Non-Hispanic 9.0 7.6 6.2 4.5
Asian & Pacific Islander, Non-Hispanic 9.0 7.6 6.2 4.5
Hispanic 10.8 9.2 5.0 5.0
Other, Non-Hispanic 9.0 7.6 6.2 4.5
West Virginia
All Origins 9.4 9.9 7.5 8.5
White, Non-Hispanic 9.1 9.6 7.2 8.4
Black, Non-Hispanic 13.4 14.1 7.5 8.5
Am.Indian, Eskimo & Aleut, Non-Hispanic 0.0 0.0 7.5 8.5
Asian & Pacific Islander, Non-Hispanic 0.0 0.0 7.5 8.5
Hispanic 0.0 0.0 7.5 8.4
Other, Non-Hispanic 0.0 0.0 7.5 8.5
Wisconsin
All Origins 9.1 8.2 6.0 6.6
White, Non-Hispanic 8.2 7.4 5.2 5.9
Black, Non-Hispanic 16.9 15.3 !3.8 12.6
Am.Indian, Eskimo & A'eut, Non-Hispanic 8.7 7.9 5.7 6.6
Asian & Pacific Islander, Non-Hispanic 8.7 7.9 5.7 6.6
Hispanic 10.4 9,4 5.2 5.9
Other, Non-Hispanic 8.7 7.9 5.7 6.6
Wyoming
All Origins 9.4 8.6 7.0 7.6
White, Non-Hispanic 9.3 8.5 7.3 7.1
Black, Non-Hispanic 0.0 0.0 7.0 7.6
Am.Indian, Eskimo & Aleut, Non-Hispanic 7.2 6.6 7.0 7.6
Asian & Pacific Islander, Non-Hispanic 6.2 5.7 7.0 7.6
Hispanic 14.7 13.4 7.0 7.1
Other, Non -Hispanic 7.2 6.6 7.0 7.6
Rates are number of infant deaths or fetal deaths per 1000 live births.
Race Specific Rates calculated from NCHS "Vital Statistics of the United States 1988 & 1989" Tables.
A-17 U
L -:'.J
Table Al. WIC Eligibility Study
United States Territories' - Infant Mortality and Fetal Mortality Rates.
Census
Extract
Racial/
Ethnic
Origin
1988 1989 1988 1989
Infant Infant Fetal Fetal
Mortality Mortality Loss Loss
Rate Rate Rate Rate
Guam
All Origins
Puerto Rico
All Origins
U.S. Virgi
AllOrigi
ids
7.7
12.5
13.1
7.6
12.3
12.9
6.8
10.0
17.1
6.8
10.0
17.1
' Data from Vital Statistics of the United States. 1988. - Mortality and Natality Data. Rates for 1989 calculated by applying the
change in the overall U.S. rates 1988-1989 to the 1988 data.
Rates are number of infant deaths or fetal deaths per 1000 live births.
A-18
fl
Sources of Data
National Center for Health Statistics, Vital Statistics of the United States, 1988. Vol I - Natality,
Tables 1-69 & 3-2, Public Health Service, Washington DC. U.S. Government Printing
Office, 1991.
National Center for Health Statistics, Vital Statistics of the United States, 1988. Vol II -
Mortality, Tables 9-9 & 9-10, Public Health Service, Washington DC. U.S. Government
Printing Office, 1991.
National Center for Health Statistics, Vital Statistics of the United States, 1989. Vol I - Natality,
Tables 1-41 & 1-51 Public Health Service, Washington DC. U.S. Government Printing
Office, 1993.
National Center for Health Statistics, Vital Statistics of the United States, 1989. Vol II -
Mortality, Public Health Service, Washington DC. U.S. Government Printing Office, 1993.
Table 2-4. Infant, Neonatal, and Postneonatal Deaths and Mortality Rates by Specified
Race or National Origin and Sex: United States, 1989. (p.4).
Table 2-8. Infant Mortality Rates by Race: United States, Each Division and State.
1987-1989 (p. 10).
Table 2-19. Infant Deaths by Specified Hispanic Origin and Race for Non-Hispanic
Origin: 47 Reporting States and the District of Columbia, 1989 (p.91).
Table 3-1. Fetal Deaths by Period of Gestation and State of Occurrence: United States
and Each State, 1989 (Page 1).
Table 3-7. Fetal-death Ratios by Race: Each Division and State, 1984-1989 (Page 8).
National Center for Health Statistics. 'Advance report of final mortality statistics, 1990.'
Monthly Vital Statistics Report, Vol. 41, No. 7, (Supplement). Hyattsville:MD: Public Health
Service. January 7, 1993.
A-19 70
Table 25. Total deaths and death rates, and infant and neonatal deaths and mortality
rates for the United States, each Division, and State; and by race and sex
for the United States (Page 43).
National Center for Health Statistics, Vital Statistics of the United States, 1989. Vol I - Natality,
Public Health Service, Washington DC. U.S. Government Printing Office, 1993.
Table 1-51. Live Births by Hispanic Origin and Race of Mother: 47 Reporting States
and the District of Columbia. 1989 (p.95).
A-20 II
Appendix B
Estimation of the WIC Categorical
and Income-Eligible Population
7JL
Appendix B. Estimation of the WIC Categorical and Income-Eligible Population
This appendix describes in detail the methods used to estimate the number of persons at
or below 185 percent of poverty in the categorical groups served by the WIC Program.
These estimates represented the number of persons in the five categorical groups at or
below 185 percent of poverty level who would have been eligible for the WIC program
in an average month in 1989 if categorical and income criteria alone were used to
determine eligibility.
Infants and Children
As noted in the text of this report, the estimates of infants under one year of age and
children ages one to five years of age were based on direct counts from the Census
data.1 The Census Bureau modified the age data for approximately 2.2 percent of the
infants and children under five years of age by checking the correspondence between
reported birth years and ages with the quarterly distribution of births at the national
level in each birth year obtained from National Center for Health Statistics data.2
Modifications were performed separately for each birth year, by sex and by race. The
procedure essentially took the reported birth year to be true and the reported month of
birth to be false. Consequently, babies born in 1990 had their ages correctly imputed to
be zero. As Table Bl indicates, the age modification for infants resulted in an increase
of more than 728,000 infants. This was 22.6 percent more than the unmodified counts.
These estimates differ from other published Census data because they use an adjusted count of infants
developed by the Census to correct for the misclassification of infants less than one year of age. The Census
Bureau found some problems with the accuracy of the ages reported in the 1990 Census. These problems were
particularly acute for infants. Respondents tended to report ages and family size as of the date they completed
the questionnaires, rather than as of April 1, 1990. Ages of infants in years may have been rounded up to age
1 to avoid reporting an age of 0 years. The latter practice would underestimate the number of infants and
overestimate the number of children 1 year of age.
'The major assumptions in the modification procedure wee that the geographic distribution of births was
the same as the national distribution, that there were no significant birthplace-sex-race differences in annual birth
distributions by quarter, and that mortality did not vary by quarter of birth.
B-l
11
Table Bl. Age Modification of 1990 Census Counts of Infants and Children
Age Unmodified Modified Change
(1) (2) (l)-(2)
0 3,217.312 3,945,974 (728,662)
1 3,949,107 3,768,154 180,953
2 3,815,040 3,701,195 113,845
3 3,683,177 3,640,012 43,165
4 3,689,807 3,702,312 (12,505)
Total 18,354,443 18,757,647 (403,204)
Source: Bureau of the Census, "Age, Sex, Race, and Hispanic Origin Information from the 1990
Census: A Comparison of Census Results with Results Where Age and Race Have Been
Modified," 1990 CPH-L-74, August 1991.
On the Census Extract file, infant data reflected the modified counts because the Census
Bureau advised that these were the best counts of infants. These counts were given for
six racial categories and five income levels (75 percent of poverty, 100 percent of
poverty, 130 percent of poverty, 185 percent of poverty, and all incomes). The numbers
of children who were categorically eligible for WIC was enumerated from the Census
data, as classified by race and income levels.
Pregnant Women
The number of pregnant women was estimated from the Census counts of women with
their own infants. These were women who were assumed to have given birth in the
3 While the general problem of age misclassification for infants also applies to children, the Census Bureau
notes that for most single years of age, inaccuracies in age reporting offset each other. As Table 3 shows,
children ages 1 and 2 were most affected by the ^classification of age. The original Census count for age 1 was
reduced by 4.5 percent and that for age 2 by 3 percent, while ages 3 and 4 were only marginally affected. The
Census Bureau only released counts of infants to FCS because of the size of the misclassification problem in that
category and the importance of a reliable estimate of infants for the WIC Program. The Census Bureau will
not release .he modified counts of older children because these data are not considered more reliable than the
unmodified counts.
B-2 7?
twelve months prior to the Census date. However, the Census undercounts all
pregnancies because it omits pregnancies that resulted in live births but for which the
infant died prior to April 1, 1990 or disappeared for other reasons such as adoption. It
also omits pregnancies that terminated without a live birth. To make the count more
accurate, the estimates of pregnant women were adjusted for infant mortality and fetal
loss. The estimates were also adjusted for infants who were not living with their natural
mothers at the time of the Census. This included infants who were no longer living with
their natural mothers as a result of adoption or placement in foster care or with
relatives. The estimation of the number of pregnant women used the following
demographic relationship:
m - C(t-l) x (1 - /Af*(f.lf) - FLR(l.lt)
where
t denotes a period of time,
t - 1 denotes the previous twelve-month period,
C(t-l) denotes the number of conceptions at a given time t-1,
N(t) denotes the number of women with infants less than one year of age at a
given time t,
IMRfajn denotes the proportion of infant deaths as a fraction of live births and
fetal deaths during the period /-/ to f, and
FLR(h]t, denotes the proportion of fetal deaths as a proportion of live births and
fetal deaths in the period (t-l,t).4
^he adjustment for fetal deaths is not a complete adjustment for "pregnancy wastage" because fetal death
rates WjR/fci/i as given in the NCHS Vital Statistics reports includes only the rate of termination of pregnancy
resulting in fetal deaths for pre{. icies at 20 weeks gestation or later. This estimate of pregnancies omits,
therefore, any adjustment for spontaneous or induced abortions occurring prior to 20 weeks gestation.
B-3
7/
The estimate of conceptions in a year is given by the equation:
C(f-1) -
(i-/*Vi.o-FL,W
The number 1 / [1 - IMR(t.]t)-FLR(Ht)) is the adjustment factor to account for infant
deaths and fetal deaths. It was used in estimating pregnancies from the number of live
births in a given year.
To estimate the number of women who were pregnant in an average month in 1989, we
adjusted the above number for the average proportion of the year that a woman was
pregnant in 19895. Becruse on average, a woman is pregnant for 9 months out of 12
months in a year, we would intuitively expect the factor to be 0.75. However, the use of
the Census Extract required a somewhat more complex derivation to yield the desired
degree of accuracy.
As seen in Table B2, women who were pregnant at any time in 1989 gave birth at any
time between January 1989 and September 1990. For mothers who gave birth to
children in the first quarter of 1989 (January through March 1989), the average duration
of pregnancy was 6.5 weeks or 12.4 percent of the year. For the mothers of infants who
were born between April 1989 and March 1990, the average number of weeks that the
mother was pregnant in 1989 was 31.1 weeks or 59.7 percent of the year. For the
mothers of infants born between April and September of 1990 (after the Census was
taken), the average duration of pregnancy was 13.6 weeks or 26.1 percent of the year.
5This adjustment was required in the present study, but was not required in the 1979 WIC Eligibility Study
I. In the 1980 Census Extract women were tabulated if they had own children (0-9 months old), but in the 1990
Census age is only given in years, i.e., infants are recorded as age zero years.
B-4
Table B2. Derivation of the Percentage of a Year that a Woman Was
Pregnant in 1989
Birth Birth Child's Age Time Mother Was
Month Year on 4/1/90
in months
Pregnant in 1989
in weeks
January 1989 14 2.3
February 1989 13 6.4
March 1989 12 10.7
April 1989 11 15.0
May 1989 10 19.4
June 1989 9 23.7
July 1989 8 28.1
August 1989 7 32.6
September 1989 6 36.9
October 1989 5 39.6
November 1989 4 39.6
December 1989 3 39.6
January 1990 2 37.3
February 1990 1 33.1
March 1990 0 28.9
April 1990 Not born 24.6
May 1990 Not born 20.1
June 1990 Not born 15.9
July 1990 Not born 11.4
August 1990 Not born 7.0
September 1990 Not born 2.7
Summary
Average
Birth Child's Weeks Percent
Date Age Pregnant of 1989
Jan 1989-Mar 1989 12-14 months 6.5 12.4%
Apr 1989-Mar 1990 0-11 months 31.1 59.7%
Apr 1990-Sep 1990 Not born on -1/01/90 13.6 26.1%
Note: Assumes birth date is midpoint of the month, and gestational period is 39.6 weeks.
B-5 77
The number of women who were pregnant in an average month in 1989 is given by:
Birffe4/>9.3/90x.597 Birthsm.3inx.l2A Births^^xMl
(1 -IMRl9t9A990-FLRm9A^) + (1 -IMRim-FLRl9u,l9^> + d-«^-HW
Using the data available from the Census Extract, this formula can be rewritten as:
{CCWOl0.nm^^nfantsadop^x.59n ATOC12.ISii4wwl>tx.l24 NOCborMI90.9/90x ?61
(l-/Af/?19891990-FZJ?19891990) + d-/Af/?1989-FZi?19M>19g9) (\-IMR1990-FLRl99t)
where
CCWOI /^.yy mom/,^ = Census Counts of Women with Own Infants who met the
WIC income eligibility criterion for the current family size less 1 ,
Infants (ado tions) = Census Counts of Infants not living with their natural mothers
(adopted, foster care, living with re