See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/26766204

Application of the Kessner and Kotelchuck

Prenatal Care Adequacy Indices in a Preterm
Birth Population

Article in Public Health Nursing September 2009

DOI: 10.1111/j.1525-1446.2009.00803.x Source: PubMed

CITATIONS READS

13 258

3 authors, including:

Joan Rosen Bloch

Drexel University
57 PUBLICATIONS 379 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Home visiting for medically high-risk pregnant women View project

All content following this page was uploaded by Joan Rosen Bloch on 14 July 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Public Health Nursing Vol. 26 No. 5, pp. 449459
0737-1209/r 2009 Wiley Periodicals, Inc.
doi: 10.1111/j.1525-1446.2009.00803.x

SPECIAL FEATURES: METHODS

Application of the Kessner and

Kotelchuck Prenatal Care Adequacy
Indices in a Preterm Birth Population
Joan Rosen Bloch, Katy Dawley, and Patricia Dunphy Suplee

ABSTRACT Objective: Healthy People 2010 goals to eliminate racial and ethnic health disparities that
persist in the utilization of prenatal care (PNC) highlight the importance of measuring PNC as a variable in
maternal and infant health outcomes research. These disparities are significantly correlated to adverse
infant outcomes in preterm birth (PTB), a leading cause of infant mortality and life-long morbidity. Cur-
rently the most extensively used PNC adequacy indices (Kessner and Kotelchuck) were developed to mea-
sure outcomes in populations consisting mostly of full-term births. It is unclear whether these PNC
adequacy indices are reliable when pregnancy is truncated due to PTB (o37 weeks). This paper compares
and demonstrates how they can be applied in a specific PTB cohort. Design and Sample: This secondary
analysis of a nested case-control study compares Kessner and Kotelchuck adequacy scores of 367 mothers
of PTB infants. Results: There were significant differences in the rating of PNC inadequacy ( po.001)
depending on the PNC adequacy index used. Conclusion: Critical evaluation is warranted before using
these PNC adequacy indices in future public health nursing and PTB research.

Key words: Kessner Index, Kotelchuck Index, methodological challenges measuring PNC, prenatal
care adequacy indices, prenatal care utilization, preterm birth.

There are tremendous variations in outcomes of
infants born prematurely (Callaghan, MacDorman,
Rasmussen, Qin, & Lackritz, 2006; Institute of Med-
icine [IOM], 2007; Moster, Lie, & Markestad, 2008).
At present, exact antenatal predictors of infants born
preterm who will have subsequent morbidity and
Joan Rosen Bloch, Ph.D., C.N.R.P., is Assistant Professor,
College of Nursing & Health Professions, Drexel Univer-
sity, Philadelphia, Pennsylvania. Katy Dawley, Ph.D.,
C.N.M., F.A.C.N.M., is Associate Professor and Program
Director of Midwifery, Philadelphia University, Phila-
delphia, Pennsylvania. Patricia Dunphy Suplee, Ph.D.,
C.S., R.N.C., is Assistant Professor and Assistant Dean of
Special Projects, College of Nursing & Health Profes-
sions, Drexel University, Philadelphia, Pennsylvania.
Correspondence to:
Joan Rosen Bloch, Drexel University,245 N.15th Street,
MS 1030, Philadelphia, PA 19102. E-mail: jrb68@
drexel.edu
mortality are unknown. Also unknown is whether
infant outcomes are improved if the mother received
adequate prenatal care (PNC). Selecting appropriate
research methods is critical as we seek to understand
why some preterm birth (PTB) infants do well, some
die, and others have life-long disabilities. The purpose
of this paper is to specically evaluate and compare how
the most commonly used PNC adequacy indices, the
Kessner and Kotelchuck Adequacy of Prenatal Care
Utilization Indices, can be applied in PTB research
aimed at meeting the goals of the nation.
One of the two overarching goals of Healthy People
2010 is the elimination of health disparities among
various segments of the population (U.S. Department
of Health and Human Services [USDHHS], 2000). The
biological, social, and environmental factors contri-
buting to the racial and ethnic health disparities in the
nation are quite complex. Poor pregnancy outcomes
are scrutinized as they are considered a litmus test for
the health of a nation (Nagahawatte & Goldenberg,
2008). Identied racial and ethnic health disparities

449
450 Public Health Nursing Volume 26 Number 5
include accessing early and adequate PNC and mini-
mizing rates of problematic pregnancy outcomes, such
as PTB (USDHHS, 2000). Despite over a decade of
research effort aimed at understanding risk factors as-
sociated with PTB, the rate of PTB has not improved
(Hamilton, Martin, & Ventura, 2007) and the PTB rate
for Black women in 2005 was 18.4% compared with
11.7% for White women (Hamilton et al., 2007). Racial
disparities in PTB rates are directly correlated to the
racial disparities in infant mortality rates in the United
States (Hamilton et al., 2007; IOM, 2007; Martin,
Hamilton, Sutton, Ventura et al., 2006). Methods of
measuring adequacy of PNC in PTB populations must
be evaluated for reliability, because these methods are
integral to research that addresses racial disparities in
receiving adequate PNC, and the potential relationship
between inadequate PNC and adverse infant outcomes
of PTB (Cokkinides, 2001).
For the last 40 years the amount of PNC has consis-
tently been measured as a key variable in maternal
infant health outcome research (Alexander & Kotel-
chuck, 2001; Conway & Kutinova, 2006; Lu et al.,
2006). Currently the most extensively used PNC
adequacy indices (Kessner and Kotelchuck) were
developed to measure outcomes in populations con-
sisting mostly of full-term births. Measurements of
PNC adequacy in research studies when a pregnancy
was truncated due to PTB (o37 completed weeks)
must be explored. This is especially important for
public health nurses who have been leaders in deve-
loping and implementing innovative population
specic community-based PNC programs reaching
out to the most vulnerable at-risk populations for
PTB (Burns, 2005; Cramer, Chen, Roberts, & Clute,
2007; Flynn, Budd, & Modelski, 2008). Public health
nurses may wish to reexamine, and perhaps revise
the way in which they measure the outcomes of the
programs since the existing methods of evaluation
have some drawbacks. Revision of outcome measure-
ments, if undertaken, should take account of current
and future research, program cost-effectiveness,
and the complex social questions related to racial and
ethnic health disparities (Conway & Kutinova, 2006;
IOM, 2007).
In fact, the United States is now launching
the largest long-term study of childrens health and
development ever to be conducted in the United Sta-
tes (in-depth information about this can be found
at http://www.nationalchildrensstudy.gov) providing
an opportune time to scrutinize how reliable tradi-
September/October 2009
tional PNC adequacy indices are when applied solely
to PTB populations. Researchers at multiple sites
throughout the country will enroll women before
pregnancy and follow 100,000 of their infants from
birth until 21 years of age (Landrigan et al., 2006).
Collecting data on women before and during preg-
nancy, and then after birth occurs, promises to
advance scientic understandings of the relationships
between maternal health, prenatal experiences and in
utero exposures with subsequent infant and child-
hood mortality and morbidity. This large-scale study
provides excellent opportunities for researchers,
including public health nurses, to critically evaluate
maternal-infant health outcomes of PNC and requires
reliable measurement tools. The existing tools, used
most widely, were created between 1973 and 1994
when access to PNC was a major focus and priority.
The application of these PNC adequacy indices in
analyzing adverse perinatal outcomes between full-
term and low birth weight/preterm infants was an
important focus of research. However, the use of these
PNC adequacy indices to explicate adverse outcomes
among the subset of preterm infants has not been
adequately addressed, despite their currently routine
use in research and program evaluation.
The History of the Traditional Methods
used to Measure Prenatal Care Adequacy
The two separate PNC adequacy indices that have
been used most frequently in perinatal research were
developed by David Kessner (Kessner, Singer, Kalk, &
Schlesinger, 1973) and Milton Kotelchuck (1994).
They have been extensively applied to populations
consisting mostly of full-term births and have been
important in elucidating risk factors for PTBs
compared with term births. In the United States,
monitoring the utilization of PNC began in 1968. The
month in which the rst prenatal visit occurred and
the total number of prenatal visits were added as
elds to the U.S. Standard Certicate of Live Birth
(Kirby, 1997). The rst and still most widely recog-
nized effort to measure the adequacy of utilized PNC
was carried out by Kessner et al. (1973) in an IOMs
study to determine the impact of health services on
reducing adverse birth outcomes. Later, Kotelchuck
(1994) developed a different PNC Index, which he
named the Adequacy of Prenatal Care Utilization
Index. Both are in use today and both measure
adequacy only in terms of number of visits.

Quantity Content
Attendance
Risk Intervention
Assessment
Initiation Frequency
of Care of Attended Visits
Content
Site Organization
Nurse Managed
Office Based Community Based
Hospital Based
Figure 1. Conceptual Framework for PNC Adequacy Research
Historically, measurements for assessing PNC
adequacy have been methodologically complicated
and challenging (Bell & Zimmerman, 2003; Rosen-
berg, Handler, & Furner, 2004). While clinicians may
not argue the importance of PNC relative to outcome,
epidemiologists and health service researchers have
disputed, for almost four decades, the effectiveness of
PNC in the prevention of low birth weight and PTB
(Conway & Deb, 2005; Conway & Kutinova, 2006;
Liu, 1998; Lu et al., 2006). An emerging consensus
found in the economic (Conway & Kutinova, 2006;
Joyce, 1999; Liu, 1998) and epidemiologic (Bell &
Zimmerman, 2003) literature reveals a failure of the
Medicaid expansions of the 1990s, which were aimed
to increase access to PNC and to decrease low birth
weight, to impact infant health. However, this may
be because of methodological problems with how
adequacy of prenatal care is measured or dened.
Current indices produce crude quantitative measure-
ments of prenatal visits obtained, rather than a multi-
dimensional analysis of content and structure of
the PNC received (Alexander & Kotelchuck, 2001).
Comprehensive PNC provides a multidimensional
program aimed at maintaining and promoting optimal
health for two patients: the mother and her growing
Bloch et al.: Measuring Prenatal Care 451
Utilization of Prenatal Care
Structure Quality
Continuity Provider Comprehensive/Holistic
Teaching Prevention & of care
Diagnostic Guidance
Competence Interpersonal Skills
Method
Provider Case Management Insurance Type
Physician Managed Nurse-Directed Social Worker
FFS HMO
Other Other Medicaid
Obstetrician Midwife NP Family Practice Doctor Collaborative MD/NP
fetus. Figure 1 illustrates dimensions of modern PNC.
Ideally, all of these dimensions should be assessed and
tested in relationship to infant outcomes in epidemio-
logy and health outcomes research. Yet, the two most
commonly used PNC adequacy indices were created
only to provide a measure of access to, and use of, PNC
services based on the American College of Obstetrics
and Gynecology (ACOG, 1989)-recommended number
of attended prenatal visits.
Compounding the unidimensional measurement
of PNC are problems of bias in the research (Bell &
Zimmerman, 2003; Koroukian & Rimm, 2002; Rosen-
berg et al., 2004). Self-selection bias is a big issue.
Women who seek PNC are considered to engage in
good health behaviors, and are generally healthier than
women who do not utilize PNC (Bell & Zimmerman,
2003; Liu, 1998; Rosenberg et al., 2004). It is very dif-
cult to factor this bias out from the overall assessment
of PNC effectiveness. If women who seek adequate PNC
are healthier and have better health behaviors, the
effectiveness of PNC may be overestimated. Conversely,
with women who obtain more PNC because they have
medical and obstetrical problems, the effectiveness
of PNC may be underestimated (Liu, 1998). In fact
research points to differences between women who
452 Public Health Nursing Volume 26 Number 5
obtain no PNC and those who obtain inadequate PNC
(Alexander & Kotelchuck, 1996; York, Williams, &
Monro, 1993). York et al. (1993) found that women
who did not access any PNC had more personal prob-
lems than women who accessed some PNC, but were
classied as receiving inadequate PNC. However, the indices factor in additional visits, now standard
when scoring with the Kotelchuck and Kessner PNC
adequacy indices, these two separate groups of women
are lumped together in the adequacy indices. Additional
bias is introduced if PTB occurs (Bloch, 2001; Rosen-
berg et al., 2004; Stringer, 1998). Mothers experiencing
a PTB have truncated pregnancies resulting in fewer
opportunities to attend PNC, thus creating a spurious
relationship between the numbers of attended prenatal
visits associated with PTBs. While the two indices
incorporate ways to minimize the effect of preterm
delivery bias, they do not compensate for the bias
against the benet of increased attendance at PNC for
the mother identied early as high risk for a PTB.
She might have actually attended more visits than the
ACOG recommendations for the period of her trun-
cated pregnancy.
Despite these methodological aws, published
comparison and criticism of the Kessner and Kotel-
chuck Indices, and creation of other indices (Alexander
& Cornley, 1987; Alexander & Kotelchuck, 2001;
Rosenberg, Handler, & Furner, 2004; Stringer, 1998),
the Kessner and Kotelchuck PNC adequacy indices
continue to be used extensively (Collins, David, Simon,
& Prachand, 2007; Cramer et al., 2007; Morris, Egan,
Fang, & Campbell, 2007). The extensive use of these
indices can be attributed to the ease with which the
PNC data are readily available from birth certicate
records (number of PNC visits, month PNC began, and
gestational length of pregnancy) along with computer-
ized software to calculate PNC adequacy scores.
Recently, Morris et al. (2007) used the Kessner Index
to measure the effects of levels of PNC on the incidence
of Down syndrome birth using the National Center for
Health Statistics dataset, and Collins et al. (2007) used
the Kotelchuck Index to assess PNC adequacy in a com-
bined analysis of Illinois vital records and U.S. census
income data to assess racial disparities in the rates of
PTB infants in Chicago. In another study that evolved
from the College of Nursing and Public Health in
Nebraska, Cramer et al. (2007) used the Kotelchuck
Index as one outcome measure in their comprehensive
evaluation of the Omaha Healthy Start community-
based, case management PNC program designed to
reduce local racial disparities in birth outcomes.
September/October 2009
The problem with the existing indices is that the
adequacy measure may not translate well from a full-
term birth population to a PTB population because
in full-term pregnancies, more PNC visits occur after
28 weeks gestation than before. Also, neither of
of care, for early prenatal screening tests because
existing indices for measuring PNC adequacy were
developed before the advent of early prenatal screen-
ing. Unexplored is how the Kessner and Kotelchuck
Indices perform when applied to populations of births
that are all preterm. Clearly, as illustrated in the above
citations, these indices continue to be used in current
perinatal research. Their application to PTB popula-
tions warrants further scrutiny. The following analysis
was undertaken to measure PNC adequacy when all
births are preterm, by applying both PNC adequacy
indices to an existing dataset, a population of mothers
who birthed live PTB infants (gestational age 19.736
weeks).
Methods
Design and sample
This secondary analysis of a nested case-control study
(n 5 484) using original data collected during the
landmark Neonatal Brain Hemorrhage Study (Pinto-
Martin et al., 1992), compares use of the Kessner and
Kotelchuck Indices in a PTB population of infants.
The dataset is from an ongoing cohort study assessing
outcomes among 1105 low birth weight infants born
in central New Jersey between 1984 and 1987
(Pinto-Martin et al., 1992). The purpose of the nested
case-control study was to determine if there was a
relationship between PNC utilization and neonatal
brain injury evidenced by a cranial ultrasound at 4 hr
after birth (Bloch, 2001). The mean gestational ages
were 29.5 weeks in the case group (brain injury by 4 hr
after birth) and 32.3 weeks in the control group
(no brain injury by 4 hr after birth). The infants
mothers ranged in age from 15 to 44 and the mean
age was 27 years. There were no statistically signi-
cant differences between mothers in the case group
and control group with respect to maternal age,
education, parity, and race.
This dataset was chosen because the entire popu-
lation of infants was born prematurely and the PNC
data were obtained directly from the mother and
prenatal records (if available), not from birth certi-
cates. Trained research nurses collected data directly
 Bloch et al.: Measuring Prenatal Care 453

from prenatal and hospital records in addition to con-
ducting face-to-face postpartum interviews (Casey,
Rieckhoff, Beebe, & Pinto-Martin, 1992). Enrolling
women postpartum captured the mothers who
received no PNC. (Studies that enroll women during
PNC exclude women who never seek this care.) While
birth certicate records are commonly used for this
type of research, the reliability of such vital statistics
has been debated (DiGiuseppe, Aron, Ranborn,
Harper, & Rosenthal, 2002; Northam & Knapp,
2006). Obtaining the data directly from the original
PNC records and the mothers is certainly a more
reliable option (Casey et al., 1992).
Measures
The Kessner Index. The Kessner Index is a
composite index consisting of three categories:
adequate, intermediate, and inadequate dened
according to a xed combination of values on (1) time
of entry into PNC, (2) number of prenatal visits, and
(3) gestational age at delivery (Table 1). To be
assigned to the adequate care category, a woman
must start PNC in the rst trimester, and attend the
number of visits recommended by ACOG, adjusting
for gestational age. The recommended ACOG sche-
dule for a woman with an uncomplicated pregnancy is
every 4 weeks for the rst 28 weeks of pregnancy, then
every 23 weeks until 36 weeks. After 36 weeks,
ACOG recommends weekly visits. Although ACOG de-
nes PNC adequacy as 13 visits, the Kessner Index
denes adequacy as 9 visits because computers, at
that time, only allowed one digit for this eld (Alex-
ander & Kotelchuck, 1996). A woman is assigned to
the inadequate category if she received no care at all, if
her care started in the third trimester, if she has miss-
ing data on PNC, or if the number of visits she re-
ceived falls far below the ACOG standards. Women
with any other combinations of values on the relevant
variables are assigned to the intermediate category
(Kessner et al., 1973; Rosenberg, 1998).
The Kotelchuck Index. In 1994, Kotelchuck
developed a different PNC Index, which he named the
Adequacy of Prenatal Care Utilization Index. For the
purpose of clarity, it will be referred to as the Kotel-
chuck Index in this paper. There are four adequacy
categories: adequate plus, adequate, intermediate,
and inadequate (Table 2). Although the categories
appear similar to the Kessner Index categories, they
are derived differently. Two subindices are generated:
one measure is the time of entry into PNC (Adequacy
of Initiation of Prenatal Care) and the other is the
percent of visits attended after care was initiated
(Adequacy of Received Services). These two indices
are the building blocks for the summary Kotelchuck
Index (Kotelchuck, 1994).
The Adequacy of Initiation of Prenatal Care Index
has four categories based on the month PNC begins:
(1) 12 months, (2) 34 months, (3) 56 months, and
(4) 79 months. In the Adequacy of Received Services

TABLE 1. Summary of Scoring for the Kessner Index for Various Gestational Ages of Preterm Birth

Category of adequacy Trimester of first visit Gestational weeks at delivery Number of visits
Adequate First (up to 12 weeks) And  13 And 1
1417 2
1821 3
2225 4
2629 5
3031 6
3233 7
3435 8
 36 9
Inadequate Third (2840 weeks) Or 1821 And 0
2229 1
3031 2
3233 3
 34 4
Intermediate All other combinations

Note. This table is a composite of tables that have appeared in print since 1973 (Kessner et al., 1973; Morris et al., 2007; Stringer,
1998).
454 Public Health Nursing Volume 26 Number 5 September/October 2009

TABLE 2. Summary of Scoring for the Kotelchuk (APNCU) Index: Worktable to Determine Score Based on Entry into Prenatal
Care (PNC) and Number of Recommended Visits Based on ACOG for Various Gestational Ages of Preterm Birth

Gestational age # of visits # of visits # of visits # of visits

Months prenatal of preterm delivery under 50% 5070% 80109% 110%
care started (weeks) of ACOG of ACOG of ACOG of ACOG
12a Inadequate PNC Intermediate PNC Adequate PNC Adequate Plus PNC
23 o1 2 34 5
24 o2 3 45 6
25 o2 3 45 6
26 o2 3 45 6
27 o2 3 45 6
28 o2 34 56 7
29 o2 34 56 7
30 o3 45 67 8
31 o3 45 67 8
32 o3 46 78 9
33 o3 46 78 9
34 o4 57 89 10
35 o4 57 89 10
36 o4 57 810 11
34a Inadequate PNC Intermediate PNC Adequate PNC Adequate Plus PNC
23 o1 2 3b 4b
24 o1 2 34 5
25 o1 2 34 5
26 o1 2 34 5
27 o1 2 34 5
28 o2 3 45 6
29 o2 3 45 6
30 o2 34 56 7
31 o2 34 56 7
32 o3 45 67 8
33 o3 45 67 8
34 o3 46 78 9
35 o3 46 78 9
36 o4 57 89 10
5 Inadequate PNC Inadequate PNC Inadequate PNC Inadequate PNC
7 Inadequate PNC Inadequate PNC Inadequate PNC Inadequate PNC

Note. Information complied from Kotelchuck (1994).

ACOG 5 American College of Obstetrics and Gynecology; APNCU 5 Adequacy of Prenatal Care Utilization; PNC 5 prenatal care.
a
Beginning of the rst month stated through to the end of the second stated month. bThese numbers are very close. The range of
80109% of three visits is 2.43.27; 110% of three visits is 3.3. Therefore if four visits were attended it could be categorized in
either the 80109% or 110% category.

Index the percent of visits is derived according to the
ACOG standards and is adjusted based on gestational
age. The percent of visits attended, a continuous vari-
able, is a ratio of observed visits to expected visits and
is subdivided into four categories: (1) o50%, (2)
5079%, (3) 80109%, and (4) 4110%, correspon-
ding to inadequate, intermediate, adequate, and
adequate plus. This ratio, together with the month
that PNC started, determines the adequacy category
score. Adequate plus is PNC begun by the 4th month
and 110% or more recommended visits. Adequate is
PNC begun by the 4th month and 80109% recom-
mended visits. Intermediate is PNC begun by the 4th
month and 5079% of recommended visits; and in-
adequate is PNC that begun after the 4th month or
o50% of recommended visits (Kotelchuck, 1994).
An important feature of the Kotelchuck Index is
the category of adequate plus reecting more inten-
sive PNC use. The adequate plus category requires
early initiation of PNC and eliminates the possibility
 Bloch et al.: Measuring Prenatal Care 455

TABLE 3. Comparison of Indices to Measure Utilization of Prenatal Care

Attributes Kessner Index Kotelchuck Index

Basis for standard
Adequate start of care
Standard computer program
Adequate # of visits at 40 weeks
Differentiates no care No
Adequacy categories
Inadequate
Intermediate
Adequate
Intensive (Adequate1)
ACOG
13 months
No
9 13
No
Missing data, 3rd trimester start, or
inadequate number of visits (range of
visits for gestational age, truncated at
9 for 36 weeks or more)
1st trimester start and intermediate
numbers of visits, or 2nd trimester start
and intermediate or adequate numbers
of visits
1st trimester start and adequate numbers
of visits
No category
ACOG
14 months
Yes
Month 59 start, or o50% of the adjusted
expected number of visits (range of visits
for gestational age, adjusted for when
prenatal care began)
Month 14 start, and 5079.9% of the
adjusted expected number of visits
Month 14 start, and 80109.9% of the
adjusted expected number of visits
Month 14 start, and 4110% of the expected
number of visits

Note. ACOG 5 American College of Obstetrics and Gynecology.

of the woman scoring adequate if she starts care
late, is high risk, and then requires excessive care due
to her high-risk status. The Kessner Index does not
have a comparable category. Therefore the Kotelchuck
Index permits more sophisticated analytic strategies.
The component indices can be used separately or
together. The observed/expected visit ratio can be
used as a continuous variable instead of a categorical
variable based on the adequacy category. In addition,
researchers can modify the breakpoints on the con-
tinuous observed/expected visit ratio. Using the two
component indices, one can also examine the distri-
bution of women with adequate and inadequate visits
within strata of time of entry into care. Table 3
presents a comparison between the two indices.
Analytic strategy
This secondary analysis of existing data obtained from
the dataset described above was approved by the
University of Pennsylvanias Review Board. The
researcher received deidentied data directly from
the primary investigator of the original Neonatal
Brain Hemorrhage Study.
Each mothers PNC use was assigned two PNC
adequacy scores, one using the Kessner Index and the
other using the Kotelchuck Index. All adequacy scores
were hand calculated by the researcher using Tables 1
and 2. For this analysis, these tables were developed
for premature gestations (o37 weeks). They can be
easily extended if needed for research with term
gestations. To determine agreement of these indices
for scoring prenatal adequacy in a group of mothers
who all had PTB infants, the kappa statistic was used.
McNemar chi-square analysis was conducted to
compare how each mothers adequacy scores varied
depending on which PNC index was used.
Results
PNC adequacy scores were calculated for 367 of the
mothers of the PTB infants. Difculty getting original
PNC records resulted in missing data for 24% of the
mothers of the nested case-control study. Missing
data was expected in this analysis as it is common in
secondary analyses of large epidemiology datasets in
which hundreds of variables are measured (Magee,
Lee, Giuliana, & Munro, 2006). Unknown was the
extent of missing data until the dataset was retrieved
and the variables of interest were scrutinized.
This samples adequacy patterns of PNC (see
Fig. 2) differed according to which adequacy index
was used (k 5 0.58, po.01). This was not surprising
because the two indices have conceptual differences
with different scoring (Kotelchuck, 1994). Using the
Kessner Index, 42.8% (n 5 157) of the mothers had
adequate PNC, 42.5% (n 5 156) intermediate, and
456 Public Health Nursing Volume 26 Number 5
45
40
35 42.8%
30
25
20
15
10 33% 22% 14.7%
5
0
Adequate Inadequate
Kotelchuck 33% 22%
Kessner 42.8% 14.7%
Figure 2. Kotelchuck and Kessner indices reflecting prenatal care utilization of the total preterm
birth sample
14.7% (n 54) had inadequate PNC. Using the Kotel-
chuck Index, 27.0% (n 5 99) had adequate plus PNC,
33.0% (n 5 121) had adequate PNC, 18.0% (n 5 66)
intermediate, and 22.0% (n 5 81) had inadequate
PNC. Combining the two adequacy categories (ade-
quate and adequate plus) of the Kotelchuck Index,
60% (n 5 220) of this sample attended adequate
amount of PNC visits compared to 42.8% (n 5 157)
when using the Kessner Index to score PNC adequacy.
Signicant differences when classifying PNC inade-
quacy ( po.001) were noted with the odds of a woman
classied as inadequate on the Kotelchuck Index
being 15.5 times more likely than a woman classied as
inadequate on the Kessner Index (95% CI: 3.9133.1).
Discussion
There are signicant differences between how the PTB
populations PNC utilization was portrayed when the
Kessner and Kotelchuck Indices were used to cate-
gorize this samples adequacy/inadequacy use of PNC.
Noteworthy is the different conclusions that can be
drawn about a populations inadequate PNC utiliza-
tion depending on the index used. The criteria for
entry into PNC are stricter when using the Kotelchuck
Index. No matter how many visits the mother
attended, if she did not start care by 20 weeks,
she scored inadequate. To score inadequate on the
Kessner Index based solely on entry into care, the
criterion is 28 weeks. From a public policy perspec-
tive this is important because reporting a populations
PNC inadequacy rate as 22.0% versus 14.7% may
effect public health resource allocations. As illustrated
by this analysis, projections of the health service
needs of a population differ based on which rate was
September/October 2009
42.5%
18% 27%
Intermediate Adequate Plus
18% 27%
42.5%
applied. It is clear that a more sensitive method is
necessary to measure PNC adequacy/inadequacy
given the potential to inuence allocation of public
health resources. When evaluating PNC adequacy for
populations of PTB infants counting PNC visits is in-
sufcient. This is especially so when the difference of
scores may be based on only one visit, for example 2
versus 3 visits. Undifferentiated by these indices is
the scenario where one pregnant woman initiates PNC
care at 10 weeks gestation. During her initial visit, she
gets an ultrasound and blood drawn for rst trimester
screening. Yet, in a different setting, another woman,
also at 10 weeks gestation initiates PNC but receives
two additional appointments for her rst trimester
screen (one appointment for the ultrasound, the other
to the laboratory to draw her blood). When these
women are interviewed postpartum about the num-
ber of PNC visits, are they counting all the antenatal
testing or just the visits with the midwife, nurse prac-
titioner, physician assistant, or doctor? What if the
woman received community-based nurse case man-
agement services? Does this get counted or does the
perinatal nursing care she received remain invisible to
the epidemiologists and policy makers? Also problem-
atic is that women who receive no PNC are not sepa-
rated out in either PNC adequacy index. Health
behavior implications for mothers who never seek
PNC are important (York et al., 1993) and requires
critical evaluation of using these indices in perinatal
research. Research that combines women who never
received PNC with women who receive less than an
adequate amount may result in ndings that underes-
timate the risk of adverse maternal and infant out-
comes among women who receive no PNC. While
the dose-effect of PNC or exactly what about PNC

improves outcomes is unclear, the evidence indicates
that women who receive no PNC have worse
pregnancy outcomes (Nagahawatte & Goldenberg,
2008). The aspects of PNC that contribute to success-
ful pregnancy outcomes may be access to appropriate
screening for infections and other conditions, or a
connection with a health care provider/system when
an emergency occurs. Receiving appropriate care for
an impending premature birth may be delayed if the
pregnant woman has not previously accessed PNC.
Theoretically, a woman enrolled in PNC would be
more likely to identify signs of a problem and how to
access care in case of an emergency. This knowledge
and appropriate action may prevent more serious se-
quelae. Thus, for PTB population research, either a
separate category no prenatal care should be added
to each index or researchers should abandon using
these indices and simply use a dichotomous (yes/no)
variable for PNC utilization. Grouping together
women who received no care with women who
received some, but an inadequate amount, potentially
masks the adverse outcomes of not receiving any PNC,
which then results in underestimating the effect of
care on PTB outcomes.
Implications for research and practice
The rate of PTB with widening racial and ethnic dis-
parities continues to increase contributing signicantly
to subsequent infant mortality and morbidity. Clearly
there are methodological aws in measurement of PNC
adequacy. Current methods conceptualize PNC uni-
dimensionally, but PNC is multidimensional. Future
research using expanded methodology is crucial if
we are going to elucidate modiable antenatal risk
factors, move toward practical translation into prac-
tice, and improve outcomes for the smallest and most
vulnerable infants.
Nurses have been instrumental in developing
innovative programs that reach out to vulnerable
marginalized women to facilitate optimal perinatal
outcomes. Since the 1920s excellent maternal and
neonatal health outcomes of nurse-midwifery care in
the United States (Summers, 2007) have been docu-
mented in landmark population-based research (Levy,
Wilkinson, & Marine, 1971; MacDorman & Singh, 1998;
Visintainer et al., 2000). Nurses must become savvy
about how population outcomes are measured and
what can be assumed and not assumed about their
programs of care. To prevent care decisions being
based on awed biased data, researchers need to cri-
Bloch et al.: Measuring Prenatal Care 457
tically evaluate available PNC adequacy indices before
choosing one for their research. Existing qualitative
studies on the process of care can guide this research
(Kennedy, Rousseau & Low, 2003; Kennedy &
Shannon, 2004). While the Kessner and Kotelchuck
Indices evaluate the amount of PNC received, they lack
specicity in evaluating content and quality of PNC.
Further development of methods that may include
tweaking of the existing indices or developing new
methods for measuring PNC is needed, especially in
populations of PTBs. With technological advances
including widespread use of electronic health records
and potential access to linked maternal-neonatal data-
sets, such as the National Childrens Study, which
has the potential to provide incredible opportunities
for public health nurses to evaluate PNC access and
outcomes of perinatal nursing care and programs, this
methodological problem may be eliminated. In the
future, nurse researchers must prudently consider
expanded possibilities to comprehensively evaluate
more dimensions of PNC utilization (as illustrated in
Fig. 1) with maternal and neonatal outcomes.
References
Alexander, G. R., & Cornley, D. A. (1987). Prenatal
care utilization: Its measurement and relation-
ship to pregnancy outcome. American Journal
of Preventative Medicine, 3, 243253.
Alexander, G. R., & Kotelchuck, M. (1996). Quantifying
the adequacy of prenatal care: A comparison of
indices. Public Health Reports, 111(5), 408419.
Alexander, G. R., & Kotelchuck, M. (2001). Assessing
the role and effectiveness of prenatal care:
History, challenges, and directions for future
research. Public Health Reports, 116(4),
306316.
American College of Obstetrics and Gynecology
(1989). Standards for Obstetric-Gynecologic
Services (7th ed.). Washington, DC: ACOG.
Bell, J. F., & Zimmerman, F. J. (2003). Selection bias
in prenatal care use by Medicaid recipients.
Maternal and Child Health Journal, 7 (4),
239252.
Bloch, J. R. (2001). The relationship between utiliza-
tion of prenatal care and neonatal brain injury
in low birth weight infants. Unpublished doc-
toral dissertation, University of Pennsylvania.
Burns, P. G. (2005). Reducing infant mortality rates
using the perinatal periods of risk model. Public
Health Nursing, 22(1), 27.
458 Public Health Nursing Volume 26 Number 5
Callaghan, W. M., MacDorman, M. F., Rasmussen, S.
A., Qin, C., & Lackritz, E. M. (2006). The con-
tribution of preterm birth to infant mortality
rates in the United States. Pediatrics, 118(4),
15661573.
Casey, R., Rieckhoff, M., Beebe, S. A., & Pinto-Martin,
J. (1992). Obstetric and perinatal events: The
accuracy of maternal report. Clinical Pediat-
rics, 31(4), 200204.
Cokkinides, V. (2001). Health insurance coverage-
enrollment and adequacy of prenatal care utili-
zation. Journal of Health Care for the Poor and
Underserved, 12(4), 461473.
Collins, J. W., David, R.J, Simon, D. M., & Prachand,
N. G. (2007). Preterm birth among African
American and white women with a lifelong res-
idence in high-income Chicago neighborhoods:
An exploratory study. Ethnicity and Disease,
17 (1), 113117.
Conway, K. S., & Deb, P. (2005). Is prenatal care really
ineffective? Or, is the devil in the distribution?
Journal of Health Economics, 24(3), 489513.
Conway, K. S., & Kutinova, A. (2006). Maternal
health: Does prenatal care make a difference?
Health Economics, 15(5), 461488.
Cramer, M. E., Chen, L., Roberts, S., & Clute, D.
(2007). Evaluating the social and economic
impact of community-based prenatal care.
Public Health Nursing, 24(4), 329336.
DiGiuseppe, D. L., Aron, D. C., Ranborn, L., Harper,
D. L., & Rosenthal, G. E. (2002). Reliability of
birth certicate data: A multi-hospital compar-
ison to medical records information. Maternal
and Child Health Journal, 6(3), 169179.
Flynn, L., Budd, M., & Modelski, J. (2008). Enhancing
resource utilization among pregnant adoles-
cents. Public Health Nursing, 25(2), 140148.
Hamilton, B. E., Martin, J. A., & Ventura, S. J. (2007).
Preliminary data for 2006. National Vital
Statistics Reports, Hyattsville, MD.
Institute of Medicine (2007). Preterm birth: Causes,
consequences, and prevention. Washington,
DC: The National Academies Press.
Joyce, T. (1999). Impact of augmented prenatal care
on birth outcomes of Medicaid recipients in
New York city. Journal of Health Economics,
18(1), 3167.
Kennedy, H. P., Rousseau, A. L., & Low, L. K. (2003). An
exploratory metasynthesis of midwifery practice
in the United States. Midwifery, 19(3), 203214.
Kennedy, H. P., & Shannon, M. T. (2004). Keeping birth
normal: Research ndings on midwifery care
during childbirth. Journal of Obstetric, Gyn-
ecologic & Neonatal Nursing, 33(5), 554560.
September/October 2009
Kessner, D., Singer, J., Kalk, C., & Schlesinger, E.
(1973). Infant death: An analysis by maternal
risk and health care. Washington, DC: Insti-
tute of Medicine and National Academies of
Sciences.
Kirby, R. (1997). The quality of vital perinatal statis-
tics data, with special reference to prenatal
care. Paediatric and Perinatal Epidemiology,
11, 122128.
Koroukian, S. M., & Rimm, A. A. (2002). The Ade-
quacy of Prenatal Care Utilization (APNCU)
index to study low birth weight: Is the index
biased? Journal of Clinical Epidemiology,
55(3), 296305.
Kotelchuck, M. (1994). An evaluation of the Kessner
Adequacy of Prenatal Care Index and a pro-
posed Adequacy of Prenatal Care Utilization
Index. American Journal of Public Health,
84(9), 14141420.
Landrigan, P. J., Trasande, L., Thorpe, L. E., Gwynn,
L., Lioy, P. J., DAlton, M. E., et al. (2006). The
National Childrens Study: A 21-year prospec-
tive study of 100,000 American children. Pedi-
atrics, 118(5), 21732186.
Levy, B. S., Wilkinson, F. S., & Marine, W. M. (1971).
Reducing neonatal mortality with nurse-
midwives. American Journal of Obstetrics and
Gynecology, 109, 5058.
Liu, G. G. (1998). Birth outcomes and the effectiveness
of prenatal care. Health Services Research,
32(6), 805823.
Lu, M. C., Kotelchuck, M., Culhane, J. F., Hobel, C. J.,
Klerman, L. V., & Thorp, J. M. (2006). Precon-
ception care between pregnancies: The content
of internatal care. Maternal and Child Health
Journal, 10(5, Suppl.), S107S122.
MacDorman, M. F., & Singh, G. K. (1998). Midwifery
care, social and medical risks factors, and
birth outcomes in the USA. Journal of Epi-
demiology and Community Health, 52(5),
310317.
Magee, T., Lee, S. M., Giuliana, K., & Munro, B. (2006).
Generating new knowledge from existing data:
The use of large data sets for nursing research.
Nursing Research, 55(2S), S50S56.
Martin, J. A., Hamilton, B. E., Sutton, P. D., Ventura, S.
J., Menaker, F., & Kirmeyer, S. (2006). Births:
Final data for 2004. National Vital Statistics
Reports, 55(1), 1101.
Morris, B. A., Egan, J. F., Fang, Y. M., & Campbell, W.
A. (2007). The relationship between utilization
of prenatal care and Down syndrome live
births. Journal of Maternal-Fetal and Neo-
natal Medicine, 20(4), 307311.

Moster, D., Lie, R. T., & Markestad, T. (2008). Long-
term medical and social consequences of preterm
birth. New England Journal of Medicine, 359(3),
262273.
Nagahawatte, N. T., & Goldenberg, R. L. (2008).
Poverty, maternal health, and adverse preg-
nancy outcomes. Annals of the New York
Academy of Sciences, 1136, 8085.
Northam, S., & Knapp, T. R. (2006). The reliability
and validity of birth certicates. Journal of Ob-
stetric, Gynecologic, & Neonatal Nursing, 35(1),
312.
Pinto-Martin, J., Paneth, N., Witomski, T., Stein, I.,
Schonfeld, S., Rosenfeld, D., et al. (1992).
The central New Jersey neonatal brain
haemorrhage study: Design of the study
and reliability of ultrasound diagnosis. Pa-
ediatric and Perinatal Epidemiology, 6(2),
273284.
Rosenberg, D. (1998). A new method for describing
patterns of prenatal care utilization. Unpub-
lished doctoral dissertation, University of
Chicago.
View publication stats
Bloch et al.: Measuring Prenatal Care 459
Rosenberg, D., Handler, A., & Furner, S. (2004). A
new method for classifying patterns of prenatal
care utilization using cluster analysis. Maternal
and Child Health Journal, 8(1), 1930.
Stringer, M. (1998). Issues in determining and mea-
suring adequacy of prenatal care. Journal of
Perinatology, 18(1), 6873.
Summers, L. (2007). Enough is enough! Journal
of Midwifery and Womens Health, 52(5),
427428.
U.S. Department of Health and Human Services
(2000). Healthy People 2010: Understanding
and Improving Health (2nd ed.). Washington,
DC: U.S. Government Printing Ofce.
Visintainer, P. F., Uman, J., Horgan, K., Ibald, A.,
Verma, U., & Tejani, N. (2000). Reduced risk of
low weight births among indigent women
receiving care from nurse-midwives. Journal
of Epidemiology and Community Health,
54(3), 233238.
York, R., Williams, P., & Monro, B. H. (1993). Maternal
factors that inuence inadequate prenatal care.
Public Health Nursing, 10(4), 241244.