Content uploaded by Arthur Greil
Author content
All content in this area was uploaded by Arthur Greil on Oct 24, 2016
Content may be subject to copyright.
Available via license: CC BY-NC 4.0
Content may be subject to copyright.
DEMOGRAPHIC RESEARCH
VOLUME 35, ARTICLE 39, PAGES 1149
−
1168
PUBLISHED 20 OCTOBER 2016
http://www.demographic-research.org/Volumes/Vol35/39/
DOI: 10.4054/DemRes.2016.35.39
Research Article
Infertility and fertility intentions, desires, and
outcomes among US women
Karina M. Shreffler
Stacy Tiemeyer
Cassandra Dorius
Tiffany Spierling
Arthur L. Greil
Julia McQuillan
©2016 Karina M. Shreffler et al.
This open-access work is published under the terms of the Creative Commons
Attribution NonCommercial License 2.0 Germany, which permits use,
reproduction & distribution in any medium for non-commercial purposes,
provided the original author(s) and source are given credit.
See http:// creativecommons.org/licenses/by-nc/2.0/de/
Contents
1
Introduction
1150
1.1
Connecting the infertility experience and identity with demographic
research on fertility
1151
1.2
What does it mean to “intend to give birth” in the context of
infertility?
1152
2
Data and methods
1153
2.1
Sample
1153
2.2
Measures
1154
2.3
Analytic strategy
1155
3
Results
1156
4
Conclusions
1161
References
1165
Demographic Research: Volume 35, Article 39
Research Article
http://www.demographic-research.org 1149
Infertility and fertility intentions, desires,
and outcomes among US women
Karina M. Shreffler1
Stacy Tiemeyer2
Cassandra Dorius3
Tiffany Spierling4
Arthur L. Greil5
Julia McQuillan6
Abstract
BACKGROUND AND OBJECTIVE
Little is known about how the experience of infertility or identification as someone with
infertility shapes women’s fertility intentions, desires, or birth outcomes. The purpose
of this paper is to help fill this gap in knowledge for fertility-intentions research.
METHODS
Using data from the National Survey of Fertility Barriers (NSFB), we use linear and
logistic regression methods to assess how infertility and parity statuses are associated
with fertility intentions and desires, as well as how statuses at one point in time predict
birth three years later.
RESULTS
We find that infertility is associated with lower fertility intentions. Women who have
experienced infertility and/or identify as a person with infertility, however, express
greater desires to have a baby and a higher ideal number of children. Women who meet
the medical criteria for infertility are less likely than fecund women to give birth,
despite greater desires.
1 Oklahoma State University, USA. E-Mail: karina.shreffler@okstate.edu.
2 Oklahoma State University, USA.
3 Iowa State University, USA.
4 Oklahoma State University, USA.
5 Alfred University, USA.
6 University of Nebraska-Lincoln, USA.
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1150 http://www.demographic-research.org
CONCLUSION
These findings have important theoretical implications for our understanding of the
meaning of fertility intentions for those who think their ability to achieve their
intentions is uncertain, as well as for empirical research on fertility.
1. Introduction
Demographers interested in fertility trends and projections have long considered the
role of infertility, particularly in analyses of developing countries (see Rutstein and
Shah 2004). When surveyed about infertility, the lay public also understands the term
‘infertility’ to mean a permanent inability to give birth (Maill 1994). Yet the medical
definition of infertility ‒ 12 months or more of unprotected, heterosexual intercourse
without conception (American Society for Reproductive Medicine [ASRM] 2008),
sometimes referred to as subfecundity ‒ does not indicate permanent involuntary
childlessness. The proportion of women in the US who experience infertility using the
ASRM definition is 7% to 15.5% in a given year depending upon specific measurement
procedures (Thoma et al. 2013). The percentage is substantially higher (51.8%) for
women who have met the ASRM criteria at some point throughout their reproductive
lifespan (Greil et al. 2011a). Little is known about how women who experience
infertility (e.g., meet the medical criteria) or identify as having a fertility problem (i.e.,
with or without meeting the medical criteria) think about their fertility intentions and
desires compared to women without infertility experiences. Further, it is not known
how infertility affects birth outcomes in countries such as the United States, where over
10% of all women trying to become pregnant seek medical treatment for infertility
(Simonsen, Baksh, and Stanford 2012). Fears of infertility and decisions about how to
respond to failures to conceive may influence fertility intentions, desires, and outcomes.
Yet scant research has examined these relationships. We therefore use the National
Survey of Fertility Barriers (NSFB) to explore the associations between infertility and
fertility intentions, fertility desires, and the ideal number of children among US women
ages 25‒45 by fertility status (not infertile, medically infertile, self-identifying as
infertile without meeting the medical definition of infertility, and self-identifying as
infertile with experience of infertility) and motherhood status (childless and having one
or more children). We further examine how infertility and motherhood predict the
likelihood of a birth in the three years after the initial interview.
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1151
1.1 Connecting the infertility experience and identity with demographic research
on fertility
Infertility is an interesting phenomenon for fertility researchers and theorists who study
fertility intentions. Though seemingly straightforward, infertility does not necessarily
have clear implications for fertility considerations. Demographers have long realized
the importance of infertility at the societal level for measures of fertility trends. For
example, Davis and Blake (1956) identified intermediate fertility variables including
sterility, which included noncontraceptive sterility, or what we think of today as
infertility or subfecundity. Calculations predicting fertility trends measure infertility in
different ways: Stover (1998) suggested including a measure of sterility in fertility
calculations that includes women in a union who are not menopausal, postpartum
amenorrheic, or pregnant but have not given birth within the past five years despite not
using contraception, as well as women who declare themselves to be infertile.
Mascarenhas and colleagues (2012) have applied this approach to Demographic Health
Survey data from 53 countries, for example, in an effort to determine the global
prevalence of infertility.
Thus, whereas the demographic approach often infers infertility from questions on
couple status, contraception use, and fertility desires (Mascarenhas et al. 2012), the
medical approach utilizes questions focused on meeting the medical criteria for
infertility (e.g., “Have you ever had a period of at least 12 months of regular,
unprotected sexual intercourse without becoming pregnant?”) and perceptions of a
fertility problem (e.g., “Do you see yourself as someone who has had trouble trying to
conceive?”). The medical approach, therefore, can identify individuals or couples who
may benefit from medical assistance to achieve conception, whereas the demographic
approach is typically seeking to identify women who are not likely to become pregnant.
In the demographic- and social-science literatures, we are perhaps most familiar with
infertile women who are involuntarily childless. This is a meaningful group for both
demographic and social scientists, as this group has important implications for fertility-
rate calculations (Bongaarts and Potter 2013) and because infertility is particularly
distressing for women who are childless and have strong fertility desires
(Schwerdtfeger and Shreffler 2009). Yet the majority of women who experience
infertility do not remain childless (Schwerdtfeger and Shreffler 2009). Further, while
we might expect a great deal of overlap between experiencing a period of infertility
(e.g., not getting pregnant after at least a year of regular, unprotected heterosexual
intercourse) and realizing that there might be a fertility problem, there are surprising
discrepancies. A substantial proportion of women (35%) who meet the medical criteria
do not view themselves as being infertile (White et al. 2006). Surprisingly, there are
also women who do not appear to meet the medical criteria for infertility but believe
that they have a fertility problem or expect to have a problem when they start trying to
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1152 http://www.demographic-research.org
conceive. A study of unmarried adults aged 18‒29, for example, found that 19% of
women believed that they were “very likely” to be infertile, and having this belief was
associated with contraceptive behaviors (Polis and Zabin 2012). Infertility, therefore,
could affect fertility in two directions. Women who meet the medical criteria for
infertility are not getting pregnant, at least temporarily. Women who identify as infertile
but do not meet the medical criteria may actually be more likely to give birth due to
their beliefs that contraception to prevent birth is unnecessary. Moreover, if women
who are infertile do not think that they will become pregnant, these births could be
classified as ‘unintended’ due to the current measurement of fertility intentions..
1.2 What does it mean to “intend to give birth” in the context of infertility?
Theoretical conceptualizations of fertility intentions suggest that they are shaped by the
desire for a particular outcome, a belief that taking an action will result in that desired
outcome, and a commitment to perform the action (Malle, Moses, and Baldwin 2003).
Ajzen and Klobas (2013) suggest that one factor affecting fertility intentions is the
degree of perceived control that individuals have over their fertility. Women who are
infertile may not believe that taking an action (e.g., stopping contraception, predicting
ovulation) will result in pregnancy. Yet the experience of infertility alone may not be
enough to influence fertility intentions and behaviors. It may require conscious
processing of this experience, resulting in identification of a fertility problem. Thus,
there needs to be a consideration of both the experience of infertility and the perception
of infertility. Not all women who meet the medical criteria for infertility are aware that
they have a problem (Greil and McQuillan 2010). Indeed, it is the perception of
infertility that matters more for psychological outcomes than merely meeting the
medical criteria for infertility (Greil et al. 2011b).
Therefore, women who have given thought to their fertility experiences and have
come to realize that they have trouble becoming pregnant may be particularly
susceptible to influences of infertility on their fertility intentions. Fertility-intentions
questions seemingly assume the ability to carry out intentions; for example, the 2006‒
2010 National Survey of Family Growth (NSFG) asked respondents, “Looking to the
future, do you intend to have (a/another) child at some time?” followed by, “How sure
are you that you (will/will not) have a baby?” Often, these questions are coded into one
measure ranging from ‘very sure, do not intend’ to ‘very sure, intend’ (e.g., Thomson
1997), which is then used to predict future fertility (e.g., Schoen et al. 1999) or
intendedness of a birth (e.g., Maximova and Quesnel-Vallée 2009).
Yet, infertility calls into question the meaning of fertility intentions when the
outcomes are uncertain, especially if respondents are asked to report intentions within a
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1153
particular time frame. McQuillan et al. (2015) suggest a stronger emphasis on identity is
needed in studies of fertility intentions in addition to the more structural (i.e., life
course) and cultural (i.e., values) approaches often used. Though they focused on
motherhood identity, we suggest that identifying as infertile may be another salient
identity in the study of fertility intentions. Because perception of infertility and meeting
the medical criteria of infertility are often not synonymous constructs, measuring them
separately allows for more-refined investigation of associations between infertility and
fertility intentions, desires, ideals, and outcomes. For example, the uncertainty of the
outcome following perceived infertility raises interesting questions about fertility
intentions. What do intentions mean for women who may strongly want to give birth or
avoid giving birth but are uncertain about their ability to be able to conceive? Do they
report strong intentions (e.g., very sure that they intend or do not intend to give birth),
or do they downgrade or downplay their intentions even if they might want to have a
baby, or alternatively, do they downplay intentions if they plan to avoid pregnancy
because it seems unnecessary to have strong intentions not to have a baby while
infertile? Do infertile women reduce their intentions, and possibly their preferred
number of children, to meet their expectations of lowered fertility, perhaps because they
are trying to be realistic about what they believe is possible? On the other hand, does
concern over infertility increase their desire to have a baby and/or preferred family size,
resulting in inflated preferences? And finally, are infertile women able to have the
number of children that they want to have, or does meeting the medical criteria for
infertility predict a lower likelihood of a birth?
2. Data and methods
2.1 Sample
Our data comes from the National Survey of Fertility Barriers (NSFB), a random digit-
dialing telephone survey of 4,712 women of childbearing ages (25–45) and a subset of
their husbands and partners. The study was designed to assess social and health factors
related to reproductive choices and fertility for US women. The first wave was collected
from 2004 to 2006, and the second three years later. The data is nationally
representative, with an oversample of Black and Hispanic women and women with
fertility problems. Analyses for this study are weighted to account for the oversamples.
Our sample for fertility intentions and desires analyses is restricted to 2,978 women
who are not surgically sterilized or in a heterosexual marriage or cohabiting relationship
with a man who has been surgically sterilized. Our sample for the analysis examining
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1154 http://www.demographic-research.org
birth odds by infertility and parity status includes 1,547 women from the first sample
who participated in the wave-two interview.
2.2 Measures
Dependent variables. In our first set of multivariate analyses utilizing wave-one data
only, we examine three dependent variables targeting fertility intentions, wants, and
ideals. In a subsequent analysis, we utilize these variables as predictor variables of
whether or not a birth occurred between survey waves. The first dependent variable,
fertility intentions, is based on two questions that are combined to create an ordinal
measure of fertility intentions. Respondents were asked, “Do you intend to have a
baby?” and “Of course, sometimes things do not work out exactly as we intend them to,
or something makes us change our minds. In your case, how sure are you that you
will/will not have a child?” Responses were coded so that low scores indicate, “Very
sure, do not intend,” (‒2) and high scores indicate, “Very sure, do intend” (+2). Women
who said they “don’t know” their intentions, who said they cannot have children, or
who said they would let God or nature decide are coded 0 (the center of the scale).
These questions are similar to those used in the National Survey of Families and
Households; we recoded the response categories so that a positive score indicates
intending and a negative score indicates not intending to have a baby. Our second
dependent variable, want a baby, is measured by a question asking, “In the future,
would you like to have a(nother) baby?” Responses are coded from 1 (definitely no) to
4 (definitely yes). Ideal number of children was measured by asking, “How many
children would you consider to be ideal for you?” and coded from 0 to 5, with 5
including preferences of five or more children. Birth between waves is an indicator
variable where 1 means that the respondent gave birth between waves.
Infertility and parity groups. We categorized women as medically infertile if
they had ever had a period of 12 months or more during which they had unprotected
heterosexual intercourse and did not get pregnant. We considered women to be self-
identifying as infertile if they answered yes to either, “Do you think of yourself as
someone who has, has had, or might have trouble getting pregnant?” or “Do you think
of yourself as someone who has or has had fertility problems?” Three distinct groups
were created: women who were medically infertile but did not self-identify as infertile,
women who did not meet the medical criteria for infertility but did identify as infertile,
and women who both met medical criteria and self-identified as infertile. A fourth
group, of non-infertile women (answering no to both the self-identification and
medically infertile questions), was also included in the study. These four groups were
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1155
further split by parity, comparing women who were childless to women who had at
least one biological child.
Sociodemographic control variables. Age is a continuous variable and ranges
from 25 to 45 in our sample. Married was measured by a question about marital status,
with 1 indicating that the respondent was married at the time of the interview.
Cohabiting was coded by either a voluntary response to the marital status question or a
yes to a follow-up question, “Are you currently living with a romantic partner?” Single
women who are neither married nor in a cohabiting relationship are the reference
category. Education (in years) is a continuous variable, ranging from 2 to 22 in our
sample. Race/ethnicity is included as a dummy variable for Black, Hispanic, and “Other
race,” with White respondents as the reference category.
2.3 Analytic strategy
We first provide an overview of the sample, comparing descriptive statistics
(means/percentages) of all study variables by infertility and parity status. To examine
differences within infertility and parity groups, we employed Bonferroni post-hoc tests.
The differences in distributions of our variables by parity motivated our decision to run
separate multivariate analyses for women with young children and women with 1 or
more children.
Our first multivariate analysis uses stereotype logistic regression (SLM) to predict
fertility intentions and desires. Both fertility intentions and fertility desires are measured
on an ordinal scale, but ordinal logistic regression assumes that the distance between
each category is the same (proportional odds) (Long and Freese 2006). We tested the
proportional-odds assumption using a likelihood-ratio test and a Brant test. Our
sensitivity analyses suggested that coefficients for both fertility intentions and fertility
desires were significantly different, which violates the proportional-odds assumption.
SLM does not require proportional odds and is therefore a more appropriate method for
variables that have unequal distances between response options.
Next, we conducted an ordinary least squares (OLS) regression analysis to
examine the predictors of one’s ideal number of children. Because ‘ideal number of
children’ is a count variable ranging from 0 to 5, we also performed a sensitivity
analysis using a Poisson modeling approach; the results were robust across modeling
strategies. We present findings from the more widely used OLS models because they
are more intuitive to interpret. Finally, we conducted a logistic regression analysis to
estimate associations between infertility statuses and intentions by parity and the odds
of giving birth in the three years between waves of the study.
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1156 http://www.demographic-research.org
3. Results
Results indicate that infertility experience and identity shape fertility intentions and
desires in interesting ways. The means and percentages of fertility intentions, desires for
a baby, and ideal number of children are provided in Table 1 by infertility- and parity-
group statuses. Findings reveal that childless women at the extremes of the infertility
continuum have the highest overall fertility intentions, including women who are not
infertile and those who are both medically infertile and self-identifying as infertile; all
other groups had negative scores on the intentions variable, indicating that the average
response was a no to the question on intentions to have a child. The ‘desire to have a
baby’ and ‘ideal number of children’ variables reveal a different pattern, however.
Childless women in every group had higher scores for wanting to have a baby than
women with children. Women without children who both identify as infertile and meet
medical criteria for infertility, however, had the highest mean score (M=3.36),
indicating their responses to wanting a baby were between “definitely yes” and
“probably yes.” Further, when asked how many children would be ideal for them,
women who already had at least one child reported more children as ideal across every
category of infertility. Among childless women, however, those who experienced
infertility and identified as infertile reported the highest ideal number of children
(M=2.35). At the bivariate level, therefore, it appears that women at the extreme end of
infertility ‒ those who are both medically infertile and identifying as infertile ‒ have
higher fertility intentions, are more likely to report wanting to have a baby, and report a
higher ideal number of children than women in other groups. Women who already had
at least one child at the first wave and were not infertile were significantly more likely
to have given birth by wave two compared to women who only met the medical criteria
and women with children who met the medical criteria and identified as infertile.
Stereotype regression analyses, presented in Table 2, indicate similar patterns.
Regardless of parity, women who have experienced a fertility problem and identify as
having a fertility problem report higher levels of fertility intentions and desire to have a
baby. Fertility status was not associated with significantly higher levels of fertility
intentions for childless women. Among women with children, those who self-identified
only and those who identified and experienced infertility had higher odds of being very
sure that they intend to have more children versus being very sure that they will not
have more children relative to women without fertility problems (OR=2.13; OR=1.64).
Among childless women, the odds of definitely wanting a child versus definitely not
increase significantly by an odds ratio of 7.62. Among mothers, identifying a fertility
problem (with [OR=4.12] and without [OR‒2.34] meeting medical criteria for
infertility) the odds of definitely wanting another child are also significantly higher as
compared to mothers without fertility problems.
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1157
Table 1: Means of dependent and sociodemographic variables by fertility and
parity status group for US women 25‒45 years of age, N=2,978
Childless
1+Child
Not
infertile
Medically
infertile
Self
-ID
infertile
Infertile
& ID
Post hoc
2
Not
infertile
Medically
infertile
Self
-ID
infertile
Infertile
& ID
Post hoc
2
Intentions and d
esires
Fertility intentions .25 ‒.02 ‒.02 .10
‒.52 ‒.63 ‒.41 ‒.39
Want to have a baby 2.93 2.93 2.81 3.36
IS>N,M,S
2.26 2.22 2.48 2.66
IS>N,M
Ideal number of children
2.01
2.04
1.89
2.35
IS>N
2.82
2.97
2.79
2.91
Birth between waves
1
.23
.09
.28
.26
.33
.16
.31
.13
IS<N; M<N
Sociodemographic variables
Age (yrs) 32.44 33.80 34.16 35.09
IS>N
33.77 33.95 35.31 35.58
IS>N; IS>M
Married (%) .31 .38 .35 .54
IS>N
.70 .59 .66 .74
IS<M; M<N
Cohabiting (%) .15 .18 .09 .15
.09 .12 .13 .09
Education (yrs) 15.81 14.88 15.31 14.21
IS<N
13.65 13.41 13.85 13.73
Black (%) .12 .14 .08 .19
.11 .23 .14 .19
IS>N; M>N
Hispanic (%) .08 .13 .11 .12
.26 .25 .19 .19
Other race/ethnicity ( %) .11 .17 .16 .11
.07 .06 .12 .06
N
608
122
79
200
948
573
80
368
Source: Wave 1 & Wave 2 National Survey of Fertility Barriers, N=2,978.
1. Birth between waves reflect a smaller subset of respondents re-interviewed at Wave 2, N=1,547.
2. N= Not infertile, M= Medically infertile only, S= Self-ID infertile only, IS= Medically infertile & Self-ID
Means are presented unless otherwise noted.
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1158 http://www.demographic-research.org
Table 2: Stereotype logistic regression analyses of the association among
fertility-status indicators, sociodemographic characteristics, with
fertility intentions and wants by parity status, N=2,978
Fertility
intentions a
Want to
have a baby b
Childless
1+Child
Childless
1+Child
OR
se
z
OR
se
z
OR
se
z
OR
se
z
Infertility status
Not Infertile (Ref)
Medically infertile only
.67
(.30)
(-.89)
.76
(.18)
(-1.15)
1.20
(.47)
(.48)
.98
(.20)
(-.11)
Self-ID only
.77
(.45)
(-.44)
2.13*
(.80)
(2.01)
1.00
(.46)
(-.00)
2.34*
(.93)
(2.14)
Infertile & Self-ID
1.37
(.56)
(.77)
1.64+
(.44)
(1.84)
7.62***
(3.44)
(4.50)
4.12***
(.96)
(6.07)
Sociodemographic variables
Age (Centered)
.69***
(.02)
(-12.58)
.78***
(.02)
(-8.28)
.80***
(.02)
(-10.43)
.84***
(.02)
(-8.66)
Unmarried (Ref)
Married
2.99**
(1.04)
(3.16)
1.24
(.34)
(.80)
1.68+
(.50)
(1.73)
.86
(.21)
(-.59)
Cohabiting
1.71
(.75)
(1.22)
1.49
(.53)
(1.12)
1.14
(.45)
(.34)
.75
(.24)
(-.88)
Education yrs (Centered)
1.13*
(.06)
(2.06)
1.15*
(.07)
(2.40)
1.03
(.05)
(.55)
1.09*
(.04)
(2.04)
White (Ref)
Black
2.04*
(.74)
(1.97)
.79
(.28)
(-.65)
1.16
(.36)
(.47)
.77
(.25)
(-.80)
Hispanic
1.04
(.41)
(.10)
.83
(.29)
(-.53)
1.46
(.58)
(.95)
1.23
(.31)
(.80)
Other race/ethnicity
2.77
(1.89)
(1.50)
1.10
(.54)
(.19)
2.75
(1.78)
(1.56)
1.11
(.62)
(.18)
N
1009
1969
1009
1969
chi2
182.59***
99.55***
136.80
***
122.43
***
Source: Wave 1 National Survey of Fertility Barriers.
Exponentiated coefficients; Standard errors in parentheses
a. Fertility Intentions Comparison of Extremes: Very Sure Intend vs. Very Sure Not Intending
b. Want to Have Baby Comparison of Extremes: Definitely Yes vs. Definitely No
+ p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1159
In Table 3 we present the linear regression analyses of the association between
fertility status, sociodemographic variables, and the ideal number of children by parity
status. Among childless women, those who were both medically infertile and self-
identified as infertile reported wanting more children than their fertile counterparts,
supporting the notion that infertility may be driving the higher preferences for
childbearing among infertile women. The pattern is not the same among women who
already had children (parity of one or more). Among mothers, those who were
medically infertile wanted significantly more children than fertile women, net of other
factors.
Table 3: OLS regression analyses of the association between fertility,
sociodemographic variables, and ideal number of children by parity
status, N=2,978
Ideal number of children
Childless 1+Child
b
se
b
se
Infertility status
Not infertile (Ref)
Medically infertile only
.06
(.13)
.17*
(.07)
Self-ID only
‒.05
(.16)
.01
(.13)
Infertile & Self-ID
.41***
(.12)
.10
(.08)
Sociodemographic variables
Age (Centered)
‒.05***
(.01)
.00
(.01)
Unmarried (Ref)
Married
.08
(.09)
.17*
(.08)
Cohabiting
‒.09
(.12)
.01
(.11)
Education yrs (Centered)
‒.02
(.02)
‒.05***
(.01)
White (R ef)
Black
.16
(.11)
‒.12
(.10)
Hispanic
.10
(.13)
.16+
(.08)
Other race/ethnicity
.13
(.16)
‒.32**
(.11)
Intercept
1.84***
(.08)
2.64***
(.08)
N
1009
1969
R2
.10
.05
Source: Wave 1 National Survey of Fertility Barriers.
+ p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1160 http://www.demographic-research.org
Are fertility and parity statuses associated with the odds of a birth between waves?
To answer this question, on Table 4, we present odds ratios from multivariate logistic
regression models exploring whether fertility status, intentions, and desires influenced
the likelihood of giving birth between waves, controlling for a rich array of
sociodemographic characteristics. We ran separate sets of models for childless women
and women with children. Our first model examines only fertility status. Among
childless women, those who were medically infertile were 66% less likely to give birth
between waves, relative to fertile women (OR=.34). And among those who were
already mothers, medically infertile women were about 60% less likely to give birth
between waves, relative to fertile women (OR=.38). Likewise, mothers who were both
medically infertile and self-identified as infertile were significantly less likely than
fertile women to have children between waves (OR=.31). Next, we added fertility
intentions and desires in model two. Lastly, our third model includes all variables.
Fertility intentions and desires were positively associated with the likelihood of having
a child between waves. Taking into account fertility intentions and desires largely
reduced the difference between fertile and medically infertile mothers. Conversely,
mothers who experienced infertility and identified as infertile were significantly less
likely to have children, even when we take intentions and desires into account. Among
childless women, those who met the medical criteria only remained significantly less
likely to give birth between waves compared to fertile childless women. Our findings
suggest that birth outcomes differ by parity. Among childless women, those who met
the medical criteria only were significantly less likely to give birth between waves.
Interestingly, we find that among mothers, women who meet medical criteria for
infertility and self-identify as infertile were least likely to meet their expectations,
despite having the highest preferences for having children. As expected, women who
meet the medical criteria for infertility have significantly fewer children on average
than women who do not meet criteria for infertility.
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1161
Table 4: Logistic regression analyses of the association between infertility,
fertility, sociodemographic variables, and birth between waves by
parity status, N=1,547
Childless 1+ Child
m1 m2 m3 m1 m2 m3
OR se OR se OR se OR se OR se OR se
Infertility and fertility variables
Not infertile (Ref)
Medically infertile only
.34*
(.16)
.29*
(.16)
.15**
(.09)
.38***
(.10)
.56+
(.18)
.53+
(.19)
Self-ID only
1.29
(.52)
1.29
(.51)
1.66
(.69)
.92
(.36)
1.20
(.47)
1.57
(.67)
Infertile & Self-ID
1.18
(.35)
1.64
(.64)
1.14
(.45)
.31***
(.08)
.39**
(.12)
.44**
(.13)
Fertility intentions W1
(Centered)
2.28***
(.38)
1.71**
(.32)
1.71**
(.30)
1.47*
(.24)
Want a baby W1 (Centered)
1.63*
(.36)
1.92**
(.49)
1.35+
(.24)
1.37+
(.25)
Ideal number of children
(Centered)
1.23
(.17)
1.19
(.19)
1.19
(.13)
1.16
(.14)
Sociodemographic variables
Age
.90***
(.03)
.91***
(.02)
Unmarried (Ref)
Married
8.32***
(3.00)
4.75***
(2.05)
Cohabiting
3.23**
(1.42)
1.35
(.83)
Education years (Centered)
.92
(.06)
.99
(.05)
White (R ef)
Black
1.06
(.47)
.86
(.29)
Hispanic
2.32+
(1.07)
2.40*
(.84)
Other Race
4.48**
(2.36)
.77
(.46)
Intercept
.30***
(.04)
.13***
(.03)
.03***
(.01)
.49***
(.07)
.34***
(.06)
.07***
(.03)
N
630
630
630
917
917
917
pseudo R2
.02
.24
.36
.04
.18
.26
Source: Wave 1 and Wave 2 National Survey of Fertility Barriers.
Exponentiated coefficients; Standard errors in parentheses.
Birth between waves reflect a smaller subset of respondents re-interviewed at Wave 2, N=1,547.
+ p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001
4. Conclusions
Despite a growing body of literature on infertility/subfecundity in demographic and
social-science fields, little is known about the associations between infertility and
fertility intentions, desires, and birth outcomes. Further, extant research typically fails
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1162 http://www.demographic-research.org
to differentiate between those who meet the medical criteria but do not realize they are
infertile, those who perceive themselves to have a fertility problem although they do not
meet the medical criteria for infertility, and those who both meet the medical criteria
and identify as infertile. A few exceptions are Greil and colleagues’ work on the
“hidden infertile” (Greil et al. 2009) and the link between perceiving a fertility problem
and distress about infertility (Greil et al. 2011b).
The findings presented here highlight the importance of both the experience of
infertility and identification as a person with fertility problems for fertility intentions,
desires, and birth outcomes. Supporting theories of intentions that suggest that an
outcome needs to be certain to formulate an intention (Ajzen and Klobas 2013; Malle,
Moses, and Baldwin 2003), we found that fertility intentions were highest for women
who had never experienced infertility and were childless at the time of the first
interview. Taken at face value, this might suggest that women who meet the medical
criteria or perceive that they are infertile are not trying to get pregnant. Yet when we
explored fertility desires further, we found that women who both meet the medical
criteria and identify as infertile report significantly greater desires to have a baby.
Among childless women, they also have a much higher ideal number of children than
other women in the sample. Of course, simply wanting a birth may not predict one,
especially for women who have met the medical criteria for infertility in the past.
Indeed, our findings indicate that, regardless of parity, women who met the medical
criteria for infertility but did not identify as having a fertility problem at the first survey
wave were less likely to give birth by the second wave. Women who identified as
infertile but did not meet the medical criteria at the first wave were actually more likely
to give births between waves than women who were not infertile, though the difference
was not significant.
These findings suggest that fertility intentions may not be the best predictors of
births for women who have experienced infertility and/or identify as someone with
fertility problems. The uncertainty of whether or not they will be able to give birth may
prevent them from reporting a strong intention to have a baby. Bachrach and Morgan
(2013) have called for greater theory development “at the intersection of cognitive
science, social science, and social demography” (page 480) to better understand fertility
intentions and their realization. In particular, they argue that fertility researchers need
improved understanding of how mental and social phenomena are related to intentions
(Bachrach and Morgan 2013). This study provides an example of when intentions may
not be as meaningful or predictive for fertility outcomes. Because of the substantial
minority of women who experience infertility each year, our findings have important
implications for theory and research regarding fertility intentions.
Our findings also highlight the importance of distinguishing between experiencing
infertility and identifying as infertile; moreover, these classifications of infertility
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1163
matter in different ways for fertility intentions vs. outcomes. Not surprisingly, identity
matters for fertility intentions; women who view themselves as having a fertility
problem, especially when they also have met the medical criteria for infertility, are
reporting lower fertility intentions but greater desires to have a baby and higher family
size preferences. This is problematic because women who believe that they have
fertility problems may be assigned lower values on fertility intentions measures despite
wanting (and perhaps trying) to have a baby, and any births may be misidentified as
unplanned or unwanted when they do give birth. Interestingly, it appears that women
who fail to realize a fertility problem when, in fact, they meet the criteria for infertility,
are least likely to give birth. This finding has important implications as well; this may
be one reason why fertility outcomes do not always match intentions. There are also
important medical implications, as women who do not identify as having fertility
problems may be less likely to seek treatment that could help them achieve a desired
birth (White et al. 2006).
Of course, the results of this study should be interpreted with caution in light of its
limitations. First, respondents’ infertility statuses relied on retrospective reports of
periods of 12 months or more during which respondents had regular, unprotected
heterosexual intercourse with no pregnancy. To our knowledge, there have been no
studies examining the validity of retrospective reports of infertility episodes. Integrating
medical records with survey data would provide a more accurate classification of
infertility episodes. Given our significant findings for infertility experience, however,
we are confident that, if anything, our findings provide conservative estimates. Second,
although the study participants were of reproductive age and not surgically sterilized,
the timing of the infertility episode(s) varied; in some cases, the infertility episodes
occurred more than 10 years before the survey interview. Again, we suspect that, if
anything, this might make our findings more conservative than a study that includes
only women currently experiencing infertility. Due to the much smaller percentage of
women who experience infertility in a given year as compared to women who ever
experience infertility, it is difficult to assess these relationships with such small
samples. Another limitation related to this issue of small cell counts when creating four
groups of infertility statuses is that we were only able to utilize two groups of parity
statuses (women with and without children).
Future research should explore infertility-group membership and possible
selectivity. It is unclear, for example, why many women who have not met the medical
criteria for infertility perceive themselves as infertile. It is also unclear why some
women meet the medical criteria for infertility but do not recognize that they might
have a fertility problem. Future prospective data-collection efforts are encouraged to
include questions that allow for the classification of women into infertility categories,
including identity and experience, and follow women over time to more definitively
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1164 http://www.demographic-research.org
illustrate the ways in which infertility affects fertility. Future research is also
encouraged to investigate the reverse relationship between fertility intentions, wants,
and ideals and perception of a fertility problem. The meaning of the period of infertility
may matter for outcomes; not all women who are not contracepting are actively
“trying” to get pregnant (Johnson et al. 2014). Whereas some women may experience a
relatively long period of unprotected sex without conception and not consider it a
problem, those who are hoping or trying for a pregnancy may self-identify a problem
before they meet the medical criteria for infertility
Despite limitations, our findings suggest that women who meet the medical criteria
for infertility and/or identify as having a fertility problem are a special case for fertility-
intentions and -outcomes research and theory. This study highlights the need for more
in-depth probing of fertility plans, desires, and contraceptive behaviors when women
believe that they may not be able to conceive. Simply asking these women if they
intend to give birth may be an inadequate question.
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1165
References
Ajzen, I. and Klobas, J. (2013). Fertility intentions: An approach based on the theory of
planned behavior. Demographic Research 29(8): 203‒232. doi:10.4054/Dem
Res.2013.29.8.
American Society for Reproductive Medicine (ASRM). (2008). Definitions of infertility
and recurrent pregnancy loss. Fertility and Sterility 90(5) supplement: S60.
doi:10.1016/j.fertnstert.2008.08.065.
Bachrach, C.A. and Morgan, S.P. (2014). A cognitive‒social model of fertility
intentions. Population and Development Review 39(3): 459‒485. doi:10.1111/
j.1728-4457.2013.00612.x.
Bongaarts, J. and Potter, R.E. (2013). Fertility, biology, and behavior: An analysis of
the proximate determinants. New York: Academic Press.
Davis, K. and Blake, J. (1956). Social structure and fertility: An analytic framework.
Economic development and cultural change 4(2): 211‒235.
Greil, A.L. and McQuillan, J. (2010). ‘Trying times’: Medicalization, intent, and
ambiguity in the definition of infertility. Medical Anthropology Quarterly 24(2):
137–156. doi:10.1111/j.1548-1387.2010.01094.x.
Greil, A.L., McQuillan, J., Johnson, K., Blevins-Slauson, K., and Shreffler, K.M.
(2009). The hidden infertile: Infertile women without pregnancy intent in the
United States. Fertility and Sterility 93(6): 2080‒2083. doi:10.1016/j.fert
nstert.2009.08.024.
Greil, A.L., McQuillan, J., Lowry, M., and Shreffler, K.M. (2011a). Infertility treatment
and fertility specific distress: A longitudinal analysis of a population-based
sample of U.S. women. Social Science and Medicine 73(1): 87‒94. doi:10.1016/
j.socscimed.2011.04.023.
Greil, A.L., Shreffler, K.M., Schmidt, L., and McQuillan, J. (2011b). Variation in
distress among women with infertility: Evidence from a population-based
sample. Human Reproduction 26(8): 2101‒2112. doi:10.1093/humrep/der148.
Johnson, K.M., McQuillan, J., Greil, A.L., and Shreffler, K.M. (2014). Towards a more
inclusive framework for understanding fertility barriers. In: Nash, M. (ed.).
Reframing reproduction: Sociological perspectives on gender, sexuality, and
reproduction in late modernity. New York: Palgrave Macmillan: 23‒38.
doi:10.1057/9781137267139_2.
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1166 http://www.demographic-research.org
Long, S.J. and Freese, J. (2006). Regression models for categorical dependent variables
using Stata. 2nd ed. College Station: Stata Press.
Maill, C.E. (1994). Community constructs on involuntary childlessness: Sympathy,
stigma, and social support. Canadian Review of Sociology and Anthropology
31(4): 392‒421. doi:10.1111/j.1755-618X.1994.tb00828.x.
Malle, B.F., Moses, L.J., and Baldwin, D.A. (2003). Intentions and intentionality:
Foundations of social cognition. Cambridge: MIT Press.
Mascarenhas, M.N., Cheung, H., Mathers, C.D., and Stevens, G.A. (2012). Measuring
infertility in populations: Constructing a standard definition for use with
demographic and reproductive health surveys. Population Health Metrics
10(17): 1‒11. doi:10.1186/1478-7954-10-17.
Maximova, K. and Quesnel-Vallée, A. (2009). Mental health consequences of
unintended childlessness and unplanned births: Gender differences and life
course dynamics. Social Science & Medicine 68(5): 850‒857. doi:10.1016/
j.socscimed.2008.11.012.
McQuillan, J., Greil, A.L., Shreffler, K.M., and Bedrous, A.V. (2015). The importance
of motherhood and fertility intentions among women in the United States.
Sociological Perspectives 58(1): 20‒35. doi:10.1177/0731121414534393.
Polis, C.B. and Zabin, L.S. (2012). Missed conceptions or misconceptions? Perceived
infertility among unmarried young adults in the United States. Perspectives on
Sexual and Reproductive Health 44(1): 30‒38. doi:10.1363/4403012.
Rutstein, S.O. and Shah, I.H. (2004). Infecundity, infertility, and childlessness in
developing countries. DHS Comparative Reports No. 9. Calverton: ORC Macro
and the World Health Organization.
Schoen, R., Astone, N.M., Kim, Y.T., Nathanson, C.A., and Fields, J.M. (1999). Do
fertility intentions affect fertility behavior? Journal of Marriage and Family
61(3): 790‒799. doi:10.2307/353578.
Schwerdtfeger, K.L. and Shreffler, K.M. (2009). Trauma of pregnancy loss and
infertility among mothers and involuntarily childless women in the United
States. Journal of Loss and Trauma 14(3): 211‒227. doi:10.1080/153250208
02537468.
Simonsen, S.E., Baksh, L., and Stanford, J.B. (2012). Infertility treatment in a
population-based sample: 2004‒2005. Maternal and Child Health Journal 16(4):
877‒886. doi:10.1007/s10995-011-0809-6.
Demographic Research: Volume 35, Article 39
http://www.demographic-research.org 1167
Stover, J. (1998). Revising the proximate determinants of fertility framework: What
have we learned in the past 20 years? Studies in Family Planning 29(3): 255‒
267. doi:10.2307/172272.
Thoma, M.E., McLain, A.C., Louis, J.F., King, R.B., Trumble, A.C., Sundaram, R., and
Buck Louis, G.M. (2013). The prevalence of infertility in the United States as
estimated by the current duration approach and a traditional constructed
approach. Fertility and Sterility 99(5): 1324‒1331. doi:10.1016/j.fertnstert.2012.
11.037.
Thomson, E. (1997). Couple childbearing desires, intentions, and births. Demography
34(3): 343‒354. doi:10.2307/3038288.
White, L., McQuillan, J., Greil, A.L., and Johnson, D.L. (2006). Infertility: Testing a
helpseeking model. Social Science & Medicine 62(4): 1031‒1041. doi:10.1016/
j.socscimed.2005.11.012.
Shreffler et al.: Infertility and fertility intentions, desires, and outcomes among US women
1168 http://www.demographic-research.org
