These findings describe how unintended pregnancy may change based on the economic environment. This may be measuring fluctuations in prospective intentions that lead to different contraceptive or sexual behavior, or a change in intentions once a pregnancy has occurred. For example, if a woman experiences an unplanned pregnancy in good economic times, she may be more likely to report that pregnancy as wanted, whereas a pregnancy that comes in times of economic uncertainty may be more likely to be classified as mistimed. This paper does not explore overall risk of pregnancy, but rather the odds of unintended pregnancy relative to intended pregnancy. Risk aversion, through income or uncertainty effects, may reduce sexual behavior or increase contraceptive use in times of economic uncertainty, leading to lower pregnancy rates, as shown in the economic and demographic literature. My findings add to this line of research by showing that women who become pregnant during recessions are less likely to intended to do so than those becoming pregnant at other times. My results can be reconciled with those of Percheski and Kimbo by assuming that the relative composition of pregnancies changed even if the numbers decreased. My results may be produced by an increase in mistimed pregnancies,cannabis grow racks or a reduction in intended pregnancies. My findings are consistent with the possibility that reported reductions in fertility during the Great Recession may, in part, arise from increases in the likelihood of abortion and/or miscarriage. This finding adds to the individual literature on abortion decision-making, which finds financial concerns among the most commonly cited reasons for abortion decisions.
The abortion rate in the US gradually declined from 1990 to 2005, leveled off until 2008, and began declining again through 2011. Jones and Jerman speculate that the economy may contribute to the declines in both the birth and abortion rate between 2008-2011, indicating that fewer women got pregnant in this period. Greater declines in intended pregnancy, as women and couples revise fertility plans, would likely cause a greater decrease in number of births than abortions. The significant effect for miscarriage could indicate a biological mechanism of increases in spontaneous abortion with stressful economic conditions, as has been suggested by previous work.This work posits that pregnant women non-consciously evaluate the likelihood of success for their offspring under varying environmental conditions. When the environment poses more of a threat, in this case through scarce resources, fewer gestations survive to birth. My results of increased odds of miscarriage compared to live births support this argument. Alternatively, the increase in reported miscarriages may arise, at least in part, from misclassification. It is well documented that the stigma of induced abortion leads some women to instead report spontaneous abortion on surveys, an outcome that does not carry the same stigma in the United States. If stigma is also trending in time, as it is reflective of the political environment, there may be differential misclassification that may bias the result. For example, if anti-abortion policies, which have increased since 2010, were also correlated in time with the economic uncertainty measures, this may have increased stigma about reporting abortion. The fact that all women were interviewed over the same period between 2006-2010 and reported abortions retrospectively makes this issue less of a concern in this analysis. Additionally, research has shown that these restrictions have little impact on the abortion rate. There are several limitations to this analysis.
As noted, if women failed to report a pregnancy that ended in induced abortion or reported it as a miscarriage instead, my results for the effect on odds of induced abortion would be conservative and odds of miscarriage would be inflated. Secular trends may be are a concern of the analysis. I chose to model the economic variables as their absolute value given the hypothesis that it is the absolute rate of uncertainty that would affect decision-making, rather than a deviation from expected. However it is possible that the models captures autocorrelation shared by the independent and dependent variable. This paper examines measures of the national economy. It is possible that regional data would suggest a different or more complex relationship than I found. It may be that states with high unemployment would be most affected by changes in the unemployment rate, because more people are sensitive to shocks in that region. Conversely, it could be hypothesized that those same states would be less affected, because the relative change in unemployment may be less significant as compared to lower endemic rates. A recent study found that states with the highest income inequality also had the highest rates of unintended pregnancy, but it is unclear how an economic shock may affect these populations . NSFG does not include questions about individual unemployment or economic stress at the time of pregnancy, thus preventing further control or examination of individual circumstances. It is possible that economic uncertainty would have different effects at different times in a woman’s pregnancy. For instance, increases in unemployment several months before pregnancy, may impact contraceptive use and sexual behavior, while unemployment in the first trimester could alter risk of miscarriage or decision to have an abortion. Economic changes at any point during pregnancy could alter women’s likelihood of reporting that pregnancy as mistimed or unwanted at conception.
I chose to assign economic uncertainty measures during the month of conception to align with the framing of the pregnancy intention questions. Future work could consider if there is a particularly sensitive period in pregnancy to these effects. These findings add to the literature on the economy and fertility. They also inform public health discussions of the determinants of unintended pregnancy. Income effects from economic downturns may extend to contraceptive use, preventing some women from using methods despite desire not to become pregnant.Alcohol is a known teratogen that causes fetal alcohol syndrome and fetal alcohol spectrum disorders, as well as a range of other harms to fetuses . Alcohol use during pregnancy is common, with approximately 21% of pregnant women reporting any alcohol use and approximately 3% reporting binge drinking in the United States . Since 1974, almost all states have enacted policies targeting alcohol use during pregnancy . These include both supportive and punitive policies. In 2016, Priority Treatment for Pregnant Women and Women with Children was the least common policy and Reporting for Treatment and Data Purposes and Mandatory Warning Signs were the most common. States continue to change their alcohol and pregnancy policies each year . Despite the proliferation of these state-level policies, few studies have assessed what impact, if any, they have . While the purpose of these policies is typically unstated,cannabis drying racks it is reasonable to assume that a primary intended purpose is to reduce alcohol use during pregnancy and thereby improve birth outcomes and longer term child well-being. For example, a recent study examining the effects of state-level policies that mandate posting of warning signs about harms due to drinking during pregnancy in locations that sell alcohol finds some support for this assumption. Specifically, this study used data from a variety of sources from 1989-2006 for selected states and found that MWS policies may be associated with less alcohol use during pregnancy, and are associated with fewer very low birthweight and very preterm births . However, to our knowledge, this is the only study to have documented a positive effect of a state-level policy targeting alcohol use during pregnancy. Another recent study found that shorter waiting time for substance use disorder treatment is associated with treatment completion for pregnant women , which implies that priority treatment could increase the number of pregnant women in need of treatment that actually receive and complete treatment; this could improve birth outcomes, though this has not yet been studied. Other research suggests that state-level policies targeting alcohol use during pregnancy might not have the intended effects, and might actually have unintended consequences. A qualitative study about barriers to prenatal care for pregnant women who used alcohol and drugs found that women who use drugs during pregnancy avoided prenatal care both out of fear that they would discover that their use had already irreversibly damaged their baby, and out of fear that their providers would report them to Child Protective Services and they would lose their children or go to jail . While the women in the qualitative study explicitly described these experiences related to drug and not alcohol use, it is plausible that they apply to alcohol as well. Thus, policies that require informing women that their substance use may have already harmed their fetus, such as MWS policies, could lead women to avoid prenatal care.
Policies that mandate reporting to CPS, that define alcohol use during pregnancy as child abuse/neglect, or that allow civil commitment for alcohol use during pregnancy could also lead women to avoid prenatal care. A positive association between prenatal care utilization and birth outcomes has been documented , and if pregnant women who drink alcohol avoid prenatal care, prenatal care providers miss opportunities to provide other health promoting interventions that 1) support women to reduce or stop drinking, 2) provide other important components of prenatal care, such as monitoring for pre-eclampsia, and 3) link them to other supportive services. Punitive policies that lead pregnant women who drink to avoid prenatal care could thus increase the chances of adverse birth outcomes. In addition, policy contexts that allow criminal justice prosecutions or require CPS reporting could also influence effectiveness of alcohol-related interventions such as screening and brief interventions, which are widely recommended for pregnant women, including at-risk drinkers . Screening in an environment where being reported to CPS is a possible outcome from disclosing substance use may make women less likely to disclose use to providers and thus less likely to get support and services to help them reduce their use . To date, however, there has been no comprehensive research examining whether and how either supportive or punitive state-level policies targeting alcohol use during pregnancy are associated with birth outcomes and prenatal care utilization. Research on this topic is crucial because 1) policies continue to be debated and enacted in individual states , 2) the federal government is now incorporating them in federal legislation , and 3) some of these laws are being challenged in state court . In addition, findings from research examining the effects of policies targeting alcohol use during pregnancy can help inform how state policymakers respond to opioid and cannabis use during pregnancy, which are timely given the opioid crisis and legalization of both recreational and medical cannabis in several states. This study combines state- and individual-level data to examine associations between state-level policies targeting alcohol use during pregnancy and birth outcomes across 50 states over 42 years. We hypothesize that each supportive policy will be associated with decreased negative birth outcomes and each punitive policy will be associated with increased negative birth outcomes. We also hypothesize that each individual punitive policy will be associated with decreased prenatal care utilization, while prohibitions against criminal prosecution will be associated with increased prenatal care. We do not expect to see associations between mandatory warning signs and prenatal care or between priority treatment and prenatal care because we do not foresee them contributing to an environment of trust or mistrust between women and providers. If there are associations, we expect associations with increased prenatal care because they may lead women to be more motivated to seek information from providers or more able to get treatment and thus have more support to engage in prenatal care. Primary outcomes were low birthweight and premature birth . Secondary outcomes were any prenatal care utilization, late prenatal care utilization , inadequate prenatal care, and an APGAR score ≥ 7. All outcome data came from birth certificates. We also took steps to address changes in data collection over time. For example, prior to 1980, NCHS did not impute continuous gestational age when the last menstrual period day was unavailable. After 1980, NCHS began imputing gestational age when the last menstrual period day was unavailable. We applied this imputation method to1972-1980 data to be able to have more complete data to construct the adequacy of prenatal care variable .