The pandemic preterm paradox: A test of competing explanations.

Epidemiologists have long argued that side effects of the stress response include preterm birth. Research reports that fear of lethal infection stressed pregnant persons at the outset of the COVID-19 pandemic and that "shutdowns" and "social distancing" impeded access to social support and prenatal care. The decline in preterm births in high-income countries, including the United States (US), during the early months of the pandemic therefore poses a paradox for science. Explanations of this "pandemic preterm paradox" remain untested. We apply time-series modeling to data describing 80 monthly conception cohorts begun in the US from July 2013 through February 2020 to determine which of 3 explanations most parsimoniously explains the paradox. We infer that "prior loss," or the argument that an increase in spontaneous abortions and stillbirths depleted the population of fetuses at risk of preterm birth, best explains data currently available. We describe the implications of these results for public health practice.

Collective Optimism and Selection Against Male Twins in Utero.

Scholarly literature claims that health declines in populations when optimism about investing in the future wanes. This claim leads us to describe collective optimism as a predictor of selection in utero. Based on the literature, we argue that the incidence of suicide gauges collective optimism in a population and therefore willingness to invest in the future. Using monthly data from Sweden for the years 1973-2016, we test the hypothesis that the incidence of suicide among women of child-bearing age correlates inversely with male twin births, an indicator of biological investment in high-risk gestations. We find that, as predicted by our theory, the incidence of suicide at month t varies inversely with the ratio of twin to singleton male births at month t + 3. Our results illustrate the likely sensitivity of selection in utero to change in the social environment and so the potential for viewing collective optimism as a component of public health infrastructure.

Cohort Variation in Selection During Pregnancy and Risk of Selected Birth Defects Among Males.

BACKGROUND: The literature theorizes, but does not test, that variation over time in selective loss in utero affects the observed count of live-born birth defects cases. We test the hypothesis that the risk of birth defects among live-born males varies inversely with the strength of selection against males in utero. METHODS: We identified a subset of six birth defect phenotypes among males from the California Birth Defects Monitoring Program, an active surveillance system for over 490,000 male singletons born in eight California counties from 1986 to 2004. We assigned each birth defect case infant to a monthly conception cohort at risk of selection in utero. We used the monthly sex ratio at birth (M:F), derived from each conception cohort, as the indicator of selection against males. We analyzed the odds ratio of birth defects with both individual-level logistic regression and aggregate time-series methods. RESULTS: Consistent with selection in utero, male infants from conception cohorts with low outlying sex ratios (i.e., stronger selectivity) exhibit fewer than expected birth defects (adjusted odds ratio [OR] = 0.86; 95% confidence interval [CI] = 0.76, 0.98). Aggregate time-series tests also yield similar findings (OR = 0.81; 95% CI = 0.71, 0.90). CONCLUSIONS: Our findings among males indicate that variation in the strength of selectivity in utero accounts for a portion of observed cohort differences in morbidity due to birth defects. These findings suggest that "revealed prevalence" of morbidity across birth cohorts varies, at least in part, from selective loss in utero. See video abstract at, http://links.lww.com/EDE/B209.

Social and demographic drivers of trend and seasonality in elective abortions in Denmark.

BACKGROUND: Elective abortions show a secular decline in high income countries. That general pattern, however, may mask meaningful differences-and a potentially rising trend-among age, income, and other racial/ethnic groups. We explore these differences in Denmark, a high-income, low-fertility country with excellent data on terminations and births. METHODS: We examined monthly elective abortions (n = 225,287) from 1995 to 2009, by maternal age, parity, income level and mother's country of origin. We applied time-series methods to live births as well as spontaneous and elective abortions to approximate the denominator of pregnancies at risk of elective abortion. We used linear regression methods to identify trend and seasonal patterns. RESULTS: Despite an overall declining trend, teenage women show a rising proportion of pregnancies that end in an elective termination (56% to 67%, 1995 to 2009). Non-Western immigrant women also show a slight increase in incidence. Heightened economic disadvantage among non-Western immigrant women does not account for this rise. Elective abortions also show a sustained "summer peak" in June, July and August. Low-income women show the most pronounced summer peak. CONCLUSIONS: Identification of the causes of the increase over time in elective abortion among young women, and separately among non-Western immigrant women, represents key areas of further inquiry. The unexpected increase over time in elective abortions among teens and non-Western immigrants in Denmark may signal important social and cultural impediments to contraception. The summer peak in abortions among low-income women, moreover, conflicts with the conventional assumption that the social and demographic composition of mothers who electively end their pregnancy remains stable within a calendar year.

Spontaneous Pregnancy Loss in Denmark Following Economic Downturns.

An estimated 11%-20% of clinically recognized pregnancies result in spontaneous abortion. The literature finds elevated risk of spontaneous abortion among women who report adverse financial life events. This work suggests that, at the population level, national economic decline-an ambient and plausibly unexpected stressor-will precede an increase in spontaneous abortion. We tested this hypothesis using high-quality information on pregnancy and spontaneous loss for all women in Denmark. We applied time-series methods to monthly counts of clinically detected spontaneous abortions (n = 157,449) and the unemployment rate in Denmark beginning in January 1995 and ending in December 2009. Our statistical methods controlled for temporal patterns in spontaneous abortion (e.g., seasonality, trend) and changes in the population of pregnancies at risk of loss. Unexpected increases in the unemployment rate preceded by 1 month a rise in the number of spontaneous abortions (ß = 33.19 losses/month, 95% confidence interval: 8.71, 57.67). An attendant analysis that used consumption of durable household goods as an indicator of financial insecurity supported the inference from our main test. Changes over time in elective abortions and in the cohort composition of high-risk pregnancies did not account for results. It appears that in Denmark, ambient stressors as common as increasing unemployment may precede a population-level increase in spontaneous abortion.

Twinning in Norway Following the Oslo Massacre: Evidence of a 'Bruce Effect' in Humans.

Emerging theory and empirical work suggest that the 'Bruce Effect', or the increase in spontaneous abortion observed in non-human species when environments become threatening to offspring survival, may also appear in humans. We argue that, if it does, the effect would appear in the odds of twins among male and female live births. We test the hypothesis, implied by our argument, that the odds of a twin among male infants in Norway fell below, while those among females rose above, expected levels among birth cohorts in gestation in July 2011 when a deranged man murdered 77 Norwegians, including many youths. Results support the hypothesis and imply that the Bruce Effect operates in women to autonomically raise the standard of fetal fitness necessary to extend the gestation of twins. This circumstance has implications for using twins to estimate the relative contributions of genes and environment to human responses to exogenous stimuli.

Does the Implementation of Reference Pricing Result in Reduced Utilization? Evidence From Inpatient and Outpatient Procedures.

Reference pricing (RP) is an insurance design that can be used to incentivize patients to use low-price settings. While RP is not intended to affect overall utilization, it could unintentionally reduce utilization. We examined whether utilization was reduced when a large employer adopted RP for selected elective surgeries, including inpatient joint replacement surgery and outpatient cataract surgery, colonoscopy, and arthroscopic surgery. Data included a treatment group subject to RP implementation and a comparison group that was not. We applied autoregressive integrated moving average analysis as comparison-population interrupted time-series analysis to determine whether there were procedure reductions following RP implementation. We find no evidence of short-term decreases (within 3 months of RP implementation). However, we find very modest declines of approximately 14 (20%) fewer arthroscopic knee surgeries 6 months after RP implementation and 129 (17.2%) fewer colonoscopies 8 months after RP implementation. There were no declines in the other procedures examined.

Transient cultural influences on infant mortality: Fire-Horse daughters in Japan.

OBJECTIVES: Parental investment theory suggests that the quality and quantity of parental care depends, in part, on assessments of whether offspring will survive and yield grandchildren. Consistent with this theory, we hypothesize that parental perception that a birth cohort will have low reproductive success coincides with higher than expected infant mortality in the cohort. We test this hypothesis in industrialized Japan in 1966 when cultural aversion to females born in the astrological year of the Fire-Horse may have jeopardized the life of female infants. METHODS: We applied time-series methods to cohort infant mortality data for Japan, from 1947 to 1976, to test whether female infant mortality in 1966 rose above levels expected from history, male infant mortality, and fertility. Methods control for the secular decline in infant mortality as well as other temporal patterns that could induce spurious associations. RESULTS: Findings support the hypothesis in that female infant mortality rises by 1.1 deaths per 1,000 live births above expected levels (coefficient = 0.0011; standard error = 0.0005; P = 0.03). The result indicates an excess of 721 female infant deaths statistically attributable to the Fire-Horse year. Findings remain robust to control for male infant mortality and the secular decline in mortality over the test period. CONCLUSIONS: The discovery of a predictable, acute increase in female infant mortality during the Fire-Horse year supports the relevance of parental investment theory to developed countries. Results should encourage further research on the health sequelae of abrupt, population-level shifts in culture.

A sex-specific test of selection in utero.

We find support for the hypothesis that changes in the monthly odds of a twin among live-born males predict subsequent and opposite changes in the odds of a twin among live-born females. The hypothesis arises from the long standing argument that natural selection has conserved mechanisms by which pregnant women in stressed populations spontaneously abort fetuses least likely to yield grandchildren. Previous attempts to empirically test this argument focus almost entirely on males. We contribute to the literature by showing that, consistent with the logic of natural selection, maternal adaptations to environmental changes likely have effects on the survival of both male and female conceptuses and fetuses.

Economic antecedents of sudden infant death syndrome.

PURPOSE: To test the hypothesis that labor market contraction is associated with an elevated number of deaths due to sudden infant death syndrome (SIDS). METHODS: We apply time-series methods to monthly counts of SIDS deaths and total employment from the state of California beginning January 1989 and ending December 2001. The methods control for trends, seasonal cycles, and other forms of autocorrelation that could induce spurious associations. RESULTS: Decreases in the number of employed persons in California preceded higher than expected monthly values of SIDS cases among black, non-Hispanic White and Hispanic infants. In addition, Blacks and Hispanics appear to respond more strongly than non-Hispanic Whites to economic contraction. CONCLUSIONS: We infer support for the hypothesis that economic contraction may inhibit salutary behavior related to SIDS. We discuss various mechanisms through which the economy may affect SIDS and recommend further investigation.

Rates of unemployment and incidence of police-initiated examinations for involuntary hospitalization in Florida.

OBJECTIVE: This study tested the hypothesis that contraction of regional economies affects the incidence of involuntary admissions to psychiatric emergency services by reducing community tolerance for persons perceived as threatening to others. METHODS: This hypothesis was tested with time-series analyses of the relationship between initial claims for unemployment in Florida between July 4, 1999, and June 28, 2003, and the weekly number of men and women presented by police to be examined for involuntary psychiatric hospitalization because of danger to others. The analyses controlled for admissions presented by mental health professionals because of danger to others and for admissions presented by police because of neglect or disability. RESULTS: When the analyses controlled for autocorrelation and other covariates, claims for unemployment insurance were significantly associated with the number of men presented by police for danger to others. During the study period, police presented 5,897 men for examination because of danger to others. Increased unemployment claims were associated with approximately 309 more men being presented for examination than expected from prior presentation rates and from the number presented by mental health professionals for danger to others and by police for neglect or disability. No such association was found for women. CONCLUSIONS: Consistent with theory, this study found that presentations for involuntary admission to psychiatric services increased after contractions in the labor market. Combining the methods of this study with econometric forecasting may allow providers to anticipate better the need for psychiatric services.

Economic antecedents of the Swedish sex ratio.

Literature describing environmental influences on human conception and gestation implies that the ratio of male to female live births should vary positively over time with the population's ability to produce and distribute goods and services. No direct test of this hypothesis appears in the literature despite its apparent importance in understanding the biological implications of collective choices. We offer a test based on Swedish data for the years 1862 through 1991. The results support the hypothesis. We argue that the findings have implications for basic science as well as for the debate over the biological effects of social policy.

Involuntary psychiatric examinations for danger to others in Florida after the attacks of September 11, 2001.

OBJECTIVE: Theories of perceived risk state that when people feel threatened, they will react more strongly than they would otherwise. This study tested the hypothesis that evaluations for involuntary psychiatric hospitalizations that were initiated by law enforcement personnel in Florida increased in the weeks after the attacks of September 11, 2001. METHODS: The authors applied interrupted time-series designs to determine whether there was a relationship between the number of involuntary psychiatric examinations initiated by law enforcement officials and the attacks of September 11, 2001. They examined the number of psychiatric evaluations of men and women who were considered to be mentally ill and harmful to others by law enforcement personnel in Florida during seven-day periods ("areal" weeks) that began with Tuesday, July 6, 1999, and ended with Monday, December 31, 2001 (because September 11, 2001, fell on a Tuesday). RESULTS: Over the 130 weeks of the study, law enforcement officials initiated examinations of an average of 25.96 men and 13.47 women per areal week. Law enforcement officials initiated examinations of approximately 14 more women than expected in the areal week that began with September 11, 2001. During the three areal weeks that began with September 18, 2001, a total of 34 more men than expected were presented for evaluation. These findings cannot be attributed to trends, seasonality, other cycles, or the tendency of the examination time series to remain elevated or depressed after high or low values in the series. CONCLUSIONS: Perceived general risk in a community may increase the likelihood that law enforcement personnel and the persons who summon them perceive persons with mental illness as imminently harmful. The public health response to any future terror attacks should include efforts to alert psychiatric service providers to the possibility of lower community tolerance for mental illness in the aftermath of an attack.

The impact of capitated financing on psychiatric emergency services.

OBJECTIVE: Studies of capitated financing of mental health services have generally focused on the cost and utilization of services. Relatively little research has addressed whether capitation has an impact on the effectiveness of the mental health system as a whole. This study examined the impact of capitation on hospital emergency department visits, a widely cited indicator of the effectiveness of the other components of the system. METHODS: In 1995 Colorado's Medicaid program instituted capitation for mental health services in two areas of the state, one in which reimbursement of not-for-profit providers was directly capitated and another in which not-for-profit providers partnered with a for-profit managed behavioral health organization. The analysis examined variation over time in the number of emergency department visits by adults who had a primary mental or substance use disorder. Using interrupted time-series methods, visits in areas where reimbursement was capitated were compared with visits in areas where providers continued to be reimbursed on a fee-for-service basis. A total of 105 weeks for each area was examined; capitation was implemented at week 53. RESULTS: The number of psychiatric emergencies treated in capitated areas declined by 814 (28 percent) below the 2,908 psychiatric emergencies expected from trends, cycles, and levels in fee-for-service areas. Findings were similar for for-profit and not-for-profit areas. The decrease persisted through the end of the first year after capitation. CONCLUSIONS: In Colorado the implementation of capitation was associated with a sustained decrease in utilization of psychiatric emergency services provided by hospital emergency departments. Our findings suggest that capitation does not necessarily reduce the quality of care provided to clients.

Post-term birth as a response to environmental stress : The case of September 11, 2001.

BACKGROUND AND OBJECTIVES: Despite growing interest in the role of maternal psychosocial stress as a determinant of preterm birth, no existing work has examined the relation between maternal stress and post-term birth (≥42 weeks). We hypothesize that prolonging gestation past term may represent an adaptive strategy to a suboptimal environment. METHODOLOGY: We examined the relationship between exposure to the September 2001 terrorist attacks and odds of post-term birth in California. We calculated the expected odds of post-term birth among conception cohorts of singleton gestations in California between October 1996 and November 2005. We used time series analysis to test for higher than expected odds of post-term birth among the 10 cohorts exposed to the attacks of September 2001 (those conceived from December 2000 to September 2001). RESULTS: The observed odds of post-term delivery among gestations at 33-36 weeks in September 2001 were higher than statistically expected for all race/ethnic and sex groups. CONCLUSIONS AND IMPLICATIONS: Our finding that odds of post-term birth were higher than expected among pregnancies exposed to the September 2001 terrorist attacks in late gestation provides initial support for the hypothesis that exposure to a psychosocial stress during pregnancy may result in prolonged gestation.

Unemployment and the detection of early stage breast tumors among African Americans and non-Hispanic whites.

PURPOSE: To test the hypothesis that high unemployment predicts reduced detection of local breast tumors among African American and non-Hispanic white women in the Detroit, Michigan and Atlanta, Georgia SEER catchment areas. METHODS: We test the hypothesis with data for the 156 months from January 1985 through December 1997. RESULTS: In situ and local breast tumors in African American and non-Hispanic white women were less likely to be detected during periods of high unemployment. CONCLUSIONS: Contracting labor markets may impede women with symptoms from getting proper medical attention or distract women from discovering symptoms they would otherwise detect. African American women appear at greatest risk of having a tumor going undetected by virtue of labor market performance.

Psychiatric emergencies after the terrorist attacks of September 11, 2001.

OBJECTIVE: Population surveys suggest that the events of September 11, 2001, resulted in psychiatric emergencies in U.S. communities. This study tested the extent of such emergencies in San Francisco. METHOD: S: Interrupted time-series designs were applied to counts of emergency calls to the police during the 424-day period beginning January 1, 2001, and of voluntary and coerced admissions to psychiatric emergency services during the 1620-day period beginning July 1, 1997. RESULTS: The number of men and women who were coerced into treatment increased significantly on Thursday, September 13, but the number of voluntary admissions was as expected. The number of telephone calls from citizens that police dispatchers judged to be mental health related increased significantly on Wednesday, September 12, and remained elevated through September 13. Several additional analyses were conducted to test the stability of the findings, and the results were essentially unchanged. CONCLUSIONS: The events of September 11 may not have induced emergent mental illness in U.S. communities at relatively great distance from the attacks. However, it is possible that persons with severe mental illness were either more evident to or less tolerated by the community.

Ambient temperature during gestation and cold-related adult mortality in a Swedish cohort, 1915-2002.

For all climatic regions, mortality due to cold exceeds mortality due to heat. A separate line of research indicates that season of birth predicts lifespan after age 50. This and other literature implies the hypothesis that ambient temperature during gestation may influence cold-related adult mortality. We use data on over 13,500 Swedes from the Uppsala Birth Cohort Study to test whether cold-related mortality in adulthood varies positively with unusually benign ambient temperature during gestation. We linked daily thermometer temperatures in Uppsala, Sweden (1915-2002) to subjects beginning at their estimated date of conception and ending at death or the end of follow-up. We specified a Cox proportional hazards model with time-dependent covariates to analyze the two leading causes of cold-related death in adulthood: ischemic heart disease (IHD) and stroke. Over 540,450 person-years, 1313 IHD and 406 stroke deaths occurred. For a one standard deviation increase in our measure of warm temperatures during gestation, we observe an increased hazard ratio of 1.16 for cold-related IHD death (95% confidence interval: 1.03-1.29). We, however, observe no relation for cold-related stroke mortality. Additional analyses show that birthweight percentile and/or gestational age do not mediate discovered findings. The IHD results indicate that ambient temperature during gestation--independent of birth month--modifies the relation between cold and adult mortality. We encourage longitudinal studies of the adult sequelae of ambient temperature during gestation among populations not sufficiently sheltered from heat or cold waves.

The contribution of attenuated selection in utero to small-for-gestational-age (SGA) among term African American male infants.

Natural selection conserves mechanisms allowing women to spontaneously abort gestations least likely to yield fit offspring. Small gestational size has been proposed as an indicator of fitness observable by maternal biology. Previous research suggests that exposure to ambient stress in utero results in more "culling" of small fetuses and therefore lower rates of small-for-gestational-age (SGA). However, African American women persistently have higher rates of SGA than non-Hispanic white women, despite experiencing more ambient stress. This paper tests whether attenuation of the stress response among highly stressed African American women, as suggested by the weathering hypothesis, may help to explain this apparent inconsistency. We apply time-series modeling to over 2 million African American and non-Hispanic white male term births in California over the period of January 1989 through December 2010. We test for the parabolic (i.e., "U" shaped) relationship, implied by an attenuated stress response, between unusually strong labor market contraction and the rate of SGA among African American term male infants, and a linear relationship among non-Hispanic whites. We find the hypothesized parabolic relationship among term male African American infants. As expected, we find a linear relationship between unexpected layoffs and the rate of SGA among term male non-Hispanic whites. These results are robust to sensitivity analyses. These results may help to explain the high rates of SGA among term male African American infants, despite greater maternal exposure to ambient stress during pregnancy.