key: cord-1050798-1rlabwgp
authors: Fruh, Victoria; Lyons, Genevieve; Scalise, Ariel L.; Gallagher, Nicola J.; Jukic, Anne-Marie; Baird, Donna D.; Chaturvedi, Uvika; Suharwardy, Sanaa; Onnela, Jukka-Pekka; Williams, Michelle A.; Hauser, Russ; Coull, Brent A.; Mahalingaiah, Shruthi
title: Attempts to Conceive and the COVID-19 Pandemic: Data from the Apple Women’s Health Study
date: 2022-05-11
journal: Am J Obstet Gynecol
DOI: 10.1016/j.ajog.2022.05.013
sha: 580b3e790b29f9e1c9371789acab0444f0cc33e9
doc_id: 1050798
cord_uid: 1rlabwgp

Background Previous studies have suggested emergent events may affect pregnancy planning decisions. However, few have investigated the impact of factors related to the COVID-19 pandemic on pregnancy planning, measured by attempting conception, and how attempting conception status may differ by individual level factors such as social status or education level. Objective To examine the effects of factors related to the COVID-19 pandemic, through March 2021, on attempting conception status. To assess effect measure modification by education level and subjective social status (SSS). Study Design We conducted a longitudinal analysis within a subgroup of 21,616 participants in the Apple Women's Health Study (AWHS) who enrolled from November 2019-March 2021, met inclusion criteria, and who responded to the monthly status menstrual update question on attempting conception status (yes/no). Those reporting hysterectomy, pregnancy, lactation, or menopause were excluded. We used generalized estimating equation (GEE) methodology to fit logistic regression models that estimate odds ratios (ORs) and 95% confidence intervals (95% CI) for the association between the proportion of those who were attempting conception each month (compared to a pre-pandemic reference month of February 2020), while accounting for longitudinal correlation and adjusting for age, race/ethnicity, and marital status. We stratified analysis by social status and education level. Results We observed a trend of reduced odds of attempting conception, with an 18% reduction in the odds of attempting conception in August and October 2020 compared to the pre-pandemic month of February 2020 (August OR: 0.82, 95% CI: 0.70, 0.97; October OR: 0.82, 95% CI: 0.69, 0.97).Participants with lower education level (no college education) experienced a sustained reduction in odds of attempting to conceive from June 2020 to March 2021 when compared to February 2020, with up to a 24% reduction in the odds of attempting to conceive in October 2020 (OR: 0.76, 95% CI: 0.59, 0.96). Among participants that were college educated, we observed an initial reduction in odds of attempting to conceive starting in July 2020 (OR: 0.73, 95% CI: 0.54, 0.99) that returned near pre-pandemic odds. We also observed a reduction in odds of attempting to conceive among those with low SSS, with a decline in odds of attempting to conceive beginning in July 2020 (OR: 0.83, 95% CI: 0.63, 1.10) and continuing through March 2021 (OR: 0.79, 95% CI: 0.59, 1.06), with the greatest reduction in odds in October 2020 (OR: 0.67, 95% CI: 0.50, 0.91). Conclusion Among women in the AWHS cohort, our findings suggest a reduction in the odds of attempting to conceive over the course of the COVID-19 pandemic, through March 2021, particularly among women of lower education level and lower perceived social status.

Previous studies have suggested emergent events may affect pregnancy planning decisions.

However, few have investigated the impact of factors related to the COVID-19 pandemic on pregnancy planning, measured by attempting conception, and how attempting conception status may differ by individual level factors such as social status or education level.

To examine the effects of factors related to the COVID-19 pandemic, through March 2021, on attempting conception status. To assess effect measure modification by education level and subjective social status (SSS).

We conducted a longitudinal analysis within a subgroup of 21,616 participants in the Apple Women's Health Study (AWHS) who enrolled from November 2019-March 2021, met inclusion criteria, and who responded to the monthly status menstrual update question on attempting conception status (yes/no). Those reporting hysterectomy, pregnancy, lactation, or menopause were excluded. We used generalized estimating equation (GEE) methodology to fit logistic regression models that estimate odds ratios (ORs) and 95% confidence intervals (95% CI) for the association between the proportion of those who were attempting conception each month (compared to a prepandemic reference month of February 2020), while accounting for longitudinal correlation and adjusting for age, race/ethnicity, and marital status. We stratified analysis by social status and education level.

We observed a trend of reduced odds of attempting conception, with an 18% reduction in the odds of attempting conception in August and October 2020 compared to the pre-pandemic month of February 2020 (August OR: 0.82, 95% CI: 0.70, 0.97; October OR: 0.82, 95% CI: 0.69, 0.97).

Participants with lower education level (no college education) experienced a sustained reduction in odds of attempting to conceive from June 2020 to March 2021 when compared to February 2020, with up to a 24% reduction in the odds of attempting to conceive in October 2020 (OR: 0.76, 95% CI: 0.59, 0.96). Among participants that were college educated, we observed an initial reduction in odds of attempting to conceive starting in July 2020 (OR: 0.73, 95% CI: 0.54, 0.99) that returned near pre-pandemic odds. We also observed a reduction in odds of attempting to conceive among those with low SSS, with a decline in odds of attempting to conceive beginning in July 2020 (OR: 0.83, 95% CI: 0.63, 1.10) and continuing through March 2021 (OR: 0.79, 95% CI: 0.59, 1.06), with the greatest reduction in odds in October 2020 (OR: 0.67, 95% CI: 0.50, 0.91).

Many factors influence pregnancy planning and the desire to conceive, including economy 32 stability, education, and access to public services. 1 circumstances. 12,13 The SSS instructs participants to rate themselves on a "social ladder", with 84 corresponding numbered social status scale rungs of 0 (lowest) to 9 (highest). 12 We categorized SSS 85 scores as low (score of 0-3), moderate (score of 4-5) and high (score of 6-9). prefer not to answer. We defined attempting to conceive as a dichotomous variable. We categorized 93 participants selecting either response 1) or 2) above as attempting to conceive. Otherwise, we 94 designated participants as not attempting to conceive (Table S1 ).

Out of 97,052 total MSMU responses from active participants over the follow-up period, 96 14,543 (15.0%) were missing responses for one or more months. We singly imputed those missing 97 J o u r n a l P r e -p r o o f 9 responses with concordant survey responses before and after the response gap (n=14,020, 14.4%).

Supplemental Table 2 summarizes response concordance among these participants (Table S2) .

Statistical Analysis 101 We used generalized estimating equations (GEE) to fit logistic regression models for the 102 association between the proportion of participants attempting conception and month of response, 103 while accounting for longitudinal correlation 14-17 . We computed month-specific odds ratios and 104 95% confidence intervals (CIs) for the dichotomous outcome of attempting to conceive (yes vs. no), 105 in reference to the pre-pandemic month of February 2020. All models were adjusted for age, 106 race/ethnicity, and marital status. GEEs allow for missing non-concordant responses before and 107 after the response gap (n=523), with the assumption of missing completely at random. We specified 108 an autoregressive covariance structure. Within main models, we assessed effect modification by 109 education status, as education may indicate personal agency over physical risks of exposure and 110 lifestyle associated with ability to work remotely during the pandemic. We stratified by SSS as a 111 secondary analysis. We used de-identified data for all analyses and described aggregate results. 112 We performed a sensitivity analysis to adjust for seasonal trends modeled by sine and 113 cosine functions, as previously described as an applicable test for seasonality. 18, 19 We conducted a 114 second sensitivity analysis that was independent of cluster size by assigning AWHS participants to 115 closed sub-cohorts by quarter. For these models, we restricted analysis of participants to the first 116 quarter in which they responded to MSMU and assessed whether the participant attempted 117 conception at least once within that quarter. We performed logistic regression and test for trend 118 analysis. We then stratified by education and SSS separately. We performed data engineering in The mean (SD) age at enrollment of participants was 32.1 (8.6) years. A majority of 123 participants were White (71.6%) ( Table 1) . Forty-five percent of participants had no college 124 education, 33.0% of participants were college educated, and 22.1% of participants had a graduate 125 school degree. Participants attempting conception were more likely to be married (66% vs. 36%) 126 and were less likely to be in school (4% vs. 12%) compared to those not attempting conception. (Table 2 ). There were 7.6% of participants attempting conception at least once over the study (Table   132 S3). Participants attempted conception for an average of 4.4 months (Table S4) . Among those not 133 previously attempting conception at their study entry, 1.0-5.9% of responding participants newly 134 attempted conception during the study (Table S5) . Education level was generally consistent across 135 months (Table S6) .

In February 2020, 6.3% of participants were attempting conception ( Figure 1 ). We observed with high SSS (SSS score of 6-9) were generally similar to those with low SSS (Table S8) . 167 Sensitivity analysis adjusting for seasonality yielded similar results to those unadjusted for sine and 168 cosine functions (Table S9) . 169 Within the additional sensitivity analysis, effect estimates for quarter cohort models were often 170 less precise than GEE models and were closer to the null following quarter 2 2020 (Table S10, 171 S11). For education stratified models, odds of attempting to conceive within the quarter continued 172 to be elevated among participants with graduate degrees through March 2021 (Table S10) , similar 173 to main models (Table S7) . Results were more attenuated among non-college educated participants.

Among college educated participants, we observed a reduction in the odds of attempts to conceive 175 in quarter 4 2020 (OR: 0.75, 95% CI: 0.54, 1.04), compared to quarter 1 2020 (Table S10 , p trend = 176 0.06), though findings were imprecise. Trends for SSS stratified models were generally consistent 177 with main models (Table S11) . Trends among participants with high SSS were generally 178 comparable to those with low SSS following quarter 2 2020. Our primary research question and the focus of this study was based on measuring attempts 201 to conceive, in effort to understand the intent to conceive over the course of the pandemic.

Although fertility and birth rates do not exclusively indicate intent to conceive, we can compare 203 these rates with our findings to assess general trends. Our results have comparable trends with 204 publicly available fertility rate data and birth rate data. For example, the U.S. National Center for Our results may potentially be attributed to increased financial security and flexibility to 213 work from home among advanced degree workers. 21 Working from home may reduce the potential 214 for exposure to COVID-19, 21 and enable participants to focus on pregnancy planning rather than 215 daily concern about virus exposure at work. [22] [23] [24] In this way, education may serve as a proxy for 216 financial security and personal agency over physical risks of exposure associated with ability to 217 work remotely, and thereby act as a potential mitigatory factor for the impact of the pandemic. In 218 contrast to our study, other research found that the desire for children is generally higher among 219 those of lower education levels 23 and that women of higher SES had lower net fertility, 22-24 220 potentially due to those of higher SES responding faster to changing family norms. 24 However, 221 these analyses were conducted with data collected from a non-pandemic period. Table 12 (Table S12) . Our study has many strengths. Our analysis allows for a more microscopic, monthly view of 236 the changes in attempts to conceive. Prior studies have had challenges in gathering population level 237 data on monthly pregnancy attempts in that they occur approximately ten months or more before 238 birth. Since our digital study was launched before the pandemic, we were able to follow 239 pregnancy attempts over the pandemic. In contrast to existing studies that are cross-sectional and 240 rely on respective reports, as described previously, our study was prospective. Additionally, we 241 collected sufficient demographic information to understand the trends in conception attempts 242 stratified by education level. Our study is one of the first of its kind to combine frequently collected 243 survey data and menstrual health indicators to study women's health.

Our study has several limitations. Generalizability of this study may be limited since the 245 AWHS cohort is restricted to iPhone users, and demographic characteristics from this cohort of Census. Due to the racial homogeneity of participants in the study population, we were not able to 250 evaluate racial disparities in COVID-19 burden and attempts to conceive. 30,31 Additionally, women 251 who are attempting conception may be more likely to enroll in the AWHS study. Yet, pregnancy 252 planners were not likely to be oversampled as they were not targeted in recruitment efforts. The Table 2 . 255 Furthermore, we did not confirm self-reported conception attempts with sexual activity 256 logging. However, the purpose of our investigation did not focus on outcomes such as time to 257 J o u r n a l P r e -p r o o f 16 pregnancy. Another potential concern is attrition bias due to the potential impact of COVID-19 and 258 pandemic related factors on fertility. If these factors did reduce fertility, we would likely not see a 259 bias due to attrition. Rather we might find increases in the level of the month specific pool of 260 participants reporting yes to attempting conception, that may potentially inflate the proportion of 261 those attempting in any given month during the pandemic.

Due to the timing of the study, we were unable to account for seasonality based on historic 263 data prior to COVID-19, but we incorporated sine and cosine functions to account for seasonality 264 and did not find varying results. Moreover, while we did not collect data on type of employment, 265 income, or work-from-home status, we were able to evaluate education level. Although education 266 may be a poor proxy for SES, education is an objective measure and may in its own way predict 267 some behavioral factors related to lifestyle such as work-from-home status. Since the SSS was a 268 subjective measure of social status and did not objectively measure education, SES or income 269 status, we included this assessment as part of a secondary analysis.

Another potential limitation may be related to informative clustering. The longer a 271 participant has attempted to conceive at any point within the study, and stays within the study, the 272 more likely the participant may be to attempt to conceive in any month thereafter. In effort to assess 273 findings independent of cluster size, we conducted a sensitivity analysis assigning participants to 274 quarter cohorts and found similar results, though less precise (Table S10, S11). Finally, we 275 experienced loss to follow-up in this cohort, though we did not observe changes in demographics 276 among participants who were vs. were not lost to follow-up. 

Resources, aspirations and first births during the Great Recession

Birth Rate Decline Linked to Recession

Perceived consequences of the COVID-19 pandemic 300 and childbearing intentions in Poland

Desire for parenthood at the time of COVID-19 pandemic: an insight into the 302 Italian situation

The impact of the COVID-19 pandemic on 304 economic security and pregnancy intentions among people at risk of pregnancy

Multiple imputation in public health research

The Impact of the COVID-309 19 Pandemic on Pregnancy Planning Behaviors

Fertility intentions among couples in Shanghai under COVID-19: A cross-312 sectional study

Pandemic: Findings from the 2020 Guttmacher Survey of Reproductive Health Experiences

The impact of the COVID-19 pandemic on 317 economic security and pregnancy intentions among people at risk of pregnancy

Design and Methods of the Apple Women's Health Study: A Digital 320 Longitudinal Cohort Study. Under Review at

Relationship of subjective and 322 objective social status with psychological and physiological functioning: preliminary data in 323 healthy white women

Subjective social status, objective socioeconomic status, and 325 cardiovascular risk in women

Worsening of emergency department length of stay during the COVID-19

GENDER SPECIFIC DIFFERENCES IN 329 NEURODEVELOPMENTAL EFFECTS OF PRENATAL EXPOSURE TO VERY LOW-330 LEAD LEVELS: THE PROSPECTIVE COHORT STUDY IN THREE-YEAR OLDS

Longitudinal data analysis for discrete and continuous outcomes

Statistical analysis of 335 correlated data using generalized estimating equations: an orientation

Relation of Prenatal Air Pollutant and Nutritional Exposures with 338 Biomarkers of Allergic Disease in Adolescence

Studying Seasonality by Using Sine

Cosine Functions in Regression Analysis

Pandemic: Findings from the 2020 Guttmacher Survey of Reproductive Health Experiences

Workers with advanced degrees were more likely to work at 346 home than those without

Socioeconomic status and fertility decline: Insights from historical transitions in 350

Socio-economic and demographic factors associated with fertility 352 preferences among women of reproductive age in Ghana: evidence from the 2014 Demographic 353 and Health Survey

Socioeconomic Status and Net Fertility during the 355 Fertility Decline: A Comparative Analysis of Canada

Subjective social status: its determinants and 358 its association with measures of ill-health in the Whitehall II study

Subjective 361 Socioeconomic Status and Presence of the Metabolic Syndrome in Midlife Community 362 Volunteers

Education and Lifetime Earnings in the United 364 States

Educational Attainment in the United States: 2020. Census

CDC. Health Equity Considerations and Racial and Ethnic Minority Groups | CDC

Employed for pay (part-time, fulltime, self-employed)

By signing this Cover Sheet, the Author, on behalf of NIH, agrees to the provisions set out below, which modify and supersede, solely with respect to NIH, any conflicting provisions that are in the Publisher's standard copyright agreement (the "Publisher's Agreement"). If a Publisher's Agreement is attached, execution of this Cover Sheet constitutes an execution of the Publisher's Agreement, subject to the provisions and conditions of this Cover Sheet.

Indemnification. No Indemnification or "hold harmless" obligation is provided by either party.

Governing Law. This agreement will be governed by the law of the court in which a claim is brought.

Copyright. Author's contribution to the Work was done as part of the Author's official duties as a NIH employee and is a Work of the United States Government. Therefore, copyright may not be established in the United States. 17 U.S.C. § 105. If Publisher intends to disseminate the Work outside of the U.S., Publisher may secure copyright to the extent authorized under the domestic laws of the relevant country, subject to a paid-up, nonexclusive, irrevocable worldwide license to the United States in such copyrighted work to reproduce, prepare derivative works, distribute copies to the public and perform publicly and display publicly the work, and to permit others to do so.

No royalty income or other compensation may be accepted for work done as part of official duties. The author may accept for the agency a limited number of reprints or copies of the publication.

NIH Representations. NIH represents to the Publisher that the Author is the sole author of the Author's contribution to the Work and that NIH is the owner of the rights that are the subject of this agreement; that the Work is an original work and has not previously been published in any form anywhere in the world; that to the best of NIH's knowledge the Work is not a violation of any existing copyright, moral right, database right, or of any right of privacy or other intellectual property, personal, proprietary or statutory right; that where the Author is responsible for obtaining permissions or assisting the Publishers in obtaining permissions for the use of third party material, all relevant permissions and information have been secured; and that the Work contains nothing misleading, obscene, libelous or defamatory or otherwise unlawful. NIH agrees to reasonable instructions or requirements regarding submission procedures or author communications, and reasonable ethics or conflict of interest disclosure requirements unless they conflict with the provisions of this Cover Sheet.

Disclaimer. NIH and the Author expressly disclaim any obligation in Publisher's Agreement that is not consistent with the Author's official duties or the NIH mission, described at http://www.nih.gov/about/. NIH and the Author do not disclaim obligations to comply with a Publisher's conflict of interest policy so long as, and to the extent that, such policy is consistent with NIH's own conflict of interest policies.

For Peer-Reviewed Papers to be Submitted to PubMed Central. The Author is a US government employee who must comply with the NIH Public Access Policy, and the Author or NIH will deposit, or have deposited, in NIH's PubMed Central archive, an electronic version of the final, peer-reviewed manuscript upon acceptance for publication, to be made publicly available no later than 12 months after the official date of publication. The Author and NIH agree (notwithstanding Paragraph 3 above) to follow the manuscript deposition procedures (including the relevant embargo period, if any) of the publisher so long as they are consistent with the NIH Public Access Policy.

Modifications. PubMed Central may tag or modify the work consistent with its customary practices and with the meaning and integrity of the underlying work.