Missing data and missed infections: Investigating racial and ethnic disparities in SARS-CoV-2 testing and infection rates in Holyoke, Massachusetts

Routinely collected testing data has been a vital resource for public health response during the COVID-19 pandemic and has revealed the extent to which Black and Hispanic persons have borne a disproportionate burden of SARS-CoV-2 infections and hospitalizations in the United States. However, missing race and ethnicity data and missed infections due to testing disparities limit the interpretation of testing data and obscure the true toll of the pandemic. We investigated potential bias arising from these two types of missing data through a case study in Holyoke, Massachusetts during the pre-vaccination phase of the pandemic. First, we estimated SARS-CoV-2 testing and case rates by race/ethnicity, imputing missing data using a joint modelling approach. We then investigated disparities in SARS-CoV-2 reported case rates and missed infections by comparing case rate estimates to estimates derived from a COVID-19 seroprevalence survey. Compared to the non-Hispanic white population, we found that the Hispanic population had similar testing rates (476 vs. 480 tested per 1,000) but twice the case rate (8.1% vs. 3.7%). We found evidence of inequitable testing, with a higher rate of missed infections in the Hispanic population compared to the non-Hispanic white population (77 vs. 58 infections missed per 1,000).

Disparities in SARS-CoV-2 Infection by Race, Ethnicity, Language, and Social Vulnerability: Evidence from a Citywide Seroprevalence Study in Massachusetts, USA

OBJECTIVES

Uncovering and addressing disparities in infectious disease outbreaks require a rapid, methodical understanding of local epidemiology. We conducted a seroprevalence study of SARS-CoV-2 infection in Holyoke, Massachusetts, a majority Hispanic city with high levels of socio-economic disadvantage to estimate seroprevalence and identify disparities in SARS-CoV-2 infection.

METHODS

We invited 2000 randomly sampled households between 11/5/2020 and 12/31/2020 to complete questionnaires and provide dried blood spots for SARS-CoV-2 antibody testing. We calculated seroprevalence based on the presence of IgG antibodies using a weighted Bayesian procedure that incorporated uncertainty in antibody test sensitivity and specificity and accounted for household clustering.

RESULTS

Two hundred eighty households including 472 individuals were enrolled. Three hundred twenty-eight individuals underwent antibody testing. Citywide seroprevalence of SARS-CoV-2 IgG was 13.1% (95% CI 6.9-22.3) compared to 9.8% of the population infected based on publicly reported cases. Seroprevalence was 16.1% (95% CI 6.2-31.8) among Hispanic individuals compared to 9.4% (95% CI 4.6-16.4) among non-Hispanic white individuals. Seroprevalence was higher among Spanish-speaking households (21.9%; 95% CI 8.3-43.9) compared to English-speaking households (10.2%; 95% CI 5.2-18.0) and among individuals in high social vulnerability index (SVI) areas based on the CDC SVI (14.4%; 95% CI 7.1-25.5) compared to low SVI areas (8.2%; 95% CI 3.1-16.9).

CONCLUSIONS

The SARS-CoV-2 IgG seroprevalence in a city with high levels of social vulnerability was 13.1% during the pre-vaccination period of the COVID-19 pandemic. Hispanic individuals and individuals in communities characterized by high SVI were at the highest risk of infection. Public health interventions should be designed to ensure that individuals in high social vulnerability communities have access to the tools to combat COVID-19.

Evaluation of the implementation of a community health worker-led COVID-19 contact tracing intervention in Chiapas, Mexico, from March 2020 to December 2021

BACKGROUND

Mexico is one of the countries with the greatest excess death due to COVID-19. Chiapas, the poorest state in the country, has been particularly affected. Faced with an exacerbated shortage of health professionals, medical supplies, and infrastructure to respond to the pandemic, the non-governmental organization Compañeros En Salud (CES) implemented a COVID-19 infection prevention and control program to limit the impact of the pandemic in the region. We evaluated CES's implementation of a community health worker (CHW)-led contact tracing intervention in eight rural communities in Chiapas.

METHODS

Our retrospective observational study used operational data collected during the contract tracing intervention from March 2020 to December 2021. We evaluated three outcomes: contact tracing coverage, defined as the proportion of named contacts that were located by CHWs, successful completion of contact tracing, and incidence of suspected COVID-19 among contacts. We described how these outcomes changed over time as the intervention evolved. In addition, we assessed associations between these three main outcomes and demographic characteristics of contacts and intervention period (pre vs. post March 2021) using univariate and multivariate logistic regression.

RESULTS

From a roster of 2,177 named contacts, 1,187 (54.5%) received at least one home visit by a CHW and 560 (25.7%) had successful completion of contact tracing according to intervention guidelines. Of 560 contacts with complete contact tracing, 93 (16.6%) became suspected COVID-19 cases. We observed significant associations between sex and coverage (p = 0.006), sex and complete contact tracing (p = 0.049), community of residence and both coverage and complete contact tracing (p < 0.001), and intervention period and both coverage and complete contact tracing (p < 0.001).

CONCLUSIONS

Our analysis highlights the promises and the challenges of implementing CHW-led COVID-19 contact tracing programs. To optimize implementation, we recommend using digital tools for data collection with a human-centered design, conducting regular data quality assessments, providing CHWs with sufficient technical knowledge of the data collection system, supervising CHWs to ensure contact tracing guidelines are followed, involving communities in the design and implementation of the intervention, and addressing community member needs and concerns surrounding stigmatization arising from lack of privacy.

Predicting facility-based delivery in Zanzibar: The vulnerability of machine learning algorithms to adversarial attacks

Background

Community health worker (CHW)-led maternal health programs have contributed to increased facility-based deliveries and decreased maternal mortality in sub-Saharan Africa. The recent adoption of mobile devices in these programs provides an opportunity for real-time implementation of machine learning predictive models to identify women most at risk for home-based delivery. However, it is possible that falsified data could be entered into the model to get a specific prediction result - known as an "adversarial attack". The goal of this paper is to evaluate the algorithm's vulnerability to adversarial attacks.

Methods

The dataset used in this research is from the Uzazi Salama ("Safer Deliveries") program, which operated between 2016 and 2019 in Zanzibar. We used LASSO regularized logistic regression to develop the prediction model. We used "One-At-a-Time (OAT)" adversarial attacks across four different types of input variables: binary - access to electricity at home, categorical - previous delivery location, ordinal - educational level, and continuous - gestational age. We evaluated the percent of predicted classifications that change due to these adversarial attacks.

Results

Manipulating input variables affected prediction results. The variable with the greatest vulnerability was previous delivery location, with 55.65% of predicted classifications changing when applying adversarial attacks from previously delivered at a facility to previously delivered at home, and 37.63% of predicted classifications changing when applying adversarial attacks from previously delivered at home to previously delivered at a facility.

Conclusion

This paper investigates the vulnerability of an algorithm to predict facility-based delivery when facing adversarial attacks. By understanding the effect of adversarial attacks, programs can implement data monitoring strategies to assess for and deter these manipulations. Ensuring fidelity in algorithm deployment secures that CHWs target those women who are actually at high risk of delivering at home.

Opioid Use After Laparoscopic Surgery for Endometriosis and Pelvic Pain

STUDY OBJECTIVE

The primary objective was to quantify postoperative opioid use after laparoscopic surgery for endometriosis or pelvic pain. The secondary objective was to identify patient characteristics associated with greater postoperative opioid requirements.

DESIGN

Prospective, survey-based study in which subjects completed 1 preoperative and 7 postoperative surveys within 28 days of surgery regarding medication usage and pain control.

SETTING

Tertiary care, academic center.

PATIENTS

A total of 100 women with endometriosis or pelvic pain.

INTERVENTIONS

Laparoscopic same-day discharge surgery by fellowship-trained minimally invasive gynecologists.

MEASUREMENTS AND MAIN RESULTS

A total of 100 patients were recruited and 8 excluded, for a final sample size of 92 patients. All patients completed the preoperative survey. Postoperative response rates ranged from 70.7% to 80%. The mean number of pills (5 mg oxycodone tablets) taken by day 28 was 6.8. The average number of pills prescribed was 10.2, with a minimum of 4 (n = 1) and maximum of 20 (n = 3). Previous laparoscopy for pelvic pain was associated with a significant increase in postoperative narcotic use (8.2 vs 5.6; p = .044). Hysterectomy was the only surgical procedure associated with a significant increase in postoperative narcotic use (9.7 vs 5.4; p = .013). There were no difference in number of pills taken by presence of deep endometriosis or pathology-confirmed endometriosis (all p >.36). There was a trend of greater opioid use in patients with diagnoses of self-reported chronic pelvic pain, anxiety, and depression (7.9 vs 5.7, p = .051; 7.7 vs 5.2, p = .155; 8.1 vs 5.6, p = .118).

CONCLUSION

Most patients undergoing laparoscopic surgery for endometriosis and pelvic pain had a lower postoperative opioid requirement than prescribed, suggesting surgeons can prescribe fewer postoperative narcotics in this population. Patients with a previous surgery for pelvic pain, self-reported chronic pelvic pain syndrome, anxiety, and depression may represent a subset of patients with increased postoperative opioid requirements.

Adapting Lot Quality Assurance Sampling to accommodate imperfect diagnostic tests: application to COVID-19 serosurveillance in Haiti

BACKGROUND

Lot Quality Assurance Sampling (LQAS), a tool used for monitoring health indicators in low resource settings resulting in "high" or "low" classifications, assumes that determination of the trait of interest is perfect. This is often not true for diagnostic tests, with imperfect sensitivity and specificity. Here, we develop Lot Quality Assurance Sampling for Imperfect Tests (LQAS-IMP) to address this issue and apply it to a COVID-19 serosurveillance study design in Haiti.

METHODS

We first derive a modified procedure, LQAS-IMP, that accounts for the sensitivity and specificity of a diagnostic test to yield correct classification errors. We then apply the novel LQAS-IMP to design an LQAS system to classify prevalence of SARS-CoV-2 antibodies among healthcare workers at eleven Zanmia Lasante health facilities in Haiti. Finally, we show the performance of the LQAS-IMP procedure in a simulation study.

RESULTS

We found that when an imperfect diagnostic test is used, the classification errors in the standard LQAS procedure are larger than specified. In the modified LQAS-IMP procedure, classification errors are consistent with the specified maximum classification error. We then utilized the LQAS-IMP procedure to define valid systems for sampling at eleven hospitals in Haiti.

CONCLUSION

The LQAS-IMP procedure accounts for imperfect sensitivity and specificity in system design; if the accuracy of a test is known, the use of LQAS-IMP extends LQAS to applications for indicators that are based on laboratory tests, such as SARS-CoV-2 antibodies.

The association of U.S. state-level abortion restrictions with medication abortion service delivery innovations during the early COVID-19 pandemic

OBJECTIVES

To assess whether state-level abortion restrictions resulted in differential uptake of innovative medication abortion practices such as changing ultrasound requirements, offering telehealth, or dispensing medications without a physical exam during the early COVID-19 pandemic.

METHODS

We used data from a prospective national survey of abortion providers to assess the association between a novel index of state-level abortion hostility and adoption of medication abortion services innovations during the pandemic.

RESULTS

Clinics in states with low or medium hostility were more likely to adopt innovative practices than those in high or extreme hostility states.

CONCLUSIONS

Clinics in abortion hostile states were less likely to adopt clinical recommendations and public health best practices for abortion care during the COVID-19 pandemic.

Machine learning for maternal health: Predicting delivery location in a community health worker program in Zanzibar

Background

Maternal and neonatal health outcomes in low- and middle-income countries (LMICs) have improved over the last two decades. However, many pregnant women still deliver at home, which increases the health risks for both the mother and the child. Community health worker programs have been broadly employed in LMICs to connect women to antenatal care and delivery locations. More recently, employment of digital tools in maternal health programs have resulted in better care delivery and served as a routine mode of data collection. Despite the availability of rich, patient-level data within these digital tools, there has been limited utilization of this type of data to inform program delivery in LMICs.

Methods

We use program data from 38,787 women enrolled in Safer Deliveries, a community health worker program in Zanzibar, to build a generalizable prediction model that accurately predicts whether a newly enrolled pregnant woman will deliver in a health facility. We use information collected during the enrollment visit, including demographic data, health characteristics and current pregnancy information. We apply four machine learning methods: logistic regression, LASSO regularized logistic regression, random forest and an artificial neural network; and three sampling techniques to address the imbalanced data: undersampling of facility deliveries, oversampling of home deliveries and addition of synthetic home deliveries using SMOTE.

Results

Our models correctly predicted the delivery location for 68%–77% of the women in the test set, with slightly higher accuracy when predicting facility delivery versus home delivery. A random forest model with a balanced training set created using undersampling of existing facility deliveries accurately identified 74.4% of women delivering at home.

Conclusions

This model can provide a “real-time” prediction of the delivery location for new maternal health program enrollees and may enable early provision of extra support for individuals at risk of not delivering in a health facility, which has potential to improve health outcomes for both mothers and their newborns. The framework presented here is applicable in other contexts and the selection of input features can easily be adapted to match data availability and other outcomes, both within and beyond maternal health.

The impact of the COVID-19 pandemic on abortion care utilization and disparities by age

Background

A variety of state-level restrictions were placed on abortion care in response to the COVID-19 pandemic, leading to drops in utilization and delays in time to abortion. Other pandemic-related factors also may have impacted receipt of abortion care, potentially exacerbating existing barriers to care. Massachusetts is an ideal setting to study the impact of these other pandemic-related factors on abortion care utilization because there was no wide-scale abortion policy change in response to the pandemic.

Objective

This study aimed to evaluate the impact of the COVID-19 pandemic on abortion care utilization and disparities in utilization by patient age in Massachusetts.

Study Design

Using the electronic medical records from all abortions that occurred at the Planned Parenthood League of Massachusetts from May 1, 2017 through December 31, 2020 (N=35,411), we performed time series modeling to estimate monthly changes in the number of abortions from the expected counts during the COVID-19 pandemic. We also assessed if legal minors (<18 years) experienced delays in time to abortion, based on gestational age at procedure, and whether minors were differentially impacted by the pandemic.

Results

There were 1725 less abortions than expected, corresponding to a 20% drop, from March 2020 to December 2020 (95% prediction interval, −2025 to −1394) with 888 less (20% reduction) abortions among adults, 792 (20% reduction) less among young adults, and 45 (27% reduction) among minors. Adults and young adults experienced significant reductions in the number of abortions beginning in March 2020, whereas decreases among minors did not begin until July 2020. The rate of abortions occurring ≥12 weeks gestational age was unchanged during the COVID-19 pandemic among minors (adjusted rate ratio, 0.92; 95% confidence interval, 0.55–1.51) and among adults (adjusted rate ratio, 0.92; 95% confidence interval, 0.78–1.09). Young adults had a lower rate of second trimester abortion during the pandemic (adjusted rate ratio, 0.79; 95% confidence interval, 0.66–0.95).

Conclusion

Despite uninterrupted abortion service provision, abortion care utilization decreased markedly in Massachusetts during the pandemic. There was no evidence of an increase in second trimester abortions in any age group. Further research is needed to determine if a decline in the pregnancy rate or other factors, such as financial and travel barriers, fear of infection, or privacy concerns, may have contributed to this decline.

Childhood immunization during the COVID-19 pandemic: experiences in Haiti, Lesotho, Liberia and Malawi

Objective

To examine changes in vaccination of children younger than 1 year during the coronavirus disease 2019 (COVID-19) pandemic (March 2020–August 2021) in Haiti, Lesotho, Liberia and Malawi.

Methods

We used data from health management information systems on vaccination of children aged 12 months or younger in districts supported by Partners In Health. We used data from January 2016 to February 2020 and a linear model with negative binomial distribution to estimate the expected immunization counts for March 2020–August 2021 with 95% prediction intervals, assuming no pandemic. We compared these expected levels with observed values and estimated the immunization deficits or excesses during the pandemic months.

Findings

Baseline vaccination counts varied substantially by country, with Lesotho having the lowest count and Haiti the highest. We observed declines in vaccination administration early in the COVID-19 pandemic in Haiti, Lesotho and Liberia. Continued declines largely corresponded to high rates of COVID-19 infection and discrete stock-outs. By August 2021, vaccination levels had returned to close to or above expected levels in Haiti, Liberia and Lesotho; in Malawi levels remained below expected.

Conclusion

Patterns of childhood immunization coverage varied by country over the course of the pandemic, with significantly lower than expected vaccination levels seen in one country during subsequent COVID-19 waves. Governments and health-care stakeholders should monitor vaccine coverage closely and consider interventions, such as community outreach, to avoid or combat the disruptions in childhood vaccination.

Disruptions in maternal health service use during the COVID-19 pandemic in 2020: experiences from 37 health facilities in low-income and middle-income countries

The COVID-19 pandemic has heterogeneously affected use of basic health services worldwide, with disruptions in some countries beginning in the early stages of the emergency in March 2020. These disruptions have occurred on both the supply and demand sides of healthcare, and have often been related to resource shortages to provide care and lower patient turnout associated with mobility restrictions and fear of contracting COVID-19 at facilities. In this paper, we assess the impact of the COVID-19 pandemic on the use of maternal health services using a time series modelling approach developed to monitor health service use during the pandemic using routinely collected health information systems data. We focus on data from 37 non-governmental organisation-supported health facilities in Haiti, Lesotho, Liberia, Malawi, Mexico and Sierra Leone. Overall, our analyses indicate significant declines in first antenatal care visits in Haiti (18% drop) and Sierra Leone (32% drop) and facility-based deliveries in all countries except Malawi from March to December 2020. Different strategies were adopted to maintain continuity of maternal health services, including communication campaigns, continuity of community health worker services, human resource capacity building to ensure compliance with international and national guidelines for front-line health workers, adapting spaces for safe distancing and ensuring the availability of personal protective equipment. We employ a local lens, providing prepandemic context and reporting results and strategies by country, to highlight the importance of developing context-specific interventions to design effective mitigation strategies.

The Financial Burden of Antiquated Laws: The Case of Massachusetts' Parental Involvement Law for Abortion

Background: A majority of U.S. states enforce parental involvement laws that require minors seeking abortion to obtain parental consent, or else obtain judicial bypass through the court system. Although such laws are widespread, the financial cost of their enforcement has yet to be documented.

Methods: We used data from a retrospective observational cohort study among adolescents (aged ≤17 years old) who sought abortion services at Planned Parenthood League of Massachusetts (PPLM) between 2010 and 2016. We assessed the direct financial burden of judicial bypass among 449 minors accounting for direct public legal costs, private professional costs, cost of lost school, and cost to the young person.

Results: The total added cost of judicial bypass in our cohort amounted to $374,982.04 (median cost of $705.14 per abortion). The direct out-of-pocket cost amounted to $84,370.23 ($179.89 per abortion). The majority of this cost was due to increased average procedure costs solely due to delays in care incurred by judicial bypass (range $0 to $5,200.50). In total, 74% of minors in our cohort were insured through Medicaid at the time of their abortion. Additional out-of-pocket costs for bypass were 20.2% of their household's maximum monthly income.

Conclusions: These analyses show that judicial bypass as a function of parental involvement laws correlates with increased costs to individual minors and to the public, with the heaviest burden placed on minors of low socioeconomic status.

Discussion on "Causal mediation of semicompeting risks" by Yen-Tsung Huang

Huang proposes a method for assessing the impact of a point treatment on mortality either directly or mediated by occurrence of a nonterminal health event, based on data from a prospective cohort study in which the occurrence of the nonterminal health event may be preemptied by death but not vice versa. The author uses a causal mediation framework to formally define causal quantities known as natural (in)direct effects. The novelty consists of adapting these concepts to a continuous-time modeling framework based on counting processes. In an effort to posit "scientifically interpretable estimands," statistical and causal assumptions are introduced for identification. In this commentary, we argue that these assumptions are not only difficult to interpret and justify, but are also likely violated in the hepatitis B motivating example and other survival/time to event settings as well.

Improving health facility delivery rates in Zanzibar, Tanzania through a large-scale digital community health volunteer programme: a process evaluation

The utilization of community health worker (CHW) programmes to improve maternal and neonatal health outcomes has become widely applied in low- and middle-income countries. While current research has focused on discerning the effect of these interventions, documenting the process of implementing, scaling and sustaining these programmes has been largely ignored. Here, we focused on the implementation of the Safer Deliveries CHW programme in Zanzibar, a programme designed to address high rates of maternal and neonatal mortality by increasing rates of health facility delivery and postnatal care visits. The programme was implemented and brought to scale in 10 of 11 districts in Zanzibar over the course of 3 years by D-tree International and the Zanzibar Ministry of Health. As the programme utilized a mobile app to support CHWs during their visits, a rich data resource comprised of 133 481 pregnancy and postpartum home visits from 41 653 women and 436 CHWs was collected, enabling the evaluation of numerous measures related to intervention fidelity and health outcomes. Utilizing the framework of Steckler et al., we completed a formal process evaluation of the primary intervention, CHW home visits to women during their pregnancy and postpartum period. Our in-depth analysis and discussion will serve as a model for process evaluations of similar CHW programmes and will hopefully encourage future implementers to report analogous measures of programme performance.

Syndromic surveillance using monthly aggregate health systems information data: methods with application to COVID-19 in Liberia

BACKGROUND

Early detection of SARS-CoV-2 circulation is imperative to inform local public health response. However, it has been hindered by limited access to SARS-CoV-2 diagnostic tests and testing infrastructure. In regions with limited testing capacity, routinely collected health data might be leveraged to identify geographical locales experiencing higher than expected rates of COVID-19-associated symptoms for more specific testing activities.

METHODS

We developed syndromic surveillance tools to analyse aggregated health facility data on COVID-19-related indicators in seven low- and middle-income countries (LMICs), including Liberia. We used time series models to estimate the expected monthly counts and 95% prediction intervals based on 4 years of previous data. Here, we detail and provide resources for our data preparation procedures, modelling approach and data visualisation tools with application to Liberia.

RESULTS

To demonstrate the utility of these methods, we present syndromic surveillance results for acute respiratory infections (ARI) at health facilities in Liberia during the initial months of the COVID-19 pandemic (January through August 2020). For each month, we estimated the deviation between the expected and observed number of ARI cases for 325 health facilities and 15 counties to identify potential areas of SARS-CoV-2 circulation.

CONCLUSIONS

Syndromic surveillance can be used to monitor health facility catchment areas for spikes in specific symptoms which may indicate SARS-CoV-2 circulation. The developed methods coupled with the existing infrastructure for routine health data systems can be leveraged to monitor a variety of indicators and other infectious diseases with epidemic potential.

Gender diversity among aspiration-abortion patients

OBJECTIVE

Describe the array of gender identities among procedural abortion patients.

STUDY DESIGN

Cross-sectional survey of abortion patients in three clinics in Massachusetts. Following aspiration abortion procedures and prior to discharge, patients self-administered a survey on a tablet.

RESULTS

From November 2017 through July 2018, 1,553 aspiration abortion patients completed the survey (participation rate: 82%). Patients reported several gender identities. Non-binary (0.4%) and agender (0.4%) were the most common identities after female (91.1%) and woman (6.0%). Overall, 2.7% of patients identified as a gender other than female or woman.

CONCLUSION

Aspiration abortion patients have a variety of gender identities. To promote quality of care for all patients, abortion providers can ensure their names, marketing materials, patient forms, and clinical environments are gender inclusive rather than focusing on women's health.

Auto-G-Computation of Causal Effects on a Network

Methods for inferring average causal effects have traditionally relied on two key assumptions: (i) the intervention received by one unit cannot causally influence the outcome of another; and (ii) units can be organized into nonoverlapping groups such that outcomes of units in separate groups are independent. In this article, we develop new statistical methods for causal inference based on a single realization of a network of connected units for which neither assumption (i) nor (ii) holds. The proposed approach allows both for arbitrary forms of interference, whereby the outcome of a unit may depend on interventions received by other units with whom a network path through connected units exists; and long range dependence, whereby outcomes for any two units likewise connected by a path in the network may be dependent. Under network versions of consistency and no unobserved confounding, inference is made tractable by an assumption that the networks outcome, treatment and covariate vectors are a single realization of a certain chain graph model. This assumption allows inferences about various network causal effects via the auto-g-computation algorithm, a network generalization of Robins' well-known g-computation algorithm previously described for causal inference under assumptions (i) and (ii). Supplementary materials for this article are available online.

State and federal abortion restrictions increase risk of COVID-19 exposure by mandating unnecessary clinic visits

Objective

To quantify the number of medically unnecessary clinical visits and in-clinic contacts monthly caused by US abortion regulations.

Study Design

We estimated the number of clinical visits and clinical contacts (any worker a patient may come into physical contact with during their visit) under the current policy landscape, compared to the number of visits and contacts if the following regulations were repealed: (1) State mandatory in-person counseling visit laws that necessitate two visits for abortion, (2) State mandatory-ultrasound laws, (3) State mandates requiring the prescribing clinician be present during mifepristone administration, (4) Federal Food and Drug Administration Risk Evaluation and Mitigation Strategy for mifepristone. If these laws were repealed, “no-test” telemedicine abortion would be possible for some patients. We modeled the number of visits averted if a minimum of 15 percent or a maximum of 70 percent of medication abortion patients had a “no-test” telemedicine abortion.

Results

We estimate that 12,742 in-person clinic visits (50,978 clinical contacts) would be averted each month if counseling visit laws alone were repealed, and 31,132 visits (142,910 clinical contacts) would be averted if all four policies were repealed and 70 percent of medication abortion patients received no-test telemedicine abortions. Over 2 million clinical contacts could be averted over the projected 18-month COVID-19 pandemic.

Conclusion

Medically unnecessary abortion regulations result in a large number of excess clinical visits and contacts.

Policy Implications

Repeal of medically unnecessary state and federal abortion restrictions in the United States would allow for evidence-based telemedicine abortion care, thereby lowering risk of SARS-CoV-2 transmission.

Robust inference on population indirect causal effects: the generalized front door criterion

Standard methods for inference about direct and indirect effects require stringent no-unmeasured-confounding assumptions which often fail to hold in practice, particularly in observational studies. The goal of the paper is to introduce a new form of indirect effect, the population intervention indirect effect, that can be non-parametrically identified in the presence of an unmeasured common cause of exposure and outcome. This new type of indirect effect captures the extent to which the effect of exposure is mediated by an intermediate variable under an intervention that holds the component of exposure directly influencing the outcome at its observed value. The population intervention indirect effect is in fact the indirect component of the population intervention effect, introduced by Hubbard and Van der Laan. Interestingly, our identification criterion generalizes Judea Pearl's front door criterion as it does not require no direct effect of exposure not mediated by the intermediate variable. For inference, we develop both parametric and semiparametric methods, including a novel doubly robust semiparametric locally efficient estimator, that perform very well in simulation studies. Finally, the methods proposed are used to measure the effectiveness of monetary saving recommendations among women enrolled in a maternal health programme in Tanzania.

Estimation of Natural Indirect Effects Robust to Unmeasured Confounding and Mediator Measurement Error

The use of causal mediation analysis to evaluate the pathways by which an exposure affects an outcome is widespread in the social and biomedical sciences. Recent advances in this area have established formal conditions for identification and estimation of natural direct and indirect effects. However, these conditions typically involve stringent assumptions of no unmeasured confounding and that the mediator has been measured without error. These assumptions may fail to hold in many practical settings where mediation methods are applied. The goal of this article is two-fold. First, we formally establish that the natural indirect effect can in fact be identified in the presence of unmeasured exposure-outcome confounding provided there is no additive interaction between the mediator and unmeasured confounder(s). Second, we introduce a new estimator of the natural indirect effect that is robust to both classical measurement error of the mediator and unmeasured confounding of both exposure-outcome and mediator-outcome relations under certain no interaction assumptions. We provide formal proofs and a simulation study to illustrate our results. In addition, we apply the proposed methodology to data from the Harvard President's Emergency Plan for AIDS Relief (PEPFAR) program in Nigeria.

Massachusetts' Parental Consent Law and Procedural Timing Among Adolescents Undergoing Abortion

OBJECTIVE

To describe individual-level delay in obtaining abortion associated with use of the Massachusetts judicial bypass system, which legal minors (aged 17 years or younger) use to obtain abortion without consent of a parent or legal guardian in the setting of Massachusetts' parental consent law for abortion.

METHODS

We conducted a retrospective cohort study of 2,026 abortions among minors at a large, statewide network of abortion clinics between 2010 and 2016. Delay was defined as the number of calendar days between the minor's first call to the clinic to schedule an abortion, and the day the abortion was received.

RESULTS

In the study population, 1,559 (77%) abortions were obtained with parental consent and 467 (23%) using judicial bypass. Abortions after judicial bypass were more common among minors identifying as Hispanic, non-Hispanic black, or other race, those of low socioeconomic status (as indicated by having Medicaid insurance) and those with a prior birth or prior abortion (all P<.05). Minors with parental consent received their abortion a mean of 8.6 days after initial contact, compared with 14.8 days for minors with judicial bypass, for an unadjusted difference of 6.1 days. In multivariable linear regression modeling adjusting for demographic differences between groups, this difference persisted: minors who obtained abortions after judicial bypass had a significantly greater delay compared with those with parental consent (adjusted mean difference = 5.2 days; 95% CI 4.3 to 6.2). Using multivariable logistic regression modeling, minors with judicial bypass also had higher odds of becoming ineligible for medication abortion between the day of first call and the day of procedure (adjusted odds ratio 1.57; 95% CI 1.09 to 2.26).

CONCLUSION

Massachusetts' parental consent law for abortion is associated with delay among minors and thereby may constrain the clinical options available to them.

Heritability of fetal hemoglobin, white cell count, and other clinical traits from a sickle cell disease family cohort

Sickle cell disease (SCD) is the most common monogenic disorder in the world. Notably, there is extensive clinical heterogeneity in SCD that cannot be fully accounted for by known factors, and in particular, the extent to which the phenotypic diversity of SCD can be explained by genetic variation has not been reliably quantified. Here, in a family-based cohort of 449 patients with SCD and 755 relatives, we first show that 5 known modifiers affect 11 adverse outcomes in SCD to varying degrees. We then utilize a restricted maximum likelihood procedure to estimate the heritability of 20 hematologic traits, including fetal hemoglobin (HbF) and white blood cell count (WBC), in the clinically relevant context of inheritance from healthy carriers to SCD patients. We report novel estimations of heritability for HbF at 31.6% (±5.4%) and WBC at 41.2% (±6.8%) in our cohort. Finally, we demonstrate shared genetic bases between HbF, WBC, and other hematologic traits, but surprisingly little overlap between HbF and WBC themselves. In total, our analyses show that HbF and WBC have significant heritable components among individuals with SCD and their relatives, demonstrating the value of using family-based studies to better understand modifiers of SCD.

Mediation Analysis for Censored Survival Data Under an Accelerated Failure Time Model

Recent advances in causal mediation analysis have formalized conditions for estimating direct and indirect effects in various contexts. These approaches have been extended to a number of models for survival outcomes including accelerated failure time models, which are widely used in a broad range of health applications given their intuitive interpretation. In this setting, it has been suggested that under standard assumptions, the "difference" and "product" methods produce equivalent estimates of the indirect effect of exposure on the survival outcome. We formally show that these two methods may produce substantially different estimates in the presence of censoring or truncation, due to a form of model misspecification. Specifically, we establish that while the product method remains valid under standard assumptions in the presence of independent censoring, the difference method can be biased in the presence of such censoring whenever the error distribution of the accelerated failure time model fails to be collapsible upon marginalizing over the mediator. This will invariably be the case for most choices of mediator and outcome error distributions. A notable exception arises in case of normal mediator-normal outcome where we show consistency of both difference and product estimators in the presence of independent censoring. These results are confirmed in simulation studies and two data applications.