Telehealth Utilization in High-Risk Pregnancies During COVID-19

Purpose: To determine how telehealth has influenced outcomes in high-risk obstetrics patients during the Coronavirus disease 2019 (COVID-19) pandemic. Methods: A retrospective chart review was conducted to identify patterns in both telehealth and in-person clinic visits among patients of a Maternal Fetal Medicine (MFM) department from the onset of the COVID-19 pandemic from March 2020 until October 2021. For the descriptive analysis, p-values were calculated using Wilcoxon rank sum for continuous variables and chi-square or Fisher exact (where cell n < 5) for categorical variables. Variables of interest were then tested for their univariate association with telehealth utilization using logistic regression. Variables found to meet the criterion of p < 0.2 in the univariate case were introduced into a multivariable logistic model with a backward elimination for determining variable retention. We aimed to analyze whether telehealth visits significantly impacted pregnancy outcomes. Results: Four hundred nineteen high-risk patients visited the clinic via in-person and/or telehealth appointments during the study period: 320 patients without telehealth visits and 99 patients with telehealth visits. Care provided by telehealth visits was not found to be related to self-reported race (p = 0.81), maternal body mass index (p = 1.0), or maternal age (p = 0.53). Patients with private insurance were more likely to have telehealth visits than patients with public insurance (79.9% vs. 65.5%, p < 0.01). In univariate logistic analyses, patients with diagnoses of anxiety (p < 0.01), asthma (p = 0.03), and depression (p < 0.01), at the time care was established, were more likely to have telehealth visits. Those patients with telehealth visits did not have any statistical differences in mode of delivery (p = 0.2) or pregnancy outcomes (p = 0.12), including fetal demise, preterm delivery, or delivery at term as compared with patients with all in-office visits. In multivariable analysis, patient conditions of anxiety (p < 0.01), maternal obesity (p < 0.01), and twin pregnancy (p = 0.04) were associated with higher rates of telehealth visits. Conclusion: Patients with certain pregnancy complications elected to have more telehealth visits. Patients with private insurance were more likely to have telehealth visits than patients with public insurance. There are benefits for patients with certain pregnancy complications to incorporate telehealth visits in addition to regularly scheduled in-person clinic visits and may be suitable in a post-pandemic setting as well. Further research in this field is needed to better understand the impact of implementing telehealth in high-risk obstetrics patients.


Introduction
Before the Coronavirus disease 2019 (COVID-19) pandemic, telehealth was increasingly promoted as an opportunity to expand health care access to rural and underserved populations. [1][2][3] In addition, telehealth was viewed as an intervention to substantially reduce health care costs. 4 However, due to the COVID-19 pandemic and the need to provide consistent reliable care to patients with the concurrent need to provide care within safe distancing guidelines, telehealth was urgently adopted as a safer means of providing regular and consistent medical care. 5 One of the Centers for Disease Control and Prevention (CDC) recommendations was the implementation of telehealth to mitigate risks of COVID-19 viral exposure to patients while ensuring continued care in both inpatient and outpatient settings. 6 The CDC emphasized and encouraged patients with a high risk of severe illness to seek telehealth options when available. The pressure to provide consistent care while not endangering those at a higher risk of viral sequela were acutely felt while caring for pregnant individuals, as pregnancy itself is considered a baseline immunocompromised state.
The COVID-19 pandemic further increased immune vulnerability in this population. A recent metaanalysis found that COVID-19 infection in pregnancy was associated with a higher rate of preterm birth, preeclampsia, cesarean delivery, and perinatal death with preterm birth being the most common outcome. 5 Telehealth has seen expanded use in prenatal care and has been shown to be both safe and effective. [7][8][9] Several studies have revealed that telehealth enhances patient satisfaction, 10-12 while maintaining comparable maternal and fetal outcomes to in-person visits. 13,14 A study by Peahl et al. suggested that ''low-risk'' patients prefer telehealth visits over in-person visits when available. 15 Further, a recent systemic review found that telehealth interventions were associated with improvements in obstetric outcomes, decreased perinatal smoking, increased breastfeeding, and improved schedule optimization for high-risk obstetrics patients and providers. 16 Telehealth may also further improve health care access for disadvantaged populations who are at a higher risk of complications due to socioeconomic status and access to health care. However, studies have been inconclusive and have instead suggested that obstetric patients with a lower socioeconomic status and/or public health insurance were less likely to access telehealth for prenatal care. 17 Further research is, therefore, needed to help clinicians determine how to integrate telehealth into practice in ways that improve patient care in disadvantaged populations.
Following the CDC guidelines, telehealth outpatient visits were implemented in our tertiary academic medical center Maternal Fetal Medicine (MFM or ''highrisk'') outpatient clinic. Our study specifically evaluates pregnant individuals referred to be seen by an MFM specialist focusing on managing diagnoses, contributing to the classification of a high-risk pregnancy. The purpose of this study was to review the impact of telehealth for high-risk obstetric patients receiving care at a single tertiary academic center during the COVID-19 pandemic. We hypothesized that care and subsequent outcomes attributed to patients who sought care via telehealth would be noninferior to patients who continued with standard all in-person office visits.

Methods
A retrospective chart review was conducted to identify all patients who had either telehealth, telehealth and inperson visits, or in-person only visits in an MFM department from the onset of the COVID-19 pandemic through the end of the study period (March 2020 until October 2021). At the onset of the pandemic, clinic visits were limited, and patients were automatically assigned a telehealth visit by providers. As time progressed, patients were given the choice to have either in-office or telehealth visits at the conclusion of their visit. Doximity (Doximity, Inc., San Francisco, CA), a Health Insurance Portability and Accountability Act-compliant audio and visual messaging software, was used to conduct telehealth visits either by combined audio-visual or by audio-only calls according to individual provider preference.
Providers included MFM physicians and nurse practitioners as well as obstetrics and gynecology resident physicians training with MFM physicians. This study received Institutional Review Board approval from the University of Tennessee Health Sciences Center (IRB# 21-08173-XP) before initiation of this study. All individuals aged 13 to 42 years old with a highrisk pregnancy seen at the MFM office during the study period were included.
Data obtained included maternal age at time of first clinic appointment, all medical diagnoses pertaining to pregnancy, self-reported race, type of insurance, number of telehealth visits, body mass index (BMI) (defined as kg/m 2 ), mode of delivery, pregnancy outcomes (live birth rate and intrauterine fetal demise), and total incidence of fetal demise. We aimed at analyzing whether telehealth significantly impacted pregnancy outcomes according to patient pregnancy complications and known medical diagnoses.
Pregnancy outcomes and covariates of interest were described by telehealth utilization status using number and row percentage for categorical variables and mean, median, interquartile range, and standard deviation for continuous variables (Table 1). In preliminary comparisons of patients with telehealth and in-clinic only patients, Chi-square or Fisher exact test (where cell n < 5) were applied to categorical variables. Wilcoxon rank sum tests were applied to continuous variables.
Variables of interest were then tested for their univariate association with telehealth utilization using logistic regressions. Those variables found to meet the criterion of p < 0.2 in the univariate case were introduced into a multivariable logistic model with a limitation for variable retention.
Also, introduced to this model were relevant clinical variables such as anemia, chronic hypertension, and diabetes. Variables were retained in the model with p-value <0.1. Odds Ratios with 95% confidence intervals were also calculated. p-Values <0.05 were used for statistical significance. All statistical analyses were performed using SAS v9.4.

Results
The initial search yielded a total of 419 high-risk patients over the study timeframe. There were 320 patients (76.4%) without telehealth visits and 99 patients (23.6%) with telehealth visits. The mean num-ber of telehealth visits per patient in the telehealth group was 1.28, SD = 0.74. Our study population selfidentified as 72.5% Black, 16.9%, White, 8.1% Hispanic, and 2.5% identified as Other. Self-reported race was not a factor in choosing telehealth visits ( p = 0.81).
Other demographic characteristics are further delineated in Table 1. Patients with private insurance coverage were more likely to have telehealth visits included in their care. In addition, maternal BMI and age were not significantly related to having telehealth visits ( p = 1.00, 0.53). Mode of delivery and pregnancy outcomes were not impacted by telehealth visits. Table 2 is a comparison of telehealth patients versus controls. Odds ratios >1 indicate that the condition is more common among the telehealth patients than the controls. Patients with asthma, depression, anxiety, or preeclampsia with severe features were more likely to have telehealth visits. In multivariable logistic modeling, anxiety, maternal obesity, and twin pregnancy were shown to be independent factors associated with having telehealth visits (Table 3).

Discussion
Although the pandemic has increased telehealth utilization, there are still challenges to reach all patient populations. In this study, patients were from a combination rural and urban populations who had care from a tertiary care center. Our study suggested that patients with certain comorbidities in a highvolume academic MFM practice were more likely to have telehealth visits. Visits were initially scheduled by Resident Physicians, Nurse Practitioners, and Providers.
However, patients were subsequently given the option of telehealth versus in-person visits. We analyzed a number of variables that granted us the opportunity to generate a comprehensive review of the benefit of incorporating telehealth visits to aid in the care of highrisk obstetric patients. Factors such as anxiety, asthma, or twin pregnancy may have increased patients' interest in making use of telehealth appointments.
Patients with anxiety may have preferred to avoid being examined in clinic during the pandemic due to viral exposure concerns; however, it is not immediately obvious why patients with asthma, obesity, or a twin pregnancy were more inclined to pursue telehealth visits. Telehealth may meet these patients' needs more readily than in-person visits and they may be prematurely accustomed to telehealth visits by previous mental health sessions, or patients with more complicated conditions such as diabetes and/or hypertension require more in-person visits due to the complexity of the visits including blood pressure or glycemic control logs.
Patients with private insurance such as Blue Cross Blue Shield, Cigna, and United Health Care were more likely to have a telehealth visit than patients with public insurance (Medicaid/Medicare). It is possible that private insurance companies provided better coverage or incentives for telehealth visits. In addition, associated social determinants of health such as barriers to stable internet connection and lack of transportation may have factored into patient decisions to  utilize or forgo telehealth appointments. Additional research is needed to further evaluate both incentivization practices of private and public insurance companies and the roles of social determinants of health on telehealth utilization. The COVID-19 pandemic has impacted life and health care in numerous ways. It forced urgent and creative implementation of telehealth to enable stable continuous care for patients while decreasing further spread of the pandemic where possible. There are limited data on how telehealth has impacted patient care in the field of general obstetrics and gynecology and even less examining high-risk obstetrics patients.
There has been an increase in telehealth offered in obstetrics clinics in many places during the pandemic. 8,9 One notable study demonstrated that after an initial telehealth appointment, at least half of the patients considered pursuing future telehealth appointments as needed on a regular basis and physicians similarly were interested in offering telehealth options after an initial visit. 10 The American College of Obstetricians and Gynecologists (ACOG) has made several recommendations regarding telehealth implementation. 9 While some providers favor the use of t telehealth, 11 some institutions still face challenges in transitioning and implementing telehealth. Challenges that were found in this transition were barriers with public insurance and additional requirements of the patient (including technological proficiency, language barriers, Wi-Fi and data access, and childcare) in the implementation of telehealth visits. 11 Fryer et al. 18 proposed a new model for perinatal health that incorporated telehealth appointments and also discussed ways one health system adapted to telehealth in the time of the pandemic, including having patients purchase blood pressure cuffs, scales, and optionally, fetal dopplers to monitor vital signs from home. 12 Another study examined how women with maternal obesity utilizing telephone intervention had less gestational weight gain than women in the usual care group. 10 Telehealth interventions overall improved obstetric outcomes related to smoking cessation and breastfeeding and decreased the need for high-risk obstetric monitoring office visits while maintaining maternal and fetal outcomes. 11,12 In a literature review of 13 publications conducted by Wu et al., 19 virtual care resulted in strong patient and provider satisfaction related to cost savings and convenience, with clinic wait times and cancellation rates also improving. The results support safety equivalence in birth outcomes between in-person prenatal care models and those that integrate virtual visits in both low-and high-risk prenatal populations. Virtual visits provided time and cost savings and reduced appointment cancellations and missed appointments. In addition, virtual visits and postpartum monitoring from home can help health care providers improve the experience for expecting and new parents. 7 Our study was limited by it being a retrospective study performed at a single academic institution. The reason that some patients chose telehealth over inperson visits is not immediately obvious and we did not question their reasoning for electing for telehealth visits. The findings should be prospectively validated at other institutions. In addition, the small number of patients with multiple telehealth visits may limit the generalizability of the study.
Strengths of this study are that it was completed at a relatively large high-risk practice and number of subjects with a significant proportion of underrepresented minorities and the racial breakdown in population. In summary, our study concludes that there may be certain patients who are more likely to have a combination of in-person and telehealth visits.
The similar delivery outcomes suggest that telehealth is a safe option to combine with in-person visits to maximize scheduling flexibility for patients, physicians, and clinic staff as well as a popular option for those that have utilized telehealth previously. These findings will assist other MFM institutions and are encouraging for the continuation of telehealth visits after the pandemic.
One area that may see improvement is in making scheduled postpartum visits, where ACOG recommends a visit within the first 3 weeks postpartum followed by ongoing care as needed up to 12 weeks, but many patients (40%) are unable to make these appointments, especially in populations with limited resources. 20 It has been shown by Wu et al. 7,19 that virtual visits and postpartum monitoring from home can help health care providers improve the experience for expecting and new parents as well as contraception management. Difficulty with lack of follow-up or inappropriate gaps between appointments due to travel, illness, or other difficulties could be ameliorated by implementing a telehealth visit option for patients where clinically appropriate.