Saturday, November 23, 2024
Virtual Heart Care
Telehealth 

Postpartum Care in the Time of COVID-19: The Use of Telemedicine for Postpartum Care

Editor


Introduction

The postpartum period is an extremely important time for maternal and infant health and wellbeing. More than half of all pregnancy-related deaths occur in the postpartum period.1 In addition, the postpartum period is a time of great transition and many women experience complications such as heavy bleeding, pain, breastfeeding difficulties, fatigue, exhaustion, depression, and urinary incontinence.2 In the past, it was recommended that women attend a postpartum visit within 6–12 weeks of delivery.2 More recently, the American College of Obstetrics and Gynecology (ACOG) has updated their recommendations to suggest that all women should have earlier contact with a medical provider—ideally around 2–3 weeks postpartum.2

Many patient populations are not yet meeting the Healthy People 2020 target of 90.8% attendance at the postpartum visit.3 This is even more so the case for socioeconomically disadvantaged populations. The National Committee for Quality Assurance tracks postpartum visit attendance rates and have found that in 2019 visit attendance was 75.2% in patients with Medicaid and 80.7% in patients with Commercial Health Maintenance Organization insurance.4

At our large urban academic medical center, the postpartum visit attendance rate was 67% in 2013 and 64% at one of our Federally Qualified Health Centers (FQHCs) in 2019. These rates are well below the recommended postpartum attendance rate.5 In response to the COVID-19 pandemic, a significant proportion of outpatient Obstetrics and Gynecology care was converted to telemedicine. At our FQHC, telemedicine visits began on March 23, 2020. The use of telemedicine in Obstetrics dates back to the 1970s.6 Several studies, mainly in developing countries and rural settings, have explored the use of telemedicine and mobile technology in health (m-Health) in the delivery of obstetric care during the prenatal and postpartum periods.7–17

In developed countries, these studies have mainly explored the role of m-Health for postpartum depression screening and treatment,7 remote blood pressure monitoring,8–11 lactation assistance12,13 and gestational weight gain and postpartum weight loss.14–17 While the results are mixed, studies suggest that telemedicine and m-health technologies have the potential to reduce barriers to accessing care for remote and underserved populations and to improve maternal and infant outcomes.7–17 However, data do not currently exist about the use of telemedicine in providing care during the comprehensive postpartum visit, especially in urban populations in developed countries.

The objective of our study was to compare postpartum visit attendance before and after the implementation of telemedicine at an urban FQHC. We hypothesized that telemedicine would result in increased attendance at the comprehensive postpartum visit. Secondary objectives in the study were to examine rates of emergency department (ED) visits in the postpartum period, hospital readmissions, contraception use at postpartum visit, social work visits, depression screening, and breastfeeding rates before and after the implementation of telemedicine.

Methods

We carried out a retrospective observational cohort study of women who received prenatal care at an urban FQHC in the Bronx, New York, and delivered at a large teaching hospital within our medical system. Our medical system is a large academic center, which serves a socioeconomically and racially diverse underserved population. The study was approved by the Albert Einstein College of Medicine Institutional Review Board on June 12, 2020.

Subjects were identified based on chart review of all patients who delivered at our institution during the time period being studied and received their prenatal care at the FQHC. All data were retrieved from the patient’s medical records. At our institution, a single electronic medical record system is used for both inpatient and outpatient care. Inclusion criteria were patients age ≥18 years who had a vaginal or cesarean delivery at ≥36 weeks gestation, received prenatal care at the FHQC, and subsequently delivered at our teaching hospital. Women were excluded from the study if they had a fetal or neonatal death or demise.

Women were divided into three cohorts based on their dates of delivery: pre-COVID, peak-COVID, and ongoing-COVID. The pre-COVID cohort included women who delivered between September and November 2019, the peak-COVID cohort included deliveries between February and April 2020, and the ongoing-COVID cohort included deliveries between June and August 2020.

The cohorts were created so that the postpartum time period coincided with the implementation of telemedicine visits at our institutions. As such, women in the peak-COVID cohort would have been six or more weeks postpartum at the time that telemedicine was started and during the height of the initial COVID-19 pandemic. Therefore, the corresponding 3-month postpartum periods for the women in the pre-COVID, peak-COVID, and ongoing-COVID cohorts were October 2019 through February 2020, March 2020 through July 2020, and August through November 2020, respectively.

We performed a chart review of the electronic medical record to collect data on patient demographics and obstetric course, including age at the time of delivery, race, ethnicity, gravidity and parity, preferred language, insurance status, gestational age at delivery, number of prenatal visits, hypertensive disorders or diabetes, and mode of delivery. Patients were counted as having a hypertensive disorder or diabetes if they had the condition before pregnancy or developed it during pregnancy or their intrapartum course. Data were collected on the type of postpartum visit—whether it was in-person, via telephone, or via video. Data were also collected on ED visits and hospital readmission rates in the first 12 weeks postpartum as well as contraception use, breastfeeding status at the postpartum visit, social work visits, and postpartum depression screening rates.

Postpartum depression screening was done at the postpartum visit using the validated Patient Health Questionnaire-2 (PHQ2) screen, which was the standard practice at the FQHC. Postpartum outcomes were also analyzed by the timing of the postpartum visit. Outcomes were compared based on if the visits occurred at <6 weeks, 6–12 weeks, or >12 weeks.

We estimated relative risk and 95% confidence intervals (95% CIs) of the primary and secondary outcomes across the three cohorts using log-binomial regression models. The pre-COVID period was the reference. For analyses combining multiple time periods where some women may have had multiple visits (<6 weeks, 6–12 weeks, and >12 weeks), we used generalized estimating equations with a Poisson distribution and log link to account for the repeated observations. The model was not adjusted as no individual level characteristics were thought to be causally related to the primary independent variable based on date of delivery. While data were collected on the group of patients who had a postpartum visit >12 weeks, the numbers were too small to be analyzed, so, results were not provided for this postpartum time period. In addition, an a priori power analysis was not performed. All analyses were done using R (v4.0.5).

Results

Three hundred fifty-nine women (n = 359) were included in the study. There were 142 women in the pre-COVID cohort, 132 women in the peak-COVID cohort, and 85 women in the ongoing-COVID cohort. The three cohorts of women were similar in respect to age at delivery, race, ethnicity, parity, preferred language, Medicaid/Medicare use, gestational age at delivery, type of delivery, and rate of hypertensive disorders (Table 1). Women in the ongoing-COVID cohort had one fewer prenatal visit than those in the pre-COVID or peak-COVID cohorts (8.0 vs. 8.8 and 9.3, respectively) (Table 1). Moreover, women in the ongoing-COVID cohort had a significantly increased rate of diabetes in pregnancy (34.9% vs. 16.2% in the pre-COVID cohort, and 15.5% in the peak-COVID cohort; p < 0.01) (Table 1).

Table 1. Cohort Characteristics
  PRE-COVID, N = 142 PEAK-COVID, N = 132 ONGOING-COVID, N = 85 p
Age 29.5 29.4 30.3 0.47
Black race 23.2% 25.0% 21.2% 0.87
Hispanic ethnicity 43.7% 42.4% 45.9% 0.88
Gravidity 3.0 2.9 3.5 0.20
Parity 1.3 1.1 1.5 0.05
Preferred language English 64.8% 70.5% 71.8% 0.98
Medicaid/Medicare 76.8% 74.2% 68.2% 0.67
Gestational age at delivery, weeks 38 38 38 0.36
Number of prenatal visits 8.8 9.3 8.0 0.03
Cesarean section 35.2% 34.1% 32.9% 0.26
Hypertension 28.9% 34.5% 28.2% 0.37
Diabetes 16.2% 15.5% 34.9% <0.01

Across the cohorts, the usage of telemedicine to provide postpartum care increased from 1.4% in the pre-COVID cohort to 59.6% in the peak-COVID cohort and 48.0% in the ongoing-COVID cohort (Fig. 1). The type of telemedicine visit performed also changed from primarily telephone visits in the peak-COVID cohort to video visits in the ongoing-COVID cohort (Fig. 1).

Fig. 1.

Fig. 1. Postpartum visit type across cohorts. In-person: patients who had in-person postpartum visits. Video: patients who had video postpartum visits. Telephone: patients who had telephone postpartum visits.

There was no significant difference in visit attendance at the comprehensive 6–12 week postpartum visit by cohort (Table 2). Fifty-two percent (52%) of patients attended a visit in the pre-COVID cohort, 43% in the peak-COVID cohort, and 56% in the ongoing-COVID cohort (Table 2). This was also true if the postpartum visit occurred outside of the 6–12 week time frame. Seventy-five percent of women in the pre-COVID cohort, 65% in the peak-COVID cohort, and 80% in the ongoing-COVID cohort attended a visit at any point in the 6-month postpartum period (Table 2). In addition, the peak-COVID cohort had the lowest attendance rates regardless of when the postpartum visit occurred (Table 2). Of note, these rates also included women who did not have a postpartum appointment scheduled, and so did not attend an appointment.

Table 2. Postpartum Visit Attendance Across Cohorts
COHORT 6–12 WEEK VISIT ANY POSTPARTUM VISIT
ATTENDANCE RATE, % RR (95% CI) p ATTENDANCE RATE, % RR (95% CI) p
Pre-COVID 52 Ref. 75 Ref.
Peak-COVID 43 0.83 (0.64–1.07) 0.14 65 0.86 (0.74–1.01) 0.07
Ongoing-COVID 56 1.08 (0.85–1.38) 0.52 80 1.06 (0.92–1.22) 0.41

For women who had a 6–12 week postpartum appointment scheduled, the show rate was increased if the visit was scheduled as a telemedicine visit, compared to an in-person visit, although this difference was not statistically significant. This was true for both the peak-COVID cohort (76% vs. 65%, RR 1.17, 95% CI 0.87–1.57) and the ongoing-COVID cohort (85% vs. 74%, RR 1.16, 95% CI 0.90–1.50) (Table 3). However, the increase in visit show rate for telemedicine visit compared to in-person visit was significant in the peak-COVID cohort if the visit occurred <6 weeks postpartum (96% vs. 57%, RR 1.70, 95% CI 1.31–2.22) (Table 3). Overall, there appeared to be a trend toward increasing visit attendance if visits were done by telemedicine compared to if they were done in-person, but this trend was not statistically significant.

Table 3. Postpartum Visit Show Rate by Type of Visit
  TYPE OF VISIT 6–12 WEEK VISIT <6 WEEK VISIT
SHOW RATE, % RR (95% CI) p SHOW RATE, % RR (95% CI) p
Peak-COVID In Person 65 Ref. 57 Ref.
Telemedicine 76 1.17 (0.87–1.57) 0.31 96 1.7 (1.31–2.22) <0.001
Ongoing-COVID In Person 74 Ref. 88 Ref.
Telemedicine 85 1.16 (0.90–1.50) 0.26 96 1.09 (0.93–1.29) 0.29

There were no significant differences in hospital readmission rates, breastfeeding rates, social work visits, or contraception use across the three cohorts (Appendix Table A1). There was a significant reduction in ED visits between the pre-COVID cohort and the peak-COVID cohort (17% vs. 8%, RR 0.49, 95% CI 0.25–0.97) (Table 4). In addition, there was a significant decrease in the percentage of patients receiving a PHQ2 depression screen in both the peak-COVID cohort (22%, RR 0.3, 95% CI 0.20–0.45) and the ongoing-COVID cohort (33%, RR 0.45, 95% CI 0.32–0.65) when compared to the pre-COVID cohort (74%) (Table 4).

Table 4. Secondary Outcomes Across Cohorts
COHORT ED VISIT PHQ2 DEPRESSION SCREEN
RATE, % RR (95% CI) p RATE, % RR (95% CI) p
Pre-COVID 17 Ref. 74 Ref.
Peak-COVID 8 0.49 (0.25–0.97) 0.04 22 0.3 (0.20–0.45) <0.01
Ongoing-COVID 12 0.7 (0.35–1.38) 0.3 33 0.45 (0.32–0.65) <0.01

Discussion

Overall, postpartum visit attendance remained low at our urban FQHC even after the implementation of telemedicine. While our hypothesis that telemedicine would result in increased postpartum visit attendance rates across cohorts was not supported by our data, there was a trend toward increasing visit show rates within cohorts if the visit was telemedicine compared to in-person. There were no changes in hospital readmission rates across the cohorts, and patients were equally as likely to breastfeed, use contraception, and have a visit with the Social Worker across the cohorts. However, there are areas of the postpartum visit that did not translate well from the in-person visit to the virtual space—namely depression screening.

The attendance rates found in our study for any postpartum visit (65–80%) were comparable to those found by Wilcox et al.5 This study, and others, have identified factors associated with decreased attendance at the postpartum visit—including having Medicaid or no insurance, being Hispanic or Latino, and being <20 years of age.5 Therefore, it is not surprising that in a similarly socioeconomically disadvantaged population, our postpartum visit attendance rates would be similarly low. While the primary purpose of this study was to investigate the use of telemedicine for postpartum care, the impact of the COVID-19 pandemic cannot be separated from the results.

We see its impact on the decrease in the number of prenatal visits in the ongoing-COVID cohort compared to the other two cohorts and in the decrease in ED visits in the peak-COVID cohort compared to the other two cohorts. Other studies have also observed an overall decrease in health care utilization during the COVID-19 pandemic.18,19 While these studies were all quantitative and thus did not specifically investigate why these reductions were seen, this is possibly secondary to reluctance of patients to engage with the health care system due to decreased access and/or fear.

A strength of this study was that we were able to collect a large amount of data over several time periods, which allowed us to investigate multiple different outcomes. In addition, this is a novel study as no other studies have examined telemedicine for the comprehensive postpartum care visit in an urban setting in the developed world. The study is also very relevant to current events as the world continues to deal with the COVID-19 pandemic while maintaining access to medical care.

The fact that this study was conducted during a pandemic and involves institutional changes that were rapidly implemented in response to a pandemic is a limitation of our study. We acknowledge that patient behavior and their decision to access health care may have been impacted by the pandemic, as some of our data suggest. It would therefore be important to validate these results outside of a global pandemic. Other limitations were the retrospective design of the study, lack of an a priori power analysis, and the relatively small sample size of the cohorts.

At our FHQC, PHQ2 screening is done by nurses at the intake portion of the visit. Nurses no longer performed intake as part of the telemedicine visit, and our data suggest that providers did not incorporate PHQ2 screening into their workflow. This likely explains our findings of the decline in depression screening and is an area that would need to be addressed when considering the longevity of telemedicine as an option for postpartum care at our institution. Women who gave birth in the ongoing-COVID cohort would have received most of their prenatal care at the peak of the COVID-19 pandemic and women in the peak-COVID cohort would have been delivering and receiving postpartum care during the peak of the pandemic as well. This likely reflects the decrease in postpartum visit attendance in the peak-COVID cohort compared to the other cohorts, although this difference was not statistically significant.

A second change that occurred at our institution because of the COVID-19 pandemic was gestational diabetes screening. We changed from a two-step method, using a 1-h 50 g glucose challenge test followed by a 3-h 100 g glucose tolerance test if the first screen was positive, to a one-step method, using a 2-h 75 g glucose tolerance test. While this alone does not explain the entire increase in diabetes rates seen in the ongoing-COVID cohort compared to the other two cohorts, the one-step approach has been shown to increase rates of patients ruling in for gestational diabetes, although the difference has not been found to be significant.20 Other possible explanations for the increase could be decreased care utilization during the pandemic and later screening and increased weight gain in pregnancy. However, we do not have the data to support these hypotheses.

Patients in the ongoing-COVID cohort had one less prenatal visit than those in the pre-COVID and peak-COVID cohorts. While this was a statistically significant difference (p = 0.03), we did not consider it a clinically significant difference. Moreover, the difference in visits likely represents the fact that patients in the ongoing-COVID cohort received their prenatal care during the peak of the pandemic and had decreased access to medical care due to barriers brought on secondary to the pandemic and not that they had differences in health care seeking behavior.

At our institution, telemedicine was rapidly implemented during the peak of the COVID pandemic and was a new concept for many of the providers. Our institution, providers, and patients had challenges with the new video platform during the first few months of use. This accounts for the lower usage of videos and higher telephone encounters noted in the peak-COVID versus ongoing-COVID cohorts.

Our study suggests that telemedicine may be a viable tool to increase postpartum visit attendance, especially in an urban underserved population. However, more research is needed to explore if this trend would reach significance with a study powered to detect this difference. Incorporating depression screening into the virtual patient encounter in a way that is effective and not burdensome for the provider is also an important future consideration. Possible considerations include having nursing staff perform intake phone calls/video visits to conduct depression screening before the scheduled postpartum telemedicine visits, allowing patients to self-complete the screening form before their visit via the electronic medical record interface, fax, or email; or incorporating the depression screening tool into the postpartum visit note template so that the provider is prompted to complete it during the video visit.

Conclusions

Regardless of its limited impact on visit attendance, telemedicine currently appears to be a potentially safe alternative to in-person visits at our institution. Although we still have to improve our postpartum visit attendance rates and further research is needed to parse out the impact on the COVID-19 pandemic on these results, telemedicine was comparable to in-person postpartum care in terms of attendance, without an increase in hospital visits or readmissions. Postpartum depression screening, however, needs to be better integrated into these telemedicine visits as it remains an important part of the comprehensive postpartum encounter and should be routinely performed.

Authors’ Contributions

A.M.A.: Conceptualization (equal); Investigation (lead); Methodology (equal), and Original draft (lead). H.W.: Formal analysis; Review and editing (supporting). F.R.Z.: Investigation (supporting) and Original draft (supporting). J.R.W.: Investigation (supporting) and Original draft (supporting). T.L.B.: Conceptualization (equal), Methodology (equal), Supervision, and Review and editing (lead).

Acknowledgments

This work has been previously presented by A.M.A., T.L.B., H.W., J.M. Samaedam, J.R.W., and F.R.Z. “Postpartum Care in the Time of COVID-19: An Evaluation of the Use of Telemedicine for Postpartum Care.” Poster presentation at ACOG District II Virtual Annual Meeting. Online. September 24, 2021.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

REFERENCES

  • 1. Kassebaum NJ, Bertozzi-Villa A, Coggeshall MS, et al. Global, regional, and national levels and causes of maternal mortality during 1990–2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014;384:980–1004. Crossref, Medline, Google Scholar
  • 2. McKinney J, Keyser L, Clinton S, Pagliano C. ACOG Committee Opinion No. 736: Optimizing postpartum care. Obstet Gynecol 2018;132:784–785. Crossref, Medline, Google Scholar
  • 3. Office of Disease Prevention and Health Promotion. MICH-19 increase the proportion of women giving birth who attend a postpartum care visit with a health care worker. Healthy People 2020. Published 2019. Available at https://www.healthypeople.gov/2020/data-search/Search-the-Data#objid=4855 (last accessed March 18, 2019). Google Scholar
  • 4. National Committee for Quality Assurance. Prenatal and Postpartum Care (PPC). Healthcare Effectiveness Data and Information Set (HEDIS). Published 2021. Available at https://www.ncqa.org/hedis/measures/prenatal-and-postpartum-care-ppc/ (last accessed October 26, 2021). Google Scholar
  • 5. Wilcox A, Levi EE, Garrett JM. Predictors of non-attendance to the postpartum follow-up visit. Matern Child Health J 2016;20(Suppl 1):22–27. Crossref, Medline, Google Scholar
  • 6. Odibo IN, Wendel PJ, Magann EF. Telemedicine in obstetrics. Clin Obstet Gynecol 2013;56:422–433. Crossref, Medline, Google Scholar
  • 7. Hantsoo L, Podcasy J, Sammel M, Epperson CN, Kim DR. Pregnancy and the acceptability of computer-based versus traditional mental health treatments. J Womens Health (Larchmt) 2017;26:1106–1113. Link, Google Scholar
  • 8. Hauspurg A, Lemon LS, Quinn BA, et al. A postpartum remote hypertension monitoring protocol implemented at the hospital level. Obstet Gynecol 2019;134:685–691. Crossref, Medline, Google Scholar
  • 9. Kalafat E, Benlioglu C, Thilaganathan B, Khalil A. Home blood pressure monitoring in the antenatal and postpartum period: A systematic review meta-analysis. Pregnancy Hypertens 2020;19:44–51. Crossref, Medline, Google Scholar
  • 10. Aquino M, Munce S, Griffith J, Pakosh M, Munnery M, Seto E. Exploring the use of telemonitoring for patients at high risk for hypertensive disorders of pregnancy in the antepartum and postpartum periods: scoping review. JMIR Mhealth Uhealth 2020;8:e15095. Crossref, Medline, Google Scholar
  • 11. Rhoads SJ, Serrano CI, Lynch CE, et al. Exploring implementation of m-Health monitoring in postpartum women with hypertension. Telemed J E Health 2017;23:833–841. Link, Google Scholar
  • 12. Kapinos K, Kotzias V, Bogen D, et al. The use of and experiences with telelactation among rural breastfeeding mothers: Secondary analysis of a randomized controlled trial. J Med Internet Res 2019;21:e13967. Crossref, Medline, Google Scholar
  • 13. Uscher-Pines L, Ghosh-Dastidar B, Bogen DL, et al. Feasibility and effectiveness of telelactation among rural breastfeeding women. Acad Pediatr 2020;20:652–659. Crossref, Medline, Google Scholar
  • 14. Mertens L, Braeken M, Bogaerts A. Effect of lifestyle coaching including telemonitoring and telecoaching on gestational weight gain and postnatal weight loss: A systematic review. Telemed J E Health 2019;25:889–901. Link, Google Scholar
  • 15. Van der Pligt P, Ball K, Hesketh KD, et al. A pilot intervention to reduce postpartum weight retention and central adiposity in first-time mothers: Results from the mums OnLiNE (Online, Lifestyle, Nutrition & Exercise) study. J Hum Nutr Diet 2018;31:314–328. Crossref, Medline, Google Scholar
  • 16. Sherifali D, Nerenberg KA, Wilson S, et al. The effectiveness of ehealth technologies on weight management in pregnant and postpartum women: Systematic review and meta-analysis. J Med Internet Res 2017;19:e337. Crossref, Medline, Google Scholar
  • 17. Gilmore LA, Klempel MC, Martin CK, et al. Personalized mobile health intervention for health and weight loss in postpartum women receiving women, infants, and children benefit: A randomized controlled pilot study. J Womens Health (Larchmt) 2017;26:719–727. Link, Google Scholar
  • 18. Roy CM, Bollman EB, Carson LM, Northrop AJ, Jackson EF, Moresky RT. Assessing the indirect effects of COVID-19 on healthcare delivery, utilization and health outcomes: A scoping review. Eur J Public Health 2021;31:634–640. Crossref, Medline, Google Scholar
  • 19. Haikin Herzberger E, Efros O, Herzberger S, Biron-Shental T, Shechter-Maor G. Differences in obstetric healthcare utilization and delivery complications before and after the COVID-19 pandemic—A retrospective study. J Matern Fetal Neonatal Med 2021:1–6. DOI: 10.1080/14767058.2021.1946506. Crossref, Medline, Google Scholar
  • 20. Saccone G, Khalifeh A, Al-Kouatly HB, Sendek K, Berghella V. Screening for gestational diabetes mellitus: One step versus two step approach. A meta-analysis of randomized trials. J Matern Fetal Neonatal Med 2020;33:1616–1624. Crossref, Medline, Google Scholar

Appendix

Appendix Table A1. Nonsignificant Secondary Outcomes across Cohorts
COHORT HOSPITAL READMISSION CONTRACEPTION USE BREASTFEEDING
RATE, % RR (95% CI) p RATE, % RR (95% CI) p RATE, % RR (95% CI) p
Pre-COVID 4 Ref. 75 Ref. 81 Ref.
Peak-COVID 5 1.29 (0.40–4.13) 0.67 74 0.99 (0.83–1.17) 0.86 75 0.92 (0.78–1.08) 0.31
Ongoing-COVID 4 1.00 (0.25–4.09) 1.00 72 0.95 (0.78–1.15) 0.60 68 0.84 (0.68–1.02) 0.08





Source link