Practice change following an interdepartmental training on early pregnancy loss management in the emergency department

Abstract

Early pregnancy loss (EPL), or miscarriage, is the most common complication of early pregnancy, and pregnant people experiencing EPL most often seek care from their primary obstetric provider or from an emergency department (ED). However, it is uncommon for providers outside obstetrician-gynecologists to be trained in EPL management, and nursing and support staff are also not often included. This study assessed change in practice following an interprofessional training intervention of self-reported use of manual uterine aspiration for management of EPL in the ED. We conducted a single arm pre-test/post-test interventional study design with a matched sample of pre- and post-test data from participants at 8 academic medical center sites that received a training intervention from the Training, Education & Advocacy in Miscarriage Management project between March 2016 and May 2017. We assessed the provision of EPL services in the ED and provider attitudes about these services. Our sample included 80 faculty physicians, resident physicians, and advanced practice clinicians from 8 academic medical center sites. Our findings demonstrate a positive association between participation in the training intervention and self-reported use of manual uterine aspirator in the ED for management of EPL, as well as increased belief that coordination between departments within the institution was smooth post-training. Our findings suggest that the training intervention was successful in influencing the practice of management of EPL using manual uterine aspiration in the ED, highlighting the importance of engaging EDs in training efforts.

Keywords:

• Interprofessional education
• interprofessional learning
• role clarity
• teamwork
• miscarriage

PUBLIC INTEREST STATEMENT

Early pregnancy loss, also called miscarriage, is a very common complication of early pregnancy, occurring in 15-20% of all clinically recognized pregnancies. When patients go to a healthcare facility for care, in most circumstances patients should be able to choose from three types of management options, since all options are effective, acceptable to patients, and no management approach has been shown to result in different long-term outcomes. People experiencing early pregnancy loss often seek care from an obstetrician-gynecologist or go to the emergency room. However, it is uncommon for providers who are not obstetrician-gynecologists to be trained in early pregnancy loss care. This study found positive change in practice in the management types offered in emergency rooms following an interprofessional training program about early pregnancy loss management, suggesting that engaging emergency department service delivery teams in training is important.

1. Introduction

Early pregnancy loss (EPL), or miscarriage, is the most common complication of early pregnancy, occurring in 15–20% of all clinically recognized pregnancies (Ventura et al., 2012). EPL is defined as a nonviable, intrauterine pregnancy with either an empty gestational sac, or a gestational sac containing an embryo or fetus without cardiac activity before 13 weeks gestation ACOG Practice bulletin no. 200: Early pregnancy loss, 2018). Accepted treatment options include expectant management, medication treatment, or surgical uterine evacuation, all of which have been shown to be reasonably effective and acceptable to patients; no management approach has been shown to result in different long-term outcomes ACOG Practice bulletin no. 200: Early pregnancy loss, 2018). To date, EPL in the United States has largely been managed by uterine aspiration in the operating room (L. H. Harris et al., 2007; Dalton et al., 2010). However, all three forms of miscarriage management can be offered in other appropriately equipped settings (Jauniaux et al., 2005), including in the emergency department (ED), which may decrease delays in obtaining care, facilitate ED throughput by circumventing the need for an operating room, and decrease costs (Kinariwala et al., 2013; Dennis et al., 2015).

EDs are often the site of first contact in the healthcare system for patients with early pregnancy bleeding. Approximately 500,000 patients per year present to EDs with vaginal bleeding during early pregnancy, accounting for an estimated 1.6% of all ED visits (Wittels et al., 2008). No data exist on the number of pregnancies diagnosed in EDs, but there are an estimated 8.6 million pregnancy tests ordered in the ED each year for women ages 14 to 21 (Goyal et al., 2013). Patients with vaginal bleeding in early pregnancy seek medical care in EDs for a variety of reasons which may include lack of established prenatal care, occurrence of symptoms outside regular clinic hours, level of perceived clinical urgency, or other barriers to accessing routine care (Baird et al., 2018). Despite evidence suggesting early pregnancy bleeding and EPL can be well-managed outside the operating room, including in the ED, efforts to increase outpatient management access have encountered challenges (Darney, VanDerhei et al., 2013; Darney, Weaver, Vanderhei et al., 2013; Darney, VanDerhei et al., 2013).

The Training, Education and Advocacy in Miscarriage Management (TEAMM) program provides tailored training interventions to assist clinical and academic sites around the United States in integrating EPL management into their services and curricula. The TEAMM project has provided EPL management trainings since 2008, primarily to individual family medicine practices, OB/GYN practices, and family medicine residency programs. The inclusion of clinical and administrative support staff in EPL management trainings at family medicine residency sites has been shown to facilitate the integration of comprehensive EPL services in the outpatient clinic setting (Darney, VanDerhei et al., 2013; Darney, Weaver, Vanderhei et al., 2013; Darney, VanDerhei et al., 2013). Comprehensive EPL services are defined as offering expectant, medication, and outpatient surgical uterine aspiration to eligible patients. Using this inclusive all-staff training model, in 2016 TEAMM began providing interdepartmental trainings at large academic medical centers with the goal of decreasing barriers to comprehensive EPL management not only in the outpatient clinic setting, but in the ED setting as well. The purpose of this study was to examine the influence of an interdepartmental training intervention on EPL management in the ED at academic medical centers. We hypothesized that interdepartmental training would lead to increased outpatient management using MVA and medication among those who participated in the training.

2. Methods

The TEAMM program led interdepartmental and interdisciplinary training sessions at eight academic medical centers from March 2016 through May 2017. Each training session consisted of two half-days and included clinical providers and clinical and administrative staff from the OB/GYN, emergency medicine, and/or family medicine departments. The first half-day was geared towards clinicians, clinical support staff, and administrative support staff and began with a didactic session on EPL that included information on the epidemiology and diagnosis of EPL, the three management options (expectant, medication, and aspiration), pain management, and follow-up. It also included a simulation workshop for practicing manual uterine aspiration (MUA) to further participants’ understanding of the procedure, using papayas as a uterine model. This was followed by an exercise that encouraged reflection on the values each member of the healthcare team brings to EPL care and highlighted the importance of providing non-judgmental care and supportive counseling to patients. An additional exercise facilitated discussion of systems barriers that participants identified as being important to address in order to implement EPL services in their setting. The second half-day focused on clinical and administrative support staff. It included a brief didactic review of management options and then addressed support staff roles as well as outpatient clinic and ED barriers to implementing the full array of miscarriage management options in those settings.

2.1. Study design

This study used a single arm pre-test/post-test interventional study design with a matched sample of pre- and post-test data from participants at 8 academic medical center sites that received the TEAMM training intervention between 2016 and 2017 to assess how the provision of EPL services in the ED setting changed following the interdepartmental, interdisciplinary training provided. All study data were collected and managed using REDCap electronic data capture tools hosted at University of Washington (Harris et al., 2009; Harris et al., 2019).

2.2. Participants

There were a total of 475 unique participants at the 8 TEAMM training sites from 2016 and 2017, of which 418 responded to the pre-test. For this study, only practicing providers (physicians, nurse practitioners, nurse midwives, and physician assistants) in OB/GYN, Emergency Medicine (EM), or Family Medicine who responded to both the pre- and 6-month post-test were included in order to prevent over-estimation of practice information. Support staff, nursing staff, and students were excluded from the analysis as were practicing providers who responded only to the pre-test. Two hundred twenty-five of 418 total respondents (53.8%) were practicing providers, and of these providers, 80 of 225 (35.5%) responded to both the pre- and post-tests. Data were linked using a unique ID number and all results were de-identified for the purpose of analysis. Pre-test data and demographics were collected in advance of each training using a paper form. Survey pre-test items included training and current practice; options offered to patients who present with EPL; locations where EPL services were offered; opinions and knowledge about EPL services; and perceived barriers to EPL management. Using contact information gathered during each training, participants were sent an online follow up survey at 6-months post-training. Items on the 6-month post-test measuring self-reported management practice were the same as those on the pre-test.

Our primary outcome was regular offering of surgical management using MUA in the ED, measured by change in who reported providing this service pre- to post-intervention. Variables related to EPL management offerings and settings were answered with “yes,” “no,” or “I’m not sure”; the “I’m not sure” responses were treated as missing data and excluded from the analysis to better measure real change between pre- and post-tests. An additional option of “referral to other providers” for these questions was excluded in our analysis and reporting as it was inadvertently left off the pre-test and thus could not be compared. Variables related to attitudes and barriers were grouped into those who agreed or strongly agreed and those who disagreed or strongly disagreed. Analyses comparing respondents by medical specialty compare only OB/GYN and Emergency Medicine (EM) respondents due to small sample sizes of Family Medicine and midwifery respondents.

2.3. Statistical analyses

We first characterized the sample using descriptive statistics. To compare changes in practice and attitudes and perceptions of barriers from pre-test to 6-month post-test, we used the McNemar-Bowker test for “within subjects” paired comparisons. Using a matched pair analysis helps eliminate the variation between samples that could be caused by other factors, for example, in this case, the many other differences between training sites that might impact practice change, such as leadership, market forces, or regional cultural differences. We also analyzed pre- and post-test data by site using summary statistics for select variables of interest as a way to confirm overall trends in our results and to confirm that no one site was inadvertently driving our overall findings. This study was reviewed by the University of Washington Human Subjects Division and considered exempt from review by their Institutional Review Board in accordance with the federal regulations under 45 CFR 46.101/ 21 CFR 56.104 in category 1 or 2. Data were analyzed using SPSS 19.0.

3. Results

3.1. Sample characteristics

A total of 80 practicing providers were included in the site comparison pre- and post-test study sample. These providers came from each of the 8 sites, though completion of the post-test ranged from 5.0% of the total sample from a specific site (N = 4) to 22.5% of the sample (N = 18) [Table 1]. The majority of respondents were physicians (N = 69, 86.3%) and most respondents worked in OB/GYN (N = 60, 75.0%) or EM (N = 15, 18.8%). Most were either faculty (N = 32, 40.0%) or trainees, i.e. residents or fellows (N = 38, 47.5%). The providers included in this analysis were similar demographically to all providers trained, i.e. all those who responded to the pre-test (data not shown).

Table 1. Demographics of respondents

		Providers who responded to both pre- and post-test (study sample, N = 80)
Site	1	18 (22.5%)
	2	4 (5.0%)
	3	6 (7.5%)
	4	15 (18.8%)
	5	11 (13.8%)
	6	14 (17.5%)
	7	8 (10.0%)
	8	4 (5.0%)
Role	Physician (MD or DO) -Attending physician -Resident/fellow	69 (86.3%) 32 (40%) 38 (47.5%)
	Advanced Practice Clinician Provider*	11 (13.8%)
Medical specialty	OB/GYN	60 (75.0%)
	Emergency Medicine	15 (18.8%)
	Other (Family Medicine, Midwifery)	5 (6.3%)
Gender	Male	14 (17.5%)
	Female	66 (82.5%)
	Other**	0 (0%)
Age	Under 30	15 (18.8%)
	30–39	42 (52.5%)
	40–49	12 (15.0%)
	50–59	6 (7.5%)
	60 and over	1 (1.3%)

*Advanced Practice Clinician Providers included advanced registered nurse practitioners, other nurse practitioners, certified nurse midwives, and physician assistants

**Other gender combined transgender, gender non-conforming, and those who selected “prefer not to say” answer categories with those who did not respond

To check for any overweighting by site in our sample, the proportion of respondents by site on the post-test survey was compared to the proportion of unique participants (pre-test respondents) by site; the maximum difference in proportion of respondents by site between tests was about 3.7% though most sites were within 1% to 1.5% difference (data not shown).

All respondents reported that they saw at least one patient per month with EPL on both the pre- and post-tests, and more than 30% of respondents indicated that they saw 5 or more patients with EPL per month at both assessments. Initial assessment of knowledge around EPL management on the pre-test was very high for all participants and remained high in the post-test (data not shown).

3.2. Prior training in EPL management

In the pre-test, most OB/GYNs reported that they had prior training in EPL management, with any technique (N = 53, 88.3%), and almost all reported familiarity with manual uterine aspiration management (N = 54, 90.0%) and medication management (N = 60, 100.0%). The emergency medicine providers reported less training in EPL management (N = 6, 40.0%), and less familiarity with both manual uterine aspiration management (N = 6, 40.0%) or medication management (N = 13, 86.7%).

3.3. EPL practice change

A significant difference was seen between the pre- and post-tests for all providers who reported regularly offering ED-based surgical management using MUA between the pre- and post-tests (42.2% vs. 65.6%, p = .001) [Table 2]. The increase in offering ED-based surgical aspiration to patients between the pre- and post-test is seen across almost all training sites; only 1 of the 8 sites saw a decrease in providers reporting offering ED-based surgical aspiration post-test and this site had a very small respondent sample (N = 4) (data not shown). No significant change between the pre- and post-test was seen for any other EPL management option or any of the other settings where services were offered, though there was a trend, particularly among the OB/GYN providers, to report feeling “that my medical setting supports office/outpatient/ED management of miscarriage” more often in the post-test (82.6% vs 92.8%, p = .065). When analyzed by provider specialty, there was a significant difference among OB/GYN providers between the pre- and post-test (80.0% vs 92.7%, p = 0.039), but not among EM providers [Table 2, EM provider data not shown].

Table 2. EPL management options and locations among all providers and OB/GYN providers, pre- and post-training

		All Providers (N = 80)				OB/GYN Providers only (N = 60)
		N*	Pre	Post	p	N*	Pre	Post	p
EPL management option regularly offered to patients	Expectant (watch and wait)	77	77 100.0%	77 100.0%	–	58	58 100.0%	58 100.0%	–
	Medication with misoprostol or cytotec	76	74 97.4%	75 98.7%	1.000	60	60 100.0%	60 100.0%	–
	Office-based surgical (aspiration using MVA)	68	56 82.4%	59 86.8%	.581	56	48 88.9%	54 96.4%	.125
	Emergency-room based surgical (aspiration using MVA)	64	27 42.2%	42 65.6%	.001	49	24 49.0%	38 77.6%	.003
	Operating- room based surgical (aspiration or D&C)	75	70 93.3%	69 92.0%	1.000	58	58 100.0%	55 94.8%	–
Services offered at this location	Office or clinic setting	75	68 90.7%	70 93.3%	.727	60	59 98.3%	60 100.0%	–
	Emergency room	65	49 75.4%	55 84.6%	.180	48	37 77.1%	43 89.6%	.109
	Operating room	74	66 89.2%	65 87.8%	1.000	58	57 98.3%	56 96.6%	1.000
In general, I feel that my medical setting supports office/outpatient/ED management of miscarriage		69	57 82.6%	64 92.8%	.065	55	44 80.0%	51 92.7%	.039

*valid N, excludes missing and “I’m not sure”

These analyses excluded any responses of “I’m not sure” as missing data; for most of these variables these missing data were minimal (1 or 2 respondents). However, 6 (7.5%) respondents pre-test and 4 (5.0%) respondents post-test were “not sure” if office-based surgical management was regularly offered to patients; and 8 (10.0%) respondents pre-test and 4 (5.0%) respondents post-test were “not sure” if ED-based surgical management was regularly offered to patients. On the pre-test, 10 (12.5%) respondents were “not sure” if EPL services were offered in the ED; however, on the post-test, only 1 (1.3%) respondent did not know the answer to this.

3.4. Attitudes about EPL management

To consider whether attitudes about EPL service provision had changed in addition to actual practice as described above, we compared those who agreed or strongly agreed with several attitudes statements and several perceived barriers [Table 3]. The greatest change between the pre- and post-test among respondents was in their agreement with the statement “coordination between departments at my institution is smooth when patients present [to the ED] with EPL” (56.6% vs. 77.6%, p = .007). Similar to the increase in offering ED-based surgical aspiration, respondents’ change in perception of coordination between departments between the pre- and post-test is seen across all but one training site with a small sample size (data not shown).

Table 3. Attitudes pre- and post-training about and barriers to EPL management among all providers and OB/GYN providers

	All Providers (N = 80)				OB/GYN Providers only (N = 60)
	N*	Pre	Post	p	N*	Pre	Post	p
Agreement or strong agreement with attitudes statements
Primary care providers should have MVA and medication miscarriage management skills	79	66 83.5%	72 91.1%	.070	59	46 78.0%	52 88.1%	.070
Primary care advanced practice clinicians (ARNPs, CNMs, PAs) should have MUA and medication management skills	78	61 78.2%	64 82.1%	.453	59	44 74.6%	48 81.4%	.219
Miscarriage can be appropriately managed with MUA in an office/outpatient clinic setting	80	80 100.0%	80 100.0%	–	60	60 100.0%	60 100.0%	—
Miscarriage can be appropriately managed with MUA in an emergency department setting	79	76 96.2%	77 97.5%	1.000	59	59 100.0%	59 100.0%	–
MUA in the outpatient setting can be appropriate for a patient experiencing miscarriage who desires to be awake	78	77 98.7%	78 100.0%	–	58	58 100.0%	58 100.0%	–
Coordination between departments at my institution is smooth when patients present with early pregnancy loss	76	43 56.6%	59 77.6%	.007	58	28 48.3%	44 75.9%	.002
Agreement or strong agreement with perceived barriers to integration in current practice setting
Lack of training	79	67 84.8%	54 68.4%	.011	60	49 81.7%	36 60.0%	.011
Lack of support from administration, colleagues, or staff	79	63 79.7%	55 69.6%	.115	60	47 78.3%	37 61.7%	.031
Scope of practice conflicts	79	11 13.9%	6 7.6%	.180	60	2 3.3%	1 1.7%	1.000
Miscarriage management perceived as induced abortion	78	33 42.3%	29 37.2%	.585	59	24 40.7%	22 37.3%	.839

*valid N, excludes missing

There was no change in agreement with statements about which types of providers should have EPL management skills or settings where EPL can be appropriately managed at the time of the post-test, though there was a trend toward agreeing that primary care providers should have manual vacuum aspiration and medication management skills (83.5% vs 91.1%, p = .070). Both “lack of training” and “lack of support from administration, colleagues, or staff” were perceived to be barriers by a majority of respondents pre- and post-test, although there was a significant change in agreement that lack of training was a barrier (84.8% vs. 68.4%, p = .011). Perception of scope of practice conflicts and perception of EPL management as induced abortion were not considered barriers to service provision by the majority of respondents and did not change between the pre- and post-tests. When results were compared between OB/GYN and EM providers, it was clear that all significant findings were driven by practice change among the OB/GYN providers [Table 3; EM provider data not shown].

4. Discussion

We report a positive association of the TEAMM training intervention with self-reported use of MUAs for EPL in the ED by providers, particularly OB/GYN, suggesting that the TEAMM intervention was successful in influencing practice patterns for management of EPL using MUA among these providers.

Despite strong evidence that health care delivered by well-functioning teams leads to better outcomes, health professionals have historically functioned as highly skilled but autonomous by role, both within and between specialties. Inter-professional education (IPE) prepares a workforce that is experienced in collaborative working relationships that can then lead to improved health outcomes through a less fragmented health care system (WHO, 2010). A 2013 review of studies looking at the effectiveness of IPE demonstrated positive outcomes in the ED related to departmental culture, patient satisfaction, collaborative team behavior and lowering of clinical error rates for ED teams (Reeves et al., 2013).

We observed significant post-intervention changes in attitudes among OB/GYNs of their perception that their setting supported outpatient and ED management of miscarriage. OB/GYN providers reported an improvement in support for outpatient and ED management of EPL. Both OB/GYN and non-OB/GYN providers reported an overall perceived improvement in coordination between specialists for patients presenting with EPL. Post-training, more respondents knew what management options were available to patients. Our interdepartmental TEAMM trainings gave attendees the opportunity to learn from each other and to share perspectives and experiences related to EPL management. This helped to eliminate the assumption that managing EPL patients was complex, paving the way for an approach to care requiring fewer referrals of EPL patients to different departments and facilitating the provision of evidence-based management in the ED.

There were several limitations to our study. First, we cannot be certain that those who responded to the post-test are representative of all practicing provider participants trained given the lower response rate to the post-test survey. It is possible that those who did change practice were also those most compelled to respond to the 6-month follow-up survey. Second, the study had an overall low response rate. While a low response rate is typical for surveys administered to health providers online, it is a limitation in the study design that cannot be overlooked as it reduces the generalizability of our results. However, the matched pre-post study design to describe individual-level change gave us more power to assess differences by analyzing several variables of interest pre-/post-test by site to confirm that no one site skewed our overall trends. Third, it is possible that other confounding factors may have influenced the changes seen between pre- and post-test that we were not able to measure or control for despite the matched-pair study design. Fourth, the study did not measure practice change among participants in the training who were not clinicians. This was intentional as a way to prevent overestimation of practice change, but it is worth noting that in this study, an assumption is made that their participation is essential to success as measured among providers. Finally, the first, second, and fifth authors of this paper were also involved in the planning and provision of the trainings, which may have introduced potential bias in the study design and interpretation of results that emphasized the benefits of the intervention.

In most circumstances, patient preference should guide treatment choice for EPL management since all options are effective, acceptable to patients, and no management approach has been shown to result in different long-term outcomes. Institutions that want to increase patient-centered EPL care and decrease fragmentation of services should consider working and training teams collaboratively to implement EPL practice change in the outpatient setting. Patients will benefit from more coordinated EPL care and progress towards this goal will occur successfully when departments and staff work collaboratively in their efforts to implement change. It is important for institutions desiring to implement practice change to educate key stakeholders, including staff, about how the MUA procedure is performed and how to communicate with patients experiencing EPL in order to decrease resistance commonly associated with offering this procedure. The training model also demonstrates the importance of establishing clear protocols and procedures identifying the roles and responsibilities of all staff that care for EPL patients. Finally, it is important to develop and clarify systems to guide the staff in managing supplies, patient scheduling, sedation needs, and patient flow.

Since many patients who experience EPL prefer to be managed outside the operating room, this study, which found benefits to interdepartmental training, can be used as a resource for institutions that want to improve their EPL management service offerings for patients in the locations to which they present.

5. Conclusions

Our findings suggest that the TEAMM interdepartmental training intervention was successful in influencing practice patterns for management of EPL using MUA. Inclusion of emergency medicine service delivery teams in efforts to implement EPL services by OB/GYN providers in the ED setting is an important facilitator to offering eligible patients diagnosed with EPL this management option, which will lead to greater patient satisfaction with care.

Acknowledgments

The authors thank Robin Supplee, Denise Johnson May, Sophia Dzilenski, Rachel Schaeffer, Sara Magnusson, and the University of Washington Department of Obstetrics & Gynecology for their invaluable contributions to this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author, EEM. The data are not publicly available due to their containing information that could compromise the privacy of research participants.

Additional information

Funding

Financial support for the study was provided by the University of Washington Department of Obstetrics & Gynecology.

Notes on contributors

Erin E McCoy

Erin E McCoy, MPH, is a Program and Research Manager at the University of Washington Department of Obstetrics and Gynecology in the Division of Complex Family Planning. She holds a Master of Public Health from Columbia University in Population and Family Health. The work of her group focuses on improving patients’ experience with early pregnancy loss care through training and education of the entire healthcare team. Their work also focuses on outcomes related to contraception and abortion.