Mate Whenua—follow-up after early medical abortion: study protocol for a randomized controlled trial

Abstract

Background

Early medical abortion (EMA) is safe and effective; an uncommon but crucial adverse outcome is ongoing live pregnancy. The best method of follow-up after EMA to detect ongoing pregnancy is a critical research gap. Few trials compare blood or urine pregnancy tests to ultrasound scans, and no trial compares these tests to each other. The aim is to evaluate the completeness of follow-up of two methods of follow-up after EMA— self-assessment with low-sensitivity urine pregnancy test result or serial serum βhCG blood tests. Secondary aims will evaluate whether self-assessment follow-up is safe and acceptable to patients and clinicians.

Methods

This is a multicentre randomised controlled trial in New Zealand. Eligible women and pregnant people having EMA will be randomised to self-assessment or blood test follow-up. The primary outcome is ‘lost to follow-up.’ To detect a decrease in ‘lost to follow-up’ rate from baseline of 15% to 7.5%, with 90% power and a two-sided type 1 error of 0.05, the sample size required is 736 participants, in a 1:1 ratio.

Discussion

If self-assessment reduces lost to follow-up, has additional clinical benefits, and is safe, cost-effective, and acceptable to women and clinicians, we anticipate change in EMA follow-up practice around the world. We think self-assessment will be welcomed as part of a patient-centred package of care following EMA.

Trial registration

This trial was prospectively registered with the Australia New Zealand Clinical Trials Registry (ANZCTR) on 21 August 2023, registration number ACTRN12623000890639;

https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=384373

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-025-09016-0.

Introduction

Background and rationale

Abortion was decriminalised in Aotearoa New Zealand (NZ) with the passing of the Abortion Legislation Act 2020 [1]. The Act removed many of the legislative/legal restrictions that had been in place for over 50 years and placed abortion care under the jurisdiction of health care. As countries reduce barriers to accessing abortion, women present earlier. Abortions performed earlier are safer, and presenting early gives women a choice as to abortion method (medications versus surgical procedure).

Early medical abortion

Medical abortion uses one or more medicines alone or in combination to end a pregnancy. The most common medicines are the hormones prostaglandin E1 (misoprostol) and mifepristone. These medicines work by softening the cervix (neck of the womb) and causing the uterus (womb) to contract. They can be given by a health practitioner in an abortion service, or taken by women at home. Failed abortion is an infrequent but important complication of medical abortion, meaning the abortion has not worked and there is an ongoing live pregnancy.

Medical abortion is a safe and effective way to terminate pregnancy in the first trimester without having a surgical procedure [2, 3]. In NZ, EMA is offered up to 10 weeks’ gestation, and a common protocol used is a 200-mg tablet of mifepristone taken orally, followed by 800 mcg of misoprostol taken 24–48 h later.

The proportion of medical to surgical abortion is increasing worldwide, due to one of the medications (mifepristone) being more widely available, and the publication of high-quality research supporting its safety, effectiveness and acceptability [2, 3]. In the last 2 years of the COVID-19 pandemic protections, there has been a further increase, likely due to the fact that EMA does not necessarily require in-person contact between the patient and the health practitioner (as surgical abortion does) [4, 5]. Figure 1 shows this trend in NZ [6].

Fig. 1.

Number of abortions by method in Aotearoa New Zealand, over time

After EMA, follow-up is recommended to check for well-being, detect complications, and offer contraception. It is also important to ensure the procedure worked, and the woman is no longer pregnant. Detecting an ongoing live pregnancy is the most critical clinical outcome of EMA [7].

Published studies of EMA report rates of ongoing viable pregnancy from 0.7% to 2.2% [3–5]. In this situation, women are recommended to have a surgical procedure to terminate the pregnancy. However, if women do not follow-up, they may not initially realise they are still pregnant and present with ongoing pregnancy symptoms weeks to months later. Therefore, the follow-up method needs to be both accurate and acceptable to the patient. Which method of follow-up after EMA is most effective is a critical research gap [8].

In the first EMA studies, women routinely had a follow-up visit with their doctor that included a history, examination and ultrasound scan. This progressed to using quantitative βhCG (human chorionic gonadotrophin pregnancy hormone) blood tests instead of an ultrasound scan, checking that the serum βhCG level dropped as expected between the day of the abortion and 7 days later. This approach is recommended practice in Canada and is supported by observational studies [9]. Half of abortion services in NZ use comparative blood tests for follow-up. However, blood tests are costly and invasive and require lots of health practitioner time to follow-up the results and contact the patient to explain the results.

An informal survey of abortion providers in NZ suggests that 10–20% of people are lost to follow-up, where they do not get the second blood test and cannot be contacted despite several attempts over several weeks. Studies report even higher rates of lost to follow-up [10, 11].

Low sensitivity βhCG urine test

An alternative innovative method of follow-up is the low sensitivity βhCG urine test. In contrast to the commonly used high sensitivity urine pregnancy tests from the chemist (positive pregnancy test at the lowest level of βhCG), the low sensitivity urine test will be positive only at a high level of βhCG. It is usually done 3 weeks after an EMA, and the test will be negative if the EMA was successful and the woman is no longer pregnant.

The low sensitivity urine test performs well. A 2019 metanalysis of four trials of EMA follow-up (5456 women) compared self-assessment using a low sensitivity urine test (with follow-up phone call) to routine clinic in-person follow-up [12]. The authors of the metanalysis found no difference in effectiveness, as measured by ongoing pregnancy (1.0% vs 1.0%), or safety, as measured by clinical adverse events. Importantly, acceptability was significantly greater for self-assessment than for clinic follow-up (82% vs 41%). Moreover, the percentage of women lost to follow-up was slightly lower in the self-assessment group (3.8% vs 6.5%).

Based on the above metanalysis, a large cohort study was designed in response to the Covid pandemic protections in the UK for women to access telemedicine abortion with self-assessment follow-up. The authors found that 96% of 2453 women who completed the survey were ‘satisfied’ or ‘very satisfied’ with their care [4].

Potential benefits of self-assessment follow-up of EMA

• Promotes autonomy

• Inexpensive and not invasive

• Easy to interpret (similar to COVID-19 rapid test and the commonly used pregnancy tests)

• Acceptable

• Fits with growing trend to telehealth

• Fits with growing trend toward self-management of health (e.g. self-monitored sleep and heart rate via smartwatch), prevention (e.g. self-testing for sexually transmitted infections, cervical screening), and disease (e.g. self-checking of blood sugars and blood pressure)

• Less in-person contact time with health practitioners and with the health care system

There are no published trials directly comparing self-assessment low sensitivity urine pregnancy testing with follow-up serum βhCG testing. Given the lack of high-quality evidence, the 2021 NZ Abortion Clinical Guideline was unable to make a recommendation about EMA follow-up and instead has a good practice point to offer all methods, including pelvic ultrasound, blood tests, or urine pregnancy test [8]. The Royal Australian and NZ College of Obstetrics and Gynaecology guideline on mifepristone does not recommend the offer of urine pregnancy tests to follow-up EMA [13].

Objectives

Primary hypothesis

Women with self-assessment follow-up (including urine pregnancy test) after EMA will have more complete follow-up compared to women with follow-up with serial blood tests.

Secondary hypotheses

1. Compared to serial blood test follow-up, women with self-assessment follow-up after EMA will be more likely to have an ongoing live pregnancy detected.

2. Women with self-assessment follow-up will not have more adverse events.

3. Women with self-assessment follow-up will be more satisfied.

4. Health practitioners looking after women with self-assessment follow-up will be more satisfied.

Primary Aim

To assess the effectiveness of self-assessment follow-up compared to serial blood test follow-up on the rate of lost to follow-up (LFU) in women having EMA.

Secondary Aims

To assess other outcomes between the two methods of follow-up:

• Adverse events

• Participant satisfaction and acceptability

• Health practitioner satisfaction and acceptability

Trial design

This is a multi-centre parallel group randomised controlled trial with a 1:1 allocation ratio and a superiority framework.

Methods: participants, interventions and outcomes

Study setting

Participants will be recruited from abortion services in New Zealand. At the time of publication, six services are already recruiting, representing over half of all abortion provision in New Zealand. Another three services are going through locality approval processes. The participating abortion services are listed at www.matewhenua.auckland.ac.nz.

Eligibility criteria

Inclusion criteria:

• Pregnant women having early medical abortion

• Able to provide written informed consent

Exclusion criteria:

• previous participation in Mate Whenua

• Gestational age > 10 + 0 weeks’

• Situation where the health practitioner feels one of the follow-up methods would be contraindicated

Who will take informed consent?

The local research team at each site will regularly screen all EMA bookings in order to identify potentially eligible women to the health practitioner.

At the point an EMA is planned, the health practitioner will discuss the trial with the woman and provide the Patient Information Sheet and Consent Form (CF). She will be given time to consider and discuss with her family as needed. Contact information for the local investigators, and the link to the study website with a 3 min video about the study, will also be provided. Information about the trial may be presented on the same day as the EMA or at a prior appointment. In each situation, the clinician will ensure that there is sufficient time for the participant to consider their decision to participate.

On the day of EMA, the health practitioner or local site researcher will confirm the woman still wishes to participate in the study and answer any final questions. If in-person, the CF will be completed online via REDCap whilst in clinic. If by Telehealth, then a personalised link to the CF in REDCap will be sent via email or text message, and the health practitioner or local site researcher will remain on the phone as the woman completes the CF. The health practitioner or local site researcher will complete a researcher declaration stating they explained the trial in full, addressed all queries/concerns, and that the participant completed the form themselves.

Additional consent provisions for collection and use of participant data and biological specimens

Not applicable; there are no biological specimens or ancillary studies.

Interventions

Explanation for the choice of comparators

Participants will be randomised to one of two follow-up protocols:

1. Self-assessment including urine pregnancy test and notification to research team

2. Comparative βhCG blood tests with contact by health practitioner for results—this is the usual practice of almost all abortion clinics in New Zealand currently, and recommended in the national clinical guideline on abortion care in New Zealand and Canada

Intervention description

1. Self-assessment:

   Participants will receive a low sensitivity urine pregnancy test to take home, with instructions to take the test 3 weeks after the EMA, and what to do if the test is positive or negative. Written and pictorial information will be provided on how to interpret the test result, alongside a list of symptoms by which the woman can assess herself that she is, or is no longer, pregnant. If participating via telehealth, the urine test and supporting information will be sent via courier following randomisation.

   As part of a holistic approach, the research nurse will touch base with participants weekly in order to help support women and their family to manage the process. At four weeks post-EMA, they will specifically ask if they did the urine pregnancy test or not. If they did not, there will be 2 further reminders after which they will be considered ‘lost to follow- up’. If the participant loses the test, a replacement test will be sent to her.

   If the urine pregnancy test at 3 weeks is positive, then the clinician should ask about pain and bleeding, symptoms of infection, and symptoms of ongoing pregnancy. If there is any concern about ectopic (especially if ultrasound confirming intrauterine pregnancy was not done before the EMA), the clinician will follow usual hospital protocol for management of suspected ectopic. If there is no suspicion of ectopic pregnancy, sepsis, or haemorrhage, then they will offer a quantitative serum βhcg. If the result is higher than expected, then they will consult the medical lead for the EMA service to decide on further management (examination in person, repeat serum βhcg and/or ultrasound scan).

2. Comparative βhCG blood tests:

   Participants will have a quantitative serum βhCG on the day of the mifepristone at the usual lab associated with the abortion service. Participants will be given a paper or electronic form to do a second quantitative serum βhCG 7 days later at their nearest lab. The health practitioner will contact the participant after the results of the second test are available. If the second test is not done, there will be 3 reminders by phone, text message or email, after which they will be considered ‘lost to follow-up.’ If participating via telehealth, the lab forms will be emailed/couriered following randomisation.

Criteria for discontinuing or modifying allocated interventions

Participants can choose to withdraw from the study at any time. Data collected up to that point will be retained and used as part of the intent to treat analysis.

At the time of withdrawal, the local lead investigator is encouraged to ask the participant if they would be willing to continue participating in a more limited way. For example, they may no longer wish to continue with their allocated method of follow-up, but they may be willing to have their data collected until they are discharged from the abortion service. This limited level of participation will be recorded in the appropriate REDCap Form.

Strategies to improve adherence to interventions

The research nurse will contact participants every week to remind them about the follow-up test.

Relevant concomitant care permitted or prohibited during the trial

Resources about mental health support and contraception will be provided by the abortion service at the time of EMA as part of usual clinical care. Clinical care to participants may be provided by multiple clinicians throughout their EMA. It is up to the clinician if they feel it is necessary to offer additional investigations, phone calls or visits to ensure safe clinical care, and/or if requested by the participant. Local guidelines can be followed if any uncertainty about whether or not the pregnancy has ended based on the allocated follow-up.

Provisions for post-trial care

There are no provisions for ancillary or post-trial care, nor for compensation to those who suffer harm from trial participation. We do not expect there to be any harm from trial participation. Ongoing live pregnancy and ectopic pregnancy are recognised adverse outcomes of EMA irrespective of the method of follow-up.

Outcomes

Primary outcome: lost to follow-up

Defined as unable to discuss the results of the follow-up pregnancy test, after three attempts in total to contact the participant after the time frame recommended for the follow-up test to be performed. This binary outcome (yes/no) will be measured at 6 weeks after EMA. The proportion of participants who are lost to follow-up will be compared between the two study groups. The outcome of ‘lost to follow-up’ is relevant because it is associated with undetected live ongoing pregnancy (failed abortion).

Secondary outcomes

1. Contraception provided (binary outcome yes/no)

   1. If yes, categorised by type (pill, depo, IUD, condoms, refer permanent, none, other)
2. Timing of follow-up test

   1. If serial blood test, then categorised as < 7 days, 7 days, 8–14 days, > 14 days
   2. If UPT, then categorised as 1–2 weeks, 3 weeks, later than 3 weeks

3. Successful abortion, defined as termination of pregnancy without need for surgery (binary outcome yes/no)

4. Incomplete abortion/retained pregnancy tissue/retained products of conception (binary outcome yes/no)

5. Haemorrhage, defined as 500mL or more (binary outcome yes/no)

   1. If yes, requiring red blood cell transfusion (binary outcome yes/no)

6. Uterine infection, defined clinically as fever, tachycardia, tender on exam, or purulent vaginal discharge, and received broad spectrum IV antibiotics (binary outcome yes/no)

7. Uterine rupture, defined as clinically significant rupture involving the full thickness of the uterine wall and requiring surgical repair (binary outcome yes/no)

8. Additional investigations (binary outcome yes/no)

   1. If yes, number of investigations (compared using median)
   2. If yes, categorised by type (blood test, pelvic ultrasound scan, other)
9. Additional community prescription dispensed (binary outcome yes/no)

   1. If yes, number of prescriptions (compared using median)
   2. If yes, categorised by type (antibiotics, painkillers, other)
10. Additional health care visit (binary outcome yes/no)

    1. If yes, number of visits (compared using median)
    2. If yes, categorised as type (abortion service, hospital emergency department, mental health service)
11. Surgical intervention (binary outcome yes/no)

    1. If yes, categorised by type (laparotomy, laparoscopy, ERPOC/D&C, EUA, hysteroscopy)
12. Hospitalisation (binary outcome yes/no)

    1. If yes, length of stay (number of hours, compared using median and mean)
    2. If yes, primary reason (free text box)
13. Admission to intensive care unit or equivalent (binary outcome yes/no)

    1. If yes, primary reason (free text box)

14. Death (binary outcome yes/no)

15. Adverse event e.g. ectopic pregnancy (binary outcome yes/no)

Participant, outcomes measured at 4–6 weeks after EMA (online survey).

1. Satisfaction/experience

   1. Satisfaction (scale of 1 to 10, categorised as satisfied, neutral, unsatisfied, and also compared as an ordinal variable as a mean)
   2. Acceptability (scale of 1 to 10, categorised as acceptable, neutral, unacceptable, and also compared as an ordinal variable as a mean)

Participant, measured at 8 months after EMA (from medical records):

1. Ongoing viable pregnancy (binary outcome yes/no)

   1. If yes, gestational age (measured in completed gestational weeks) at detection
   2. If yes, outcome of pregnancy (abortion, ectopic, miscarriage, stillbirth, live birth)

Participant, outcomes measured at 12 months after EMA (online survey):

1. Contraception use (binary outcome yes/no)

   • ◦ If yes, same as one year ago (binary outcome yes/no)
   • ◦ If no, reason discontinued (free text box)
2. Pregnancy (binary outcome yes/no)

   • ◦ If yes, intended or unintended
   • ◦ If yes, categorise by outcome (miscarriage, abortion, birth, other)

     • if abortion, categorise by type (medical or surgical)
3. Seen the GP, nurse, midwife (binary outcome yes/no)

   • ◦ If yes, main reason (women’s health, mental health, whānau health, relationship health, other)

Abortion services staff, outcomes measured at 6 months from start of recruitment (online survey).

1. Satisfaction/experience

   1. Satisfaction (scale of 1 to 5, categorised as satisfied, neutral, unsatisfied, and also compared as an ordinal variable as a mean)
   2. Stress (scale of 1 to 5, categorised as more stress, neutral, less stress, and also compared as an ordinal variable as a mean)
   3. Workload (scale of 1 to 5, categorised as more work, neutral, less work, and also compared as an ordinal variable as a mean)

Participant timeline (SPIRIT figure)

	Enrolment	Allocation	Study period
Timepoint	Up to 7 days in advance	Day of EMA: Day 0	5–7 days	3 weeks	4–6 weeks	12 months	6 months after study start
Enrolment:
Eligibility screen	X
Informed consent	X
Eligibility confirmation	X
Baseline data collection		X
Randomisation		X
Follow-up:
Self-assessment (urine test at home)				X
Comparative blood test 1		X
Comparative blood test 2			X
Assessments:
Results of pregnancy tests:			X	X
Follow-up questionnaire 1					X
Follow-up questionnaire 2						X
Staff satisfaction questionnaire							X

Sample size

At the largest abortion service in NZ (EDU), where standard follow-up is comparative blood tests, an audit reported a ‘lost to follow-up’ rate of 10% (unpublished data). In one trial of EMA in Massachusetts, comparing blood tests to ultrasound, the ‘lost to follow-up’ rate was 12% in the blood test group [11]. An informal 2022 survey of all 19 abortion care providers, through the Abortion Providers Group of Aotearoa New Zealand (www.apganz.org.nz), found that half used comparative blood tests as their preferred follow-up method, and of these, on average, the estimated ‘lost to follow-up’ rate was 15% (unpublished data). Based on informal consultation with local and Australian abortion providers (accessed through APGANZ), we felt that a 50% relative risk reduction would be clinically meaningful. To detect a decrease in ‘lost to follow-up’ rate from 15% to 7.5%, with 90% power and a two-sided type 1 error of 0.05, the sample size required would be 368 participants for each study group (total 736).

This sample size is also sufficient to evaluate one of the secondary outcomes that of detecting ongoing live pregnancy, which is an important safety check. To show no difference in detection of ongoing pregnancy of 1.5% in both groups (the middle of the range of published trials), for 90% power, one-sided significance level = 2.5%, non-inferiority margin = 5% between the two groups, and inflating the sample size for 20% lost to follow-up, the sample size required would be 156 participants for each group (313 in total).

Recruitment

Six abortion clinics are already recruiting into the study, representing over 50% of all abortion provision in New Zealand. Three more clinics are undergoing locality approvals to increase the recruitment rate. We have paid nurse research assistants at each clinic helping with recruitment, monthly meetings of the steering committee and local research teams, newsletters, website, recruitment video and local celebrations of recruitment milestones.

We are tracking how many eligible women are being approached to participate, and up to this point, we are meeting our target of > 80%.

Assignment of interventions: allocation

Sequence generation

The randomisation schedule was prepared by the Trial statistician, National Institute for Health Innovation, University of Auckland. Randomisation has no restrictions. Allocation is on a 1:1 ratio. Participants are randomised to one of two study groups using stratified block randomisation and stratified by centre.

Concealment mechanism

Centralised online randomisation occurs using an independent randomisation tool within the trial’s REDCap database.

Implementation

After obtaining written consent from the patient, the health practitioner or local researcher enrols the participant, performs the web-based randomisation, and assigns the participant to the interventions.

Assignment of interventions: blinding

Who will be blinded

Participants, clinicians, outcome assessors and data analysts are not blinded to treatment allocation; however, the outcomes are objective, and most data are routinely collected for clinical care.

Procedure for unblinding if needed

Not applicable—there is no blinding in the trial.

Data collection and management

Plans for assessment and collection of outcomes

On the day of the EMA, the health practitioner informs the participant of her study group and asks her to complete the Enrolment Survey, which includes questions on demographics and contact details. The health practitioner collects baseline data. Demographic and clinical data include: age (from DOB), gestational age (calculated by health practitioner), ethnicity (self-reported by patient and then prioritised by research team as per MoH protocol), and New Zealand Index of Deprivation (NZDep) (from postcode and then checked by research team).

Local lead investigators or abortion service health practitioners will collect data on participants at four time points. The first is at time of EMA; the second is at time of completed follow-up (6 weeks after EMA); the third is 8 months after EMA; and the fourth is 12 months after EMA.

Some questions will be asked directly to women. Most data will be collected contemporaneously by the health practitioners looking after the woman; most of these data are collected already for usual clinical care or formal legal notification to the Ministry of Health. Some data may need to be collected retrospectively from the medical records if study forms are incomplete. Data are input directly into the REDCap online database; there are no paper data collection forms.

A participant questionnaire will be carried out at 4 weeks after the EMA. It will be done by email direct to participants through REDCap, followed by two automatic email reminders at 5 and 6 weeks if needed. The research team can also ring or text participants as a reminder and to ensure the email address is correct. If on their enrolment form the participant has indicated a different preference for receiving the questionnaire, then it is the responsibility of the local investigators to arrange the survey by telephone or print/post with a stamped return envelope.

An anonymous health practitioner survey will be carried out at 6 months from the study starting at each participating centre. The Study Coordinator will send a REDCap link to the survey to the local research team, who will then circulate internally.

All researchers have had one-on-one formal training in the use of REDCap and have received their own user logon/password. All researchers have had Good Clinical Practice training. Data collection forms can be provided on request.

Plans to promote participant retention and complete follow-up

Participants will be contacted regularly after their EMA during their follow-up by the nurse research assistant. We anticipate complete follow-up of the trial’s main outcomes given the short time frame from the time of recruitment to the time of completion (maximum 4–6 weeks). The research assistants will use participants’ hospital medical records system to update contact details as needed for the 12-month questionnaire.

Whilst this study protocol will implement additional strategies to maximise participant retention that could bias the primary outcome, this study will standardise follow-up procedures across both study arms to avoid influencing differential loss to follow-up to mitigate any impacts on the primary outcome.

Data management

Data are entered into REDCap by the Research Assistants at each research site. The REDCap database was developed with range checks for data values and has been modified during the study as needed. REDCap data are checked and cleaned on a regular basis by the Trial Manager and Trial Statistician. Reports are run on a 6-monthly basis for review by the Data and Safety Monitoring Committee. Trial data will be extracted from REDCap and stored electronically on a secure password-protected University of Auckland server for 10 years. Data management procedures can be accessed as a Supplementary File.

Confidentiality

Participants’ privacy and confidentiality will be respected through the protection of their data. The Investigator will comply with all legal and regulatory requirements regarding the privacy and confidentiality of participants’ data.

Data will be entered into REDCap by the local research team using the participant’s unique Study ID Number. REDCap is a secure web application for building and managing online surveys and databases. Identifiable data will be entered into a contact details form in REDCap and kept separately from the rest of the case report forms to ensure confidentiality. Data will be analysed using de-identified data and findings reported in aggregate.

Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use

Not applicable—no biological specimens collected.

Statistical methods

Statistical methods for primary and secondary outcomes

Data analyses are specified a priori in the statistical analysis plan prepared by the trial statistician (and agreed by all members of the TSC). The data will be extracted into SAS version 9.4 (SAS Institute Inc. Cary NC) for analysis. A p-value of 0.05 will be considered statistically significant. Baseline demographic and clinical characteristics of each study group will be described. Analyses will follow the principle of intention-to-treat, where participants will be analysed according to the assigned treatment group at randomisation. Chi-squared tests, relative risks and absolute risk differences (with 95% CIs) will be calculated for the binary outcomes. The main analyses for the primary outcome will be unadjusted, and adjusted sensitivity analyses will be conducted using log-binomial regression and will adjust for the stratification factor (center) and other variables if required (for example, ethnicity).

Non-inferiority for the ongoing live pregnancy outcome will be evaluated by observing whether the lower bound of the two-sided 95% confidence intervals for the risk difference between the two groups is above the non-inferiority limit of − 5. If non-inferiority is evident, assessment as to whether the intervention group has effectiveness superior to that of standard care will be carried out using the same approach but comparing to a zero difference.

Interim analyses

There is no planned interim analysis and there are no stopping guidelines. This was a decision agreed on by the Trial Steering Committee and the Data Safety Management Committee (DSMC) before the trial started.

Methods for additional analyses (e.g. subgroup analyses)

There is no planned subgroup analysis.

Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data

Per-protocol sensitivity analyses will be conducted for the primary outcome, excluding participants who had major protocol violations. Since the primary outcome is lost to follow-up, there should be no missing data. For secondary outcomes (unless otherwise stated), participants with missing outcome data will be excluded from the analyses.

Plans to give access to the full protocol, participant-level data and statistical code

Full protocol, participant-level data and statistical code can be provided by the principal investigator on request.

Oversight and monitoring

Composition of the coordinating centre and trial steering committee

The Trial Steering Committee (TSC) has overall responsibility for, and provides high-level academic oversight of, the entire study. Terms of Reference are available on request. The TSC is comprised of the principal investigator (PI) who is a gynaecologist providing abortion care and Associate Professor at the University of Auckland; the co-investigator, who is a sexual health physician and academic researcher at the University of Victoria; two abortion care providers (one nurse and one doctor); a Māori midwife and researcher; a GP who is on the Executive of the Abortion Providers Group of Aotearoa New Zealand; and a Professor in women’s health research at the University of Victoria.

The Trial Manager manages the day-to-day organisation of the trial, with the support of the PI.

The data management team trains new researchers in REDCap, manages the dataset (data checking and cleaning) and prepares the reports for the Data Safety Monitoring Committee (DSMC).

Local research teams have been appointed at all participating sites, comprising the paid research assistant and one or more health practitioners and managers from the site.

Members of the TSC, the local research teams and the trial statistician meet monthly by videoconference. We also have annual planning in-person workshops. Meetings are organised and minuted by the Trial Manager.

Composition of the data monitoring committee, its role and reporting structure

The DSMC is responsible for safeguarding the interests of trial participants by assessing the safety of the follow-up methods during the trial and monitoring the overall progress and conduct of the trial. The DSMC is comprised of the Chair (K Antony) who is an Associate Professor in women’s health at University of Auckland, a gynaecologist providing abortion care, and a Māori obstetrician.

The DSMC membership has been restricted to individuals independent of the trial and who are free of apparent significant conflicts of interest. The DSMC members should not serve on DSMCs of similar concurrently active trials. DMSC members must act in accordance with the HRC Policy on Conflict of Interest.

The DSMC functions independently of all other individuals and bodies associated with the conduct of the trial. The DSMC has an advisory role; while all DSMC recommendations will be considered by the TSC, the TSC has final responsibility for the conduct and completion of the trial. Progress reports are prepared 6-monthly by the Trial Statistician and provided to the DSMC for review.

Terms of Reference can be provided on request from the PI.

Adverse event reporting and harms

Data are systematically collected by local investigators on pre-specified adverse events at 4–6 weeks and 8 months after the EMA. AEs include: live ongoing pregnancy past the first trimester and ectopic pregnancy. AEs are reported by the local investigators to the PI within 24 h of finding out about their occurrence, reviewed, and then reported to the Chair of the DSMC within 48 h. The DSMC will commence review of the AE within 72 h of receiving the AE report from the PI and will complete review and report to the TSC within the next 72 h. The DSMC will determine for each AE how likely it was that the trial intervention was a causative factor in the AE.

AEs are included in the 6-monthly reports created by the Trial Statistician. AEs will be reported in the publication of trial results.

We will also review the Comments free-text boxes for each participant to determine if any other adverse events or harms were spontaneously reported.

Frequency and plans for auditing trial conduct

The Trial Statistician reports monthly to the Trial Steering Committee and local investigators about recruitment numbers by site and in total.

The DSMC will meet every 6 months to audit trial conduct, and minutes will be taken. All meetings will be closed meetings, open only to DSMC members. Additional open meetings between the DSMC and TSC and/or PI may be arranged as required. DSMC reports containing data by treatment group, whether masked or not, will be reviewed only in closed meetings. The DSMC will provide a written report to the TSC within 2 weeks of conducting this meeting.

Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees)

Important protocol modifications will be shared with local investigators at monthly meetings and included in the Minutes of those meetings. The Registry will be updated by the PI. The Ethics Committee has a published list of minor and major amendments; if an amendment is minor, then it will be included in the annual progress report, and if it is major, then a formal written amendment will be submitted. We do not have planned communications with trial participants.

Dissemination plans

We plan to communicate trial results to health care professionals by publishing our findings in a peer-reviewed international scientific journal aimed at Obstetrician Gynaecologists, sexual and reproductive health specialists and abortion providers. Authorship will be as per ICMJE guidelines. The PI plans to present about trial progress at the annual education day of the Abortion Providers Group of Aotearoa New Zealand, at other relevant local and national conferences, and to the New Zealand Ministry of Health and Health New Zealand abortion teams. The PI plans to work with the University of Auckland communications team to disseminate a media release to help share results with the public. Findings will be published on a public website www.matewhenua.auckland.ac.nz. We plan to communicate trial findings to participants using their contact details from REDCap, in the form of an infographic and a link to the website. Participant level de-identified data can be shared in future on request to the PI.

Discussion

This study has the potential to significantly change EMA follow-up practice in NZ and around the world. Paramount to all other outcomes from this study is that a simple intervention, such as a switch to the use of self-assessment with a urine pregnancy test as the default method of follow-up, may improve completeness of follow-up and reduce the risk of undetected live ongoing pregnancy. The existing high-quality evidence on this research question is minimal. For the intervention of self-assessment, although there are safety and effectiveness data in large international cohort studies, the only randomised trials are small and compare in-person clinic follow-up with ultrasound scan, which is no longer standard of care in New Zealand or most other developed countries. For the intervention of serial blood tests, there are no published trials at all; rather, it is based on observational data of the rate of serum beta hcg dropping in a population of women having EMA. This trial answers an important contemporary research question evaluating methods used currently around the world.

We hypothesise that women choosing EMA are more likely to prefer self-assessment and find it more acceptable to follow up at home rather than going to a lab for a blood test and having to wait to hear the results from a health practitioner. We also hypothesise that health practitioners will find it more acceptable not to have to follow up with patients multiple times to get the blood test done and explain the result. This has the potential to save time and resources on the part of the abortion service and may be a cost-effective strategy.

Trial status

First participant recruited on 27 October 2023. Anticipated recruitment completion in July 2025. Current protocol: Version 11, dated 11 December 2023.

Abbreviations

ANZCTR

Australia New Zealand Clinical Trials Registry

βhCG

Human Chorionic Gonadotrophin pregnancy hormone

CF

Consent Form

DSMC

Data Safety Monitoring Committee

EMA

Early medical abortion

HDEC

Health and Disability Ethics Committee

HRC

Health Research Council

LFU

Lost to follow-up

PI

Principal investigator

PIS

Patient Information Sheet

TSC

Trial Steering Committee

NZ

New Zealand

Authors' contributions

MRW is the principal investigator; she conceived the study, led the proposal and protocol development, wrote the first draft and subsequent revisions. EJM, ST, AK and BL contributed to study design and to development of the proposal and protocol. VP was the lead trial methodologist and contributed to development of the proposal and protocol. SS, AOB and SA contributed to the development of the protocol. All authors read and approved the final manuscript.

Funding

Funding was granted by the Health Research Council (HRC) of New Zealand (22/713/A) in December 2022. The funding body has no role in the design of the study, data collection, analysis, interpretation of data, or in writing the manuscript.

Data availability

The trial statistician, trial manager and PI will have access to the final trial dataset. There are no contractual agreements that limit access for investigators.

Ethics approval and consent to participate

This trial received ethics approval through the Health and Disability Ethics Committee (HDEC) New Zealand full review pathway (Northern B Health and Disability Ethics Committee) granted on 15 June 2023 (Reference: 2023 FULL 13189). Protocol amendment was granted on 15 January 2024. Progress reports are provided regularly to HDEC. Written, informed consent to participate will be obtained from all participants.

Consent for publication

The consent form given to participants can be provided on request to the PI.

Competing interests

AK reports a conflict of interest as a director and shareholder of Istar Ltd. (registered charity CC20298). Istar Ltd. imports both Mifegyne (mifepristone) and CheckToP (low sensitivity urine pregnancy test). AK receives a nominal annual director’s fee from Istar Ltd. but does not receive any other dividends from the charity.

The remaining investigators declare no conflict of financial or other competing interest.