We have used and adhered to the revised standards for quality improvement reporting excellence (SQUIRE 2.0) to report this initiative .
Given the major shortage of midwives in Liberia, the first aim of this initiative, was to assess the feasibility of educating women in labour to assess their FHRs with a portable doppler monitor and to alert the midwife if they detected changes in the FHR which might indicate fetal distress.
The second aim of this initiative was to facilitate system changes to educate and enable the attending midwife, who would likely be caring for several individual women in labour, to initiate an immediate remedial course of action should she/he be alerted by a woman in labour who detected changes in her FHR. The questions posed, and the measured outcomes of this intervention, are shown in Table 2.
In Liberia, over the past 6 years, a partnership between the Ministry of Health and Social Welfare (MOHSW), The World Health Organization (WHO), The United Nations Population Fund (UNFPA), the Liberian Board for Nursing and Midwifery (LBNM) and the registered charity (not-for profit organization), Maternal and Childhealth Advocacy International (known as MCAI), has been established to train senior birth attendants (19 midwives and 2 physician assistants) in advanced obstetrics to become obstetric clinicians, able to undertake emergency clinical procedures to expedite the delivery of a distressed fetus, such as vacuum delivery and Caesarean section . Currently, ten obstetric clinicians have completed training, 10 more are currently in their final year of training and 9 more have just been appointed following an examination and interview.
More recently, the partnership initiated a new programme to improve neonatal resuscitation and hospital neonatal care by training selected nurses and midwives to become advanced neonatal nurse practitioners (neonatal clinicians) and is now based in both of the two hospitals involved. Currently three have completed their training, 6 are in their final year and 8 have just been appointed following an examination and interview.
The fetal monitoring initiative took place in CB Dunbar Hospital in Bong County and CH Rennie Hospital. In Margibi County. The main training hospital for both obstetric clinicians and advanced neonatal nurse practitioners (neonatal clinicians) is CB Dunbar Maternity Hospital. During this initiative and working in these two hospitals, there was one fully qualified and licensed obstetric clinician, 4 trainee obstetric clinicians and 9 trainee neonatal clinician practitioners at CB Dunbar Hospital. There were 5 trainee obstetric clinicians at CH Rennie Hospital but no neonatal clinicians at the time of this feasibility study.
Management committee for this health care improvement initiative
This initiative represented an integrated project between obstetrics and neonatology. Both qualified and trainee obstetric clinicians and trainee neonatal clinicians and their trainers were involved.
The management committee oversaw the day to day running of the initiative and was responsible for the data analysis. If there were any problems identified, in particular involving the need for improving the recording methods, the management committee would make the necessary changes to the initiative as appropriate.
As members of the management committee, two trainee Obstetric Clinicians (KP and NJ), and the international advanced neonatal nurse practitioner (AK) managing the recently established neonatal intensive care unit, led the day to day running of the fetal monitoring component of this initiative.
The fetal monitoring project started as a service delivery intervention; when authors realized the possible significance of the outcomes, we decided to approach the work from a research angle for publication. As a programmatic intervention, it was fully within the domain of the Ministry of Health to implement. Re-framed as a research project, it was relevant to seek National Research Ethics Board approval.
Because of poverty and complex social circumstances, it is common for mothers in Liberia to be aged under 16 years. Also, young mothers frequently attend the hospitals in labour by themselves, with no legally assigned family members in attendance and sometimes only a traditional birth attendant for support. Many women and adolescent girls do not know their age, and the illiteracy rate in Liberia is high (63%).
Over a 15-month period (from 31 July 2017 to 24 October 2018), 474 women admitted in the active first stage or second stages of labour, who were not experiencing any complications such as haemorrhage, severe pre-eclampsia or obstructed labour, were invited to participate. Before participation, each mother was asked to give her informed consent by a trainee or fully trained obstetric clinician. Initially (given a female literacy rate from UNICEF data 2011-2016  of those aged 15 to 24 years of approximately 37%) this consent was requested verbally and recorded and later (after 27th April 2018) a dedicated consent form was signed (either by fingerprint or in writing).
From the start, all authors considered this work as the implementation of a new project that would improve health service delivery, in particular regarding fetal monitoring during labour. The severe lack of midwives and the infrequency of adequate completion of the partograph, indicated to the authors that this project was an essential part of health service delivery rather than a research study. The project had the prior approval of the Liberian Ministry of Health. As the focus was improved health service delivery and did not involve any invasive procedures, authors initially sought verbal consent from all potential participants. However, as the initial results of the project were so encouraging, authors recognised the value in sharing our experience of this project through a peer-reviewed publication and so changed the consent process to require written consent from all participants. Informed consent, written or verbal, was therefore obtained from all participants.
Throughout, this non-invasive project was supervised by the Liberian Ministry of Health (the original Minister of Health, BD, up to January 2018 and current Minister, WJ, from January 2018 are both authors of this paper).
The seeking of consent, whether verbal or written, was dependent on each woman understanding what was being requested of her. The consent form (part of the Additional File 1) was read out and discussed with each woman before she was enrolled in the study. Only women or adolescent girls (those potential participants aged under 18 years) who fully understood what they were being asked to undertake were recruited. The obstetric clinicians who gained consent were skilled health workers with years of experience in managing pregnant women in Liberian public hospitals. All had undergone a medical ethics and professional standards course as part of their training. Therefore, these health workers were well placed to ascertain if each woman who was invited to participate understood what was required. Each woman could change her mind at any time, and her wishes were always respected.
Plan of investigation
Between contractions, each participating woman was shown and educated by an obstetric clinician in how to use a re-chargeable, battery operated, fetal doppler heart sound monitoring device (a sonicaid), including the best position to place the probe on the mother’s abdomen. This training usually lasted between 10 and 15 minutes. The mother was also educated in what was a normal heart rate and what was slow or fast, by tapping out a rhythm. Mothers were asked to monitor for approximately a 60 second period immediately following the end of every contraction. However, it wasn’t, in the authors’ opinion, of crucial importance or appropriate to insist that participating women monitor for exactly 60 seconds, which may be difficult. In addition, it is/was the beginning of the 60 second period that we considered of most clinical relevance to detecting changes in the FHR that might indicate fetal distress.
Having determined what was a normal rate, the mother was then asked to identify and immediately inform a midwife of any changes in the FHR that she detected. In the ward areas in the hospitals undertaking this initiative, there is no buzzer or bell alert and so if a participating woman thought that she had detected a change in FHR, she called out for a midwife or obstetric clinician to come and confirm possible changes. As the labour wards in both hospitals were small, this system worked well. Sometimes, the mothers reported actual heart rates (if literate) but mostly identified a fall or increase in heart rates without counting the numbers involved. The midwives and obstetric clinicians who responded to the mother, recorded and reported the actual heart rates. Additional File 1 contains the form that was used for the recording process which was completed as soon as possible after delivery.
Additional file2 is a Table containing the comments made on the monitoring process by the mother, either written directly or transcribed for illiterate mothers by the obstetric clinician completing the form. To help with anonymity, age groups for each participant have been used instead of actual ages.
In all participating women, the partograph was supposed to be completed according to WHO recommendations by a midwife on the labour ward every 30 minutes during the first stage and every 5 minutes during the second stage of labour [20,21]. However, in reality, the fetal section of the partograph was inconsistent and rarely completed due to time pressures on the small number of midwives available.
If a participating woman decided that she could not continue to monitor the FHR, for whatever reason, her wishes were respected and documented. If a woman's condition during labour made it difficult for her to continue making recordings, a midwife or obstetric clinician would, if possible, and if time allowed, assist or take over the monitoring from her with the woman’s permission.
If a woman declined to undertake the monitoring herself but still wanted her fetus to be monitored, with the mother’s permission, the monitoring could be done by a midwife or by an obstetric clinician, provided they had time available given their heavy workloads.
Actions by midwives and/or obstetric clinicians
The midwives working in the two hospitals were informed of the plans and scope of the initiative by the management committee team and educated in their role and responsibilities regarding the initiative, especially that they must, if possible, respond immediately when alerted by the participating women to a change in FHR and to examine the woman. If the FHR was potentially of concern, the midwife would then immediately notify the obstetric clinician and/or doctor on duty who would assist the midwife in the management of labour until the FHR had either recovered or the baby was safely delivered (as per the birth asphyxia prevention protocol: see below). This remedial protocol was already in place on the maternity units should potential fetal distress be detected. However, unfortunately, FHR monitoring as part of the partograph was rarely undertaken prior to this initiative for reasons outlined in the earlier “Background” to this project. At CB Dunbar Hospital, the neonatal clinician on duty would also be asked to be present at the delivery to resuscitate the neonate if necessary.
The Birth Asphyxia Prevention Protocol
- if a participating woman in labour detected a possible FHR change (bradycardia or tachycardia) she immediately notified an available midwife.
- The midwife checked the FHR and the mother’s condition.
- If a bradycardia (<120 beats/minute) or tachycardia (> 160 beats/minute) were present, the midwife immediately notified the obstetric clinician or doctor on duty.
- If the FHR was normal, the midwife immediately began a period of continuous fetal monitoring including the time up to the next contraction. She / he also listened to the FHR during the whole of the next contraction and immediately following that contraction.
- If a suspicious FHR was detected during the one minute following the next contraction, the midwife took the following actions:
(i) ensured the mother was not lying flat on her back by providing left lateral tilt, examined the liquor if membranes had ruptured, and noted whether meconium staining of amniotic fluid was present.
(ii) ensured that the obstetric clinician and/or doctor on duty was present
(iii) provided additional inspired oxygen (when available), secured an intravenous cannula and gave a bolus of either 0.9% saline or Ringer Lactate solution and where there was suspicion of ketosis, added a bolus of 50% dextrose to the IV infusion 
6. If there was evidence of fetal distress (late decelerations, persistent bradycardia, persistent tachycardia, meconium stained liquor) then the mother was assessed by the obstetric clinician or doctor to manage any maternal obstetric problem that could be responsible for the fetal bradycardia/tachycardia and assess for urgent immediate delivery as follows:
(i) If the cervix was fully dilated and there were no contraindications, a vacuum delivery was undertaken.
(ii) if the cervix was not fully dilated, then the woman was prepared for an emergency Caesarean section (CS) and taken to the operating theatre where she was re-examined. If the fetus was still alive and the cervix still not fully dilated, a CS was performed. An abdominal ultrasound scan was often helpful at this time, not only to check whether the fetus was still alive, but also related to other clinically relevant issues, such as to determine the lie and position of the fetus and confirm the position of the placenta.
7) Under either circumstance outlined in 6 above, a neonatal practitioner or midwife experienced in neonatal resuscitation, was immediately made available for when the baby is delivered. She/he would ensure that the equipment needed for resuscitation was available and functioning.
The outcome for the mother and baby.
The clinical condition of the baby at birth, the need for neonatal resuscitation, admission to the neonatal unit and any signs of subsequent birth asphyxia (also known as Hypoxic Ischaemic Encephalopathy: HIE) were documented. A clinical summary of the pregnancy and delivery was also documented.
The views of the mothers on the fetal monitoring process.
From 23rd October 2017, through written or verbal comments (the latter transcribed by the midwife or obstetric clinician), mothers were asked for their views on the monitoring process.
153 out of 154 participants who had consented at CH Rennie hospital (where the project started on 21st March 2018) were approached for comment (on admission one potential participant’s fetus was found to be dead during her training in monitoring).
This request for comments was not in place for the 33 of the consented mothers, who had enrolled before 23rd September 2017 at CB Dunbar Hospital.
Two authors (DS and RM), categorized the comments retrospectively by first agreeing the common themes after reading all of the comments provided by the participating women. Any discrepancies between DS and RM were discussed and agreed by consensus.
The comments in Table 3 were selected as being representative of the basic themes, as subjectively agreed by the two authors RM and DS. The full set of comments from all participants are available in the Supplementary Information (Additional file 2).
MCAI was responsible for the data analysis and undertook a descriptive analysis.
The lead obstetric clinician in each hospital (KP and NJ) was responsible for charging the batteries and for ensuring gel, consent, and monitoring forms were constantly available. Forms were scanned and sent to the UK by MCAI programme and finance managers in Liberia.