“If there is no data, how do we improve?” Exploring health workers’ perspective on stillbirth recording in the Ashanti Region of Ghana

Abstract

Background

Health workers play a critical role in documenting the estimated 2 million stillbirths that occur annually. From the moment a stillbirth occurs, a health worker is responsible for recording the birth outcome. The reliability of stillbirth data for informing global and national-level strategies on stillbirths depends on the information recorded by the health worker at the point of care. This study aimed to gain insights into the health worker practices and challenges related to stillbirth recording and reporting.

Methods

The qualitative study explored three objectives using an a priori framework: 1) experiences, perceptions, and attitudes; 2) barriers; and 3) support mechanisms available to health workers for stillbirth recording and reporting. Semi-structured interviews were conducted with 28 health workers, including midwives, medical officers, physician assistants, and health information officers. The study was conducted across four secondary and four primary care facilities in the Ashanti Region of Ghana. All health facilities are government owned. Thematic analysis was performed.

Results

Under experiences, perceptions, and attitudes, inconsistent definitions were used to describe stillbirths. Health workers described stillbirths using various gestational age thresholds, including 24-,28-,36-, and 38-weeks. Some health workers did not reference gestational age when describing stillbirths. Pre-service education shaped knowledge on stillbirths and its recording, with limited opportunities for in-service training. The motivation to record stillbirths was influenced by both intrinsic, driven by the moral imperative to do what is right, and extrinsic factors, influenced by district-level standards. Misclassifications and omissions of stillbirths occurred due to a higher workload, a large volume of forms requiring completion, limited knowledge and experience, and a deliberate effort to minimize facility mortality rates, especially in cases of macerated stillbirths. For barriers to stillbirth recording, midwives reported that blame was evident at three levels: blame from the broader health system, blame within the organizational facility-level, and individual-level blame. The failure to implement audit recommendations was identified as a bottleneck perpetuating negative attitudes toward collecting stillbirth data. The engagement of clinical staff in audit reviews and training was identified as support available to health workers.

Conclusion

We need to understand the health worker experiences, perceptions, and attitudes that underpin stillbirth data to reduce the stillbirth burden. The study suggests several recommendations, including socializing the national stillbirth definition, and reviewing audit protocols.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-026-08651-y.

Background

In the last twenty years, progress has been made in maternal and child health outcomes. A 2.9% reduction in neonatal mortality was observed, and maternal mortality decreased by 2.9% between 2000 and 2017 [1]. Despite these improvements, neonatal and maternal mortality remain significant public health concerns. Stillbirths, a related dimension of the maternal and child health life course, is a global public health issue, with nearly 2 million babies stillborn after 28 weeks of gestation each year. Several low-and-middle income countries including Ghana are at risk of missing the global target of 12 stillbirths per 1 000 total births by 2030. Currently, Ghana’s stillbirth rate is 12.3 per 1 000 total births [2]. To accelerate efforts in ending preventable stillbirths, a key strategy identified by the World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF) is to improve health worker skills on recording and reporting of fetal deaths [1].

Health workers are the foundation of high-quality data. Their critical role involves accurately collecting and reporting stillbirth data at the point of care, which in turn informs both national and global actions. This reliable data is critical for addressing the implementation and knowledge gaps necessary to end preventable maternal and neonatal deaths.

From the moment a stillbirth occurs, a health worker is tasked with recording the birth outcome. However, gaps persist in health workers' ability to properly document stillbirths [3–5].These challenges often stem from insufficient knowledge, limited training, and a lack of a supportive work environment [6]. Additional barriers include fear of accountability, lack of understanding about the importance of recording stillbirths, and a culture of blame [7–9]. Furthermore, the absence of data review mechanisms such as perinatal audits, along with high workloads, can impact the accuracy of stillbirth measurement.

Even when stillbirth data is collected, issues such as non-standard definitions, omissions, under-reporting, and misclassification often compromise the data’s quality [8–11]. For instance, a recent study found that in 62 countries, accounting for 29% of all stillbirths in 2019, no high-quality stillbirth data was available. The study called urgently for improvements in the recording of stillbirth timing and causes.

Data quality issues are compounded by the limited knowledge amongst some health workers, both clinical and non-clinical staff, on stillbirth [12]. While many studies focus on the experience of care, clinical management or the socio-cultural aspects of stillbirths, few address the role of health workers in stillbirth data recording [6, 11, 13–19]. Additional studies from various regions identified health workers' attitudes and beliefs as perceived barriers to adequately recording pregnancy outcomes [20–22].

Against this backdrop, our study aimed to gain insights into health workers’ knowledge and practices related to stillbirth recording and reporting in the Ashanti Region, Ghana.

Methods

Specifically, we aimed to explore three objectives: 1) experiences, perceptions, and attitudes of health workers towards stillbirth recording and reporting; 2) understand barriers related to stillbirth recording and reporting among health workers; and 3) identify support mechanisms available to health workers to enable stillbirth recording and reporting.

Study setting

The qualitative study using key informant interviews involved data collection in government health facilities between January to April 2023 in the Ashanti Region of Ghana. Four districts in the Ashanti region were selected for the study. The rationale for the selected region and study districts (within which the study facilities operate) are published elsewhere [23]. Briefly, the region and districts were selected based on contextual factors related to stillbirth. The Ashanti Region, the most populous region in Ghana, has a high stillbirth rate as well as high neonatal and maternal mortality rates. The selection of districts was informed by the volume of deliveries, the stillbirth rate or number, and the absolute number of maternal deaths or maternal mortality ratio.

Within the districts, we focused on secondary-care level (hospitals) and primary-care level (health centers) government health facilities as our inclusion criteria for facility selection. The primary care-level is the first point of contact that many women have with the health care system. It is often the basis for referring complicated pregnancies to the secondary level. We excluded the one tertiary-level teaching hospital in the Ashanti Region due to previously published studies on stillbirths [14, 24, 25]. Private facilities, faith-based facilities and quasi-government (partly owned by government and another institution, usually a university or religious body) facilities were excluded because they operate with a degree of autonomy and do not follow the same ethical approval processes as government-owned facilities. Table 1 outlines the total number of hospitals and health center facilities available in the study districts.

Table 1.

Total number of secondary and primary care facilities in selected districts

Eight government health facilities were selected for this study, consisting one hospital and one health center per each of the four districts for the study. The rationale for facility selection was informed by four indicators – total number of deliveries, stillbirths, neonatal deaths and maternal deaths (Table 2).

Table 2.

Rationale for facility selection

2022 data as reported in the District Health Information Management System 2 (DHIMS- 2)

*2022 self-reported data during interviews and follow-up communications.

An earlier related study with the District Health Directorate (DHD) provided insights into frontline health workers who deal with stillbirths and stillbirth data [23]. At the health facility-level, our primary focus was health workers who interact with stillbirth data including midwives, as they are the ones primarily responsible for stillbirth cases and the associated data. In addition, we interviewed medical officers and physician assistants who provide support to midwives in stillbirth cases and health information officers who play a critical role in verifying stillbirth data. We planned to interview 4 staff members per facility, amounting to 32 interviews. Table 3 illustrates the distribution of the four types of health workers in each facility.

Table 3.

Distribution of the four types of health workers in each facility and the study sample

2022 data as reported in the District Health Information Management System (DHIMS- 2) for January—June 2023. * Indicates self-reported data from interviews and follow-up communications. † a midwife serves as the health information officer

Procedure

Data collection was conducted in January to April 2023, by the first author who has professional experience in qualitative research and is knowledgeable in stillbirth recording and reporting, and health systems. The DHD helped introduce this study to the health facilities [23]. Once at the health facility, ethical approval documentation was presented to the facility director and the matron in charge before commencing interviews. Convenience sampling based on health worker availability during the interview period was applied.

On interview days, participants received consent forms, either read aloud or provided in print for self-reading. Interviews were conducted in person, in a neutral and comfortable setting after participants provided verbal consent. The interview guide (Supplementary Document 1), developed for the study was informed by literature reviews and conversations with stillbirth measurement experts [26]. A visual aid was also shared with health workers to further elicit perspectives on stillbirths (Supplementary Document 2). The interview guide was pretested and refined with the in-country team in Ghana. Interviews, lasting between 30 to 45 min, were conducted by the first author in Twi, the native language to the Ashanti Region, or English. Only the researcher and study participant were present during interviews. Responses were paraphrased and repeated several times for data validation and improved credibility. As such, no transcribed text was shared with participants following the interviews. All interviews were audio-recorded, transcribed verbatim, translated, and stored in a password-protected computer by the first author. No repeat interviews were carried out. Data saturation was achieved when further interviews did not contribute any new insights.

Data analysis

Thematic analyses were conducted following the Braun and Clark 6-step approach. This included familiarization with the data, generating initial codes, searching for themes, reviewing themes, defining and naming themes and report production [27].

Three major a priori objectives–: 1) experiences, perceptions, and attitudes; 2) barriers; and 3) support mechanisms to stillbirth recording and reporting, were identified based on literature reviews. Themes were added as they emerged from the data. Initial coding was done by the first author, with 35% of interviews reviewed by a second coder. The authorship team provided feedback on coding and theme identification. Any differences emerging during data analysis, were resolved through discussion.

Field notes were cross-referenced with the identified to enhance and expand upon the findings. NVivo qualitative software facilitated data management. The study follows the Consolidated Criteria for Reporting Qualitative Research (COREQ).

Reflexivity

Having a deep contextual awareness of Ghana, as well as being a native of the Ashanti Region, may have influenced how the first author (NAMA) phrased questions in the native dialect of the Ashanti Region, and which questions from the semi-structured interview guide NAMA chose to probe further. The cultural familiarity of NAMA likely shaped the interactions with study participants, potentially leading them to respond in ways they believed the researcher, as a fellow Ghanaian, would understand or approve of. To address potential biases, NAMA asked questions in various ways to cross-check the consistency of responses and reduce the impact of her own assumptions. Additionally, as a Ghanaian female, NAMA may have been biased toward interviewing more females (particularly midwives) than males working in health facilities. This feeling could be linked to shared gender experiences and an implicit comfort level, which might have influenced the dynamics of the interviews. To ensure a more comprehensive sampling, NAMA made a conscious effort to include a diverse range of health workers involved in stillbirth measurement, recognizing that different cadres may have unique insights that could otherwise be overlooked. Furthermore, NAMA’s advanced training in public health (holds an MSc), health systems (worked for the World Health Organization at the time of the study) and extensive knowledge of maternal and child health issues, including stillbirth, may have influenced the data analysis, particularly in determining which thematic areas to focus on. The expertise of NAMA in health systems could have led to the prioritization of certain findings over others (for example, a focus on systemic issues rather than clinical care) guided by professional background and prior experiences. This potential for bias was mitigated by continuously reflecting on how NAMA’s positionality influenced the research process. Additionally, 35% of interviews were reviewed by a second coder and additional feedback on coding and theme identification were provided by the authorship team. Furthermore, the role of power dynamics is also acknowledged. Being viewed as an "outsider," i.e., an external researcher inquiring about the sensitive topic of stillbirths, may have influenced the responses provided by study participants. Active listening (listening without interruption) and adjusting interview techniques (such as how questions are phrased) helped address some of the potential biases.

Results

A total of 28 health workers were interviewed (out of an original planned 32 individuals), comprising 17 midwives, 5 health information officers, 4 physician assistants and 2 medical officers (Table 3). Not all health facilities had a health information officer. Some health centers had 1–2 midwives and no medical officer. Two health workers declined to participate in the study, stating that they needed to attend to an emergency, which was not apparent to the researcher at the time of the interviews.

Saturation of data was attained after 28 interviews, revealing the following themes aligned with our a priori objectives (Fig. 1). Under the “experiences, perceptions and attitudes to stillbirth recording and reporting” objective, the study identified themes related to awareness of stillbirths and stillbirth documentation drivers. For “barriers to stillbirth recording and reporting” themes identified included blame across levels of the health system, and tools and resources. Finally, for “support mechanisms available to health workers for stillbirth recording and reporting” themes included engagement of facility-level staff in audits and the role of training as support mechanisms available to health workers for stillbirth recording and reporting.

Fig. 1.

Study objectives and themes

Direct quotes by health worker cadre are presented to substantiate findings with the number of years of experience (YoE) indicated.

Experiences, perceptions, and attitudes to stillbirth recording and reporting

Awareness

Health workers were generally aware of the importance of recording and reporting stillbirths. This view was summarized by one midwife:

“Recording a stillbirth really matters because … if there is no data how do we improve? (Midwife, 12YoE).”

Health workers highlighted that recording a stillbirth not only serves as an important step towards ending preventable stillbirths but also forms an integral part of the auditing process. Particularly, the processes of recording and auditing a stillbirth were perceived as interconnected. Recording a stillbirth provides the data that underpin audits.

“When we record a stillbirth, we have to go for audit and find out the causes…some causes are preventable. You end up knowing the actual cause to prevent mothers from losing a baby whether during pregnancy or during labor (Midwife, 7 YoE).”

Another midwife shared that recording helps to identify and address health system gaps promptly to prevent stillbirths.

“Recording a stillbirth helps us to know the number of times a stillbirth occurs in a particular month or year so that we can prepare ourselves well, and train our staff to prevent it (Midwife, 3 YoE).”

Inconsistent application of stillbirth definition

The inconsistent application of stillbirth definitions was a notable issue. Respondents demonstrated variations in their understanding of gestational age, a key data element needed to determine length of pregnancy. There were inconsistencies in the gestational age cut-off in weeks used to define stillbirth. Additionally, a few health workers were unable to describe stillbirths using gestational age, and instead, reference was made to death upon delivery.

Table 4 provides an overview of the definitions shared by health workers.

Table 4.

Stillbirth definitions as described by some health workers

Assessment of the fetal skin using fresh and maceration was also used to describe stillbirths.

“Fresh stillbirth is when a baby dies immediately it is born and the macerated is when the baby dies in the uterus (Midwife, 1 YoE)

Pre-service education

Education acquired during midwifery school emerged as a key source of knowledge and information that profoundly shapes individuals’ attitudes and perspectives about stillbirths. In cases where health workers received training on preventable causes of stillbirth and the definition, along with its recording, this education primarily came from midwifery and nursing training school or tertiary universities.

“You know, we got the training from school. When you come to the facility, nobody will teach you, they know you have learnt it. Every trained midwife knows the definition of a stillbirth (Midwife, 8 YoE).”

“We haven’t gone for further education on stillbirths but during the period of schooling that is when I learned something about it. (Midwife,6YoE).”

Documentation

All health workers had a shared understanding on the facility-level processes required following a stillbirth. Once a stillbirth occurs, the attending midwife records the birth outcome in the record books, before informing the matron or midwife in-charge. The district health directorate is informed within 24 hours to initiate the audit processes within 7 days.

Motivation for reporting

In recording a stillbirth, some health workers expressed their motivation to record a stillbirth for two primary reasons. Firstly, it arises from a deep sense of compassion, driven by their commitment to doing what is morally right.

“…Personally, I won’t feel well that a baby died during my line of duty...It will hurt me but you have a responsibility so I have to report (Midwife, 1 YoE).”

Secondly, data verification procedures established by the Ghana Health Service make it challenging for health workers to disregard a stillbirth, as the system eventually identifies, and records missed cases.

“…if there are any discrepancies in the hard-copy, I sit down with the midwives who did the report of that particular month and find out what went wrong (Health Information Officer).”

Omissions

Increased workload among midwives was reported as a factor contributing to omissions in stillbirth reporting.

“Sometimes it is not reported because when you have a stillbirth case and you have a lot of clients coming in for antenatal care, if you don’t take care, you might switch and attend to another mother without recording what happened. With the workload, sometimes, you may forget (Physician Assistant, 5 YoE).”

The requirement to complete various record books when recording a stillbirth outcome, including but not limited to, admissions and discharge form, maternity form, antenatal care booklet, Midwife Returns Form, death notification form and the delivery form, sometimes leads to midwives omitting key information required on stillbirths in record books.

“ … recording in too many books, we sometimes forget some…Mostly the stillbirth sheet, we always forget it. It is later that we recall and get back to fill it (Midwife, 1YoE).”

Knowledge surrounding how to properly complete recording forms and legibility of handwritten forms were flagged as additional reasons for omissions.

Misclassifications

Misclassifications between fresh and macerated stillbirths, as well as stillbirths and neonatal deaths, were reported as recurring issues. Health workers identified limited knowledge as a key factor contributing to these classification gaps.

“… I believe misclassifications occur because the person might not have adequate knowledge about what the stillbirth is or the difference between the stillbirth and the neonatal death (Midwife, 2 YoE).”

Similarly, some health workers attributed insufficient in-service experience with a likelihood to misclassify.

“Normally if you have a new staff, like those who have just been posted, sometimes you have APGAR of one, then later on the baby will be gone. This is where sometimes they misclassify between stillbirth and neonatal death (Midwife, 14 YoE).”

Barriers

Blame

As reported by midwives, blame manifests in three layers: blame from the wider health-system level, blame at the organizational facility-level and individual-level blame.

Health system-level blame

Audits are mandated to be conducted within a week of a stillbirth occurrence and involve a multi-disciplinary team of health workers from the Regional Health Directorate, DHD and health facilities. However, midwives shared that while audits are intended to be a learning experience aimed at addressing identified shortcomings, challenges persist in their execution. These challenges relate to the language used during audit investigations.

“The auditors shouldn’t blame midwives when such an incident happens. Due to the blame, if something happens and the midwife can help, she will not risk it. She will just refer to avoid all the questioning (Midwife, 8YoE).”

“Nobody likes audits to be honest. Most of the midwives do not like audits because sometimes they are made to feel incompetent (Medical Officer, 4YoE).”

As part of the audit process, health facilities that refer cases to the higher-level are invited to participate in the audit process. Some health workers placed blame on the referring facilities for a stillbirth outcome.

“…most of the stillbirths we were having, were from our peripheral facilities. That is the primary care, so the level of care was not that adequate (Medical Officer, 4 YoE)

Some health workers noted that blame can arise for various reasons, many of which are linked to macerated stillbirths. In particular, midwives may choose not to report a stillbirth to avoid inflating their facility’s mortality statistics.

“You know initially, like macerated for instance, they didn’t want to record because they felt that this baby didn’t come with fetal heart rate so if I should record, I am going to have higher numbers and they are going to say that I am not doing my work well (Midwife, 12 YoE).”

Some midwives held the belief that lower-level facilities sometimes refer stillbirth cases to higher-level facilities late to avoid the need to document stillbirth outcomes within their own facilities.

“Sometimes the recordings are not done because like for instance, someone will be attending antenatal somewhere in a different facility. During delivery, the person might come in here with the stillbirth. Due to that, we don’t want to record to spoil our data, so sometimes we might miss that one (Midwife, 10 YoE).”

Organizational facility-level blame

Some midwives highlighted that senior staff tend to blame the attending junior midwife when a stillbirth occurs.

“When a stillbirth happens, they blame you as if you intentionally did it (Midwife, 3YoE).”

The way some supervisors communicate with midwives was identified as an obstacle to reporting stillbirths.

“I think the superiors should have a cordial relationship with their juniors so that when something occurs, they (junior staff) will be okay to report (Midwife, 8 YoE).”

Individual-level blame

All midwives shared that witnessing a stillbirth is traumatic. There is a general sense of guilt and blame when a pregnancy outcome is a stillbirth.

“When you deliver a dead baby, it makes you feel like you did not complete your work. That is not why you came here. You came here to deliver babies. Live babies not dead babies so it is quite sad (Midwife, 1 YoE).”

The feeling of blame can affect the decision to record a stillbirth or not.

“A lot of us go through a whole lot of trauma and other things when stillbirths happen. Because of that, most midwives do not want to record when they get a stillbirth (Midwife, 7 YoE).”

Some midwives flagged the long-term impact associated with delivering a stillbirth, which is often overlooked.

“You need to train yourself. We do not have a psychologist here to speak to. They (supervisors) will say, it is part of the work. Nobody talks to you about it so mentally you fight through and let the process pass. I remember almost 10 years ago I had a death. For more than a year, I couldn’t be myself, even though I knew it wasn’t my fault (Midwife, 12 YoE).”

Tools and resources

Data capture

Tools are needed to facilitate timely and accurate recording of a stillbirth. In some health facilities, we observed that the facilities had run out of their supply of recording books. Thus, facilities improvised using a normal notebook. However, the improvised record book had key data missing, including information related to stillbirths. This observational finding was further corroborated by perspectives shared by some health workers.

“For now, the standard Form A (Midwife Returns Form) has finished, and we had to improvise. You know the indicators in the standard one, are too many so at times we skip some of them. Not all the indicators can be recorded in the improvised one. In the improvised one, they (health care workers) select a few to report on (Health Information Officer).”

Audit recommendations

Implementation of audit recommendations emerged as a key bottleneck to address gaps uncovered within the system, often impending efforts to avoid preventable stillbirths and improve birth outcomes.

“…We will go, sit there and talk about it. They (audit team) will tell us to go and see the authorities to purchase. But, in the long run, we will not get it. When the audit recommendation involves equipment or tools, we don’t get (Midwife, 7 YoE).”

A few midwives flagged that though audits are useful, the documentation of audit recommendations can be challenging. One midwife shared that the audit form should be made available to all midwifery units to serve as a guidance tool during delivery.

“We don’t have the audit sheet. If I didn’t go for my colleague’s audit, I wouldn’t have seen that book. I think every facility needs to have that book. In a way it guides you to know what you are doing. Before you even go for the audit, you know where you were at fault… (Midwife, 1 YoE).”

Support mechanisms

Facility-level engagement

All health workers emphasized the importance of audits in ensuring improvements in service delivery. Particularly, the feedback received during audits was highlighted as an educational moment to close gaps recorded in the system.

“Just recently, I think we had one stillbirth, so we had an audit and through that, we educated ourselves about things to do so that it does not happen again (Midwife 10YoE).”

An additional support mechanism aimed at facilitating improvements within the wider health system is the engagement of health workers from the receiving and referring health facilities in audits. Health workers stressed the importance of coordination and communication within the district to improve overall quality of care.

“ When we have an audit, we notify each facility with letters to participate. Each facility is supposed to bring a representative to learn from the process so that it (stillbirths) does not occur at the peripheral levels (Midwife, 12 YoE).”

Training

Capacity building was shared as an avenue for supporting health workers in strengthening their skills in maternal and perinatal health. However, most of the capacity-building needs are focused on training related to clinical care and audits, with little to no training provided on appropriate recording of stillbirths.

“Yes, we have had trainings on helping baby breath and other aspects to reduce stillbirth…We talk about audits...fetal heart monitoring and the steps involved...(Midwife, 12YoE).”

Some experienced health workers alluded to the fact that should training opportunities be made available to new midwives, this would ease some of the gaps with data recording.

“An experienced person sees and acts differently from a new one so definitely experience counts a lot. If the new midwives avail themselves for training, there will not be any problems. They will also have the courage to do things to avoid stillbirth (Physician Assistant, 20 YoE).”

However, a noted challenge echoed by relatively newly posted midwives is that training opportunities are mostly available to older tenured midwives.

“I haven’t gone for stillbirth training before. Those who go for the trainings are those who have worked here for a longer period e.g., 10 years…Me? I have been here for just 3 years and never had a training (Midwife, 3YoE).”

Discussion

The study findings indicate that health workers had a good understanding of the importance of recording a stillbirth. However, variations were observed in the gestational weeks cut-off used to define stillbirths. Perspectives on stillbirths were influenced by pre-service education. Further, the findings highlight some reasons for omissions and misclassifications of stillbirths, shedding light on the motivations driving the recording and reporting of stillbirths. Challenges emerged, including blame shifting across different levels of the health system and a lack of adequate tools and resources. Further, the findings underscored the importance of active participation of health workers in audit processes and the availability of relevant training opportunities, as supportive mechanisms for improving the accuracy of stillbirth recording and reporting.

Experiences, perceptions and attitudes to stillbirth recording and reporting

A common issue surrounding stillbirths is its definition. Countries use different reporting criteria and thresholds for gestational age and birthweight [28–31]. The varying definitions used across countries, and sometimes within the same country [32], as revealed in this study, create challenges in accurately recording stillbirths and understanding the true scope of the population mortality. Despite the substantial burden of nearly 2 million third trimester stillbirths annually, stillbirths remain invisible in many national and global-level conversations. Lack of consistency in applying definitions can lead to incorrect classification of adverse pregnancy events – potentially omitting (if misclassified as miscarriage), or misclassifying stillbirths (if misclassified as neonatal deaths) in routine data [33]. Acknowledging this gap, the United Nations Inter-agency Group for Child Mortality Estimation (UN IGME) and the WHO recently updated global guidance on defining a stillbirth in the International Classification of Disease (ICD) 11th revision; defining stillbirth as a baby born with no signs of life at 22 or more completed weeks of gestation, and distinguishing between early gestation stillbirth (at 22 to 27 completed weeks of gestation) and late gestation stillbirth (at 28 or more completed weeks of gestation) [34].Communicating better with countries on this new definition for stillbirths will assist countries to consistently and universally apply the definition to facilitate the accurate reporting and recording of stillbirths.

Both intrinsic and extrinsic motivation were shared by health workers as reasons for recording stillbirths. Intrinsic motivation aligns with the personal will to do the right thing i.e., recording the birth outcome of a stillbirth. This commitment to do what is right may stem from a commitment to provide quality health care [14, 16, 35] or professional ethics enshrined in many pledges, declarations or oaths undertaken by health care professionals [36–39]. Extrinsic motivation to record a stillbirth outcome is encouraged by external factors. In this study, we found that adhering to standards set by the Regional Health Directorate and the DHD was the crucial factor. This finding was aligned with past studies on district health management teams (DHMTs) [40, 41], as DHMTs teams conduct data quality checks on health facility data.

Our findings, consistent with previous studies, demonstrate that pre-service education plays a crucial role in shaping midwifery practices within maternity health service delivery [42, 43]. Consequently, there was no significant difference between responses from health workers in hospitals and those in primary care facilities. Reflecting on the varied definitions shared by midwives in this study, it is important for the Ghana Health Service and the Nursing and Midwifery Council of Ghana to review the pre-service education curricula. This review is necessary to ensure the latest definition of stillbirth alongside its appropriate recording and reporting are thoroughly incorporated into the educational programme [44]. Well-trained midwives are more inclined to record and report stillbirth cases thereby facilitating the implementation of necessary evidence-based interventions to address gaps.

Barriers to stillbirth recording and reporting

A perinatal audit is the process of capturing information on the number and causes of mortality with a view to improve the care provided to mothers and babies [45]. Key elements of this audit cycle include identifying the death, reporting the death, reviewing the death and taking action. Health workers acknowledged that “if there is no data, how do we improve,” yet taking action to address gaps remains challenging, particularly in procuring essential tools. Perinatal audits are means to improve the quality of health service delivery and birth outcomes [46]. Thus, without taking action, pre-existing unfavorable attitudes towards stillbirth data collection and a lack of data ownership may persist, hindering the progress to improve stillbirth recording and reporting [47]. The impact of perinatal audit relies on successfully closing the audit loop. Mere recording of stillbirths and its causes is insufficient. Improvement in quality of care occurs when the solutions to identified problems are effectively implemented. Key strategies proposed within existing literature to promote a positive culture for perinatal audits include explaining the purpose, process and how to conduct blame-free audit with implementation tools available at all levels of the health system [48].

The importance of a blame-free culture for health workers is well documented in the literature [41, 48]. At the broader health system level, we found that blame was an essential element of audit processes. This finding contradicts the intended purpose for which audits were designed. When blame and punishment exist, health workers shared they feared reporting mortalities. A blame environment assigns blame rather than understanding the root causes of mortalities and focuses on individual errors rather than systemic issues. A shift in culture from one of blame to one of learning and improvement is needed to ensure that health workers feel comfortable to record, report, review and implement key actions to improve birth outcomes.

Another component of blame was at the referral-level. Some health workers blamed lower-level facilities for referring stillbirth cases at the last moment for fear of punitive measures or avoiding recording such cases in their own facilities. This culture of downstream blame aligns with two case studies from India [49]. One study found that mortality was under-reported due to fear of punitive actions. This fear led to late referrals, mainly to prevent punishment [50]. Higher facilities struggled with patient referrals, and if a patient experienced adverse outcomes within their facility, they often shifted the blame onto providers from lower-level facilities. The Ghana Health Service is currently addressing this downstream blame culture by engaging referral health facilities within the audit process.

At the organizational health facility-level, blame culture was identified as one of the areas needing urgent attention [48]. Health workers flagged issues of blame arising from other clinical staff. Of note, most senior midwives blamed issues on stillbirth recording on junior staff. Junior midwives, in contrast, expressed their reluctance to report stillbirths due to the punitive language used by senior staff members. When such a culture exists, studies have shown that health workers often chose not to report issues on quality of care because reporting an incident brings the risk of conflict with other colleagues [51]. Literature suggests that blaming people, focusing on mistakes, and yelling are demotivating factors for obstetric care [52]. However, successful supervision during clinical care was associated with higher clinical experience. In the Northern Ghana, researchers found that a positive relationship between supervisor and student creates a positive learning environment [53]. In such cases, health workers are open to reporting mortalities and discussing the outcomes with supervisors.

At the individual level, blame can lead to demoralized health workers. This can affect motivation, job satisfaction, and overall performance, impacting the quality of health services. A blame environment promotes a culture of omitting or under-reporting stillbirths.

Support mechanisms for stillbirth recording and reporting

Health workers acknowledged the critical role played by learning and feedback mechanisms as part of audit processes. However, to realize this, a collective shift from a blame-oriented culture to one focused on learning and improvement is required [41]. One study argued that individuals should be willing to continually seek lifelong learning. Simultaneously, at the health systems level, policies and protocols should be designed to cultivate an environment promoting professionalism and continuous development [54].

Disparities in in-service training opportunities, determined by years of service, influenced the selection of participants for training sessions. Junior midwives expressed their interest in receiving training; however, these opportunities were not consistently provided, often being prioritized for senior staff members. A study examining task-sharing in the African region recommends ensuring equal access to in-service training for all staff members [55]. Equal opportunities to in-service training ensures that all midwives are trained on standardized protocols and best practices for stillbirth recording.

Limitations

This study exclusively focused on government health facilities despite, private health facilities constituting 40.2% of health facilities in Ghana. We recommend conducting further studies involving health workers in the private sector. Furthermore, our study focused on one region in Ghana. To obtain a comprehensive understanding of the situation in the country, further research across diverse regions is needed, noting that health workers frequently undergo rotations across various regions in Ghana.

Implications for action

Five actions are proposed to improve stillbirth recording and reporting at the health facility level. Proposed actions necessitate close coordination with health facility staff and the district health management teams.

Firstly, strengthen capacity of midwives and health workers on the definition for stillbirth, the types of stillbirths and its appropriate recording and reporting. Additionally, this recommendation includes a review of the pre-service nursing and midwifery education curricula. Secondly, periodically review audit protocols to ensure that a non-punitive approach is emphasized during reviews/audit meetings, trainings, and supportive supervision visits. Thirdly, review the health workforce training tracking system/sheet to ensure that equal opportunities are provided to all midwives for capacity strengthening opportunities. Fourth, when recommendations are made during audits, ensure that effective action is taken to close the audit loop. Finally, emphasize as part of audit process shared accountability, coordination, and communication between levels of the health system.

Conclusion

This study underscores the critical role of frontline health workers in recording stillbirths. To reduce the burden of stillbirths, it is important to understand the perspectives of health workers that inform stillbirth data, which drives national and global-level action. This is the first in-depth analysis of the issue. Comprehensive global efforts are needed to fully understand why stillbirths remain significantly underrepresented in global data tracking, which obscures the true magnitude of the problem.

Abbreviations

COREQ

Consolidated Criteria for Reporting Qualitative Research

DHD

District Health Directorate

DHIMS-2

District Health Information Management Systems 2

DHMT

District Health Management Team

GHS

Ghana Health Service

ICD

International Classification of Disease

LSHTM

London School of Hygiene and Tropical Medicine

UN IGME

United Nations Inter-agency Group for Child Mortality Estimation

UNICEF

United Nations Children’s Fund

YoE

Years of Experience

Authors’ contributions

NAMA, YBO, HB, and DJ did the study conceptualization. NAMA provided formal analysis. NAMA conducted the research investigation. NAMA, YBO, HB, and DJ designed the methodology. YBO, HB, and DJ provided supervision. NAMA and YBO conducted data validation. NAMA provided virtualization. NAMA wrote the original draft. DJ, HB, YBO, KFO, EKA, and RLR performed writing–review & editing

Funding

No specific funding was provided for this study however, funding for DJ was provided by the Takeda Foundation as part of the Takeda Chair in Global Child Health at the London School of Hygiene and Tropical Medicine.

Data availability

The datasets used and/or analyzed during the study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was provided by Ghana Health Service (GHS, Reference: GHS-ERC 025/07/22) and the London School of Hygiene and Tropical Medicine (LSHTM, Reference: 28017). Approval was also granted by one of the study hospitals, functioning as a regional referral hospital (KSH./RESH-50).

Study participants were provided with a consent form (Supplementary Document 3) and an information sheet (Supplementary Document 4) outlining the intended scope of the research. Informed consent was obtained from all subjects. Interviews were conducted in a neutral and comfortable setting after participants provided verbal consent. The research was conducted in accordance with the Declaration of Helsinki.

Findings from the research was shared with the Ashanti Regional and District Health Directorate, as well as the Ghana Health Service.

Consent for publication

Not applicable.

Competing interest

The authors declare no competing interests.