The midwife leaders strengthening obstetric triage in Ghana

In recent years, Ghana has made significant progress in maternal and newborn health, including large increases in institutional deliveries. Yet major challenges remain in ensuring that hospitals can accommodate high caseloads while maintaining high-quality care for mothers and babies. In 2020, the average maternal mortality rate in Ghana was 263 per 100,000 live births, and the average neonatal mortality rate was 23 per 1,000 live births. These remain well above global averages and Sustainable Development Goals to reduce maternal mortality to less than 70 per 100,000 live births and neonatal mortality to 12 or fewer deaths per 1,000 live births.

Most maternal and neonatal deaths are preventable if quality care is provided by skilled health professionals before, during and after childbirth. However, many hospitals still face shortages in the critical resources needed to treat high-risk patients. Furthermore, spreading knowledge on practice improvements to the whole workforce is challenging, particularly in resource-constrained settings.

Evidence suggests that informal leaders in the workplace are well placed to diffuse information, accelerate learning and encourage their peers to adopt new practices. Yet there is limited evidence on how to select workers who are credible, respected and influential enough to spread knowledge effectively among their peers.

The Obstetric Triage Implementation Package

A study part funded by Thrive aimed to address this evidence gap. The intervention we study is Ghana’s Obstetric Triage Implementation Package (OTIP) protocol. Developed by the NGO Kybele in partnership with the Ghana Health Service (GHS), OTIP was introduced to improve the quality of care for mothers in labour across Ghana’s high-volume hospitals (those with more than 1,200 deliveries in 2022). It was rolled out nationally between 2023 and 2025.

OTIP’s aim is to enable midwives to promptly and accurately identify the severity of pregnant women’s conditions, determine the appropriate level of care, and ensure that those with the most critical needs receive immediate attention. Under the protocol, women arriving at the hospital should be assessed within 10 minutes in a dedicated obstetric triage area. Midwives use a standardised assessment form to evaluate obstetric and medical history, vital signs and labour progress, then categorise patients as high, intermediate or low risk using colour-coded wristbands. A care plan is then developed and documented based on the diagnosis and risk status.

In each hospital, a small group of midwives receive one-week-long, on-site training from OTIP national trainers, who are Ghana Health Service obstetrician-gynaecologists and midwives. The training covers clinical risk assessment, practical triage skills, OTIP-specific procedures and leadership skills needed to support implementation. These trained midwives are known as Champions.

Champions are expected to lead implementation of OTIP within their hospitals. Their responsibilities include training colleagues, disseminating knowledge on triage practices, encouraging adherence to new procedures, and helping establish and maintain dedicated triage areas. Despite taking on these additional responsibilities, Champions do not receive extra monetary compensation.

To be considered for the role, Champion candidates are expected to demonstrate leadership ability, strong teamwork, flexibility, the ability to handle stressful situations, good communication skills, an understanding of compassionate care and the ability to take initiative in difficult situations.

This cascade model is attractive because it does not require every midwife to be trained directly by national trainers. However, it depends on whether the right people are selected to spread knowledge through peer networks.

Research questions and design

Our research team partnered with the GHS and Kybele to embed a large-scale experiment during the final phase of the OTIP rollout – which started in September 2024 and covered 25 hospitals in the Greater Accra, Central and Western regions (with more than 1,400 midwives employed in total). The study tested two questions. First, did OTIP improve midwives’ knowledge and triage practices? Second, did the way Champions were selected influence programme adoption and effectiveness?

While OTIP relies heavily on peer influence and was designed with a collaborative leadership approach in mind, Champions are traditionally appointed by hospital management – a hierarchical selection process that may miss the benefits of leveraging peer networks. Our study introduced an alternative of peer election, where Champions were democratically selected by their fellow midwives.

Hospitals were randomly assigned to receive OTIP earlier or later, allowing us to rigorously estimate OTIP’s impact. They were also randomly assigned to one of two approaches for selecting Champions: appointment by hospital management, referred to as top-down selection, or election by midwife peers, referred to as bottom-up selection. In the bottom-up hospitals, midwives nominated candidates through an online platform; the response rate was 80%, and the top nominees were chosen as OTIP Champions. In each hospital, 10 Champions (representing roughly 16% of the midwifery workforce on average) were selected for training.

Study results

We collected data through a census, surveys with midwives and mothers, and novel behavioural games to measure attitudes and behaviours. OTIP improved clinical knowledge, encouraged better adherence to recommended processes and enhanced midwives’ ability to identify high-risk cases in challenging scenarios. It also improved aspects of clinical practice, including the provision of drugs in response to complication during delivery.

These gains suggest that OTIP is a promising foundation for strengthening obstetric triage. Sustained support, including reinforcement of triage protocols, periodic refresher training and practical simulation exercises, may help consolidate these improvements over time.

Crucially, the study also revealed that the way Champions were chosen was central to the programme’s success. Hospitals using the bottom-up process achieved significantly larger improvements than hospitals where Champions were appointed by management. Clinical knowledge was 0.17 standard deviations higher in bottom-up hospitals than in top-down hospitals, and knowledge of the recommended maximum waiting time for assessment was 11 percentage points higher.

The stronger impacts in bottom-up hospitals do not appear to be explained by differences in training exposure, qualifications, baseline knowledge or midwifery experience. Instead, peer-elected Champions had attributes that made them especially effective at spreading knowledge.

Compared with top-down Champions, bottom-up Champions held characteristics that are advantageous for diffusing new practices, such as lower neuroticism and greater openness. They also experienced greater legitimacy among colleagues, strengthening their ability to motivate and influence others. The nomination process suggests that midwives were not simply choosing friends. The most common reasons given for nominating Champions were that they ‘fit the criteria’ and ‘would be good trainers’. Fewer than 5% of respondents said friendship was a reason for their nomination.

This recognition and social legitimacy appear to have strengthened the ability of bottom-up Champions to motivate colleagues and diffuse knowledge. The results suggest that frontline workers may be especially well placed to identify informal leaders who are trusted and influential within their own networks.

Policy implications

Training roughly one in six midwives and tasking them with passing knowledge on to peers was effective in catalysing organisational learning. For hospitals operating under fiscal and staffing constraints, cascade training can therefore be a promising strategy for expanding workforce knowledge at low marginal cost.

The study also shows that the design of cascade training matters. A bottom-up selection process that uses peer knowledge can significantly amplify the impact of the intervention.

Future work should examine whether these gains are sustained and whether the mechanisms documented in the experiment translate into longer-run improvements in outcomes for mothers and babies.

Find out more about the study in this briefing and working paper.

Banner image: USAID on Flickr