Open Access Research article

A root-cause analysis of maternal deaths in Botswana: towards developing a culture of patient safety and quality improvement

Farai D Madzimbamuto12*, Sunanda C Ray12, Keitshokile D Mogobe3, Doreen Ramogola-Masire4, Raina Phillips4, Miriam Haverkamp4, Mosidi Mokotedi3 and Mpho Motana5

Author Affiliations

1 School of Medicine University of Botswana, Gaborone, Botswana

2 Department of Anaesthesia and Critical Care Medicine, University of Zimbabwe College of Health Sciences, Mazowe Street, Belgravia, Harare, Zimbabwe

3 School of Nursing University of Botswana, Gaborone, Botswana

4 Government of Botswana/University of Pennsylvania [Botswana-UPenn] Partnership, Gaborone, Botswana

5 Medical Education Partnership Initiative [MEPI] Maternal Mortality Project, Gaborone, Botswana

For all author emails, please log on.

BMC Pregnancy and Childbirth 2014, 14:231  doi:10.1186/1471-2393-14-231

Published: 16 July 2014



In 2007, 95% of women in Botswana delivered in health facilities with 73% attending at least 4 antenatal care visits. HIV-prevalence in pregnant women was 28.7%. The maternal mortality ratio in 2010 was 163 deaths per 100 000 live births versus the government target of 130 for that year, indicating that the Millennium Development Goal 5 was unlikely to be met. A root-cause analysis was carried out with the aim of determining the underlying causes of maternal deaths reported in 2010, to categorise contributory factors and to prioritise appropriate interventions based on the identified causes, to prevent further deaths.


Case-notes for maternal deaths were reviewed by a panel of five clinicians, initially independently then discussed together to achieve consensus on assigning contributory factors, cause of death and whether each death was avoidable or not at presentation to hospital. Factors contributing to maternal deaths were categorised into organisational/management, personnel, technology/equipment/supplies, environment and barriers to accessing healthcare.


Fifty-six case notes were available for review from 82 deaths notified in 2010, with 0–4 contributory factors in 19 deaths, 5–9 in 27deaths and 9–14 in nine. The cause of death in one case was not ascertainable since the notes were incomplete. The high number of contributory factors demonstrates poor quality of care even where deaths were not avoidable: 14/23 (61%) of direct deaths were considered avoidable compared to 12/32 (38%) indirect deaths. Highest ranking categories were: failure to recognise seriousness of patients’ condition (71% of cases); lack of knowledge (67%); failure to follow recommended practice (53%); lack of or failure to implement policies, protocols and guidelines (44%); and poor organisational arrangements (35%). Half the deaths had some barrier to accessing health services.


Root-cause analysis demonstrates the interactions between patients, health professionals and health system in generating adverse outcomes for patients. The lessons provided indicate where training of undergraduate and postgraduate medical, midwifery and nursing students need to be intensified, with emphasis on evidence-based practice and adherence to protocols. Action plans and interventions aimed at changing the circumstances that led to maternal deaths can be implemented and re-evaluated.

Root-cause analysis; Maternal mortality; Contributory factors; Botswana