Quality improvement collaboratives as part of a quality improvement intervention package for preterm births at sub-national level in East Africa: a multi-method analysis.

BACKGROUND: Quality improvement collaboratives (QIC) are an approach to accelerate the spread and impact of evidence-based interventions across health facilities, which are found to be particularly successful when combined with other interventions such as clinical skills training. We implemented a QIC as part of a quality improvement intervention package designed to improve newborn survival in Kenya and Uganda. We use a multi-method approach to describe how a QIC was used as part of an overall improvement effort and describe specific changes measured and participant perceptions of the QIC. METHODS: We examined QIC-aggregated run charts on three shared indicators related to uptake of evidence-based practices over time and conducted key informant interviews to understand participants' perceptions of quality improvement practice. Run charts were evaluated for change from baseline medians. Interviews were analysed using framework analysis. RESULTS: Run charts for all indicators reflected an increase in evidence-based practices across both countries. In Uganda, pre-QIC median gestational age (GA) recording of 44% improved to 86%, while Kangaroo Mother Care (KMC) initiation went from 51% to 96% and appropriate antenatal corticosteroid (ACS) use increased from 17% to 74%. In Kenya, these indicators went from 82% to 96%, 4% to 74% and 4% to 57%, respectively. Qualitative results indicate that participants appreciated the experience of working with data, and the friendly competition of the QIC was motivating. The participants reported integration of the QIC with other interventions of the package as a benefit. CONCLUSIONS: In a QIC that demonstrated increased evidence-based practices, QIC participants point to data use, friendly competition and package integration as the drivers of success, despite challenges common to these settings such as health worker and resource shortages. TRIAL REGISTRATION NUMBER: NCT03112018.

How people-centred health systems can reach the grassroots: experiences implementing community-level quality improvement in rural Tanzania and Uganda.

BACKGROUND: Quality improvement (QI) methods engage stakeholders in identifying problems, creating strategies called change ideas to address those problems, testing those change ideas and scaling them up where successful. These methods have rarely been used at the community level in low-income country settings. Here we share experiences from rural Tanzania and Uganda, where QI was applied as part of the Expanded Quality Management Using Information Power (EQUIP) intervention with the aim of improving maternal and newborn health. Village volunteers were taught how to generate change ideas to improve health-seeking behaviours and home-based maternal and newborn care practices. Interaction was encouraged between communities and health staff. AIM: To describe experiences implementing EQUIP's QI approach at the community level. METHODS: A mixed methods process evaluation of community-level QI was conducted in Tanzania and a feasibility study in Uganda. We outlined how village volunteers were trained in and applied QI techniques and examined the interaction between village volunteers and health facilities, and in Tanzania, the interaction with the wider community also. RESULTS: Village volunteers had the capacity to learn and apply QI techniques to address local maternal and neonatal health problems. Data collection and presentation was a persistent challenge for village volunteers, overcome through intensive continuous mentoring and coaching. Village volunteers complemented health facility staff, particularly to reinforce behaviour change on health facility delivery and birth preparedness. There was some evidence of changing social norms around maternal and newborn health, which EQUIP helped to reinforce. CONCLUSIONS: Community-level QI is a participatory research approach that engaged volunteers in Tanzania and Uganda, putting them in a central position within local health systems to increase health-seeking behaviours and improve preventative maternal and newborn health practices.

Illness recognition and care-seeking for maternal and newborn complications in rural eastern Uganda.

BACKGROUND: To enhance understanding of the roles of community-based initiatives in poor rural societies, we describe and explore illness recognition, decision-making, and appropriate care-seeking for mothers and newborn illnesses in two districts in eastern Uganda where in one implementation district, a facility and community quality improvement approach was implemented. METHODS: This was a cross-sectional study using qualitative methods. We conducted 48 event narratives: eight maternal and newborn deaths and 16 maternal and newborn illnesses. Additionally, we conducted six FGDs with women's saving groups and community leaders. Qualitative data were analyzed thematically using Atlas.ti software. RESULTS: Women and caretakers reported that community initiatives including the presence of community health workers and women's saving groups helped in enhancing illness recognition, decision-making, and care-seeking for maternal and newborn complications. Newborn illness seemed to be less well understood, and formal care was often delayed. Care-seeking was complicated by accessing several stations from primary to secondary care, and often, the hospital was reached too late. CONCLUSIONS: Our qualitative study suggests that community approaches may play a role in illness recognition, decision-making, and care-seeking for maternal and newborn illness. The role of primary facilities in providing care for maternal and newborn emergencies might need to be reviewed.