Quality improvement initiatives: the case of the sick newborn
In 2018 an excellent review article was published, summarising all the strategies that have been used to improve the quality of care of newborns.
As you would remember, newborn mortality is the element of under-five mortality with the least improvement, and now makes up some 40% of all under-five mortality. The paper, Quality improvement initiatives for hospitalised small and sick newborns in low- and middle-income countries: a systematic review, reports on a systematic review to examine strategies that improve the quality of newborn care.
The authors looked at 28 studies that covered 23 low and middle-income countries.
They divided the interventions into the level at which they worked: the macro (above district level), meso (interventions that work at the clinic and district level) and micro (those interventions that change the provider-client interaction). As you might expect, most of the interventions worked at the meso (clinic or district) level. The interventions described at each level are summarised in the following table (some studies involved more than one strategy).
Table 1. Strategies and health system level
|Level||Strategy||Number of studies|
|Micro||Distribution of reference materials to providers||8|
|Meso||Strengthening facility infrastructure||6|
|Continuous quality improvement||7|
|Macro||Regulation and governance||1|
As you can see from the table the most common intervention involved in-service training of health providers, followed by reorganising services (e.g. adding more streamlined triage, or changing the patient flow and room arrangements in a Neonatal Intensive Care Unit). The macro approaches, which reflect a more systems approach, included undertaking a national service gap assessment and then authorising task shifting to address the identified gaps, and a national level focus on capacity building (using a range of strategies) to embed improved practice. It is interesting to see the relative lack of system wide quality improvement interventions.
The authors then looked at the evidence for the effectiveness of these quality improvement strategies. ‘Effectiveness’ was most commonly measured by newborn mortality, and some other indicators like length of stay in hospital. None of the studies they found had looked at whether any changes in equity resulted from the interventions.
What do you think the findings were?
If you said ‘mixed’ then you would be correct – there were many studies that demonstrated improved outcomes, but also many studies where there was no change, or even poorer outcomes.
The authors then did a more interesting analysis of the enabling and constraining factors to effective quality improvement, and the results are summarised in this list:
Promoters of quality
- Motivation of key individuals
- Continuing monitoring throughout
- Interdisciplinary collaboration
- Abandonment of unnecessary practices
- Schemes tailored to participants
- On-site support
- Refresher programmes
- Formal training in QI methods
- Low cost of intervention
- Relationships between health workers, community leaders and district officials
- High-quality national data collection
- Formal health service support
- NGO collaboration initiatives
Barriers to quality
- Overburdened staff
- Lack of sufficient equipment
- High changeover of workforce
- Defects in staff knowledge and practice
- Unmotivated staff
- Multiple QI measures/audits simultaneously
- Insufficient funding
- Insufficient health services relative to demand
- Government redistribution of staff
- Inadequate documentation
- Confounding health policy changes
What can we learn from this review?
We think there are a number of lessons. We tend to focus on single interventions in quality improvement rather than looking at system-wide quality improvement, which is harder, more expensive and more difficult to measure and attribute change. We come back to many of the themes that have run throughout this course: we don’t get the same outcome when we apply the same intervention in a different context; we are constrained by our data and the quality of our HMIS in measuring improvement; we are more comfortable with training interventions and we invest less in system approaches.
The authors concluded:
“Going forward, we recommend more rigorous evaluation of quality improvement in neonatal hospital care. Interventions are commonly at the meso level and educational in nature, and more focus is required around macro- and micro-level interventions; other study designs should be explored, with direct investigation of barriers and promoters. This should be linked to programmatic efforts where possible, in order to combine implementation and research. Small and sick hospitalised newborns in LMICs are a population at the highest risk—they should be one of the prime beneficiaries of quality of care interventions and investments.” (Zaka et al., 2018, pg. 18)
In your context:
- Do you find that training health workers is often the first (and maybe ONLY) quality improvement strategy?
- What does this say about our understanding of the importance of health system interventions?
Share any thoughts and experiences in the comments section below, then, take some time to read how your response compares to others.
Zaka, N, Alexander, EC, Manikam, L, Norman, IC, Akhbari, M, Moxon, S, Ram, PK, Murphy, G, English, M, Niermeyer, S & Pearson, L, 2018. ‘Quality improvement initiatives for hospitalised small and sick newborns in low-and middle-income countries: a systematic review’, Implementation science, vol. 13, no. 1, p.20.
© Nossal Institute for Global Health at the University of Melbourne