Readers will be familiar with my use of terms such as systematic review and meta-analysis when describing medical research. They are very much the lingua franca of evidence-based medicine.
But what do they mean and how do they affect the healthcare you receive as a patient?
The authors of a meta-analysis combine the data from multiple previous studies and perform a new statistical analysis. In a systematic review, the authors collect and critically analyse multiple studies using criteria that were set before the review starts; it differs from a literature review which includes any research papers that provide evidence for a particular point.
However, we need this cumulative information to move from academia to the bedside. This is achieved through a process called evidence synthesis which aims to provide a rigorous and transparent knowledge base for translating research into decisions about your health.
The advantage of evidence synthesis is that all studies on a topic are assessed together in context. Individual studies may make dramatic headlines, but they cannot form the basis of decision-making until they are placed in context of what is already known.
Which makes a new €2 million, all-island investment to improve ways of gathering and combining evidence and using it effectively to inform health and healthcare decisions most welcome. The initiative, Evidence Synthesis Ireland (ESI), is being funded for three years by the Health Research Board (€1.5 million) and the research and development division of the Public Health Agency in Northern Ireland (€500,000). (Full disclosure: I am involved in the project.)
Decision-making
Of note, ESI supports a key action in the Sláintecare report, which highlights that evidence should be at the core of routine decision-making across the health system. Prof Declan Devane, professor of midwifery at NUI Galway and director of the HRB – Trials Methodology Research Network, who leads Evidence Synthesis Ireland, says he is excited by the potential the project has to make a difference to healthcare decisions across the island. "Healthcare decisions should be based on the full range of information that is available on a topic; it can't rely on one or more individual pieces of information."
What happens if you don’t perform up to date systematic reviews and ensure they are communicated to healthcare professionals?
In 1972, the first randomised controlled trial of a short, inexpensive course of a steroid given to women about to give birth prematurely, was carried out.
It suggested strongly that steroids reduce the risk of babies dying from the complications of immaturity. By 1991, seven more trials had been reported, and the evidence had become even stronger. Cumulatively, the treatment was shown to reduce the odds of the babies dying from the complications of immaturity by 30 to 50 per cent. But because no systematic review of these trials had been published until 1989, most obstetricians had not realised that the treatment was so effective.
More access
As a result, tens of thousands of premature babies have suffered and some may have died unnecessarily. The human cost of failing to synthesise this evidence was significant.
Despite the rapid increase in the amount and availability of healthcare evidence, there is a lack of capacity to conduct evidence synthesis. There is a particular need to train more people in evidence synthesis skills. ESI will do this through workshops, webinars and fellowships to drive capacity. Another important aim is to bring the findings of evidence syntheses closer to the general public by making them more accessible and easier to understand.
Failure to use synthesised evidence in decision-making is known to lead to, at best, waste, and, at worst, harm to the public, including increased mortality. It’s great that more people with evidence synthesis skills will be available to ensure that when you are being offered a particular medical treatment in the future, it will reflect the safest, most comprehensive and up to date information available.
mhouston@irishtimes.com