“The notion of applying simple linear and reductionist methods to find answers to complex public health issues disappears once you start talking about prevention on a grand scale,” Professor Kelly told members of our Research Network (early- to mid-career researchers).
Professor Kelly was Director of the Centre for Public Health at the National Institute of Health and Care Excellence (NICE) from 2005–2014. In this role, he applied systems thinking to developing public health guidelines covering complex issues such as the obesity epidemic.
He cautioned that this approach was far from easy and required courage to face the complexity of the problem.
”It is a mixture of rocket science and brain surgery … don’t be frightened of that,” he said. “As scientists and as co-producers of knowledge, we do our best to understand how it works. It is a gradual process, not one where simple solutions will manifest themselves quickly.”
Professor Kelly warned against ‘fast thinking’1 and the tendency to default to simple solutions. “When we default to fast thinking we are more likely to repeat old errors,” he said.
An alternative option was to stop and question your assumptions or bias about an action, situation or policy.
He said it was important to understand the deeply beliefs and values under which a system usually operated before attempting to intervene, as these were the hardest to change but often the most effective.
He experienced this first hand in his role at NICE, where he led the development of guidelines on how to help pregnant women quit smoking.
Even though the knowledge and evidence about dangers to the developing foetus and woman were incontrovertible, the midwives and health professionals believed that confronting pregnant women about their smoking would destroy existing relationships.
The problem was explored in greater detail to discover that midwives were trained in general communications but not in difficult communications. The solution: incorporate difficult communications into the training.
“We sometimes assume that professionals and researchers have similar if not the same goals but in reality it is often different,” Professor Kelly said. The falling out between the two parties forced the change from the normal state of the system.
Another benefit of a systems approach was embracing failure: looking for what doesn’t work; being “anti-fragile”; and ultimately building resilience.
“If we can be more humble and accept both error and failure, we can then consider a broader spectrum of likely outcomes from any other model,” he said.
- Kahneman D. Thinking, Fast and Slow. Penguin Books, 2012.