complex systems complexity Public Health

Interventions to influence SYSTEM change. Complexity part 3

Following on from previous blogs part 1 on context & part 2 on actions, this third one follows a seminar delivered by Prof Rutter focusing on interventions to influence SYSTEMS.

At best, our institutions are only partially equipped to deal with complex problems

Our seminar focused oh how to deal with this and what to do next.

Many of the examples focused on obesity. The principles can be applied to anything.

1. Interventions within a system.

Strong tendency to consider one bit of the jigsaw vs the whole of the jigsaw. The line that “There’s no evidence this will solve the system problem”. Of course there isn’t, there won’t be as few places have done it + the intervention is 1 piece in a 400 piece jigsaw puzzle.

Can solve big complex problems if we have a powerful enough lever (eg We can solve obesity in 12m if we introduce food rationing)

When we do research we take one sandbag from a wall of sandbags and test whether it solve the problem we are trying to solve – floods. Mostly with the wrong form of research methodology to test the system problem

Often we operate in, and try to influence a very decentralised system, with no one actor in overall control. Often massive inertia to change, change is too difficult, counter to the vested interests in maintaining status quo. Thus either strong central control or regulation becomes the default desire.

2. Context and framing matters – context in which individually focused interventions takes place determines how successful they will be.

Weight watchers in the context of a high street full of chicken shops

Imagery drives stigma, stigma shifts blame – or perceptions of it – onto individuals. This drives a policy response that often makes it worse. Mental health patients committing crime, obesity.

The Rose hypothesis – Reducing salt consumption across whole pop will shift population BP by an amount less than the standard error of measurement of a BP cuff… but will prevent far more strokes than focusing on individuals. Same concept can be applied to pretty much any context

The dangerous olive of evidence. If we only focus on cost effective interventions we will do the wrong thing if the evidence base is loaded towards the wrong thing / individuals.

Interventions that are highly effective at individual level are completely irrelevant at population level. And vice versa – things that are irrelevant at individual level are massively important at population level.

How to change, challenge the methodological approaches that lead us in the wrong space

Evidence based policy vs evidence led policy

3. Framing matters

Aim of sugar tax isn’t to “solve obesity”, but to 1) create a level playing field, 2) set the rights context for other upstream interventions and 3) re internalise the externalities into the price.

This is a massively important framing of a policy point.

Getting the framing wrong will lead the policy implementation in the wrong way

Get the framing of the term “evidence” right

We’ve reached the limit of where simple easily generatable evidence is.

Need to develop different methodological concerns

4. Complex adaptive system.

a. Complex v complicated

Building a rocket is complicated. Linear chain of cause and effect. (Can land on the moon with significantly more precision that picking up your 15yr old daughter from a disco).

Peace in the Middle East presents a (perhaps THE) complex problem. COIN dynamics. There isn’t a single solution – many moving prices, no control. Interventions, political complexity, institutional capacity and capability, internal and external influences. All at once. Often with little to no quantification of the costs and benefits (or disbenefits) of competing options.

Any one thing you do will have an impact on other system dynamics. Can’t solve the problem. But can reshape some aspects of parts of the system to get to a different system response

b. Adaptive

SSB tax was thought to bring in £500m a year

But massive reformulation and industry adaption – which in itself is positive – was unintended.

This has affected implementation

Strong lobbying and industry funded front groups to undermine credibility of the policy, the evaluation. the lobbying was anticipated, the impact was unpredictable.

Each step in a chain will be attenuated – By events, by political and media discourse

Experiment. Introduce some stuff, test it. See how it goes

No plan survives contact with the enemy

Assume all assumptions are dynamic

5. What are the factors involved in change. Think about it.

Levers – what are they

Inflection points over long term, medium term, short term

What is the thing that creates the social will to change (Amsterdam approach to travel – 450 kids died in a year + fuel crisis. If we’d asked the Dutch to show effectiveness of cycle network in 1970s……The perturbation that builds political will,

What are the other social and political interventions that build will for change.

6. Where there is conflict or trade offs what to do.

Eg – Conflict between politics and science is greater for climate change than for anything else. (I don’t wholly buy it is all about politics, more about decentralised system, no control, inertia, vested interests)

Talk about trade offs, internalise the externality, do the maths from societal perspective. Shine a light on it.

Walk to school as an example of unwitting harm

Aim is to normalise walking, set up healthy life habits. But if we design an intervention that denormalises it – making them wear high vis tabard – this isn’t normalising, and might have counter effect.

Spin off of doing it right is to teach kids and parents to want safe, healthy streets and generate evidence base and political mandate for the right thing to happen

7. Time factor is often ignored.

Draughts – generally think 1 move ahead

Chess – good players think 10 moves ahead. But your opponent reacts so you need to adapt. 20 yr vision, 5yr strategy, 1yr plan

What % change can be achieved within a single political cycle. Over a long term, small % change year on year. Keep the long term aim in mind.

8. Backstage v front of house

Backstage still needs to be transparent and open

Shiny things happen on the stage. Be able to describe both

9. We need to be in the left hand side of the Swinburn diagram

But we are relentlessly pulled to the right side

Opportunities vs barriers. Focus on the opportunities. High quality individuals that can exert leverage in a system

No simple answer. Deal with it

Political capital and evidence weight increases as we go further downstream

Hostile scrutiny by media of political and officer actions makes the wrong thing more likely. Puts politicians in an impossible condition

The obesity foresight system map is loaded on the food side in terms of interventions

The reason for this is that more people that turned up to the food workshop and more post it notes were put on the table

Evidence base and practice is structurally biased towards short term impacts of tightly defined individual level interventions. This drives a response aimed at proximal risk factors and ignores the population point of Rose

10. Insights and foundational actions to change SYSTEMS

People adopt a normal response to the environment & other cues around them.

Problem framing matters. Individual will power vs inevitable response to environmental cues.

the framing of the problem is critical.

Target interventions – Political and public discourse around market failure, correcting externalities. Is equally valid than a simplistic reductionist model. It’s kind of like building foundations for the right things to happen in the future. But there will never be an evidence base to support this.

The time dimension is often ignored. Focus on 1, 5, 20yr plan.

No plan survives contact with the enemy and your Assumptions are dynamic

Backstage v front of house. Make sure you can tell a story on both.

Building the will and narrative for other forms of interventions. Create will and foundations for social change

Complex systems approaches can help us through some of these issues

Chess not draughts. Think 10 moves ahead, with an end game in mind.

What others have said on SYSTEM interventions

Donella Meadows wrote beautifully on points of intervention in systems decades ago. It’s still relevant today.

Most powerful influencers and points of intervention

1. The power to transcend paradigms

2. The mindset or paradigm out of which the system—its goals, structure, rules, delays, parameters—arises

3. The goals of the system

Least powerful

9. The lengths of delays, relative to the rate of system change

10. The structure of material stocks and flows (such as transport networks, population age structures)

11. The sizes of buffers and other stabilizing stocks, relative to their flows.

12. Constants, parameters, numbers (such as subsidies, taxes, standards)

This was developed further by @DTFinegood into an Intervention Level Framework – five levels

Table 1 of Malhi is also useful

Collective reshape the ask – focus on systems that set the context for setting the framework for the right interventions

3 replies on “Interventions to influence SYSTEM change. Complexity part 3”

Thanks Greg – the best blog yet Andrew Rix Independent Research and Evaluation Consultant Honorary Researcher Swansea University School of Medicine

2 Brynau Rd Cardiff CF15 7SA mob:07778057927 home 02920 214374

*Gmail automatically scans emails and attachments for viruses when sending and receiving. *

Liked by 1 person

Thanks very much for your blogs on this. I like the translation into something practical. Raises questions about how many change initiatives should be going on (far too many, the urge to act overwhelms the need to reflect). Also, about QI – too much focuses on the event and structure view of the system not on the mental models. Oh we’ll just get all the surgeons to comply with the guidelines….. Its why we need to integrate all the change people – OD, QI, transformation, project management and see how they learn to make change happen. Although on the other hand, people can only change things within the scope of their power. Hmmm….

Also really shapes what you do at a policy level. If every system is uniquely complex, who can say what change you need to implement? I guess focus on outcomes helpful here.

I really like your rethink of upstream/downstream. Also, current narrative tends to think of complexity for work that is happening across multiple organisations (and what that means for leadership hence all the systems leadership stuff going on) when in fact if its a socio-technical system at any scale then its complex or if you can’t find the answer on Google.

I think that far more energy and time needs to go into the discovery and design of system interventions and we still need systematic approaches to implementation of systemic theories of change.

Research makes systems unreal. Gives you some interesting data but no substitute for the real world context and testing what works incrementally.

Finally, systems thinking has pushed me to think much more about what goes on in the spaces between, the connectivity and what you don’t see when you break things apart. The sum of the parts really is far far greater than the whole. Explore the distinctions people use and see what they are part of, a wider whole.

Am trying to apply some of this thinking to implementing a wellness model in a health economy.


Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s