Without the right coordination, health reforms will be manoeuvres in the dark

As a part-time musician, I’ve always been fascinated by how other musicians get their act together to produce something greater than the sum of its parts.

Consider two of the musical giants of the 20th century, James Brown and Miles Davis. Both could call on phenomenally talented players, but had completely different ways of getting the best out of them.

Subjected to James Brown-type command and control, most musicians would probably say, “What’s the point? This isn’t fun any more”

In putting together the landmark recording Kind of Blue, Miles Davis gave minimal direction to his collaborators, who included John Coltrane, Bill Evans and Cannonball Adderley.

The magic happened as they worked out how to collectively use this vast space for experimentation.

In contrast, James Brown knew exactly the result he wanted. In order to get it, he imposed a rigid level of discipline and control over his players, such as fining them if they hit their note a fraction late.

These diametrically opposite approaches both produced timeless masterpieces.

However, most mortal musicians wouldn’t do well in either scenario.

Without the musical chops to do what Davis’ band could, maximum autonomy usually descends into a rabble.

And, when subjected to James Brown-type command and control, most musicians would probably say, “What’s the point? This isn’t fun any more.”

Dilemmas about how to coordinate actions are relevant at any scale of human activity. Such dilemmas have long been of interest to sociologists, anthropologists, economists and political scientists studying how organisations work internally and with one another.

Whatever the starting point, the typical options boil down to three basic archetypes: hierarchy, markets and networks.

Each constructs the social world in a particular way.

Hierarchy entails relationships between bosses and subordinates.

Markets consist of buyers or sellers.

Networks require collaborators who bring different types of knowledge and expertise, and a level of trust between them.

Hierarchical coordination works really well when there is a known recipe for achieving an objective that can be broken down into clear components, allocating responsibility for each component.

Infection control protocols in surgical wards are a good example.

But hierarchy also can produce the well-known side effects of rigidity and inflexibility.

Markets and networks tend to be far more flexible, as neither requires a central, coordinating brain. But markets entrench inequalities of power, and collaborative networks are prone to opportunism and can just as often dissolve into talkfests with few tangible results.

Social coordination is also very “sticky”.

Most of us are not adept at suddenly pivoting from being a boss or subordinate to being a colleague on equal footing in the context of a particular relationship.

The same is true for organisations.

Relationships are built up from habits of interactions, referred to by social scientists as “institutions”, and habits can be awfully hard to change, even when the world around us changes.

This stuckness can go both ways.

Network coordination based on trust is deeply rooted in how health professionals relate to one another, but institutions based on trust are often weak or late at picking up bad behaviour (as with serial killer Harold Shipman, who had been a much-loved GP).

On the other hand, the combination of hierarchical and market coordination that underpinned the redesign of New Zealand government in the 1990s is manifestly unhelpful when it comes to dealing with complex and wicked problems, in which tasks cannot be neatly divvied between people, or tightly specified in contracts.

It’s tempting to think that all we need to do is know what type of activity needs coordinating, then work out which type of coordination to use. Maximum collaboration and autonomy is unlikely to produce “Get On the Good Foot”, and maximum command and control will never produce Kind of Blue.

But it’s often difficult to distinguish between contexts best suited to particular types. In many parts of the health system, we often witness a cycling between different types of coordination.

Take the example of after-hours medical services in Auckland.

In 2010, the approach was highly contractual, based on hierarchical and market coordination, but all parties were dissatisfied.

DHBs thought they were not getting what they were paying for, and providers (PHOs and general practices) knew the contracts were too inflexible to deal with the shortage of health professionals at the time.

From 2011 to 2015, a switch was made to network coordination, with lots of negotiations and creative problem-solving.

The downside was an enormous commitment to meetings and the inevitable exclusion of some players from the network.

Since 2016, the DHBs have reverted to a more traditional procurement approach.

We can’t assert that any type of coordination is inherently superior.

Health-system bandleaders just have to choose which problems they can live with, manage the inevitable tensions between the imperatives of control and flexibility, and avoid the worst pathologies of each type.

This brings me to another reason I am so dissatisfied with the final report of the Health and Disability System Review Panel.

There is precious little awareness or acknowledgement that these tensions and dilemmas around coordination even exist.

I read the report’s recommendations on how to steer the health system, as akin to saying we want health-system virtuosos of the calibre of John Coltrane, but we are going to put James Brown in charge.

What could possibly go wrong?

Tim Tenbensel is associate professor, health policy, in the School of Population Health at the University of Auckland