Places to intervene in a health system

From SystemsWiki

Jump to: navigation, search

Health Examples of Systems Archetypes


Intervention in one place in a system to fix a problem

  • symptom vs complex causal network?)
  • Linear progress slows over time (specialty vs generalist care technical vs pt experience)
  • Fix creates a problem elsewhere (eg outliers, diverting resources and demand from primary, aged, social and community care)
  • The fix overshoots the goal Medical workforce boom bust cycles
  • Things seem to oscillate endlessly (wider swings by close coupling eg app for wait times)
  • In time the problem returns
  • Underlying cause not being addressed (dynamic capacity balancing long AND short term investments, dynamic systems view)
  • Dependence on the fix develops (ED for After hours Care)
  • Over time there is a tendency to settle for less (long waits or poor quality)

Meanwhile others are trying to promote growth

  • Growth leads to decline elsewhere (primary care, trauma replaced by aged acute on chronic (eg medications))
  • Partners for growth become adversaries (hospital spec vs ED spec and generalists)
  • Growth slows over time
  • Limited resources are shared by others
  • More than one limit to be addressed ?
  • The limit is insufficient capacity (flexible, dynamic based on capability reconfig)
  • Tendency to let standards slip
  • See Systems_Archetypes_Relationships for Video and more references

Places to Intervene in a System

Applying the Systems Approach

  • See Applying_the_Systems_Approach
  • Limitations of Analytic Approaches:
  • Need for Realist evaluation Context Mechanism Outcomes
  • Multiscale dynamic modelling and simulation (Rosie)
  • Systemic and dynamic multiscale perspectives CSH CST
  • Potential Problems How might this situation go wrong?

ED Overcrowding Interventions

  • IM-6167 Unintended effects of reducing time in ED
  • IM-1010 Hospital fixes that fail

Policy resistance or fixes that fail

Controlling health care expenditure. Goals of subsystems are different from and inconsistent with each other. One persons expenditure is another person’s income. Moves and countermoves produce a standoff with little change. Everyone is pulling in different directions. You can try to overpower policy resistance, if you can gain and keep enough power by centralised control over resources. But it builds up resentment and possible explosive backlash if the power is ever weakened. The counterintuitive approach is to let go and give up ineffective policies in order to be able to find time to align the various goals of the subsystems. E.g. patient focussed care in hospitals can let go of more narrow goals of patient throughput and budget control and consider the long term welfare of the entire system.

Tragedy of the commons

Bounded rationality. I fight hard to grow my staff and resources. But there is a limit to the overall budget or capacity to pay. Missing or too late feedback from the resource to the growth of users of the resource. Widespread availability of antibiotics grew the uses generating antibiotic resistance, with more powerful strionger antibiotics used reducing the comon shared resource of antibiotic senstive bacteria. Overuse continues resource is destroyed and all the users are ruined. A commons system makes selfish behaviour much more convenient and profitable than behaviour that is responsible to the whole community. It can be avoided by education and exhortation, privatising the commons if it can be or regulation. Mutual coercion, mutually agreed upon. Gains and losses fall on the same decision maker. Fines and rewards. Need to be informed strong and represent the common good.

Drift to low performance

We know what we ought to do, for some reason we just don’t do it. Eroding timeliness of service in the ED. Or accepting poor diet and lack of exrecise. Or boiled frog syndrome. Eroding goals over time people cleared from ED by the end of the shift accept htem taking more than one shift. Adjust expectations little by little so expect conflict at work, waiting in ED, mistakes to just happen Success = Perceptions minus expectations Keep performance standards absolute. Even better, let standards be enhanced by the best actual performance instead of being discouraged by the worst. Optimism can be The better things get the harder Im going to work to make them even better.


Each actor takes its desired state from the other’s perceived state and ups it. Wages and conditions Resarcher to gain public attention with miracle breakthroughs. They expect it and it becomes commonplace Hospitals outdoing each other with expensive diagnostic machines. Can lead to extremes very quickly. Can refuse to compete or negotiate/regulate to control escalation Success to the Successful High tech medicine vs humanising care and bedside manner. Specialists crowding out the generalist or primary care. Inverse care law. Rich get more services poor get less. Progressive taxation to fund universal health care; donation of time and money. Diversification allows losers to get out of the game and start a new one. Holistic care in primary care. Strict limits on the amount that anyone can win. Policies that level the playing field such are knowledge sharing

Shifting the burden to the Intervenor – Addiction

(see p135 primer summary description) Dependence on government to fix our health through intervening doctors and medicines rather than the practices and lifestyles of each individuale.

Taking drugs to enhance perception of wellbeing. Rescue fosters depencence on the intervenor Care of the aged has shifted from families to health and aged care. Now families no longer have the space, time skills or willingness to care for their elderly members. Confronting the real state of the system and taking the actions the addiction allowed us to put off. Beware the symptom relieving or signal denying policies or practices that don’t really address the problem. Take the focus off short term relief and put it on long term restructuring

Rule beating

Rules to govern a system can lead to rule beating – perverse behaviour that gives the appearance of obeying the rules or achieving the goals but that actually distorts the system. Patient throughput spending the budget reprting performance Documenting rather than caring for/communicating with the patient Need to redesign the rules to release creativity in achieving the purpose of the rules

Seeking the wrong goal

Be careful what you ask systems to produce. Patient throughtput producing effort not result by seeking the wrong goal because the goals are defined inaccurately or incompletely. The indicators and goals need to reflect the real welfare of the system

Questions & Comments to Geoff McDonnell
Personal tools