System Properties

1
Session 8
2
Why systems analysis

• Integration rather than reduction
• Cannot predict output by separate analysis of
  individual elements
• Understand interactions
• Eliminate unimportant factors
• Make informed predictions
• Propose changes, remediation to influence health
• Allow for powerful visual presentations

3
System Properties

• Nonlinear
• Stochastic (non deterministic)
• Multifactorial interrelationships between
  factors
• e.g. hurricane/resources/policy
• e.g. spending 75 of income on health leads to
  greater poverty
• Multidimensional (health/wealth)
• e.g. changes not along curve but curve changes
  position what is the extra dimension?

4
Different vantages

• Hierarchical
• Individual family neighborhood city -
  -region global
• Each hierarchical level may interact with other
  levels
• Differences between developed and developing
  countries
• Each level influenced by personal and social
  behavior, cultural and spiritual background,
  education
• Each level influenced by external factors
• Factorial, time dependent
• Nutrition, obesity, water, waste, crime,
  accidents, war etc.
• Factors at each level may be different
• Influenced by historical information and
  perspective

5
Requirements

• What other analyses are there
• Can they assist in our task?
• Can they be integrated into ours
• How not to duplicate other efforts?
• Sensitivity analysis
• Which are the important factors (may not be
  obvious from the start)

6
Next?

• What to focus on
• Agree on definition of health
• Agree on initial set of parameters to be included
• Agree on basic needs for health
• Agree on goals of program, on target audience
• Factors that can make significant change to
  improve health
• Factors that can influence (policy makers,
  practitioners)
• Do we have quality data? What data can be
  included in a systems approach? What are the
  gaps, what new data should we seek?
• Outcome measures (monitoring)
• Have we made a difference