Skip to content

tomasaragon/cdph

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

80 Commits
 
 
 
 

Repository files navigation

CDPH is building a Decision Intelligence Unit

Decision Intelligence

Tomás Aragón, MD, DrPH
State Public Health Officer and Director
California Department of Public Health
[email protected] (email)
https://drtomasaragon.github.io/ (blog)

Decision-making is our most important activity.1 "A decision is a choice between two or more alternatives that involves an irrevocable allocation of resources." Decisions drive vision, strategy, policy, and transformational change. Every decision has causal assumptions, predictions, trade-offs, and an opportunity cost—the lost benefit of the better option(s) not chosen or not considered. Some decisions have extreme time constraints under deep uncertainty.

Decision intelligence is using ethics, science, and technology to improve individual and team decisions for finding and solving problems, and achieving objectives and key results in challenging, including VUCA,2 environments (Figure 1).

Decision Intelligence

The CDPH Office of Policy and Planning will have a Decision Intelligence Unit (DIU) to develop, implement, and improve strategic decision-making methods for high stakes, high cost, and high impact public health decisions. Methods will include cost-benefit and cost-effective analyses, and decision analysis. These positions have permanent funding. We will secure grants and develop collaborative partnerships with academic institutions. The major areas of initial focus, development, and collaboration will be

  • Decision making under uncertainty, including deep uncertainty (ie, risks not calculable)
  • Methods to optimize policy and regulatory decisions; conduct cost-effectiveness and cost-benefit analyses; and program budgeting and marginal analysis (PBMA)
  • Computational algorithms for decision making (using Bayesian networks (including decision networks [influence diagrams]), Markov Decision Processes (MDPs), Partially Observable Markov Decision Processes (POMDPs), Reinforcement Learning, Agent-based modeling)
  • Leaders/managers as decision architects:3 Design and improve of team decision meetings that incorporate cognitive psychology and behavioral economics

The DIU will have a trans-disciplinary team:

  • Health economist with expertise in population health cost-benefit and cost-effectiveness analyses
  • Behavioral economist with expertise in decision implementation science
  • Computational decision scientist with expertise in programming (Julia, Python, R)
  • Population health data scientist
  • Biostatistician

The DIU is part of CDPH's longer term strategy to build our capability in population health data science (Table) and to move us to Levels 4 and 5.

Level Analysis Description
1 Description a. surveillance and early detection of events
b. prevalence and incidence of risks and outcomes
2 Prediction a. early prediction and targeting of interventions
3 Explanation4 a. discovery and testing of new causal pathways
b. estimation of intervention efficacy/effectiveness
4 Simulation a. modeling for epidemiologic or decision insights
5 Optimization5 a. optimizing decision, effectiveness, or efficiency metrics

Here is a graphical depiction of the Table.6 We aim to embrace our "reasoning" to make better decisions.

Population Health Data Science

To learn more about books relevant to decision intelligence, here is my list.

Appendix

Here is an alternative, but very informative, view of decision intelligence (see two figures below):7

Footnotes

  1. Tomas Aragon (2021). Decision intelligence: An Introduction. Available from: https://drtomasaragon.github.io/posts/2021-03-13-decision-intelligence01/

  2. volatile, uncertain, complex, and ambiguous

  3. John Beshears and Francesca Gino (2015). Leaders as Decision Architects: Structure your organization’s work to encourage wise choices. Harvard Business Review. Available from: https://hbr.org/2015/05/leaders-as-decision-architects

  4. Causal inference. See Judea Pearl. The Book of Why: The new science of cause and effect. Penguin 2019

  5. Decision analysis, cost-effectiveness/benefit analysis, mathematical modeling, operations research, etc.

  6. Source: https://www.bayesia.com/articles/#!bayesialab-knowledge-hub/1-introduction

  7. Source: https://wequity.tech/decision-intelligence-and-design-thinking/

About

CDPH public documents

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published