Causal Loop Diagram

Causal loop diagrams map the causal relationships between pairs of elements within a system and identify feedback loops. These loops can either be reinforcing (vicious cycle) or balancing (goal-seeking) and complex interactions between loops can lead to unintended consequences.

The arrows in the diagram describe the directions of effect. A causal link from one element (A) to another (B) is positive (+) when a change in A leads to change in B in the same direction; an increase in A leads to an increase to B and a decrease in A leads to a decrease in B. Conversely, a causal link from A to B is negative (-) when a change in A leads to a change in B in the opposite direction. Figure 1 describes dynamic factors affecting dispensing errors in a pharmacy setting. One reinforcing loop in red reveals that higher schedule pressure increases the number of dispensing errors, which leads to more rework to be done and schedule pressure eventually spirals up. On the other hand, a balancing loop on the left-hand side in blue reveals that schedule pressure can increase productivity, which then leads to less work to be done and decreased schedule pressure.

 

CLP-pharmacy
Figure 1 Causal Loop Diagram of Dispensing Errors (adapted from Shire, 2017)

By understanding dynamic interactions between loops, causal loop diagrams allow us to have a more in-depth understanding of interdependencies underpinning a control structure and impact of changes.

Causal loop diagrams can never be comprehensive and are also never final, but always provisional. The diagrams evolve as analyst’s understanding improves and the purpose of modelling effort evolves.

Causal loop diagrams can be used as part of System Dynamics modelling and simulation approach, which additionally use stock flow diagrams to quantitatively model and simulate system’s dynamic behaviour. The simulation allows various scenarios to be tested.

Causal loop diagrams are conceptually simple, but not easy to apply and use without experience/support. Causal loop diagrams have been used to evaluate unintended effects of policies in various domains, e.g. public health (Leveson et al., 2012), patient safety (Guo et al., 2013), mining (Goh et al. 2012), military accident (Minami et al., 2009) and construction safety (Han et al. 2014).

References

  • Guo, S., Roudsari, A., Garcez, A., 2013. A Causal Loop Approach to the Study of Diagnostic Errors. Studies in Health Technology and Informatics. 2014;205:73-7.
  • Goh, Y.M., Love, P.E.D., Brown, H., Spickett, J., 2012. Organizational Accidents: A Systemic Model of Production versus Protection. Journal of Management Studies. 49, 52–76.
  • Han, S., Saba, F., Lee, S., Mohamed, Y., Peña-Mora, F., 2014. Toward an understanding of the impact of production pressure on safety performance in construction operations. Accident Analysis and Prevention. 68, 106–16.
  • Leveson, N., Couturier, M., Thomas, J., 2012. Applying system engineering to pharmaceutical safety. Journal of Healthcare Engineering, 3 (3), 391-414.
  • Minami, N. a., Madnick, S., 2009. Dynamic analysis of combat vehicle accidents. System Dynamics Review. 25, 79–100.
  • Shire, M., Jun, G. T., Robinson, S., under review, The Application of System Dynamics Modelling to System Safety Improvement: present use and future potential, Safety Science.