Exploring Behavior and Other Risk Factors in Vulnerable Populations during an Extreme Heat Event through Use of Participatory Systems Dynamics Models - CANCELLED (June 17)

BACKGROUND: Heat-related morbidity and mortality can be prevented by protective heat health behaviors yet barriers exist to their implementation.  In particular, the behaviors and attitudes of vulnerable populations are key, yet the relative importance and interplay of social and physical factors in an extreme heat event (EHE) are poorly understood.  Systems dynamics modeling is a methodology for users to explore the relative contributions of physical, economic, and social factors to heat health and identify important intervention points. Systems modeling can be an effective method of participatory exploratory research that engages the community and aids in understanding complex relationships.

METHODS:  

The Mid-Michigan Heat Model (MMHM) is an exploratory systems model of social and biophysical dynamics during an EHE.  It was created using STELLA software and calibrated with published data on heat deaths and hospitalizations from the 1995 Chicago heat wave. Other model variable relations were derived from behavioral surveys among 3,000 low-income residents of Washtenaw County and Lansing, Michigan, and from key informant interviews of eight local public health officers, state health and environmental officials, and academics. MMHM was constructed as a hypothesis development tool with these key inputs in the interface:  number of people at risk (0-5 million); proportion of population with air conditioning (0-100%); a ‘brownout’ reducing electrical power (yes/no); number of cooling centers (1-20); proportion of cooling centers accommodating pets (0-100%); proportion of population receiving heat media messaging (0-100%); urban heat island effect reduction (0-4.3 ^oF); and access to public transportation (0-100%).  MMHM was demonstrated at a workshop where participants manipulated the model and provided feedback on its utility in facilitating thinking, understanding and discussion of factors influencing heat health outcomes during EHEs.

RESULTS:   The simulation assumed 60 deaths and 64 hospitalizations among a population of 820,000 during an EHE similar to Chicago’s heat wave. Sensitivity analysis identified brownout occurrence, the proportion of homes with air conditioning and a media campaign to have the largest effect on health outcomes, while the number of cooling centers, whether they accommodated pets, and providing transportation had no measurable effect. Workshop participants found manipulating the model helpful in exploring the relative impacts of EHE inputs and were comfortable with the results, which were intended to facilitate exploration and discussion of interrelationships. Further work is needed before MMHM can be used for quantitative prediction of EHE inputs and outputs.

CONCLUSIONS: Participatory systems modeling is an effective tool for understanding EHEs and identifying key intervention points.