Heat adaptation measures in private households: an application and adaptation of the protective action decision model

Climate change is increasing the frequency and intensity of heatwaves, posing a significant threat to human health, particularly in urban areas experiencing urban heat island effects. The elderly are especially vulnerable. While research highlights risk perception, age, gender, income, and education as influencing factors in heat adaptation behavior, protective actions in German households remain low. Existing frameworks like the Theory of Planned Behavior (TPB), Health Belief Model (HBM), Value-Belief-Norm (VBN), and Protection Motivation Theory (PMT) have been applied but often lack the inclusion of external stimuli like temperature exposure. This study uses the PADM, which incorporates external stimuli, to investigate heat adaptation behaviors in private German households, aiming to identify key influencing factors and guide targeted communication strategies.

Several theoretical frameworks have been used to study heat or climate change adaptation. The TPB considers attitude, perceived behavioral control, and subjective norms as determinants of intention, leading to behavior. The VBN focuses on values, beliefs, and norms, while the HBM emphasizes perceived susceptibility, severity, and threat. The PMT combines threat and coping appraisals to explain protective action. However, these models often lack the incorporation of external environmental stimuli. The PADM, successfully applied to other hazards, addresses this limitation by considering hazard exposure in addition to individual factors like efficacy-related and resource-related attributes. This study applies the PADM to the context of heat adaptation, filling a gap in the literature.

This study used a mixed-methods approach in Augsburg, Germany. Data was collected from July 1 to September 1, 2019, during which the highest temperature reached 34.4°C. Over 500 households participated, providing indoor bedroom temperature data via data loggers (accuracy ±0.5°C) and completing questionnaires assessing heat risk perception, subjective heat stress (SHS), knowledge about heatwaves, and adaptation behaviors (e.g., using air conditioning, installing shadings, using fans). The questionnaires also gathered sociodemographic information (age, gender, income, employment, etc.). Quantitative data analysis involved one-way ANOVAs, linear regression analyses, and hierarchical regression modeling to test hypotheses derived from the PADM. Heat adaptation behavior was operationalized as a composite score from responses about adaptation measures and behaviors.

Descriptive statistics revealed that 15.5% of participants were 65 years or older, 60.6% were female, and 58.9% held a university degree. Analysis revealed that: 1. **Knowledge significantly impacted heat adaptation behavior:** Individuals with higher knowledge levels demonstrated a greater likelihood of implementing adaptation measures. 2. **Heat risk perception was a significant predictor of heat adaptation behavior:** Higher perceived risk was associated with more frequent adaptation actions. 3. **Age significantly influenced heat risk perception:** Younger participants exhibited higher heat risk perception than older participants. 4. **Contrary to expectations, efficacy-related attributes (self-efficacy), income, and employment did not significantly predict heat adaptation behavior.** 5. **Hierarchical regression analysis revealed that indoor temperature, subjective heat stress (SHS) at home, and experienced health implications during heatwaves were significant predictors of adaptation behavior, along with age and self-efficacy.** The model explained 21.7% of the variance in heat adaptation behavior. The adjusted PADM shows that higher indoor temperatures and higher SHS lead to more adaptation.

The findings partially support the PADM, highlighting the importance of knowledge and heat risk perception in influencing adaptation behavior, consistent with prior research. The lack of significant influence from income and employment suggests that adaptation is not solely determined by socioeconomic factors, implying that communication strategies should target a broad range of demographics. The unexpected finding that indoor temperatures did not directly influence heat risk perception challenges the initial PADM conceptualization. The strong influence of experienced health implications and SHS highlights that personal experience significantly motivates adaptation. Age, negatively correlated with adaptation, suggests a focus on communication strategies tailored to older individuals.

This study demonstrates the applicability of the PADM to understanding heat adaptation behavior, though requiring modifications based on the findings. Knowledge, risk perception, age, efficacy, indoor temperature, SHS, and health implications are key factors. Future research should replicate this study across different cities and cultures, and refine the knowledge construct to enhance its reliability. Targeted interventions based on these factors can enhance heat adaptation and reduce heat-related health risks.

The study's limitations include its focus on Augsburg, Germany, potentially limiting generalizability. The voluntary nature of participation might introduce selection bias. The online survey's timing may have affected responses, and the lack of a pre-test for some constructs is a limitation. Furthermore, the Cronbach's alpha values for some constructs were lower than ideal, suggesting potential measurement error or issues with instrument design; specifically, the fragmented knowledge regarding heat risks might have caused a low Cronbach's alpha for the corresponding construct. Further research should address these limitations by conducting the study in other locations and refining data collection methodologies.