Temperature and place associations with Inuit mental health in the context of climate change

Background: Climate change has important implications for mental health globally. Yet, few studies have quantified the magnitude and direction of associations between weather and mental health-related factors, or assessed the geographical distribution of associations, particularly in areas experiencing rapid climatic change. This study examined the associations between air temperature variables and mental health-related community clinic visits across Nunatsiavut, Labrador, Canada, and the place-specific attributes of these associations. Methods: Daily de-identified community clinic visit data were collected from the provincial electronic health recording system and linked to historical weather data (2012-2018). A multilevel, multivariable negative binomial regression model was fit to investigate associations between temperature variables and mental healthrelated community clinic visits across the region, adjusting for seasonality as a fixed effect and community as a random effect. A multivariable negative binomial model was then fit for each Nunatsiavut community, adjusting for seasonality. Results: Mental health-related visits contributed to 2.4% of all 228,104 visit types across the study period; this proportion ranged from 0.6% to 11.3% based on community and year. Regionally, the incidence rate of mental healthrelated community clinic visits was greater after two weeks of warm average (i.e. above -5 degrees C) temperatures compared to temperatures below -5 degrees C (IRR-5 <= 5 degrees C = 1.47, 95% CI = 1.21-1.78; IRR6 <= 15 degrees C = 2.24, 95% CI = 1.66-3.03; IRR>15 degrees C = 1.73, 95% CI = 1.02-2.94), and the incidence rate of mental health-related clinic visits was lower when the number of consecutive days within -5 to 5 degrees C ranges (i.e. temperatures considered to be critical to land use) increased (IRR = 0.96; 95% CI = 0.94-0.99), adjusting for seasonal and community effects. Community-specific models, however, revealed that no two communities had the same association between meteorological conditions and the incidence rate of daily mental health-related visits. Discussion: Regionally, longer periods of warm temperatures may burden existing healthcare resources and shorter periods of temperatures critical to land use (i.e. -5 to 5 degrees C) may present enjoyable or opportunistic conditions to access community and land-based resources. The heterogeneity found in temperature and mental health-related clinic visits associations across Nunatsiavut communities demonstrates that place quantitatively matters in the context of Inuit mental health and climate change. This evidence underscores the importance of place-based approaches to health policy, planning, adaptation, and research related to climate change, particularly in circumpolar regions such as Nunatsiavut where the rate of warming is one of the fastest on the planet.