Mean elevation of inhabited areas reveals that coarse district-level analyses overestimate the altitudinal shift of dengue in Nepal.

The potential altitudinal expansion of dengue fever under climate change has received increasing attention. In Nepal, rising dengue incidence across all districts has been interpreted as evidence of spread into higher elevations. However, such conclusions are often based on district-level aggregated data, which can misrepresent transmission patterns in a highly heterogeneous mountainous landscape. District mean elevation frequently includes extensive uninhabited high-altitude areas, leading to overestimation of the elevation at which dengue transmission occurs. We combined Shuttle Radar Topography Mission (SRTM) elevation data with gridded population datasets to derive mean elevation of inhabited area, providing a more realistic proxy of inhabited environments. District-level dengue data (2006-2025) in Nepal were analyzed using both conventional mean elevation and mean elevation of inhabited areas. Substantial discrepancies were observed between these measures. For example, Kaski District shifts from 2608 m (rank 19) to 987 m (rank 49) when population weighting is applied. Overall, population distribution is strongly concentrated in river valleys at lower elevations, with a sharp decline above 2000 m. When population distribution is accounted for, the apparent expansion of dengue into high-elevation districts is markedly reduced indicating previously reported altitudinal shifts of dengue in Nepal are partly driven by methodological artifacts of coarse spatial aggregation. Our results highlight a critical methodological issue in analyzing climate-sensitive diseases in mountainous settings and suggest that mean elevation of inhabited areas approaches provide a more accurate representation of exposure and hence should be prioritized.