Heat waves raised kidney disease deaths across 2,790 Chinese counties by 3.5%

Heat and the kidney have an obvious mechanical relationship — dehydration, reduced renal perfusion, electrolyte disturbance — and a surprisingly thin evidence base at population scale. Most heat-mortality research reports all-cause or cardiorespiratory deaths, where chronic kidney disease is folded into a larger total.

A study in the April issue of Biomedical and Environmental Sciences separates it out, using one of the few mortality datasets large enough to support the analysis [s1].

What was analysed

The authors used chronic kidney disease mortality records from 2,790 counties and districts across China for 2004 to 2022, collected by the Chinese Center for Disease Control and Prevention, paired with meteorological data from the fifth-generation European Reanalysis Land dataset for the same period [s1]. In total, 236,260 CKD deaths were included [s1].

The design was a time-stratified case-crossover combined with a distributed lag nonlinear model [s1]. Case-crossover designs compare each death against control days for the same person in the same month, which controls by construction for anything that does not change within a person over weeks — age, sex, comorbidity, socioeconomic status, chronic exposures. It is a strong design for short-term environmental triggers, and a weak one for anything that acts slowly.

What was found

Compared with non-heat-wave days, CKD mortality increased by 3.48% during heat waves (95% CI 1.67% to 5.33%) [s1]. Within heat-wave days, the risk escalated by a further 2.48% for each additional 1°C (95% CI 0.12% to 4.91%) [s1].

The second figure is the more fragile one. Its lower confidence bound is 0.12% — barely clear of zero. The temperature-gradient effect is directionally consistent with the main finding but is not precisely estimated.

Three subgroups showed higher excess risk [s1]:

• Women: 5.52% (95% CI 2.71% to 8.40%)
• People aged 65 and older: 4.60% (95% CI 2.30% to 6.96%)
• People in mesic and cold regions: 6.20% (95% CI 1.13% to 11.53%)

The third is the counterintuitive one, and it is the most policy-relevant result in the paper. The largest excess risk was not in China's hottest regions. It was in its cooler ones.

Why cooler places fare worse in heat

This pattern recurs across heat-mortality literature, and the usual explanation is adaptation: places where heat waves are rare have less air conditioning, less acclimatised physiology, weaker heat-health warning systems and populations less practised at responding. A given absolute temperature is a larger departure from normal, and departure from local normal is what kills.

The study reports the association; it does not test the mechanism, so the explanation above should be read as the standard interpretation rather than as something this analysis established.

The projection

The authors also projected future CKD mortality attributable to heat waves under climate and population scenarios. Under SSP5-8.5 — the high-emissions pathway — the attributable fraction of CKD mortality rises from 0.64% in the 2020s (95% CI 0.52% to 0.78%) to 2.44% in the 2090s (95% CI 1.97% to 2.95%) [s1].

The heaviest projected burden falls on southeastern China: Hainan at 3.31% (95% CI 1.66% to 5.02%), Yunnan at 3.05% (95% CI 1.46% to 4.75%) and Guangdong at 2.84% (95% CI 1.24% to 4.41%) [s1].

Projections of this kind carry compounding uncertainty that the confidence intervals do not capture. They hold the exposure-response relationship fixed across seven decades, which assumes no adaptation — no additional air conditioning, no improvement in heat warning systems, no change in CKD treatment. Given that the study's own subgroup finding suggests adaptation strongly modifies risk, holding it constant is a conservative assumption in one direction and an implausible one in another. SSP5-8.5 is also the highest-emissions scenario, not a central estimate.

Read the projection as a statement about what happens if nothing changes, which is the only thing a projection of this design can say.

What it adds

Three things.

First, it puts a number on a cause of death that heat research usually leaves inside an aggregate. Chronic kidney disease is not a rare outcome — 236,260 deaths in one country's surveillance system over 19 years — and a 3.48% excess on heat-wave days across a population that size is not a small absolute count [s1].

Second, the subgroup structure identifies who to protect: older adults, women, and residents of regions that do not think of themselves as hot [s1].

Third, it is a reminder that the health cost of warming does not distribute the way intuition suggests. The provinces projected to carry the heaviest attributable burden are in the southeast [s1], but the sharpest per-event excess risk sits in the cooler regions [s1] — a distinction between total burden and individual vulnerability that heat-health planning has to hold separately.

What to watch

Whether comparable analyses in other large national mortality datasets reproduce the cool-region finding. It is the result with the widest confidence interval in this study (1.13% to 11.53%) and the largest implication for where heat-health infrastructure gets built.

This article is informational and does not constitute medical advice.

Sources

• [s1] Guo MG, Liu JM, He GH, et al., "Heat Wave and Mortality Burden of Chronic Kidney Disease: Current Assessment and Future Projection in China," Biomedical and Environmental Sciences, vol. 39, April 2026 issue. https://doi.org/10.3967/bes2026.013