Objective:

The article examines associations between mean daily ambient temperature and heart failure (HF) hospitalizations among US adults aged 65 years or older.

Approach:

• Study: Kuhrt et al. analyzed daily mean ambient temperature and HF hospitalizations among adults aged 65 years or older using a 50% random sample of Medicare fee-for-service beneficiaries from 2007–2022.
• Design and exposure assessment: The analysis included approximately 1 million patients and 1.5 million hospitalizations. It used a case-crossover and time-series approach, assessed lagged associations and regional variation, and measured daily temperature at the ZIP-code level.

Key Findings:

• Compared with 75 °F, cumulative exposure to moderate cold (25th percentile) was associated with 16% higher odds of HF hospitalization; extreme cold (5th percentile) was associated with 12% higher odds.
• The largest cold-associated estimates occurred in the West North Central, West, Southwest, and Northwest regions.
• Hot-temperature estimates (99th percentile versus 75 °F) were imprecise and crossed the null, except in the South, where exposure was associated with 5% lower odds of hospitalization.
• The analysis did not show a U-shaped or J-shaped temperature–hospitalization relationship.

Interpretation:

The article discusses several possible explanations for the heat findings, including use of daily mean temperature, adjustment for relative humidity, and death as a competing event. Regional cold estimates may not be directly comparable because the study used a fixed 75 °F reference despite differing regional temperature distributions. The authors of the commentary note that moderate cold was more consistently associated with increased HF hospitalization than extreme heat.

Limitations:

• Daily mean temperature may not capture physiologically relevant heat stress; humidity and other environmental factors affect thermal load.
• Adjustment for relative humidity may bias heat estimates toward the null if humidity lies on the causal pathway.
• Death may compete with hospitalization and contribute to apparently imprecise or protective heat associations.
• A single 75 °F reference temperature across regions limits direct comparison of regional estimates.
• Ambient temperature may not represent personal exposure, particularly where climate control is used.

Conclusion:

Kuhrt et al. found that moderate cold exposure was associated with increased odds of HF hospitalization among older adults, while estimates for extreme heat were imprecise. The commentary calls for further investigation of temperature–hospitalization exposure-response patterns and notes that moderately cold conditions may be more consequential than uncommon extreme-cold episodes because they occur more often.

Sources:

• Source article context