COVID-19 control measure effects suggest excess winter mortality is more sensitive to infection control than warmer temperatures.

BackgroundExcess winter mortality (EWM) has been attributed to both seasonal cold exposure, and to infectious disease. In 2020, New Zealands border management and lockdown measures successfully eliminated community transmission of SARS-CoV-2, and also largely eliminated influenza and many other respiratory viruses. This study investigates the contribution of infections and temperature to EWM and typical extended winter (May to October) deaths in this natural experiment created by New Zealands COVID-19 pandemic response. MethodsWe used age-standardised weekly deaths to measure EWM 2011 to 2019, then used historical patterns to estimate high, medium and low scenario 2020 EWMs. We then modelled typical year and 2020 heating degree day: mortality relationships to estimate relative contributions of cold temperature and infection to typical EWEDs. ResultsEWM 2011 to 2019 averaged 14.7% (low 11.4%, high 20.9%). In contrast, 2020 EWM was estimated at 1.6%, 2.7%, or 3.8% under high, medium, and low spring-summer mortality scenarios. Between 2011 and 2019, temperature was estimated to explain 47% of extended winter deaths, and infection 27%, with the remaining 26% attributable to the interaction between infection and temperature. DiscussionThe society-wide response to COVID-19 in 2020 resulted in a major reduction of winter mortality in this high-income nation with a temperate climate. In addition to influenza, other respiratory pathogens likely also make a significant contribution to EWM. Low cost protection measures such as mask wearing (eg, in residential care facilities), discouragement of sick presenteeism, and increased influenza vaccine coverage, all have potential to reduce future winter mortality. Research in contextO_ST_ABSEvidence before this studyC_ST_ABSExcess winter mortality (EWM) is a widely observed phenomenon, commonly attributed to physiological responses to short and long-term outdoor and indoor cold exposure (and associated increased air pollution); other seasonal physiology changes; and higher incidence of some infectious diseases. Previous estimates of EWM in New Zealand range from 10.3% to 25.6%, with influenza estimated to make up roughly a third of that excess. Internationally, deaths attributable to cold temperatures are also found outside the traditional winter period, with influenza making a large contribution to cold temperature deaths. Added value of this studyThis study finds that following a successful COVID-19 elimination strategy, which simultaneously prevented the annual winter influenza season, and likely other winter respiratory infections, New Zealand is likely in 2020 to experience less than a third of the usual winter mortality excess. Further, this study for the first time estimates the relative contributions of cold temperature and infection, and the interaction between the two, to New Zealand winter deaths. We estimate that of the 9.5% fewer deaths than in typical years recorded between 1 May and 31 October 2020, 92.5% were prevented by infection control measures; 1.4% by the 1.14{degrees}C warmer than average winter; and 6.1% by the interaction between infection and low temperature. Implications of all the available evidenceInfluenza and other infectious respiratory pathogens appear to make a much larger contribution to winter mortality than previously recognised. Low cost protection measures such as mask wearing (eg, in residential care facilities), discouragement of sick presenteeism, and increased influenza vaccine coverage, all have potential to reduce future winter deaths, and lower overall annual mortality rates.