Uncoupling of all-cause excess mortality from COVID-19 cases in a highly vaccinated state.

Since March, 2020, excess mortality—the number of all-cause deaths exceeding the baseline number of expected deaths—has been observed in waves coinciding with COVID-19 outbreaks in the USA and worldwide.1, 2 However, after February, 2022, the reported number of COVID-19-associated deaths decreased despite a notable spring wave of infections primarily due to omicron subvariants (BA.2, BA.2.12.1, BA.4, BA.5). Until now, it has been unknown whether the spring, 2022, COVID-19 wave in Massachusetts, USA, was associated with all-cause excess mortality.

Accordingly, we assembled population data (2014–19) and weekly mortality data (January, 2015–February, 2020) provided by the Massachusetts Registry of Vital Records and Statistics (MRVRS) and applied seasonal autoregressive integrated moving averages to project the weekly number of expected deaths for the state for the pandemic period (Feb 3, 2020–June 26, 2022). We summed age-specific mortality to create state-level estimates and additionally corrected for the lower-than-expected state population owing to cumulative excess mortality recorded during the pandemic (for a more detailed description, see appendix p 1).4, 5, 6 Weekly observed deaths provided by the MRVRS are more than 99% complete for all study weeks. Case, wastewater, and hospitalisation data were accessed from publicly available databases.7, 8 Analyses were conducted with R (version 4.1.2). The MRVRS deemed the study exempt from institutional review board review.

In the 18-week period after BA.2, BA.2.12.1, BA.4, and BA.5 subvariants became prevalent (week ending Feb 27, 2022; appendix pp 2–3), there have been 0·1 excess deaths per 100 000 person-weeks, corresponding to 134 excess deaths (95% CI –921 to 1189; appendix pp 2–3), despite at least 226 857 newly recorded cases, as evidenced by corresponding substantial spikes in SARS-CoV-2 wastewater levels and changes in testing volume (appendix pp 2, 4–5). This value corresponds to a 97·3% reduction in excess mortality compared with the 8-week initial omicron (B.1.1.529) wave, during which excess mortality was 4·0 per 100 000 person-weeks (2239 excess deaths; 95% CI 1746–2733), and a 92·7% reduction in excess mortality compared with the combined 26-week delta (B.1.617.2) and delta-to-omicron transition periods, during which excess mortality was 1·5 per 100 000 person-weeks (2643 excess deaths; 95% CI 1192–4094). However, new COVID-19-associated hospitalisations continued to occur during this period (figure B ), either reflecting background community prevalence or that new COVID-19 cases were less severe or exacerbated chronic medical illnesses enough to require emergency care but not to cause proximate mortality.

Figure

All-cause excess mortality, COVID-19-associated hospitalisations, and confirmed COVID-19 cases in Massachusetts, USA

(A) Shows the cumulative all-cause excess mortality (blue line) plotted against weekly confirmed new COVID-19 cases (light blue line) in Massachusetts, USA, from February, 2020, through June, 2022. (B) Shows the cumulative number of COVID-19-associated hospitalisations (purple line) plotted against weekly confirmed new COVID-19 cases (light blue line) in Massachusetts, USA, from February, 2020, through June, 2022; hospitalisation data did not become available until May, 2020.

Since the initial COVID-19 outbreak in Massachusetts, USA, there have been periods without excess mortality, corresponding to times of low prevalence (figure A). However, we also have observed two substantial outbreaks not accompanied by excess mortality. The first instance, (late February–June, 2021) corresponded to the phased vaccine rollout period, during which the mean age of newly infected people dropped precipitously and prevalence among people older than 60 years was low (appendix pp 2, 6), probably temporarily reflecting exceptionally high vaccine-conferred protection against SARS-CoV-2 infection among the vaccine-eligible population. The second instance occurred late February–June, 2022). Unlike February–June, 2021, the mean age of newly infected people did not fall during the corresponding 2022 period, and in fact rose (appendix pp 2, 6).

The uncoupling of excess mortality and new COVID-19 cases, in the absence of decreases in the mean age of infected individuals (appendix pp 2, 6), suggests that in our highly vaccinated state, current levels of immunity are considerable, leaving many, if not most, individuals at high risk with substantial protection against the most severe outcomes of SARS-CoV-2 infection. However, given newly emerging variants and the unknown duration of protection from infection and vaccination, further monitoring is warranted.

HMK reported receiving consulting fees from UnitedHealth, Element Science, Aetna, Reality Labs, F-Prime, and Tesseract/4Catalyst; serving as an expert witness for Martin Baughman law firm, Arnold and Porter law firm, and Siegfried and Jensen law firm; being a cofounder of Hugo Health, a personal health information platform; being a cofounder of Refactor Health, an enterprise health-care, artificial intelligence-augmented data management company; receiving contracts from the Centers for Medicare & Medicaid Services through Yale New Haven Hospital to develop and maintain performance measures that are publicly reported; and receiving grants from Johnson & Johnson outside the submitted work. All other authors declare no competing interests. We thank the Registry of Vital Records and Statistics, Office of Population Health, Massachusetts Department of Public Health, for assistance with data acquisition.