Abstract
Introduction
The COVID-19 pandemic has been associated with substantial rates of all-cause excess mortality. The contribution of external causes of death to excess mortality including drug overdose, homicide, suicide, and unintentional injuries during the initial outbreak in the United States is less well documented.
Methods Using public data published by the National Center for Health Statistics on February 10, 2021, we measured monthly excess mortality (the gap between observed and expected deaths) from five external causes using national-level data published by National Center for Health Statistics; assault (homicide); intentional self-harm (suicide); accidents (unintentional injuries); and motor vehicle accidents. We used seasonal autoregressive integrated moving average (sARIMA) models developed with cause-specific monthly mortality counts and US population data from 2015-2019 and estimated the contribution of individual cause-specific mortality to all-cause excess mortality from March-July 2020.
Results From March-July, 2020, 212,825 (95% CI 136,236-290,776) all-cause excess deaths occurred in the US). There were 8,540 excess drug overdoses (all intents) (95% CI 5,106 to 11,975), accounting for 4% of all excess mortality; 1,455 excess homicide deaths (95% CI 708 to 2202, accounting for 0.7% of excess mortality; 5,492 excess deaths due to unintentional accidents occurred (95% CI 85 to 10,899, accounting for 2.6% of excess mortality. Though a non-significantly 135 (95% CI -1361 to 1,630) more MVA deaths were recorded during the study period, a significant decrease in April (525; 95% CI -817 to -233) and significant increases in June-July (965; 95% CI 348 to 1,587) were observed. Suicide deaths were statistically lower than projected by 2,067 (95% CI 941-3,193 fewer deaths).
Meaning Excess deaths from drug overdoses, homicide, and addicents occurred during the pandemic but represented a small fraction of all-cause excess mortality. The excess external causes of death, however, still represent thousands of lives lost. Notably, deaths from suicide were lower than expected and therefore did not contribute to excess mortality.
Introduction
All-cause excess mortality has been closely tracked during the COVID-19 pandemic. However, the contribution of external causes of death during the initial months of the US outbreak is incompletely understood.1–3 We analyzed monthly trends in drug overdose deaths, homicide, suicide, unintentional injuries, and motor vehicle accidents (MVAs) from 2015-2020, focusing on the first five months of the COVID-19 outbreak.
Methods
We measured monthly excess mortality (the gap between observed and expected deaths) from five external causes using national-level data published by National Center for Health Statistics (NCHS) including provisional mortality figures through July 2020 (released February 2021). Data from March-July, 2020 was aggregated by the NCHS into five groups: drug overdose (all intents), International Classification of Diseases, Version-10 codes X40-44, X60-64, X85, Y10-Y14; assault (homicide), U01-U02, X85-Y09, Y87.1); intentional self-harm (suicide). U03, X60-X84, Y87.0; accidents (unintentional injuries), V01-X59, Y85-Y86); motor vehicle accidents, V02– V04,V09.0,V09.2,V12–V14,V19.0–V19.2,V19.4–V19.6, V20–V79,V80.3–V80.5, V81.0– V81.1,V82.0–V82.1,V83–V86, V87.0–V87.8,V88.0–V88.8, V89.0,V89.2.4,5
To forecast all-cause and cause-specific expected monthly deaths from March-July, 2020, we used seasonal autoregressive integrated moving average (sARIMA) models developed with cause-specific monthly mortality counts and US population data from 2015-2019. We used the Akaike information criterion to select the best models. We plotted observed and expected deaths monthly with 95% CI estimated from sARIMA models. We estimated the contribution of individual cause-specific mortality to all-cause excess mortality by dividing cause-specific mortality by total all-cause excess mortality from March-July 2020.
This study used publicly available data and was not subject to institutional review approval and adhered to STROBE guidelines.
Results
From March-July, 2020, 212,825 (95% CI 136,236-290,776) all-cause excess deaths occurred in the US (1,409,826 observed; 1,197,001 expected). There were 8,540 excess drug overdoses (all intents) (95% CI 5,106-11,975, Figure 1a), accounting for 4% of all excess mortality; 1,455 excess homicide deaths (95% CI 708-2202, Figure 1b), accounting for 0.7% of excess mortality; 5,492 excess deaths due to unintentional accidents occurred (95% CI 85-10,899, Figure 1c), accounting for 2.6% of excess mortality. Though a non-significantly 135 (95% CI -1361-1,630) more MVA deaths were recorded during the study period, a significant decrease in April (525; 95% CI -817 to - 233) and significant increases in June-July (965; 95% CI 348-1,587) were observed (Figure 1d). Suicide deaths were statistically lower than projected by 2,067 (95% CI 941-3,193 fewer deaths), (Figure 1e).
Figure Title:
Cause-specific mortality in the United States, January 2015 through July 2020.
Figure Legend:
The solid black line indicates raw cause-specific death counts from January 2015-July 2020; the dashed green line represents the point estimate for cause-specific expected deaths from March-July 2020 using the seasonal adjusted model; the gray shaded area represents projected range (95% CI) of cause-specific deaths expected to occur during March-July 2020 using the seasonal adjusted model.
1A. Drug overdose deaths (all intents), United States, January 1, 2015 – July 31st, 2020.
1B. Homicide deaths, United States, January 1, 2015 – July 31st, 2020.
1C. Unintentional accident deaths, United States, January 1, 2015 – July 31st, 2020.
1D. Motor vehicle accident deaths, United States, January 1, 2015 – July 31st, 2020. 1E. Suicide deaths, United States, January 1, 2015 – July 31st, 2020.
Discussion
Using provisional mortality data, we find that excess deaths from external causes occurred during the pandemic but represented a small fraction of all-cause excess mortality. The excess external causes of death, however, still represent thousands of lives lost exceeding pre-pandemic trends. Notably, deaths from suicide were lower than expected and therefore did not contribute to excess mortality.
Explanations for these excess external deaths are uncertain. Drug overdoses and homicides may be related to economic stress. Pandemic-associated changes in access to substance use disorder treatments could be exacerbating mortality from overdoses.6 Decreases in MVA deaths in April coincided with less traffic, despite increases in drivers testing positive for drugs and alcohol and lower seatbelt use.3 Increases in MVA deaths in June-July occurred as traffic increased (though still below 2019 levels), likely reflecting higher-risk behaviors.3
A limitation of this study is that all 2020 data published by NCHS are considered preliminary. However, substantial changes to March-July 2020 data are unlikely.
The increases in external causes of death may have been preventable. These findings suggest the need for public health strategies, both now and in future outbreaks, to include attention on these and other socio-behavioral risks. These data suggest that enhanced safety measures may prove beneficial, not just during stay-at-home periods but during subsequent re-opening phases as well.
Data Availability
Dr. Faust had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The data are available to the public.
Data Statement and Author Contributions
Dr. Faust had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Faust, Barnett, Krumholz
Acquisition, analysis, or interpretation of data: Faust, Du, Mayes, Li, Lin
Drafting of the manuscript: Faust
Critical revision of the manuscript for important intellectual content: All authors.
Statistical plan and analysis: Faust, Du, Li, Lin
Administrative, technical, or material support: Mayes.
Supervision: Faust, Lin, Krumholz
Funding Statements
Faust: None.
Du: None.
Mayes: None.
Li: None.
Lin: Dr Lin reported working under contract with the Centers for Medicare & Medicaid Services.
Barnett: Dr Barnett reported being retained as an expert witness by government plaintiffs in lawsuits against opioid manufacturers. No other disclosures were reported.
Krumholz: In the past three years, Dr Krumholz received expenses and/or personal fees from UnitedHealth, IBM Watson Health, Element Science, Aetna, Facebook, the Siegfried and Jensen Law Firm, Arnold and Porter Law Firm, Martin/Baughman Law Firm, F-Prime, and the National Center for Cardiovascular Diseases in Beijing. He is an owner of Refactor Health and HugoHealth, and had grants and/or contracts from the Centers for Medicare & Medicaid Services, Medtronic, the U.S. Food and Drug Administration, Johnson & Johnson, and the Shenzhen Center for Health Information.