key: cord-0842903-z1nz3sf9
authors: Hansen, Julia; Hawkins, Devan
title: Increase in Massachusetts deaths with ICD-10 codes associated with COVID-19 during the first four months of 2020
date: 2022-02-08
journal: Dialogues in Health
DOI: 10.1016/j.dialog.2022.100004
sha: d7306c351d55d5dcb5bdd624d55c6216fd69035c
doc_id: 842903
cord_uid: z1nz3sf9

Background Previous research has suggested that some COVID-19 infections and deaths have gone unrecorded, especially in the early days of the pandemic. Therefore, it is likely that people in Massachusetts were exposed to, infected with, and died from COVID-19 before the first death was recorded and that other deaths in early 2020 may have been due to COVID-19, but were not coded that way. This study sought to determine the number of deaths in the first 4 months of 2020 that may have been due to COVID-19, by comparing deaths with selected ICD-10 codes to the same time frame in 2019 and 2018. Methods Death certificate information was obtained for the first 21 weeks of 2018, 2019, and 2020. We calculated and compared the number of deaths for specific ICD-10 codes that may be related to COVID-19 during this time period for each year. Results There was a notable increase in deaths potentially related to COVID-19 between the 11th and 17th weeks of 2020 in comparison with the same time period in 2018 and 2019. Conclusions Even after Massachusetts began recording deaths as being due to COVID-19, the number of deaths that may have been due to the disease was higher than would have been expected based on data from the two preceding years. These findings may indicate that some COVID-19 deaths were not being recorded or that the pandemic was exacerbating other health issues.

Massachusetts reported its first death from COVID-19 on March 20. However, due to restrictions on testing availability and unclear information about how the illness presented, it is likely that there were more cases and deaths than were officially recorded. 1 A model by researchers at Northeastern University estimated that though Boston reported its second case on March 2, the city may have already had 2,300 cases at that time. 2 The possibility that there may have been increases in deaths related to COVID-19 that were not captured by standard reporting methods is supported by one study that found there had been nearly 300,000

excessive deaths beyond what would be expected by October 15, 2020 nationwide. Only twothird of these deaths were accounted for by . 3 Similarly, Krieger et al. found a surge in mortality beyond what would be expected from previous years in Massachusetts, which was not accounted for by COVID-19 deaths alone. 4 Some of these deaths may be due to indirect effects J o u r n a l P r e -p r o o f of the pandemic, such as individuals foregoing care due to financial strain or fear of infection. 5 However, some of the excessive deaths may also have been due to COVID-19 deaths not being correctly diagnosed and recorded. This study sought to determine the number of deaths in the first 4 months of 2020 that may have been due to COVID-19, by comparing deaths with selected ICD-10 codes to the same time frame in 2019 and 2018.

Death certificate information for all deaths occurring in Massachusetts (MA) during the first 21 weeks of 2018, 2019, and 2020 were obtained from the MA Registry of Vital Records and Statistics. Medical information on death certificates in MA is completed shortly after deaths by the medical professionals who either pronounced the death of or last treated the patient, hospital medical officers, or the MA medical examiner. Information on the death certificate is often obtained from the patients' medical records or speaking with family members and/or friends.

Because this study was concerned with deaths that may have been due to COVID-19, but not recorded as such, all deaths that included the ICD-10 code for COVID-19 (U07.1) listed as a primary or contributory cause of death were identified. The additional causes of deaths for these were reviewed. From this review, a list of 41 ICD-10 codes found to be associated with COVID-19 diagnosis were selected. These codes are show in supplementary table 1. These codes were grouped into four categories: infectious disease (B codes), respiratory disease (J codes), respiratory symptoms (R0 codes), and those not belonging to the previous categories.

These codes were chosen as those most likely to be recorded in cases where COVID-19 was the cause of death, but that diagnosis was unknown. Codes related to comorbidities that increased the likelihood of mortality due to COVID-19, such as cardiovascular disease and diabetes were not considered for the purposes of this study. It should be noted that deaths that included codes from multiple categories were counted once for each categorye.g. a person J o u r n a l P r e -p r o o f whose death was coded with both J18 and A419 is counted once in both the J and other categories, while a person whose death was coded both J18 and J209 is only counted once in the J category. Therefore, the sum of deaths in each category is greater than the actual number of deaths. Because this is true of the data for all three years, this will not contribute to a misleading trend. For 2020, the confirmed COVID-19 deaths were removed from analysis, as the data of interest were potential COVID-19 cases that had not been identified as such.

Each record includes the date of death, which were then categorized by week to make direct comparisons between the three years. For 2020, the first 21 weeks of the year were from 

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Unspecified mental disorder due to brain damage and dysfunction and to physical disease G93.1 Anoxic brain damage, not elsewhere classified R41.0 Disorientation, unspecified R43.0 Anosmia R50.9 Fever, unspecified R53.1 Weakness R53.8 3

Other fatigue R57

Shock, not elsewhere classified R69

Illness, unspecified R99

Ill-defined and unknown cause of mortality X59

Exposure to unspecified factor Z20.8 28 Contact with and (suspected) exposure to other viral communicable diseases 

All-Cause Excess Mortality and COVID-19-Related Mortality Among US Adults Aged

Hidden Outbreaks Spread Through U.S. Cities Far Earlier Than Americans Knew, Estimates Say