Coronavirus disease-19 mortality among solid organ transplant recipients in the United States during June and December 2020: Comparison of Organ Procurement and Transplantation Network and National Death Index data

To the Editor:

As of February 17, 2023, 1.1 million deaths due to coronavirus disease-19 (COVID-19) were documented in the United States.¹ Although solid organ transplant recipients (SOTRs) are highly susceptible to COVID-19 mortality due to immunosuppression and medical comorbidities,² the number of COVID-19 deaths occurring among SOTRs in the United States is not well quantified. Studies reporting on COVID-19 mortality among SOTRs in the United States have relied on data from the Organ Procurement and Transplantation Network (OPTN).^{3 , 4} Although the OPTN covers all organ transplants in the United States, causes of death (CODs) are unadjudicated and missing for a substantial proportion of reported deaths.⁵

To assess how well OPTN data capture the COVID-19 mortality burden among SOTRs, we linked a sample of deaths reported by the OPTN in the Scientific Registry of Transplant Recipients (SRTR) to the National Death Index (NDI). We included deaths occurring from June 1, 2020 (2 months after an OPTN COD code for COVID-19 was introduced) through December 31, 2020 (last date when NDI data were available).

Because SRTR COD data are organized by transplant, there were 14 237 death records for 12 608 deceased people during this period (11 183 single-organ and 1425 multiple-organ recipients). Half (n = 7151) of these death records were missing an SRTR COD (Table 1 ). We submitted 2375 records randomly selected within 6 COD categories for NDI linkage, where the sampling fraction varied by SRTR COD category, from 7.9% to 100.0% (Table 1). The NDI returned valid matches for 2303 death records (97.0%). Among death records with an SRTR code or free text indicating a COVID-19 death as the primary COD, NDI linkage confirmed COVID-19 as the COD in 97.9% and 93.3%, respectively. In addition, 77.8% of death records with COVID-19 listed as a secondary/contributing COD in the SRTR had an NDI COD of COVID-19. Importantly, 7.9% to 15.4% of deaths in other sampled categories were identified by the NDI as due to COVID-19 (Table 1).

Table 1.

Comparison of SRTR and NDI COD during June and December 2020, N = 14 237 death records among 12 608 people.

SRTR COD category	Analyses based on one death record per transplant				Analyses based on one death record per personc
	Total death records	Records sampled for NDI linkage (% sampled)a	NDI records returned	NDI COD indicated on death record as COVID-19,bn (%)	Total deathsc	Estimated COVID-19 deathsc,d
Primary COD
COVID-19	1267	100 (7.9)	94	92 (97.9)f	1226	1200
“Other, specify”, with free text specifying COVID-19	120	64 (53.3)	60	56 (93.3)f	115	108
Respiratory or multiorgan failure, or other/unspecified infection	895	470 (52.5)	458	60 (13.1)	740	97
Any other cause	4804	740 (15.4)	723	57 (7.9)	4137	326
Unknown cause or death from non–OPTN source	7151	1001 (14.0)	968	149 (15.4)	6310	971
Secondary/contributing COD
COVID-19	85e	85e (100.0)	81e	63 (77.8)e,f	80	62
Total	14 237	2375 (16.7)	2303	414 (18.0)	12 608	2764

Abbreviations: COD, cause of death; COVID-19, coronavirus disease-19; NDI, National Death Index; OPTN, Organ Procurement and Transplant Network; SRTR, Scientific Registry of Transplant Recipients.

^aWe sampled fewer records (n = 100, 64, and 85) in the SRTR COD categories where we believed almost all records would be identified as COVID-19 by the NDI, and we sampled more death records (n = 470, 740, and 1001) in all other SRTR COD categories.

^bCOVID-19 deaths were identified based on International Classification of Diseases, Tenth Revision (ICD-10) code U07.1.

^cFor the 95 multiple-organ recipients who had SRTR CODs that varied across their death records, we assigned a single COD by preferentially selecting the SRTR COD that had the highest proportion confirmed to be COVID-19 by the NDI. The order of preference was as follows (all CODs are primary CODs unless otherwise noted): COVID-19; “Other, specify”, with free text specifying COVID-19; COVID-19 as a secondary/contributing COD; unknown cause or death from non–OPTN source; respiratory or multiorgan failure, or other/unspecified infection; and, finally, any other cause.

^dFor each COD category, the estimated number of COVID-19 deaths was obtained by multiplying the percentages that appear in the column labeled “NDI COD indicated on death record as COVID-19” by the figures that appear under “Total deaths” (eg, 0.979 × 1226).

^eThe death records in this category overlap with those in the primary COD categories and are then excluded from the total counts.

^fConsidering NDI as the gold standard, these percentages are the positive predictive values for the SRTR primary or secondary/contributing COD for identifying COVID-19 deaths.

SRTR CODs were consistent across all death records for 1330 (93.3%) of the 1425 multiple-organ recipients. To estimate the total number of COVID-19 deaths, the proportions of CODs confirmed to be COVID-19 by the NDI were applied to the number of deceased SOTRs in each SRTR COD category and summed across categories (Table 1). We thereby estimated that COVID-19 was the COD in 2764 (21.9%) of the 12 608 SOTRs who died during June and December 2020. Our estimate of 2764 is comparable with that of Massie et al⁵, who calculated the number of excess deaths over the same period (2550). By contrast, only 1421 COVID-19 deaths were captured as primary or secondary/contributing SRTR CODs.

In conclusion, COVID-19 was estimated to be the COD for more than 1 in 5 deaths among SOTRs during the period investigated. Our analysis demonstrates that OPTN/SRTR data did not capture approximately half of COVID-19 deaths because of missing and incorrectly assigned CODs. Our findings indicate that (1) future analyses of COVID-19 mortality among SOTRs should not be solely based on OPTN/SRTR COVID-19 codes, and (2) there is an opportunity for the OPTN/SRTR to improve COD ascertainment.

Funding

This work was funded by the Intramural Research Program of the National Cancer Institute. This work was conducted under the auspices of the Hennepin Healthcare Research Institute (HHRI), contractor for the Scientific Registry of Transplant Recipients (SRTR), as a deliverable under contract no. 75R60220C00011 (United States Department of Health and Human Services, Health Resources and Services Administration, Healthcare Systems Bureau, Division of Transplantation). The United States Government (and others acting on its behalf) retains a paid-up, nonexclusive, irrevocable, worldwide license for all works produced under the SRTR contract, and to reproduce them, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government. The data reported here have been supplied by HHRI as the contractor for SRTR. The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy of or interpretation by SRTR or the United States Government.

Disclosures

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.