Joint UCLA-Harvard study led by PhD student Jay Xu finds substantial racial/ethnic disparities in years of potential life lost attributable to COVID-19

A joint UCLA-Harvard study, led by UCLA Biostatistics PhD student Jay Xu, quantified
racial/ethnic disparities in years of potential life lost (YPLL) attributable to COVID-19, finding a
consistent pattern across states of non-Hispanic Blacks and Hispanics experiencing
disproportionately high and non-Hispanic Whites experiencing disproportionately low COVID-
19-attributable YPLL.

The team comprised of researchers at the UCLA Fielding School of Public Health and the
Harvard T.H. Chan School of Public Health. UCLA faculty on the study include Tom Belin
(Biostatistics), Ron Brookmeyer (Biostatistics, Dean of the UCLA Fielding School of Public
Health), Marc Suchard (Biostatistics, Human Genetics, Computational Medicine) and Christina
Ramirez (Biostatistics).

Using years of potential life lost – an epidemiological quantity that measures mortality via the
difference between an upper reference age (75 for this study) minus age at death if positive and
0 otherwise – to measure mortality, Xu’s team analyzed COVID-19 mortality data from the
National Center for Health Statistics as of December 30, 2020, characterizing YPLL-based
disparities in two ways: (1) estimating percentages of total YPLL by race/ethnicity, contrasting
them with their respective percent population shares, and (2) estimating age-adjusted YPLL rate
ratios (RR) – anchoring comparisons to non-Hispanic Whites – in each of 45 states and the
District of Columbia (D.C.).

To estimate these quantities, an adaptation of a novel Monte Carlo simulation procedure to
quantify the uncertainty of YPLL-based estimates that was recently developed by Xu, Belin,
and Ramirez [1] was used. For comparison, Xu’s team also calculated the corresponding
percentages total deaths and age-adjusted mortality rate ratios. The results of the analysis
revealed that racial/ethnic disparities in the COVID-19 mortality burden are generally greater in
magnitude when measuring mortality in terms of YPLL compared to (age-irrespective) death
counts, reflecting the greater intensity of the disparities at younger ages. Xu’s team also found
substantial state-to-state variability in the magnitudes of the estimated racial/ethnic disparities,
suggesting that they are driven in large part by social determinants of health whose degree of
association with race/ethnicity varies by state.

The associated paper is currently undergoing peer review, but a preliminary non-peer reviewed
report is available on medRxiv, accessible at

[1] Xu, Jay J., Thomas R. Belin, and Christina M. Ramirez. "Uncertainty quantification of years
of potential life lost-based estimates from mortality data summarized as death counts within age
intervals." Annals of Epidemiology 55 (2020): 1-3. Available at: