Study population and setting
Authors grouped health conditions shown to be associated with increased risk of severe COVID-19 into 11 categories using WHO guidelines and the Global Burden of Diseases, Risk Factors, and Injuries Study (GBD). Authors also included older, healthy (i.e., no underlying condition) adults as a proxy for frailty. Then, using UN mid-year population estimates, authors estimated the number of individuals with underlying conditions by sex, age, and country for 188 countries. Authors estimated the percentage of each country’s population at increased risk with and without age standardization and adjusted for individuals with more than one condition using results from prior multimorbidity studies in China and Scotland. Authors estimated degree of risk among those at increased risk by defining “high risk” as those who would require hospitalization if infected, and assumed males were twice as likely to be at high risk compared to females.
Summary of Main Findings
Overall, authors estimated that 1.7 billion individuals, or 22% of the global population, have at least one underlying condition that places them at increased risk of severe COVID-19. If healthy adults (i.e., with no underlying health conditions) 50 years and older are included as a proxy for frailty, the percentage increases to 34%. Chronic kidney disease, cardiovascular disease, chronic respiratory disease, and diabetes were the most common among individuals 50 years and older. According to results from non-age standardized estimates, the population proportions at risk ranged from 16% of the population in Africa to 31% of the population in Europe. Authors estimated that 349 million, or 4% of the global population, were at high risk of severe COVID-19. This varied considerably by age group: the proportion of those at high risk among individuals 20 years and younger was 1/900 and the proportion among individuals 70 years and older was 1/5.
Authors generated uncertainty bounds by running sensitivity analyses, varying values for country population size, disease prevalences, and the fraction of each population with more than one condition.
Although GBD provides the prevalence of individuals with each underlying condition, it does not provide prevalence for individuals with more than one condition. Authors corrected for this in the estimates, but methods relied on data from Scotland and China, which may not be applicable to all locations included in the analyses. All hypertension conditions were included in the heart disease category, but this may have diluted the association between risk of severe COVID-19 and other diseases caused by hypertension, such as chronic kidney disease. Furthermore, the proportion of individuals with an underlying condition – and therefore at increased risk – may be underestimated if conditions are not diagnosed or unreported. Authors estimated the proportion of individuals at high risk (i.e., would require hospitalization if infected), but were unable to estimate the probability of such individuals ever being infected. Although included in supplementary material and discussed by authors, age-specific risks are largely uncommunicated in results. As authors indicate, this could misconstrue the risk for individuals in certain parts of the world with overall younger populations. In Africa, for example, although the overall proportion of the population at risk is lower than that in Europe, age-specific risks are similar or higher for several conditions.
To date, studies have largely focused on assessing underlying health factors that put individuals at increased risk of COVID-19 (e.g., diabetes, chronic respiratory conditions). To aid and inform policy makers regarding interventions for vulnerable populations, this study expands beyond these aims by quantifying the number and percentage of individuals at increased risk.
This review was posted on: 28 October 2020